editorial statement: the first year of the european journal of government and economics european journal of government and economics volume 1, number 2 (december 2012) issn: 2254-7088 103 editorial statement: the first year of the european journal of government and economics diego varela, university of a coruña, spain giacomo benedetto, royal holloway, university of london, united kingdom jose manuel sanchez-santos, university of a coruña, spain abstract in this editorial statement we present a balance of the first year of life of the european journal of government and economics. we discuss the main developments that concern the journal’s indexation by academic databases. we also comment on the approval of a code of publication ethics and malpractice. finally, we emphasise the dangers of excessive technical sophistication and the need to keep an integrated approach between the fields of political science and economics, according to the spirit of the journal. jel classification a11; a12; l17 keywords academic journal; economics, indexing; open access; political science european journal of government and economics 1(2) 104 with the publication of this second issue, it is a good time to present a balance of the first year of the journal. we thought that an editorial statement might be a good way to communicate with our fellow editorial board members, authors and readers. there have been some substantial changes since the first issue was published in june. we obtained an issn and took the first steps for indexation of our journal’s contents. the purpose was to increase the visibility and ease of use of the journal, thereby promoting its potential impact (see varela, 2012). these efforts gave some early fruits, and our journal contents are already indexed by google scholar, repec, worldcat and bielefeld academic search. another index that we would like to join but is taking longer than expected is the directory of open access journals (doaj). a particularity of this index, as its name indicates, is that it only lists journals that use a funding model that does not charge readers or their institutions for access. other than that, the formal requirements are that journals must have an issn and at least five research papers published. in practice, however, there is another factor that is making this an increasingly selective index. the doaj itself is not only open access, but nor does it charge indexed journals or publishers. as a consequence, there is great demand from journals to join this index, which cannot always be met by the limited staff at lund university in sweden, which is currently responsible for the maintenance of the index. as a result, there is backlog of journals to be indexed, among which ejge. this is a particular instance of rationing by waiting (see barzel, 1974), in which waiting can exclude new projects that do not have a firm vocation of continuity. other established indices, such as econlit, which used to be a very popular index in economics, the isi web of science, or scopus, also require a publication record of two or three years prior to indexation. the difference with the doaj is that the waiting periods and associated publication record requirements are not fixed but endogenous. in addition to the traditional requirements about periodicity and peer review, scopus also requires journals to publish a statement about their publication ethics and malpractice policy. we have published such a statement on our website, in which we cover the duties of authors, reviewers, and editors, and deal with issues such as replicability, acknowledgement of sources or conflicts of interest (ejge, 2012). writing this formal statement, basing ourselves on the experience of other major journals, publishers and associations, made us reflect about some of those important issues. this experience, which we initially thought of as merely a formal requirement, proved helpful when preparing this second issue, which includes an article about legislator shirking that is based on experiments with human subjects. the nine research papers included in the first volume of this journal include a fair geographical representation. our authors have come from germany, belgium, the us, the uk, france, spain, norway, italy, and hungary. so far as topics are concerned, we have covered many important issues, most of which are about the link between government and economics and thus can be of potential interest to political scientists and economists alike. however, we are somewhat concerned that the methodological differences between economists and political scientists may be creating an undesirable cleavage between both traditions, which we would like to avoid. in particular, some of the articles have a high component of technical sophistication, which may be discouraging some readers and potential authors. we believe that it is important to have a high standard of analytic rigour, but also think that technical sophistication should not be an end in itself. therefore, we would like authors to make an effort to convey their messages in a way that can be understood by less technically oriented readers. this view is consistent with that of some prominent economists that see a dangerous slip in their discipline. as steven levitt points out: european journal of government and economics 1(2) 105 there’s a lot of intellectual firepower going into economics today. at the same time, i think the profession has gone off the rails. the fetishistic infatuation with technicality and mathematical difficutlty is extremely unhealthy… i think that will make us increasingly irrelevant to the real world (interview with levitt in bowmaker, 2012: 246-7). therefore, we shall make an effort to reduce the unnecessary technical oversophistication of published papers, while retaining analytical rigour. we wish to encourage more political scientists to make submissions to our journal, which may require a rewording of the journal’s aims and scope. we believe this will focus the papers on important questions, and help reduce the current cleavage between economists and political scientists that goes against the spirit of our journal and social science in general. references barzel, yoram (1974) ‘a theory of rationing by waiting’, journal of law and economics 17(1): 73-95. ejge (2012) ‘editorial policies’, url (consulted dec. 2012): http://www.ejge.org/index.php/ejge/about/editorialpolicies. bowmaker, simon w. (2012) the art and practice of economics today: lessons from leading minds. cheltenham: edward elgar. varela, diego (2012) ‘the contribution of isi indexing to a papers’ citations: results of a natural experiment’, european political science advance online publication 5 october 2012, url: http://dx.doi.org/10.1057/eps.2012.29. editorial statement: why do we need a european journal of government and economics? european journal of government and economics volume 1, number 1 (june 2012) issn: 2254-7088 5 editorial statement: why do we need a european journal of government and economics? diego varela, university of a coruña, spain giacomo benedetto, royal holloway, university of london, united kingdom josé manuel sánchez-santos, university of a coruña, spain abstract in this editorial statement we explain the reasons for the creation of the european journal of government and economics. we argue that there is a general shortage of academic journals. although the new journal we are starting will not solve or significantly reduce this problem, it will represent a marginal step in the right direction. we also explain our views regarding some important aspects of the journal’s policy, such as indexing and open access. doi: https://doi.org/10.17979/ejge.2012.1.1.4273 european journal of government and economics 1(1) 6 introduction when we informed our colleagues about our intention to create the european journal of government and economics (ejge), we found mixed reactions. on the one hand, established academics that had no difficulty in publishing their research in major journals wondered why we would need another journal, but they were ready to invest in the new project. on the other hand, junior academics that had difficulties in getting their work published could potentially benefit more from the creation of a new journal, but at the same time were more concerned about the risks involved in such an initiative. but all of them offered their support for the new venture. the list of colleagues that have accepted to be members of the editorial board of the journal provides a good representation of both kinds of profiles, and of the journal’s cross-disciplinary and international vocation. the list includes young and well as more established researchers, political scientists and economists working in academia and public policy. it includes colleagues from a dozen different countries. we are grateful to all of them. so far as the scope of the journal is concerned, ejge aims to be an international journal for peer reviewed research on all aspects of government and economics. ejge is particularly interested in current issues regarding the interrelationship between the fields of government and economics, from the influence of government on the economy (economic policy) to economic explanations of government and politics (public choice). it is also particularly interested in questions directly or indirectly related to europe. contributions will be valued by the substantial relavance of their research questions as well as their originality. the latter may consist in the development of new models to explain existing data, the aplication of existing models to new data sets, or a combination of both. some of the points raised above merit additional clarification. on the one hand, it should be noted that the european focus of the journal is not limited to eu-level issues, but also includes national and subnational questions, and the papers included in this first number are a good indication of this. on the other hand, although the journal aims to strengthen the link between theory and empirical evidence by encouraging model-based research, there is nothing intrinsically quantitative or qualitative about this approach. but a basic premise is that the papers should deal with some empirically falsifiable hypothesis about the observable world, be it political or economic. the shortage of academic journals the creation of this new journal is based on the belief that there is a general shortage of journals, and that this shortage is hindering the dissemination of existing research and, what is more serious perhaps, discouraging research in the first place. the shortage or abundance of journals is measured by price and in the case of academic publishing, the price for authors is not only easy to measure because it is not only paid in money, but also in terms of waiting. of course, this general shortage is not felt in the same way by everyone, as the reactions to the announcement of the journal indicate. it is particularly younger academics or academics from non-anglophone countries that have greater difficulties to get their work published by internationally recognised high-impact journals. academic impact is often only computed for a limited number of high quality journals. this does not mean that other journals do not make an impact. it just means that they are excluded because thay do not meet a minimum standard. currently, such minimum standard is the inclusion in isi’s journal of citation reports (jcr). much like minimum wage requirements do not affect individuals and regions equally, jcr standards will have a differential effect on authors and european journal of government and economics 1(1) 7 regions. the jcr minimum standard will have no effect on many established academics from anglophone countries, whose contributions will be valued by their academic impact. however, for those academics and whole regions for which the bar is too high, such minimum standards will mean a sort of censorship. it would be naïve to think that the creation of this journal will resolve the general shortage of academic journals or have any significant impact on the price of publishing. the main reason is the journal’s broad scope and international vocation encompassing political and economic science, which means that it will represent a negligible fraction of the relevant market. as a mere price taker, ejge does not aspire to change the world but to have a marginal contribution in the right direction. in order to do so, it will have to comply with the existing quality standards, and proper indexing is one of them. indexing in a recent paper one of the editors investigates to what extent isi web of science (wos) indexing contributes to the impact of an academic paper by analysing the results of a natural experiment (varela, 2012). the statistical tests indicate a significant effect of isi indexing on the number of citations received by individual papers, which could be increased by a factor of two. the conclusion is that isiindexing does not simply provide an objective measure of academic impact, but also affects academic impact itself. this is so even in the case of papers that are covered by other major indexing and search services such as scopus or google scholar, which means that in spite of the increasing competition from alternative providers, there is still no full substitute for the isi web of science. if a journal indexing service has the power to multiply the academic impact of a journal by two, and indexing by other providers cannot compensate for that, then it has an extraordinary amount of power. this result may have policy implications for public authorities that go beyond the scope of this editorial statement, but also has practical implications for us as editors, and we take good note of them. open access one of the distinctive features of this journal is its open-access (oa) policy, which is believed to have a significantly positive effect on academic impact. most analyses compare the citation rates of articles that have been made open access by their authors with other articles published in the same journals that have not been made open access, finding significant statistical difference in the average citation rates, and conclude that there is a causal effect of open access on the academic impact of a paper (eg antelman, 2004). but the results of such analyses may be biased if assignment to the treatment group (in this case, open-access) is not random but based on self-selection by the authors, which is not independent from potential outcomes in terms of academic impact. as a result, the observed difference in average academic impact may not be a good indicator of the average treatment effect on the treated because it will also include what is known as selection bias (angrist and pischke, 2009). it could happen, for instance, that authors who choose to make their papers open-access may be those most concerned with their academic impact, and probably use other means in parallel to increase the impact of their papers, whose effect will be confused with that of open access. randomised experiments such as those used in the natural sciences play an important role in uncovering causal effects, because random assignment solves the selection problem by making selection independent of potential outcomes (angrist and pischke, 2009). a recent randomized controlled trial of open access publishing, involving 36 participating journals in the sciences, social sciences, and humanities, reports that articles placed in the open access condition (n=712) european journal of government and economics 1(1) 8 received significantly more downloads and reached a broader audience within the first year, yet were cited no more frequently, nor earlier, in isi-indexed journals than subscription-access control articles (n=2533) within 3 years. this contradictory result may be explained by social stratification, a process that concentrates scientific authors at a small number of elite research universities with excellent access to the scientific literature (davis, 2011). but open-acces journals also have disadvantages vis-à-vis those hosted by commercial publishers. prominently among them is the fact that open-access journals give up a precious source of finance, namely the subscription fees paid by university libraries. better finance allows for better services such as proofreading and copyediting. it is not for ideological but for practical reasons that we start this project as an open-access journal, and we do not exclude that at any point in the future the journal, if successful and if the market allows, might migrate to a commercial publisher's platform. that is happening at the moment with some old spanish journals, but thinking about that at this early point might be like the milkmaid's tale. at this moment we are only thinking about adding an option for colleagues to be able to publish their research and doing it as well as we can so that the journal has a high impact, and open access is just an aid in this regard. conclusion with this editorial statement we have wanted to express the reasons that have led us to the creation of this new journal and explain some of its main defining features. however, the journal has a life of its own. its evolution will be determined by the contributions of colleagues. references angrist, joshua d. and jörn-steffen pischke (2009) mostly harmless econometrics: an empiricist's companion. princeton, nj: princeton university press. antelman, kristin (2004) 'do open-access articles have a greater research impact?', college & research libraries 65(5): 372-82. davis, phillip m. (2011) 'open access, readership, citations: a randomized controlled trial of scientific journal publishing', the faseb journal 25(7): 2129-34. varela, diego (forthcoming) ‘the contribution of isi indexing to a paper’s citations: results of a natural experiment’, european political science. microsoft word ejge_02_02_043.doc european journal of government and economics volume 2, number 2 (december 2013) issn: 2254-7088 95 editorial statement: the first two years of ejge diego varela, university of a coruña, spain and alexandru ioan cuza university of iasi, romania giacomo benedetto, royal holloway, university of london, united kingdom jose manuel sanchez-santos, university of a coruña, spain abstract in this editorial statement we explain the developments in the european journal of government and economics in its second year of existence. in this period, the journal has continued its international expansion, entered some new indices and started to experience its impact, as measured by the number of citations in other journals. after two years of publication, we have also learned some lessons that should help us redefine the role of ejge. jel classification a12; a30; p27 keywords academic publishing; academic journal; internationalisation; indexing; academic impact european journal of government and economics 2(2) 96 introduction one year has passed since our last editorial, in which we discussed our efforts to include the journal in some indices in order to improve its visibility and impact. we also discussed the geographical base of the journal, as well as our intention to make it accessible to a broader economic and political science audience by avoiding unnecessary technical oversophistication (varela et al., 2012a). in the current year, we have continued our work along the lines of expanding the journal base, both geographically and methodologically, and working for inclusion by abstracting and indexing services. we have also harvested our first fruits in terms of academic impact measured by citations received by our papers in other major publications. in our first editorial we also discussed the shortage of academic journals and the role of ejge in filling this gap (varela et al., 2012b). the experience of two years of publication allows us to test this statement and redefine the role of our journal in the current context. the rest of this editorial statement is structured to cover three major issues: the expansion of the journal base, the efforts to improve the journal’s indexing and impact, and the contribution of ejge to the academic world. expanding the journal base since the start of this project two years ago, we aimed to be an international journal, as we highlighted in the journal’s aims and scope published on our website. in our editorial statement of december 2012 we pointed out that the nine research papers in the first volume featured authors from germany, belgium, the us, the uk, france, spain, norway, italy, and hungary (varela et al., 2012a). the nine research articles in this volume include authors from four new countries: finland, romania, nigeria and russia. this means that, not without difficulty in some cases, we are trying to expand the geographical base of the journal. in our last editorial statement we also aimed to increase the methodological base of the journal in order to make it more accessible to a broader audience. in the first issue of this volume we included a qualitative analysis of the european commission’s role in three pieces of legislation concerning the regulation of financial services (hartlapp and rauh, 2013). in the same issue we also included a interesting paper on ceausescu’s population policy based on the historical analysis of archival material, including the minutes of the meetings of the central committee of the romanian communist party in 1966 (soare, 2013). the topic would seem outdated if it were not because only a few days ago the spanish government approved a legislative proposal to restrict abortion to an extent not since before 1985. many would like to have access to the minutes of the spanish council of ministers that approved the proposal but, like in the romanian case, we shall have to wait many years until the secret expires. the current issue of the journal also includes a qualitative analysis of the european union’s common agricultural policy and a historical paper on russian natalism. indexing and impact when we started the journal, we set proper indexing as one of our major objectives, in order to assure the visibility of the journal’s contents and increase their academic impact (varela et al., 2012b). in our second editorial statement one year ago we reported the developments in this respect, and mentioned the directory of open access journals (doaj) as our next target (varela et al., 2012a). well, after one year we can say that the objective was accomplished, and we are now indexed and abstracted by the doaj at a journal and article level. in the meantime, our journal was also indexed by econis (a german index) and rated by index copernicus (a polish journal index). we also applied for inclusion in varela, benedetto and sanchez-santos ● the first two years of ejge 97 latindex (a directory of latin american, spanish and portuguese journals) and are currently waiting for their assessment. our next major target is scopus. scopus is the second most selective index, after web of science, and one of the best ones to increase the visibility of articles and measure their academic impact. one of the general requirements to apply for inclusion in scopus is to have a publication record of two years, which we meet with the publication of this issue. as in the case of the doaj, the assessment procedure will take several months, so in practice it is possible that the two-year requirement will become a three-year one. the other minimum criterion was to have a publication ethics and malpractice statement, which we introduced one year ago. scopus selection is carried out by experts using both quantitative and qualitative measures and the selection is partly based on sample documents from the title. the selection criteria are grouped into five categories: journal policy, content, journal standing, regularity, and online availability. the criteria included in each of these broad categories are summarised in table 1. table 1. scopus journal selection criteria category criteria journal policy convincing editorial policy type of peer review diversity in geographical distribution of editors diversity in geographical distribution of authors content academic contribution to the field clarity of abstracts quality and conformity with stated aims and scope readability of articles journal standing citedness of journal articles in scopus editor standing regularity no delay in the publication schedule online availability content available online english language journal home page available quality of journal home page source: scopus.com. we agree with the general desirability of the different criteria and we aim to fare well in all of them irrespective of scopus selection policy. two of the criteria are clearly quantitative. one of them is the geographical distribution of authors, an area in which we feel we are particularly strong, at least compared to other journals already listed in scopus. the second quantitative criterion that deserves special attention is the citedness of journal articles in scopus. our journal is still young to assess accurately the number and quality of citations received by its papers. a two year window after the year of publication is generally used as an indicator, but it is clear that in most journals the peak of citations comes later. for instance, in the case of european union politics, the peak came in the fourth year (varela, 2013). in our case, as we are not yet indexed by scopus or the web of science, we use google scholar to monitor the citations received by our papers, and, at the moment of writing, five of them have received citations. the paper by fiorino, galli and petrarca (2012) on corruption and growth merits special attention. at the time of writing, it already counts five citations, three of which are in journals listed in scopus (petrarca and ricciuti, 2013; caruso and baronchelli, 2013; padovano, 2013). this is a remarkable achievement considering that this paper was published less than a year ago. all in all, the fact that five out of 14 papers have started to receive citations so early is very promising. redefining the role of ejge the experience of two years of publication, and the fact that we have had to do everything by ourselves without the help of a professional publisher, has helped us acquaint ourselves with the tasks involved in running an academic journal. it is not european journal of government and economics 2(2) 98 until one has to do everything by oneself that one realises how much work is involved, and one starts to understand the reasons why it is so expensive for authors (sometimes in terms of money and often in terms of revisions and waiting) to get their work published. it has been hard, but it would be impossible to get the same experience otherwise. one of the lessons we learned in this process is that reputation building is a slow process, and the quality of the work originally submitted by authors and their willingness to work on the revisions is directly related to this reputation. this is one of the reasons why the journal has depended on the good will of editorial board members who have submitted papers or encouraged others to submit in order to maintain the quality of the first issues of the journal (special thanks to them). and it is also one of the reasons why we have had to work so hard in the constructive review of other papers (special thanks to our reviewers too). for people outside this inner group, the journal is not more than a set of indicators. that is why every little step counts (a new issue published, a new index joined, a new citation received…). we hope that this new issue will also be a small step in this direction. but irrespective of the initial push from our contacts and the progressive reputation increase of the journal, it is clear that we shall have to be ready to fish in less travelled waters. the need to expand our journal base beyond its current limits should come at no surprise to colleagues familiar with the current geographical distribution of journals, and the current (and projected) demographics of potential authors and readers around the world. there are countries and entire regions (some of them in continental europe) that are currently undersupplied of quality academic journals. getting out of the confort zone may be disturbing at first, but research has always benefited from such endeavours. references caruso, raul and adelaide baronchelli (2013) 'economic aspects of the complementarity between corruption and crime: evidence from italy in the period 1996–2005', international journal of monetary economics and finance 6(2): 24460. fiorino, nadia, emma galli and ilaria petrarca (2012) 'corruption and growth: evidence from the italian regions', european journal of government and economics 1(2): 126-44. hartlapp, miriam and christian rauh (2013) 'the commission’s internal conditions for social re-regulation: market efficiency and wider social goals in setting the rules for financial services in europe', european journal of government and economics 2(1): 25-40. padovano, fabio (2013) 'distribution of transfers and soft budget spending behaviors: evidence from italian regions', public choice: 1-19. petrarca, ilaria and roberto ricciuti (2013) 'the historical economics of corruption and development within italy', international journal of monetary economics and finance 6(2): 186-202. soare, florin s. (2013) 'ceausescu’s population policy: a moral or an economic choice between compulsory and voluntary incentivised motherhood?', european journal of government and economics 2(1): 59-78. varela, diego, giacomo benedetto and jose m. sanchez-santos (2012a) 'editorial statement: the first year of the european journal of government and economics', european journal of government and economics 1(2): 103-5. varela, diego, giacomo benedetto and jose m. sanchez-santos (2012b) 'editorial statement: why do we need a european journal of government and economics?', european journal of government and economics 1(1): 5-8. varela, benedetto and sanchez-santos ● the first two years of ejge 99 varela, diego (2013) 'the contribution of isi indexing to a paper's citations: results of a natural experiment', european political science 12(2): 245-53. microsoft word ejge_03_02_014.doc european journal of government and economics volume 3, number 2 (december 2014) issn: 2254-7088 138 are political support-driven policies always bad? the case of large interest groups gael lagadec, university of new caledonia, new caledonia abstract the action of active interest groups (lobbies) has been traditionally considered to be a source of harmful waste for the economy which reduces social well-being. can this analysis be adapted to the case of large unorganised groups which do not ask for anything directly? or, on the contrary, does the setting up of policies which improve the situation of these large groups permit an improvement in social welfare? we start from classical (public choice) analyses of lobbying and rentseeking developed since the 1970s, closely linked with the hypothesis of reelection-seeking politicians, before extending our analysis also to consider nonsector-specific policies and passive interest groups (notably those too large to meet the olsonian condition of efficient collective action). then the research question to be answered becomes whether promoting the interest of large groups can deliver social welfare as defined by the incumbent’s social welfare function. we refer to the political cycles’ evidence to consider that no social welfare objective can motivate the favouring of large groups. jel classification d71; d72. keywords interest groups; lobbies; collective action; social welfare. acknowledgements financial support from the fonds pacifique of the french overseas ministry is gratefully acknowledged. the usual disclaimer applies. l lagadec ● are political support-driven policies always bad? 139 introduction free trade is generally considered to be the optimal policy for a small open economy. if free trade is associated with lump sum transfers between individuals, it is then in the interest of each individual in an economy (e.g. hillman 1989: 1), and is a pareto optimum at the international level. nevertheless it has traditionally been recognised that a few exceptions to this rule exist (e.g. protection of emerging industries, optimal tariff), but for which the practical application can appear very limited (e.g. magee, 1997: 532). if social welfare is maximised for zero protection, positive protection is suboptimal. how can this situation be explained, as in reality international exchanges are hindered by numerous barriers, and no state practices free trade? (magee, 1997: 542) why would a country choose a suboptimal situation for itself rather than an optimal solution? firstly, the case of a suboptimal situation in which each of the agents finds themselves in a situation inferior to that of the optimal situation, must be excluded. in fact, whichever way economic and trade policy are formed, they come from at least one individual, if not from a majority. an effective suboptimal policy thus advantages certain agents and disadvantages others. this policy is thus implemented by the agents who benefit from it. the idea of a benevolent government must therefore be dismissed; as social welfare is not the criterion which guides the policy (otherwise free trade would prevail). a protectionist trade policy, whatever the degree, is therefore shaped by private interests. thus protectionism must be the result of a political process.1 trade policy is the privileged instrument of political support; the other instruments at the disposal of government have the disadvantage of being too transparent (grossman and helpman, 1995: 680). however there is no reason to think that trade policy is the only lever of political support. in this article the concept of endogenous policy is extended to economic policy in the widest sense. in this context the effects on welfare of policies implemented by re-election seeking politicians, or more simply those looking for political support, will be discussed to answer this question: in the case of interest groups driven political support would the traditional negative link between lobbying and welfare be reversed? in the first section the principle of lobbying will be summarised and the concept of political support will be extended to unorganised interest groups. in the second section the commonly acknowledged effects of lobbying on social welfare will be presented and the link between the satisfaction of unorganised interest groups and social welfare will be emphasised. the last section shows the empirical relevance of our approach. lobbies and interest groups: a new look at endogenous policies small versus large groups for private interests to shape economic policy, it has been traditionally considered that they need to organise themselves into pressure groups (lobbies). apart from the pressure itself, the activity of the lobby is rent-seeking, which corresponds to obtaining a (beneficial) non-competitive situation. in the case of trade policy, tariffs are endogenous when they are established in response to the demands of lobbies, 1 this has been acknowledged for a long time; see baldwin (1984: 674), grossman and helpman (1994: 835), hillman (1992: 7), mayer (1984: 983), peirce (1991: 276), pincus (1975: 763), wellisz and wilson (1986: 369). european journal of government and economics 3(2) 140 and we can then talk about endogenous protection2 (for empirical evidence of the endogenous aspect of protection see hillman, 1989: 133-149). a distinction must be made between pressure groups (lobbies) and interest groups. a lobby is primarily an interest group but, unlike the latter, the members of the group have joined forces to organise a collective action in order to collectively pursue their individual common interest. in fact, according to baldwin (1982) there are a certain number of basic factors for a lobby to appear (necessary conditions, especially including olsonian conditions3 related to small-size groups) but, according to him, a triggering factor is necessary to initiate the change from potential lobby (that is the interest group) into active lobby (pressure group). this "trigger" could be a crisis, or a series of crises striking the common interests of the members of the potential lobby (baldwin, 1982: 279). this means that the individuals organise themselves more naturally or more efficiently to counter losses than to seek income (realistic hypothesis that an individual perceives a loss more acutely than the loss of potential gain). once the lobby is set up, it can become more efficient and organise itself, not only against losses or risks of losses, but also to obtain advantages (which particularly lead to theories of "rent seeking"). conversely, interest groups have not overcome the obstacles related to collective action, which is most often due to their characteristic of large size. in the literature on political support it is generally considered that everything that is endogenous comes from lobbies. an original feature of this paper is to consider taking the interests of large groups into account on the grounds of political support, which thus establishes endogenous policies which are not sector-based ("global policies"). if this idea is not widespread it is nevertheless not new, and it seems that this demarcation criterion between endogenous sector-based or global policies could be credited to an article by travis (1968). in this way travis showed that american tariffs specific to labour protection were prohibitive whatever the sector concerned (travis, 1968: 640). this study meant that if one considered a large unorganised group obtaining protection, the protection could be explained other than in terms of lobbying. the large group could in fact be protected by the government because it represents significant electoral weight. as browning (1974: 376) then pointed out the government can elaborate legislation to win votes, even without lobbying. one may think considering political support in terms of number of ballots would make political support and social welfare converge (the political process being a way to aggregate individual preferences, despite its weaknesses known since condorcet’s paradox). however, legislation made to win votes from an identified population group is the expression of an "endogenous" policy in that it is not driven by the search for social welfare (which comes back to the fact that social welfare is not an addition of individual well-being or of community interest in a benthamian way). 2 the counterpart of the protection is either promises of financial contributions given to political candidates by the lobbies before the election (a typical approach to electoral competition, pioneered by brock and magee, 1978) or recurring financial contributions given to the incumbent by the lobbies to fine tune the shape of protection they benefit (a typical approach to "influence driven contributions" pioneered by peltzman, 1976, and formalised by helpman and grossman, 1994, 1995). 3 the cases in which collective action could emerge are the cases where at least certain production factors are not mobile ("specific" production factors). these specific production factors form the common interest of the different producers of the same product. however the existence of specific production factors is not a sufficient condition for the formation of lobbies (grossman and helpman, 1995: 681). if the specific factor is concentrated among few members, the incentive for collective action is strong, which is close to the condition of small size for the formation and efficiency of lobbies. see olson (1965) for this condition of efficiency through small size. l lagadec ● are political support-driven policies always bad? 141 the pioneer analysis of travis thus corresponds to "endogenous" policies because they are guided by unorganised interest groups (which are therefore not pressure groups). this means that in terms of political support, interest groups should be taken into account by their electoral weight (passive influence) and not by their lobbying, which would necessitate collective action.4 from large groups to endogenous economic policies the object of this paper is to extend the concept of endogenous economic policy to its widest sense, and not simply limit it to trade protection, to understand whether in this context the traditional negative link between lobbying and welfare would be reversed (see introduction for the sub-optimality of protection)? the definition of an endogenous economic policy is thus close to that of endogenous protection, but it must however be extended in two complementary and related dimensions: (1) by considering that political support in its widest sense is at the base of endogenous policies; and (2) by extending the analysis beyond the perimeter of sector-oriented policies, thus distinguishing global endogenous policies and sector-oriented endogenous policies. these two dimensions are interrelated in that (unorganised) interest groups are most often large groups (cf. the small size condition for the success of collective action). this feature of size automatically has the effect of increasing the number of agents involved and thus increasing the number of sectors concerned leads to global and not exclusively sector-oriented policies. in the model of endogenous protection, an economic policy is defined as endogenous as soon as it is implemented according to identified interest groups (interest groups or pressure groups/lobbies)5 and not for the consideration of social welfare (classical model of the benevolent dictator). if it is interest groups, it is more in an electoral context (the groups are not taken into consideration for their contribution like in peltzman (1976), but for their electoral weight during the future election); if it is lobbies, it corresponds to a classical case of lobbying thus including the role of contributions. the first case corresponds to what we call "global endogenous policies", the second corresponds to "sector-oriented endogenous policies". lobbying, interest groups and social welfare classical analysis of the effects on social welfare the analysis of the effects of lobbying on the social welfare was initiated by browning (1974). a source of inefficiency thus appears as lobbying leads to diversion of resources in non-productive activities.6 this diversion corresponds to the contributions received by politicians to finance their campaigns, because it is money taken from a productive use. besides these contributions (standard peltzmanian case), rodrik (1986) identified three types of social welfare costs generated by lobbying: (i) the direct cost of lobbying, that is total amount of work diverted from productive activities; (ii) the absolute loss due to excess production 4 denzau and munger (1986) also showed that unorganised voters can be indirectly represented in the making of economic policy just because a contrario the lobbies prefer to work with legislator politicians who don’t have informed voters opposed to their views in their home districts. 5 for an historical example of the antagonism between private interests and social welfare in the case of endogenous trade policy, see the economist (2008). the article describes how the american free trade policy was challenged by the "tariff act" of 1930, voted following the initiative of willis hawley (congressman) and reed smoot (senator), in spite of a petition from president hoover signed by more than a thousand american economists. 6 these are dups popularised by bhagwati (1980), then bhagwati and srinivasan (1980). european journal of government and economics 3(2) 142 from the sector in competition with imports; and (iii) the loss in consumption caused by the increase in the domestic price (rodrik,1986: 295-296). following this, the social cost of lobbying was measured. for example laband and sophocleus (1988) showed that profit-seeking drastically reduced the growth of the american gnp by 22.6 percent for the year 1985 (also see horgos and zimmermann, 2009a on this point, and desai and olofsgard, 2011, for the specific case of developing countries). lobbying has been traditionally seen as being used to protect oneself against less expensive imports from the rest of the world. however the protection itself creates a situation of second best. bhagwati was the first to analyse the problem of second best in a world with lobbying. bhagwati (1980) considered that trade protection was either exogenous (established without lobbying), or endogenous (established with lobbying) and, in this context, showed that this departure from free trade is not necessarily synonymous with a decline in social welfare. indeed, the addition of one distortion (lobbying) to another (protection) means that lobbying can have a negative social cost (bhagwati, 1980: 357).7 this paradox was later criticised, particularly by tullock (1981), who showed that from the time one also considers the matter of customs revenue from the tariff, the paradox is reduced. finally the paradox disappears if one considers the hypothesis which is widespread in the literature, that the activity of profit-seeking is perfectly competitive.8 such a hypothesis has a corollary of dissipation of the income through the expenses involved in obtaining it; the optimum second best aspect thus disappears. anam and katz (1988: 220) showed it starting from monopoly rent, by emphasizing that in this case, the marginal cost of production in the monopolistic sector (including the cost of rent seeking) is reflected in the price. unorganised interest groups and social welfare let us now come back to the distinction between "global" and "sector-oriented" policies. the difference is not only semantic (in which case it would suffice to consider "global" and "partial" policies so that the full sense of this distinction can be understood. the distinction can indeed be considered to be deeper. on the one hand, one can consider the analysis of sector-oriented policies which ask the classical questions of aggregated preferences, or of groups of agents who work towards the general interest in their own way, but whose respective interests are in conflict. on the other hand, an analysis of the balance of political power within an economy, with the help of global policies, can be taken into consideration. in the latter case, the interests at stake are not necessarily conflicting and can thus all the more work towards the general interest because taking them into account allows the incumbents to have a function of social welfare (fsw) in keeping with the structuring of society, or at the very least not disconnected from social reality. furthermore, unorganised groups carry weight by their vote and not by the contributions (cf. travis, 1968). in this way, they do not divert labour resources like rodrik, or (especially) contributions like peltzman. one can therefore consider that the influence of interest groups on economic policy does not include any negative element. figure 1 shows that the curve of utility possibilities confronts the two individuals. the figure’s shape comes from the fact that the incentives to invest vary according 7 that means that the factors diverted from their productive use have a negative shadow price. this point is explained in detail by hillman (1989, pp. 53-58) and more especially taken up again by hillman and katz (1994: 104). 8 cf. foster (1971: 177), for a definition of competitive rent seeking and cf. corcoran and karels (1985: 227), for the general nature of this concept. more precisely, monissen (1991: 125) studies the rent dissipation conditions. l lagadec ● are political support-driven policies always bad? 143 to redistribution – see wolfelsperger (1985) for example. point a corresponds to a benthamian fsw (addition of utilities), point c to rawls’ solution, point d to the egalitarian solution and point b to any fsw. figure 1: utilities and function of social welfare let us consider point b, which could correspond to taking unorganised interests into account. one is thus in an exogenous context, in which a benevolent dictator imposes his fsw that authoritatively settles questions of intransitivity linked to the collective aggregation of individual preferences. the information here thus goes from interest groups to the incumbent, who best calibrates his fsw. therefore, although interest groups are passive, they can have a positive impact in terms of social welfare. one thus faces, in some ways, the issue developed in the theories of informative lobbying (austen-smith and wright, 1992; bennedsen and feldmann, 2006; or lagerlöf, 2006). however, whatever the function of social welfare considered, it corresponds to the social preferences of the incumbent and can thus not vary according to the electoral cycles. if it is noted that one group is more assisted just before an election that means that policy is not driven by anf fsw. as soon as the ruler is rational, if his objective function is not an fsw, it can only be a function of political support (possibly camouflaged as an fsw outside the election period: political speech advocating equity, social justice, etc.). so, one is actually in a context of endogenous policy. private interests, possibly conflicting, thus influence the incumbent who takes them into account according to his function of political support and not according to an fsw. empirical relevance what do we see in practice? if one looks for an increased activity of the lobbying kind before an election, one will get plenty of examples that simply match the opportunist government model.9 the problem is to differentiate systematically what relates to actions in favour of lobbies from what relates to actions in favour of (large) interest groups. there is a way to discriminate between these types of actions, which has been developed in recent works that are mainly empirical and some of which have not been published yet. lagona and padovano (2008) show for the italian case that parliamentary activity is connected to the political cycle with more laws being voted at the end of the cycle (just before the next election). brechler and geršl (2011) analyse the political legislation cycle in the czech republic and bring out a two-peak cycle with 9 aidt, veiga and veiga (2011: 35) have also recently demonstrated that the opportunistic pre-election distortions are connected to the prospect of winning elections: the smaller the win-margin, the higher pre-election distortions. european journal of government and economics 3(2) 144 legislation activity increasing after the elections followed by a decrease around the middle of electoral cycle. the interpretation of brechler and geršl (2011: 20) is first, that incumbents keep their pre-election promises (the first peak after elections), and second, that they maximise political support by voters (the second peak just before next elections). as the analysis focuses on laws connected to transfer payments, that should concern large groups and not specific interests. so, there is a strong presumption that this empirical evidence confirms the relevance of our analysis that large interest groups passively shape political support-driven policies. other recent works allow a finer analysis of the differences between favours to the lobbies and favours to the interest groups. thus padovano and petrarca (2012) make an explicit distinction between legislation in the form of laws visible to all agents and decrees visible only to special interest groups. they consider that in order to collect election campaign resources from special interest groups, a legislator generates a pre-electoral cycle of the targeted good by distorting upwards the production of decrees. but, mainly, they show that the legislator generates an electoral cycle of the general public good at the end of the legislature by distorting upwards the production of laws to increase his probability of being reelected (padovano and petrarca 2012: 4). this analysis is tested by padovano and gavoille (2013) on french data covering the monthly counts of legislative production from 1959 to 2012 and providing detailed characteristics of the composition of the government as well as personal information about the ministers. the highest peaks of legislative production (and not decrees) occur towards the end of the legislature. all other things being equal, the legislative production increases by roughly 17 percent in the last year of the legislature, while this increase reaches 13 percent during the last six months of the presidential mandate. as the constitutional reform of 2000 gave the political mandates a time stability by synchronizing the legislative and presidential period, the authors also note that the peak of legislative production is particularly high when the legislature lasts its natural duration length. (p. 8). it means that the 17 percent increase does not come from a "rush to the end" (padovano and gavoille, 2013: 8 and 17) padovano and petrarca (2012) and padovano and gavoille (2013) allow us to discriminate between actions in favour of the lobbies and actions in favour of unorganised large interest groups. it is then shown that the legislator changes his agenda to favour the large groups at the end of his mandate and thereby distorts his economic policy before the election to make his policy closer to the interest (although not expressed) of large groups. these empirical analyses demonstrate that, independently of questions of intransitivity, it is certain that the influence of unorganised groups does not contribute to social welfare, even if they do not generate diversion of productive resources (dups), since the latter, whatever the form it is given, is not the target of the incumbent (for that we can work on social welfare without being blocked by the arrow’s theorem). in this way, contrary to what it could seem, taking the interests of large groups into account tends to reduce social welfare. conclusion the activity of interest groups is a source of waste of resources. in exchange for contributions paid (on the whole corresponding to waste), these activities aim at obtaining advantages from public authorities, particularly corresponding to obtaining or keeping market power (a classical case of endogenous protection: lobbies organise themselves to apply pressure with a view to customs tariffs which protect them from international competition). therefore the economy can no longer be at the optimum and social welfare deteriorates. apart from a few paradoxical situations, this point is commonly accepted in the literature. in this type of situation, l lagadec ● are political support-driven policies always bad? 145 the efficiency of lobbying is inversely proportional to the size of the group (olsonian condition): small groups are advantaged (particularly in that their incentives for collective action are stronger, particularly because their per capita gain is higher). this is why large groups are inefficient and do not manage to organise themselves into lobbies. this is especially the case for consumers whose associations have a very weak influence in relation to the number of agents they represent. nevertheless these large groups can also be taken into account by those in power in a perspective of direct political support,that is through votes (endogenous economic policies in the widest sense influenced by active or passive groups). moreover it is noted that the attention these groups benefit from depends on the political cycle (padovano and petrarca, 2012; padovano and gavoille, 2013). as the function of social welfare of the incumbent cannot be subject to these uncertainties (preferences are by definition exogenous), this means that actions in favour of large groups are driven by political support.10 it must be concluded that policies driven by political support, which take the diverse interests of big groups into account, move the economy away from the optimum and reduce social welfare, in much the same way as classical lobbying would do. references aidt, toke s., francisco j. veiga and linda g. veiga (2011) ‘election results and opportunistic policies: a new test of the rational political business cycle model’, public choice 148: 21-44. austen-smith, david a. and john, r. wright (1992) ‘competitive lobbying for a legislator's vote’, social choice and welfare 9(3):229-257. baldwin, robert e. (1982) ‘the political economy of protectionism’, in jagdish n. bhagwati (ed.), import competition and response, chicago: university of chicago press, pp. 263-292. bennedsen, morten and sven e. feldmann (2006) ‘informational lobbying and political contributions’, journal of public economics 90(4-5): 631-656. bhagwati, jagdish n. (1980) ‘lobbying and welfare’, journal of publicl economics 14(3): 355-363. bhagwati, jagdish n. and t. n. srinivasan (1980) ‘revenue seeking: a generalization of the theory of tariffs’, journal of political economy 88(6):10691081. brechler, josef and adam geršl (2011) ‘political legislation cycle in the czech republic’, ies working paper 21/2011, ies fsv, charles university. browning, edgard k. (1974) ‘on the welfare cost of transfers’, kyklos 27(2): 374377. corcoran, willian. j. and gordon v. karels (1985) ‘rent-seeking behavior in the long-run’, public choice 46 (3): 227-246. denzau, arthur t. and michael c. munger (1986) ‘legislators and interest groups: how unorganized interests get represented’, american political science review 80(1): 89-106. farvaque, étienne and sonia paty (2009) economie de la démocratie, bruxelles: de boeck edition université. 10 furthermore, considering that taking large groups into account can be seen as informative lobbying (which increases social welfare) also comes up against the olsonian condition of efficiency (small size): passive large groups do not appear to be able to efficiently produce information aiming at the promotion of their interests. european journal of government and economics 3(2) 146 desai, raj m. and anders olofsgard (2011) ‘the costs of political influence: firmlevel evidence from developing countries’, quarterly journal of political science 6(2):137–178. foster, edward (1971) ‘the treatment of rents in cost-benefit analysis’, american economic review 71(1): 171-178. grossman, gene m. and elhanan helpman (1994) ‘protection for sale’, american economic review 84(4): 833-850. grossman, gene m. and elhanan helpman (1995) ‘trade wars and trade talks’, journal of political economy 103(4): 675-708. hillman, arye l. (1992) ‘international trade policy: benevolent dictators and optimizing politicians’ public choice 74: 1-15. hillman, arye. l. and eliakim katz (1994) ‘risk-averse rent seekers and the social cost of monopoly power’ the economic journal 84: 104-110. hillman, arye l. (1989) the political economy of protection, london: harwood academic publishers. horgos, daniel and klaus w. zimmermann (2009a) ‘interest groups and economic performance: some new evidence’, public choice 138:301-315. horgos, daniel and klaus, w. zimmermann k. w. (2009b) ‘it takes two to tango: lobbies and the political business cycle’, working paper series, nr./no. 98, september, department of economics; fächergruppe volkswirtschaftslehre. katz, eliakim and jacob rosenberg (1989) ‘rent-seeking for budgetary allocation: preliminary results for 20 countries’ public choice 60: 133-144. laband, david n. and john p. sophocleus (1988) ‘the social cost of rentseeking: first estimates’ public choice 58: 269-275. lagerlöf, johan n. m. (2006) ‘a theory of rent seeking with informational foundations’, discussion paper series, cepr n° 5893. lagona, francesco and fabio padovano (2008) ‘the political legislation cycle’, public choice 134: 201–229. magee, stephen p. (1997) ‘endogenous protection: the empirical evidence’, in dennis c. mueller (ed.), perspectives on public choice, cambridge: cambridge university press, pp.526-561. mayer, wolfgang. (1984) ‘endogenous tariff formation’, american economic review 74(5): 970-985. monissen, hans g. (1991) ‘rent-seeking in general equilibrium: a didactic illustration’, public choice 72: 111-129. olson, mancur (1965) the logic of collective action: public goods and the theory of groups, cambridge, ma: harvard university press. padovano, fabio and ilaria petrarca (2012) ‘a theory of cyclical production of laws and decrees’, working paper 2012-09, condorcet center for political economy, university of rennes 1. padovano, fabio and nicolas gavoille (2013) ‘the dual political legislation cycle in france’, working paper 2013-02, condorcet center for political economy, university of rennes 1. peirce, williams s. (1991) ‘after 1992: the european community and the redistribution of rents’, kyklos 44(4): 521-536. peltzman, sam (1976) ‘toward a more general theory of regulation’, the journal of law and economics, 19(2): 211-248. l lagadec ● are political support-driven policies always bad? 147 pincus, jonathan j. (1975) ‘pressure groups and the pattern of tariffs’, journal of political economy 83(4): 757-778. rodrik, dani (1986) ‘tariffs, subsidies, and welfare with endogenous policy’, journal of international economics 21(3-4): 285-299. tollison, robert d. (1982) ‘rent seeking: a survey’, kyklos, 35(4): 575-602. travis, william p. (1968) ‘the effective rate of protection and the question of labor protection in the united states’, journal of political economy 76(3): 443-461. tullock, gordon (1981) ‘lobbying and welfare: a comment’, journal of international economics 16(3): 391-394. wellisz, stanislaw and john wilson (1986) ‘lobbying and tariff formation: a deadweight loss consideration’, journal of international economics 20(3-4): 367375. wolfelsperger a. (1995) economie publique, paris: puf microsoft word ejge_02_02_043.doc european journal of government and economics volume 2, number 2 (december 2013) issn: 2254-7088 137 value added tax and price stability in nigeria: a partial equilibrium analysis marius ikpe, federal university, ndufu-alike, ikwo, nigeria alwell nteegah, university of port-harcourt, nigeria abstract the economic impact of value added tax (vat) that was implemented in nigeria in 1994 has generated much debate in recent times, especially with respect to its effect on the level of aggregate prices. this study empirically examines the influence of vat on price stability in nigeria using partial equilibrium analysis. we introduced the vat variable in the framework of a combination of structuralist, monetarist and fiscalist approaches to inflation modelling. the analysis was carried out by applying multiple regression analysis in static form to data for the 1994-2010 period. the results reveal that vat exerts a strong upward pressure on price levels, most likely due to the burden of vat on intermediate outputs. the study rules out the option of vat exemptions for intermediate outputs as a solution, due to the difficulty in distinguishing between intermediate and final outputs. instead, it recommends a detailed post-vat cost-benefit analysis to assess the social desirability of vat policy in nigeria. jel classification e03; e62; e63 keywords value added tax; price stability; partial equilibrium analysis; static model; nigeria european journal of government and economics 2(2) 138 introduction one of the means by which government increases its internally generated revenue is value added tax (vat). this is a tax on the supply of goods and services which is eventually borne by the final consumer, but collected at each stage of the production and distribution chain. vat as a concept was first introduced by france in 1954, and has over time been embraced by well over 70 countries. it has in recent time become a major source of revenue in many developing countries, including the sub-saharan african countries. shalizi and squire (1989) found that vat accounted for about 30 percent of total tax revenue in ivory coast, kenya and senegal in 1982. bogetic and hasan (1993) found that indonesia introduced vat in 1983, and by 1988, the ratio of vat revenue to gdp had risen to 4.5 percent. this impressive record in virtually all countries where it was introduced clearly influenced the decision to introduce vat in nigeria in january 1994 as a replacement for the existing sales tax. it was imposed on all goods manufactured in nigeria, as well as on the goods that were imported and sold domestically. the federal inland revenue service (firs), the agency in charge of tax administration in nigeria, pointed out that vat is a consumption tax that is relatively easy to administer and difficult to evade, and has been embraced by many countries (firs, 1993a; 1993b; 1993c). in this context, it becomes necessary to empirically examine the likely macroeconomic impact of vat administration. evidence so far supports the view that vat is already a significant source of revenue in nigeria. for instance, vat revenue in the year of its inception (1994) was n8.194 billion, which was 36.5 percent greater than the projected n6 billion for that year (ajakaiye, 1999). however, the members of the organised private sector have been voicing their reservations in the sense that vat is taking a toll on the prices of their products. from an economic point of view, one expects the price of goods subject to vat to rise, however, beyond this expected rise, businesses are taking advantage of the existence of vat to increase prices of goods and services arbitrarily. according to aruwa (2008), the resulting price increase has led to higher inflation. this may have prompted mclure (1989) to state that policy makers should be concerned about the macroeconomic impact of vat, especially on prices, output, income and consumption, before considering its adoption. a few empirical works on the subject exist in the context of the nigerian economy. ajakaiye (1999) undertook the most detailed study for nigeria, including an extensive investigation of the impact of vat on key sectoral macroeconomic aggregates, by using a computable general equilibrium (cge) model of the nigerian economy. unfortunately, the study was carried out when vat was only six years old in nigeria, too early to get reliable conclusions on its impact on other macroeconomic aggregates. besides, from 1999 to date, the economic environment in nigeria has undergone a number of changes. for instance, there was a transition from a military to a democratic regime. as a result of the aforementioned factors, there is need for a re-examination of the possible macroeconomic impact of vat in nigeria, especially on the price level. the results of investigations of this kind will provide a basis for minimising the adverse effects of vat, while consolidating its benefits. this study seeks to assess the macroeconomic impact of vat on general price levels in nigeria by means of a partial equilibrium analysis. in particular, the study specifically seeks to estimate the price elasticity of vat. the paper first gives an overview of vat and inflation in nigeria and reviews the empirical literature. next, it deals with methodology, conceptual issues and data, before presenting the empirical results and a discussion. finally, the conclusion follows and presents some policy implications and a research agenda. ikpe and nteegah ● value added tax and price stability in nigeria 139 overview of the nigerian economy overview of vat and its administration in nigeria vat was introduced in nigeria in 1993 by the vat act no. 102 of 1993 as a replacement for the sales tax that was in operation in the federal capital territory. it was the outcome of dr. sylvester ugo’s study group on indirect taxation in november 1991. it was designed as a consumption tax payable on goods and services consumed by individuals, government agencies or business organisations. nigeria operates a vat rate that does not synchronise with the economic community of west african states (ecowas) protocol. ecowas adopted a uniform vat protocol due to the constant movement of people and goods across the countries in the region, and the need to be subject to similar conditions. under nigeria’s influence, the ecowas advisory rate has been reduced to 10 percent, while nigeria, despite being a signatory of the protocol, currently operates the lowest vat rate across the sub-region, at 5 percent. it should be noted that, since its inception in 1993 and subsequent implementation in 1994, the income tax burden has been reduced twice (company income tax from 35 to 30 percent and personal income tax from 30 to 24 percent), but the vat rate has remained static at 5 percent. exemptions granted to reduce the distortions of vat create a lot of complexities, lack of transparency and arbitrariness in terms of application and enforcement. hence, the government is affected by high levels of exemptions and a low vat rate. the nigerian gross product vat model is one that tries to maximise tax by disallowing cost. it however allows for restrictions on the recovery of vat paid on capital terms given that the cost of capital is amortised and spread across items. currently, 17 categories of goods and 24 categories of services are vat-eligible and all imports are vat-eligible, whether raw materials or finished goods. the benefit of vat in nigeria since its inception cannot be overemphasised. it has increased the revenue of nigeria over time, so that it is now the third highest revenue earner for the federal government, next to company income tax and petroleum profit tax. secondly, it has also reduced the tendency of tax evasion. however, there are increasing complaints from various quarters, especially the organised private sector, about the effects of vat on their operating costs and the prices of their products (ajakaiye, 1999).1 inflation and central bank of nigeria policy response inflation had its bitter toll on the nigerian economy, and monetary and fiscal policies among others have been developed to reduce it. the central bank of nigeria (cbn) has the statutory responsibility of formulating and implementing monetary policy with an emphasis on price stability. the inflationary trend has been cyclical since the mid-1970s, peaking in 1988, 1989, 1992, 1993, 1994, 1995, 1996, 2001 and 2005. the implementation of the recommendations of the udoji committee of 1975 and the hosting of black and african festival of arts and culture (festac) in 1977 constitute two events that had significant impacts on the history of inflationary trend in nigeria. the udoji committee, which doubled the basic minimum wage in the public sector in 1975, represents a climax in inflationary tendencies that led to widespread strikes and unrest in the private sector, on which the udoji recommendations were not binding. this cost push factor further crippled productivity and increased inflation as the economy’s level of productivity could not match increased money supply and aggregate demand. on the other hand, the hosting of festac in 1977 helped to compound the problem of macroeconomic instability, as the resulting public spending undermined the government’s traditional 1 kakaki online magazine, 4 august 2011; peoples daily newspaper, 5 april 2013. european journal of government and economics 2(2) 140 objective of ensuring fiscal discipline, and the resulting inflationary pressure continued unabated (fatukasi, 2005). in the past, the government under various administrations has adopted different measures for dealing with the situation. one of the measures taken by the monetary authorities (cbn) in recent times was a move from its traditional monetary policy to inflation targeting. since the last obasanjo regime, the primary objective of the authorities has been the attainment of a single digit inflation rate. with this, the authorities have been able to bring down the rate of inflation to a relatively mild level. the inflation-vat nexus the current study critically examines and analyses the existing relationship between inflation and vat using trend analyses for each of the aggregates. the result of this is shown in figure 1 below. figure 1. evolution of inflation and vat revenue, 1994-2010 0 10 20 30 40 50 60 70 80 19941995 19961997 1998 1999 2000 20012002 2003 2004 20052006 2007 2008 2009 2010 year ra te ( % ) 0 50000 100000 150000 200000 250000 300000 m il li o n s o f n g n inflation (left axis) vat (right axis) vat revenue has been on the increase over the whole study period ever since its introduction into the nigerian economy in 1994. inflation, on the other hand, is observed to have decreased steadily to a lower level in the year 2000, from a previously higher level in 1994 and 1995. it increased again in 2001, but decreased thereafter in 2002, after which it remained stable up to the year 2005. beyond 2005, inflation decreased to an all time low in 2007, increased again in 2008, but then remained stable until 2010. from the analysis above, which fleshed out the trend of vat within the study period, and the behaviour of inflation that follows, a naïve analyst will quickly be drawn to the conclusion that inflation does not respond to vat policy outcomes under the nigerian context within the period under investigation. however, it may not be the case, given the fact that inflation in modern economies is never left to take its natural course, but is rather subjected to various anti-inflationary measures aimed at maintaining stability in prices. in other words, inflationary statistics reflect more the impacts of measures by the monetary authorities aimed at ensuring price stability. the conclusion drawn from the foregoing is the fact that the trend analysis might not truly have reflected the existing possible connection between vat policy and the level of prices in nigeria within the period being investigated. this might be due to the influence of monetary anti-inflationary measures by the cbn to ensure stable prices. ikpe and nteegah ● value added tax and price stability in nigeria 141 empirical literature few empirical works exist on the subject of vat especially in developing countries. extensive studies have been done on the impact of indirect taxation in developing countries in general, and nigeria in particular. naiyeju (1996) argues that the benefits from any tax depend on the extent to which it is properly managed. how tax law is interpreted and implemented, as well as its publicity, will determine how a particular tax is able to meet its objectives. the concern about the economic impact of vat led shoup (1989) to argue that it is all the more important because it may cause consumers to reduce their consumption of certain commodities that have direct and/or indirect effects on labour productivity. at a different level, some authors have raised arguments in favour of the computable general equilibrium (cge) model, which they argue is preferred to partial equilibrium analysis for the assessment of the impact of vat on any given economy. mclure (1989) was the first to highlight his preference for the cge model, which is an economy-wide framework that incorporates the interactions and feedbacks among demand, production and income. within this model, the relevant variables are adjusted until production and consumption decisions are consistent. following the same line of argument, ajakaiye (1999) undertook an analysis of the impact of vat on key sectoral macroeconomic aggregates, using a cge model that he argues to be suitable for nigeria. from the results of its model simulations for three scenarios, he concludes that the scenario where vat is treated in a cascading manner (i.e. where it is viewed as a cost) by the vat-eligible organisations, and vat revenue is re-injected into the economy, price, consumption, expenditure, output and income effects will be most deleterious, and this best approximates the nigerian situation. in a study of instability in government revenues and expenditures in less developed countries, lim (1983) observed that tax revenue instability was the major cause of expenditure instability within the period 1965 to 1973. bleaney, gemmel and greenaway (1995), with particular reference to sub-saharan africa, analysed the sources and the consequences of revenue instability in developing countries, and found that revenue instability is more common in poor, more open and more inflationary economies. furthermore, ebeke and ehrhart (2010) in a study on the sources and consequences of instability in tax revenues in sub-saharan african countries, using panel data for 39 countries over the period 1980 to 2005, give credence to bleaney, gemmel and greenaway (1995), guillanmont et al (1999), fatas and mihov, (2003), telvi and vegh (2005), furceri (2007), loayza et al (2007), thornthon (2008) and diallo (2009). ebeke and ehrhart (2010) argue that tax revenue instability in sub-saharan africa leads to public investment and government consumption instability, which in turn generates a lower public investment ratio, and is therefore detrimental to long-term economic growth. the need to specifically focus on the domestic economy, and vat in particular, may have led owolabi and okwu (2011), as reported by worlu and nkoro (2012), into the examination of the contribution of vat to the development of lagos state using simple regression models as abstractions of the respective sectors considered in the study. the study, which considered a vector of development indicators as dependent variables, regressing each on vat revenue, found that revenue contributed positively to the development of the respective sectors considered. it was however found to be statistically significant in the development of the agricultural sector only. unegbu and irefin (2011), also focusing on the domestic economy, carried out research on the impact of vat on economic and human development of emerging nations from 2001 to 2009, using regression discriminant analysis and anova. the outcome of the investigation revealed that vat has a significant impact on the expenditure pattern of adamawa state within the study period. european journal of government and economics 2(2) 142 from the foregoing analysis, ajakaiye (1999) remains the closest attempt at extensive examination of the macroeconomic impact of vat in nigeria. the study however is limited by the issue of the insufficient time lag necessary for proper alignment of the policy with other macroeconomic aggregates that is needed for better evaluation of vat policy in nigeria. this paper tries to overcome this limitation and some others, such as regime change, and it focuses on the impact of vat on the price level. methodology variables and data the dependent variable is inflation (inf), as measured by the annual rate of growth of the consumer price index (cpi), which is the most commonly used index in nigeria. the national bureau of statistics (nbs) computes the cpi for nigeria. the independent variables are inflation fundamentals, such as the fiscal deficits as a percentage of gdp (fd), the growth rate of the money supply (grm2), the real interest rate (rint), the value added tax (vat), and the real exchange rate (rexg). the data employed in the analysis span over the period 1994 to 2010. the inflation rate (inf) and fiscal deficit as a ratio of gdp (fd), and vat as well as broad money supply are sourced from the cbn’s annual statistical bulletin. the cbn publishes processed data on these aggregates annually. the real interest rate (rint) and real exchange rate (rexg) are sourced from the african institute of applied economics (aiae). processed data on these aggregates are available in the data bank of aiae (2010). broad money (m2) is adopted as the concept of money supply. its growth rate is calculated as a ratio of the amount by which the current period level of money supply differs from that for the preceding period level. model specification in modeling the impact of vat on the price level, we follow fatukasi (2005) in combining the structuralist, monetarist and fiscalist approaches to inflation modelling. the econometric equation is specified in dynamic logarithmic form as loginft = β0 + β1logfdt = β2loggrm2t + β3logrintt + β4logvatt + β5logrexgt + vt where log stands for the natural logarithm, v for the error term, t for the time parameter, and β1, β2, β3, β4, β5 are the elasticities of rates of fiscal deficits as a ratio of gdp, growth rates of money supply, real interest rate, and real exchange rates, respectively. a priori expectation is that β1, β2, β4, β5 >0, while β3<0. empirical analysis estimation results because macroeconomic variables are known to exhibit random-walk behaviour, we had to examine the stationarity properties of the variables in the model by means of an augmented dickey-fuller (adf) stationarity test. the results are presented in table 1. ikpe and nteegah ● value added tax and price stability in nigeria 143 table 1: results of augmented dickey-fuller (adf) stationarity test variables adf test statistics critical values % order of integration inf 5.121 1 i (1) fd 5.198 1 i (1) grm2 6.087 1 i (1) rint 3.710 1 i (1) vat 3.833 1 i (1) rexg 4.012 1 i (1) results of the adf test show that all the variables are stationary after first differentiating. this implies that the variables are integrated as order 1 (i.e. i(1)), and should enter the model in their growth forms (i.e in their differentiated forms). this is required by the fact that regressions using non-stationary variables lead to results that cannot be relied upon for predictions, so they must be differentiated. because the sample size for the study is not large enough to be considered for long-run analysis, the study avoided being drawn into the conventional test for long-run equilibrium relationships between the dependent and independent variables which would have been the case, given the coincidence of order of integration between the dependent variable and the set of independent variables. as a result, analysis herein is based on the static (multiple regression) model. moreover, when forecasting we do not bother about time frame, but about stationarity of data used, so that the data can forecast well, irrespective of the sample size. table 2: determinants of inflation (inf) independent variables/ constant coefficients t values c 60.879 1.248 fd 0.437 1.076 grm2 1.987 1.054 rint 1.082 12.385 vat 5.777 2.696 rexg 4.812 0.899 r2 = 0.97 adj-r2 = 0.95 f-statistic = 66.62 d–watson = 1.75 the ordinary least squares (ols) regression technique was employed in the estimation of the model. the estimation was carried out after making sure that the variables in their behaviours conform to the assumptions of the classical normal linear regression model (cnlrm). efforts were also made to ensure that the model adheres to the principle of parsimony using the akaike information criterion (aic) and the swartz bayesian criterion (sbc). results of the estimation as presented in table 2 show that two (rint and vat) out of the five explanatory variables are statistically significant at the conventional 5-percent level of significance, and have the theoretically expected signs. the coefficient of multiple determination (adjusted r2) is very strong at 0.95, which indicates the power of the explanatory variables in explaining variations in the rates of inflation. the model has very high f-statistics, showing that all the independent variables are non-zero at 95 percent level of confidence. the value of the durbin–watson statistics on the other hand reveals the presence of negative auto-correlation in the model that can be attributed to the quality of the data used. the result of the white heteroscedasticity test failed to reject the hypothesis of homocedasticity in the data, which variance can be assumed to be constant over time. from the correlation matrix, the model was observed to be free from any multicollinearity problem, given that all the pair-wise correlation values in the model fell well below the 0.8 rule of thumb mark. the jarque-bera (jb) test of normality failed to reject the hypothesis that the residuals are normally distributed. the jb statistic is given as 0.78276, and the probability of such a high value is 0.701778 . the conclusion from this is the fact that the residuals are normally distributed. european journal of government and economics 2(2) 144 discussion of results the analysis reveals that the main variable of interest, which is vat, exerts an upward pressure on price levels within the period studied. it is statistically significant and strong in influencing the rate of change in the price level. the strong positive impact on prices is most likely due to vat charges on intermediate outputs. it is however difficult to draw lines between these and final outputs in view of the fact that the final result of a given production process could be the input of yet another production process, whose outputs are successively subject to vat. when this happens, multiple vat burdens become the norm, and this translates to increases in prices of goods and services. from another perspective, the burden of vat for producers whose inputs are intermediate outputs amounts to an increased cost of production. the consequence of this is a reduction in the output level, and subsequent increases in unit prices, given the level of aggregate demand. ajakaiye (1999) has earlier reported that vat is more deleterious when viewed as a cost. furthermore, the result reveals a negative significant impact of real interest rates. the strength of this effect is, however, very weak in influencing the rate of change in price levels. this outcome reveals that although real interest rates significantly affect inflation in nigeria within the period studied, it is not a major factor to be considered when it comes to regulating the level of prices. an examination of this outcome buttresses the need for one to apply caution given that the weak impact of real interest rates (rint) on prices might have been as a result of the influence of funds from alternative sources of investment financing in nigeria. many investors are exploiting this to their benefit. notable among these sources of finance for investments are microfinance banks. these are banks created to meet the credit needs of a social segment for which the high cost of borrowing from the conventional banks (deposit money banks (dmbs)) does not adequately cater. credits offered by microfinance banks are often high-jacked by big time investors who now use it as means of getting funds at low rates of interest to compliment the high cost of funds available to them from dmbs. secondly, remittances constitute another alternative source of investment financing. nigeria is a country that both gives and receives remittances. greater proportions of remittance inflow are channelled into investments at little or no cost upon receipt, without necessarily having to pass through the banks. by this, more goods and services are produced at relatively low costs of production, which helps to reduce inflationary pressures at given levels of aggregate demand to the extent that the effect of interest rates on prices is weakened. apart from these variables (vat and rint), other variables are found to be insignificant in determining the rate of changes in the level of prices. among fd, grm2, and rexg, fiscal deficits as a percentage of gdp (fd) and the growth rate of money supply (grm2) are highly correlated in their individual effects on prices. from the theory, money supply is one means of financing deficits. degree of the impact of fiscal deficits on price level depends on the extent to which funds set aside for deficit financing are channelled to productive investments through the development of need and maintenance of existing social infrastructures. it is however unfortunate that rather than the development of new infrastructures and maintenance of existing ones, deficit funds are often diverted into private investments abroad, while some percentages of it are used mainly for external debt servicing. by these means, funds are drained away from the nigerian system. as a result, rather than stimulate the level of economic activities and possibly induce inflation given the level of aggregate demand, fiscal funds instead create a dampening effect on the level of economic activities, hence the negative impact of each of the macroeconomic aggregates on the level of prices in nigeria. in a similar development, the real exchange rate (rexg) is not significant in influencing the rate of changes in the price level. this is not unconnected with the influences of remittances already highlighted above, and the practice of quality trade-offs by nigerian importers in collaboration with their foreign producers. as a ikpe and nteegah ● value added tax and price stability in nigeria 145 way of getting over possible charges of money laundering activity, nigerians abroad often use foreign goods as a vehicle for the transfer of huge sums of foreign currency to nigeria. such goods add to the existing level of goods and services, and are often sold at “give away” prices, thereby neutralising the expected significant impact of rexg on level of prices. similarly, price is one factor that does not obey the law of gravity (“what goes up, must come down”) – prices when up seldom come down. in a market economy, price increases have the tendency of reducing profits for producers/sellers especially for products whose demand is elastic. the common trend among nigerian importers, in collaboration with their foreign producers, is the practice of successively reducing product quality as a way of relatively maintaining existing market prices and profits. over time, consumers are lulled into a false sense of security about the prices of products. little do they know that there has always been a downward trend in the quality of the same products they have been consuming. those that compute the cpi may not have taken this into consideration for the necessary adjustments, even if they are aware of the existing practice. conclusions it is perceived that the vat policy embraced by nigeria in 1994, which has ever since become a significant source of revenue for growth and development, has done more harm than good to the nation’s overall growth and development. the policy’s critics argue that vat induces an upward pressure on prices, the negative consequences of which outweigh whatever gains are derived from the revenue generated, especially as the management of these revenues by the federal government has been problematic. attempts to examine the validity of these claims have led to this study. findings from the empirical analysis reveal that vat has a strong positive significant impact on prices, as it shown by a very high coefficient. the findings support the results of the cge model employed by ajakaiye (1999), which has earlier discovered that vat is more deleterious when viewed as a cost. the implication of this is the instability that it introduces into the nigerian economic system through upward movements in prices. furthermore, when examined from a different perspective, the partial equilibrium analysis adopted herein, by means of a multiple regression model, allows us to explain the effect of vat in isolation from those of other control variables included in the model. besides, our partial equilibrium analysis has the advantage of providing simultaneously the short and long-run impacts of the variables included in the model. further analysis of these findings leads one into the discovery that the strong positive impact of vat policy on prices is in most cases due to multiple vat burdens exerted on individuals and corporate bodies in nigeria, through vat charges on intermediate outputs. the ugly development is one that the agency in charge of vat administration (firs) may not find easy to get over, considering the structure of the vat payment system in nigeria. as a result, it becomes difficult to draw a line between final and intermediate outputs, should one consider tax exemptions for intermediate outputs as a possible option for remedying the situation. such a development leads one into the conclusion that vat as a policy in nigeria, though able to generate much revenue for the government and reduce the problem of tax evasion to a great extent, is not without serious inflationary consequences. the consequence is one that the authorities cannot afford to ignore given the primary role of prices in the modern economic system. it is on the basis of these that the study, as a remedy, recommends further investigation into the macroeconomic effect of vat policy in nigeria. studies on post-vat cost-benefit analysis are needed to ascertain the social desirability of vat policy in nigeria. where vat is discovered to be socially desirable, the result from such a study is european journal of government and economics 2(2) 146 one that can further make provision for better management of adverse situations while consolidating its benefits. in the case of subsequent investigations into the area of macroeconomic problems, partial equilibrium analysis is preferred, in view of its feature of being able to disaggregate effects of given aggregates into short and long-run impacts, necessary for a more detailed evaluation of outcomes. furthermore, this paper advocates the inclusion of remittances in the inflation models of developing economies, from which they are mostly net recipients. remittances constitute an important factor in the inflation models of such economies. references african institute for applied economics (2010) data bank for nigeria's economics statistics, enugu. ajakaiye, dele olu (1999) 'macroeconomic effects of vat in nigeria: a computable general equilibrium analysis', african economic research consortium papers, n. 92. aruwa, suleiman a. s. 82008) 'the administration and problems of value added tax in nigeria' finance and accounting research monitor, 2(2) available at http://ssm.com/abstract=1418661. bleaney, michael; gemmell, norman and david greenaway (1995), 'tax revenue instability, with particular reference to sub-saharan africa', journal of development studies 31 (6): 883-902. bogetic, zeljko and fareed hassan (1993), “determinants of value-added tax revenue: a cross sectoral analysis” world bank working paper-series no. 1203. diallo, oumar (2009) “tortuous road, towards countercyclical fiscal policy: lessons from democratized sub-saharan africa”, journal of policy modelling 31(1): 36-50. ebeke, christian and helene ehrhart (2010) “tax revenue instability in subsaharan africa: consequences and remedies, cerdi working paper no.1192. fatás, antonio and iljan mihov (2003) “the case for restricting fiscal policy discretion”, quarterly journal of economics 118 (4): 1419-1447. fatukasi, bayo (2005) “determinants of inflation in nigeria: an empirical analysis”, international journal of humanities and social sciences 1(18): 262-271. federal inland revenue service (1993a) value added tax decree no. 102. federal inland revenue service (1993b) 'value added tax', firs information circular no. 9304. federal inland revenue service (1993c) 'vat on imports', firs information circular no 9305. gillis, makolm; shoup, carl and sicat, g. (eds.) (1989), value added taxation in developing countries. washington d.c.: the world bank. guillaumont, patrick; jeanneney, sylviane and jean-franciscois brun (1999) 'how instability lowers africa’s growth', journal of african economies 8 (1): 87-107 lim, daniel (1983) 'instability of government revenue and expenditure in less developed countries', world development report, 11(5): 447-450. loayza, norman; ranciere, romain; sirven, luis and jaume ventura (2007) 'macro economic volatility and welfare in developing countries: an introduction', world bank economic review 21(3): 343-357. ikpe and nteegah ● value added tax and price stability in nigeria 147 mclure, charles e., jr. (1989) 'income distribution and tax incidence under vat' in gillis, makolm, carl s. shoup and g. p. sicat (eds) value and taxation in developing countries. washington d.c.: the world bank. naiyeju, j.a. (1996), 'administration of vat in nigeria' firs enlightenment workshop paper, national theatre, lagos. owolabi, oyeleke a. and andy t. okwu (2011) 'empirical evaluation of contribution of value added tax to development of lagos state economy', journal of middle eastern finance and economics 1(9): 24-34. shalizi, z and lynd squire (1989) 'consumption taxes in sub-africa: building on existing instruments' in gillis, makolm, carl s. shoup and g. p. sicat (eds) value and taxation in developing countries. washington d.c.: the world bank. shoup (1989), 'changing among types of vats' in gillis, makolm, carl s. shoup and g. p. sicat (eds) value and taxation in developing countries. washington d.c.: the world bank. talvi, ernesto and carlos vegh (2005) 'tax base variability and pro-cyclical fiscal policy in developing countries', journal of development economics, 78(1): 156 – 190. thornton, john (2008), 'explaining procyclical fiscal policy in african countries', journal of african economies, 17(3): 451-464. unegbu, angus and david irefin (2011) 'impact of vat on economic development of emerging nations', journal of economics and international finance 3 (8): 492503. worlu, christian and emeka nkoro (2012), 'tax revenue and economic development in nigeria: a macroeconomic approach', academic journal of interdisciplinary studies 1(2): 211-23. give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia vol.8 • no.1 2019 issn: 2254-7088 european journal of government and economics 8(1), june 2019. european journal of government and economics issn: 2254-7088 number 8, issue 1, june 2019 doi: https://doi.org/10.17979/ejge.2018.8.1 the evaluation index improvement for the relocation of research equipment 6-29 donghun yoon residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) 30-47 josé ramón cardona, daniela álvarez bassi, maría dolores sánchez-fernández social capital and banking system profitability: a survey of european union countries 48-62 arash nayebyazdi governance and domestic investment in africa 63-80 chimere okechukwu iheonu economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt 81-95 marwa shibl biltagy give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia 96-109 jaroslav mihálik, peter horváth, martin švikruha https://doi.org/10.17979/ejge.2018.8.1 european journal of government and economics 8(1), june 2019, 96-109. european journal of government and economics issn: 2254-7088 give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia jaroslav mihálika *, peter horvátha , martin švikruhaa a university of saints cyril and methodius in trnava, slovakia * corresponding author at: jaroslav.mihalik@ucm.sk article history. received 12 april 2018; first revision required 5 october 2018; accepted 3 december 2018. abstract. the aim of this study is to focus on fiscal decentralization and regional financial autonomy in slovakia. the public administration reform and fiscal decentralization process in slovakia ought to increase the autonomy of the multilevel governance and to decrease its dependency on state budgeting and other transfers. we argue that regional territorial governance is greatly dependent on financial incentives from central state financing, which limits the expected effect of decentralization and regional fiscal autonomy. by collecting quantitative data on state transfers and revenues to regional governments we demonstrate the limits of spatial, financial and decision-making autonomy within particular regions. the selected time frame 2005-2016 reflects all stages, reforms and changes in the fiscal decentralization of the second level governance in slovakia. keywords. fiscal decentralization, fiscal autonomy, regional government financing, slovakia. doi. https://10.17979/ejge.2019.8.1.4573 1. introduction regional governance is not the end in itself; it is how a goal is sought (barnes – foster, 2012). public administration reform and the process of fiscal decentralization should lead to an increased autonomy of territorial self-government and to reduce their dependence on the state budget as well as other government transfers or between government levels. the hollowing out of the state is not only in the political and competence area but also financial. the rationale is thus to motivate local and regional governments to manage public finances and resources and to use them to increase public welfare (pierre and peters, 2000; sellers and lidström, 2008, provazníková, 2015). it is a progressive process of transforming the government into the governance model (rhodes, 2007). this may include several levels of control, depending on the model to be applied. according to halásková (2009), significant differences can be found between the individual eu countries in the structure of the division of territorial self-government. single-tier territorial selfgovernment, consisting of a municipal level is typical for slovenia or lithuania. two-tier governance in which the local and regional level coexists applies in slovakia, the czech republic or hungary. for poland, as well as for italy, the three-level territorial self-government is https://10.0.70.59/ejge.2019.8.1.4573 jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 97 characterized by the fact that there is a level represented by the districts alongside with the local and regional levels or provinces. multi-level systems of territorial self-government are present, for example also in france, the united kingdom and other european countries. in addition to the most important self-governing level organized within cities and municipalities, gradual development also led to an increase in the importance of a second level of territorial selfgovernment, which is the regional tier. as loughlin, hendriks and lidström (2011) argue, the world has been modified since the 20th century which was typical mainly by the dominant interest in national sovereignty and centralized governments. the current period is characterized by new systems of governance strengthening the role of sub-national levels of public policymaking as well as the relationship of representation and collective self-determination. in this regard, the effectiveness and democratic aspect of policymaking in multi-level governance have attained specific and very important status in the political arrangements of most european countries and have led into deep restructuralization of the regional and local governments. such trends may be observed across multiple european countries and according to authors (sturm – dieringer, 2005), the decentralization and regionalization tendencies are present in all european countries during the last decades. deriving from this, it is evident that this process is currently not only from the slovak perspective but is more of a global and standard feature. in the slovak perspective, the discussion about the regional governance is generally limited to arguments about improper areas and sizes of individual self-governing regions specifically due to historical and natural reasons (nižňanský, 2005; klimovský, 2006; sloboda, 2006; machyniak, 2013; nižňanský and hamalová, 2013; mikuš, 2014). according to šteiner and kozlayová (2010) the major problem of regional governance is not vested only in numbers and spatial issues but a more relevant aspect – the competention area. they point out, in particular, the low-level use of regional development instruments, either in the area of policy-making or cooperation with other actors at this level, while also presenting recommendations for improving the management of regional development. some authors are concerned with partial problems in the form of funding and regional differences in their studies (sloboda, 2006; fabianová, 2010; bobáková, 2015). a recent analysis made by the international monetary fund (caselli and ralyea, 2017) stipulates that slovakia has one of the highest regional disparities among oecd and eu countries. the authors argue that although the country has received significant eu transfers devoted mainly to foster regional convergence, the absorption of eu funding during the 20072013 period has been greatly delayed. yet they agree that a greater degree of fiscal decentralization helps increase the rate of eu funds absorption. this is much in line with our presented research on the fiscal decentralization in slovakia which brings forward the issue of regional budgeting dependency on state budget allocation and transfers established from the central government. although regional governments are responsible, pursuant to law, for the comprehensive integrative development of the respective region, their own direct impact over the economic, social and environmental development is substantially low. while the competences decentralization process reflects the subsidiarity principle of administrative units, fiscal decentralization has become a vital point of discussion due to its measures and limits. jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 98 in our research, we argue that fiscal decentralization and financial autonomy of slovak regions cannot be accomplished based on the empirical findings which demonstrate the high levels of financial dependence of the regions multiplied by the effects of recent changes in the composition of the regional budget. the research study is conducted as a case study informing about the regional fiscal decentralization and autonomy in slovakia. we analyze and develop the approaches, legal norms and practice of fiscal decentralization while considering the criteria of financial transfers from the central government. particularly, we focus on the developments in fiscal decentralization together with the recent changes in budget allocations to regional governments which came into force in 2014. this had caused the change in nature of the research rationale from measuring the regional fiscal autonomy to fiscal dependence. the data collection reflects the time periods from the beginning of the process of fiscal decentralization in 2005 until the recently available final balance sheets of the 8 regional territorial units in 2016. 2. an overview of the regional self-government financing in slovakia to be able to talk about functional regional self-government, it is essential it manages sufficient financial resources. in terms of the slovak republic, the framework of regional financing has undergone continuous changes. these are made through legal norms determining the financial rules applicable on the regional level. there is a full spectrum of legislative acts involving the financial distribution and transfers allocation while it has been outlined in article 65 of the constitution. it is to be noted that financial resources should cover the performance of transferred as well as original competences (sivák et al., 2007). this requirement, apparently, has not been fulfilled which according to our research has direct impact on the creation of the financial framework of the regional self-governments. this principle of financial responsibility which is based on the shift of responsibility in lowerlevel decision-making to the provision of public goods and services and their funding is also highlighted by litvack and seddon (1999). according to them, if a local or regional government has a good capacity to fulfil its functions and tasks, it must have an adequate own income or income from the central government, as well as the power to decide on its use. the very same article of the constitution defines that self-governing region (also known as higher territorial unit htu) as a legal entity that, under the conditions stipulated by law independently manages its own assets and financial means (article 65, constitution of the slovak republic). the financing of the htu is based mainly on own revenues and state subsidies while the law determines which taxes and fees create the income. the funding base of the htu is budget prepared for a period of one marketing year (tej, 2007). self-governing regions draw up and approve the budget broken down into the current, capital budget and financial operations. this is an important document in which the htu has the duty to ensure fulfilment of all obligations under specific regulations. the budget creates the main means jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 99 of financial management of self-governing regions and financing the tasks and functions of selfgoverning regions. in the reporting process, a key role is played by the final balance sheet alongside the budget. since its existence, the structure of regional self-government revenue has undergone several modifications. in the short period after the establishment of the second level of territorial selfgovernment, the method of its financing was based on the principle of close interdependence with the central level, which provided municipalities with funds in the form of special grants. since 2005, regional governments have a higher degree of financial autonomy when they can independently decide on the use of tax revenue (cíbik, 2014) which consists of a share tax or a personal income tax and, until 2014, a tax on motor vehicles. the share of personal income tax is the most significant item of municipalities in terms of their income and currently the only tax revenue of self-governing regions. this tax is levied by the state, which distributes it to individual self-governing regions on the basis of the criteria set out in figure 1. figure 1. criteria determining the reallocation of funds from the state budget to htus. source: government regulation no. 668/2004 coll. about the distribution of revenue from the income tax of the territorial self-government the system of financing introduced in 2005 was based on fixed criteria according to which the personal income tax is distributed among individual municipalities, cities and self-governing regions and becomes their own income. another major part of the revenues was the tax on motor vehicles, which was actually the own income of self-governing regions. the amendment to the act on taxes from motor vehicles of august 2014 has a change in this section, which came into force on 1 january 2015 and from that date does not include this tax in the tax revenue portfolio of self-governing regions. the higher degree of financial autonomy of self-governing regions was manifested not only in the fact that the regional authorities could decide on the actual amount of the motor vehicle tax but also on whether the tax applied by the legislator would be imposed on the territory of the self-governing region. as a result of these steps, the responsibility of local selfjaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 100 government for the management and development of individual regions (čavojec and sloboda, 2005) was increased in comparison to the period 2002 – 2004. the facultative nature of the taxes that the self-governing regions enjoyed since 2005, in particular, the tax on motor vehicles, is reflected in an increase in the tax powers of self-governing regions and, ultimately, as one of the objectives of fiscal decentralization. yet another change in financing has been introduced by fiscal decentralization – share tax and income tax revenue. the income tax of persons has so far gone through several changes, the overview of which is shown in table 1. in the period 2005 to 2011, the self-governing region was allocated a share of 23.5% of the total amount of this tax (act no. 564/2004 coll., §3). the remainder is further distributed among municipalities, cities and the state. since 2012, the fiscal decentralization system has determined that the total amount of income tax belongs to the selfgoverning region, with a yield of 21.9%. at time there was a change when part of the income tax was transferred to the state. the share of self-governing regions in the personal income tax thus decreased by 1.6% compared to the previous period. at the beginning of 2015, there was a further change in the proportion of income tax revenue for self-governing regions from 21.9% to 29.2%. this substantial increase in the percentage is due to a change in the fiscal system, the essence of which is to compensate for the fact that the state has taken over the self-government of their own income from the tax on motor vehicles. table 1. the distribution of income tax since 2005 in the multi-level governance time frame local level regional level central level 1.1. 2005 – 31.12. 2011 70,3% 23,5% 6,2% 1.1. 2012 – 31.12. 2014 65,4% 21,9% 12,7% 1.1. 2015 – 31.12. 2015 68,5% 29,2% 2,3% 1.1. 2016 70% 30% 0% source: authors´ processing according to act no. 564/2004 coll. on the budget determination of the income tax on local government income and on amendments to some acts, act no. 548/2011coll. amending act no. 595/2003 coll. on income tax, as amended, and amending certain laws, approved by the government bill amending act no. 595/2003 coll. on income tax as amended, and amending certain acts of 30 october 2014, act no. 361/2014 coll. on the taxation of motor vehicles and on the amendment of certain laws. the third stage of the financial system of regional self-government in slovakia is characterized by the withdrawal of a single local tax (motor vehicle tax) from the exclusive competence of the htu and its subsequent assignment to the central national tax collection portfolio for which the tax authorities are responsible. at the same time, we are experiencing an increase in the redistribution coefficient from the share tax for self-governing regions to compensate for the outflow of motor vehicle tax. jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 101 table 2. development of the tax share for regional self-government in the slovak republic from 2013 to 2016 (in €) year income tax motor vehicle tax total tax revenue 2013 411 845 497,30 139 116 513,20 550 962 010,50 2014 426 657 757,47 166 351 882,48 593 009 639,95 2015 626 393 932,80 13 785 662,77 640 179 595,57 2016 666 615 180,00 666 615 180,00 source: authors´ processing according to final balance sheets of the htus 2013-2016 the current structure of income and revenues for htus is composed as follows: non-taxable income from the ownership and transfer of ownership of the assets of the htu and from the activities of the higher territorial unit and its budgetary organizations; interest and other income from the funds of the htu; penalties for violation of financial discipline imposed by the htu; donations and revenues of voluntary collections for the benefit of a htu; shares in other taxes of the state administration under a separate regulation; subsidies from the state budget for reimbursement of costs incurred by the state administration in accordance with the law on the state budget for the respective financial year and subsidies from state funds; other subsidies from the state budget in accordance with the law on the state budget for the respective financial year; purposeful subsidies from the municipal budget or from the budget of another htu for the implementation of contracts according to special regulations; funds from the european union and other funds from abroad provided for a specific purpose; other income provided for by special regulations (act no. 583/2004 coll.). a separate category in this area is the government tax rate. this is a personal income tax, that is, a share tax whose income is currently distributed between the two levels of territorial selfgovernment, as shown in table 1. the slovak legislation regulates it as an own income but this does not correspond to the international standards of tax autonomy. its current setting is the reason why this share tax does not meet the parameters of own income or autonomous revenue. foreign authors (sutherland, price and joumard, 2005; blöchliger and king, 2006; sacchi and salotti, 2014) and the european commission divide tax revenues into two areas. the first is the autonomous tax, the revenues of which the territorial unit can correct or change the key parameters. the second group is share tax where the regional governments have no impact over its revenues. by default, personal income tax belongs to share taxes rate because the htus are not able to directly influence important variables such as tax rates, tax bases or tax deductions. self-governments´ revenues, in general, serve to quantify how much income the territorial units actually have. the basic feature of own income is the ability to actively influence their height. in the case of own tax revenues, the basis lies in the fact that the territorial unit determines the jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 102 tax rate, the tax base and tax concessions or autonomous exceptions (blöchliger and king, 2006). from this point of view, we can assume the htus own revenues as those stipulated by law except for the share of income tax. the share of current own revenues in total income then reflects the degree of autonomy of individual units of territorial self-government. the percentage of financial independence demonstrates how much of the total revenue the regional authority can actively influence its decision, independent of the central level. this provides an opportunity to assess autonomy in the revenue area and expresses the extent to which the municipality or the htus are independent. 3. data and methods looking at the final accounts of individual htus as well as their budgets, we can observe that ordinary incomes constitute a relatively large income component of their budgets. however, if we clean current own revenues from funds flowing from income tax, we find that own revenues make up only a small part of the total income of the htus. first of all, we had the ambition to demonstrate this fact by computing the financial independence of the htu from 2005 to 2016 and capturing it graphically. consequently, we realized that a more appropriate way to demonstrate the minimum proportion of own revenues in the budgets of self-governing regions would be to modify the formula for calculating financial independence. instead of financial independence, we focused on the degree of financial dependence on current domestic grants, transfers and subsidies in public administration, and we also included the share tax. financial dependency identifies the degree of binding the budgets of self-governing regions to the state budget without the possibility to actively influence the amount of these financial flows. regional self-government is fully dependent in this area and depends on central level decisions, which, as the only one, has the power to change the amount of funds that go to the budgets of self-governing regions through ordinary subsidies, grants and transfers within the public administration system. the same is true for the revenue from the share tax. in determining financial dependency, we started from the indicator of financial autonomy modified for our needs. this means, when comparing financial independence we focus on the ratio of own revenues to total income currently, and in the case of financial dependence we have decided to examine the value of the antagonistic variable standard foreign/external incomes. by identifying financial dependency, we can illustrate the degree of dependence of self-governing regions on current subsidies, grants and transfers that occur in the financial system of public administration in slovakia. these financial resources come from the central level and its institutions and represent the flow of resources that primarily serve to perform the transferred competences from the state to the self-governing regions. in this respect, the htus represent the "deconcentrates" of the central government which carry out a range of delegated competencies and their implementation is paid from the state budget through the budgets of individual ministries. in such a case, the central government uses the structure of regional self-governments to implement the transferred competencies from the state administration, while 100% оf their jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 103 performance is paid in the form of special transfers. financial dependency reflects the extent to which the regional self-government budgets are linked to current transfers within the public administration. however, the htus cannot actively influence this financial flow making it fully dependent on central government decisions. in the calculation of financial dependency, we also included the proportion of self-governing regions in the revenues from income tax among the current transfers within public administration. the reason is our belief that setting it up in its present form rather advises it to non-specific subsidies. we focus on the fact that in the evaluation of tax autonomy, we could not associate the share tax in slovakia with the category of own tax revenues which form the core of standard incomes (blöchliger and king, 2006; jílek, 2008). the financial dependence itself is calculated using the following formula, which will be followed by specific data from final balance accounts and budgets of htus: we have included standard grants, subsidies and transfers from the state budget to cover the costs of the transferred state administration in accordance with the state budget law for the respective budgetary year and subsidies from state funds, transfers from the state budget in accordance with the state budget law to the competent state budget year, specific subsidies from the municipal budget or from the budget of another htu for the implementation of contracts according to special regulations. the income tax revenues represent the revenue from the share tax in the state administration under a special regulation. in the discussion, objections may arise in the case of inclusion of income tax revenues for regional governments in ordinary non-specific subsidies. if the income from the share tax was perceived as own income of the htu, it would radically increase financial autonomy while reducing the degree of financial dependency. also, the argument based on the fact that the income tax funds are not primarily intended for the performance of the transferred competences may also appear. this is true; however, the practice has shown that the funds from income tax are mainly paid for normal, i.e. operating expenses. the revenue from the share tax is used to cover the costs of securing the running of the administration and covers the costs of energy, wages, insurance, materials and the like. total revenue is the sum of current income, capital gains and income from financial operations for a particular financial year. subsequently, we have adjusted the final accounts and budgets of all self-governing regions for the period 2005 2016 in the modified formula for the calculation of the financial dependency. at the time of processing our work, the self-governing regions have not yet processed the final account balance for 2016, so the calculation starts from the expected reality of the fulfilment of individual self-governing regions. jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 104 4. the fiscal (in)dependence and autonomy of regional governance in slovakia table 3 summarizes the financial dependency ratio of specific slovak regional governments for the period 2005-2016. table 3.the rate of financial dependence of htus during the years 2005-2016 (%) year 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 average htu bsk 64,6 65,2 64,9 60,2 58,3 69,2 59,2 61,1 58,8 63,2 88,1 87,9 66,7 bbsk 79,4 77,1 76,8 73,4 75,7 73,8 72,9 73,2 68,7 71,9 86,4 85,3 76,3 ksk 84,3 94,8 73,4 68,7 71,3 69,9 67,1 69,9 70,0 69,2 76,2 73,9 74,0 nsk 74,4 73,8 76,4 71,4 68,2 68,9 57,6 67,7 69,0 67,2 79,9 85,5 71,6 psk 85,7 84,5 83,8 86,2 74,3 67,9 64,1 71,5 73,1 71,2 77,1 86,4 77,3 tsk 84,9 82,8 81,5 84,4 63,8 80,6 79,8 72,7 76,8 78,2 89,8 92,5 80,6 ttsk 72,5 48,8 66,2 70,9 72,6 84,3 82,8 71,1 zsk 72,1 72,9 79,3 71,3 82,1 62,7 71,1 70,6 68,6 57,9 83,8 72,1 72,0 average 77,9 78,7 76,5 73,6 70,5 70,6 65,0 69,1 69,4 68,9 83,2 83,3 73,8 legend: bsk – bratislava htu, bbsk – banská bystrica htu, ksk – košice htu, nsk – nitra htu, psk – prešov htu, tsk – trenčín htu, ttsk – trnava htu, zsk – žilina htu. the higher the value we have found, the more the slovak self-governing regions are financially dependent on transfers from the state budget, while the central government keeps the decisionmaking about the amount and the yield of these transfers. our research reveals significant interregional and time differences in financial dependency. for example, bratislava htu has developed as the least dependent on standard transfers, subsidies and grants within the public administration in slovakia. its average value of financial dependence for the last twelve years was 66.7%. on the other hand, three htus (trnava, nitra and žilina) follow a significant distance within their average values. trenčín htu, particularly, has been the most dependent region on funds from standard subsidies, transfers and grants. its average rate of dependence on current transfers from the state budget over the monitored period has been more than 80%. the development of the average rate of financial dependency from 2005 to 2016 of all eight self-governing regions is shown in figure 2. at the same time, it illustrates the average level of financial dependency of all regions over the twelve-year period, i.e. the entire segment of regional self-governance during the years 2005-2016. financial dependency rates have fallen since 2006. the lowest average rate of financial dependence of self-governing regions was recorded in 2011. subsequently, by 2014, the average rate of financial dependency increased slightly. in 2015, we identified an unusually high increase in financial dependency, an increase of 14.3% over the previous year. this huge leap was caused by the reform of tax revenues for htus. the tax from motor vehicles was deducted which was offset by an increase in the redistribution coefficient of the share tax. this had, on the one hand, a negative impact on the reduction of financial autonomy and, on the other, it contributed to increasing the financial dependence of self-governing regions on the central government. the financial dependence on current financial means from the central jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 105 government for the whole segment of regional self-governance in slovakia in the period 20052016 was determined by analysis and calculation at 73.8%. this means that, during this period, the revenue component of the regional self-government budgets was closely tied to state budget funds, funds that are not decided by the htus, but are fully in the hands of the central government. figure 2: the rate of financial dependence of htus in a given year and the average rate of financial dependency for the years 2005-2016 (%). source: authors´ processing however, one should add that, if the average financial dependence of htus is at 73.8%, it does not automatically mean that the regional self-government is on average 26.2% autonomous in the area of its income (figure 2). the difference between these values is based on revenue operations in the area of capital income and financial operations, with which both indicators measure the financial situation of self-governing regions. in fact, the degree of financial independence, excluding income tax shares, was down to 2014 at a low level, with an average median of 11-16% based on partial calculations. after removing the tax on motor vehicles from the portfolio of exclusive regional government powers, this already low share of own revenue fell dramatically. on the basis of the process of fiscal decentralization and subsequent financial dependency calculation, it can be stated that the regional self-governance in slovakia shows a high financial dependence over funds coming from the state budget in the form of standard subsidies, grants and transfers. their height can actively influence the central government level on which "good will" the regional self-governing units are fully dependent. we could also take a slight look at the fact that the self-governing regions perform the activities of state administration, which have been identified and funded by the central government. with a very high degree of financial dependence, such as territorial self-government the self-governing element of a self-governing unit, its original competences and its own income base, is pushed into the background. therefore, elected 77.9% 78.7% 76.5% 73.6% 70.5% 70.6% 65.0% 69.1% 69.4% 68.9% 83.2% 83.3% 73.8% 73.8% 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 htus financial dependence in given year htus average financial dependence jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 106 representatives, with a precisely defined portfolio of delegated competencies and whose performance is accountable to a central government that fully finances them. consequently, it is questionable whether it is important for long-term maintenance of the situation where regional self-government fails to fulfil its basic self-government functions, because it does not have a sufficiently profitable income of its own and, on the contrary, is burdened with the exercise of a large number of competences by the state administration. 5. conclusion based on the analysis of the financial dependence it can be stated that in the case of the regional level of territorial self-government, the principles stemming from the fiscal decentralization are not applied. we believe that the current state is even contrary to the theory of fiscal decentralization (musgrave and musgrave, 1994; oates, 2008; jílek, 2008; cíbik, 2014; provazníková, 2015). these principles were crucial for the implementation of the public administration reform and the establishment of regional level self-government. regional self-government shows a high financial dependence from state resources, which we consider to be a problem because the self-governing nature of this territorial unit is threatened as it fulfils the tasks that are determined and financed by the central government. the self-government aspect represented by the original competencies and own income base is minimized. the dependence of regional self-government on central power is manifested not only in the financial sphere but also in the area of competence. requirements for addressing both transferred and original competencies are often ignored or rejected by the central government and its ministries. a key issue in the area of independence is the fact that regional self-government is not perceived as a central power partner, but in several cases, it has rather elements of a subordinated entity. the current state of independence of regional self-government is most strikingly characterized by the current model of its financing. our analysis points out that the extent of the competencies is insufficiently covered by the volume of funds, while the state of the property, with which the competencies were transferred to the particular htus have created the modernization debt. as a result, desirable efficiency shall also be achieved as well as the pressure to effectively deal with funds and the use of multi-source funding. on the other hand, there may be an increase in indebtedness and a poor quality of the given competencies or their non-provision, insufficient capacities management and low standards set by regional self-government. an unfavourable moment is the financing model mentioned above, as regional governments receive a major share of the tax that is shared between local and regional governments. this tax shows six distribution criteria, where the resulting value is still multiplied by the coefficient set for the given htu. the coefficient was set so that the economically stronger regions had lower distribution than economically weaker ones. this is gradually being balanced, but the gap between the regions has not been significantly narrowed yet; on the contrary, they have even deepened. therefore, we propose to modify the system of financing of self-governing regions, which should aim at gradually releasing the competence of self-governing regions by reducing specific state subsidies jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 107 linked to performance of specific functions and their replacement by general state subsidies individual self-governing regions could decide independently, in what sphere and to what extent will use the funds. it can be concluded that, since its inception, regional self-government suffered from uncertainty as to how much funding will be transferred to its budget from the income tax, and despite the fiscal decentralization that has taken place, this situation persists to this day. instability and changes in determining the percentage are reflected in the exercise of their powers and strategies as well as the removal of one's own tax. the local government has lost the real possibility of influencing financial flows, and this has affected the quality and quantity of services provided. it is, therefore, necessary to make adjustments that would increase the economic sovereignty of regional self-governments and create real conditions to address the long-standing problems of the financial system with a view to stabilizing financial resources over a longer period by adjusting fiscal relations between state and local levels. we believe that an increase in resources, as well as the possibility of participation in the creation of central financial plans for taxation and subsequent redistribution, should be the basis for better functioning. the process of deinstitutionalization and privatization, i.e. the transition from the construction of facilities to the provision of services, is essential, in order to create an environment in which a citizen can choose from a wide variety of different service providers wherever it is efficient and appropriate. decisions on changes in the competency and financial framework must also discuss the interdependence of the two levels of self-government, with which the issue of a comprehensive reform of territorial self-government is closely linked. references act no. 564/2004 coll. on the budget determination of the income tax on local government income and on the amendment of some laws. act no. 583/2004 coll. on the financial rules of territorial self-government and on the amendment of some laws. barnes, w.r. & foster, l. a. (2012). toward a more useful way of understanding regional governance. a paper for presentation at the september 2012 conference of the european urban research association. vienna: austria. blöchliger, h. & king, d. (2006). the fiscal autonomy of sub-central governments. paris: oecd. bobáková, v. (2015). zadlžovanie regionálnej samosprávy, slovak journal of public policy and public administration, 2015, vol. 2, no. 2, pp. 153 – 166. caselli, f. & ralyea, j. (2017). slovak republic: selected issues. imf country report no. 17/72. washington, d.c.: international monetary fund. https://doi.org/10.5089/9781475587937.002 cíbik, l. (2014). slovensko a fiškálna decentralizácia. acta oeconomica universitatis selye, 2014, vol. 3, no. 2, pp. 24 – 32. čavojec, j. & sloboda, d. (2005). fiškálna decentralizácia a obce. bratislava: konzervatívny inštitút m. r. štefánika. https://doi.org/10.5089/9781475587937.002 jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 108 fabianová, k. (2010). finančná samostatnosť vyšších územných celkov na slovensku. in šoltés, v. (ed.). national and regional economics viii. košice: technical university, pp. 227 – 234. halásková, m. (2009). veřejná správa v evropské unii. opava: optys. horváth, p., cíbik, l., & švikruha, m. (2018). finančná závislosť územnej samosprávy na slovensku. scientific papers of the university of pardubice, series d. 2018, vol. 25, no. 42 (1/2018), pp. 67 – 79. jílek, m. (2008). fiskálni decentralizace, teorie a empirie. praha: aspi wolters kluwer. klimovský, d. (2006). regionálna samospráva v rokoch 2002 – 2005: skúsenosti a súvislosti v politickej reflexii. in mesežnikov, g. (ed.). regionálne voľby 2005 – súvislosti a výsledky. bratislava: ivo. litvack, j. & seddon, j. (1999). decentralization. briefing notes. washington, dc: world bank institute. loughlin, j., hendriks, f., & lidström, a. (2011). the oxford handbook of local and regional democracy in europe. new york: oxford university press. https://doi.org/10.1093/oxfordhb/9780199562978.001.0001 machyniak, j. (2013).transformácia verejnej správy na slovensku po roku 1989 výzvy a realita. in ondrová, d. et al. (ed.).teória a prax verejnej správy. košice: univerzita pavla jozefa šafárika v košiciach. mikuš, d. (2014). vybrané aspekty prvej etapy decentralizácie verejnej správy na slovensku. acta oeconomica universitatis selye, 2014, vol. 3, no. 2, pp. 102 – 108. musgrave, r. a. & musgrave, p. b. (1994). veřejné finance v teorii a praxi. praha: management press. nižňanský, v. (2005). decentralizácia na slovensku. bilancia nekonečného príbehu 1995 – 2005. bratislava: úrad vlády sr. nižňanský, v. & hamalová, m. (2013). decentralizácia a slovensko. bratislava: všemvs. oates, w. e. (2008). on the evolution of fiscal federalism: theory and institutions. national tax journal, 2008, 61(2), pp. 313-343. https://doi.org/10.17310/ntj.2008.2.08 pierre, j., & peters, b. g. (2000). governance, politics and the state (political analysis). new york: palgrave macmillan. provazníková, r. (2015). financování měst, obcí a regionů: teorie a praxe. praha: grada publishing. rhodes, r. a. w. (2007). understanding governance: ten years on. organization studies, 2007, vol. 28, no. 8, pp. 1243 – 1264. https://doi.org/10.1177/0170840607076586 sacchi, a. & salotti, s. (2014). how regional inequality affects fiscal decentralisation: accounting for the autonomy of subcentral governments, environment and planning c. politics and space, 2014, vol. 32, no. 1, pp. 144 – 162. https://doi.org/10.1068/c1241r sivák, r. et al. (2007). verejné financie. bratislava: iura edition. sellers, j. m., & lidström, a. (2007). decentralization, local government and the welfare state. governance, 2007, vol. 20, no. 4, pp. 609 – 632. https://doi.org/10.1111/j.14680491.2007.00374.x sloboda, d. (2006). slovensko a regionálne rozdiely. teórie, regióny, indikátory, metódy. bratislava: konzervatívny inštitút m.r. štefánika. https://doi.org/10.1093/oxfordhb/9780199562978.001.0001 https://ucm.dawinci.sk/?fn=*recview&uid=50483&pageid=resultform https://ucm.dawinci.sk/?fn=*recview&uid=50483&pageid=resultform https://doi.org/10.17310/ntj.2008.2.08 https://doi.org/10.1177/0170840607076586 https://doi.org/10.1068/c1241r https://doi.org/10.1111/j.1468-0491.2007.00374.x https://doi.org/10.1111/j.1468-0491.2007.00374.x jaroslav mihálik et al. / european journal of government and economics 8(1), 96-109. 109 sturm, r., & dieringer, j. (2005). the europeanization of regions in eastern and western europe: theoretical perspectives. regional & federal studies, 2005, vol. 15, no. 3, pp. 279 294. https://doi.org/10.1080/10438590500223251 sutherland, d., price, r., & joumard, i. (2005). fiscal rules for sub-central governments: design and impact. oecd economics department working paper 465. https://doi.org/10.2139/ssrn.870443 šteiner, a., & kozlayová, a. (2010). dobré spravovanie rozvoja regiónov výzva pre slovensko. košice: karpatský rozvojový inštitút. tej, j. (2007). správa a manažment. prešov: manacon. žárska, e., & kozovský, d. (2008). teoretické a praktické aspekty fiškálnej decentralizácie. bratislava: ekonóm. https://doi.org/10.1080/10438590500223251 https://doi.org/10.2139/ssrn.870443 front81 ejge oficial 8-1 contents81 number 8, issue 1, june 2019 4573 abstract. the aim of this study is to focus on fiscal decentralization and regional financial autonomy in slovakia. the public administration reform and fiscal decentralization process in slovakia ought to increase the autonomy of the multi-level gove... keywords. fiscal decentralization, fiscal autonomy, regional government financing, slovakia. doi. https://10.17979/ejge.2019.8.1.4573 microsoft word ejge_02_02_043.doc european journal of government and economics volume 2, number 2 (december 2013) issn: 2254-7088 148 modernity and natalism in russia: historic perspectives nina kouprianova, russian federation abstract less than desirable indigenous birth rates in western europe have generated interest toward examining the question of natalism — an organised state initiative to manage and promote reproduction, child rearing, health, as well as related neotraditional cultural values — from a comparative perspective. this paper reviews the history of natalism in the ussr and contemporary russia within the greater framework of modernity, by focusing on sweeping questions of ideology and geopolitics as well as current historic models. economic stability is not an unimportant factor, yet it is authentic traditional culture that is of equal, if not greater, importance, even if expressed through state policies. jel classification z; j13; j11 keywords ussr; russia; natalism; modernity; demographics; neo-traditionalism kouprianova ● modernity and natalism in russia 149 introduction “i can’t believe i wanted an abortion,” states a rather blunt soviet poster from 1961, depicting a woman looking fondly upon her newborn.1 those with a popular knowledge of russian history associate the natalist drive in the ussr with the iconic heroic-mother imagery in the era of joseph stalin. yet natalism first appeared in 1920s soviet union and has not left since (perhaps, with certain exceptions in the 1990s-early 2000s). this was — and remains — an organised state initiative to manage and promote reproduction, child rearing, health, education, and socio-cultural mores, as they pertain to family relations at large, not just the fertility drive in the strict demographic sense of the word. the government’s attempts to boost the population size in the ussr and the russian federation, respectively, appeared to coincide with modest to positive results, whether the stalinist programme of overt propaganda and awards that were paralleled by a natural population increase after the second world war or the contemporary vladimir putin era campaign that combines financial reward and cultural projects. historically, natalism is a product of modernity. pre-modern governments in the west and russia certainly maintained an interest in the people they governed in terms of taxation, military service, and territorial control, among others. but it was not until the post-enlightenment period that the state sought to make its respective populations “legible” as scholar james scott put it. it accomplished the latter by dividing its subjects into manageable segments, thereby monopolizing its ability to intervene, and demonstrating a progressively greater investment in its subjects’ health, reproduction, education as the marriage between technical capabilities and social capital, and other formerly privately held categories within a particular ideological framework (in the 20th century, these were liberalism, communism, and fascism) (scott 1998: 78). today, many scholars do not question state involvement in these matters, per se, only their extent. by and large, criticism of the attempts to boost fertility comes from ecologists and those concerned with global population growth in relation to the planet’s limited resources. in terms of birth rates, specifically, modern states’ most basic motivation thereof stemmed from the need to replenish those lost in war. there were also vital geopolitical concerns, i.e., maintaining a population large enough to defend a country’s borders, especially applicable in the case of resource-rich ussr and its heir. another relevant factor was the relationship between workforce size, social services, and taxation levels (hidden or otherwise) both in historically industrial socialist countries and in contemporary serviceand consumption-oriented liberal democracies of the west.2 in this context, european welfare states may seem benign by comparison to their authoritarian counterparts, but their keen interest in the subject of birth rates, education, and other related matters in europe-proper is guided by the same overarching principle of population “legibility.” in the decades since the 1970s, the reasons for mass immigration into europe have been economic, demographic, and postcolonial.3 that is to say, its immediate purpose has been to ensure that an adequate number of taxpayers is available in order to maintain all the social programs along with other tax-funded projects, not to mention guaranteeing sufficient support for a western european liberal democracy in its variants. 1 see poster selection “materinstvo v illiustratsiiakh,” materinstvo.ru (consulted 10 march 2013): materinstvo.ru/art/6214/. 2 in terms of ideology, it would be more accurate to use the term “post-liberal” after 1991, as i explain below. however, i cover large periods of time in this paper, hence the usage of “liberalism” to avoid confusion. 3 some argue that migration is primarily economically driven, but one cannot divorce the importance thereof from the western european liberal democratic ideology per se. european journal of government and economics 2(2) 150 supranational bureaucratic bodies like the united nations regularly issue reports with thinly veiled threats about raising the retirement age to 75, for instance, if certain migration levels are not maintained in order to support the current system.4 first, it is questionable as to whether the migrants can theoretically support the low indigenous birthrates, ultimately leading to systemic failure. second, the new europeans’ relocation comes at a price for both the migrants and the hosts. of course, the implicit (or, at times, explicit) population management technique here comprises a strong socio-cultural component, ensuring that the said newcomers adhere to the dominant secular liberal ideology — which, at times, is quite foreign to their own background — under the faulty assumption that cultures are interchangeable. on the other hand, they undermine the ethno-cultural cohesion of the dominant indigenous peoples. considering the attention paid to the failures of this brand of multiculturalism in the european and american media in the recent years, it is unsurprising that there is an interest in somewhat alternative models like the russian natalist variant.5 when it comes to natalism in the ussr, specifically, western liberal historians generally place it in the context of “neo-traditionalism.” they describe it as a phenomenon, in which modern secular states take over the role of authentic, spontaneously occurring tradition in pre-modern societies, realizing the sheer power of the latter as a mobilisation tactic. in this view, “neo-traditionalism” in the ussr ranged from the limited reintroduction of patriarchal gender roles to the usage of historic and folk heroes for agitational purposes (martin 2000: 348-367). this framework is not inaccurate, and is quite practical. however, it emphasises the powerful state, while somewhat downplaying the vast support thereof among the masses. of course, western scholars of russian history have rejected the original black and white “totalitarian” model of a near-omnipotent government and a submissive, oppressed population, but continue to debate the particular relationship between the state and its subjects. in order to understand the longevity of natalist initiatives and their apparent successes (or lack thereof), we must examine their roots. let us briefly review the facts both in the ussr and contemporary russia. each historical period will be accompanied by an analysis focused on ideology, as a partial imposition or borrowing of a foreign, i.e., western european value system, as well as geopolitics, surpassing ideology. we will then discuss the benefits and drawbacks of current historic models and offer an alternative view that incorporates russian folkic undercurrents in the era of modernity and postmodernity. in other words, the primary focus here is the history and theory of natalism in the ussr and russia within a socio-cultural framework, not specific fertility behaviour, policies, and other heavily statistical information, although some applicable examples will be given. history of natalism in the ussr the earliest soviet posters advocating natalism appeared at the beginning of the 1920s immediately after the revolutionary and civil wars, when the new regime consolidated its military and political power just enough to focus on social and cultural issues affecting the new political entity. they pointed both to ideological concerns — the increased role in child rearing on the part of the state as one of the goals to create a new generation of productive soviet citizens — along with practical replenishment of those who died in combat. while this concern found its 4 “new report on replacement migration issued by un population division” (17 march 2000), united nations press release (consulted 7 july 2013): http://www.un.org/news/press/docs/2000/20000317.dev2234.doc.html. 5 see, for instance, “merkel says german multicultural society has failed,” bbc (consulted 8 march 2013): http://www.bbc.co.uk/news/world-europe-11559451. kouprianova ● modernity and natalism in russia 151 greatest expression throughout the 20th century, it had even earlier roots within the modern western liberal paradigm which undoubtedly affected pre-revolutionary russia. this was the state’s growing attention to the bodies of its subjects, which at the end of the 19th and the beginning of the 20th century pertained to questions of cleanliness, hygiene, and public health particularly in the case of the lower classes. arguably, one of the better known initiatives in public memory was that of jacob riis, whose photographs of new york slums were used to garner interest in social reform (riis 1890). in the nineteenth century russian empire, the state’s involvement in the matters of the body pertained by and large to the question of corporal punishment and military service. otherwise, family matters were generally a private issue as compared to modern and contemporary macroand micro-management; there was no unified system of school education, for instance. large families were desirable for two reasons. first, much like elsewhere, having several children ensured that at least some of them survived into adulthood at a time when many detrimental healthrelated factors, such as epidemics, could strike populations. second, the russian empire was generally a religious society, making the patriarchal family structure and large family size preferable. for religious adherents, regardless of denomination, there was a certain sense of shared responsibility to having children, which was not one of the state, but one of faith and cultural traditions. in the last hundred years, if any generalisations could be made about the purpose of natalist policies in the ussr and russia from the point of view of the state, then it is two-fold. first comes geopolitics in order to maintain a population large enough to defend the country’s borders. second comes social engineering within the general context of modernity in order to provide adequate workforce-taxpayer numbers among responsible patriotic citizenry. when it comes to geopolitics, natalism served as a method for producing an adequate number of citizens to defend the territorial integrity of the ussr and the vast natural resources within — oil, gas, minerals, etc. in modern-day russia, this issue is even more pressing if one were to consider the rising commodity prices. in light of the ussr’s and contemporary russia’s uneven population distribution with high concentration in the european part of the continent, the state engaged in a number of development and relocation initiatives in siberia and the far east in order to reduce vulnerability in those areas (e.g., the creation of entire industrial cities in the ural mountains, such as magnitogorsk, in the soviet period and economic and infrastructural projects in the contemporary russian far east, albeit lesser by comparison, like rosneft’s mining operations in sakhalin). this frontier drive can be traced back to russia’s earliest expansion eastward and yermak’s conquest of siberia in the late 1500s, not only as the means to access important trade routes, but also to control the newly acquired lands militarily and through the presence of a sufficiently sized subject population. in the modern period, this vulnerability has manifested on numerous occasions — from the british attempts to check russia during the so-called great game in central asia and the crimean wars in the late imperial period to the thinly veiled attempt to move into resource-rich siberia and the far east during the allied “intervention” on the part of anglo-americans along with canada, as well as france, china, and japan in 1918, nazi germany’s concept of lebensraum, americans’ so-called “containment” policy during the cold war, and the ongoing encirclement of russia in the post-soviet space, for instance, in central asia. these seemingly unrelated events can be explained by the general angloamerican geopolitical trajectory, surpassing ideology, from such theorists as alfred mahan at the end of the 19th century to zbigniew brzezinski at the end of the 20th. terminological fluctuations notwithstanding, the general thrust of this strategy is the control of the heartland, to use halford mackinder’s name for russian-ruled eurasia, in order to gain access to the majority of the world’s natural resources european journal of government and economics 2(2) 152 (mackinder 1904: 421-437). thus, having a large and more evenly distributed population size can be seen as a matter of national survival for the russians. in terms of ideology, after the bolshevik revolution, the soviet government exhibited considerable interest in natalist thinking, although its specific treatment changed. much like the natalist concerns in the modern states of the west in the interwar period, the ussr sought to manage its citizens to an unprecedented extent, ranging from setting an adequate level of education, culture, and taste to dictating attitudes toward their bodies, including the questions of hygiene and reproduction. at first, in the 1920s, the bolsheviks attempted to adhere to classic marxism. in the arena of familial relations, this meant getting rid of patriarchal gender roles, legalizing abortion, and making divorce procedures simpler, among others. certain radical figures, such as aleksandra kollontai, viewed the family institution and traditional christian morality itself as remnants of the bourgeois past that had to be discarded, with the worker-state gaining a prominent role in raising children communally. the state did become more involved: the earliest soviet poster advertising targeting social matters depicted orphanages featuring classical architecture, which stated, “children are the future of soviet russia”; others promoted the link between the health of the mother and her baby through government-issued healthcare (snopkov et al 2006). in the 1930s, however, the egalitarian measures from the previous decade were replaced by top-down “neo-traditionalism,” as western historians refer to it. here, the state acted as the civiliser, including the reintroduction of ethnic heroes and classics in the arts and education, in light of the dramatically increased literacy level for which it was responsible; it also made vast improvements in the area of hygiene, fighting diseases, and infant mortality. for the family, this meant a return to somewhat patriarchal family values (but not in terms of pre-revolutionary property rights of the patriarchs or stay-at-home bourgeois housewives), sobriety, as well as natalist propaganda coupled with a ban on abortion, difficult-to-obtain divorces, and mockery of irresponsible husbands.6 this change was especially noticeable in visual culture. 1920s-early 1930s imagery depicted women in a rather gender-neutral manner. by contrast, by the mid-to-late 1930s, the emphasis on femininity re-entered social advertising (hoffmann 2003). furthermore, whereas this subject is beyond the scope of this paper, it is worth pointing out that other countries in the soviet bloc, such as romania, exhibited certain comparable elements of state-imposed neo-traditionalism, such as restrictive abortions (soare 2013: 59-78). in part, the state’s initiatives were a practical way to address the drop in fertility, creating a population crisis in the 1930s, which greatly concerned soviet demographers. the crisis was caused by a number of factors, including the 193233 famine and the elimination of female unemployment, as a result of industrialisation and the collectivisation drive in agriculture. with women entering the workforce in such a radical manner in a short time period, the government took over the role of child care, starting as early as nursery schools (pre-kindergarten) around the tender age of two (hoffmann 2003). along with successes in the area of health and medicine, specifically infant-mortality reduction, the availability of such early childhood education facilities comprised the most significant solutions. whereas we obviously cannot argue for the presence of a direct, causal link between this kind of social advertising and fertility, it does reflect traditional cultural mores at large. 6 it is important to note that the ussr, like the russian empire before it and the russian federation afterward, was a multiethnic entity. cultural specifics notwithstanding, non-slavic regions, such as the traditionally muslim-dominated areas in northern caucasus or central asian republics, could be described as exhibiting an even greater degree of societal traditionalism. kouprianova ● modernity and natalism in russia 153 the next decade saw the establishment of the heroic mother (literally, “motherheroine”) medal (1944) for women with 10 or more living children (naturally born or adopted), as well as two other lesser medals for similar achievements; the lowest award, the medal of motherhood, commemorated those with five or six children. this type of recognition symbolised one of the ways in which the soviet government chose to combat the massive loss of life (24-26 million) during the course of wwii. after the war, the ratio of men to women was rather skewed, which caused additional problems. at this time, the state increased financial help to women who were pregnant or raising children, in general. natalist propaganda intensified, creating or, at least, emphasizing the atmosphere of communal support. poster advertising featured images of wholesome, but serious mothers surrounded by numerous children of all ages, including teenagers already serving in the soviet armed forces. others promoted orphan adoptions to mitigate the effects of war.7 the latter was a unifying experience for ussr’s citizens, which approached the need to rebuild on a national basis as a common goal. for instance, at the end of the war, approximately 41% of orphaned children was adopted or in foster care.8 figure 1. total fertility rate, 1950-2011 1.0 1.2 1.4 1.6 1.8 2.0 2.2 2.4 2.6 2.8 3.0 1950 1960 1970 1980 1990 2000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 data source: institute of demography at the national research university 'higher school of economics', demoscope.ru. natalist social advertising continued throughout the majority of the soviet period. many examples targeted proper child care, in which the government played an active role, while the mother was at work, on the way to raising the new generation of well-educated, cultured, responsible, and healthy soviet citizens. abortion was legalised once again after stalin’s death (1953), as part of the general khrushchev era liberalisation. somewhat lower birth rates in the 1970s have, in part, been explained by the rise in the standard of living and related embourgoisement of the brezhnev period. soviet authorities provided reasonable time for maternity leave, maintaining salary and employment, as well as free post-secondary education.9 this meant that there was little conflict, at least in theory, between having several children and being able to provide for them. 7 see poster selection “materinstvo v illiustratsiiakh,” materinstvo.ru (consulted 10 march 2013): materinstvo.ru/art/6214/; “gosudarstvennye nagrady za materinstvo,” ria novosti (consulted 10 march 2013): http://ria.ru/society/20090708/176642182.html. 8 “istoriia usynovleniia v rossii,” usynovlenie.ru (consulted 7 july 2013): http://www.usynovite.ru/experience/history/chapter1/. 9 112 total calendar days before and after birth, maintaining salary and employment were introduced as early as 1917, with the possibility to extend the leave (maintaining employment without maintaining full salary) up until the point when the child reaches 1 year. additional measures were taken for women with multiple children (“otpusk po beremennosti i rodam,” bolshaia sovetskaia entsiklopediia (moscow: sovetskaia entsiklopediia, 1969-1978)); see same at http://demography.academic.ru/2158/отпуск_по_беременности_и_родам. european journal of government and economics 2(2) 154 natalism and modernity in scholarship before moving onto contemporary natalist initiatives in the russian federation, it is worth having a brief look at western historiography about the subject of “neotraditionalism” in the ussr, some of which was already used above. it is also important to consider the limitations of current western scholarship linked to its own assumptions analyzing foreign subjects from a liberal (and now, post-liberal) perspective through the prism of the individual and the seemingly ever expanding “human rights” at the expense of authentic folkic communities. the earliest texts about 1930s ussr provided a rather straightforward ideological critique, arguing that the stalinist regime backtracked on the egalitarian achievements of the bolshevik coup d'état, such as women’s rights. these texts can be traced to leon trotsky’s revolution betrayed (1937), in which the exiled politician used marxist analysis in order to highlight the conservative turn within the soviet revolution with the rise of stalin. this strict ideological reading made him describe 1930s “neotraditionalist” programme as a general embourgoisement of the soviet society, including patriarchal family structure and the emphasis on having several children (trotsky 1937; timasheff 1946; dunham 1976). in contrast to this analysis from a strictly marxist ideological perspective, in the last 20 years or so, the modernity paradigm has become quite prominent in historical circles, bridging the gap between the different ideologies of the 20th century and drawing from different scholarly publications, including the work of michel foucault and james scott’s idea of modern population management, as mentioned above (foucault 1975; scott 1998). born during the enlightenment era, modernity spread throughout europe, russia, and beyond taking on culturally specific forms and thus resulting in an entire spectrum of “modernities.” its three main ideologies of the 20th century, liberalism, communism, and fascism, can also be described as three political theories. each theory had a different historical subject: the individual in liberalism, class in communism, and the state or race in fascism, affecting the specifics of political expression. these ideologies challenged each other as the best representation of modernity, and liberalism ultimately triumphed.10 at the same time, they all shared several fundamental similarities, such as secularism, rationalism, materialism, economic determinism, and the belief in unidirectional, infinite progress as part of its unequivocal trust in science. in fact, the latter notion of continuous progress, not cycles of growth and decline, was conceived as the main trajectory of social and cultural development and the principal driving force in history (dugin 2009). scholars of russian history in the west have used the modernity spectrum to examine the issue of natalism and beyond, arguing that liberal, communist, and fascist states were all concerned with the bodies of their subjects, including, but not limited to: birth rates, health and hygiene, education as a form of social engineering, population replenishment after war; taxation levels (hidden or otherwise) as being linked to workforce size and social services of the western welfare state or industry, agriculture, and the military in the historically socialist states; subject treatment on a largely ideological level, such as promoting or discouraging abortion as an issue of women’s civil rights or traditional family values, and even aesthetics as they pertained to the question of race in nazism and that of class in communism. this type of reading acknowledges cultural variation, but prefers the comparative framework of the generalised modernity paradigm. western historians especially gravitated toward what they saw as the most “natural” comparison between nazi germany and stalinist ussr as the two most 10 the triumph of liberalism (for the sake of convenience, in 1991 with the soviet collapse) as the most accurate representation of modernity also coincides with the beginning of postmodernity. kouprianova ● modernity and natalism in russia 155 extreme cases of modern population management (hoffmann 2003). however, other studies demonstrated that this phenomenon was not limited to what these historians defined as “authoritarianism” (previously, “totalitarianism”). for instance, mary louise roberts examined the natalist drive in france in the wake of the first world war and its massive loss of life, including suppressing contraception, linking it to vichy regime era conservative gender relations (roberts 2003: 91-101). others analysed the authoritarian streak within nominally liberal regimes, as in the case of gordon chang’s study of the japanese internment camps in the united states during wwii (chang 2003: 189-204). depicting the natalist drive in the west and russia as one of the many population management techniques used by a powerful and intrusive state is not inaccurate, but provides only one piece of the puzzle. it certainly does not fully explain the support of the modern state on the part of the masses. for instance, what was the relationship between the soviet government and nearly half a million women — i.e., the total number of those awarded the heroic mother medal — who were motivated to have more than ten children, each?11 western historians have realised the sheer complexity of this question when they linked it to the subject of resistance — that of individuals against the state. one of the difficulties here is rooted in divergent views of what comprises resistance, active and passive, and how to explain the support of a powerful state outside of various types of coercion (viola 2002).12 a recent study by the carnegie endowment fund concluded that contemporary russians are “confused” because stalin’s popularity has grown among them, which does not fit into this think tank’s liberal world view.13 the latter is part of a greater trend, often ignoring cultural diversity and specificity and measuring non-western societies by liberal standards as mentioned above. there has to be something else that could adequately explain the relationship between the modern state, the supportive masses (exceptions notwithstanding), and the institution of “neo-traditionalism,” which stood in seeming contradiction to the early years of the soviet regime, when it attempted to adhere to marxist dogma as closely as possible. recently, russian philosopher alexander dugin attempted to explain that very “something.” admittedly, he ventured beyond the subject of history and used gilbert durand’s two-tier method — the sociology of the imaginary, which synthesises the work of carl jung on the collective unconscious and that of david émile durkheim on the collective consciousness. this sociological research area refers to the sum total of overt cultural features as the logos, whereas mythos represents the symbolic and archaic undercurrents pushed into the unconscious in the rationalist, secularist, science-focused modern period. ideally, there should be a strong connection between the logos and mythos, and the former should be an expression of the latter. the sociology of the imaginary has existed for a number of decades, but is a rather new direction for examining one’s own history for russian scholars. its primary usefulness lies in its ability to supplement the aforementioned historic models (dugin 2010). in fact, dugin asserts that the last several hundred years of russian history, by and large, comprised the imposition or borrowing of a foreign western european logos — and the transition from one such logos to another. the latter can be contrasted 11 between 1944-1991, according to “presidium of the supreme soviet of the ussr adopted the first decree ‘on awarding the honorary title of ‘mother hero’ to mothers of many children’,” president’s library (consulted 10 march 2013): http://www.prlib.ru/en-us/history/pages/item.aspx?itemid=296. it would be too simplistic to reduce these numbers to the lack of birth control. 12 the question of resistance in the ussr is beyond the scope of this paper. 13 de waal, thomas et al (2013) “the stalin puzzle: deciphering post-soviet public opinion,” the carnegie endowment fund for international peace (consulted 1 march 2013): http://www.carnegieendowment.org/2013/03/01/stalin-puzzle-deciphering-post-soviet-public-opinion/. european journal of government and economics 2(2) 156 with the fixed unconscious level of russian culture, including myths, symbols, and dreams. under the influence of the rationalist enlightenment paradigm, there has been a tendency to suppress one’s own myths (whether in the case of crushing the folk ways by the europeanizing policies on the part of peter the great or the attempts to root out orthodox christianity during the early bolshevik days), pushing them into the unconscious tier. this approach unifies and interprets the seemingly diverse processes and regime changes in russian history in this broad time period under the umbrella of modernity, during which gradual secularisation occurred (dugin 2010: 73). the russian version of the mystical nocturnal regime, in which the country’s collective imagination functions, is associated with dreams of unity, absorption, and amalgamation, hence the appeal of the communal aspects of communism to russians as a whole.14 as a result, dugin asserts that the popular support for communism in the ussr was based on the marriage between the rational marxist logos and the irrational russian mythos, thereby establishing the soviet system, which was oriented toward achieving this mystical unity in the future. in other words, many aspects of the soviet system embodied the realisation of the people’s collective unconscious. in fact, despite its rationalist, secularist roots in the enlightenment period, the communist myth itself was a “unique and special archetype of the collective unconscious, a dream about an earthly paradise, in which the eschatological messianic cult opposed the official christian religion,” writes dugin (dugin 2010: 22, 60-65, 77-78). ultimately, the myth of progress within modernity at large — leading toward this kind of a paradise — was, too, akin to irrational religious sentiment. let us apply dugin’s argument about mythic undercurrents under the apparently secular and rational communist regime to cover the subject of “neo-traditionalism.” we can suggest that the partial return to certain traditional aspects of russian culture, albeit within the framework of a modern state, was not merely a calculated population management method. the authorities themselves functioned as carriers of the strong folkic undercurrents in russian culture, and, after the first dozen years under the strict marxist — foreign — logos, the system morphed to include those aspects that were more naturally acceptable to the russian society, such as those elements of “neo-traditionalism” described above.15 thus, we can summarise the selective turn to traditionalism within the modern framework as follows. revolutionary regimes that came into power through violence and instituted a radical social, political, and cultural programme cannot continue to function in the same extreme mode for an indefinite time period. eventually (shortly), they arrive at normalisation, which considers the cultural particulars of its subject population as a kind of negotiation between them and the ruling elite. in the case of twentieth century russia, an almost 10-year period comprising the naked violence of wwi, the bolshevik revolution, and the civil war was accompanied (or followed) by a short period of turbulent social and cultural changes when the new soviet state implemented a programme closely reflecting the foreign marxist doctrine. by the mid-1920s, the state began to gradually introduce policies, haphazard at first, that were closer linked to the cultures comprising the ussr, resulting in “neo-traditionalism” by the mid-1930s. roughly speaking, the case with the late 1980s-1990s was similar in terms of the attempts 14 dugin’s assertion is based on several culturological factors, such as the predominance of powerful female characters in russian fairy tales. 15 there is even evidence for dugin’s assessment about the feminoid nature of russian culture, with links to powerful female folk-tale characters, in 1930s social advertising. despite the seeming reintroduction of patriarchal family life, the latter frequently depicted women as the centre and the glue of the family, whereas men were portrayed as irresponsible and prone to mistreating wives and children. kouprianova ● modernity and natalism in russia 157 to rapidly import another foreign liberal-capitalist system — with every imaginable negative consequence — followed by relative normalisation of the putin era. post-soviet russia with that in mind, let us have a look at the last 20 or so years of russia’s demographic policies and cultural perspectives, informed by an understanding that post-soviet russia has partially adopted the liberal-capitalist system. for instance, its current constitution is based on western european models, and the international monetary fund (imf) puts pressure on the country to tighten its fiscal policy and increase the retirement age, as it does in europe. whereas liberalism takes root at the same enlightenment source as communism, its focus on the individual is even less of a fit for communally oriented russians. indeed, the tension between russia’s traditionalist (authentic or state-sponsored) perseverance in the face of western liberal onslaught has even been implicitly mentioned in recent demographic studies about the country. sergei zakharov notes, for instance, that russian completed-fertility levels of real cohorts differ little from europe, yet russia exhibits “neo-traditionalism” by displaying an early age for marriage and having the first child along with preferences for the woman’s primary role as a mother (zakharov 2008: 250). in addition, a more “clan-like” structure, in which grandparents and other relatives take an active role in child rearing — in contrast to the western ideal of a “nuclear” family — is also quite prevalent. ultimately, despite all these reasons, his argument is that russia is heading the way of europe if trends persist. but to what extent are these conclusions drawn from the constraints of their ideological framework? after all, for demographers and policy makers, it is customary to discuss the latter in the context of familiar keywords, “development” and “progress” (here, explicitly comparing russia to western europe). such discussions are based on two major assumptions: that there is a single linear path of “development,” and that this path ultimately ends up with western europe, perhaps with regional peculiarities that are trivial in the grand scheme of things; that progress necessarily has positive value, as its connotation itself implies. this assumes that each subsequent generation supersedes and improves upon the previous one: in other words, our ancestors are necessarily believed to be lesser versions of us because they did not reduce their existence to an economic-based, consumerist standard of living. scholarly debate is normally framed in these particular terms without questioning the extent to which population management along with “development” in the direction of the contemporary west are even necessary; if any, criticism comes from ecologists concerned with climate change, population growth, and limited resources. indeed, it is possible to view the russian case in the opposite way: the persistence of “neo-traditionalism” as a way of resisting this kind of “development”; resilience in the face of modernity and postmodernity both under soviet communism and partial liberalism. in terms of fertility, there was noticeable decline toward the end of the soviet period around the time of perestroika with deaths exceeding births shortly after the collapse of the ussr, according to russia’s statistical agency, rosstat at gks.ru. in that decade, in addition to such detrimental factors as lowered life expectancy, abortion and divorce rates had been described as the highest in the world by the united nations’ web resource unstats.un.org. considering the sheer turmoil of that period, ranging from questionable privatisation of public assets during the so-called “shock therapy” years to hyperinflation of the currency, causing the destruction of lifetime savings, the immediate assumption on the part of the government has been that the steady improvement of economic conditions would create an increase in birth rates. in 2006, putin made russian demographics one of the central issues of his presidency during his address to the nation. european journal of government and economics 2(2) 158 since then, the solutions to russia’s dire population predictions have followed a two-tier approach involving straightforward financial rewards along with legal components as well as a socio-cultural initiative. the best known example of the former is the maternal-capital programme, which became operational in 2007. the latter involves various types of subsidies for women with more than one child, naturally born or adopted. this government aid package can only be spent on the child’s education, pension investment, or the improvement of one’s living conditions. residential real estate in the largest cities in russia (moscow, st. petersburg) remains near the top of the most expensive list in the world. this factor alone impedes certain urbanites from having more than one or two children. in terms of the judicial aspect, the state’s attempts to combat relatively high mortality (as compared to western europe) among men — both due to disease and accidents — have involved increased punishment for reckless or inebriated driving, limits to the sale of beverages with high-alcohol content, and others. socio-cultural propaganda ranges from billboard advertising and holidays to television shows. there are parliamentary suggestions to reintroduce more modest versions of the “motherhood” medals.16 even western-modeled popular television dating shows such as let’s get married! (davai pozhenimsia!) systematically portray young female contestants who reaffirm their desire of marriage and children as their top priority, with careers being secondary, despite coming from urban environments. one particular cultural initiative is the all-russian day of family, love, and faithfulness celebrated on 8 july. this is a typical example of a contemporary “neo-traditionalist” project by the state — established in 2008 — but with roots going back to older russian culture, named after ss. peter and fevronia, medieval patron saints of marriage and love. and while the holiday was instituted by a secular government, its very focus points to another player in this matter, the russian orthodox church. a traditional pillar of society, the orthodox church has experienced a true revival in the last 20 years, and has also been active in the demographic initiative from an overarching socio-cultural perspective. in addition to conventional measures such as the construction of thousands of new places of worship and the creation of church-based orphanages, the church has also embraced the new media, such as social networking, to spread its message of traditionalism. this institution sees the russian population decline as part of the same trend affecting europe (i.e., other historically christian nations), naming the ideology of materialist, individualist liberalism as the culprit. furthermore, in the last decade, there has been a certain sense of rapprochement between the orthodox church and other faiths, such as the indigenous muslim clerics in russia, who advocate similar goals for their respective regions. overall, the last 20 years, since the transition into a post-soviet environment, demonstrate lower natural population gains than those of the soviet period as a whole. of course, we cannot discount the economic factors in their entirety. in general, russians no longer have the multitude of social guarantees that they enjoyed as soviet citizens. in addition to that, in the challenging decade of the 1990s, the loss of lifetime savings as a result of hyperinflation in the painful transition to a market-based economy made some russians cautious, particularly in urban environments, in fear of being unable to provide for their families. as a result, they limit themselves with having one or two children. the years of putin’s leadership, specifically, have coincided with a slight natural population increase as compared to the previous decade, although still below desirable levels. however, it is difficult to attribute the latter to government policies, as a cause and effect, rather than the overall stability of the first decade of the 21st century. yet economic factors are only part of the equation, and it is the socio 16 “gosduma vernet zvanie ‘mat’-geroinia’,” izvestiia (consulted 10 march 2013): http://izvestia.ru/news/545459. kouprianova ● modernity and natalism in russia 159 cultural dynamic that is likely to be of greater importance. despite its selfproclaimed universalism, the contemporary individualist liberal-capitalist value system is specific to the west, and is even further removed from traditional russia than hybridised soviet socialism. the real solution to the country’s demographic issues lies in the realisation of its authentic culture on political and social levels. to paraphrase dugin, russia’s politics — social and otherwise — must embody the country’s own logos, and the latter must be intimately connected to its mythos. references antonov, anatolii (2010) 'demografiia v eru depopuliatsii,' demograficheskie issledovaniia, vol. 1 (consulted 7 july 2013): http://demographia.ru/articles_n/index.html?idr=19&idart=78. chang, gordon h. (2003). 'social darwinism vs. social engineering: the ‘education’ of japanese americans during world war ii.' in landscaping the human garden: twentieth-century population management in a comparative framework, ed. amir weiner, 189-204. stanford: stanford university press. dugin, aleksandr (2009) chetvertaia politicheskaia teoriia. st. petersburg: amfora. dugin, aleksandr (2010) logos i mifos. moscow: akademicheskii proekt. dunham, vera (1976) in stalin's time: middleclass values in soviet fiction. cambridge: cambridge university press. foucault, michel (1975) surveiller et punir. paris: gallimard. hoffmann, david (2003) stalinist values: the cultural norms of soviet modernity, 1917-1941. ithaca: cornell university press. mackinder, halford j. (1904) 'the geographical pivot of history,' the geographical journal 23(4): 421-437. martin, terry (2000). 'modernization or neo-traditionalism? ascribed nationality and soviet primordialism.' in stalinism: new directions, ed. sheila fitzpatrick, 348367. new york: routledge. riis, jacob (1890) how the other half lives: studies among the tenements of new york. new york: charles scribner’s sons. roberts, mary louise (2003). 'the dead and the unborn.' in landscaping the human garden: twentieth-century population management in a comparative framework, ed. amir weiner. stanford: stanford university press, pp. 91-101. soare, florin stanica (2013) 'ceausescu’s population policy: a moral or an economic choice between compulsory and voluntary incentivised motherhood?' european journal of government and economics 2(1): 59-78. scott, james c. (1998) seeing like a state: how certain schemes to improve the human condition have failed. new haven: yale university press. snopkov, aleksandr et al. (2006) materinstvo i detstvo v russkom plakate. moscow: kontakt-kul’tura. timasheff, nicholas (1946) the great retreat: the growth and decline of communism in russia. new york: e. p. dutton & company. trotsky, leon (1937) the revolution betrayed, max eastman (tr.). new york: doubleday. viola, lynne ed. (2002) contending with stalinism: soviet power & popular resistance in the 1930s. ithaca: cornell university press. zakharov, sergei (2008) 'russian federation: from the first to second demographic transition,' demographic research 19(24): 907-972. microsoft word ejge_03_01_010.doc european journal of government and economics volume 3, number 1 (june 2014) issn: 2254-7088 47 interest group influence in micro-states: the role of networking skills direnç kanol, university of siena, italy abstract this paper argues that an interest group’s networking skills in micro-states may be as important, if not more important than other variables discussed in the interest group influence literature. this argument is based on the recent literature on democratisation in micro-states which shows that politics in these states is personalistic in nature. the argument is supported by expert interviews undertaken in the republic of cyprus and the turkish republic of northern cyprus. jel classification d72; d73, d74. keywords cyprus; interest groups; interest group influence; lobbying; micro-states; networking skills. european journal of government and economics 3(1) 48 introduction what determines interest group influence? this question is crucial to all who value democracy. in theory, one-person, one-vote rule should ensure that each citizen has an equal say in public policy. this is, however, far from the truth. organised interests try to influence policy and at times, they achieve their goals. hence, the public policy literature focuses on how influential organised interests are and why one group may be influential whereas another may not be. the predominant belief in the literature is that providing information and offering citizen-support to the policy-makers enable interest groups to exert influence. the level of influence also depends on the preferences of the policy-makers, public opinion and issue salience. interest group literature, however, does not pay enough attention to the role of networking skills. this paper argues that an interest group’s networking skills may be as important, if not more important than other variables discussed in the interest group literature. this argument is based on the recent literature on democratisation in micro-states which shows that politics in these states are personalistic in nature. personal ties are crucial for getting things done. the argument is supported by empirical evidence gathered from expert interviews in the republic of cyprus and the turkish republic of northern cyprus. the following section reviews the interest group literature and shows that the role of networking skills as a determinant of influence is critically under-researched. the section after that describes the nature of politics in micro-states and generates a proposition regarding interest group influence in these states. after discussing the method and presenting the findings from the in-depth interviews, the final part discusses the implications of this study. interest group influence: a literature review some authors argue that interest groups lobby ‘friendly’ legislators, those who already agree with them, and the legislators use the interest groups to acquire much needed information to attain their goals (bauer et al, 1963; milbrath, 1963). this view has been criticised by some who suggest that lobbying sympathisers serves only to counteract the effect of other groups trying to push the policy in the opposite direction, and lobbying ‘foes’ is not an exception but a norm (austensmith and wright, 1994). also, we should not forget that there are a significant number of scholars who argue that lobbying ‘swing’ legislators is also common (smith, 1984; wright, 1990; schlozman and tierney, 1986). if the first group of authors is right, this either means that interest groups are mere information conveyors to policy-makers and they do not exert much influence on policy-makers (hall and wayman, 1990) or they select to convey information to ‘friends’ in power in order to achieve their own objectives that is congruent with the public actors’ objectives (hall and deardorff, 2006). although the nature of interest group influence is still a puzzle, exploring the determinants of interest group influence is equally important. the most parsimonious theory that can be used to explain the relationship between interest groups and decision-makers to date is the resource dependence theory. the roots of this theory can be found in the work of the sociologists levine and white (1961). an important extension to the resource dependence perspective was presented by pfeffer and salancik (2003). studying health organisations, levine and white (1961) argued that no organisation is able to control all the resources it needs to achieve its goals. this is why organisations are dependent on other organisations to accomplish their tasks (levine and white, 1961). according to pfeffer and salancik (2003), interdependence is the key word for explaining exchange between organisations. dependence of one organisation on another can be measured by the importance of the resource that the organisation receives through interactions with the other, measured in turn by the relative magnitude and the nature of the resource (pfeffer and salancik, 2003). if an organisation absolutely needs a certain kanol ● interest group influence in micro-states 49 type of resource to attain its goals, then this resource is essential for that organisation (pfeffer and salancik, 2003). dependence on the environment refers to the importance of the resource to the organisation demanding that resource as well as the availability of the resource from other sources (jacobs, 1974; pfeffer and salancik, 2003). based on this understanding, one may argue that organisations are expected to be more responsive to other organisations that control the most problematic/critical resource (pfeffer and salancik, 2003). as a priority, we need to start the discussion about governmental organisations by acknowledging their considerable power when choosing which interest groups to contact and which ones not to contact. nevertheless, like any other organisation, governmental organisations do not possess all the resources they need to accomplish their tasks. so, it is inevitable that the criticality of the resources possessed by interest groups will have an effect on governmental authorities’ decision to engage in resource exchange with these groups. it should come as no surprise that scholars studying access and influence of interest groups to governmental actors have benefited from this theory. scholars working on the european union produced interesting research that strived to explain the logic of access and influence by using the resource dependence theory (bouwen, 2004; klüver, 2013). the vast majority of research on interest groups concentrates on the power of information as a determinant of influence regardless of the method or discipline. we can speak of two types of information: political information and technical information. plausibly, most technical information would also entail some political aspects but what matters is that technical information gives predictions about policy consequences in the technical sense. political information, on the other hand, is about signalling mobilisation consequences of the group or other groups as a result of an action that could be taken by a policy-maker (heitshusen, 2000, potters and van winden, 1992; austen-smith, 1998). governmental officials’ time is restricted and they are neither able nor willing to give access and influence to each and every interest group. in line with the resource dependence theory, one can argue that policy-makers grant access and influence to informative groups (reenock and gerber, 2008; crombez, 2002; bouwen, 2004; klüver, 2013). another crucial resource is citizen-support (klüver, 2013). policy-makers try to understand the magnitude of support for or opposition against a policy-proposal because if policy-makers enact laws that are highly unpopular, they pay for this in the next election (kingdon, 1995; fiorina, 1989; mayhew, 1974; arnold, 1990). with re-election prospects in mind, public actors look to have the endorsement of organisations with the support of a sizeable number of citizens (klüver, 2013). therefore, based on the resource dependence perspective, citizen support can be interpreted as a good for access and influence where electoral support is the causal mechanism that gives meaning to this relationship. citizen support does not only affect the politicians with the ambition to get re-elected but also the bureaucrats. despite the fact that bureaucrats may have more manoeuvring space as they do not run in the elections, they look for public endorsement and legitimacy (poppelaars, 2009). the higher the support of citizens an organisation has that gives its endorsement to a bureaucrat, the higher the perceived legitimacy of that bureaucrat will be. therefore, citizen support is also expected to affect bureaucrats as well as politicians. although resource dependence theory suggests a parsimonious framework for explaining influence, we cannot oversee the varying nature of political context and its impact on the opportunity to exert influence. political opportunity structures (kitschelt, 1986; meyer and minkoff, 2004) and political mediation (amenta et al, 1992; amenta et al, 1994; amenta et al, 2005) theories that were developed by scholars studying social movements took this aspect into account and suggested that the characteristics of the political system and the politicians have a considerable amount of impact on social movement outcomes. various studies european journal of government and economics 3(1) 50 show that interest groups are more likely to exert influence, for instance, when the politicians are friendlier towards the cause of the lobbyists (kriesi et al, 1995; giugni, 2004; 2007; soule and olzak, 2004; mcveigh et al, 2003). this finding suggests that interest group influence also depends on which politicians are in power and interest groups may have to adjust their strategies depending on the political context. public opinion is also important. re-election motivated politicians take what the public thinks about a certain issue and how strongly they feel about this issue into account (kingdon, 1995; fiorina, 1989; mayhew, 1974; arnold, 1990). there are mixed findings in the literature. some research found that policy-makers are not responsive to public opinion (jacobs and shapiro, 2000; 2002; cohen, 1997). however, the predominant belief in the literature based on empirical findings is that public opinion matters. politicians are more likely to pass laws that are popular (page and shapiro, 1983; burstein, 2003; erikson et al, 2002; monroe, 1998; lax and phillips, 2009). a lobbyist that advocates an issue that is clearly opposed by the public may not have much success. we should also take issue salience into account. baumgartner et al (2009) argues that most issues do not find a place on the agenda of the policy-makers. policymaking process does not resemble an incrementalist one as lindblom (1959) argued. according to lindblom (1959), the policy-making process involves many actors and moves in small steps. a rather recent major work shows that this argument could be false (baumgartner et al, 2009). baumgartner et al’s (2009) results are based on baumgartner and jones’ (1993) punctuated-equilibrium model which argues that most issues remain stable for most of the time, and where policy-change occurs, it follows a pattern that is contrary to the logic of incrementalism. policy-change occurs very rapidly in few issue areas (baumgartner and jones 1993). salience of an issue may have a positive impact on the likelihood of an interest group’s success in exerting influence on the policy agenda (burstein, 1981; page and shapiro, 1983; haider-markel, 1996; 2003). once an issue becomes very salient and is placed on the political agenda, we may observe a rapid change of policy. what is clearly lacking in the interest group influence literature is a discussion about the role of networking skills. the role of networking skills is only very indirectly mentioned in a few studies such as carpenter et al (1998), who argue that interest groups exist in a network. a group which possesses weak ties with a sizeable number of groups in this network is able to receive and convey information, which in turn, provides it access to the governmental actors. heaney (2006) shows that playing a brokerage role between disconnected lobbying coalitions and political parties enhances a group’s influence in the policy-making process. finally, beyers and braun (2014) argue that an interest group’s ability to bridge different coalitions increases its chances of gaining access to the policymakers. however, none of these studies directly deal with the capacity of the individuals within interest groups to establish direct relationships with the policymakers. the nature of politics in micro-states and its implications for interest group influence different authors used different population sizes to define which states can be defined as micro-states (anckar, 2010). using certain numbers, however, can be problematic. consider, for instance, that we use 500.000 people as a ceiling to determine if a state is indeed a micro-state or not. can we really assume that theoretical arguments that apply to a country of 490.000 people do not apply to a country of 510.000 people? this paper uses a more deductive approach to defining micro-states that serve the topical purpose of this paper. a micro-state is a very small country with a very small number of population size and with a very large kanol ● interest group influence in micro-states 51 number of people knowing a large part of the rest of the population. the proverbial ‘everyone knows everyone’ is a crucial definition of the social and political life in micro-states. this deductive understanding would corroborate with the classification schemes of scholars who keep fiji with a population of 780.000 within the discussion about micro-states but exclude papua new guinea, with a population of 4 million (anckar, 2002a). anckar (2003; 2008) uses the 1 million ceiling and suggests that there are 42 micro-states in the world. therefore, the small size of these states should not mean that careful inquiries about their society and politics can be neglected. unlike the earlier studies which suggested that smallness is good for democracy (dahl and tufte, 1973; hadenius, 1992; srebrnik, 2004), the recent literature suggests that the relationship between smallness and democracy is ambiguous at best and small size may, in fact, have a negative effect on the quality of democracy (veenendaal, 2013a; 2013b; 2013c; corbett, 2013). the earlier literature points to homogeneity, a sense of fellowship and community, a lack of complexity in solving social and political issues, and the proximity between the rulers and the citizens as mechanisms to establish a positive link between smallness and quality of democracy (dahl and tufte, 1973; ott, 2000; anckar, 2002a; 2002b). the recent literature, however, suggests that politics in micro-states is more ‘informal’ in nature. almost everything depends on personal relationships. obligations to family and kin precedes the obligations to the society as a whole (corbett, 2013). thus, civic culture and social capital, which are argued to be essential parts of embedded democracies are not found in micro-states. the proximity between the rulers and the citizens which is argued to be a boon can actually be a bane. consider, for instance, the decision-making process for public appointments. it is very hard for the elected officials to be impartial when family, kin and friends are extremely important in social life and when there are many individuals that could be classified as either of these three. it is, also, almost impossible to avoid people one falls apart with. therefore, elected officials try to live up to the expectations of their close ones and acquaintances and the citizens try to keep good relationships with their government at all times. politics is conducted outside of the public domain. this causes issues with nepotism and clientelism (corbett, 2013; veenendaal, 2013a; 2013c). if politics in micro-states is personalistic in nature, then the organised interests should master the art of establishing personal relationships, in order to exert influence on the policy-making process. in these states, who you know may be much more important than what you know (information), which runs counter to the decades of theorising in the interest group literature. networking skills, therefore, may be particularly important for exerting influence in micro-states. by networking skills, i mean the ability of an individual to meet and establish good contacts with the governmental actors. a person may have the intellectual resources to generate and provide information to the governmental actors. however, this does not mean that he/she has the skills to establish weak or strong ties with these governmental actors. once a relationship between an interest group and a governmental actor is established, interest groups can ask for ‘favours’ from their contacts in governmental posts which is shown to be the main form of policy-making in the highly personalistic and clientelistic political decision-making process in microstates. therefore, networking skills are expected to be a strong determinant of interest group influence in micro-states. method expert-interviewing is a valuable technique when it comes to discovering what affects interest group influence. influence is the primary goal of lobbying. by practicing lobbying for long years, experts learn what works, what does not work and why. therefore, they are in a good position to suggest what tactics work and what the exogenous factors that may affect their influence may be. different european journal of government and economics 3(1) 52 interest groups may have different opinions about these exogenous factors. the goal of expert-interviewing method is not to get as many interviews as possible, but to get as many interviews from people who have the most in-depth information as possible. therefore, a policy field is chosen first and the names of the organisations that have the most in-depth information about how to exert influence are obtained. the civil society organisations directory (2011) provides a good source to obtain the emails of a population of interest groups, both in the republic of cyprus and the turkish republic of northern cyprus on the island of cyprus. the republic of cyprus was established in 1960 as a partnership state between the greek cypriots and the turkish cypriots. after civil and political conflict intensified between these two communities, it has become a greek cypriot state in 1963. the turkish republic of northern cyprus was established in 1983. however, it is only recognised by turkey which makes it fall under the de facto state classification. these are micro-states with population sizes of 862,000 (statistical service of the republic of cyprus, 2011) and 294,906 (devlet planlama örgütü, 2011) respectively. therefore, concentrating on lobbying in these two states offers an opportunity to analyse the potential impact of networking skills for exerting influence in micro-states. i focused on the field of peacebuilding for the purposes of this paper since the experts in this field could be identified easily. a list of the population of organisations that potentially deal with peace related issues during the months of july and august 2012 is created. from august 2012 to october 2012, i conducted a survey that asked these organisations to name 3 well known peacebuilding organisations that are active in cyprus in order increase the number of observations. it should be stressed that an interest group in this paper refers to any non-governmental group that tries to influence public policy either by inside lobbying or outside lobbying tactics. therefore, non-governmental organisations (ngos) which engage in lobbying the governmental actors classify as interest groups. overall, 41 civil society organisations were found to be active in the field of peacebuilding with varying degrees of intensity. these organisations were emailed a questionnaire followed by a reminder email in case no response after a month. 25 organisations replied to the emails with a response rate of 61 percent. these organisations were asked if they were able to initiate contact with the decisionmakers at the governmental level face-to-face, via internet, phone or by other similar means in order to promote peace in the last year. out of these 25 organisations that responded to the survey, 14 of them answered ‘yes’ to this question. considering also the list of organisations that were named by other organisations as one of the 3 well known peacebuilding organisations, 6 peacebuilding organisations out of these 14 organisations were identified as true experts in the field of peace lobbying.1 i also included a turkish cypriot policymaker in the sample who was responsible for deciding the main aspects of the peace negotiations with the greek cypriots. kudret özersay, the chief turkish cypriot negotiator responsible for the negotiations with the greek cypriots from april 2010 to june 2012 was included in the sample as an expert policy-maker who is not a lobbyist but has expert knowledge on lobbying. özersay was lobbied by various organisations with respect to the reunification/peace issue. managers and/or project managers of these organisations and the policy-maker were interviewed face to face with the open-ended interview method. they were asked about the possible ways to gain influence and prompts and probes were posed depending on the discussion that took place. interviews lasted approximately 30 1 these organisations are: association for historical dialogue and research, the management centre of the mediterranean, ngo support-centre, cyprus academic dialogue, cyprus 2015 and the cyprus community media centre. kanol ● interest group influence in micro-states 53 minutes on average. the interviewees agreed to be recorded and identified for the academic purposes of this project. results in this section, the findings from the interviews are presented. evidence for the factors that were discussed in the literature review section are presented under separate sub-headings. the role of networking skills is discussed at the end of this section. information the experts suggested that the main factors discussed in the literature are all relevant. findings from the in-depth interviews provide evidence for the value of conveying information to the policy-makers. the common viewpoint of the interviewees is that technical information that could be of use to the policy-makers can make a big difference. in line with the resource dependence theory, the consensus is that the type of information matters. if peacebuilding organisations possess information that is valuable to the policy-makers and if it cannot be offered to them by other sources, the likelihood of influence increases. the experts suggested that providing information to the policy-makers is very important for gaining access. they also believe that it can increase the likelihood of exerting influence: even the policy-makers at the highest levels whom you might assume to have a network of people supplying them information, they are not aware of certain things… they do not know… so if you give them information that they did not know about, they may realise the usefulness of this information. let me give you an example... we did research on teachers’ perception on history teaching about three years ago and we found out that teachers were positive towards reform in history teaching. the policy-makers, on the other hand, thought on the contrary. so, they were afraid of something that did not exist. if you can bring new information in a rather scientific or objective way, you have the space to influence the policy-maker during the interaction process. (expert 1) from our experience, i think the more you contribute to the immediate interest of the decisionmakers, the more access and influence you will have. so if you are giving them something they can use and this is something they don’t have and they find this valuable, then the relationship becomes more of a mutual benefit kind. why do we have access as cyprus 2015? because we say listen; if you give up güzelyurt (morphou) or maraş (varosha), this is how much reaction you will get, or if you accept more than fifty thousand people of turkish origin to become citizens of a united cyprus, this is how much reaction you will get. how many greek cypriots will accept, not to be reinstated to their old properties? this is very valuable information... decision-makers need information. we try to bring expert knowledge to the attention of the policy-makers that they usually lack because they are surrounded by people who are not really experts… this can be technical or political information but information should be relevant with the needs of the policy-makers. (expert 2) citizen support information, however, is not the only thing the policy-makers need. politicians need votes. in other words, they need citizen support. bureaucrats need legitimacy and they get this via citizen support as well. how many members a peacebuilding organisation is representing or how many people can it mobilise? these are very important questions. the interviewees suggest that citizen support is pivotal for exerting influence: because the number of members an organisation is representing means votes for them, when the politicians are judging whether they should give you an appointment or take your ideas into consideration, they always ask themselves; ‘i am going to give him my half an hour or so; in return, what am i going to get from this guy? first, i can gain some information, but secondly and sometimes more importantly, it is very important for me to accept this guy because he may be representing 1,000 votes. i have to take him into consideration!. (expert 3) european journal of government and economics 3(1) 54 political allies although the exchange argument may be valid, we should also consider that influence is decided by exogenous political factors. the experts were of the same opinion that it is much more difficult to convince policy-makers that have a more ethno-nationalist approach to the cyprus problem. similarly, influence is limited when peace-building organisations lobby the policy-makers that are not favourably disposed toward civil society participation in the policy-making process. this argument is in line with the political opportunity structures and political mediation theories that argue that having political allies increases the likelihood of influence: naturally, the ‘leftist’ policy-makers that work on reconciliation are more inclined to listen to us. i’d lobby both friendly and non-friendly politicians though. because, in fact, you would need a good mix of people. the positive ones will most likely take you further. the negative ones, on the other hand, if you never approach them, then you would never have a chance to have an attempt to convert them. (expert 4) when you have a certain character who is negative towards civil society participation in policy-making and sees you as a nuisance and thinks that you are rocking the boat, he/she is going to shy away from you and it will be much harder to deal with him/her. (expert 5) public opinion the experts also mentioned the possible impact of public opinion. based on the findings in the literature, it is expected that if public opinion is not in line with the demands of the peacebuilding organisations, then these organisations are less likely to be able to exert influence. interviewees suggest that this may be the case. however, what is clear is that most issues that the lobbyists work on are more technical in nature and the preferences of the policy-maker himself/herself, irrespective of the public opinion, could be a more decisive exogenous factor that determines if influence takes place or not: of course, politicians, who want to be re-elected are not going to tell you ‘yes’ if you want something that a large number of voters are against. but there are ways around this. many times, we work on small technical issues that do not catch the eye of the public. therefore, politicians are less concerned with the public opinion in such cases. (expert 3) issue salience regarding issue salience and the likelihood of success, some interviewees suggested that lobbyists are more likely to be successful when there is a dynamic for resolving the problem at the highest political level. when there are some prospects with respect to the solution of the cyprus problem and the issue is salient, policy-makers are more likely to be willing to engage with the lobbyists: one very important issue is the general conjecture that you are living in. for example, if there is a situation when the reconciliation issue is becoming quite hot in the sense that we are at a stage of critical point in the negotiation process, it affects your work. in those times, peace issue occupies a very high spot on the agenda of the policy-makers and it is easier to influence those people because all of a sudden, they start thinking about reconciliation and peace and they want some ideas or some technical knowledge. when there is no possibility and everybody is very pessimistic about things, policy-makers just put the issue at the bottom of their agenda and they don’t want to hear anything about it. (expert 3) networking skills the experts, however, stressed that none of these variables may be as important as networking skills. in a small country, it is very important that people know the policy-makers. most individuals working for powerful interest groups have at least a few contacts, controlling the gates of political power. personal relationships make access and influence much more likely. it is much easier to establish long-term relationships with the governmental actors and ask for certain policies to be enacted if one possesses the skills to establish these contacts. interest group literature does not pay enough attention to the role of networking skills. the kanol ● interest group influence in micro-states 55 following quotations provide evidence for the critical nature of personal relationships and the skills to establish these relationships: personal contacts are very important. it is easier for the civil society representatives who know people to access the authorities and it is easier for the authorities to have access to civil society as well. (expert 6) i am an academic myself, and i happen to know the cyprus problem quite well, let’s say, since this is my research area and all that, i have a specific name that some people know, especially by the decision-makers. then, for example, i, as a member of a cso, if i want to have access, to want to go and see, to talk to a decision-maker, compared to somebody, let’s say an assistant at the eastern mediterranean university, of course, it makes a very big difference. (expert 2) i mean networking is an issue here ok… so, if an ex-policy maker establishes an ngo then obviously, he knows the channels or she knows the channels and she knows how to make a move and hence that ngo has … i’m giving an extreme situation… let’s say two academics and two politicians and two business advisors get together and form an ngo then that becomes a very strong structure immediately. because there is enough weight and networking substance that will increase access and influence. so by extension, networking is very important. (expert 4) i would stress the word capacity because you might have a hundred staff and not one of them may have the capacity to exert influence. you know when we talk about capacity you are also talking about people having good networks, not just their education and experience, but you know they might have good networks and you know they can pull the strings, this is quite important. (expert 3) discussion and conclusion this paper argues that networking skills are an important determinant of interest group influence in micro-states. since the role of networking skills is underresearched, the paper aimed to contribute to the literature on the determinants of interest group influence by providing evidence from the interviews undertaken in the republic of cyprus and the turkish republic of northern cyprus. the qualitative data obtained from peace advocacy experts in these two states provided some evidence for this hypothesis. a couple of points should be made for guiding future research on this topic. expert interviews conducted for this paper provided valuable evidence for the role of networking skills in the influence game. however, the absence of expert interview data from large states prevented testing a possible difference between interest group influence in micro-states and large states. networking skills may actually be more valuable for exerting influence in micro-states than large states when the highly personalistic nature of politics in micro-states is taken into consideration. future research could benefit from a most similar systems design (mssd) analysis and conduct a larger number of interviews both from a micro-state and a large state to see if networking skills are significantly more important for influencing political actors in micro-states than in large states. a careful reader may also question if the personalistic and clientelistic nature of politics in the republic of cyprus and the turkish republic of northern cyprus is indeed due to its smallness rather than other factors. after all, it is a fact that the countries in southern europe show similar political characteristics with these two micro-states when compared to the northern european countries. there is, however, much evidence that can suggest that smallness of these two states has an independent effect other than the geographical location of these micro-states. this is due to the opportunity available for us to compare the nature of politics in other micro-states with the two micro-states analysed in this paper. brilliant indepth research from micro-states all over the world suggest that similar phenomena exist in most micro-states in regard to the political process (see corbett, 2013; veenendaal, 2013a; 2013b; 2013c). at the expense of sounding tautological, it should be stressed that more research in different countries and other policy areas is needed to substantiate the argument european journal of government and economics 3(1) 56 put forward. nevertheless, this study aims to contribute to the literature on interest group influence and the nature of politics in micro-states. references amenta, edwin, neal caren, elisabeth chiarello and yang su (2010) ‘the political consequences of social movements’, annual review of sociology 36: 287-307. amenta, edwin, neal caren, and sheera joy olasky (2005) ‘age for leisure? political mediation and the impact of the pension movement on us old-age policy’, american sociological review 70 (3): 516-538. amenta, edwin, bruce g. carruthers and yvonne zylan (1992) ‘a hero for the aged? the townsend movement, the political mediation model, and us old-age policy, 1934-1950’, american journal of sociology 98 (2): 308-339. amenta, edwin, kathleen dunleavy and mary bernstein (1994) ‘stolen thunder? huey long's "share our wealth," political mediation, and the second new deal’, american sociological review 59 (5): 678-702. anckar, dag (2002a) ‘democratic standard and performance in twelve pacific micro-states’, pacific affairs 75 (2): 207-225. anckar, dag (2002b) ‘why are small island states democracies?’. the round table: the commonwealth journal of international affairs: 91 (365): 375-390. anckar, dag (2003) ‘direct democracy in microstates and small island states’, world develpment 32(2): 379-390. anckar, dag (2008). ‘microstate democracy: majority or consensus; diffusion or problem-solving?’, democratization 15(1): 67-85. anckar, dag (2010) ‘small is democratic, but who is small?’, arts and social sciences journal: 1-10. arnold, r. douglas (1990) the logic of congressional action. new haven: yale university press. austen-smith, david (1998) ‘allocating access for information contributions’, journal of law, economics and organization 14 (2): 277-303. austen-smith, david and john r. wright (1994) ‘counteractive lobbying’, american journal of political science 38 (1): 25-44. bauer, raymond a., ithiel de la sola pool and lewis a. dexter (1963) american business and public policy: the politics of foreign trade. new york: atherton press. beyers, jan and caelesta braun (2014) ‘ties that count: explaining interest group acces to policymakers’, journal of public policy 34(1): 93-121. baumgartner, frank r., jeffrey berry, marie hojnacki, david c. kimball and beth l. leech (2009) lobbying and policy change: who wins, who loses, and why. chicago: university of chicago press. baumgartner, frank r. and bryan d. jones (1993) agendas and instability in american politics. chicago: university of chicago press. bouwen, pieter (2004) ‘exchanging access goods for access: a comparative study of business lobbying in the european union institutions’, european journal of political research 43 (3): 337-369. burstein, paul (1981) ‘the sociology of democratic politics and government’, annual review of sociology 7: 291-319. burstein, paul (2003) ‘the impact of public opinion on public policy: a review and an agenda’ political research quarterly 56 (1): 29-40. kanol ● interest group influence in micro-states 57 carpenter, daniel p., kevin m. esterling and david m. j. lazer (1998) ‘the strength of weak ties in lobbying networks: evidence from health-care politics in the united states’, journal of theoretical politics 10(4): 417-44. civil society organizations directory. (2011). the management centre of the mediterranean and the ngo support centre. url (consulted on 05.07.2012) http://www.csodirectory-cyprus.eu/. cohen, jeffrey e. (1997) presidential responsiveness and public-policy making. ann arbor: the university of michigan press. corbett, jack (2013) ‘everybody knows everybody: practicing politics in the pacific islands’.democratization doi:10.1080/13510347.2013.811233. crombez, christophe (2002) ‘information, lobbying and the legislative process in the european union’, european union politics 3 (1): 7-32. dahl, robert alan and edward r. tufte (1973) size and democracy. stanford: stanford university press. devlet planlama örgütü (2011) kktc nüfus ve konut sayımı. url (consulted 09.03.2014) http://www.devplan.org/nufus-2011/nufus%20son_.pdf. erikson, robert, s., michael b.mackuen and james a.stimson (2002) the macro polity. new york: cambridge university press. fiorina, morris p. (1989) congress: keystone of washington establishment. new haven: yale university press. giugni, marco (2004) social protest and policy change: ecology, antinuclear, and peace movements in comparative perspective. lanham, md: rowman & littlefield. giugni, marco (2007) ‘useless protest? a time-series analysis of the policy outcomes of ecology, antinuclear, and peace movements in the united states, 1977-1995’, mobilization: an international quarterly 12 (1): 53-77. hadenius, axel (1992). democracy and development. cambridge: cambridge university press. haider-markel, donald p. and kenneth j. meier (1996) ‘the politics of gay and lesbian rights: expanding the scope of conflict’, the journal of politics 58 (2): 332-349. haider-markel, donald p. and kenneth j. meier, (2003) ‘legislative victory, electoral uncertainty: explaining outcomes in the battles over lesbian and civil rights’, review of policy research 20 (4): 671-690. hall, richard l. and alan v. deardorff, (2006) ‘lobbying as legislative subsidy’, american political science review 100 (1): 69-84. hall, richard l. and frank w. wayman (1990) ‘buying time: moneyed interests and the mobilization of bias in congressional committees’. american political science review 84 (3): 797-820. heaney, michael t. (2006) ‘brokering health policy: coalitions, parties, and interest group influence’, journal of health politics, policy and law 31(5): 887-944. heitshusen, valerie (2000) ‘interest group lobbying and us house decentralization: linking informational focus to committee hearing appearances’, political research quarterly 53 (1): 151-176. jacobs, david (1974) ‘dependency and vulnerability: an exchange approach to the control of organizations’, administrative science quarterly 19 (1): 45-59. european journal of government and economics 3(1) 58 jacobs, lawrence r. and robert y. shapiro (2000) politicians don’t pander: political manipulation and the loss of democratic responsiveness. chicago: chicago university press. jacobs, lawrence r. and robert y. shapiro (2002) ‘politics and policymaking in the real world: crafted talk and the loss of democratic responsiveness’, in jeff manza, fay lomax cook, and benjamin i. page (eds) navigating public opinion: polls, policy and the future of american democracy. oxford, oxford university press, pp. 54-75. kingdon, john w. (1995) agendas, alternatives, and public policies. new york: harper collins. kitschelt, herbert (1986) ‘political opportunity structures and political protest: antinuclear movements in four democracies’. british journal of political science 16 (1): 57-85. klüver, heike (2013) lobbying in the european union: interest groups, lobbying coalitions, and policy change. oxford: oxford university press. kriesi, hanspeter., ruud koopmans, jan willem duyvendak and marco g. giugni (1995) the politics of new social movements in western europe. a comparative analysis. london: university of minnesota. lax, jeffrey r.and justin h. phillips (2009) ‘gay rights in the states: public opinion and policy responsiveness’, american political science review 103 (3): 367-386. levine, sol and paul e. white (1961) ‘exchange as a conceptual framework for the study of interorganizational relationships’, administrative science quarterly 5 (4): 583-601. lindblom, charles e. (1959) ‘the science of "muddling through"’. public administration review, 9(2): 79-88. mayhew, david r. (1974) congress: the electoral connection. new haven, ct: yale university press. mcveigh, rory, michael r. welch and thoroddur bjarnason (2003) ‘hate crime reporting as a successful social movement outcome’, american sociological review 68 (6): 843-867. meyer, david s. and debra c. minkoff(2004) ‘conceptualizing political opportunity’, social forces 82 (4): 1457-1492. milbrath, lester w. (1963) the washington lobbyists. chicago: rand-macnally. monroe, alan d. (1998) ‘public opinion and policy, 1980-1993’. public opinion quarterly 62 (1): 6-28. ott, dana (2000). small is democratic: an examination of state size and democratic development. new york: garland. page, benjamin i. and robert y. shapiro (1983) ‘effects of public opinion on policy’, american political science review 77 (1): 175-190. pfeffer, jeffrey and gerald r. salancik (2003) the external control of organizations: a resource dependence perspective. stanford: stanford university press. poppelaars, caelesta (2009) steering a course between friends and foes: why bureaucrats interest with interest groups. delft: eburon. potters, jan and frans van winden (1992) ‘lobbying and asymmetric information’, public choice 74 (3): 269-292. kanol ● interest group influence in micro-states 59 reenock, christopher m. and brian j. gerber (2008) ‘political insulation, information exchange, and interest group access to the bureaucracy’, journal of public administration research and theory 18 (3): 415-440. schlozman, kay lehman and john t. tierney (1986) organized interests and american democracy. new york: harper collins. smith, richard a. (1984) ‘advocacy, interpretation, and influence in the u.s. congress’, american political science review 78 (1): 44-63. soule, sarah a. and susan olzak (2004) ‘when do movements matter? the politics of contingency and the equal rights amendment’, american sociological review 69 (4): 473-497. srebrnik, henry (2004) ‘small island nations and democratic values’, world development 32(2): 329-341. statistical service of the republic of cyprus (2011) new publication: demographic report, 2010-2011. url (consulted 09.03.2014): http://www.mof.gov.cy/mof/cystat/statistics.nsf/all/c6ece5795eac1ed8c2257ab 70038fcd4?opendocument&sub=1&sel=1&e=&print. veenendaal, wouter (2013a) ‘democracy in microstates: why smallness does not produce a democratic political system’. democratization doi:10.1080/13510347.2013.820710 veenendaal, wouter (2013b) ‘size and personalistic politics: characteristics of political competion in four microstates’, the round table: the commonwealth journal of international affairs 102 (3): 245-257. veenendaal, wouter (2013c) ‘size and institutional legitimacy: the case of st kitts and nevis’, commonwealth & comparative politics 51 (3): 243-361. wright, john r. (1990) ‘contributions, lobbying and committee voting in the u.s. house of representatives’, the american political science review 84 (2): 417438. microsoft word ejge_03_02_014.doc european journal of government and economics volume 3, number 2 (december 2014) issn: 2254-7088 148 the role of confidence in the evolution of the spanish economy: empirical evidence from an ardl model pablo castellanos-garcía, universidade da coruña, spain indalecio pérez-díaz-del-río, universidade da coruña, spain jose manuel sánchez-santos, universidade da coruña, spain abstract the aim of this paper is to verify the existence and to determine the nature of longterm relationships between economic agents’ confidence, measured by the economic sentiment index (esi), with some of the "fundamentals" of the spanish economy. in particular, by modeling this type of relations, we try to determine whether confidence is a dependent (explained) or independent (explanatory) variable. along with confidence, in our model we incorporate variables such as risk premium of sovereign debt, financial market volatility, unemployment, inflation, public and private debt and the net lending/net borrowing of the economy. for the purpose of obtaining some empirical evidence on the exogenous or endogenous character of the above mentioned variables an ardl (autoregressive-distributed lag) model is formulated. the model is estimated with quarterly data of the spanish economy for the period 1990-2012. our findings suggest that: (a) unemployment is the dependent variable, (b) there is an inverse relationship between esi in spain and unemployment; and (c) the granger causality goes from confidence to unemployment. jel classification c22; d84; e37. keywords confidence; economic sentiment index; ardl models. castellanos et al. ● the role of confidence in the evolution of the spanish economy 149 introduction the so-called confidence crisis, understood as a sharp deterioration in expectations and a dramatic increase in economic uncertainty, has been usually considered as a factor playing a key role in the last global crisis in most of the peripheral economies. one of the main reasons that explain that role has to do with the fact that confidence is a variable that contains qualitative information which directly measures the expectations of economic agents. these expectations, through their impact on decisions of consumption, saving, credit, investment, etc., can exert a significant influence both in the state of the economy and in financial stability. the role of confidence in promoting economic activity has received considerable academic attention by social scientists. among early economic researches addressing the role played by confidence indicators in the real economy and economic fluctuations can be mentioned fuhrer (1993), matsusaka and sbordone (1995), golinelli and parigi (2004), ludvigson (2004) and taylor and mcnabb (2007). some other authors also provide empirical evidence supporting the influence of sentiment of investors on aggregate corporate investments (arif and lee, 2014), stock market developments (otoo, 1999; jansen and nahuis, 2003; lee et al., 2002), market price of risk (verma and soydemir, 2009), yield spreads (ferreira et al. 2008), stock market reaction to monetary policy (kurov, 2010) and even on the probability of future financial distress (mishkin, 1978). confidence indices are statistical tools that are obtained on a regular basis in order to know, through sampling, perceptions of individuals (consumers or businesses) on the state of the economy, their own economic situation and their expectations in the medium and long term.1 for this reason, confidence indices are usually considered as leading indicators that could somehow incorporate "advanced" information in relation with the one facilitated through national accounts and/or macroeconomic data. in this sense, a large number of papers have investigated the predictive capacity of survey data for consumer spending or output. for instance, carroll et al. (1994) and easaw et al. (2005) investigate the predictive power of consumer confidence or sentiment and its usefulness as a forecast of the growth of households consumption. in a similar line, claveria et al. (2007) analyse the possibility of improving the forecasts for some selected macroeconomic variables for the euro area using the information provided by business and consumer surveys. beyond the explanatory and predictive power of confidence indicators, it is important to take into account that the behaviour of this variable in turn might also be conditioned by the evolution of macroeconomic data. some authors as vuchelen (1995) seek for determinants of economic agents’ sentiment in macroeconomic data. particularly, this author focuses on the influence of political events in consumer confidence. ghonghadze and lux (2012) examine the dynamics of european union economic sentiment indicator, providing some evidence about the appropriate modelling of expectations formation. in this sense, as katona (1951, 1975) argues, it should be taken into account that the consumer sentiment contains information that cannot be deduced from economic and financial variables. in this paper our aim is to shed some light on the nature of the relationship between confidence and some "fundamentals" of the spanish economy, namely unemployment, inflation, levels of public and private debt and net lending/net 1 due to the extreme complexity of measuring such a subjective dimension as confidence is, any confidence index has a series of limitations, so there is a wide controversy on its use. in this sense, see for instance dominitz and manski (2004). european journal of government and economics 3(2) 150 borrowing capacity of the economy. in particular, modeling this type of relations, we try to determine whether confidence is a dependent (explained) or independent (explanatory) variable. this is a key issue, mainly because if confidence is an independent variable, the policy maker should focus on the effects of that variable on the macroeconomic performance. once identified those effects, the efficiency of economic policies could be improved considering their influence on the confidence indices. we use as a "proxy" of the confidence the economic sentiment indicator (hereinafter, esi) elaborated by the european commission for the region as a whole, for its constituent economies, and by type of economic activity. this is a composite indicator that reflects and represents judgments and attitudes of producers and consumers. this indicator incorporates information from both the confidence and perception that the different sectors of the economy have about the current economic situation well as its future expectations. particularly, with the purpose of dealing with the role of confidence in the evolution of the spanish economy, we develop an empirical analysis in three steps. first, we try to verify whether confidence is explained by economic fundamentals or whether, by contrast, it should be considered as an exogenous variable. secondly, once identified the existence of a long-term relationship between economic fundamentals and confidence through a cointegration analysis, it is discussed how the dependent variables react to changes in the regressors, both in the short and long term. this discussion is based on the empirical evidence derived from an ardl (autoregressive-distributed lag) model that is estimated with quarterly data of the spanish economy for the period 1990-2012. finally, we establish the direction of the causality between variables through granger causality tests. the rest of the paper is organised as follows. the second section describes in detail the variables and the data set. the empirical methodology and the main results are discussed in the third section. finally, the fourth section summarises the conclusions. variables and data in order to carry out our empirical analysis we consider two subsets of variables. on the one hand, we include qualitative variables proxying the economic agents’ confidence (economic sentiment index), the perception of risk (risk premium of sovereign debt) and the levels of uncertainty (volatility of financial markets). on the other hand, we consider some macroeconomic variables reflecting the spanish economic fundamentals (unemployment, inflation and debt). the sample on which our empirical analysis is based consists of quarterly data for the above-mentioned variables covering from the fourth quarter of 1990 to the second quarter of 2012, which represents a sample size of 87 data. table 1. descriptive statistics and data sources variables max. min. mean sd url pri_deb 226.87 71.32 135.04 58.90 www.bde.es/webbde/es/estadis/infoest/bolest3.html pub_deb 75.95 35.48 54.15 10.14 www.bde.es/webbde/es/estadis/infoest/htmls/cdp.html esi_sp 114.70 73.93 99.74 10.12 ec.europa.eu/economy_finance/db_indicators/surveys/time _series/index_en.htm inf 6.73 -1.07 3.37 1.51 www.bde.es/webbde/es/estadis/infoest/indeco.html unem 24.63 7.95 16.22 5.32 www.bde.es/webbde/es/estadis/infoest/sindi.html spr 5.66 0.01 1.51 1.71 www.bde.es/webbde/es/estadis/infoest/indeco.html bor_cap 2.40 -11.00 -3.18 3.18 www.bde.es/webbde/es/estadis/infoest/indeco.html vix 58.60 11.03 20.49 7.68 www.cboe.com/micro/vix/historical.aspx table 1 shows the detail of the main descriptive statistics of the variables. as we mentioned before, the esi is a composite index published on a monthly basis2 2 the compilation of statistics on the esi is part of the joint harmonised eu programme of business and consumer surveys (bcs). data of this indicator are published monthly by the directorate general for economic and financial affairs, organism which depends on the european commission. castellanos et al. ● the role of confidence in the evolution of the spanish economy 151 made up of five sectorial confidence indicators with different weights: industrial confidence indicator (40%), service confidence indicator (30%), construction confidence indicator (5%), consumer confidence indicator (20%) and retail trade confidence indicator (5%). confidence indicators are seasonally adjusted data obtained from a survey with specific questions for each sector. esi values are derived from qualitative questions about current and future behaviour of economic agents. in this sense, the confidence indicator is calculated for each sector as an arithmetic mean of the responses (seasonally adjusted) to specific questions related to a reference variable whose trajectory is tried to be anticipated (e.g. industrial production in the case of the industrial confidence indicator). the responses from the surveys are added as "balances" which are calculated as the differences between the percentages of respondents who provided positive and negative responses. in numerical terms, the esi is calculated as an index value with an average of 100 and a standard deviation of 10 for a given sampling period. indicators for each of the sectors would be added in the socalled esi. in this study, since all other variables are expressed quarterly, we proceeded to calculate a quarterly value of esi of spain to make the appropriate econometric tests, denoted as esi_sp. the chicago board options exchange market volatility index, also known as vix, is an index that measures the price volatility of the options on the standard & poors 500 index (spx). this index is considered a benchmark for measuring market volatility, collecting in an indirect way the "feeling" of agents operating in it. when volatility is high, the vix reaches a high figure and correlates with falls of s&p 500, pointing out that there is fear and pessimism in the market, while a minimum vix reflects in an atmosphere of optimism and confidence. in sum, the vix is an indicator that provides information on feelings, perceptions or expectations for the future. we incorporate this index in our empirical analysis because it has recently gained acceptance as a summary indicator of global uncertainty or financial stress. in this line, authors such as bloom (2009), show that the vix is strongly correlated with measures of uncertainty, including from financial variables. this lends support to its use as a measure of global financial stress. since this indicator is calculated on a daily basis, it has proceeded to get its correspondent quarterly value. the yield spread of spanish and german government bonds, also known as risk premium or spread (hereinafter, spr) was calculated as the difference between the spanish and german government bonds yield to ten years in domestic markets. theoretically speaking, a high risk premium would show lack of confidence in the possibilities of an economy to pay its public debt. from this perspective, the risk premium can be considered as an extra compensation demanded by investors for holding an asset that involves not only a predictable and measurable risk (specific to the economy in question), but also an uncertainty component difficult to quantify. in this sense, the german long-term debt is used widely as a proxy for the return of a risk-free asset and, consequently, their differential with respect to the bond yields of other eurozone economies is considered as a measure of default risk associated with the sovereign debt of a country. this spread is also obtained on a daily basis and was transformed quarterly for the study using an arithmetic mean. these three variables (the esi_sp, the vix and the spread) have in common (each in its field, with its own characteristics) that they collect qualitative information on the economy in relation to the “sentiment”, the state of opinion or expectations of the economic agents whose decisions determine the economic progress of the country. among the fundamentals of the spanish economy we incorporate to our analysis the following ones: · unemployment rate (unem). the unemployment rate is one of the most important for not only economic but also social situation in a country. traditionally the level or change in the unemployment rate is used as a proxy, european journal of government and economics 3(2) 152 or interpreted as such, for future economic conditions as well as for uncertainty surrounding these conditions. · inflation rate (inf). since the inflation rate used in this paper will be the rate or percentage change in the cpi spain general variable provided monthly, series has been transformed on a quarterly basis in order to allow comparison with other variables. · private debt (pri_deb). households and private nonfinancial corporations debt has been growing steadily over the time period analysed and it could be used as a proxy of the internal solvency of the private sector in a country. the value of the spanish private debt used in this research is the result of adding loans to households and non-profit institutions plus loans to non-financial corporations. · public debt (pub_deb). the value of spanish government debt is the result of adding the debt of the central government, the regional governments, the social security administrations and the local corporations. that is, the debt embodied in cash and deposits, debt securities and loans, in nominal value and consolidated. this series is relativised by the spanish gdp series referenced in the bank of spain for the calculation of public debt following the excessive deficit procedure (edp). · borrowing capacity (bor_cap). it is the sum of the value of the current account balance plus the capital account balance (calculated on a quarterly basis) in proportion to gdp. this ratio has a remarkable economic importance, as it measures in relative terms the borrowing capacity of spain in relation to other countries. empirical methodology and results description of method in an auto-regressive distributed lag (ardl) model, the variable of interest is assumed to be a function of the past values of itself (auto-regressive) and the current and past values of other variables (distributed lag). the basic form of an ardl regression model, denoted ardl (p, q), is: yt = β0 + β1 yt-1 + … + βk yt-p + α0 xt + α1 xt-1 + … + αq xt-q + εt (1) where y is the dependent variable, x is an explanatory variable and εt is a random disturbance term, which we assume is “well-behaved” in the usual sense. the ardl/bounds testing methodology of pesaran et al. (2001) has some advantages over conventional cointegration testing. for instance, it can be used with a mixture of i(0) and i(1) data; it involves just a single-equation setup, making it simple to implement and interpret; different variables can be assigned different lag lengths as they enter the model. the basic steps of this method are the following: (1) check that all the variables are not i(2); (2) estimate an ‘unrestricted’ (also called ‘unconstrained’ or ‘conditional’) error correction model (uecm), which is a particular type of ardl model. for this, we have to determine the appropriate lag structure, applying an adequate criterion (r2, aic, bic, hq) and analysing the behaviour of the results (serially independent errors, dynamic stability…); (3) perform a ‘bounds test’ to look if there is evidence of a long run relationship between the variables.; (4) if the outcome at step 3 is affirmative, estimate a long run “levels model”, as well as a separate “restricted” error correction model (ecm); (5) use the results of the model estimated at step 4 to measure short-run dynamic effects, and the long-run equilibrating relationships among the variables, and (6) analyse the granger causality among variables (optional). castellanos et al. ● the role of confidence in the evolution of the spanish economy 153 unit root tests taking into account that only i(0) and i(1) variables can be included in an ardl model reflecting the relationships between confidence and some of the fundamentals of the spanish economy, unit root tests are performed. in this sense perez et al (2013) provide evidence of the existence of a unit root for variables such as public debt/gdp, private debt/gdp, inflation rate, unemployment rate, borrowing capacity/gdp and spain vs german yield debt spread. in this paper we complete the analysis developing unit root tests to check the order of integration of esi_sp and the vix. according to the results of unit root tests, esi_sp and vix are stationary variables (table 2). table 2. results of unit root tests variable s adf test pp test kpss test tests with breaks additive outlier innovative outlier esi_sp not stationary not stationary stationary stationary stationary outliers: 1q95, 2q09 outliers: 2q93, 2q07 desi_sp stationary stationary stationary no outliers stationary outliers: 2q07, 3q08 vix stationary stationary stationary non-stationary no outliers outlier: 2q08 ho: existence of a unit root except kpss (h1 is the existence of unit root). level of significance = 5%. bic criterion in the adf test. more specifically, augmented dickey-fuller (adf), phillips-perron (pp) and kwiatkowski-phillips-schmidt-shin (kpss) tests results clearly show that the vix is a stationary variable. in relation to esi_sp, although adf and pp tests identify this variable as i(1), unit root with structural breaks tests reveal that i(0) would be its correct order of integration. details of the procedures and tests performed can be obtained at perez et al (2013). pss test results. bound or pesaran, shin and smith (pss) tests are conducted with the aim of establishing whether the variables incorporated in the model have to be considered as exogenous or endogenous. the abovementioned results of unit root tests allow us to claim that all the considered variables meet the requirements for their inclusion in a bound test. these tests are implemented making use of relevant alternative statistics f and t and considering a maximum of four lags (pesaran and pesaran, 1997). the variables considered in the implementation of f and t tests are private debt/gdp (pri_deb), public debt/gdp (pub_deb), esi spain (esi_sp), inflation (inf), unemployment rate (des), risk premium (spr), borrowing capacity/gdp (bor_cap) and vix (vix). pss test results are reported in table 3 considering each of the variables as the dependent variable and the other ones, as regressors. european journal of government and economics 3(2) 154 table 3. results of bound tests or pss dependent variable of the model regressors of the model f test t test pri_deb pub_deb, esi_sp, inf, unem, spr, bor_cap, vix 1.6684 pub_deb pri_deb, esi_sp, inf, unem, spr, bor_cap, vix 4.2775 esi_sp pri_deb, pub_deb, inf, unem, spr, bor_cap, vix 2.3978 -1.9748 inf pri_deb,pub_deb,esi_sp, unem, spr, bor_cap, vix 1.6891 unem pri_deb,pub_deb,esi_sp, inf, spr, bor_cap, vix 4.8347 spr pri_deb,pub_deb,esi_sp, inf, unem, bor_cap, vix 3.6413 -1.9827 bor_cap pri_deb,pub_deb,esi_sp, inf, unem, spr, vix 1.4470 vix pri_deb,pub_deb,esi_sp, inf, unem, spr, bor_cap 5.0427 critical values pesaran et al. (2001) f test t test i(0) i(1) i(0) i(1) 10% 2.035 3.153 -2.57 -4,23 5% 2.365 3.553 -2.86 -4,57 1% 3.027 3.233 -3.46 -5,19 critical values narayan (2005) f test i(0) i(1) 10% 2.129 3.289 5% 2.476 3.746 1% 4.296 4.760 following pesaran et al. (2001) in cases in which f tests are not conclusive we apply a t test. according to the results reported in table 3, in the model in which the esi of spain is included as dependent variable, the null hypothesis of the absence of a cointegration relationship between the variables could not be rejected. therefore, esi_sp should not be considered as dependent variable. in summary, according to f and t tests, the ardl could be applicable in models in which the public debt/gdp, unemployment rate or the vix were dependent variables. estimation of an ardl model generally speaking, an ardl is a general–to-specific approach that allows to include all the variables that a priori may influence the behaviour of a dependent variable and later on it reveals which ones are relevant and which are not. thus it minimises the probability of omitting relevant variables, a problem that would be more serious than to include irrelevant variables (hendry and ericsson, 1991). moreover, it allows for the use of variables without a prior knowledge of their order of integration and, therefore, facilitates the joint inclusion of both i(0) and i(1) variables (sultan, 2010). finally, the formulation of this model will offer also an insight into the short and long term behaviour of the variables, estimating the speed of adjustment of the dependent variable to a long-run equilibrium after a shock in the independent variables. in our particular case, the ardl model is formulated with each of the dependent variables that the bound test indicated as significant for the case of four lags: unemployment, public debt/gdp ratio and vix. the formulation of the ardl when the public debt/gdp is incorporated as dependent variable shows problems of serial autocorrelation, lack of significance and instability of long term coefficients. the specification with the vix as dependent variable reveals serial autocorrelation, incorrect functional form, non normality and heteroscedasticity in the residuals which lead us to reject this formulation. as can be seen in table 4, the choice of an ardl model with the unemployment rate as the dependent variable does not reveal problems of serial correlation, misspecification, normality or heteroscedasticity in any of the tests performed with the four criteria. castellanos et al. ● the role of confidence in the evolution of the spanish economy 155 table 4. diagnostic tests considering unemployment rate as dependent variable statistical test r2 selection criterion lm version f version lagrange test (serial correlation) chsq(4) = 10.7435 [0.030] f(4, 56) = 2.0816 [0.095] reset test (functional form) chsq(1) = 1.2180 [0.270] f(1, 59) = 0.8787 [0.352] normality test chsq(2) = 1.3382 [0.512] not applicable heteroscedasticity test chsq(1) = 3.1514 [0.076] f(1, 81) = 3.1968 [0.078] aic selection criterion lm version f version lagrange test chsq(4) = 8.7208 [0.068] f(4. 59) = 1.7317 [0.155] reset test chsq(1) = 2.3370 [0.126] f(1. 62) = 1.7963 [0.185] normality test chsq(2) = 0.3591 [0.836] not applicable heteroscedasticity test chsq(1) = 1.8399 [0.175] f(1. 81) = 1.8363 [0.179] bic selection criterion lm version f version lagrange test chsq(4) = 3.6464 [0.456] f(4. 64) = 0.7352 [0.571] reset test chsq(1) = 2.7220 [0.099] f(1. 67) = 2.2717 [0.136] normality test chsq(2) = 0.2876 [0.866] not applicable heteroscedasticity test chsq(1) = 0.0455 [0.831] f(1. 81) = 0.0444 [0.834] hq selection criterion lm version f version lagrange test chsq(4) = 8.7208 [0.068] f(4. 59) = 1.7317 [0.155] reset test chsq(1) = 2.3370 [0.126] f(1. 62) = 1.7963 [0.185] normality test chsq(2) = 0.3591 [0.836] not applicable heteroscedasticity test chsq(1) = 1.8399 [0.175] f(1. 81) = 1.8363 [0.179] additionally, cusum and cusumq stability tests did not reflect instability of long-run coefficients of the variables. thus, after the completion of these tests, the general ardl model would be established as follows: tit i i i iti i iti it i iit i iit i i it i iit i it unemvixcapbor sprinfspesi debpubdebiunem                         4 1 4 0 4 0 4 0 4 0 4 0 4 0 4 0 0 _ _ __.pr (2) where all the variables have been previously defined. t corresponds to the random disturbance, 0 is the intercept of the equation and i i, i, i, i,i, i and i correspond to the coefficients of the regressors (including lags of the dependent variable). the corresponding ardl models are selected taking into account four criteria, namely, the r2, the akaike information criterion (aic), the schwarz bayesian information criterion (bic) and the hannan-quinn information criterion (hq), and setting a maximum of 4 lags as previously indicated. thus, as it can be seen in table 5, the number of lags to be considered is identical in all four criteria, except for pri_deb (mostly two lags), esi_sp (mostly four lags), bor_cap (mostly 3 lags) and vix (mostly no lags) where some small differences were detected. likewise, we emphasise that the results obtained under the hq criterion are equal to those obtained in the aic, so only the results of the first three criteria (r2, aic and bic) will be specified hereinafter. european journal of government and economics 3(2) 156 table 5. number of lags to be considered for each criterion and for each variable nº lags unem pri_deb pub_deb esi_sp inf spr bor_cap vix r2 criterion 3 3 0 4 0 0 3 2 aic 3 2 0 4 0 0 3 0 bic 3 2 0 0 0 0 2 0 hq criterion 3 2 0 4 0 0 3 0 long and short term models if two of more variables are cointegrated, two types of models can be estimated. first, an ols regression model using the levels of the data. this model will provide the long-run equilibrating relationship between the variables. second, and errorcorrection model (ecm), estimated by ols too, but using the differences of the data. this model will represent the short-run dynamics of the relationships between the variables. once established the existence of a long-term relationship between the variables through the bound test, the ardl model allows us to analyse relationships in both the long and short term. long-term model the coefficients of the long-term model obtained for each of the estimations are shown in table 6: table 6. ardl coefficients in the long term model variable r2 criterion aic bic pri_deb -0.0537*** (-4.7174) -0.5621*** (-4.1903) -0.0422 (-1.1052) pub_deb 0.3316*** (5.2437) 0.3233*** (4.3905) 0.1139 (0.4779) esi_sp -0.5347*** (-5.8873) -0.5976*** (-5.3191) -1.1600*** (-1.9309) inf -0.8710** (-3.3095) -0.8841** (-2.9335) -1.6693 (-1.5124) spr 0.4548 (1.2591) 0.5723 (1.3861) 0.9490 (0.8149) bor_cap -0.9932** (-3.2426) -1.0212** (-2.8412) -1.0300 (-0.9740) vix 0.0191 (0.3307) 0.0805 (1.4517) 0.2818 (1.2228) c 59.4954*** (6.2485) 65.1116*** (5.5120) 132.1281** (2.0086) * indicates significance at 10%, ** at 5% and *** at 1%. in brackets, values of the t statistic the signs of the numbers in table 6 show the kind of long term relationship that each of the regressors has with the dependent variable, i.e., the unemployment rate (unem). from their analysis we can draw the following preliminary evidences previously to the results of the causality tests that will be commented later. -the only variable that is significant in all four criteria is the confidence measured through the economic sentiment indicator in spain. in this sense we could confirm that increases in the confidence are associated with decreases in the level of unemployment in this country. -neither risk premium nor vix volatility are significant variables in any of the criteria considered, that is, neither the yield of spanish debt versus german debt nor vix should be considered as regressors when determining long-term unemployment in spain. short term model an error-correction model (ecm) was formulated in order to determine the shortterm behaviour of the variables included in our analysis. the formulation of the castellanos et al. ● the role of confidence in the evolution of the spanish economy 157 ecm, allows us to analyse the impact of a "shock" on the variables of the model as well as the adjustment time needed to return to a situation of equilibrium. this model can be expressed as follows: tit i i i iti i iti it i iit i iit i i it i iit i itt unemvixcapbor sprinfspesi debpubdebieunem                          4 1 4 0 4 0 4 0 4 0 4 0 4 0 4 0 10 _ _ __.pr (3) where et-1 is the error correction lagged one period, t corresponds to the random disturbance, 0 is the intercept of the equation and i i, i, i, i,i, i and i are the coefficients of the regressors (including lags of the dependent variable). first of all, in order to use the results of the estimation, we have to check both the significance of the error correction lagged one period, et-1, and its sign, which has to be negative (see table 7). note that the coefficient of the lagged residual in the ecm shows the speed of adjustment towards the equilibrium following a shock to the syste. table 7. ardl coefficients in the short term model (ecm) variable r2 criterion aic bic dpri_deb -0.0210 (-0.9495) -0.0292 (-1.3704) -0.0108 (-0.4904) dpub_deb 0.0656*** (3.4323) 0.0548** (3.0104) 0.0064 (0.4251) desi_sp -0.0606** (-2.5609) -0.0689** (-2.9709) -0.0654*** (-6.006) dinf -0.1723** (-3.0494) -0.1499** (-2.8043) -0.941*** (-1.6890) dspr 0.0900 (1.1829) 0.0971 (1.2727) 0.0535 (0.7395) dbor_cap -0.1077** (-3.1386) -0.1146*** (-3.3466) -0.1010** (-2.814) dvix 0.0119 (1.2822) 0.0136 (1.6175) 0.0159** (2.0143) dc 11.7676*** (7.4258) 11.0430*** (7.2127) 7.4484*** (6.2846) et-1 -0.1978*** (-4.7096) -0.1696*** (-4.3203) -0.0564*** (-1.9163) * indicates significance at 10%, ** at 5% and *** at 1% level respectively. in brackets, values of the t statistic as regards the qualitative variables on which we focus, the estimated coefficients of the ecm show the following results. the ecm is valid as the error correction term is significant and negative in the three criteria. in the short term, increases in confidence measured through esi_sp have a positive relationship with the decrease in the unemployment rate in spain, according to all criteria. as in the long term, spread has no significant result with any of the criteria, in the short term. the vix index only appears as significant according to the bic criterion, showing in this case a positive relationship with the unemployment rate. to sum up, the following aspects can be highlighted in both the short and long term. the variable esi_sp used as a proxy of confidence is significant in both the short and long term in a model with the unemployment rate as the dependent variable. both variables are inversely related, that is, increases in business confidence and economic sentiment will be associated with decreases in the spanish unemployment rate and vice versa. the vix index is only significant in the short term and with the bayesian information criterion, showing in this case a positive relationship with the unemployment rate. the interest rate spread is not a european journal of government and economics 3(2) 158 significant variable neither in the short nor in the long run with any of the criteria used. for the shake of simplicity, table 7 really does not include all the estimates of the short-run models, we only included the first term of the lag polynomial of each variable and the error correction term. confidence-unemployment: results of the causality test taking into account that the results of the cointegration tests (which check if there is a long-term relationship between the variables) are not sufficient and at the same time they are necessary to determine the existence of causality in the granger sense, we proceeded to perform causality tests applying toda and yamamoto (1995) procedure to test for granger causality – this procedure is applicable in situations where not all the variables are stationary, as it is the case here. the information criteria suggest that we should have a maximum lag length of 3 for each variable. the application of the lm test for serial independence against the alternative of ar(k)/ma(k), for k = 1, …, 12 shows that there is no problem using 3 lags (serial correlation is removed). johansen’s trace test and maximum eigenvalue test both indicate the presence of cointegration between the two series, at the 5% level. as m (maximum order of integration of these two variables) is 1, we re-estimated the var levels with one extra lag of each variable in each equation and we undertook the granger non-causality testing. according to the results of this procedure, we cannot reject the null hypothesis of no causality from unem to esi_sp. furthermore, the null hypothesis of no causality from esi_sp to unem can also be rejected at the 1% significance level. thus, a causal link between the two variables is found, that is, increases in the confidence in the economy causes decreases in the level of unemployment in a country and not vice versa. the confidence measured by the esi seems to cause reductions in the unemployment rate: i.e. a qualitative magnitude linked to expectations as mentioned earlier in this paper, would play a key role in the activation of the labour market. this result would come to endorse the idea that the degree of confidence or "feeling" about the economic situation which the different economic actors have is fundamental in the evolution of unemployment. therefore, in a situation of increased confidence in the present and future state of the economy, more employees would be hired, whereas worse expectations on the economic situation in the future would reduce employments (increasing the unemployment accordingly). although this study does not conclude that the confidence is the only or the most important variable to consider in a model explaining the evolution of unemployment in the spanish economy, the results obtained show that the influence of "qualitative" factors such as confidence directly impacts on the economic sentiment and causes a decrease in the level of unemployment. 4. conclusions the relationship between economic agents’ confidence and the real economy is a subject of great interest to policymakers. on the one hand, confidence, insofar as it affects economic choices of consumption, saving, investment, etc., exerts an influence on both the current state of the economy and on the perspectives of economic growth. at the same time, the behaviour of confidence indices may also be conditioned by the evolution of certain economic and financial indicators. in this paper we investigate the role of confidence in the evolution of the spanish economy. particularly, we try to determine whether confidence should be considered as a explained or a explanatory variable. with this purpose we have incorporated to our analysis some of the fundamental variables of the spanish economy (unemployment, inflation, private debt/gdp, public debt/gdp, and castellanos et al. ● the role of confidence in the evolution of the spanish economy 159 borrowing capacity/gdp) along with other variables that contain qualitative and prospective information measuring the expectations of the economic agents (confidence, risk premium and market volatility). with the aim of determining the existence of long-term relationships between confidence and some of the fundamentals of the spanish economy and of exploring the nature of these relationships, we have developed a three step process. first, the application of a bound test for studying the relationship between the variables considered in the analysis enables us to identify unemployment as the dependent variable in a cointegration model that includes the aforementioned variables. second, we have specified an ardl, which incorporates both lagged independent variables and the lagged dependent variable (the unemployment rate) as regressors. among the most relevant results shown by the estimated ardl model it can be emphasised that the yield spread of spanish public debt (risk premium) has no significant impact on the unemployment rate. this model also reveals that the esi of spain is a variable that shows an inverse relationship with the level of unemployment, that is, esi_sp increments are closely associated with long-term unemployment decreases and vice versa. additionally, the results obtained from the mce show that increases in the confidence measured by the esi_sp have a positive relationship with the decrease in the unemployment rate in spain also in the short term. finally, since the existence of a long-term relationship is a necessary but not sufficient condition for the existence of causality, we also proceeded to analyse the relationship between unemployment and confidence through a test of granger. this test shows unidirectional granger causality from esi_sp towards the unemployment rate for the case of four lags. these results lead us to consider the confidence as a variable with a long-term relationship and that causes unemployment in the granger sense. this is a result that does not fully explain the behaviour of the unemployment rate variable (whose evolution is also due to other factors), but it would emphasise the importance of a qualitative variable such as the confidence of economic agents in relation to job creation. furthermore, this finding corroborates the importance of the role played by confidence in the spanish economy. on the one hand, the existence of this causality justifies the need of including the confidence in macroeconomic models that seek to explain the evolution of unemployment in spain. on the other hand, policy implications also emerge from the results obtained, which highlight the importance of implementing economic policies that encourage confidence among economic agents if the ultimate goal is to reduce unemployment. thus, empirical evidence reveals that one of the criteria to be considered when evaluating economic policy decisions would be their influence on confidence. references arif, salman and charles m.c. lee (2014) `aggregate investment and investor sentiment´, the review of financial studies 27(11): 3241-3279. bloom, nicholas (2009) ‘the impact of uncertainty shocks’, econometrica 77(3): 623–68 carroll, christopher d., jeffrey c fuhrer and david w. wilcox, (1994) `does consumer sentiment forecast household spending? if so, why?´ american economic review 84 (5): 1397-1408. chicago board options exchange (2014) introduction to vix options and futures, url (consulted nov. 2014): http://www.cboe.com/micro/vix/vixintro.aspx claveria, oscar, ernest pons and raúl ramos (2007) `business and consumer expectations and macroeconomic forecasts´, international journal of forecasting 23(1): 47-69. european journal of government and economics 3(2) 160 dominitz, jeff and charles f. manski (2004) ‘how should we measure consumer confidence?’, journal of economic perspectives 18(2): 51-66. easaw, joshy z., dean garratt and saeed m. heravi, (2005) `does consumer sentiment accurately forecast uk household consumption? are there any comparisons to be made with the us?´, journal macroeconomics 27(3): 517-532. european commision (2004) `the joint harmonised eu programme of business and consumer surveys. user guide´. directorate-general for economic and financial affairs, brussels, url (consulted nov.2014): http://ec.europa.eu/economy_finance/db_indicators/surveys/documents/bcs_user_ guide_en.pdf european council (2009) council regulation (ec) no 479/2009, official journal of the european union, url (consulted nov. 2014): http://epp.eurostat.ec.europa.eu/cache/ity_sdds/annexes/ei_naga_a_esms_an2 .pdf ferreira, eva, maria i. martínez, eliseo navarro and gonzalo rubio (2008) `economic sentiment and yield spreads in europe´, european financial management 14(2): 206-221. fuhrer, jeffrey c. (1993) `what role does consumer sentiment play in the u.s. economy?´ federal reserve bank of boston new england economic review, january-february: 32-44. ghonghadze, jaba and thomas lux (2012) `modelling the dynamics of eu economic sentiment indicators: an interaction-based approach´, applied economics 44(24): 205-3088. golinelli, roberto and giuseppe parigi (2004) `consumer sentiment and economic activity. a cross country comparison´, journal of business cycle measurement and analysis 10(2): 147-170. granger, clive.w.j. (1969) `investigating causal relationships by econometric models and cross spectral methods´, econometrica 37(3): 425-435. hendry david f. and neil r. ericsson (1991) `modeling the demand for narrow money in the united kingdom and the united states´, european economic review 35(4): 833-866. jansen, jos w. and niek j. nahuis. (2003) `the stock market and consumer confidence: european evidence´, economics letters 79(1): 89-98. katona, george (1951) psychological analysis of economic behavior. new york: mcgraw-hill (reprinted 1977). katona, george (1975) psychological economics. amsterdam: elsevier. kurov, alexander (2010) `investor sentiment and the stock market’s reaction to monetary policy´, journal of banking & finance 24(1): 139-149. lee, wayne y., christine x. jiang and daniel c. indro (2002) `stock market volatility, excess returns, and the role of investor sentiment´, journal of banking & finance 26(12): 2277-2299. ludvigson, sydney c. (2004) `consumer confidence and consumer spending´, the journal of economic perspectives, 18(2): 29-50. matsusaka, john g. and argia m. sbordone (1995) `consumer confidence and economic fluctuations´ economic inquiry 33 (2): 296-318 mishkin, frederic s. (1978) `consumer sentiment and spending on durable goods´, brookings papers on economic activity 1: 217-232. narayan, p.k. (2005) `the saving and investment nexus for china: evidence from cointegration tests´, applied economics 37(17): 1979-1990. castellanos et al. ● the role of confidence in the evolution of the spanish economy 161 otoo, maría (1999) `consumer sentiment and the stock market´, board of governors of the federal reserve system, finance and economics discussion series 60. perez, indalecio, pablo castellanos and josé m. sanchez-santos (2013) `risk premium as an economic policy objective: the spanish case´, economics and business letters, 2(3): 94-104. pesaran, m. hashem and bijan pesaran (1997) working with microfit 4.0: interactive econometric analysis. oxford university press. pesaran, m. hashem, yongcheol shin and richard.j. smith (2001) `bounds testing approaches to the analysis of level relationships´, journal of applied economics 16(3): 289-326. sultan, riad (2010) `short-run and long-run elasticities of gasoline demand in mauritius: an ardl bounds test approach´, journal of emerging trends in economics and managements sciences 2(1): 90-95. taylor, karl and robert mcnabb (2007) `business cycles and the role of confidence: evidence for europe´, oxford bulletin of economics and statistics 69(2): 185–208. toda, hiro y. and taku yamamoto (1995) ‘statistical inference in vector autoregressions with possibly integrated processes’, journal of econometrics 66(12): 225-250. verma, rahul and gökçe soydemir (2009) `the impact of individual and institutional investor sentiment on the market price of risk´, the quarterly review of economics and finance 49(3): 1129-1145. vuchelen, jef (1995) `political events and consumer confidence in belgium´, journal of economic psychology 16(4): 563-579. microsoft word ejge_04_2015-011.docx european journal of government and economics volume 4, number 2 (december 2015) issn: 2254-7088 118 bed capacity and surgical waiting lists: a simulation analysis manel antelo, universidade de santiago de compostela, spain francisco reyes santias, universidade de vigo, spain adela martinez calvo, universidade de santiago de compostela, spain abstract waiting time for elective surgery is a key problem in the current medical world. this paper aims to reproduce, by a monte carlo simulation model, the relationship between hospital capacity, inpatient activity, and surgery waiting list size in teaching hospitals. inpatient activity is simulated by fitting a normal distribution to real inpatient activity data, and the effect of the number of beds on inpatient activity is modelled with a linear regression model. analysis is performed with data of the university multi-hospital complex of santiago de compostela (santiago de compostela, spain), by considering two scenarios regarding the elastiticity of demand with bed increase. if demand does not grow with an increase on bed capacity, small changes lead to drastic reductions in the waiting lists. however, if demand grows as bed capacity does, adding additional capacity merely makes waiting lists worse. jel classification i10; i12. keywords hospital beds; inpatient activity; length of stay; monte carlo simulation; surgery waiting lists. acknowledgements the authors thank two anonymous reviewers for useful comments and suggestions. they also acknowledge discussions with seminar participants at xv applied economics meeting held in a coruña (spain) on june 7-8, 2012. ma acknowledges funding from the galician regional government (xunta de galicia) through grant gpc 2013-045. the work of am-c was supported by the spanish ministry of science and innovation (project mtm2008-03010), the galician regional government (pgidit07pxib207031pr), and the spanish ministry of economy and industry (10mds207015pr). european journal of government and economics 4(2) 119 1. introduction a waiting list for healthcare is a queue of patients who have been given a care procedure but, due to reasons beyond their control, must wait to be served within a variable time period (sampietro and espallargues, 2001). though waiting lists are common in different healthcare fields (both primary and specialty care), healthcare levels (outpatient and hospital care), and therapeutic procedures (surgical and nonsurgical), diagnostic and rehabilitative (churruca, 2000), surgical waiting lists are those commonly discussed. these are often studied due to their major problems of morbidity and mortality, as well as their greater economic impact (richards, 1999; silber et al., 1996). with the aim of assessing the influence of waiting lists on survival rates, richards (1999) reviewed 87 studies published in different countries, and showed that survival of patients with over a three-month delay in surgical treatment of breast cancer was reduced by 10 per cent at 5 years when compared to patients who benefited from earlier operations. in the same line, silber et al. (1996) examined waiting list-related risk, and pointed out that the risk of mortality in patients awaiting coronary by-pass was 1.3 per cent per month, and that by-pass should be performed within the first week after coronary angiography diagnosis to minimize death risks in the waiting list. elective surgery in spain works through waiting lists to such an extent that a public opinion barometer identified waiting time for elective surgery as the leading source of public dissatisfaction with inpatient services. one of the main reasons behind waiting lists for elective surgery is the fact that spanish hospitals face serious problems of productive capacity (siciliani and hurst, 2003). although productive capacity usually refers to resources such as staff, beds, operating theatres, and community-based health centers, just to name a few, the two main resources in hospital production function are personnel and number of beds (often considered a rough proxy for capital endowment).1 this paper is focused on analysing the situation in one of these hospitals, namely the university multi-hospital complex of santiago de compostela2 (santiago de compostela, spain) specifically, we examine the impact of bed capacity—a scarce and expensive input in healthcare—on daily inpatient activity, patients’ length of stay and, consequently, waiting lists in a teaching hospital like the above mentioned. the high cost of academic health centres and other teaching hospitals is largely attributed to the unique missions pursued by these institutions—including graduate medical education, biomedical research, and the maintenance of standby capacity for highly specialized patients (commonwealth found, 1997). medicare is often viewed as the primary payer of graduate medical education in view of its size and use of explicit payment adjustments for teaching hospitals (koening et al., 2003). however, experimentation with the real system would cause a lot of trouble both for patients and staff. for that reason, we are obliged to perform a simulation approach, which represents the system and can be manipulated with no daily healthcare practice disruptions. indeed, one of the often-mentioned reasons for using simulation as a tool is the experimentation with non-existing systems (law and kelton, 1991). once validated, the simulated model can yield accurate estimates of the behaviour of the real system and help to understand and clarify complex dynamic processes (yamaguchi et al., 1994). finally, simulating a process such as admission to elective surgery can also help to identify bottle-neck and congestion points. 1 the common wisdom that the lack of hospital resources other than beds (equipment, for example) may determine waiting lists rather than beds themselves is considered here, by assuming a direct relationship between beds and equipment in such a way that beds represent roughly the amount of capital input of the hospital. 2 throughout the paper, we will maintain the galician name of “complexo hospitalario universitario de santiago de compostela” and its chus acronym. antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 120 besides, the simulation model can be useful to monitor hospital system performance and assess the relative effectiveness of alternative policies aimed at coping with historical or statistically-generated patient load. in sum, simulation is a recommended tool to solve the problems created by complex systems where mathematical models are not operational. for this reason simulation is widely used to analyse hospital problems, as these problems are considered a complex system comprising many variables and different random events. for example, in surgical services, everett (2002) developed a decision support tool to evaluate various policies on surgey waiting lists and bed occupancy. akkerman and knip (2004) used simulation to allocate beds to cardiac surgery in order to reduce waiting times. denton et al. (2007) applied simulation to examine optimal timing of surgery.3 likewise, vanberkel and blake (2007) built a discrete event simulation model to evaluate surgical waiting times and support capacity-planning decisions. although the optimization approach is different in each case, it is aimed at improving all services through optimal resource use. to examine the pattern of waiting lists size in programmed surgery and reproduce the behaviour of daily inpatient activity, the length of stay and, consequently, waiting lists, a known distribution was fit to each variable. such distribution then allows us to generate new values for daily inpatient activity and patient length of stay by means of the monte carlo method. finally, after generating new observations of inpatient activity and length of stay, the corresponding simulated waiting list and daily percentage of occupied beds (occupancy rate) can also be created. in this context, waiting list variations with increased number of hospital beds (a rough proxy for capital input) can also be studied by examining the impact of the number of beds on inpatient activity, length of stay and, consequently, waiting lists. with this purpose, the simulation process was replicated for various increased percentages in the number of available beds in two alternative scenarios. first, the increased number of beds is assumed to lead to no modification of the inpatient activity pattern (no beds effect). then, alternatively, a change in the number of beds was assumed to modify the behaviour of inpatient activity in the amount given by the inpatient activity-beds elasticity estimated by kroneman and siegers (2004) in a comparative study for 10 european countries. all computational programming was performed with free statistical software r. the current research provides two main results. firstly, if there is no beds effect, i.e., if the number of beds affects the admission pattern, even remarkable extensions in the mumber of beds lead to no significant difference in hospital occupancy rates or waiting lists. secondly, if the elasticity of the beds-inpatient activity relationship, which is over 1 (kroneman and siegers, 2004), is taken into account, then no significant differences in terms of occupancy rates and surgery waiting lists are observed, as the potential number of beds in the hospital increases. in other words, when elasticity is included in the simulation model, waiting lists are found not to drop when the hospital disposes of a higher amount of physical capital (beds) in its production function. our findings suggest that, under certain conditions, the number of hospital beds is the driver of waiting lists in galician hospitals like chus. this, obviously, has a key consequences at both policy and managerial levels: hospital configurations (evaluation of bed usage policies, creation of new services, revision of elective admissions…) allocation plans, new view of measures of performance, etc. 3 in spite of the fact that the use of simulation models in healthcare is not new, a survey on 200 healthcare simulation models proves that their results were implemented in only 16 cases, showing limited acceptance (wilson, 1981).the primary reason for the healthcare industry reluctance to accept simulation was managerial reluctance to reduce complex healthcare processes to a model representation (alvarez and centeno, 1999). european journal of government and economics 4(2) 121 finally, it is worth to note that our results are close to those of kroneman and siegers (2004) and zeraati et al. (2005), among others. in their analysis of the effect of hospital beds on their use for 10 european countries, kroneman and siegers (2004) found that admission rates seem sensitive to bed supply, and show a positive and over-1 elasticity (in particular, 1.44). this means that admission rates are higher in those countries whose hospitals have a high number of beds, but the amount of hospital beds itself seems to have no significant impact on average length of stay. in the same line, zeraati et al. (2005) suggested that an increase in the number of beds tends to generate additional healthcare demand, either in the form of higher patient admission, longer-treated patients, or some combination of both. this fact reflects the so-called roemer’s law of demand (shain and roemer, 1959; roemer, 1961), indicating that a sudden increase in in the amount of hospital beds in a given country, with no changes in other factors, leads to a sharp increase in usage rates. the rest of the paper is laid out in four sections. section 2 outlines the model, while simulation analysis and the main results of the paper are presented in section 3. the results achieved are discussed in section 4. finally, section 5 concludes the paper. 2. the model 2.1. data on inpatient activity and length of stay our approach is applied to the university multi-hospital complex of santiago de compostela (chus), a teaching hospital-complex located in the city of santiago de compostela (nw spain). the chus is composed of three hospitals with 1,100 beds, 29 surgical theatres and 705 full-time physician equivalents. in 2008 the number of chus-treated inpatients was 34,953 vs. 352,930 inpatient days that correspond to 9,525 first-time visits and 149,406 emergencies.4 our working variables are daily inpatient activity and length of stay for all patients admitted in chus in 2007. inpatient or residential activity is the number of hospital patients who, daily, receives lodging and food as well as treatment divided by the number of beds, the length of stay, on the other hand, refers the number of days that a patient is treated in hospital, including readmissions. to fit an inpatient activity pattern from real data and select adequate fitting distribution, data were split into two groups: working days (i.e. monday to friday, except july, august, and december), and holidays (i.e. saturdays and sundays plus july, august and december). the reason behind this splitting is that, as plotted in figure 1, clear differences in inpatient activity values were observed between working days and holidays. 4 consellería de sanidade, xunta de galicia (2008), memoria 2007 sistema público de saúde de galicia. available at www.sergas.es/publicaciones/detallepublicacion.aspx?idpaxina=40008&idcatalogo=1732. antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 122 figure 1. box-plot (left panel) and histogram (right panel) of daily inpatient activity. in fact, quartiles are 96.5 (first quartile), 116 (median) and 126 days (third quartile) for working days, while they are 53 (first quartile), 80.5 (median) and 94 days (third quartile) for holidays. a two-sample kolmogorov-smirnov test applied to data shows that the statistic value is 0.5667 and the associated p-value is below 0.05, so the null hypothesis that inpatient activity for working days and holidays is drawn from the same distribution of probability can be rejected. hence, we decided to analyze inpatient activity in working days and holidays separately. both the box-plot and the histogram of patient length of stay are shown in figure 2. here, most data can be observed to correspond to short stays, as the first quartile is 3 days, the median is 6 days, and the third quartile is 12 days, although some outliers correspond to longer hospital stays. figure 2. box-plot (left panel) and histogram (right panel) of length of stay. 2.2. the monte carlo simulation method the monte carlo simulation method allows generating new values for daily inpatient activity and patient length of stay. for the former, data were considered independent observations of a continuous variable and then fitted a normal distribution for working days holidays 4 0 6 0 8 0 1 0 0 1 2 0 1 4 0 working days 40 60 80 100 140 0 .0 0 0 0 .0 0 5 0 .0 1 0 0 .0 1 5 0 .0 2 0 holidays 40 60 80 100 140 0 .0 0 0 0 .0 0 5 0 .0 1 0 0 .0 1 5 0 1 0 0 2 0 0 3 0 0 0 100 200 300 0 .0 0 0 .0 1 0 .0 2 0 .0 3 0 .0 4 european journal of government and economics 4(2) 123 working days and holidays separately.5 in other words, the two parameters (the mean, , and the standard deviation, ) that determine a normal distribution for working days and holidays were determined. after the fitting procedure, the normal distribution )45.20ˆ,02.112ˆ(  wwn  , (1) was selected for inpatient activity in a working day (denoted by subscript w), and the normal distribution )08.24ˆ,72.76ˆ(  hhn  , (2) was chosen for inpatient activity in a holiday (denoted by subscript h). to decide whether the behaviour of the inpatient activity variable comes from a normal distribution, a kolmogorov-smirnov test was completed. this offered 0.0972 as the value of the statistics for working days (p-value=0.0503), and 0.1047 as the value of the statistics for holidays (p-value=0.0481). data are then accepted to be normally distributed for significance level 01.0 . regarding the length of hospital stay, a normal and other distributions were fitted like a poisson distribution, but none of them rendered good fitting. a possible cause could be the presence of outliers corresponding to large stays, which leads the distributions to not have heavy enough tails to fit real data adequately. hence, an alternative approach based on the kernel density estimator was chosen, for which the length-of-stay variable is implicitly assumed to be continuous. denote the real observations of the length-of-stay variable in 2007 by sample ),...,,( 21 nxxx drawn for some distribution with an unknown density f, where n represents the sample size. since we are interested in estimating the shape of the density function of the length-of-stay variable, f can be estimated by the following kernel density estimator (wand and jones, 1995)             n i i n i ih h xx k nh xxk n xf 11 11 )(ˆ , (3) where k is a kernel (i.e. a non-negative real-valued integrable function, satisfying the following two requirements:     1)( dxxk (4) and )()( xkxk  , for all values of x.6 (5) on the other hand, h is a positive number called bandwidth related to the obtained estimator's smoothness. specifically, the standard normal density function was adopted as the kernel function, and the bandwidth related to the obtained estimator’s smoothness was estimated as 74.0ˆ h . 5 although we have tried to fit distributions other than normal distribution, the normal distribution rendered the best fit. 6 the first requirement ensures that the kernel density estimation method results in a probability density function, and the second one ensures that the mean of the corresponding distribution equals that of the sample used. k is usually a unimodal symmetric probability density function. antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 124 2.3. generating new observations once the distributions for daily inpatient activity and patient length of stay were chosen, data were generated for both variables. on one hand, a new observation for inpatient activity is drawn from the normal distribution stated in (1) on a working day, and from the normal distribution postulated in (2) on a holiday. on the other hand, the new observations for patient length of stay were obtained from the kernel estimator given in (3). this was carried out by means of a threestage method. in the first stage, the real length of stay in 2007 was denoted as (x1, x2,..., xn), and an element x from dataset },...,,{ 21 nxxx was randomly selected. in the second stage, z was generated from the standard normal distribution )1,0(nz  . finally, in the third stage, a new length of stay, x*, was built by using the rule )ˆ(* zhxroundx  , (6) where ĥ stands for the bandwidth estimated for the kernel density estimator (3), and )(round rounds the simulated values x* to zero decimal places. 3. simulating inpatient activity and patient length of stay the activity of the chus can now be simulated for 2007, when the total number of beds (roughly speaking, the amount of physical capital) was 1,100. to avoid starting with an “empty” system (i.e. a hospital without patients), the simulation process begins on august 1st, 2006 (i.e. before the period of analysis).7 then, for each day of 2007, the simulation process followed a three-stage procedure. in the first stage, an inpatient activity value was generated taking into account both new and waitinglist patients. in the second stage, we detect the number of free beds in the hospital and we decide to occupy them with patients for which generate the length of their stays. finally, in the third stage, if daily inpatient activity exceeds the number of available beds, the remaining patients are included in the waiting list. the results of this 3-step model are depicted in figure 3, which plots the hospital's actual inpatient activity in 2007 (black line), and the 500 runs of model-simulated inpatient activity (grey lines). 7 in the simulation literature, this is called the start-up problem (law, 1983). european journal of government and economics 4(2) 125 figure 3. actual inpatient activity (solid black line) and simulated inpatient activity (grey lines) in 2007. on the other hand, figures 4 and 5 depict, respectively, the 500 replications of simulated daily waiting lists and occupancy rate. antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 126 figure 4. simulated daily waiting list (grey lines) in 2007 waiting list quartiles are highlighted: 25 per cent (dashed black line), 50 per cent (solid black line) and 75 per cent (dotted black line). figure 5. simulated daily occupancy rate (grey lines) in 2007 occupancy rate quartiles are highlighted: 25 per cent (dashed black line), 50 per cent (solid black line), and 75 per cent (dotted black line). the simulated values can be concluded to fit well the real-data pattern for this period. besides, the hospital waiting list grows until june, disappears in july and august, and rises again until december. there are then two critical moments before summer and christmas with very high peaks. a similar pattern is observed for daily occupancy rate. the results obtained for the simulated daily waiting list data and the simulated daily occupancy rate can be summarized as follows. thirty-eight days out of a hundred days a year show no waiting list, while 62 days show a variable-size surgical waiting list. this size ranges from 1 to 100 patients (21 per 100 days), from 100 to 200 patients (14 per cent of the analysed period), from 200 to 350 patients (13 per cent of the time considered), and over 350 patients (14 per cent of the time). the differences observed in waiting list time and size would be explained by the seasonal trend observed in hospital admissions. for instance, the admissions of chronicdisease patients in autumn has a strong impact on hospitals' waiting lists, as well as the appearance of flu outbreaks from november to january. likewise, the impact of spring on some patients with respiratory or allergic disorders can also explain these waiting list differences. on the other hand, the simulation allows us to conclude that, bed occupation rate ranges between 70 and 80 per cent of hospital capacity in 7 out of 100 days, during 12 per cent of the days this rate ranges between 80 and 90 per cent, and during 19 per cent of the days it ranges between 90 and 99 per cent. that is, 28 out of 100 days the hospital is working below its capacity, while 62 out of 100 days the hospital generates a widely variable-length waiting list. thus, wide variability is observed in the use of hospital capacity. the fact that bed occupancy rate shows a strong seasonal behaviour may be due to increased healthcare pressure in some months due to the impact of some climate conditions on certain disorders. 3.1. beds-inpatient activity elasticity in this subsection we examine what happens in the simulation analysis if the number of hospital beds increases. to this end, two alternative scenarios are assumed. first, in scenario 1 the number of beds has no impact on inpatient activity and the stays pattern. this assumption is then removed in scenario 2. in both cases, we consider the actual number of beds (1,100), as well as successive increases: 55 beds more european journal of government and economics 4(2) 127 (5% capacity increase), 110 beds more (10%), 165 beds more (15%), 220 beds more (20%), 330 beds more (30%), 440 beds more (40%) and 550 beds more (50%). 3.1.1. scenario 1 here daily inpatient activity is simulated as described above, i.e. w̂ and w̂ ( h̂ and h̂ ) are estimated using the inpatient activity sample of working days (holidays). this implies that no increase in the number of hospital beds, irrespective of the magnitude of the increase, modifies the generation process of new values for inpatient activity or length of stay. in this context, figures 6 and 7 show, respectively, both the median of the 500 simulated waiting lists and the median of the 500 simulated occupancy rates for various amounts of beds. figure 6. median of simulated waiting lists. no beds effect figure 7. median of simulated daily occupancy rate. no beds effect as shown in figure 6, increased number of beds, although slightly increased, would lead to significantly reduced waiting lists. for instance, a 5 per cent increase in the number of available beds would double the period with no waiting lists (from 38 to 76 per cent). besides, the remainding waiting list would also be considerably smaller. particularly, waiting lists comprising 1-100 patients would prevail only 18 per cent of the time (vs. 21 per cent), 100-200 patients would correspond to 5 per cent of the time (vs. 14 per cent), and those comprising over 200 patients would disappear. if the number of beds increased by 10 per cent, the period with no waiting list would extend up to 95 per cent, as there would be a below-100 waiting list during 4 per cent of the time. finally, if the increase in the number of beds were 15 per cent, there would be no days with waiting lists in the analysed period. in short, a below 15 per cent increase in hospital production capacity would be enough to eliminate waiting lists. the data shown by figure 7 allow deducing that a 5 per cent increase in the number of hospital beds would lead to: (i) the number of days of full hospital occupancy (therefore involving waiting lists) would be reduced from 62 to 25 per cent; and (ii) extended periods in which the occupancy rate is below 100 per cent. at the same time, a 10 per cent increase in the number of beds would reduce the number of days in which hospital occupancy rate is 100 per cent up to only 5 per cent per year and would significantly increase the number of days in which occupancy rate remains below 100 per cent. for instance, while the current number of beds leads to hospital occupancy rates between 90 and 99 per cent during 19 per cent of the year, this occupancy rate would extend up to 40 or 42 per cent with either a 5 or a 10 per cent increase in the number of beds, respectively. if the number of hospital beds is increased by 20 per cent, maximum occupancy would not exceed 99 per cent antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 128 throughout the year, while a 30 or a 40 per cent increase would lead maximum occupancy not to exceed 90 or 70 per cent. the results obtained under scenario 1 can be formally recorded as follows. proposition 1. if bed capacity has no impact on inpatient activity, even a minor increase in the number of beds leads waiting lists to decrease, and occupancy rate to drastically reduce. 3.1.2. scenario 2 in this case we take the median of daily inpatient activity in working days (holidays) and the number of beds in working days (holidays) to fit a linear regression. it is well known that, if the number of hospital beds increases, inpatient activity tends to adapt itself to the hospital’s new productive capacity. this is the so-called roemer's law of demand (shine and roemer, 1959; roemer, 1961). in our case, this effect between capacity and utilization is taken into account by adopting the value that kroneman and siegers (2004) obtained for a set of hospitals from 19 european countries. these authors found that a 1 per cent increase in the number of beds leads inpatient activity to increase 1.44 per cent. thus, inpatient activity seems sensitive to bed supply. this value is taken for the simulations completed in scenario 2 with the analysed hospital. since we are modeling inpatient activity as a normal distribution, the pattern of the variable is determined by mean and standard deviation. then we examine whether both parameters, mean and standard deviation, are related to the number of beds. to avoid pernicious effects from outliers, median and median absolute deviation (mad) are considered as a robust estimator of dispersion.8 the fitted regression model, where the covariate is the number of hospital beds and the response is the median of daily inpatient activity, is given by       holidaysin ,0643.02362.4 days in working ,0972.09619.3 activity) (inpatientmedian b b (7) where b denotes the number of beds. hence, there is a reasonable linear relationship between the median of inpatient activity and beds. indeed, the determination coefficients for both fittings are over 0.9. in addition, the fitted regression models, where the covariate is the number of hospital beds and the response is the mad of daily inpatient activity, is       holidaysin ,0181.02984.2 days in working ,0165.03747.2 activity) (inpatient mad b b (8) where, once more, a reasonable linear relationship can be observed between mad of inpatient activity and the number of beds. the determination coefficients for both fittings are over 90 per cent. therefore, we can use a modified simulation process where the inpatient activity values for working days are drawn from the normal distribution )ˆ,ˆ( wwn  , where bw 0972.09619.3ˆ  and bw 0165.03747.2ˆ  (9) 8 the mad is the median of the absolute deviations from the data’s median. for example, for a dataset as {2,2,3,4,12}, the median is 3, so the absolute deviations from the median are {1,1,0,1,9} (reordered as {0,1,1,1,9}) with a median of 1, in this case unaffected by the value of the outlier 12. hence, the mad is 1. european journal of government and economics 4(2) 129 in turn, inpatient activity values for holidays are drawn from the normal distribution )ˆ,ˆ( hhn  , where bh 0643.02362.4ˆ  and bh 0181.02984.2ˆ  (10) figures 8 and 9 sum up the main results provided by the validated model in scenario 2. figure 8. median of simulated waiting list with a beds effect figure 9. median of simulated daily occupancy rate with a beds effect figure 8 shows that now waiting lists cannot be eliminated, not even when the increase in the number of beds is high. in this scenario in which the admission rate adjusts itself to the number of available beds, a 5 per cent increase in the number of beds is observed not to reduce but to increase waiting lists. from here onwards, waiting lists are reduced if the number of beds keeps growing over 5 per cent. although the reduction is not proportional, every 5 per cent increase in the number of beds can be said to lead to a 1 per cent waiting list reduction. waiting list increases are due to the fact that an increase in the hospital’s inputs leads to an increased number of patients included for admission. besides, given that the elasticity is positive and over-1, greater input amounts lead to greater demand and therefore to longer waiting lists. though consecutive increases in the number of beds do not shorten waiting lists at the same pace, they do show some effect in this sense, probably due to the appearance of growing scale performances from a particular hospital size which compensate for the effect of (over-1) admissiondemand elasticity relative to inputs. the simulation results achieved in scenario 2 can be summarized in the following proposition. proposition 2. if we consider the beds effect on inpatient activity, an increase in the number of beds leads to no significant differences in waiting list and occupancy rate (even for huge increases of beds). 4. discussion the expansion of hospital physical capacity (building new surgical units, for example) is a long-run policy that may require time to be implemented. the increase of the health workforce may be even slower, since physicians and specialists need several years' training before becoming active. although staff can be recruited from abroad, such staff may face assimilation difficulties and such a policy can therefore antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 130 also take time. this means that the different ways of increasing supply will generally involve different costs and require different time scales. in the short run, purchasing extra activity from public facilities at low marginal cost may be possible if there is spare capacity. however, if public facilities are already working close to full capacity, purchasing extra activity in the short run will only be possible at high marginal costs. in the medium to longer term, it may well be cheaper to expand activity by expanding public capacity. for example, denmark adjusted its public capacity to respond to the upsurge in demand for coronary revascularisation procedures more rapidly than england in the 1990s. consequently, waiting times for revascularisation fell in denmark, whereas they rose steeply in england (siciliani and hurst, 2003). it is argued that, in principle, waiting times can be reduced through supply-side policies, if the volume of surgery is not considered adequate, or through demandside policies, if the volume of surgery is considered adequate. supply-side policies include raising production capacity by increasing the number of beds and specialists, or by using the available capacity in other (private) hospitals. they also include increasing productivity by funding extra activity, fostering day-surgery, and linking doctors' and hospitals' remuneration system to their performance (hurst and siciliani, 2006). however, it is common to take measures aimed at reducing waiting times by increasing activity, and then find that, after a brief period, demand has increased and waiting times have reverted to levels similar to those before the application of measures. such responses may be hard to overcome, since demand responds positively to reductions in waiting times. this is the result rendered by the present study. by means of simulation analysis, we observed that if the demand pattern is not modified relative to the increased number of beds, a (slight) increase in the number of beds would remarkably help to reduce surgical waiting lists. particularly, a 10 per cent increase in the number of available beds would reduce variable-size surgical waiting lists from 62 to 5 per cent of the studied period. in this context, this production capacity increase would be highly effective. however, if the demand responds positively to increased supply, then every increase in the number of beds (no matter how large) would have scarce impact on waiting list reduction. the outflow (supply) of elective surgery depends on both public and private surgical capacity, and the productivity with which capacity is used. econometric evidence (cross-sectional and at national level) suggests that higher capacity —in terms of increased numbers of beds and physicians— is associated with lower waiting times. evidence on the impact of capacity is provided by martin and smith (1999) through an english database from the hospital episode statistics in 1991/92. these authors showed that waiting time is negatively associated with the number of available beds. they particularly found that elasticity equals -0.242. similarly, lindsay and feigenbaum (1984) found out that waiting times are negatively associated to both the number of available doctors and beds. furthermore, álvarez and centeno (1999) used a simulation analysis for the washington adventist hospital to evaluate the impact of an expansion in the number of beds in the emergency room. an expansion in the number of beds was found to result in a 0.6-hour reduction of average length of stay. kirtland et al. (1995) used simulation to improve performance by reducing patient time in the system and determining appropriate staffing levels. eleven alternatives were studied and resulted in a reduction of 38 minutes on average. however, larger increases in capacity may have a different impact on waiting times according to the level of excess demand and initial waiting time. countries with low supply and high initial waiting times are likely to have elastic demand to waiting-time variations. for this reason, the effect of even larger increases in capacity on waiting times may be rather modest (hurst and siciliani, 2006). in general, supply-side policies may well succeed in their aim of raising the rate of elective surgery, but they may be disappointing in their effects on waiting times. that european journal of government and economics 4(2) 131 is because an increase in supply may follow rather than lead an increase in demand, or may be overtaken by fresh increases in demand. moreover, any reduction in waiting times may encourage an increase in admission rates to lists due to lowered clinical thresholds (hurst and siciliani, 2006). moreover, many commentators suggested that an increase in the supply of hospital beds tends to generate additional demand either in the form of more admitted patients or patients treated for longer time periods, or some combination of both (zeraati et al., 2005). formerly, shain and roemer (1959) found very close correlations between the availability of short term general hospital beds per 1,000 population, and rates of use as measured by hospital days per 1,000 population. later, roemer (1961) also reported on a natural experiment where a sudden increase in hospital beds in a country—with no changes in other factors—led to a sharply increased utilization rates. using data from 10 european countries hospitals, kroneman and siegers (2004) found out that inpatient activity increases 1.44 per cent for every 1 per cent increase in the number of hospital beds. taking into account this beds-inpatient activity elasticity, the results regarding the impact of beds on waiting times offered by our study widely differ from those obtained without considering the impact of bed capacity on inpatient activity. now, an increase in bed capacity leads to no significant differences in waiting lists and occupancy rates, even for huge bed-capacity increases. our simulation model then reveals the small effect that supply policies may have on waiting lists when healthcare demand adapts itself to the new supply conditions. 5. concluding remarks waiting time for elective surgery is a significant problem in the current medical world. this paper aims at reproducing, by means of a monte carlo simulation, how the number of beds (a rough measure of a hospital’s physical capital) affects inpatient activity, length of stay and, consequently, waiting lists in teaching hospitals like the university multi-hospital complex of santiago de compostela (chus). inpatient activity is simulated by fitting a normal distribution to the real impatient activity data observed in 2007 and the effect of the number of beds on inpatient activity is modelled with a linear regression model. analysis is performed firstly assuming that the number of beds has no impact on inpatient activity or length of stay (absence of beds capacity effect), and then assuming that the number of beds has certain impact on inpatient activity by adapting itself to the new supply conditions. this research allows us to evaluate drops in waiting lists due to a potential increase in the number of beds. we add empirical evidence showing that if the effect of the number of beds on hospital admission rates is neglected, then an increase in the number on hospital beds, even the slightest increase, drastically reduces waiting lists. by contrast, if the hospital’s admission patterns become affected by the number of beds, every extension in the number of beds, even remarkable extensions, leads to no significant differences in hospital occupancy rates or waiting lists. references akkerman, renzo and marrig knip (2004) ‘reallocating of beds to reduce waiting time for cardiac surgery’, health care management science 7(2): 119-126. alvarez, adriana m. and martha centeno (1999) enhancing simulation models for emergency rooms using vba, winter simulation conference, 1685-1693. churruca, silvia (2000) ‘los líderes en gestión de listas de espera advierten: un suceso no puede condicionar la estrategia’, diario médico, 15/06/2000. available at http://www.diariomedico.com/sanidad/listas/debate.html (last accessed september 2000). antelo, reyes santias, martinez calvo ● bed capacity and surgical waiting lists 132 commonwealth fund (1997), leveling the playing field: financing themissions of academic health centers, report of the commonwealth fund task force on academic health centers new york: commonwealth fund, may. denton, brian, james viapiano and andrea vogl (2007) ‘optimization of surgery sequencing and scheduling decisions under uncertainty’, health care management science 10(1): 13-24. everett, jeffrey e. (2002) ‘a decision support simulation model for the management of an elective surgery waiting system’, health care management science 5(2): 8995. hurst, jeremy and luigi siciliani (2003) tackling excessive waiting times for elective surgery: a comparison of policies in twelve oecd countries, oecd health working paper, no. 6. kirtland, anthony, karen poisker, linda stamp and patricia wolfe (1995), simulating an ed ‘is as much fun as?’, proceedings of winter simulation conference, c. alexopoulos, k. kang, w.r. lilegdon and d. goldsman (eds.), institute of electrical and electronics engineers, washington, 1039-1042 koenig, lane, allen dobson, silver ho, jonathan m. siegel, daniel s. blumenthal and joel s. weissman (2003) ‘estimating the mission-related costs of teaching hospitals’, health affairs 22(6): 112-122. kroneman, madelon and jacques j. siegers (2004) ‘the effect of hospital bed reduction on the use of beds: a comparative study of 10 european countries’, social science & medicine 59(8): 1731-1740. lapierre, sophie d. and angel b. ruiz (2007) ‘scheduling logistic activities to improve hospital supply systems’, computers and operations research 34(3): 624641. law, averill m. (1983) ‘statistical analysis of simulation output data’, operations research 31(6): 983–1029. law, averill m. and w. david kelton (1991) simulation modelling and analysis, mcgraw-hill, new york, 2nd ed. lindsay, cotton m. and bernard feigenbaum (1984) ‘rationing by waiting lists’, american economic review 74(3): 404-417. martin, stephen and peter c. smith (1999) ‘rationing by waiting lists: an empirical investigation’, journal of public economics 71(1): 141-64. r development core team, r (2010), a language and environment for statistical computing, r foundation for statistical computing, vienna, austria. (available at <http://www.r-poject.org>.) roemer, milton i. (1961) ‘bed supply and hospital utilization: a natural experiment’, hospitals 35, 988-993. sampietro, laura and mireia espallargues (2001) ‘nuevas fórmulas para dar solución a las listas de espera’, el médico, 15 june of 2000: 6064. http://www.diariomedico.com/sanidad/listas/debate.html shain, max and milton i. roemer (1959) ‘hospital costs relate to the supply of beds’, modern hospital 92(4): 71-73. siciliani, luigi and jeremy hurst (2003), explaining waiting times variations for elective surgery across oecd countries, oecd health working papers 7. silber sigmund, h. mühling, r. dörr, g. zindler, a. preuss and a. stümpfl (1996) ‘waiting times and death on the waiting list for coronary artery bypass operation. experiences in munich with over 1,000 patients’, herz 21(6): 389-396. european journal of government and economics 4(2) 133 street, andrew and stephen duckett (1996) ‘are waiting lists inevitable?’ health policy 36(1): 1-15. vanberkel, peter t. and john t. blake (2007) ‘a comprehensive simulation for wait time reduction and capacity planning applied to general surgery’, health care management science 10(4): 373-385. wand, m.p. and m.c. jones (1995), kernel smoothing, chapman & hall, london. wilson, j.c. tunnicliffe (1981) ‘implementation of computer simulation projects in health care’, journal of the operational research society 32(9): 825–832. yamaguchi naoito, tomohide tamura, tomotaka sobue, suminori akiba, megu ohtaki, yoshinobu baba, shigeto mizuno and shaw watanabe (1994) ‘evaluation of cancer prevention strategies by computerized simulation model: methodological issues’, environmental health perspectives 102, suppl. 8, 64-71. zeraati, hojjat, farid zayeri, gholamreza babae, navid khanafshar and fatemeh ramezanzadeh (2005) ‘required hospital beds estimation: a simulation study’, journal of applied sciences 5(7): 1189-1191. the impact of port governance and infrastructures on maritime containerized trade on the west coast of latin america european journal of government and economics 7(1), june 2018, 85-101. european journal of government and economics issn: 2254-7088 the impact of port governance and infrastructures on maritime containerized trade on the west coast of latin america beatriz lopez-bermudeza, *, maría jesús freire-seoanea, ignacio de la peña zarzueloa a universidade da coruña, elviña, 15071 a coruña, spain * corresponding author at: universidade da coruña, elviña, 15071 a coruña, spain. email: beatrizlopezbermudez@gmail.com article history. received 27 february 2018; first revision required 7 may 2018; accepted 23 may 2018. abstract. latin american countries have historically had a strong dependence on trade, and are mostly characterized by being exporters of raw materials and importers of manufactured products. this fact has brought about a less negative impact of the world crisis on economic growth, mainly because of the high prices of raw materials. this paper focuses on this geographical area (the west coast of latin america) between 2008 and 2015, and adds to the literature by assessing institutional, port-related and economic factors that influence maritime transport. the analysis makes use of panel data models with fixed and random effects where the hausman test has been applied in order to define a solid specification of all the ports, as well as to discount the particular peculiarities of each country. it is shown that the analysis of maritime transport requires the analysis of a number of variables apart from trade (volume of teus), infrastructures, superstructures (number of calls, gantry cranes), and that other variables, such as port governance, which are sometimes difficult to quantify, need also to be taken into account. keywords. port governance; maritime trade; latin america; panel data. jel classification. c33; n76; r42 1. introduction latin america represents 8% of the world's gross domestic product. from 2008 to 2016, the yearly growth rate of its gdp has been 2.08%, and its population represents 8.5% of the world’s total (world bank, 2017). the world crisis of this century has produced a recession in all develop economies, severely affecting the economies of the united states and the european union, and causing a fall of world growth and trade. however, the latin american countries have managed to avoid major consequences from the global crises, and they have experienced some economic growth. this is mainly explained by the intense trade of raw materials with asian countries, especially china (eclac, 2014). according to unctad (2017), 80% of international trade is carried out by sea. in addition, the importance of seaports for latin america's growth is rooted in the region's colonial history and its natural endowment (serebrisky et al., 2016). latin america’s economy has long doi: https://doi.org/10.17979/ejge.2018.7.1.4334 mailto:beatrizlopezbermudez@gmail.com b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 86 depended on international seaborne trade (agricultural products and extractive industry) and imports of consumer goods, purchased with the capital accrued from those commodities (serebrisky et al., 2016; eclac, 2014). the continued growth of maritime trade across latin america and the increase in vessel sizes have both contributed to the expansion of container handling facilities, and enabled new institutional reforms to accommodate the increasing demand. the objective of this paper is to identify the institutional factors and the main port characteristics that favor the movement of containerized goods in the ports on the west coast of latin america located in chile (antofagasta, arica, coronel, iquique, lirquén, mejillones, san antonio, san vicente and valparaíso), colombia (buenaventura), costa rica (caldera), ecuador (esmeraldas y guayaquil), el salvador (acajutla), guatemala (p. quetzal), mexico (ensenada, lázaro cárdenas and manzanillo), nicaragua (corinto), panama (balboa) and peru (callao, matarani and paita). this paper is structured in six sections: an introduction; the second section, a bibliographic review; the third, a description of the variables that are going to be used and the modelling; the fourth, the results of the model; and, finally, the conclusions and the bibliography. 2. background studies of international economics, especially on economic growth, underline the relationship between growth and trade (romer 1990; rivera-batiz and romer 1991). the related empirical literature is quite recent, and in particular, studies that relate maritime transport and economic growth are relatively scarce (grossmann et al., 2007). radelet and sachs (1998), redding (2002), and the series "review of maritime transport" of unctad as of 2007 (unctad, 2017) have begun to address these issues. containerized maritime transport of goods, which enjoy the benefits of the economies of scale and, therefore, reduced costs incurred in the transportation of large volumes of goods, constitutes a good reference for assessing the impact of maritime traffic on the economic growth and development of any country (bernhofen et al., 2016; rodrigue et al., 2013). in addition, it should be considered that containerized cargo comes mainly from the secondary and tertiary sector activity (guisan, 2013; kenessey, 1987). indeed, corbett and winebrake (2008) find a very strong association, around 90%, between the us gross domestic product and the volume of containerized goods. the eclac carried out a first attempt to highlight the economic importance of the port infrastructures of latin america and the caribbean in the complex and changing global scheme of maritime routes in 2004 (in full rise of the price of raw materials), by analysing the supply and demand for maritime services, the price of freight, ownership of the fleet and the regime of port governance (sánchez, 2004). unctad (2017) and fay and morrison (2006) also highlight for the first time the central role that latin america could play in relation to international shipping patterns. these works emphasize the need to take advantage of the economic impulse provoked by the high prices of raw materials; something which, according to some authors, has b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 87 not been taken advantage of properly in terms of reducing inequality and increasing the quality and efficiency of infrastructures (bitar, 2016). research efforts on different port governance systems resulted from the "port devolution" process that began around the year 1945 in the united kingdom (baird, 1995, 1999, 2000; goss, 1998; baird and valentine, 2006; pettit, 2008). baird (1995, 1999, 2000), which formulated the "port function privatization matrix", wherein he made a classification based on the public and private nature of what he considered to be the most important factors. later, baltazar and brooks (2001), taking baird as a reference, presented the "port devolution matrix", where three port functions are specified (regulator, owner and operator), classified into public, private or mixed. in 2007, the world bank published the "port reform toolkit", which proposes one of the simplest and clearest classifications of port reforms, and permits to identify the public or private nature of the institutions involved carrying out a series of port functions. under these hypotheses, the results are four models of port governance: public service port, toolport, landlord port and private service port. in each of them, the following functions are to be analyzed: port administration, nautical administration, infrastructure (nautical or port), superstructure (equipments and buildings), cargo handling, and other port facilities such as pilotage, dredging, tugboats, mooring, etc. in this research, they also record the balance between private risk and the importance of the mandatory regulations that exist in each port model according to its private or public dimension. the public service port model refers to those ports whose ownership, planning, management and operation are entirely in the hands of the public sector. that is, besides providing all port services, the state also owns the land, infrastructure and superstructure. it is a model in declining use (world bank, 2007), which worked in latin american countries before what we could call the port "modernization"; that is, prior to the enactment of the port laws during the last two decades, which usually coincide in this case with obsolete labour regimes and lacking new technologies (hoffman, 1999). in the toolport model, although the public sector is the owner of the infrastructure and in charge of the operations and management of the port, they grant, however, some operational services, such as stowage, pilotage, supply, storage, etc. to legal business units with some percentage of representation either by the state or purely private. some latin american countries have used this intermediate modality to exercise port governance during periods of transition between the promulgation of port modernization laws, which in certain cases can, together with the starting of the new management, take up to ten years. in the landlord model, a state entity is the owner of the land (individual port authority or supra-local agency/institution), but it transfers via a concession contract the operation and management of the port infrastructures to a second entity, which may be a private, public or mixed company, depending on the legislation applicable to the case (cabrera et al., 2015). the transition from the toolport model to the landlord model takes place in the 1980s, and is typical in european union countries (hoffmann, 1999), as well as in the port system of those countries signatories of the north american free trade agreement. -nafta(fawcett, 2006). b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 88 the private service port model involves the privatization of all elements, including ownership of the land, with the public sector only retaining a standard regulatory supervision power. the first case of application of this model took place in 1981 with the privatization of all british ports, under the control of the associated british ports holding (pettit, 2008). at present, we can mention other countries with governance tensions tending to favour the total privatization of this service, such as argentina, chile, colombia, malaysia, mexico, new zealand, the philippines and venezuela (pettit, 2008). another manifestation of this model is called terminalization, or the assumption of terminal operators once established the granting of the functions of the port authority, which is relegated to the role of mere spectator (verhoeven, 2010; slack, 2007). rodrigue et al. (2013) point out that the concept of port governance is transcendental when analysing a port; in fact, he remarks it as the third essential pillar to define the function of a port. the concept of port governance arises out of the need of the ports to present a clear organizational structure and efficient management in the transport service. this definition is currently used by most researchers. sánchez et al. (2015) present the challenges and opportunities that the seaborne transport and port development represent for latin america and the caribbean, introducing the theme of sustainability as a transversal axis in the improvement of infrastructures. in addition to the improvements in infrastructures it is important to highlight the importance of other factors such as the characteristics of the port terminal or the governance system, factors that many authors have investigated in relation with the concept of efficiency (serebrisky et al., 2016; chang and tovar, 2014; núñez-sánchez and coto-millán, 2012; ramos-real and tovar, 2010; coto millán et al., 2000; roll and hayuth, 1993). 3. methodology to analyze the relationship between teus, gdp and port variables, a regression panel data model has been used with the aim of achieving consistent and efficient estimators of the partial effects of the observable independent variables on the dependent one, expressed as β = ∂e[yit | xit]/∂xit (greene, 2001). estimates are made in the statistical program stata 13. in addition, the structure of the data facilitates the use of this methodology, since the time variable is smaller than cross-sectional observations (n <t). the choice of the eight-year period (2008-2015) was based on the progress of maritime transport both in terms of development of infrastructures and increase in the frequencies of commercial routes in the geographical area subject to this analysis (ais, 2017). 3.1. data a description of the variables used for this study is presented in annex 1. this is information extracted from eclac (2017), annual volume of containerized merchandise, in fact the only database that provides information on the number of teus operated in each of the ports of latin b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 89 america; world bank (wb, 2017), growth indicator; ais (2017), number of ships that have made calls at the port; (several sources1) dummy of governance and the existence of container cranes, for the period analyzed. teus: container movements are expressed in teus (wb, 2016) of ports on the pacific coast of latin american countries. in 2008, the ports with the largest number of container movements were balboa (2,167,977 teus), manzanillo (1,409,782 teus) and callao (1,203,315 teus). in 2015, the ports with the largest movement of containers are the same, but they have experienced significant increases: balboa with an interannual increase of 6.49% (3,294,113 teus), manzanillo with 9.30% (2,458,135 teus) and callao with 7.24% (1,900,444) teus). gdp: gross domestic product in constant dollars of 2011 in power of purchasing parity (hereinafter, ppp) for the countries analyzed. in 2008, the countries with the highest gdp were mexico (1,050 ·109 $), colombia (272·109 $) and chile (208·109 $). in 2015, the countries with the highest gdp are again mexico with an interannual growth of 1.90% (1,210·109 $), colombia with a growth of 2.83% (359·109 $) and chile with 3.31% (263·109 $). dg1: dummy variable that takes value 1 when ports are fitted with gantry cranes and 0 when not. in 2008, the ports equipped with this type of cranes for the container loading/unloading operations were balboa, buenaventura, callao, ensenada, guayaquil, lazaro cardenas, manzanillo, san antonio and valparaíso. in 2015, the ports of coronel and paita will be linked to the previous list. nb: number of annual vessels calling at the ports analyzed. in 2008, the ports with the highest figures were balboa (729), manzanillo (652) and guayaquil (439). in 2015, the ports with the highest number of calls are the same, with balboa showing an increment of 2.49% (874), manzanillo continues to be second with an interannual rate of -0.19% (642) and callao takes a third position with an interannual increase of 3.43% (455). governance: dummy variable that takes value 1 when the governance system has a landlord structure and 0 otherwise. in 2008, 18 of the 23 ports analyzed had a landlord structure, whereas acajutla, callao, corinto, esmeraldas and puerto quetzal had a toolport or service port governance system. in 2015, there will be 20 ports with callao and puerto quetzal, while acajutla, corinto and esmeraldas will continue with a toolport or service port system. 3.2. model the models with panel data demand the premise that the transverse variables (n) must be higher than the temporary variables (t), because otherwise we would not be in a case of multidimensional time series (cross-sectional time-series), that is, n> t (in this analysis, 23> 8). the main advantage of the panel data if compared with other analysis is that they allow the estimation of multiple regression coefficients, which could not be carried out either with crosssectional data or with time series data (arellano, 1992). 1( ep arica, 2017; tpsa, 2017;epi, 2017;angamos, 2017;epa, 2017;epv, 2017;epsa, 2017;epsv, 2017; svti, 2017; lirquén, 2017; coronel, 2017;tp euroandinos, 2017; apm terminals, 2017; dp world callao, 2017; tisur, 2017;a.p. esmeraldas, 2017;ap guayaquil, 2017; sprbun, 2017;tcbuen, 2017;plp, 2017; spcaldera, 2017; pcorinto 2017;cepa, 2017; epq, 2017;tcq, 2017; sct, 2017) b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 90 in the models with panel data, the observations repeat themselves over time for a sample of individual units (arellano, 1992). therefore, it can be stated that these models combine both temporal and transversal dimensions; that is, we happen to find variables with a temporary nature which evolve over time and other transversal variables as the different port terminals. panel data models can be classified into two large groups: static and dynamic. in this paper a static panel is used. the first researchers to estimate a model with static panel data were pitt and lee (1981), who used the maximum likelihood technique. later on, schmidt and sickles (1984) compiled the different possibilities offered by the panel data: fixed effects, random effects and maximum-likelihood. this last method of statistical estimation requires t → ∞, and for this reason it will not be used in this investigation. the basic specification of a model with panel data can be expressed as follows (greene, 2005): yit = x’itβ + z’iα + ɛit =x’itβ+ci+ ɛit, where • yit is the dependent variable or explained • x'it is a vector of independent or exogenous variables (k ∙ 1) • z'iα are the individual effects, where zi contains a constant term and a series of individual or group variables, which may be observable or not observable • β is the slope vector of the equation; • t refers to the time series that reaches the period t (t = 1,2, .., t); • i refers to the ports, the last port being n (i = 1,2, .., n); • ɛit is the random term. as it has been defined, it is a classic regression model. if zi is observable for all individuals, the model can be treated as an ordinary linear model, and an adjustment by ordinary least squares could be made. the problem arises when zi is not observable, as happens in most cases. the main objective of this analysis is to achieve consistent and efficient estimators of the partial effects of the independent variables on the dependent variables, expressed as (greene, 2005): β = ∂e [yit | xit] / ∂xit the fixed-effects approach takes αi as a specific constant term for each transversal unit and implies that the differences between transversal units can be captured with differences in the constant term (greene, 2005). in order to clarify the structure of the relation between growth and containerized maritime transport, we shall take the econometric specification as starting point for all the ports subject to this analysis: b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 91 lteusi, t =βi,0 t +βi,1 t ⋅lgdp+βi,2 t ⋅lnbi t + βi,3 t · governance + βi,4 t ·dg1+u i t where • t: 2008 to 2015; • i: represents each of the ports considered in the model; • lteusi t: natural logarithm of the volume of containers operated in port i during year t (wb, 2017); • lgdpi t: natural logarithm of the gdp of every country considered in the study in year t, expressed in ppp (constant dollars 2011), (wb, 2017); • lnbi t: natural logarithm of the number of ships that have called in every port considered (ais, 2017); • governancei t: dummy representing the governance system, 1 for the landlord model and 0 for the rest; • dg1i t: dummy that represents the presence of gantry cranes, with the values 1 or 0 depending on whether or not those terminals are equipped with cranes of this type; • ui t: error term. 4. estimation results table 1 presents the results for the models with panel data fixed and random effects, additionally calculating the hausman test, which is the statistical instrument that indicates the preference for the fixed-effect model (hausman, 1978; hausman and taylor, 1981). this allows to define a solid specification of the set of analysed ports, independently of the peculiarities of each port and country. in the first place and according to the resolution conditions, the two basic tests h0:{ β k=0} (statistic f for fixed effects and wald chi2 for random effects) reject the null hypothesis of coefficients equal to zero. on interpreting the results obtained, the hausman test (lee et al., 1998; arellano and bover, 1995) is rejected, so the regression of fixed effects is used for this purpose. there is an evident existence of unknown intrinsic characteristics acting on the relationship (rho = .9246). the covariance between the error term and the variables of the model has a value (cov(vj,xi,j)=-.6429). despite the relatively high value obtained, it could be said that the influence of the idiosyncratic factors of each port is under control. increase gdp over teus the economic growth of a country expressed in gdp as constant dollar 2011 ppp, should we keep constant the number of calls made by ships, the port governance system and the commercial policy, increases the volume of teus by e(0.6193)-1= 0.8576. the evolution of the global containerized supply chain implies that the countries carry out b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 92 important works of improvement or expansion in their port infrastructures (notteboom and winkelmans, 2001). these investments are usually focused on civil engineering objectives: installation of new gantry cranes, extension of berthing lines, construction of new docks, deepening of the draft by dredging, automation of the terminal, increase in the storage surface, installation of multimodal exchange devices, etc. (tsinker, 2014; tang et al., 2011). increase in the number of vessel calls over teus the first port variable to take into account in the model is the number of ships' calls in the port during the year, so that an increment of one unit produced an increase of e (0.5048)-1= 0.6566, in the teus. the number of calls derived from the increase in the volume of the container ship fleet (unctad, 2017) does not strongly affect the volume of operated teus. however, there is a strong relationship in the geographical area under analysis, which indicates that these increases in capacity have not occurred in most of the ports. gantry cranes in port terminals the second port variable being analyzed is the existence of gantry cranes in the port to carry out the stowage operations. with the available data it is worth mentioning that, with the exception of balboa, buenaventura, callao and lazaro cardenas (in some years of the analyzed period), no other port has more than 10 gantry cranes. moreover, many of them are just equipped with only one unit. the results of the model show that the existence of a gantry crane in the ports results in an increase in the volume of operated containers of e(1.4604)1=330.76%. the effect of the governance system over teus the dummy variable represented by the landlord port governance system is significant in the analysis, what indicates its relevancy in relation to the volume of containers moved in the ports. besides, the implementation of this system in a port gives rise to an increment in the volume of containerized goods of e(0.2877)-1=33.33% teus. the impact of the landlord model is weak, and it would be necessary, in the future, to continue this process of modernization of the governance systems in latin america. the modernization of the port sector does not necessarily end with the privatization of infrastructures, but resides, as many researchers point out, in the correct organization of institutions (notteboom et al., 2013; lee and song, 2010; verhoeven, 2010; de langen and pallis, 2007). the structured made by verhoeven (2010) is the key in this sense. verhoeven (2010) creates a state entity in the form of an agency (community manager) that provides flexibility to the national port system, fostering competition among different maritime actors, being aware of the need to integrate all the stakeholders involved in the logistics chain. in b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 93 addition, the different entities of the port system (owner, operator, regulator and community manager) are assigned actions and lines that should be carried out to respond to the needs of port industry. this responds to the structure that the port system must present in order to solve the problems that arise between the different actors of the port logistics chain. researchers have pointed out different entities such as the community manager of verhoeven (2010), the cluster manager of de langen and pallis (2007) that are circumscribed in an environment of path dependence (notteboom et al., 2013) and coopetition (lee and song, 2010) that seek to modernize and update existing governance systems. table 1. fixed effects and random effects model. dependent variable: lteus lteus fixed effects random effects constant 𝛽𝛽0,𝑗𝑗 -66.369 12.93 p>|t| 0.348 0.000 lgdp 𝛽𝛽1,𝑗𝑗 0.619 -0.161 p>|t| 0.027 0.051 lnb 𝛽𝛽2,𝑗𝑗 0.504 0.648 p>|t| 0.000 0.000 governance 𝛽𝛽3,𝑗𝑗 .2877 0.288 p>|t| 0.098 0.076 dg1 𝛽𝛽4,𝑗𝑗 14.604 11.586 p>|t| 0.000 0.000 n 184 184 f 72.20 p>f 0.000 wald chi2 298.64 p>chi2 0.000 cov(vj,xi,j) -0.6429 0(def) rho .9246 .5587 r2 within .5181 .4779 between .5028 .8414 overall .4921 .7990 hausman test chi2 26.58 p>chi2 0.000 conclusions the panel data model used to analyse the relationship between the volume of teus moved in the port shows a positive relation with all the variables used; the gross domestic product of each country; the number of ships’ calls; the use of gantry cranes for the stowage and the existence of a landlord governance system. the estimation by fixed-effects can, with the b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 94 structure of data panel, be made for the sample of ports, which means that the analysis provides consistent estimators despite the idiosyncratic differences existing between ports. the relationship between the variation of the gross domestic product of each country and the trade measured in terms of maritime containerized transport is present in this analysis, since gdp produces an effect on the teus of e(0.6193)-1= 0.8576. moreover, the two characteristics related to port infrastructure and superstructure are the number of ships’ calls and the existence of gantry cranes. first of all, the number of calls made at the port can be inferred as a measure of the berth line, and this is a real fact, since taking advantage of economies of scale we can see that the present increase in the capacity of the vessels calls for longer lines of berthing and thus less port calls. the model shows that the increase in the number of calls causes an effect on the teus of e(0.5048)-1= 0.6566. in the same way, the existence of gantry cranes is important, especially for the region to which we are referring and with recent investments in infrastructure and port superstructure. besides, most of the analysed ports have less than 10 container cranes for their stowage/unstowage operations, and consequently, just the existence of at least one gantry crane means an increase in the volume of teus moved in port of e(1.4604)-1= 3.3076. in addition, the transport sector is a highly regulated worldwide sector, not only for its importance and impact on countries’ economies, but also for the fact that in most countries the port infrastructure is state property, and therefore its correct operation depends on government’s behaviour. huge private capital investments are made based on exploitation concession contracts on account of the so-called regulation landlord, giving rise to an effect on the teus of e(0.2877)-1= 0.33333 in those countries under analysis. it is necessary to establish a legislative framework that provides the public institutions necessary for the proper functioning of the port industry and that includes all the stakeholders of the multimodal transport logistics chain. this means that it is imperative to generate a correct association between public-private entities in the global supply chain. finally, and based on the results obtained, it can be verified that the analysis of (the) maritime transport requires considering other aspects on the west coast of latin america, not only trade (volume of teus), but also the infrastructures and superstructures of ports (number of calls, gantry cranes), as well as other variables which are sometimes difficult to quantify, such as (the) port governance. this analysis is necessary for the public authorities of the west coast countries, which should promote investments in maritime transport infrastructures, as well as carry on with the development of the legislative framework of the port governance system, in order to strengthen trade and economic growth. b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 95 references ais (2017). ihs markit: maritime portal. acceso a través de: http://maritime.ihs.com/ angamos (2017). web del puerto angamos. chile. retrieved from: http://www.puertoangamos.cl/ ap esmeraldas (2017). web autoridad portuaria esmeraldas. ecuador. retrieved from: http://www.puertoesmeraldas.gob.ec/ ap guayaquil (2017). web de la autoridad portuaria de guayaquil. ecuador. retrieved from: http://www.apg.gob.ec/ apm terminals (2017). web del consorcio apm terminals callao. perú retrieved from: https://www.apmterminalscallao.com.pe/ arellano, m. (1992). introducción al análisis econométrico con datos de panel (p. 9222). banco de españa: servicio de estudios. arellano, m., & bover, o. (1995). another look at the instrumental variable estimation of errorcomponents models. journal of econometrics, 68(1). https://doi.org/10.1016/03044076(94)01642-d baird, a. (1995). uk port privatisation: in context. in proceedings of uk port privatisation conference. scottish transport studies group, 21. baird, a. (1999). analysis of private seaport development: the port of felixstowe. transport policy, 6(2), 109-122. https://doi.org/10.1016/s0967-070x(99)00012-8 baird, a. j. (2000). port privatisation: objectives, extent, process, and the uk experience. international journal of maritime economics, 2(3), 177-194. https://doi.org/10.1057/ijme.2000.16 baird, a. j., & valentine, v. f. (2006). port privatisation in the united kingdom. research in transportation economics, 17, 55-84. https://doi.org/10.1016/s0739-8859(06)17003-1 baltazar, r., & brooks, m. r. (2001). the governance of port devolution: a tale of two countries. canadian public administration, 42(1), 108-132. bernhofen, d.m., el-sahli, z. & kneller, r. (2016). estimating the effects of the container revolution on world trade. journal of international economics, 98, 36-50. https://doi.org/10.1016/j.jinteco.2015.09.001 bitar, s. (2016). las tendencias mundiales y el futuro de américa latina. eclac. cepa (2017). web comisión ejecutiva portuaria autónoma, el salvador. retrieved from : http://bit.ly/2c3jovo chang, v., & tovar, b. (2014). efficiency and productivity changes for peruvian and chilean ports terminals: a parametric distance functions approach. transport policy, 31, 83-94. https://doi.org/10.1016/j.tranpol.2013.11.007 corbett, j.j. & winebrake, j. (2008). the impacts of globalisation on international maritime transport activity. globalisation, transport and the environment. oecd. coronel (2017). web del puerto coronel. chile. retrieved from : http://www.puertodecoronel.cl/ coto-millan, p., banos-pino, j., & rodriguez-alvarez, a. (2000). economic efficiency in spanish ports: some empirical evidence. maritime policy & management, 27(2), 169-174. https://doi.org/10.1016/0304-4076(94)01642-d https://doi.org/10.1016/0304-4076(94)01642-d https://doi.org/10.1016/s0967-070x(99)00012-8 https://doi.org/10.1057/ijme.2000.16 https://doi.org/10.1016/s0739-8859(06)17003-1 https://doi.org/10.1016/j.jinteco.2015.09.001 https://doi.org/10.1016/j.tranpol.2013.11.007 b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 96 https://doi.org/10.1080/030888300286581 de langen, p. w., & pallis, a. a. (2007). entry barriers in seaports. maritime policy and management, 34(5), 427-440. dp world callao (2017). web de dp world callao. perú. retrieved from : http://www.dpworldcallao.com.pe eclac (2014). statistical yearbook for latin america and the caribbean. santiago de chile: united nations publications. eclac (2017). perfil marítimo y logístico. retrieved from : http://bit.ly/2fv5bzs ep arica (2017). web de la empresa portuaria arica. chile. retrieved from : http://www.puertoarica.cl/web/index.php epa (2017). web de la empresa portuaria antofagasta. retrieved from : http://www.anfport.cl/ epi (2017). web de la empresa portuaria iquique. chile. retrieved from : http://www.epi.cl/ epq (2017). web empresa portuaria quetzal. retrieved from : http://bit.ly/1p1gogw epsa (2017). web de la empresa portuaria san antonio. chile. retrieved from : http://www.sanantonioport.cc.cl epsv (2017). web de la empresa portuaria san vicente –talcahuano. chile. retrieved from : http://www.puertotalcahuano.cl/ epv (2017). web de la empresa portuaria valparaíso. chile. retrieved from : http://www.puertovalparaiso.cl fawcett, j. (2006), port governance and privatization in the united states: public ownership and private operation. research in transportation economics, 17, 207-235. https://doi.org/10.1016/s0739-8859(06)17010-9 fay, m. & morrison, m. (2006). infrastructure in latin america and the caribbean: recent developments and key challenges. washington: world bank. https://doi.org/10.1596/978-08213-6676-9 goss, g. s. (1998). u.s. patent no. 5,828,653. washington, dc: u.s. patent and trademark office. greene, w. (2001). estimating econometric models with fixed effects. department of economics, stern school of business, new york university. grossmann, h., otto, a., stiller, s. & wedemeier, j. (2007). growth potential for maritime trade and ports in europe. intereconomics, 42(4). https://doi.org/10.1007/s10272-007-0223-x guisan, m. c. (2013). macro-econometric models of supply and demand: industry, trade and wages in 6 countries, 1960-2012. applied econometrics and international development, 13(2), 45-56. hausman, j. a. (1978). specification tests in econometrics. econometrica: journal of the econometric society, 1251-1271. https://doi.org/10.2307/1913827 hausman, j. a., & taylor, w. e. (1981). panel data and unobservable individual effects. econometrica: journal of the econometric society, 1377-1398. https://doi.org/10.2307/1911406 hoffman, j. (1999). las privatizaciones portuarias en américa latina en los 90: determinantes https://doi.org/10.1080/030888300286581 https://doi.org/10.1016/s0739-8859(06)17010-9 https://doi.org/10.1596/978-0-8213-6676-9 https://doi.org/10.1596/978-0-8213-6676-9 https://doi.org/10.1007/s10272-007-0223-x https://doi.org/10.2307/1913827 https://doi.org/10.2307/1911406 b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 97 y resultados, in transport unit, eclac. trabajo presentado en el seminar of world bank 1999. kenessey, z. (1987). the primary, secondary, tertiary and quaternary sectors of the economy. review of income and wealth, 33(4). https://doi.org/10.1111/j.1475-4991.1987.tb00680.x lee, e. s., & song, d. w. (2010). knowledge management for maritime logistics value: discussing conceptual issues. maritime policy & management, 37(6), 563-583. https://doi.org/10.1080/03088839.2010.514959 lee, m., longmire, r., matyas, l. & harris, m. (1998). growth convergence: some panel data evidence. applied economics, vol. 30, nº 7. https://doi.org/10.1080/000368498325336 lirquen (2017). web del puerto lirquén. chile. retrieved from : http://www.puertolirquen.cl/ notteboom, t. e. & winkelmans, w. (2001). structural changes in logistics: how will port authorities face the challenge? maritime policy & management, 28(1). https://doi.org/10.1080/03088830119197 notteboom, t., de langen, p., & jacobs, w. (2013). institutional plasticity and path dependence in seaports: interactions between institutions, port governance reforms and port authority routines. journal of transport geography, 27, 26-35. https://doi.org/10.1016/j.jtrangeo.2012.05.002 núñez-sánchez, r., & coto-millán, p. (2012). the impact of public reforms on the productivity of spanish ports: a parametric distance function approach. transport policy, 24, 99-108. https://doi.org/10.1016/j.tranpol.2012.07.011 pcorinto (2017). web puerto de corinto. retrieved from : http://bit.ly/2kh2jff petit, m. g. (1988). u.s. patent no. 4,760,837. washington, dc: u.s. patent and trademark office. pettit, s. j. (2008) united kingdom ports policy: changing government attitudes, marine policy, 32(4), 719-727. https://doi.org/10.1016/j.marpol.2007.12.002 pitt, m. m., & lee, l. f. (1981). the measurement and sources of technical inefficiency in the indonesian weaving industry. journal of development economics, 9(1), 43-64. https://doi.org/10.1016/0304-3878(81)90004-3 plp (2017). web portal de logística de panamá. retrieved from : http://bit.ly/2jxrolu radelet, s. & sachs, j. d. (1998). shipping costs, manufactured exports, and economic growth. american economic association meetings, harvard university, mimeo. ramos-real, f. j., & tovar, b. (2010). productivity change and economies of scale in container port terminals a cost function approach. journal of transport economics and policy (jtep), 44(2), 231-246. redding, s. (2002). specialization dynamics. journal of international economics, vol. 58, nº 2. rotterdam, elsevier. https://doi.org/10.1016/s0022-1996(01)00169-6 rivera-batiz, l.a. & romer, p. m. (1991). economic integration and endogenous growth. the quarterly journal of economics, 106(2), 531-555. https://doi.org/10.2307/2937946 rodrigue, j. p., comtois, c. & slack, b. (2013). the geography of transport systems. london: routledge. https://doi.org/10.1111/j.1475-4991.1987.tb00680.x https://doi.org/10.1080/03088839.2010.514959 https://doi.org/10.1080/000368498325336 https://doi.org/10.1080/03088830119197 https://doi.org/10.1016/j.jtrangeo.2012.05.002 https://doi.org/10.1016/j.tranpol.2012.07.011 https://doi.org/10.1016/j.marpol.2007.12.002 https://doi.org/10.1016/0304-3878(81)90004-3 https://doi.org/10.1016/s0022-1996(01)00169-6 https://doi.org/10.2307/2937946 b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 98 roll, y., & hayuth, y. e. h. u. d. a. (1993). port performance comparison applying data envelopment analysis (dea). maritime policy and management, 20(2), 153-161. https://doi.org/10.1080/03088839300000025 romer, p. m. (1990). endogenous technological change. journal of political economy, 98(5-2), 71-102. https://doi.org/10.1086/261725 sánchez, r. (2004). puertos y transporte marítimo en américa latina y el caribe: un análisis de su desempeño reciente. eclac. sánchez, r., jaimurzina, a., wilmsmeier, g., pérez-salas, g., doerr, o. & pinto, f. (2015). transporte maritimo y puertos: desafíos y oportunidades en busca de un desarrollo sostenible en américa latina y el caribe. eclac. schmidt, p., & sickles, r. c. (1984). production frontiers and panel data. journal of business & economic statistics, 2(4), 367-374. sct (2017) web secretaría de transporte, gobierno mexico. accedido a través de: http://bit.ly/1ligixb serebrisky, t., sarriera, j. m., suárez-alemán, a., araya, g., briceño-garmendía, c., & schwartz, j. (2016). exploring the drivers of port efficiency in latin america and the caribbean. transport policy, 45, 31-45. https://doi.org/10.1016/j.tranpol.2015.09.004 slack, b. (2007), the terminalisation of seaports, in international workshop on ports, cities and global supply chains, hong kong, 2005. spcaldera (2017). web sociedad portuaria puerto caldera. retrieved from : http://www.spcaldera.com/ sprbun (2017). frecuencias y tiempos de transito. servicios regulares de naves portacontenedores. retrieved from : http://www.sprbun.com tang, l. c., low, j. m. & lam, s. w. (2011). understanding port choice behavior—a network perspective. networks and spatial economics, 11(1). https://doi.org/10.1007/s11067-0089081-8 tcbuen (2017) web de la terminal de contenedores buenaventura. colombia. retrieved from : http://www.tcbuen.com/ tcq (2017). web terminal contenedores quetzal. retrieved from : https://www.tcq.com.gt/ tisur (2017). web puerto de matarani. accedido a través de: http://bit.ly/2krn3wu tp euroandinos (2017). web de terminales portuarios euroandinos. perú. retrieved from : http://www.puertopaita.com/ tpsa (2017). web del terminal puerto arica s.a. chile. retrieved from : http://www.tpa.cl/ tsinker, g. (2014). handbook of port and harbor engineering: geotechnical and structural aspects. springer. unctad (2017). review of maritime transport series. retrieved from : http://bit.ly/1vnw1kg verhoeven, p. (2010). a review of port authority functions: towards a renaissance? maritime policy and management, 37(3), 247-270. https://doi.org/10.1080/03088831003700645 world bank (2017). world bank open data. retrieved from : https://data.worldbank.org/ https://doi.org/10.1080/03088839300000025 https://doi.org/10.1086/261725 https://doi.org/10.1016/j.tranpol.2015.09.004 https://doi.org/10.1007/s11067-008-9081-8 https://doi.org/10.1007/s11067-008-9081-8 https://doi.org/10.1080/03088831003700645 https://data.worldbank.org/ b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 99 annex 1. data section descriptive statistics mean values (year: 2008) teus gdp nb gov dg1 acajutla 157415 2.1813e+10 39 0 0 antofagasta 76685 2.08e+11 56 1 0 arica 116720 2.08e+11 34 1 0 balboa 2167977 2.691e+10 729 1 1 buenaventura 743295 2.72e+11 211 1 1 caldera 169827 3.5858e+10 67 1 0 callao 1203315 1.35e+11 357 0 1 corinto 58885 8711343160 28 0 0 coronel 968 2.08e+11 11 1 0 ensenada 110423 1.05e+12 75 1 1 esmeraldas 54885 6.6808e+10 13 0 0 guayaquil 874955 6.6808e+10 439 1 1 iquique 334302 2.08e+11 193 1 0 lázaro cárdenas 524791 1.05e+12 103 1 1 lirquén 231397 2.08e+11 70 1 0 manzanillo 1409782 1.05e+12 652 1 1 matarani 19824 1.35e+11 24 1 0 mejillones 97226 2.08e+11 97 1 0 paita 138993 1.35e+11 57 1 0 puerto quetzal 280281 3.9975e+10 231 0 0 san antonio 687864 2.08e+11 165 1 1 san vicente 604560 2.08e+11 125 1 0 valparaíso 946921 2.08e+11 225 1 1 b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 100 mean value (year: 2015) teus gdp nb gov dg1 acajutla 190708 2.3606e+10 78 0 0 antofagasta 77467 2.63e+11 70 1 0 arica 226893 2.63e+11 90 1 0 balboa 3294113 4.2242e+10 874 1 1 buenaventura 911533 3.59e+11 280 1 1 caldera 235268 4.4415e+10 92 0 0 callao 1900444 1.86e+11 455 1 1 corinto 138006 1.1246e+10 48 0 0 coronel 471426 2.63e+11 108 1 1 ensenada 193424 1.21e+12 121 1 1 esmeraldas 59413 8.6639e+10 26 0 0 guayaquil 1704730 8.6639e+10 380 1 1 iquique 227099 2.63e+11 113 1 0 lázaro cárdenas 1068747 1.21e+12 403 1 1 lirquén 164994 2.63e+11 38 1 0 manzanillo 2458135 1.21e+12 642 1 1 matarani 20002 1.86e+11 6 1 0 mejillones 223124 2.63e+11 70 1 0 paita 214483 1.86e+11 42 1 1 puerto quetzal 389329 4.9883e+10 30 1 0 san antonio 1170184 2.63e+11 259 1 1 san vicente 456176 2.63e+11 83 1 0 valparaíso 902542 2.63e+11 136 1 1 b. lopez-bermudez et al. / european journal of government and economics 7(1), 85-101 101 mean values teus gdp nb gov dg1 acajutla 163017.125 2.224e+10 50.875 0 0 antofagasta 87538.25 2.3463e+11 77 1 0 arica 169788.625 2.3463e+11 69 1 0 balboa 2928101 3.3831e+10 961.5 1 1 buenaventura 783770.875 3.125e+11 219.75 1 1 caldera 180315.25 3.9631e+10 70.75 0.875 0 callao 1602763 1.6038e+11 402.375 0.75 1 corinto 86173.125 9653495299 38.875 0 0 coronel 248021.375 2.3463e+11 65.75 1 0.875 ensenada 136824 1.1088e+12 94.375 1 1 esmeraldas 69158.5 7.6771e+10 28.375 0 0 guayaquil 1322721.5 7.6771e+10 404 1 1 iquique 254012.875 2.3463e+11 151.75 1 0 lázaro cárdenas 903142.375 1.1088e+12 306.875 1 1 lirquen 190730.125 2.3463e+11 55.375 1 0 manzanillo 1839708.75 1.1088e+12 671.875 1 1 matarani 18952.5 1.6038e+11 18.75 1 0 mejillones 146695 2.3463e+11 97 1 0 paita 161617.125 1.6038e+11 67.5 1 0.375 puerto quetzal 313757.5 4.4081e+10 99.625 0.75 0 san antonio 968246.5 2.3463e+11 241.25 1 1 san vicente 482269.125 2.3463e+11 103.875 1 0 valparaíso 905290.375 2.3463e+11 205.375 1 1 source for all descriptive statistics: data eclac (2017), wb(2017), ais(2017); own elaboration abstract. latin american countries have historically had a strong dependence on trade, and are mostly characterized by being exporters of raw materials and importers of manufactured products. this fact has brought about a less negative impact of the w... keywords. port governance; maritime trade; latin america; panel data. jel classification. c33; n76; r42 european journal of government and economics volume 2, number 1 (june 2013) issn: 2254-7088 25 the commission’s internal conditions for social re-regulation: market efficiency and wider social goals in setting the rules for financial services in europe miriam hartlapp, social science research centre berlin and university of bremen, germany christian rauh, social science research centre berlin, germany abstract the european union is often considered as a prime example of a liberal regulatory state. we argue, however, that being limited to the regulatory policy does not prevent the european commission from pursuing political aims going beyond market efficiency. we draw up two ideal-type perspectives of market regulation – being either efficiency or equality enhancing – that differ systematically in terms of rationale, degree of intervention, patterns of stakeholder access and conflict within the regulator. we trace these aspects in three financial services initiatives on the registration and supervision of reinsurers, equal treatment in financial services and the regulation of consumer credit. our analyses suggest that there is scope for equality-enhancing re-regulation when proactive agents proceed decidedly on the basis of social-treaty concerns and frame regulatory beneficiaries as market participants as well as when they seek the redistribution of rights instead of resources. jel classification g28 keywords european union; regulation; financial services; equality enhancement acknowledgements the paper is part of a multi-annual, collaborative research project on ‘position formation in the eu commission’ being carried out at the social science research centre berlin. funding from the volkswagen foundation is gratefully acknowledged. more information on the project can be found at http://www.wzb.eu/en/research/civil-society-conflicts-and-democracy/ position-formation-in-the-eu-commission. we are also indebted to our colleague julia metz for fruitful discussions. an earlier version of the paper was presented at the workshop entitled ‘the future of the regulatory state: adaptation, transformation, or demise?’ held at the bi school of management, oslo (15-16 september 2011). we thank the participants and especially david levi-faur for helpful comments. european journal of government and economics 2(1) 26 introduction limited possibilities for taxation and meagre budgets severely restrict the eu regarding the pursuit of wider social goals through governmental redistribution of existing resources. rather than as a positive state, the eu has been characterized as the leading example of a liberal regulatory state that focuses mainly on the correction of various market failures (majone, 1993, 1997). the fact that eu regulation emerged increasingly as a response to the privatization of utilities (grande, 2011) has added to the view that the union’s first and foremost regulatory aim is the efficiency of the single european market. however, the traditional welfare state at the national level has also partly relied on regulatory components (mabbett, 2009; leisering, 2011a) while the definition of the regulatory does not necessarily imply particular political goals (levi-faur, 2012: esp. 14-16). influential authors have identified a european social policy, although being of the regulatory type (leibfried, 2010; falkner et al., 2005). clearly, eu regulation covers increasingly broad social themes, including, for example, consumers, the environment and energy (knill et al., 2003; vogel, 2003; coen et al., 2008). moreover, the demand for european regulation with a more social face has increasingly been on the agenda since the failed referendums in 2005 and since the financial crisis that put the spotlight on the downsides of unfettered market forces. so while eu decision-making has a strong regulatory thrust, regulation is not automatically tied to the (political) ambition for liberalization, but may also be a vehicle for attaining distributive goals. consequently, it is relevant to ask for the conditions under which eu market regulation pursues social goals that go beyond mere market efficiency. we tackle this question with regard to the european commission. the long-term perspective of its regulatory staff, its unique access to broad policy expertise and, especially, its formal monopoly of legislative initiative all provide the commission with substantial influence over the contents of european regulation (princen, 2009; tsebelis et al., 2000). however, against the background of external constraints – most notably the provisions of eu treaties, as well as political preferences in the council of ministers and the european parliament – the commission must keep up the appearance of impartiality and thus has reason to mask the outright pursuit of social goals. following an ideal-type distinction between efficiency-enhancing and equality-enhancing regulation, we argue in section 2 that the commission’s ambitions can nevertheless be identified if we take the systematic differences between the regulatory processes into account. this claim is supported by comparing three short histories of drafting processes on financial services regulations presented in section 3. initially, the european financial services market could have been realized through liberalization and the simple application of the freedoms of movement enshrined in the eu treaties. however, especially since the publication of the eu green book on the financial services market in the late 1990s, european regulation of the sector abounds (posner et al., 2010: 400). in shifting from the coverage of purely commercial activities to retail financial services for private consumers, ‘the fsap [financial service action plan] represented a change in eu strategy away from market opening measures towards common regulatory measures’ (quaglia, 2010: 1007). these efforts may only be a consequence of further market integration, but they still make financial services a promising policy area for studying regulatory capacities that go beyond market efficiency and the regulatory processes that push the commission towards the pursuit of wider social concerns. we trace the theoretically developed characteristics with respect to three regulatory processes on financial intermediaries, equal treatment in service markets and consumer credit. hartlapp and rauh ● the commission’s internal conditions for social re-regulation 27 the article generates two expectations regarding the conditions under which the european commission places a more social european market on the legislative agenda. while its capacity remains limited to the regulatory section of the overall social policy spectrum, the commission can and does proactively redistribute rights 1) where this is pushed by specific internal actors who are in turn able to draw on a decidedly social-treaty base and frame the societal beneficiaries as market participants, and 2) where rights rather than resources are at stake, typically to the benefit of diffuse stakeholder interests. under these conditions, it becomes possible to propose equality-enhancing regulation that goes beyond the aim of more efficient competition. conceptualizing regulation: market efficiency and wider social equality this article seeks to contribute to the debate on whether and under which conditions europe’s markets can be embedded ‘within a broader set of social and political rules and cultural understandings that make them work not only more efficiently but also more equitably, with greater security for market and nonmarket participants and in tune with a variety of other social purposes’ (caporaso et al., 2009: 579). existing research on this question differs along two perspectives. firstly, it focuses on the implications of eu decisions for national welfare-state models. eu integration’s lack of social embeddedness is ascribed to the systemic features of supranational decision-making, which prevent the reproduction of welfare states at the supranational level (scharpf, 1999, 2010). our interest here is more circumscribed, focusing on the re-regulatory capacity of day-to-day policymaking in brussels. in other words, we accept the premise that the eu is a regulatory state with very limited capabilities for redistributing resources directly and that its regulatory instruments are constrained by the nature and interests of the inter-institutional process. however, any european market regulation must start with a policy proposal by the commission, which in turn defines the issues as well as the strategic preferences that govern inter-institutional negotiations (princen, 2009; tsebelis et al., 2000). so if commission proposals are not a priori equality enhancing when adopted by the commission, the chances are low that such a policy orientation will be produced later in the policy cycle. focusing on europe’s central agenda-setter is thus crucial for understanding the scope for and the degree to which eu regulation can be embedded in a social dimension. a second dividing line in the literature runs between arguments about the natural or spontaneous emergence of market embeddedness and contributions highlighting the necessity for political construction in the pursuit of distributive goals. the former perspective is often evoked in relation to ecj rulings in the area of antidiscrimination policies or cross-border health care (hervey et al., 2007; more critical höpner et al., 2012), but here we are concerned with intentional political construction. we start from the premise that market embedding is not an autonomous process, rather depends on specific conditions linked to decisive regulatory actors such as the european commission. these decisions already highlight the fact that the commission only occupies the middle section of a larger social policy continuum that spans from completely unfettered markets, on the one side, to an autocratic distribution of resources along social justice criteria, on the other. strictly speaking, only a section somewhere in between these extremes is filled with regulatory policy where authoritarian intervention is based on rules stated in general terms. but also within this section, the regulator will explicitly or implicitly decide on ‘who will be indulged and who deprived’ (lowi, 1964: 690-1). in this view, regulation makes a socially relevant difference (braithwaite et al., 2007) and the narrow section should already provide crucial insights concerning the conditions under which a policy outcome will be situated more towards one or the other end of the entire social policy continuum. even when we only look at regulatory means, the outcome can be markets of very european journal of government and economics 2(1) 28 different social quality. we start from a rough distinction between two ideal-type perspectives on the organization and purpose of markets as a mode of social interaction: they may be either seen to produce the optimal allocation of resources in the equilibrium of economic supply and demand or as a socially constructed mechanism that balances competing values and demands. the classical perspective of welfare economics starts from the utility of individual actors in the marketplace. it assumes that the socially optimal allocation of resources is attained when no individual can be made better off without disadvantaging others. such a pareto-efficient outcome is reached through the equilibrium of supply and demand, which in turn can only be achieved in a perfectly competitive market where the individual market participants try to maximize their individual welfare (stiglitz, 1986). but even from this perspective, some regulation is required to preserve the market (weingast, 1995), for instance with respect to property rights or contract certainty. in practice, markets are hardly commensurate with a model of perfect competition, for example where information is asymmetrically distributed among the market participants (akerlof, 1970). thus, classical welfare economics considers market regulation to be warranted where markets fail, externalities accrue or other imperfections prevent optimal decisions on the part of individual market participants (see, e.g.,majone, 1996: esp. 28-30). in this perspective, regulation is seen only as a second-best solution to the outcomes a fully competitive and undistorted market would produce. a second perspective on market regulation criticizes this purely economic view from normative points of departure (see prosser, 2006 or; mcvea, 2005). approaches in this vein argue that market efficiency is qualitatively different from other, equally legitimate purposes of social interaction, such as social justice. here, regulation does not target perfect competition in markets, but rather aims to alter their very quality by changing the rules under which individual transactions will be carried out. instead of guaranteeing the mere pursuit of individual welfare, regulation addresses misallocation from a moral point of view, and is thus seen as a ‘balancing of competing values setting out the sort of society we wish to live in’ (prosser, 2006: 375). this is consistent with the distinction between economic and social regulation that has figured prominently in the us debate on regulatory market interventions (cf. reagan, 1987: esp. ch. 2). in the context of the eu, this approach has also been described as ‘market cushioning policies’ (sbragia et al., 2008: 133; sbragia, 2000; similarly, ‘market shaping’ quaglia, 2010; ‘extended regulation’ leisering, 2011b: 6). while economic regulation aims at freeing competitive forces, social regulation is primarily motivated by problems that often cannot be solved by an increase in economic competition alone. rather than targeting the individual pursuit of welfare, social regulation aims at protecting broader societal groups from systematic disadvantages they may encounter in competitive markets (ramsay, 1995). we argue that the two perspectives should result in different models of regulation. of course, in practice regulators often simultaneously face demand to create efficient markets that minimise burdens on industry and to create societies that are more just, equal and socially cohesive – and thus differ from the outcomes produced by market forces alone. in practice, thus we expect to see various mixtures “of interconnected and interdependent social and economic values, and that the distinction […] lies in the primacy of the values it is designed to advance and the purpose it is designed to achieve” (windholz et al., 2013: 25), yet, while analytical boundaries might be very fuzzy at times deriving ideal-types allows us to point out more specific empirical implications which ultimately lead us to a finergrained perspective on the european commission’s choices and options in embedding european markets in wider social concerns. the discussion so far yields the following two ideal types of regulation: in the first perspective – termed ‘efficiency enhancing’ – regulation is perceived as necessary to increase the immediate functioning of markets by countering market hartlapp and rauh ● the commission’s internal conditions for social re-regulation 29 failures or imperfections understood as non-optimal transactions. regulation should be expected where markets face a lack of competition due to monopolies, negative externalities or informational asymmetries. the beneficiaries of regulation are individual market participants. in the second perspective – termed ‘equality enhancing’ – market regulation is perceived as necessary to deliver broader societal goals and operates with a more long-term perspective on market gains. such regulation is motivated by other goals than purely increasing competition. structurally weaker parties in market transactions are the beneficiaries of regulation. most importantly, distinguishing these two ideal-types is at odds with the seminal perspective on the regulatory state, which is most often portrayed as the polar opposite of the positive (welfare) state (majone, 1997: 149; levi-faur, 2012: 1215). in this account, the main function of the positive state is keynesian macroeconomic stabilization and redistribution. majone’s regulatory state by contrast puts administrative and economic performance first – it is ‘efficiency enhancing’. in between these extremes, the ‘equality-enhancing’ perspective we develop here implies that there is a socially and analytically relevant middle ground. and this middle ground should differ from both the liberal regulatory state and the positive state not only with respect to long-term outcomes ‘but with regard to access, procedures and organization’ (leisering, 2011b: 258, original emphasis). however, what follows from majone’s original distinction for a ‘state’ model that sits in between, and that starts with a clear (re)distributive function operates through regulation only? we argue that if the goals of regulation differ, as the original distinction of the ideal types suggests, then also the process of rulemaking at the eu level should deviate from the typical regulatory state model in at least four respects. first, in his seminal conceptualization of the regulatory state, majone argues that redistributive consequences result from ‘potential policy constraints rather than policy objectives’ for regulators (majone, 1997: 162). in this efficiency-enhancing world, policy-makers try to avoid costs for any market participant by striving for a solution that is beneficial to all. in contrast, the equality-enhancing perspective implies that actors will focus on the creation of benefits for particular market participants that may or may not imply costs for others – this is in line with taxing and redistribution under the positive state. rather than a constraint, (re)distributive consequences should become the rationale of regulation in an equality-enhancing model. secondly, the envisaged structure of the regulatory interventions should differ. in an efficiency-enhancing world, and in line with the goal of individual choice, as many decisions as possible are delegated to the market participants themselves, so that ‘indirect government’ best describes the mode of regulatory production as well as the regulatory contents (majone, 1997: 146-8). by contrast, the equalityenhancing perspective implies that markets produce skewed outcomes so that the regulators will take as many decisions as possible themselves and try to set conclusive rules for the issue in question. thirdly, the two ideal types should differ with respect to the underlying conflict dimension and the related integration of external stakeholders into the regulatory process. especially in the eu multilevel system, stakeholder integration is best understood as support from allies. the positive state is typically associated with corporatist patterns structured around class conflict. capital and labour are granted access at the level of organized interests (cioffi, 2009: 244). under such representational monopolies, access to the regulator is biased (schmitter et al., 1979). majone’s regulatory state, by contrast, moves away from this central political dynamic and implies competition among single-issue interest groups. following pluralist theory (truman, 1951), regulators grant equal access to the different sides of the debate and act as neutral arbiters searching for the equilibrium solution. the european journal of government and economics 2(1) 30 influence of external stakeholders might be biased, but the bias should not result from prefixed access patterns, rather from competition between the interests. under the equality-enhancing model, stakeholder access should cease to be neutral without, however, focusing only on the conflicts underlying the capitalistlabour divide. where normative imbalances among pluralist contending interests are the starting point of regulation, regulators should grant privileged access for those groups whose interest they see specifically at stake. the equality-enhancing model is thus more consistent with the neo-pluralist view on interest-group integration (cf. dunleavy et al., 1987). lastly, the two possible worlds of regulation should differ concerning the way actors within the commission interact. as the eu’s agenda-setter, the commission is not a unitary actor when proposing regulation, rather different interests prevail. the position finally proposed is the outcome of a long internal process characterized by multiple and often conflictual interactions (hartlapp et al., 2013). in the seminal conception of the regulatory state, governmental actors ‘usually cast their arguments in the language of 'regulatory science' rather than in the more traditional language of interest or class policies’ (majone, 1997: 157). arguments should focus on technically ‘best’ solutions, and actors should adhere to ‘a problem-solving rather than a bargaining style of decision-making’ (ibid: 162). by contrast, the equality-enhancing perspective emphasizes normative differences in the quality of markets and should thus be much more prone to political conflict and more fundamental political exchange among actors. table 1: ideal-type models of regulation in between unfettered markets and the positive welfare state unfettered markets efficiency-enhancing regulation equality-enhancing regulation positive state rationale competition market failure redistribution of rights redistribution of resources intervention none least intrusive rule-making encompassing and conclusive rule-making taxing/spendi ng stakeholder access irrelevant pluralist: self-selection neo-pluralist: privileged access corporatist: class-based representation conflict structure price expertise driven norm driven party driven summarized in table 1, we can now identify equality-enhancing regulation as a distinct option for rule-makers. it is located in between unfettered markets and the idealized positive welfare state and, more importantly, it distinguishes itself from the efficiency-enhancing regulation usually associated with the regulatory state not only in terms of policies proposed, but also with regard to the regulatory processes at work. on the basis of these procedural differences, we can analyse whether the european commission actually pursues equality-enhancing regulation and what scope there is for the social embeddedness of europe’s internal market. tracing equality-enhancing regulation in the european commission the outcome we are interested in is the extent to which a policy is equality enhancing, whereas the market-failure perspective in the classical regulatory state model of the commission serves as our baseline expectation. in other words, our empirical starting point is the commission’s regulatory output and, more specifically, the degree to which a commission proposal deviates from the regulatory status quo in the name of efficiency or social equality. in order to highlight how the regulatory processes leading to these outcomes differ systematically we selected three cases from a broader project that studies decision-making within the commission on the basis of semi-structured interviews hartlapp and rauh ● the commission’s internal conditions for social re-regulation 31 conducted in brussels between 2006 and 2009 with involved commission officials from different directorates-general and hierarchical levels.1 firstly, the proposed directive on financial intermediaries, by regulating registration and consumer rights, seeks to establish an efficient single market for insurers distributing financial products on their own account. secondly, the proposed directive on equal treatment in the access to and supply of goods and services for men and women and for all persons irrespective of religion or belief, disability, age or sexual orientation addresses the disadvantages facing specific groups in financial markets, for example with respect to insurance contributions and benefits or the conditions for loans. and, thirdly, the final act seeks to protect vulnerable consumers by regulating creditor obligations concerning the advertisement, conclusion and termination of consumer credit agreements. whereas the first proposal follows an efficiency-enhancing logic, the second two cases lean towards the equality-enhancing side. tracing the three regulatory processes, we work our way through the four dimensions of the two ideal types derived above. while the very notion of ideal-types excludes the possibility of finding them empirically in pure form, we should find evidence that decision-making in the commission deviates from the classical regulatory-state model by following the logic of distributing political benefits, neo-pluralist and intentionally skewed interest group access, as well as substantial internal conflict instead of efficiency-enhancing problem-solving. the models in practice: drafting european regulations on financial services a first case that provides insights on the commission’s internal conditions for reregulation is the regulation of reinsurers and financial intermediaries as institutions that redistribute financial products on their own account – such as travel agencies or car dealers.2 although a forerunner directive with a narrower scope as well as a non-binding recommendation already existed, the regulatory logic was one of interface management rather than common-market regulation. the size of national markets differed, with in some cases more than 50 percent of insurance being distributed via intermediaries (the figures are especially high in belgium, the netherlands, the uk and ireland, but are low in denmark). national rules varied ‘with regard to the access and purchase of the activity, in particular with regard to financial capacities [and] professional relations’ (com109), de facto confining intermediaries to their national markets. especially with respect to consumer interests, member states typically had national rules in place. accordingly, commercial actors faced substantial uncertainty regarding which rules to apply in cross-border markets (dassesse, 1997:108). commissioner monti first announced the new regulation in february 1997 (ae, 1997a) and the proposal was adopted in 2000. for dg markt, the rationale was to boost the retail market in insurance for private individuals through further liberalization, much along the lines this had been achieved for the wholesale market of large industrial and commercial risks. to this end, one part of the proposed directive established a single registration system in the member state of origin which harmonized the recognition of insurance agents and brokers throughout europe (a regulatory logic also applied to other directives, quaglia, 2010: 1030) and went hand in hand with professional requirements and supervision through the country of origin. a drafting official summarized this rationale as follows: ‘we did it, because it was worth for the industry and also to enforce the internal market. it was a very heavy case, where not a lot of work was 1 in order to preserve interviewee anonymity, we refer to the interviews as ‘com1’, ‘com2’, etc. 2 proposal for a directive of the european parliament and of the council on insurance mediation (com(2000)511). european journal of government and economics 2(1) 32 necessary from the internal market viewpoint’ (com109). the other group of provisions considered a necessary ingredient for making retail markets for intermediaries function efficiently was to strengthen the position of consumers by introducing information and disclosure rights as well as facilities for complaints: ‘[i]t will increase the choice of insurance products available to customers and help ensure they can trust the advice they are getting from intermediaries’ (commissioner bolkestein, cited in: ae, 2002, likewise com109). clearly, in this case, consumer protection was not sought as a right per se, but to prevent the failure of a common market for financial mediation and thus qualifies as efficiency enhancing. different sources provide evidence that drafting did not take place behind closed doors and that external stakeholders were relevant in explaining both the choice of a binding instrument with a broad scope that included small intermediaries, as well as the other substantial provisions of the proposal (ae, 1997a, com109). following the efficiency-enhancing model, dg markt consulted broadly not only across several member states and the ep, but also across stakeholders from the financial services industry, notably the european insurance committee (eic), the association of european cooperative insurers (aeci), intermediaries such as the international association of insurance and reinsurance intermediaries (bipar) and customers represented by the european bureau of consumers’ unions (beuc; com109). these single-issue interest groups finally all found their positions represented in the proposal. the binding provisions clearly open the market for cross-country provision of services and thereby foster competition between professional insurance intermediaries more strongly than earlier, nonbinding recommendations. and they do so ‘within a clear framework of common rule to protect the consumer’ (beuc director jim murray, cited in ae, 1997a; similarly bipar in ae, 1997b). drafting inside the commission stretched over a period of five to six years and different portfolios were involved (com109). although dg sanco, especially, was reported to have argued for consumer interests to be reflected, the actors involved did not perceive the interaction as being conflictual and reported that all internal comment was taken on board (com109). in combination with the obviously negligible interest at the political level for this rather technical directive, the process followed a problem-solving logic. the second case is the promotion of equality of opportunity for disadvantaged groups in services, especially financial services. traditionally, at the eu level antidiscrimination regulation had been developed as part of labour law. legislation on actuarial factors differed starkly across member states, with the unisex principle mainly in place in the scandinavian countries, france and slovenia. the uk also had unisex tariffs for private health insurance, whereas car coverage showed sharp differences (ae, 2004). after the amsterdam treaty had introduced a new basis for legal action with article 13, the commission proactively sought to extend equivalent principles to other areas, including the financial services market. the commission proposal3 is equality enhancing in that it does away with practices of unequal treatment in terms of access to the insurance cover provided or in the level of benefits paid out. such practices – most pronounced in motor, health and life insurance – are efficient from a purely market point of view, but they disadvantage societal groups on the basis of factors that lie beyond personal control. against this background, in 2003 the commission had proposed a directive implementing the principle of equal treatment between women and men regarding access to and supply of goods and services that was adopted as directive 2004/113/ec. in 2008, a more far-reaching proposal was drafted by dg empl. if it 3 proposal for a council directive implementing the principle of equal treatment between women and men in the access to and supply of goods and services (com(2003)657). hartlapp and rauh ● the commission’s internal conditions for social re-regulation 33 becomes law, this proposal would extend the quest for equal treatment between the sexes in services to all persons, irrespective of religion or belief, disability, age or sexual orientation.4 taking the effect of the directive already adopted and the pending proposal together would mean that, in principle, insurers and banks will no longer be able to use age, sex or disability in a way that discriminates against groups with these particular characteristics. this will increase costs both for insurers and for those consumers who have hitherto benefitted from the unequal market relationship. moreover, the very fact that the commission has intervened to explicitly extend the market to those who were previously excluded by law or in practice underlines the goal of creating more equal markets above and beyond what could be considered economically efficient. with respect to the rationale characterizing the regulatory process in this second case, we find that while market inefficiencies were invoked in the debate, the core ambition of the drafting dg was clearly to alter the quality of this market in accordance with a political reasoning targeting greater equality. enhancing the efficiency of markets by responding to changing environments and countering unfair practices was certainly on the agenda of the commission, especially in view of persisting implementation difficulties (com125, ae, 2007). however, we also note clearly voiced criticism that under the envisaged regulations the opposite outcome would be the case, because premiums would ultimately be forced upward (minder, 2003b; parker, 2003). besides, the goal of increasing equality in markets was also explicitly voiced by the commission. the proposal is described in official documents as a necessary ‘signal of solidarity between women and men in european society’ (secrétariat général, 2003:21) and as the ‘result of a real moral and legal obligation’ (proposal, p.4). commissioner diamantopoulou stressed that there were ‘political reasons for not accepting gender as an element justifying a differentiation of the treatment of individuals in the access to the supply of goods and services’, as well as the signalling function of the act for european society (secrétariat général, 2003:21). as suggested by our ideal-type model, this case was characterized by biased stakeholder access to favour presumably weaker parts in the regulated market relationships. in the run-up to the negotiation process in 2003, dg empl had very actively used and encouraged changes in interest-group organizations to create a broad front against opposition from those who wanted to keep more expensive discrimination grounds out of the discussed acts. already prior to concrete drafting, the commission had incentivized existing and scattered ngos representing minorities to join forces in an umbrella organization which was not only consulted but also substantially influenced the legislative proposal. ‘it was certainly strategic to involve civil society groups more than social partners […] we saw the potential victims of discrimination as being people who were at risk of some sort of social marginalization as a result of discrimination, and they are the ones that tended not to have a voice in social partnership’ (com151). industry, by contrast, was consulted only to provide information that would allow the commission to follow its goal of greater market equality. to gain a better understanding of where the technical requirements of risk-based assessments for capital adequacy end and individual business interests begin, the drafting dg empl consulted industry early on in the drafting process for the 2008 proposal (com42 and com19). on the basis of the acquired knowledge, the commission was able to argue that the question as to whether existing rules and practices had to be changed depended on the product and on whether justification could be provided ‘on the basis of accurate and up-to-date data’ (commission of the european communities, 2008: 42). the proactive and selective distribution of access for interests from different 4 proposal for a council directive on implementing the principle of equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation (commission proposal com(2008)426). european journal of government and economics 2(1) 34 societal segments by dg empl clearly contradicts the presumed neutrality of the regulator that a purely efficiency-enhancing model would imply. finally, with respect to the conflict structure inside the commission, both the decision to have equal treatment legislation on financial services and the more concrete implications this would have for insurers in their actuarial practices were debated among central actors inside and outside the commission, reflecting substantial differences in the underlying understandings of markets. inside the commission, dg markt, especially, defended the positions of industry – ‘various financial sector federations’ (com42) and those member states where ‘the practice of differentiating insurance costs [...] were very, very strong. uk, ireland, but also germany’ (com21) – by trying to carry out ‘a damage limitation exercise’ (com19). dg entr was fundamentally opposed to having a binding act at all, and especially the broad 2008 directive, because they did not consider this a regulatory issue to be tackled at community level and feared costs for enterprises through the intervention (com81). for these actors inside the commission, the ideal regulatory position was for ‘insurance to be [kept] completely out’ (com21). in contrast, dg empl, holding the pen and largely acting on normative grounds, was strongly in favour of the acts (com21, commissioner diamantopoulou in minder, 2003a), but had to overcome opposition from the critical portfolios inside the commission as well as from opposing member states and organized interests. the third case is the initiation of consumer credit regulation proposed by the commission in 2002.5 supranational regulation had existed beforehand, but it had left member states with the possibility of protecting their consumers unilaterally, so that myriad different regulations on consumer credit persisted in europe. these barriers became increasingly relevant as consumer credit became the ‘lubricant of economic life’ (com89), amounting to 7 percent of the eu-15 gdp and exhibiting an upward trend (com119, commission of the european communities, 2002: 3). for the commission’s internal market directorate-general (dg markt), differing national regimes hampered growth and thus needed to be removed (com119, com89). however, unlike most other regulations on financial services, it was not dg markt but the newly established consumer-policy dg sanco, that was assigned responsibility for the regulation, and the latter dg pursued a rationale that went way beyond the facilitation of cross-border trade. in fact, dg sanco officials aimed at protecting the ‘weak consumer’ (com89). drafting was based on the assumption that the consumer was not able to fully grasp the implications of credit agreements and thus needed to be protected against the exploitation of that very fact by the banking or retail industry. this position was informed by observing individual cases of immensely over-indebted consumers, but also by the hype surrounding ‘behavioural economics’ at the time (com89). this resulted in the redistributive ‘ambition […] to create a very comprehensive, very exhaustive consumer credit regulation which would be burdensome for industry’, and dg sanco officials transferred the most stringent national regulations they could find to the proposal (com89), thereby increasing consumer protection rather than harmonizing at the level of the lowest common denominator. dg sanco’s equality-enhancing approach can be highlighted by means of some of the key regulatory provisions, in which especially the principle of responsible lending was contested. the proposal fully obliged creditors to assess before the agreement whether the consumer was actually able to honour the credit obligations (commission of the european communities, 2002: article 9). the banking industry 5 proposal for a directive of the european parliament and of the council on the harmonization of the laws, regulations and administrative provisions of the member states concerning credit for consumers – com(2003)443. this proposal was intended to repeal the existing supranational law in the area, as entailed in directive 87/102/eec. hartlapp and rauh ● the commission’s internal conditions for social re-regulation 35 was shocked (esbg, 2003: 7-9), whereas consumer associations were highly satisfied. in addition, the responsible lending principle stirred up significant internal conflict with dg markt (com89; com90). another heatedly contested issue was the way in which the total costs of credit should be communicated to the consumer, encapsulated in the annual percentage rate (apr). compared to the regulatory status quo, dg sanco’s proposal included a range of additional cost elements such as notary fees or even insurance premiums, meaning that creditors would have to disclose costs that went beyond their control. this rule, in particular, highlights the fact that dg sanco was pursuing a regulatory approach involving direct and conclusive rules governing the market. the banking industry defied this as ‘unworkable’ (esbg, 2003: 10), and was internally supported by dg markt (com89). finally, for credits linked to particular consumer goods, the proposed regulation would render creditors liable for defective products purchased on credit (commission of the european communities, 2002: article 19) – a regulatory approach borrowed from the uk consumer credit act which was considered ‘totally outlandish’ in other eu states (com89). the banking industry ‘strongly opposed’ it (esbg, 2003: 12), again backed internally by dg markt (com119). but dg sanco stuck once more with the most stringent rule it could find amongst the member state regimes, did not try to balance competing interests and was applauded by the consumer side (euro coop, 2003; also beuc, 2002: 14-5). in sum, dg sanco proposed a regulation that was geared towards equal protection for consumers in the european market. however, this regulation could only come about because inefficiencies had persisted in that market beforehand. these provided the basic justification for regulation, but the intentional focus on weak market participants and dg sanco’s need to demarcate its competences from those of the internal market dg resulted in a proposal that clearly aimed at redistributing rights in line with an equality-enhancing motivation. not least, the high level of protection entailed in the regulation was driven by the political aim of protecting consumers in credit markets. merely enhancing efficiency by removing regulatory barriers would ‘deliver a bad message’ politically and would be ‘a very difficult position for the commission’ to be in (com119). the clear focus on only one societal segment was also mirrored in interest-group access. it is important to note that stakeholder meetings on the issue indicated a ‘clear desire’ from the consumer associations (com111) as well as ‘unanimous’ opposition from the industry side. in response to this outright industry disagreement, ‘further consultation was entirely avoided’ by dg sanco (esbg, 2003: 2-3), so that ‘the industry felt that it had been very much disregarded in the process’ (com119). this intentionally skewed interest-group involvement also affected internal interaction with dg markt, which ‘had more the ear […] to the banking sector’ (com111). dg markt officials felt that banking interests were ‘not being properly consulted’ (com119), which pushed the directors-general of both dgs to the drafting table early on and resulted in a blockade of the formal inter-service consultation. as in the earlier case, these more fundamental questions were subject to exchange among commissioners about the quality of markets, so that ‘in the end it [was] a political decision’ (com89). as predicted by the equality-enhancing model of regulation, redistribution of rights among producers and consumers was an aim rather than a constraint, detailed rules rather than self-regulation along predefined objectives were prescribed, and a neo-pluralist style of communicating with external stakeholder contacts coincided with a political bargaining process among the relevant actors within the commission. european journal of government and economics 2(1) 36 conclusion: social equality through the market? the case studies emphasize that regulatory decision-making in the eu can follow an equality-enhancing type of regulation and thereby take an identifiable middle ground between traditional welfare-state models and purely efficiency-focused regulation. against this view, the social embeddedness of europe’s internal market cannot be considered as automatic as contemporary writing often seems to imply. instead, our analysis suggests that market retrenchment at the supranational level depends on the outcome of political struggles across, and especially within the regulatory institutions of the eu. where the regulator pursues social equality rather than solely focusing on market efficiency, the regulatory processes differ strongly in terms of their basic rationale, the chosen intervention, the type of stakeholder access and the internal conflict during rule formulation. initially, prevailing inefficiencies in the european single market were the starting point of rule-making in all three cases examined. and in the two equality-enhancing processes, the favoured societal group had to be framed as active participants in the regulated market. but rather than addressing such market failures by means of the least intrusive rule-making geared towards self-regulation, the regulators responded with encompassing and conclusive rules. furthermore, the respective leading commission dgs fostered and then counted on the support of consumer associations or a broad coalition of anti-discrimination ngos and intentionally disregarded the input of other affected market stakeholders. stakeholder access followed a neo-pluralist pattern and deviated from unbiased access in the classical conception of the regulatory state, but also from the corporatist pattern ascribed to the positive state. and, finally, the equality-enhancing processes led to strong internal opposition from other commission dgs that would have preferred less intrusive regulatory measures. several implications follow for the broader question as to under which conditions market regulation in the eu pursues wider social goals. initially, we noted that both the equal-treatment and the consumer-policy cases were driven by the agency of actors within the commission. equality or efficiency enhancement is always pursued by some and not by others. to answer the question as to whether we can expect brussels to intervene only where markets fail or also where social interests demand regulation, we need a disaggregated view on europe's central regulator. on the one hand, this refers to entrepreneurial actors in the drafting dgs. but for both cases it was also important that the directorates-general held a decidedly social mandate due to their portfolios and could rely on treaty bases justified along equality-enhancing lines. delineating their own competences from traditionally efficiency-oriented dgs, political entrepreneurs within the lead departments explain the push for more equality, while the available institutional mandates explain how the aim of social equality could be pursued. this finding suggests that a politicization of the selection of key personnel (hartlapp, 2014, forthcoming) and the layout of portfolios within the commission may significantly advance the social embeddedness of european integration. but against this hope, it has to be noted that the apparent need to frame beneficiaries of equality-enhancing regulation as market participants still limits the scope for european re-regulation. compared to the positive state at the national level, where social policy often targets those not able to participate in the marketplace, doubts remain as to whether the regulatory processes at the supranational level can really work to the benefit of market outsiders. related to this, one has to keep in mind that the equality-enhancing model is also limited to the regulatory section of the overall social policy spectrum. in other words, equality enhancement in the commission context means a redistribution of individual rights rather than resources. in a positive account, many see the fundamental rights agenda as completing the political union (caporaso et al., 2009). given the eu’s institutional difficulties in enacting actual resource transfers, aggravated by the hartlapp and rauh ● the commission’s internal conditions for social re-regulation 37 increasing heterogeneity of political preferences in the council, individual rights provide much better opportunities for social embeddedness because they cut across traditional conflict lines. what is more, the proactive redistribution of rights can help to make the eu ‘loveable’ (to allude to the famous phrase by delors that ‘you cannot love a market’). however, from a more critical perspective, the shift towards regulating individual rights calls into question the traditional logic of welfare provision in the ‘positive’ state through social security, corporatist actors and macroeconomic stabilization. individual rights increase the autonomy of citizens in a market setting, but they may endanger the notion of solidarity across social groups. in times when most western societies are characterized by ever-growing disparities, this might be a fatal development even more so where the emphasis on individual rights undermines solidarity institutions that have evolved historically, for example in national labour or tax law (scharpf, 2010; höpner et al., 2012). in sum, this article highlights the fact that political actors can exploit european regulation as a means to achieve broader societal goals, but are forced to take a detour via market efficiency. put differently, while social policy in the positive state sought to decommodify individuals from the market, eu regulatory policy reflects actor interests in order to change the process and outcome of existing market transactions. in conclusion, there is scope for a social embedding of the european internal market in principle, but it is constrained by its institutional foundation, the existence of proactive agents within the commission, a focus on market participants and the mere redistribution of rights rather than resources. references ae (1997a) 'eu/insurance: beuc and bipar call for european directive liberalising activities of insurance intermediaries', agence europe. ae (1997b) 'eu/insurance: commission announces the forthcoming presentation of a directive on insurance intermediaries bipar pleased', agence europe. ae (2002) 'eu/insurance: council adopts insurance broker directive', agence europe. ae (2004) '(eu) eu/employment council/social: political agreement anticipated on gender equality outside workplace public debate on social service of general interest', agence europe. ae (2007) '(eu) eu/social/equality: vladimir spidla launches european year for equal opportunities for all 2007', agence europe. akerlof, george (1970) 'the market for "lemons": quality uncertainty and the market mechanism', quarterly journal of economics 84(3): 488-500. beuc (2002) credit for consumers beuc position. proposal for a directive of the european parliament and of the council on the harmonisation of laws, regulations and administrative provisions of the member states concerning credit for consumers. com (2002) 443. bureau européen des unions de consommateurs. brussels. braithwaite, john, cary coglianese and david levi-faur (2007) 'can regulation and governance make a difference?', regulation and governance 1(1): 1-7. caporaso, james a. and sidney tarrow (2009) 'polanyi in brussels: supranational institutions and the transnational embedding of markets', international organization 63(04): 593-620. cioffi, john w. (2009) 'adversarialism versus legalism: juridification and litigation in corporate governance reform', regulation and governance 3(3): 235-58. european journal of government and economics 2(1) 38 coen, david and mark thatcher (2008) 'network governance and multi-level delegation: european networks of regulatory agencies', journal of public policy 28(01): 49-71. commission of the european communities (2002) proposal for a directive of the european parliament and of the council on the harmonisation of the laws, regulations and administrative provisions of the member states concerning credit for consumers. com(2002)443. brussels. commission of the european communities (2008) commission staff working document. proposal for a council directive on implementing the principle on equal treatment between persons irrespective of religion or belief, disability, age or sexual orientation impact assessment. sec(2008) 2180. dassesse, marc (1997) 'comments on the european commission green paper 'financial service: meeting consumers' expectations'', journal of financial regualtion and compliance 5(2): 107-14. dunleavy, patrick and brendan o'leary (1987) theories of the state. houndmills, basingstoke: palgrave macmillan. esbg (2003) on the commission's proposal for a directive of the european parliament and of the council on the harmonisation of laws, regulations and administrative provisions of the member states concerning credit for consumers com (2002) 443 final, european savings banks group. brussels. euro coop (2003) eurocoop position paper on the revision of the consumer credit directive com(2002)443, european community of consumer cooperatives. brussels. falkner, gerda, oliver treib, miriam hartlapp and simone leiber (2005) complying with europe. eu minimum harmonisation and soft law in the member states. cambridge: cambridge university press. grande, edgar (2011) 'de-regulation and re-regulation of public utilities: the new regulatory state in the european system of multi-level governance', in lutz leisering (ed.) the new regulatory state. regulating pensions in germany and the uk. basingstoke: palgrave macmillan, pp. 57-79. hartlapp, miriam (2014, forthcoming) 'politicization of the european commission: when, how and with what impact?', in michael w. bauer and jarle trondal (eds) the administrative system of the european union. basingstoke: palgrave. hartlapp, miriam, julia metz and christian rauh (2013) 'linking agenda setting to coordination structures: bureaucratic politics inside the european commission', journal of european integration 53(4): 425-441. hervey, tamara and louise trubek (2007) 'freedom to provide health care services in the eu: an opportunity for ‘hybrid government’', columbia journal of european law 13: 623-47. höpner, martin and armin schäfer (2012) 'embeddedness and regional integration.waiting for polanyi in a hayekian setting', international organization 66(3): 429-55 knill, christoph and andrea lenschow (2003) 'modes of regulation in the governance of the european union: towards a comprehensive evaluation', european integration online papers (eiop) 7(1). leibfried, stephan (2010) 'social policy. left to the judges and the markets?', in helen wallace, mark pollack and alasdair r. young (eds) policy-making in the european union. oxford: oxford university press, pp. 253-81. leisering, lutz (2011a) 'transformations of the state: comparing the new regulatory state to the post-war provider state', in lutz leisering (ed.) the new hartlapp and rauh ● the commission’s internal conditions for social re-regulation 39 regulatory state. regulating pensions in germany and the uk. houndmills, basingstoke: palgrave macmillan, pp. 254-74 leisering, lutz (2011b) 'transformations of the state: comparing the new regulatory state to the post-war provider state', in lutz leisering (ed.) the new regulatory state. regulating pensions in germany and the uk palgrave macmillan. levi-faur, david (2012) 'states making & market building for the global south: the developmental state vs. the regulatory state?', jerusalem papers in regulation and governance. working paper no. 44 (july 2012). lowi, theodore j. (1964) 'review: american business, public policy, casestudies, and political theory', world politics 16(4): 677-715. mabbett, deborah (2009) 'supplementary pensions between social policy and social regulation', west european politics 32(4): 774 91. majone, giandomenico (1993) 'the european community between social policy and social regulation', journal of common market studies 31(2): 153-70. majone, giandomenico (1996) regulating europe. london: routledge. majone, giandomenico (1997) 'from the positive to the regulatory state: causes and consequences of changes in the mode of governance', journal of public policy 17(02): 139-67. mcvea, harry (2005) 'financial services regulation under the financial services authority: a reassertion of the market failure thesis?', cambridge law journal 64(02): 413-48. minder, raphael (2003a) 'gender banned from use in setting premiums', financial times. minder, raphael (2003b) 'insurers to fight brussels over bar on sex discrimination', financial times. parker, george (2003) 'eu targets sex bias in media and insurance: ambitious move to enshrine 'respect for human dignity' could force up premiums', financial times. posner, elliot and nicolas véron (2010) 'the eu and financial regulation: power without purpose?', journal of european public policy 17(3): 400-15. princen, sebastiaan (2009) agenda-setting in the european union. houndmills: palgrave. prosser, tony (2006) 'regulation and social solidarity', journal of law and society 33(3): 364-87. quaglia, lucia (2010) 'completing the single market in financial services: the politics of competing advocacy coalitions', journal of european public policy 17(7): 1007 23. ramsay, ian (1995) 'consumer credit law, distributive justice and the welfare state', oxford journal of legal studies 15(2): 177-97. reagan, michael (1987) regulation: the politics of policy. boston: little, brown and company. sbragia, alberta m. (2000) 'governance, the state, and the market: what is going on?', governance 13(2): 243-50. sbragia, alberta and francesco stolfi (2008) 'key policies', in elizabeth bomberg, john peterson and alexander stubb (eds) the european union: how does it work. oxford: oxford university press?, pp. 116-37. european journal of government and economics 2(1) 40 scharpf, fritz w. (1999) governing in europe: effective and democratic? oxford: oxford university press. scharpf, fritz w. (2010) 'the asymmetry of european integration, or why the eu cannot be a 'social market economy'', socio-economic review 8(2): 211-50. schmitter, philippe c. and gerhard lehmbruch (eds) (1979) trends towards corporatist intermediation. london: sage. secrétariat général (2003) minutes of the 1633rd meeting of the commission held in brussels (breydel) on wednesday 5 november 2003 (morning), pv(2003)1633 final. brussels. stiglitz, joseph e. (1986) economics of the public sector. new york: norton. truman, david b. (1951) the governmental process: political interests and public opinion. new york: alfred a. knopf. tsebelis, george and geoffrey garrett (2000) 'legislative politics in the european union', european union politics 1(1): 9-36. vogel, david (2003) 'the hare and the tortoise revisited: the new politics of consumer and environmental regulation in europe', british journal of political science 33(04): 557-80. weingast, barry r. (1995) 'the economic role of political institutions: marketpreserving federalism and economic development', journal of law, economics, and organization 11(1): 1-31. windholz, eric and graeme a. hodge (2013) 'conceptualising social and economic regulation: implications for modern regulators and regulatory activity', jerusalem papers in regulation & governance. working paper no. 49 (feburary 2013). can the modernisation of a public employment service be an effective labour market intervention? the hungarian experience, 2004-2008 european journal of government and economics volume 1, number 2 (december 2012) issn: 2254-7088 145 can the modernisation of a public employment service be an effective labour market intervention? the hungarian experience, 20042008 zsombor cseres-gergely, institute of economics, cers, hungarian academy of sciences, hungary* abstract the public employment service often delivers much of the employment policy including active labour market programmes in many member states in the eu, yet we know little about its effectiveness in general. this paper provides a quantitative assessment of the potential impact of the modernisation programme of the hungarian public employment service between 2004 and 2008. using data at the level of local offices, i calculate programme effects using a difference-in-difference estimator. results show that the programme has increased re-employment rates significantly, by 6%. the modernisation was thus a moderately effective but relatively inexpensive intervention, similar in terms of cost-effectiveness to the better active labour market programmes in hungary. jel classification j14; j23; j45 keywords older workers, youth employment, labour demand, public sector acknowledgements this work was developed further from the evaluation study delivered by budapest institute and ifua horváth&partners ltd. in fullfilment of an order within a framework-agreement by the national development agency to evaluate the second phase of the modernisation of the hungarian public employment service. the original project was led by ágota scharle and produced a report downloadable from http://www.budapestinstitute.eu. i would like thank the constructive comments that ágota scharle and gábor kézdi made on the first versions of the paper and the chance of thinking about the problem together. i would like to thank the support and the precise work of the people at the national employment office. i also appreciate the apt research assistance given by bálint szőke to a previous version of the paper. i am truly grateful to the referees for their useful comments which i hope helped the paper to improve. naturally, all remaining errors are mine. * address for correspondence : cseres-gergely.zsombor@krtk.mta.hu. european journal of government and economics 1(2) 146 introduction the public employment service (pes) is an important player in the labour market of almost all european countries, yet we know very little about the effectiveness of its operation and development. within the matching theories of blanchard and diamond (1989) and pissarides (2000), the aim of a pes is to facilitate the operation of the matching technology. this can involve activities such as administering payments to clients, negotiating and supervising job-search agreements, counselling, observing compliance and the administration of labour market programmes. the pes has always been an important building block of the european employment strategy and forms part of the europe 2020 strategy (european commission, 2010). the pes is a central institution in the so-called flexicurity framework and is assigned the role of supporting the transition between labour market states (wilthagen, 2008). although most member states spent more than 0.1 per cent of their gdp on the pes and administration in 2010 (oecd data, lmpexp), there is relatively scarce evidence on the effect of pes services on outcomes for the unemployed. this paper addresses the question of improving effectiveness by looking at the modernisation of a pes on one of its main outcomes, opportunities for re-employment. the evaluation of active labour market programmes is recognised today as very important and the field has accumulated an impressive body of evidence. these studies nevertheless address the issue of programme effectiveness for the most part and are strongly separated from the operations of the institution which delivers it. one exception is “services and sanctions”, an intervention delivered by and integrated with the pes and shown by kluve (2010) to be one of the most successful labour market programmes. the situation is no different in hungary, a country where the employment rate is among the lowest in the eu. with a budget of around huf 20 billion (around eur 70 million) per year, the hungarian pes serves 450000-600000 registered jobseekers, which is about 11 to15 percent or the active population. the institution has absorbed more than huf 10 billion in the 2000s in order to modernise its operation, but no quantitative assessment has been made about this process so far. the contribution of this paper to the scarce literature on pes effectiveness and also to the better developed literature on the evaluation of labour market programmes is delivering evidence using econometric techniques on the potential effectiveness gains through the modernisation of a pes. the questions i would like to answer are whether and to what extent the modernisation project contributed to increasing the chances of its clients in finding a job. in what follows, i use a difference-in-difference (did) econometric estimation method to estimate the possible effect of the developments between 2004 and 2008, using aggregate data relating to the local pes offices. first i briefly describe the development programme itself. then i move onto the theoretical and methodological considerations and the data used for the analysis. thirdly i describe the estimation results. finally, i put them in context and present some conclusions. evidence on pes effectiveness expenditure on labour market policy and the pes is relatively high in the eu, but also varies considerably, from 3.48 and 0.51 per cent of the gdp respectively in belgium to 1 and 0.1 per cent respectively in new member states (oecd data, lmpexp). despite such differences in spending, the need for a public employment service was mostly supported since the post-second world war period (baldwin 1951) to today (oecd, 2006), and reinforced after the onset of the economic crisis (ilo, 2009). critiques of the pes often refer to the ineffectiveness of the institution and such doubts are reflected in both spending on and organisation of the pes. doubts european journal of government and economics 1(2) 147 about pes effectiveness have their roots in theory. zweifel and zaborowski (1996) ask the question if public or private employment services are better. this theoretical analysis suggests that private agencies might fare better largely because the public agency is not allowed to be selective in its user base and not as worried about placements as private agencies. similar analysis also suggest that structure of the market for employment-related institutions matter for effectiveness (campens and tanguy, 2006). in the presence of privately-owned employment offices, otherwise positive performance of public ones can decrease. the best arrangement for the employment services is sought after in practice in various ways. denmark has reformed its entire pes by privatising a large part of it (koning 2004). although it does not directly recommend a complete privatisation, the oecd is advising its members to introduce market-based signalling systems so that contracting-out of services (bruttel, 2005) can help improving overall effectiveness (fay, 1997). a less disruptive step towards taking outcomes into account is the introduction of management by objectives (mosley, schütz, and breyer, 2001), a reform followed by many pes, including the hungarian one. management of performance is possible without a major change of organisational structure, but it requires constant and appropriate monitoring and evaluation of pes-related operations (oecd, 2005). very few pes monitor effectiveness directly (or at least openly), perhaps because such measurement has its own problems. among the few studies embarking on direct effectiveness measurement, we find vassiliev et al. (2006) and ramirez and vassiliev (2006). both use a form of production frontier methods to identify offices falling below the expected effectiveness threshold, and interpret finding offices below this as those potentially able to increase their output given their inputs. besides administering and delivering active labour market programmes, the pes often provides “services and sanctions, a category comprising all measures aimed at increasing job search effectiveness, such as counselling and monitoring, job search assistance, and corresponding sanctions in case of non-compliance” (kluve, 2010). although relatively few, 21 of the observed 137 studies look at this type of programme, the meta-analysis of the author yields robust results indicating that “services and sanctions” is one of the most effective and less costly types of almp. out of 3+5 specifications, services and sanctions always increases the likelihood of an evaluation finding positive treatment effects and almost always highly significantly. this performance is rivalled only by private sector incentive schemes, such as wage subsidies, while direct employment programmes for example perform much worse. these results are quite important from the point of view of the pes, as unlike other programmes, the performance of services and sanctions depend highly on its competencies. institutional background the hungarian pes plays the role of both an authority paying financial support to the registered unemployed and that of a supporting organisation by providing counselling and delivering various active measures to the clients. beside its core duties, it also performs many tasks somewhat loosely related to unemployment, including the administration of casual work, administering a large part of the rehabilitation process of disabled workers, or assisting public employment. resources of the institution did not however keep up with the proliferation of duties. as part of the austerity measures, the number of employees in local offices had begun to shrink already in 2006 and did not increase after the onset of the 2008 crisis either. while one officer attended an average of 206 clients in 2006, the same number had increased to 273 by 2009 (figures from direct hpes communication). the pressure on the pes was relieved to a certain extent by a series of development projects started in 2003 and still on going. its main aim was to carry european journal of government and economics 1(2) 148 out a general reform of operations in order to boost its performance in improving clients’ re-employment potential. at the end of the 1990s, the operation of local pes offices was characterised by very formal and unfriendly spatial arrangements, out-dated it infrastructure and further difficulties. officers were lacking a general overview of competency areas, while clients were not only served by but also dependent on them, interested mostly in collecting unemployment insurance and benefit payments. such a situation constrains the ability of the pes to improve outcomes for clients and was therefore important to change. the modernisation efforts have dealt with all the above-mentioned areas in 20, 60 and another 60 local offices in the three phases of the modernisation process, respectively. it also affected the national employment office, the methodological and coordination centre of the pes. here i focus in the so-called hrdop 1.2 measure (with a budget of huf 9.3 billion), which was the second among the three phases of modernisation. the aims of the development process are mapped onto projects – often overarching actual measures or programmes – whose combined effect is what i consider here as the intervention to be analysed. a total of 89 projects were targeted on introducing a new model of service provision with client profiling, internal remodelling of the local offices, installing self-help computer terminals, introduction of a quality assurance system, staff training and the introduction of an integrated information system. we expect that all of these had an effect on participating offices, while development of the integrated information system had an effect on the operations of the whole pes, regardless of programme participation. for we need a comparison group to measure programme effects, only outcomes for participating offices can be measured. the principle for assessing impact of the modernisation the current analysis aims at measuring the effect of the programme on a specific indicator – such as the re-employment chance of the registered unemployed – on the office-level, similarly to the structure put forward by nagy (2006) in an earlier proposal to estimate programme effects. being interested in the actual outcome of the programme, here i look at the average treatment effect on the treated (att). such an indicator and approach helps us to answer questions and clarify doubts raised by hárs and nagy (2009) in relation to the original indicators of the programme. because the analysis is focussed on local labour markets in which the pes offices are located, this measurement provides us with an estimate of the net effect of the programme, which is the combined effect of the direct effect and indirect effects. it also does not count with the possible displacement effects the programme generates. given however that the programme effects extend to every registered unemployed, this error is likely to be modest therefore the gross effect is likely to be not very different. i calculate programme effects using a difference-in-differences (did) method from raw data, later correct it using linear regression, first applied directly to the affected groups of offices, then using matching to homogenise them. the motivation for introducing a technique for sample homogenisation is that despite the indication of available information, the comparison group might be different from the treated group and matching is a preferred method of getting rid of a part of these differences explained by observable characteristics (heckman, ichimura and todd, 1998). the total of four versions of the estimates can be used to cross-check them, similarly to the ones in the dated but comprehensive evaluation of active labour market programmes in hungary (o’leary, 1998). in order to get rid of the time-invariant effects possibly correlated with programme participation, i have written the estimating equation in differences-form: european journal of government and economics 1(2) 149 ∆outit = τ + δpi + β∆xit + uit, where outit is the indicator of outcomes, τ is a constant measuring the autonomous rate of change in this, pi is an indicator of programme participation, xit is a set of variables indicating relevant observable characteristics of the local offices, while uit summarises characteristics that are not correlated with these observables. the relationship is defined over pes offices observed in different time-points, i being an index for an office, t being an index for a specific time period. the difference (∆) operator takes time-difference of a variable between the same month in the before and after period – note that office-specific fixed-effects has already been swept from this equation. our interest centres on δ, the coefficient on the pi indicator for programme-participation, which delivers the programme effect in this context. one can show that the equation in this form is a direct implementation of the did idea, generalised to the multiple-regression case. the first set of estimates come from ols regression and is based on the assumption that the participant and non-participant groups are similar. this assumption relies on information on programme design which attempted to select offices for modernisation from every county, without direct connection to the state of the particular office. variable ∆xit ensures that we take into account the differences developing over time between the participant and non-participant group of offices, and thus we do not confuse these with the effect attributable to the programme itself. not taking these into differences would give us the simplest raw did estimates of the programme effect. the estimation strategy explained above runs into difficulties if the assumption about the initial similarity of the groups fails and there are differences between participant and non-participant groups that are correlated with the ∆xit variables or with the indicator of participation. we can treat this problem if we have a sufficiently large number of variables at our disposal that can actually explain these differences to a great extent. given that that are detailed data on the clients in each office, this seems to be possible in our case. to produce the second set of estimates, i first perform propensity-score matching following the idea of rosenbaum and rubin (1983). this procedure amounts to predicting programme participation using a large set of pre-participation variables and use the predicted propensity to find observations for every programme participant that are close to it in some way. after this, i create a counterfactual realisation as a weighted average of the neighbours of the given office and compare the actual outcome to that. a substantial difference of this approach compared to the previous one is that the equation is estimated on a database containing data for the treated group and the counterfactual realisation generated from the control group. we get a reliable did estimate of the att under the current working assumptions by subtracting participants’ time-difference in outcomes from the same difference for their set of pairs assigned through matching, then averaging them (heckman, ichimura and todd, 1998). it is important to stay on the common support of the observable characteristics during matching, that is only those observations should be matched that actually have similar values of variables. although a similar effect could be achieved by homogeneising with ols – that is including the variables used for matching in levels to the estimating equation – the technical power of matching over ols is the flexibility that it introduces hidden non-linearity in controlling for differences. because of this however, the usual way of calculating standard errors for the estimator would be misleading. although there is no clearly preferred solution to this problem to my understating, bootstrap estimates are often used and this is what i am calculating. matching with simple averaging produces an estimate that is similar to a simple did estimate in that it does not take care of the over-time changes in characteristics between groups happening after the onset of the programme, that is european journal of government and economics 1(2) 150 without the ∆xit variables. if a control for over-time changes is needed, it has to be done before matching. as a result, we are not matching on original re-employment rates, but rather on residuals from a first-step regression similar to the one used in simple ols estimation but without the programme-participation indicator (this process is analogous to the residual-regression approach in the simple ols context). standard errors have to be calculated using bootstrap as before, for which the presence of the generated variables are another reason here. the above techniques allow us to control for differences between the treatment and the control group both before and after the introduction of the programme in observable characteristics. there can however still be correlation between observed and unobserved effects. i assume the lack of such effects which i cannot prove, only argue for later. should this argument fail in practice, the amount of the inconsistency in such cases depends greatly on the size and direction of effects governing such selection. data and the estimation method this study uses data primarily from the central database (integrated system, in hungarian: “integrált rendszer”, ir) developed within the framework of the hrdop 1.2 measure itself, containing individual data on the registered unemployed since 2000. in order to estimate the programme effect, i was provided with these data aggregated at the level of the local offices, done by the employment office. using spatial identifiers of the offices, i have attached to these records data relating to local labour markets, coming from the database of the hungarian central statistics office in individual settlements (t-star). individual-level data on registered clients in the ir contain information on sex, age, education, and occupational code of previous job as well as an indicator of disability. aggregate indicators calculated from these data play the role of x variables in the estimating equation, characterising the pes offices (with their postprogramme values) on the one hand as well as the role of controlling for initial observable differences between participant and non-participant groups (with their pre-programme values) in the matching process on the other. the indicators are all defined as the share of a particular type of registered client within all registered clients. in the case of the registered unemployed staying in touch with the local pes office, we know the direction of exit at the end of the registered status. the possible directions of exit are the following: (1) employment (open market); (2) public works; (3) supported employment (various forms of wage subsidy); (4) training; (5) not known due to lack of cooperation with the pes. the share of registered clients exiting towards either of these directions is an estimate of exit probability, an indicator of a certain outcome. given that the primary goal of the pes is to facilitate matching on the labour market, the most directly relevant measure of effectiveness is the share of clients exiting the registry towards unsupported employment on the open labour market, the rate of reemployment. even though the data are aggregate, they are directly related to individual behaviour: the number of exits relative to the unemployment pool is an analogy of the individual probability of exit. because the pes produced the exit data for different groups of the registered unemployed defined over individual characteristics such as age, education or disabled status, i run the regressions for all of these groups. differentiating behaviour on the basis of these groups allowed me to assess the heterogeneity in the impact of the programme even with aggregate data. a great advantage of using data coming from the administrative records of the pes is that they are part of a complete account of the event history of the registered unemployed, but the administrative nature has drawbacks too. based on the european journal of government and economics 1(2) 151 register only, we know little about the labour market history of those unemployed who have not been eligible for financial support. also, because not all jobseekers are strictly required to report the direction of exit, this information is not available for everyone. contact and thus reporting is required in principle and one loses eligibility for financial benefits administered by the pes without it. yet, penalty is not severe if the client does not contact the office either due to no initial eligibility in the first place or due to having exhausted such benefits, as the unemployed loses eligibility for benefit for 3 months. in relation to the current analysis, this means that direction of exit is measured without error only for those eligible for benefit. we have to note also that this error can be correlated with the factors determining the chance of exit and we have no outside information to assess its size. if the measurement error is strong, it lowers the value of the outcome variable by not counting every successful exit to the open labour market. however, we have no reason to suppose that the reverse can happen, so i expect that the error does not increase the counted number of exits. in order to apply the did method here, we have to chose an appropriate before and after period. considering that the hrdop 1.2 measure was rolled out between the second half of 2004 and the first half of 2008 and also that the effects of the economic crisis were very apparent in the third quarter of 2008, i chose the first 6 months of 2004 to be the before and that of 2008 to be the after period. the evidence presented in card, kluve and weber (2010) suggests that in order to assess the programme effects fully, one should ideally follow and observe programme participants for years after the end of the programme – this is definitely a longer period than what is possible here. on the one hand, the effect of the economic crisis was very asymmetric regionally and extending the observation period would run a risk of confounding the effect of the crisis with that of the programme. on the other hand, development of the pes has continued through the srop 1.3.1 project, and this has basically eliminating the control group. there are 158 local pes offices in the analysis – only those existing both in the first half of 2004 and 2008. i have omitted two outlier offices, the one specialised in helping homeless people (on haller street) and the one specialising in helping higher education graduates (the one on andrássy street) in budapest. i have omitted also those two offices where all steps of the modernisation were completed in the previous round. if we are looking only at the formal definitions, we can consider offices modernised during the hrdop 1.2 measure as participants and those not modernised in either during the previous phase or during the hrdop 1.2 measure itself as nonparticipants. however, as only 7 offices adopted the quality assurance framework during the previous phase and the rest (13) did so only during the hrdop 1.2 measure, i consider also the latter as participants for current purposes. note that this does not affect the number of non-participants, as those having participated in the previous phase are not counted towards them. the end result is that out of the total 158 offices, we have 71 participants and 85 non-participants as their controls. as i have already mentioned, participant offices for the development project were selected from smaller and larger towns in every county, providing some randomness in selection to a certain extent. however, interviews conducted during a broader evaluation exercise indicated that participation chances were biased to some extent towards offices in worse shape. this observation warrants caution towards estimates that do not take such differences into account and prompts at least a comparative estimation with homogenisation of treatment and control groups based on initial differences between offices. because it proved to be impossible to collect comprehensive information on the actual condition of the buildings or on a similar indicator for the pre-programme period, i used the characteristics of the clients as a proxy during matching. european journal of government and economics 1(2) 152 figure 1: monthly exit rates between 2004 and 2008 0.000 0.020 0.040 0.060 0.080 0.100 0.120 j a n 2 0 0 4 a p r 2 0 0 4 j u l 2 0 0 4 o ct 2 0 0 4 j a n 2 0 0 5 a p r 2 0 0 5 j u l 2 0 0 5 o ct 2 0 0 5 j a n 2 0 0 6 a p r 2 0 0 6 j u l 2 0 0 6 o ct 2 0 0 6 j a n 2 0 0 7 a p r 2 0 0 7 j u l 2 0 0 7 o ct 2 0 0 7 j a n 2 0 0 8 a p r 2 0 0 8 j u l 2 0 0 8 o ct 2 0 0 8 employment supported employment: other public works training not known number of exiting individuals in a group divided by the number of persons in the group prior to exit average, all offices. source: own calculation from the ir based on aggregated individual data. as the programme elements were delivered at almost the same time to all offices (excluding the 13 offices where the self-help terminals and the new service model was already in place), we can look at only the combined effect of the installation of self-help terminals, the introduction of client profiling and adopting the quality assurance system. this means that if we do measure an effect, we cannot tell from which programme element it comes. however, if we do not measure an effect, we cannot tell if this means that all elements were ineffective or that there are powerful effects at work pointing at the opposite directions. figure 1 shows monthly exit rates from the unemployment register between 2004 and 2008. based on this evidence, the two major exit directions are 1) open market employment, with around 4 per cent rate by the end of the period, 2) not known due to lack of reporting back to the pes, with an average of around 8 per cent. these figures are similar to those observed in eastern european countries on average (kuddo, 2009). besides the slight increase in exit to employment, we can observe a much stronger decrease in the exit rate to the not known state. this effect is already present from 2000 on (not visible on the graph) which indicates that this decrease is not to be attributed to the modernisation process. exit rates towards all destinations also appear to show seasonal cyclicality. figure 1 shows that re-employment rates grow particularly strongly during the summer and decreases during the winter – the reason for this is partly that a large number of seasonal jobs are offered during the summer and subsidies are made available during the spring, take-up rate topping by the end of summer. even though the process started well before 2004, the large and trending decrease in the rate of exit towards an unknown state raises the question if there was indeed measurement error present in the indicator of the exit route. such a measurement error affects this analysis if the change in exit rates is correlated with the error affecting our chosen effectiveness indicator. in that case, those who were likely to report an unknown destination are becoming more likely to report exit to an unsupported job over time, for example. the most likely cause of this lack of information is the lack of motivation to keep in touch with the office. clients are motivated either directly, when contact is required for benefit payment, among others, or indirectly, through the provision of services desirable to the client. the european journal of government and economics 1(2) 153 participation of an office does not affect administrative rules directly, but clients registered with modernised offices can feel contact to be more useful. this can lead to more frequent contact between the office and the client, to greater likelihood of reporting exit to a job and ultimately to the decrease of the measurement error. such a process creates a negative correlation between participating in the modernisation programme and the measurement error in the exit rate, leading to the overestimation of the programme effect (by usual omitted variable arguments). although we can be sure that such a distortion exists, i suspect that its size is likely to be small. it probably also appears together with other noises distorting the estimates in an unknown way – this is a reason i use different estimation methods and specifications for measurement. in the absence of this negative correlation, measurement error would merely decrease the precision of the estimates (appearing on the left hand side of the equation). table 1 shows re-employment rates in the preand post-programme period based on office-level data, weighted by the number of the number of unemployed registered with the given office. the re-employment rate has increased greatly from 2004 to 2008. programme participants experienced a 1% point increase, while the same was 0.8% point in the case of the control group. using the did method, the programme-effect is the difference between these two numbers, 0.23. this number is not small compared to the overall re-employment rate, but is not significantly different from zero. looking at the same thing from a different angle, we see that the initial gap between the participant and non-participant offices in re-employment rates in 2004 has basically vanished by 2008. table 1: average re-employment rates at the pes offices by hrdop 1.2 programme participation status and time-period year participant? difference no yes 2004 0.0414 0.0386 –0.0028 2008 0.0496 0.0491 –0.0005 difference 0.0082 0.0105 0.0023 note: without participants of the first phase of the modernisation process; averages are weighted by the number of unemployed registered with the local office. source: own calculations using data aggregated from the ir of the pes using data aggregated to the level of the whole country, figure 2 shows the changing share of vulnerable client groups over time. these include those without a maturity exam, those aged above 50 (the 50+), labour market entrants and disabled persons (counting them multiply, hence proportions add up to more than 100). the most pronounced change is the growth of the share of the 50+ among the registered clients, being a mere 15% in 2000, but growing by 5% points in 10 years. this is partly explained by the rise in retirement age, partly by the autonomous increase in their level of education. the share of those without a maturity exam decreased slowly but steadily, showing a strong seasonal pattern: it decreased rapidly during the summer months providing seasonal jobs, but increased during the winter. the share of labour market entrants shows more muted, but still strong seasonality, with a reversed time-pattern: their share increases mostly during the summer. their record high share was 10% during 2006. the share of the disabled unemployed is stable below the level of 5% from 2002 on. these changes do vary substantially at the level of local offices and are an important part of the external effects we have to control for during estimation. european journal of government and economics 1(2) 154 figure 2: average composition of the local pes offices between january 2004 and december 2008 0 0.05 0.1 0.15 0.2 0.25 ja n04 m ar -0 4 m ay -0 4 ju l-0 4 s ep -0 4 n ov -0 4 ja n05 m ar -0 5 m ay -0 5 ju l-0 5 s ep -0 5 n ov -0 5 ja n06 m ar -0 6 m ay -0 6 ju l-0 6 s ep -0 6 n ov -0 6 ja n07 m ar -0 7 m ay -0 7 ju l-0 7 s ep -0 7 n ov -0 7 ja n08 m ar -0 8 m ay -0 8 ju l-0 8 s ep -0 8 n ov -0 8 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 50+ disabled labour market entrants without maturity exam (secondary y axis) source: own calculations using data aggregated from the ir of the pes when using a did method, it is important to have sufficiently similar participants and non-participants on average so that the latter form a valid control group of the former. table 2 shows the average of indicators of offices’ characteristics in the beginning of 2004, just before programme participation. there are three types of indicators: one set includes the characteristics of the registered unemployed, the second includes their exit rates towards different directions and the third includes characteristics of the local labour market. the latter were obtained from the on-line public database of the hungarian statistics office on municipalities. given that more than one municipality belongs to one local pes office, i have aggregated these data and then assigned them to the record of the appropriate local office using the matching file provided by the employment office. i have considered budapest the capital as one labour market, so the same trends are matched to all offices there. these data enable us to control for external – such as business cycle – effects not captured already by the changing composition of the registered unemployed pool. participating and non-participating local offices appear to be very similar: there is no real difference either in re-employment chances, or in local labour market in terms of group means. the main difference is that there are almost twice as many clients registered with participating offices on average than in the case of nonparticipants whereas the share of better educated clients is larger in the latter case (with very low absolute shares). not only the mean values are very similar, but also the spread of the indicators (not shown in the table), therefore the requirement of staying on the common support during the did analysis with matching was easy to satisfy. european journal of government and economics 1(2) 155 table 2: main observable characteristics of participating and non-participating local pes offices in the beginning of 2004 participant non-participant average number of clients of the local office 1531 3090 re-employment rate in the given subgroup of the registered unemployed age: 15–25 4.3 4.1 age: 26–50 4.4 4.1 age: 50+ 3.1 2.9 education: without maturity exam (including lower secondary vocational education) 3.9 3.6 education: with maturity exam 4.9 5.0 education: higher education 5.5 5.3 education: lower secondary vocational education 5.1 4.8 disabled 2.5 2.6 not labour market entrant 4.5 4.2 the share of the given subgroup among all the registered unemployed age: 15–25 14.7 14.5 age: 50+ 19.4 19.3 education: without maturity exam (including lower secondary vocational education) 75.9 77.5 education: higher education 3.2 2.4 not labour market entrant 7.3 6.8 characteristics of the local labour market registered unemployed/15–65 year olds (%) 8.2 8.5 total personal income tax collected/taxpayer (in thousand forints, on 2008 prices) 262 249 number of flat built/1000 inhabitant 5.8 5.1 enterprise/1000 inhabitant 152.6 143.6 non-profit organisation/1000 inhabitant 10.4 10.2 share of children in créches (among the 0-2 year olds, %) 7.9 7 net in-migration/1000 inhabitant –0.7 –1.3 source: own calculations using data aggregated from the ir of the pes and tstar data from the hungarian statistics office i work with aggregate data during estimation, in which observations appear more than once and this has a direct effect on the calculation of standard errors of the estimates. in order to take seasonal effects into account and increase efficiency at the same time, i use observations for 6 months separately for each office in the period before and after the programme, respectively. this way every observation contributes six times to the estimation, and the final estimate will be an average of the monthly effects. since there is a high degree of autocorrelation between the time-periods, i calculate clustered standard errors. aggregation of units with different number of observations in them creates a well-known form of heteroskedasticity, therefore i weight the regressions by the number of registered individuals. explanatory variables in the binary model for programme participation include the 2004 january values of the variables characterising local labour markets in the parametric estimating equations, as well as levels, squares and cross-products of outflow rates towards unsupported employment and unknown direction. i have calculated z-statistics using the bootstrap method, with 100 replications. i have used the psmatch2 stata module for matching (leuven and sianesi 2003). i experimented with different averaging methods such as 1:1, k-nearest neighbour, kernel and local linear matching. estimation results i start presenting results with estimated coefficients from simple ols regression of the differenced estimating equation, using the method explained earlier, including restriction to the common support obtained from the participation equation in the matching estimator. it is worth noting that without this restriction, results are stronger than we shall see. estimates related to the re-employment chances of an european journal of government and economics 1(2) 156 average registered unemployed person are shown in table 3, the programme effect being the coefficient on the participation indicator in the first row. results from the simplest specification (1) merely echo the results seen in table 1, indicating a programme effect of 0.16%point, or 5% (the numerical difference is due to slight differences in aggregation). this estimate is not significant at conventional levels. besides the lack of certain control variables, this can be due to the negative bias caused by measurement error. although i am interested in the precise estimation of the programme effect rather than modelling of the change itself, it is interesting that the equation explains very little of the variation in the change in effectiveness. specification (2) improves upon this situation by including separate indicators for all months to filter out seasonal effects. although this has increased explanatory power to 14%, indicating the importance of seasonality in reemployment, neither the estimate of the programme effect nor its precision has changed. specification (3) includes even more information, most importantly the (difference of the) share of registered clients with particular characteristics: age, education, and labour market entrant status. besides the rise in explanatory power, we observe an increase in the programme effect to 0.3% point and an improvement in precision that makes the estimate significant. the size of the effect is close to the one obtained with matching (see later), but is somewhat larger than the raw estimate. variables included in this specification capture the changes in clients’ composition over four years. the coefficients attached to specification (3) should be interpreted with a caveat: their own effect is biased by their possible role of a proxy for the measurement error and i am not able to disentangle the two. in a further specification for checking robustness (not shown here) i included the rate of exit towards unknown state as a proxy for measurement error. this variable carries a lot of extra information and being a dominant share in the same population, it is very likely to “over-control” the programme effect. including this variable, the programme effect has increased marginally and its significance decreased (but results remained significant). in the next step, i have included variables in the estimation that are meant to capture the characteristics of the local labour market. results obtained with the new specification (4) are similar to what we have got in the first two, with not only the programme effect, but significance dropping too. along with the increase in the explanatory power of the model, this shows that there is insufficient information for this extension of the model: multicollinearity between the variables decreases precision more than the value of the extra information they bring in. although the new estimate of the programme effect is smaller than it was before, the confidence interval around it is wide enough to include the previous estimate. for this reason, i use specification (3) as my preferred one in what follows. european journal of government and economics 1(2) 157 table 3: results from did ols regressions in various specifications (1) (2) (3) (4) hrdop 1.2. participant 0.0016 0.0016 0.0030* 0.00167 (0.431) (0.433) (0.0952) (0.342) age: 15–25 –0.000919 0.000540 (0.991) (0.994) age: 50+ 0.0716 0.111** (0.225) (0.0357) education: without maturity exam (including lower secondary vocational education) 0.0259 0.0214 (0.703) (0.729) education: higher education 0.367** 0.240* (0.0123) (0.0780) not labour market entrant –0.188*** –0.137** (0.00339) (0.0247) disabled –0.154** –0.0648 (0.0264) (0.336) registered unemployed/15–65 year olds (%) –0.313*** (8.50e–05) total personal income tax collected/taxpayer (in thousand forints, on 2008 prices) –3.26e–05 (0.419) number of flat built/1000 inhabitant 0.0223 (0.965) enterprise/1000 inhabitant 0.0306** (0.0419) non-profit organisation/1000 inhabitant 2.077 (0.216) share of children in créches (among the 0-2 year olds, %) –0.0391 (0.473) net in-migration/1000 inhabitant –0.153 (0.217) constant 0.00852*** 0.00874*** 0.00653*** 0.00980** (6.36e–09) (1.06e–09) (0.00859) (0.0221) n (on common support)/all observations 834/948 834/948 834/948 834/948 r2 0.02 0.14 0.22 0.28 estimated coefficients with p-values within parentheses below them. * significant at 0.10 level, ** significant at 0.05 level, *** significant at 0.01 level. after the completely parametric estimates, i turn to matching to take into account possible initial differences between the local offices, which i have assumed away so far. table 4 shows estimated programme effects and associated bootstrap zstatistics from simple did matching using various methods. the programme effect is the average effect on the treated, the average time-difference of the average difference between re-employment rates of participant local offices and their synthetic counterfactual realisations. estimates constrained to the common support, programme effects are positive in all cases, but are insignificant and smaller in magnitude than raw effects in the case of averaging methods using all data, such as the kernel and local linear methods. based on lessons from scenario 4 of frölich (2004) relating to the analysis of groups of small and similar size, results using these non-parametric methods are the most credible and i consider these as a preferred specification. table 4: raw programme effects estimated with matching 1:1 5 nearest neighbour local linear kernel estimated programme-effects 0.0039 0.0059** 0.0019 0.0016 bootstrap z-statistics 1.41 2.03 0.69 0.71 ** significant at the 5% level. local linear estimation uses epanechnikov kernel with the default bandwidth of 0.06. european journal of government and economics 1(2) 158 the next step is to combine matching and parametric estimation in order to control for both initial differences and changes in characteristics during the programme period using the two-step method outlined earlier. based on earlier results, i use the kernel method in matching and include the parametric residual-generation in the bootstrap procedure used for the z-statistics of significance of the parameters. table 5: programme effect calculated with matching on residuals obtained from did ols regressions (1) (2) (3) (4) estimated programme-effects 0.0016 0.0051* 0.0048* 0.0043 bootstrap z-statistics (0.71) (1.81) (1.76) (1.55) * significant at the 10% level. matching is performed using the kernel averaging method. regression preprocessing uses specification (3) from table 3. results from controlled matching are shown in table 5 following the structure adopted in table 3. controlling for seasonality made the smallest difference here, the programme effect increasing to 0.5% point and being significant at the 0.01 level. this value decreases slightly but not significantly after including the composition of the client pool of the local offices. including characteristics of the local labour markets has a similar effect, decreasing both the level and the significance of the programme effect. for the same reasons explained earlier, the most credible and thus preferred results come from specification (3). replacing raw numbers with those coming from a multivariate did method combined with matching has thus small but significant net effects. these estimates benefit from correcting for both initial differences and those developing over time, and can thus be regarded as more credible than those ignorant of such differences. working with numbers aggregated over the whole client pool, we could not so far look at the heterogeneity of the programme effect. although we cannot separate effects that are attributable to different types of interventions due to the lack of appropriate data, we can attempt to estimate this composite effect local to different subgroups of clients. given that some parts of the programme have targeted some types of clients, we can actually obtain some information that can be related to specific parts of the programme. one example of this is self-help terminals which are more targeted on the better educated clients: obtaining a positive programme effect of the latter makes it more likely that elements targeted at them could have worked better. the new service model on the other hand is more likely to benefit the less able, where we can apply the same argument. this makes it worthwhile to replicate the above analysis using figures that are aggregated for a specific type of client only. another dimension of the heterogeneity of the programme effect is the direction of exit. it is possible for example that the new service model is more effective in directing clients towards training, but not so effective in directing them towards employment. for this reason, it is also worth replicating the analysis for different outcome indicators. in order to take a look at the effect of modernisation on different groups of clients and with regard to different outcome indicators, i have replicated the analysis for all combinations of these using different populations and outcome indicators. table 6 shows the essence of the results as a collection of estimates of the programme parameters from a regression specified as version (3) in table 3. figures marked with a star show coefficients that are significant at a minimum of 0.1 level. european journal of government and economics 1(2) 159 table 6: programme effects for different subgroups of registered unemployed and for different exit directions employment (open market) public works other active labour market programme training unknown direction age: 15–25 0.0025 –0.0002 –0.0009 –0.0005 –0.0046 age: 26–50 0.0034* 0.0012 0.0000 –0.0002 –0.0016 age: 50+ 0.0018 0.0006 –0.0003 –0.0001 –0.0027 education: without maturity exam (including lower secondary vocational education) 0.0029 0.0011 0.0001 –0.0001 –0.0019 education: with maturity exam 0.0028 –0.0004 0.0001 –0.0005 –0.0028 education: higher education –0.0021 0.0010 –0.0053* –0.0001 –0.0023 education: lower secondary vocational education 0.0032 0.0004 0.0003 –0.0002 –0.0020 disabled 0.0038 –0.0001 –0.0015 –0.0002 –0.0073 not labour market entrant 0.0057* –0.0011 0.0009 0.0007 –0.0019 education: lower secondary vocational education 0.0034 0.0009 –0.0003 –0.0002 –0.0026 * significant at least at the 0.1 level. estimated using specification (3) from table 3, weighted with the number of registered clients within the target group, constrained to the common support the main conclusion from table 6 is that the programme helped open-market employment the most: two out of three significant effects are estimated in this case. the table shows that the 0.3% point average estimate of the programme effect is an average of larger and significant impacts on a few subpopulations with a large number of members on the one hand and smaller and less significant effects on more subpopulations with fewer members on the other. while we do not see a significant effect in the case of the young and the 50+ in the case of openmarket employment, the effect for the prime-age group is well above the average at 0.38% point. there is no real difference in terms of education attainment, but coefficients are rather imprecisely estimated in that case. finally, the effect for those already on the labour market is significantly larger than the average. other coefficients are not significant at conventional levels, except the exit rate towards almps for those with higher education, where the programme effect is negative and significant. if this effect is real, it can be attributed to the better information provided and the selection mechanism put to work and can suggest that less participation in almps might be appropriate for this group. conclusions this study has provided additional evidence to the effectiveness of a public employment service as an employment policy measure. it presented a quantitative evaluation of the effect of the 2004-2008 phase of the modernisation of the public employment service in hungary, using exit chances from the unemployment registry as outcomes. the analyses used matching to control for initial differences between participant and non-participant local offices in terms of the composition of their clients and also takes over-time changes into account. based on the results, we can conclude that the modernisation had a significant positive effect on reemployment chances and this is robust to various changes in the specification. although data restrictions did not allow me to separate the effects of different programme elements, analysis of subgroups revealed that the programme effect was strongest in the case of prime-age workers. european journal of government and economics 1(2) 160 the final numerical results include several corrections and are slightly larger than the one obtained from averaging raw numbers. during the period between 2004 and 2008, re-employment chances have risen from 3.86% to 4.91% in local offices that participated in the modernisation programme. i estimate that out of this almost 1 percentage point change, the impact of the programme was around 0.30-0.48 percentage points. in the first half of 2008, the number of registered unemployed was 450000, of which 263000 were registered with programme participant local offices. given that 5% of them became employees on the open labour market after one month, approximately 800-1200 of them became employed as a result of the development programme. the approximately 5% exit rate measured in 2008 means that the average length of such a spell is 100/5 = 20 month (assuming a constant hazard of exit). in the counterfactual case of the programme not being rolled out, based on the change in exit probability as a result of the programme, we can calculate this duration to be 100/(5 0.3) = 21.3 to 100/(5 0.48) = 22.1 months. this means that the length of the unemployment spell was shortened by 1.3-2.1 months by the programme for clients registered with the participating local offices. because the development of the pes can be considered as a labour market programme similar to almps, one might want to ask the question how the benefits from the modernisation effort compare to costs and to alternative programmes. the first question is not easy to answer, because parts of the programme are difficult to separate and even if this were possible, their costs are difficult to account for. in the extreme case of interpreting the programme as an impulse that creates an everlasting effect, the expense of huf9100 million is equivalent to an annual cost of huf273 million spent forever. if we assume that the modernisation is a “programme” in which all unemployed registered with the participant local offices benefit and divide this cost among them, this spending amounts to a yearly huf1038 thousand, a monthly huf86 thousand cost. comparing this to monthly costs of training programmes and subsidies for self-employment, being a monthly huf101000 and huf177000 per capita respectively, the modernisation is not only effective, but does not appear to be costly either (see table 12.4 and 12.5 in bálint, 2012). references baldwin, george b. (1951) 'tulamusa: a study of the place of the public employment service', industrial and labor relations review 4(4): 509–526. bálint, mónika (2012) 'statistical data', in károly fazekas and gábor kézdi (eda) the hungarian labour market 2012. budapest: research centre for economic and regional studies, hungarian academy of sciences and national employment nonprofit public company. http://econ.core.hu/kiadvany/mt.html. blanchard, olivier jean, and peter diamond (1989) 'the beveridge curve', brookings papers on economic activity (1): 1–76. bruttel, oliver (2005) 'contracting-out and governance mechanisms in the public employment service', wzb berlin discussion paper no. 2005-109. campens, etienne, and solenne tanguy (2006) 'on the (in)efficiency of the public employment agencies', url: http://campens.free.fr/paper/empagencies.pdf. card, david, jochen kluve, and andrea weber (2010) 'active labour market policy evaluations: a meta-analysis', the economic journal 120(548): 452–477. doi:10.1111/j.1468-0297.2010.02387.x. european commission (2010) 'europe 2020 an european strategy for smart, sustainable and inclusive growth', url: http://ec.europa.eu/eu2020. european journal of government and economics 1(2) 161 fay, r. g. (1997) 'making the public employment service more effective through the introduction of market signals', oecd occasional papers labour market and social policy no. 25. frölich, markus (2004) 'finite-sample properties of propensity-score matching and weighting estimators', review of economics and statistics 86(1): 77–90. hárs, ágnes, and katalin nagy (2009) a hefop 1.2 intézkedés indikátorvizsgálata. budapest: tárki. heckman, james j., hidehiko ichimura, and petra todd (1998) 'matching as an econometric evaluation estimator', the review of economic studies 65(2): 261294. ilo (2009) 'public employment services responses to the global economic crisis', url: http://www.ilo.org/wcmsp5/groups/public/---ed_emp/--ifp_skills/documents/publication/wcms_117382.pdf. kluve, jochen (2010) 'the effectiveness of european active labor market programs', labour economics 17(6): 904–918. koning, j. de (2004) the reform of the dutch public employment service. erasmus university, sociaal-economisch onderzoek rotterdam bv (seor). http://hdl.handle.net/1765/1991. kuddo, arvo (2009) 'employment services and active labor market programs in eastern european and central asian contries', world bank social protection and labor discussion paper no. 0918. leuven, edwin, and barbara sianesi (2003) psmatch2: stata module to perform full mahalanobis and propensity score matching, common support graphing, and covariate imbalance testing. mosley, hugh, holger schütz, and nicole breyer (2001) 'management by objectives in european public employment services', wzb berlin discussion paper no. 2001-203. nagy, gyula (2006) javaslat a hefop 1.2. intézkedés indikátorainak további elemzésére. budapest. o’leary, christopher (1998) 'evaluating the effectiveness of active labor programs in hungary'. upjohn institute technical reports. upjohn institute for employment research. oecd (2005) 'public employment services: managing performance', in oecd employment outlook 2005, pp. 209-233. oecd. oecd (2006) boosting jobs and incomes policy lessons from reassessing the oecd jobs strategy. oecd. pissarides, christopher a. (2000) equilibrium unemployment theory 2nd edition. the mit press. ramirez, josé v., and anatoli vassiliev (2006) 'an efficiency comparison of regional employment offices operating under different exogenous conditions', ssrn elibrary. http://ssrn.com/paper=942188. rosenbaum, paul r., and donald b. rubin (1983) 'the central role of the propensity score in observational studies for causal effects', biometrika 70(1): 41-55. vassiliev, anatoli, giovanni ferro luzzi, yves flückiger, and josé v. ramirez (2006) 'unemployment and employment offices’ efficiency: what can be done?', socio-economic planning sciences 40(3): 169-186. wilthagen, ton (2008) mapping out flexicurity pathways in the european union. tilburg: tilburg university, flexicurity research programme. european journal of government and economics 1(2) 162 zweifel, peter, and christoph zaborowski (1996) 'employment service: public or private?', public choice 89(1-2): 131-162. microsoft word ejge_03_01_010.doc european journal of government and economics volume 3, number 1 (june 2014) issn: 2254-7088 75 vertical externalities with lump-sum taxes: how much difference does unemployment make? diego martinez, universidad pablo de olavide, spain tomas sjögren, umeå university, sweden abstract this paper analyses how the existence of unemployment affects the conventional approach to vertical externalities. we discuss the optimality rule for the provision of public inputs both in a unitary and in a federal state. our findings indicate that decentralising spending responsability on public inputs in the presence of unemployment allows output to be closer to the first best level. moreover, we describe the inability of the federal government, behaving as a stackelberg leader, to replicate the unitary outcome, unless there are new policy instruments at government's disposal. jel classification j2; h4; h7. keywords public inputs; unemployment; vertical externalities. acknowledgements we are grateful to the participants at the 20th spanish meeting on public economics, three anonymous referees and the editor for their comments. the authors would like also to thank the spanish ministry of science (projects eco2010-15553 and eco2010-21706), junta de andalucia (projects sej-02479 and sej-6882), the bank of sweden tercentenary foundation (stiftelsen riksbankens jubileumsfond), the swedish council for working life and social research (fas) and the national tax board (skatteverket) for research grants. european journal of government and economics 3(1) 76 introduction the standard approach to vertical externalities establishes that sharing taxes between different levels of government has an impact on efficiency. from the seminal contribution by keen (1998), a number a papers has dealt with this issue, offering various solutions which may internalise this problem (see, for instance, boadway and tremblay, 2006). a common issue in all these contributions is assuming distortionary taxation. another common feature in this literature is the assumption of a competitive labour market, with the labour force matching exactly the demand for labour. papers such as dahlby and wilson (2003) and kotsogiannis and martinez (2008) give a central role to the supply and demand for labour in determining equilibria but always with labour market clearing. in such a world, there is no scope for one of the conventional fiscal policies aimed at fighting unemployment, namely the provision of public inputs. in fact, to the best of our knowledge, no paper so far has dealt with vertical expenditure externalities (caused by the provision of productivity-enhancing public expenditures in a federal context) in the presence of unemployment. this has not been the case when horizontal externalities are involved (ogawa et al, 2006). this paper precisely combines vertical externalities and labour market imperfections in a single model. indeed, we build a theoretical framework in which the federal government is in charge of unemployment benefits and the states provide a public input with positive effects on demand for labour. such a theoretical framework finds some real parallells in the international evidence so long as in countries like spain, the us or australia (just to name a few) the allocation of spending responsibilities and taxes across tiers of government is of the kind we are considering here. taxes are assumed to be lump-sum because we are interested in focusing on the efficiency implications derived from the expenditure side of government decisions rather than on vertical tax externalities. anyway, we will show that ignoring distortionary taxation as a policy variable may play a crucial role for correcting the vertical externality. the results in the paper may be summarised as follows. first, we show that, in spite of using exclusively lump-sum taxes to finance governments, a vertical expenditure externality arises when unemployment exists. this confirms a previous result found in the literature (dahlby and wilson, 2003; martinez, 2008), namely, that both vertical (tax and expenditure) externalities are independent of each other. the provision of public inputs creates a positive vertical impact on federal revenues as long as this type of public spending increases the demand for labour and, therefore, it reduces the resources needed at federal level for paying unemployment benefits. and this occurs without the co-occupancy of elastic tax bases. moreover, we also see how the rule for the provision of public inputs at state level is closer to the production efficiency condition than that corresponding to a unitary country with a non-clearing labour market. obviously, this does not imply that decentralisation is closer to optimality than centralisation but the fact that the assignment of distorting instruments (i. e., the public input) across different levels of government really matters for assessing optimality. second, we have analysed whether the federal government is able to replicate the outcome of a unitary country. as is standard in much of the earlier literature, we have assumed that the upper level government knows the states' reaction functions and therefore behaves as a stackelberg leader vis-a-vis the state governments. our results deviate from the earlier studies so long as the federal government has not sufficient policy instruments at its disposal to implement the unitary equilibrium. this is a consequence of using lump-sum instruments. martinez and sjögren ● vertical externalities with lump-sum taxes 77 in a sense, these results are related to the discussion initiated by sato (2000), who has shown that the ability of the federal government to replicate second-best results depends on what type of policy instrument is available to the federal government. precisely, as result of taking into consideration a new (non-lump-sum) policy instrument, i. e., a public input provided by the federal government that is complement to that offered by the states, the upper level of government is able to replicate the second-best outcome of a unitary country. the structure of the paper is as follows. section 2 describes the main features of the model and the different versions of the optimality rule for the provision of public inputs. sections 3 and 4 evaluate the ability of the federal government to replicate the unitary outcome. finally, section 5 concludes. the basic model this section aims to show two results. first, we characterise the equilibrium in a centralised country with unemployment. this allows us to see how the optimal rule for the provision of public inputs is modified in the presence of unemployment. it will also serve as a benchmark scenario with which we can compare the federal equilibria to be analysed below. second, the benchmark model presented in this section also highlights that the fiscal decisions taken by one level of government (particularly that with spending responsibilities on public inputs) will affect the other levels of the public sector. as a consequence, vertical expenditure externalities will arise even though only lump-sum taxes are used to raise revenue in the public sector. the theoretical framework consists of firms, households and two different tiers of government: the federal level and k subnational states.1 firms are identical across the country and, for the sake of simplicity, we assume that their number is normalised to one in each state. all of them produce a single good on the basis of the following production function:   ,,, 1  gkngknf  (1) where n is labour, k a fixed factor and g a public input. such a production technology allows us to qualify the public input as factor-augmenting.2 in this context, the public spending will increase the return to the fixed production factor k , which we normalised to one, in which case the profit can be expressed as:   ,, wngnf  (2) where w is the wage rate.3 the first-order condition for profit maximisation reads ),( gnfw n , and it implictly defines the demand for labour as:        1 1 11 1 , wggwn (3) 1 we do not denote the states by sub-indexes for making easier the notation and given that they are assumed to be identical. 2 an alternative approach would imply a production function with constant returns to scale in all the inputs (private and public). this would be the case of firm-augmenting public input. it would create economic rents that, in terms of the model we develop here, would not exhibit substantial differences with respect to what we obtain below. 3 the return to labour is not affected by the public input, although this would be the normal situation with factor-augmenting public inputs. this is not the case here because we are interested in considering the impact of the public input on employment, and the demand for labour we obtain below implies that the wage rate is independent of ),( gmw . in a model with full-employment, however, we should set up ),( gmw . european journal of government and economics 3(1) 78 by combining equations (2) and (3), we obtain the profit function:  gw, (4) we assume that all households have the same preferences for consumption c across the federation and these preferences are described by a utility function )(cu , which is increasing in c . each state is populated by three types of consumers: a firm-owner as well as employed and unemployed workers, respectively. these consumer types are denoted by the superindices "f", "e" and "u". the firm-owner is endowed with a fixed factor of production k , which the firmowner hires to the firm in return for the firm´s profit. his budget-constraint is defined by ffc   , where f is a lump-sum tax. regarding the other two types of consumers, we make a distinction between the total labour force available for working, m , and the number of households that effectively are employed n . obviously, full employment is characterised by nm  . the budget constraint for an employed worker is ee wc  , where e is a lump-sum tax, while the budget constraint facing an unemployed worker is given by bcu  , where b denotes a net of tax unemployment benefit. in a centralised country, for the policy variables  gbef ,,, , the government maximises a utilitarian welfare function   fue kuunmkknuw  (5) subject to the following budget constraint:   0 bnmkkgkkn fe  (6) in a situation where there is no unemployment, the first-order conditions are as follows:     ff ufoc  : (7)     ee ufoc  : (8)   1: gfgfoc (9)   ,0:  gnfoc fe  (10) where  is the lagrange multiplier associated with the government´s budget constraint. the first two equations show the standard result from optimisation with lump-sum taxes and transfers: the private marginal utility (of each type of consumer) must be equal to the social welfare cost of taxation, which here is captured by the lagrange multiplier  . equation (9), in turn, is the standard production efficiency condition in the provision of public inputs. finally, (10) is the budget constraint of central government, where the last term of lhs in (6) can be dropped as there is full employment and nm  . let us now turn to the equilibrium with unemployment. it is assumed that the existence of a minimum wage, ow , which exceeds the market-clearing wage, ew , creates unemployment so that nm  . this modification will have an impact on the optimal provision of g . in addition, the first-order condition for the unemployment benefit b must also be taken into consideration. the first-order conditions for b and g are martinez and sjögren ● vertical externalities with lump-sum taxes 79     uubfoc : (11)     .1:   gg e g ue g fbnn uun gfoc   (12) let us now modify the model to include different tiers of governments. we assume that the federal level is in charge of providing the unemployment benefit while the states provide the public inputs.4 both levels of government share the tax on employed workers (with the tax rates et and et chosen by the federal and state governments, respectively; eee tt  ). the revenues collected from the tax on profits are assigned in a proportion  (which is exogenously determined) to the states )10(   , while the tax rate f is exclusively decided by the federal government. in such a framework, let us assume that the states behave as nash players, that is, each subnational government ignores the impact of its fiscal decisions on federal revenues. therefore, the optimisation problem to be solved by the states is:         , , 0.. )( eo o fe ffbee ww gwnn sgntts ubunmwnuwmax       (13) where s is a vertical lump-sum from the federal government to states. the last inequality refers to the distortion existing in the labour market, which is the reason for unemployment. the first-order conditions for et , g and  give:     ee utfoc : (14)       01: g e g ue g ftn uun gfoc  (15)    0: sgntfoc fe  (16) observe that we, for later use, define the first-order condition for g to be a function  whereas we also use the short notation  for the state government´s budget constraint. equation (14) sets up the same rule for choosing the optimal tax rate on employed workers in a centralised country as in a world with two tiers of government. this is a direct consequence of using lump-sum taxes. even in the presence of tax sharing between different levels of government, if household behaviour is not affected by taxes, there is no scope for vertical tax externalities. by contrast, and leaving aside the discussion on the optimal levels of g (see martinez and sanchez (2010) for a graphical analysis and martinez and sjongren (2013) with a similar model to this one), let us begin by comparing the provision of the public input in a two-tier public sector (equation (15)) with the provision in a 4 this distribution of spending responsabilities is not crucial for the results, which would be symmetric with an inverse vertical assignment of public expenditures. anyway, the scheme we follow here is in line with the mainstream of theory of fiscal federalism. european journal of government and economics 3(1) 80 unitary public sector (12). the term e gtn differs from its equivalent in (12), namely, )( bn e g  . as long as the federal government implements a nonnegative lump-sum tax et on employed workers, it follows that since the states decide over g , this tends to reduce the over-provision bias that the presence of unemployment creates in the provision of public inputs. in other words, expression (15) is closer to (9) than equation (12).5 in this regard, and contrary to the conventional view in the literature on vertical externalities, we propose here that more federalism may lead to more production efficiency. to see this in an extreme case, assume that all rent taxes accrue to the states ( 1 ); the federal government needs to be financed by a negative fiscal grant (from states) and/or by charging a positive tax rate et on workers. this latter solution involves an optimal rule for the provision of public inputs closer to the production efficiency condition, minimising the differential effect that the presence of unemployment creates in the discussion on optimality. consequently, the behaviour of federal government becomes a crucial issue to determine the effect of unemployment on the achievement of the production efficiency condition in the provision of public inputs. this is what we study in the next section. the ability of the federal government to replicate the centralised outcome a common way of correcting vertical (tax and expenditure) externalities is to assume that the federal government acts as a stackelberg leader vis-a-vis the lower level governments. in such a context, the sequence of the game is as follows. first, the federal government decides on et , f , s and, residually, on b , while taking into consideration the states' reactions to changes in federal policy variables. second, the state governments determine g and et , treating as exogenous all the decision variables of the upper-level of government. despite the fact that the federal government is the first-mover, we are not here strictu sensu in the presence of a pure stackelberg leader since it lacks enough policy instruments that are strategically complementary (or substitute) to the instruments of the follower(s). under the new conditions, the optimisation problem facing the federal government can now be stated as follows:      ffbee kubunmkwknuwmax   )( (17) 0)()1(..  ksbnmkkkntts fe  ),,,,,( nmstgg fe  (18) ),,,,,( nmsttt feee  (19)  gwnn , .eo ww  5 it is straightforward to show that with full employment no vertical (tax and expenditure) externalities appear. martinez and sjögren ● vertical externalities with lump-sum taxes 81 expressions (18) and (19) are the states' reaction functions.6 as written above, given that we have set up lump-sum taxes on labour income, there is no scope for vertical tax externalities. in other words, the changes in the federal tax rates do not affect the marginal cost of the public funds perceived by the state governments and, consequently, the state tax rates. therefore, for solving the federal problem, some information on comparative statics of these reaction functions is required. to do that, we start from the first-order conditions of states (15) and (16).7 differentiating the functions  and  (and ignoring superindex "e" for sake of simplicity in the notation) produces: 0 f sttg ddsdtdtdg f   (20) 0 f sttg ddsdtdtdg f   (21) this two-equation system can be expressed using a matricial form as follows (and after solving for dg and dt ):                                  fst st tg tg d ds dt dt dg f f   1 (22) from (22), we can retrieve: )( ttttt ag dt dg  (23) )( ststs ag ds dg  (24) )( fff ttf ag d dg   (25) )( tgtgt at dt dt  (26) )( sgsgs at ds dt  (27) )( fff ggf at d dt   (28) where a is tgtg  1 . returning to the federal government´s problem, it is clear that its budget constraint can be written as    nm snt fe b   1 . plugging this into the objective function (17), we obtain the first-order conditions for the policy variables of the federal government: 6 both equations are not fundamentally reduced forms since ),( gwnn  ; the next technical developments are, however, fully consistent with the complete definition of the functions involved. 7 the first-order condition (14) can be ignored in this analysis. in a sense, this expression does not admit any influence from federal variables and, consequently, it does not matter at this point. anyway, expression (14) can be easily inserted in (15) without modifying substantially the analysis below. european journal of government and economics 3(1) 82 0)()( )())(()1()(:)(   tg b tgtgn f tgttt b t ee gnugwngnfu gbtbbunmtuntfoc (29a) 0)()()( )())(()()(:)(   f ggn f gt b g bef ugwngnfu gbtbbunmgnutunfoc ff fffff    (29b) 0)()( )())(()()(:)(   sgsgn f sg b sgsts b s e gwngnfu gnugbtbbunmtunsfoc (29c) 0)()1(:)(  ksbnmkkkntfoc fe  (29d) taking into account that b can be residually obtained from the above four equation-system, we simplify (29a)-(29d), which gives us the following results: 0tt (30) n t f    (31) , 1 n ts  (32) where ),( gnfw n , (23)-(25) and the corresponding partial derivatives of  and  (according to (15) and (16)) have been used. what is implicitly established in (30)-(32) is the inability of federal government to affect states' behaviour. one conclusion is that the federal tax rate on employed workers, et , has no effect on the corresponding state tax rate, et (equation (30)), but also none of the policy variables of the upper level of government has any impact on the state provision of public inputs. indeed, from expressions (23)-(25), it is clear that 0 st ggg f , that is, there is no way through which the federal government can modify the provision of public inputs. the unique impact of the federal policy variables ( f and s on et ) is trivial: an increase (decrease) in some of them reduces (increases) the state tax rate in a magnitude given by the number of employed workers n . therefore, the highest level of government is not able to replicate not only the first-best outcome of (9) but also the optimality rule for the provision of public inputs in a unitary country with unemployment.8 new instruments for the federal government: complementary public inputs things may be different if the federal government is also in charge of providing a public input, fg , which is assumed to be complementary with the state public input (now denoted sg ). this modification means that the model becomes more realistic in the sense that different levels of government are now jointly involved in financing complementary public infrastructure projects. this modification of the model implies that the production, profit and labour demand functions are modified as follows: 8 anyway, we must be aware that the first-best values for et and et are guaranteed in each scenario as long as they are lump-sum taxes. martinez and sjögren ● vertical externalities with lump-sum taxes 83       ggknggknf fsf  1,,, (33)   ,,, wnggnf sf  (34)              1 1 111 1 ,, wggggwn sfsf . (35) as before, combining (34) and (35), the profit function can be written as follows:  .,, sf ggw in a unitary country, the government maximises   fue kuunmkknuw  , subject to   0 bnmkkgkgkkn sffe  . in a situation characterised by full employment ( nm  ), the optimal provision of public inputs is given by the standard production efficiency condition: 1 sf gg ff . by contrast, when unemployment appears as a result of non market-clearing wage rate, the first-order conditions w.r.t. fg and sg are, respectively:     1:   fff f gg e g ue gf fbnn uun gfoc   (36)     1:   sss s gg e g ue gs fbnn uun gfoc   (37) it is straightforward to show that when the government is concerned with the level of employment, if the effect of, say, the state public input on labour demand is higher than the equivalent effect by the federal public input ( fs gg nn  ), then the optimal amount of sg will exceed fg . in a decentralised environment, in which both the federal and the state governments behave as nash competitors, the first-order conditions for fg and sg are, respectively:     1:   fff f gg e g ue gf fbntn uun gfoc  (38)     1:   sss s gg e g ue gs fbntn uun gfoc  (39) comparing these expressions with (36) and (37), it is clear that both types of public inputs will be underprovided if governments set positive tax rates on employed workers. under these circumstances, the levels of fg and sg will be below the optimal ones derived from a centralised setting. in other words, each level of government decides a level of public input without considering its impact on other jurisdictions. consequently, there now exists a double vertical expenditure externality from each level of government to the other. the question now is whether the federal government, behaving as stackelberg leader, is able to replicate the second-best outcome. recall that the answer to this question was "no" in a setting in which the federal instruments were et , f , s and b . with the federal government also providing a public input, its optimisation problem is now: european journal of government and economics 3(1) 84       0)()1(.. )(   kskgbnmkkkntts kubunmkwknuwmax ffe ffbee   (40) ),,,,,,( ffes gnmstgg  (41) ),,,,,,( ffeee gnmsttt  (42)  fs ggwnn ,, .eo ww  note that the states' reaction functions (41) and (42) now include a new argument: the federal public input fg . as before, we first need to know some comparative statics of these functions. equations (23)-(28) are still valid in the new context -with a slight change: the term a must be substituted by a -(see below)and we only have to add the corresponding response of sg and et to the new federal policy instrument fg . particularly, we can write: )( ff gtgtf s a dg dg  (43) ),( fsfs ggggf a dg dt  (44) where a is tsgtsg  1 . in this regard, a significant difference appears when comparing these new results with the previous ones. while in section 3 the federal government only had a very limited (and trivial) impact on state tax rate t (recall expressions (30)-(32) and the fact that 0 st ggg f ), things are now quite different. consider first the case of f s dg dg ; after some algebra manipulations it can be seen that the effect of changes in the federal public input on the state provision of public inputs is given by:       .0 / /     ssssss fsfsfs gggg ue gg gggg ue gg f s ftnuun ftnuun dg dg   this means that an increase in the federal public input encourages the provision of the state public input. that is, there is an additional channel through which the federal government can affect states' behaviour. in the case of the state tax rate, something similar happens: 0       fs gf s gf n dg dg n n t dg dt . but here the effect of federal public input on state policy variable is not so clear. indeed, an increase in the federal public input may lead to either an increase or a decrease in the state tax rate on employed workers. anyway, it is worth noting that again federal government may affect states' behaviour, which was not possible under the previous assumptions. given this, the first-order condition for the optimal provision of fg is as follows: martinez and sjögren ● vertical externalities with lump-sum taxes 85       .1)( :      f s ggf s gg f ue dg dg g ue gf dg dg nnbt dg dg ntn dg dt n uunuun gfoc sfsf f s sf  (45) after some algebra manipulations, it can be shown that to replicate the secondbest condition (36) requires to hold: .fs g ue f s g fb uu dg dg n           (46) if the individual utility is assumed to be linear ( ccu )( ), it is straightforward to prove that both sides of expression (46) have the same sign. hence, to replicate the second best outcome for the provision of public inputs is a real possibility when federal government can spend money in public inputs which are complementary to state public inputs. concluding remarks vertical externalities usually involve challenges for efficiency in federal countries. sharing taxes between different levels of government or the provision of certain public expenditures with effects on other tiers of government revenues imply deviations from the optimality rules, which would be obtained in a centralised world. however, the presence of vertical (tax and expenditure) externalities can be disregarded if lump-sum taxes are used. indeed, the idea of governments affecting fiscal decisions taken by others requires distorting taxes able to modify households' behaviour. all these general statements have to be qualified in the presence of unemployment, and this has been what we have done in this paper. particularly, we have built a simple model with lump-sum taxes and unemployment in which the optimal rule for the provision of public inputs depends on whether the structure of the country is federal or not. indeed, while there is no scope for vertical tax externalities (the fact of using lump-sum taxes here is crucial), a deviation from the second-best outcome takes place when states are in charge of the provision of productivity-enhancing public factors and the federal government finances unemployment benefits. we have confirmed that the optimality condition for the provision of public inputs must consider the impact of this type of public expenditure on employment and, consequently, on public spending in unemployment benefits. as the production efficiency condition for public inputs is not satisfied even in the case of a centralised country, we have analysed what would occur when states behaving as nash players take part in the game. since sub-national governments do not take into account the effect of their public expenditures on unemployment benefits, the over-provision of public inputs (compared to the first-best case with full employment) is lower with a federal structure than in a unitary country. when we have wondered about the capability of federal government to replicate the outcome of a unitary country, we have assumed that the upper level of government behaves as a stackelberg leader, considering the states’ reaction functions. under such a scenario, we have concluded that, unlike previous papers, federal government is not able to internalise the vertical expenditure externality. federal policy variables have no impact on states' decision variables. in part, this is caused by using lump-sum taxes; indeed, distortionary taxation can affect agents' behaviour and this is the way through which all the effects of public inputs can be internalised. by contrast, when the federal government is also in european journal of government and economics 3(1) 86 charge of providing a public input which is complementary to the state public input, it is possible to replicate the second-best outcome for the optimal provision of such as public inputs. a number of issues arise for further research. asymmetries at regional level in the federation can be taken into consideration. given our federal budget constraint, the characterisation of equilibria may then involve that not all the resources collected by the upper level of government in a region must be spent in such territory; consequently, some possibilities for horizontal redistribution arise and even for explicit equalisation schemes. also under this framework, in the presence of mobile production factors, phenomena of tax competition may take place, with the consequent effects on efficiency and regional labour markets. a third avenue for further research could be based on an alternative way for including unemployment in the model. let think us, for instance, of a system of generous enough unemployment benefits and in which the workers take labour decisions. as a policy implication we would underline how important the coordination of different levels of government is to attain social welfare gains. indeed, the design of federal and state fiscal policies must take into account the magnitude of their cross effects on the tax revenues of other tiers of government. particularly, this is true in the case of public infrastructure because this type of government expenditure is very vulnerable to public spending cuts and its benefits are very visible. increasing the coordination in the provision of public transport infrastructure (some roads may be provided at the regional level, whereas railways may be provided by the federal government) translates into more social welfare, part of which will be in terms of employment. this would be especially appropriate in countries like spain, where the unemployment rates have reached extraordinarily high values in the aftermath of the great recession. this is especially pertinent in the context of fiscal austerity of many developed countries in the aftermath of this recession. growth-enhancing policies such as those related to public infrastructure projects are able to generate positive fiscal externalities in the form of reduced public spending in unemployment benefits and increased tax revenues from labour taxation. this is a valuable benefit that should not be underestimated when designing economic policies. references boadway, robin and jean-françois tremblay (2006) ‘a theory of fiscal imbalance’, finanzarchiv/public finance analysis 62 (1): 1-27. dahlby, bev and leonard s. wilson (2003) ‘vertical fiscal externalities in a federation’, journal of public economics 87(5/6): 917--930. keen, michael (1998) ‘vertical tax externalities in the theory of fiscal federalism’, imf staff papers 45 (3): 454-485. kotsogiannis, christos and diego martinez (2008) ‘ad valorem taxes and the fiscal gap in federations’, economics letters 99 (3): 431-434. martinez, diego (2008) ‘optimal federal taxes with public inputs’, finanzarchiv /public finance analysis 64(4): 422-433. martinez, diego and a. jesús sanchez (2010) ‘a note on the optimal level of public inputs’, social choice and welfare 34(3): 363-369. martinez, diego and tomas sjongren (2013) ‘can labor market imperfections cause overprovision of public inputs?’, journal of economic research, 18 (2): 135 146. ogawa, hikaru, yasuhiro sato and toshiki tamai (2006) ‘a note on unemployment and capital tax competition’, journal of urban economics 60(2): 350-356. martinez and sjögren ● vertical externalities with lump-sum taxes 87 sato, motohiro (2000) ‘fiscal externalities and efficient transfers in a federation’, international tax and public finance, 7:119-139. microsoft word ejge_03_01_010.doc european journal of government and economics volume 3, number 1 (june 2014) issn: 2254-7088 24 collective action problems: disentangling possible feedback loops between government policies and the public’s value-change eivind hoff-elimari, common cause network, norway anat bardi, royal holloway university of london, united kingdom simon matti, luleå university of technology, sweden kristina östman, kth royal institute of technology, sweden abstract solving collective action problems, such as poverty reduction or climate change, depends on interactions between governments' and voters' preferences regarding pro-social actions. this paper examines whether the overall direction of change in pro-social public policy precedes public value-change, rather than the other way around. we examine change in the public’s pro-social values in six european countries, as measured by the european social survey (ess) during 2002-2012. in these countries, we conducted an expert survey to rate governmental policy that expresses these values over the same period, thereby examining value-change in governmental policy. the chronological comparison of value-change of the public with that of respective governments suggests that changes in pro-social government policies may drive public value-change rather than vice versa. this complements previous studies focused on the opinion-policy connection. possible political implications are discussed. the promising findings of this initial study point to the importance of conducting larger-scale future studies. jel classification z18. keywords values; policy feedback; democracy. acknowledgements we would like to thank wwf-uk, in particular tom crompton, for the financial support and advice that enabled this research. we are also grateful to ela polek of university college dublin for valuable advice and encouragement. last but not least, we thank the anonymous political scientists who filled in the survey on governments' political goals. hoff-elimari, bardi, matti and östman ● collective action problems 25 introduction this research addresses the feedback loop between national political agendas and value-priorities of the mass public in an attempt to further our understanding of how political space for addressing large-scale collective action problems evolves. characterised by a temporal and spatial mismatch between costs and benefits, as well as between problem causes and problem effects, a range of large-scale collective action problems (e.g. duit, 2011; ostrom, 2000; hodge & mcnally, 2000) or bigger-than-self problems (crompton, 2010) exist as consistent challenges for modern societies. for example, deficits of public pension regimes reward the present beneficiaries at the expense of future tax-payers, and the impact of co2 emissions today will be felt fully through climate change only in a few decades. addressing these problems present a significant challenge for contemporary democratic governments. they cannot be addressed by appeals to individual (or national) self-interest alone but rather require an appeal to, and an activation of, the public’s moral-normative concerns to trigger both public policy support and subsequent large-scale cooperative behaviour. the interaction between public policy and public opinion has been a long standing interest within social science research. as a key characteristic of democracy (e.g., dahl, 1989), an abundance of previous research has focused on the effects of opinion on policy, conceptualised as democratic responsiveness. although the magnitude and stability of these effects are still unknown, most scholars agree that policy-makers, particularly with regard to high-salience issues, do not allow policy decisions to flow too far away from public preferences (page & shapiro, 1983; stimson, mackuen & erikson, 1995; glynn et al., 1999; burstein, 2003; wallner, 2008). however, when policy matches the opinions of the mass public, is this match a sign of a responsive policy or of a responsive public? addressing the latter, researchers have demonstrated how political decision-makers regularly attempt to impact public opinion by emphasising certain values (feldman, 1988; caprara, schwartz, capanna, vecchione & barbaranelli, 2006); framing their messages (gilbert & lindzey, 1998; rein & schön, 1993); and making “emotive appeals” believed to increase legitimacy (wallner, 2008) when presenting policy proposals. hence, although changes in mass-opinions give rise to new policy suggestions by responsive decision-makers, public opinion may also be affected by previous policy decisions. indeed, an emerging literature on this interaction argues that the public policy – public opinion interaction, rather than focusing on a single direction of causality, should be thought of as a two-way system of feedback loops, comprising both a dynamic policy representation and a dynamic public responsiveness (e.g. the thermostatic model of responsiveness, cf. wlezien, 1995, 1996, 2004; soroka & wlezien, 2004, 2010). drawing on easton’s (e.g. 1953) traditional model of the political system, in which public reactions to policy outcomes constitutes new inputs to the system, these feedback loops can be both positive (self-reinforcing policy change) or negative (leading to policy reversal) (skocpol, 1993; soss & schram, 2007; soroka & wlezien, 2010). for example, pierson (1993) categorised these feedback effects between public policies and the mass public as resource and incentive effects and interpretive effects. the former have to do with the direct material impacts of public policies, which may create vocal constituencies that will defend existing policies or trigger opposition from those who lose out as a result of the policies. this is akin to research on retrospective voting (e.g., fiorina, 1981) suggesting that past performances is a significant factor for evaluating political alternatives and candidate support in upcoming elections. the interpretive effects, on the other hand, change the public’s own cognitive processes. soss (1999) has demonstrated how one type of interpretive effect is political learning through individuals’ direct experience with government programmes. svallfors (2010) added that public policies may also trigger a normative feedback mechanism, which may be european journal of government and economics 3(1) 26 independent of direct experience of public policies. seen this way, public policy may impact on publicly established views on how the world ought to be and thereby contribute to creating or expanding political space for addressing also the moral-normative issues and long time-horizons inherent in large-scale collective action problems. in short, new policy may indeed be conducive to new politics (e.g., schattschneider, 1935). however, although the notion of a causal effect from policy to opinion has received increased attention during the past decades, research has primarily focused on how specific policies give rise to specific changes in opinion on the particular issue addressed in policy (soss & schram, 2007; soroka & wlezien, 2010). in most of the empirical literature, therefore, the broader political effects of policy – opinion interactions have been largely overlooked. recognising the link between large-scale collective action problems and human values and value change as documented by e.g. schwartz (2006) or schultz et al. (2005), the primary aim of this article is to explore whether the value-priorities expressed through national policy are associated with subsequent changes in the value-priorities of the mass public, in the case of large-scale collective-action problems. thereby, this article contributes to further exploring one of the empirical links between public policy and public opinion: public responsiveness. to fulfil this aim, we use a combination of ess data and expert surveys to compare how national policy goals and mass public values have changed in six european countries during 2002-2012. we do not seek to explain differences in the absolute levels of value priorities across countries but focus on changes in value scores. in the next section, we further theorise the proposed feedback mechanism between national political agendas and mass public values, and outline a framework for mapping and comparing the two on the same canvas. the following section describes our method, survey and measurement strategies. the final two sections present our results and discuss their implications. policy feedback and values if there are strong correlations between public policies and voters’ preferences, how can the directions of effects between the two be established? it is tempting to consider policy and opinion as changing through “mutual adjustment, subject to a never-ending endogenous feedback loop” (svallfors, 2010, p. 121). although this may be true in many cases, in particular when viewed in a long-term perspective, it does not help much in understanding the drivers of changes to public policy and mass public opinion. this requires that each mechanism is studied independently. therefore, in this paper, we explore the suggestion that changes in the value priorities reflected through governmental policy are associated with later changes in the public’s value-priorities, implying that governments might have the ability to consciously influence the values of the public. what is the basis to expect that value emphases of national policy would be associated with subsequent changes in the value-priorities of the mass public? as mentioned above, recent developments in political science research on responsiveness have demonstrated how the public indeed reacts to changes in policy. although these reactions, in part, can be due to the real life effects of a policy (e.g. pierson, 1993), public reactions to policy can also be a factor of the political information surrounding implementation processes, allowing the public to form opinions even without being familiar with the specifics of a policy change (jacobs & shapiro, 2000; mackuen, erikson & stimson, 1992; zaller, 1992). according to, for example, the thermostatic model of responsiveness (wlezien 1995 & 1996; soroka & wlezien 2010), the public continuously gauge the difference between their preferred level and direction of policy and the reality of political decision-making, resulting both in short-term demands for policy hoff-elimari, bardi, matti and östman ● collective action problems 27 adjustments as well as in long-term preference-modification in reaction to policy change. the most prominent psychological process of value change to date is the process of adaptation (see detail and review in bardi & goodwin, 2011). that is, personal values adapt to life circumstances such that for most values, those values that can be readily pursued and fulfilled become stronger, and those values whose pursuit is blocked become less important. this mechanism has been used in the past to explain the value profiles of citizens of former communist countries in europe as an adaptation to life under communism (schwartz & bardi, 1997). furthermore, examining sub-populations, the value profiles of polish migrants to the united kingdom have become more similar to the values of united kingdom residents the longer these migrants have remained in the country, probably as part of a process of adaptation (bardi, buchanan, goodwin, slabu, & robinson, 2014). a similar process of adaptation may lead values of the public to change and adapt to the values that governments pursue in their policies. hence, it is plausible that values, as expressed implicitly in governments’ policies, actions, and discourse, would later become more important in the public as part of adaptation to the current regime. in order to explore and compare policy feedback loops, the primary task is to translate public policies and public opinion into something comparable across countries and over time. basic human values offer such an object as both mass opinion and governmental policy can be conceptualised as expressing a coherent set of value-priorities. public policy, defined broadly as the output of the political system (e.g. dror, 1973; premfors, 1989; cochran & malone, 1995; easton, 1953; hjern, 1987), is to its essence constructed by a system of values specifying its ultimate goals and guiding the choice of strategies for reaching them (cf. amara, 1972; jenkins-smith & sabatier, 1993; kingdon, 1995). values are thereby to be considered the “backstop” (tetlock et al., 1996) of public policy, and the way in which value-conflicts and value trade-offs are handled throughout the political system thereby determining its content (stewart, 2009; thatcher & rein, 2004). similarly, we know from both social psychology and public opinion research that people's opinions constitute a function of their general values (cf. rokeach, 1973; converse, 1964; glynn et al., 1999) and that value-priorities serve as a shortcut for the formation of preferences on a range of political issues (e.g. zaller, 1992; jacoby, 2006; altemeyer, 1998; rasinski, 1987; brewer & gross, 2005; feldman, 1988; hurwitz & peffley, 1987). schwartz (e.g., 2011) has shown that both individuals and cultures on all continents share a common structure of values, understood as guiding principles in life, arranging a set of ten motivational value-types. the proposed structure of individuals’ values has been examined in 75 nations worldwide (schwartz, 2011) and is widely referred to as a theoretical starting point in research on a number of socio-political issues (e.g., davidov, schmidt & schwartz, 2008; caprara et al., 2006; barnea & schwartz, 1998; rohan & zanna, 1998 & 1996; sagiv & schwartz, 1995; stern et al., 1995). according to schwartz (1992), basic human values are organised in ten value types whose reciprocal relations can be represented as a circumplex (see figure 1), where proximity on the circumference indicates that values are mutually compatible and that strengthening a value will strengthen neighbouring values, whereas diametrical opposition indicates the opposite: strengthening a value on one side of the circle will weaken the value on the opposite side. this representation of basic values as a circle allows the identification of four higher-order value types according to two oppositions: openness to change versus conservation (otc/c), focusing the conflict between willingness for change and the motivation to keep things as they are, and self-enhancement versus selftranscendence (se/st), focusing the conflict between prioritising selfish interests and prioritising the interests of others or one’s group (schwartz, 1992, p. 43). european journal of government and economics 3(1) 28 figure 1: schwartz value types source: holmes, blackmore, hawkins & wakeford (2011) priorities amongst the basic values vary from individual to individual, from society to society and over time. the aggregate value priorities across societies form the very heart of culture (e.g., hofstede, 1980; schwartz, 1994) and are correlated with the public’s political preferences (e.g., caprara et al., 2006). for instance, selftranscendence values such as equality and unity with nature are closely related to pro-social and cooperating behaviour (e.g. schwartz, 2006; de groot & steg, 2008; stern & dietz, 1994) as well as to environmental concern (schultz et al., 2005). similarly, the values that prevail in society are used by political leaders to set policies. for example, welfare laws are likely to prevail in countries where values of self-transcendence, such as justice and equality, are emphasised (bardi & sagiv, 2003). although the otc/c-dimension has been suggested as closely associated with general political-ideological reasoning, thereby predicting the formation of support for political parties and programmes primarily on matters concerning the stateindividual relationship, e.g. between conservatism and liberalism (cf. schwartz, 1992; barnea & schwartz, 1998; caprara et al., 2006), the focus of this article is on how values of the se/st-dimension are materialised in public policy and through public opinion. as this paper focuses on policies addressing large-scale collective action or bigger-than-self problems, which require for their amendment that people prioritise the good of others rather than personal utility maximisation, the st/se focus on priorities concerning individual or collective benefits is the appropriate choice. in the next section we shall therefore examine empirically how six european countries moved in different directions on the se/st-dimension in the early 2000s and how their governments’ corresponding policy goals evolved over the same time period. hoff-elimari, bardi, matti and östman ● collective action problems 29 method and data method value-priorities expressed through public opinion the longest time series for regular measurements of schwartz values at national level is the european social survey (ess). it is conducted every second year since 2002 (2003 for some countries) across europe (including russia, turkey and israel). each national survey consists of representative samples of between 1200 and 2800 respondents. a core module of questions has been included in all rounds. this module includes schwartz’ shortened 21-item portrait values questionnaire (pvq) to map values. the questions describe a characteristic of a fictitious person and the respondent is asked to evaluate how much the described person resembles him or her on a scale from 1 (very much like me) to 6 (not like me at all). each of the ten value-types is expressed by two or three items. for example, one of the items for universalism is: “she/he strongly believes that people should care for nature. looking after the environment is important to her/him.” for each respondent, a mean score of all pvq items was first computed and subtracted from 7 to reverse the scale in order to make its interpretation more intuitive: a high score means a respondent generally felt much like the person portrayed. for each of the value types, power, achievement, benevolence and universalism, each respondent’s mean score across the two or three items of the value type was calculated and the scale again reversed by subtracting it from 7. the respondent’s overall mean score across all the pvq items was subtracted from each of these values and 5 were added to the figures for each respondent to eliminate any negative values. each respondent’s mean of power and achievement represent her self-enhancement score and the mean of universalism and benevolence represent her self-transcendence score. we applied the demographic weights provided by ess to compensate for over and under-representation of certain age groups and household types to calculate national mean self-transcendence and self-enhancement scores. the difference between the two is the compound indicator we used to compare public opinion over time. we refer to this as the public opinion net self-transcendence score. value-priorities expressed through governmental policy the first five rounds of ess gives a time series from 2002-2003 to 2010-2012 of changing value priorities of public opinion on the se/st-dimension. to study feedback effects, we also need a mirror measure of changes to governments’ value priorities over the same time period. one option would have been to study changes to material effects of public policies of most obvious relevance to value priorities on the self-transcendence vs. self-enhancement opposition, such as marginal income tax, immigration or environmental policies. however, we found that official statistics were deficient in many fields before 2004 for newer eu member states (e.g., on immigration or environmental policies) and/or they covered a very small part of what basic values might impact on (such as in the case of marginal income tax). furthermore, there may be significant time lags – the material effects of public policies may appear long after they have been debated and adopted. we did searches of the frequency of keywords associated with the se/st-dimension on websites of heads of government over time (using google advanced search) in order to infer governments' value priorities, but we found that results were extremely sensitive to the choice of single words and that it was not suited to compare governments. we decided instead to construct a compound indicator of what we could infer as being the different governments’ value priorities on the se/st-dimension, based on experts’ observations. european journal of government and economics 3(1) 30 as noted by svallfors (2010), the normative feedback mechanisms of public policies that we are concerned with may be independent of direct experience of public policies: it matters not only what governments do (and don't), but also how they motivate their decisions. in other words, what matters is the sum of what the public perceives of government action and inaction. due to the lack of consistent opinion polls on these matters, we decided to approach academic political scientists, whom we presumed would have followed national politics with interest since the first ess round in 2002 whilst having relatively non-partisan views. we constructed a questionnaire (political scientists survey – pss) with 12 political goals representing typical government priorities that would be relevant across europe over the last decade. this questionnaire was sent to political scientists in the six countries with the most significant variations in the public opinion net selftranscendence score in the period from 2002-2003 to 2010-2012. we used our networks to identify active researchers (with at least a phd) at universities with a good reputation in political science. respondents were asked to rate the extent to which their countries’ governments from 2002 to 2011 (we defined a government as the tenure of a head of government) prioritised each of the 12 goals (six reflecting se-values, six reflecting st-values) using a 5-point scale. the complete list and details of the methodology can be found in the appendix. mean scores for each of the 12 items for each government were calculated and then combined into a government net self-transcendence score for each government, defined as: finally, we combined for each of the selected countries the graphs of net selftranscendence scores from ess and from pss to allow for a discussion of directions of effects between value change in public opinion and in government. data ess analysis our aim was to verify if the public’s value change may be associated with similar change in governments' value emphases, and to see chronologically in which direction these changes have occurred. for this particular aim, we focused our analysis on those eu countries with the most significant changes to the net selftranscendence score, both between the ess1 and ess5 and between two consecutive ess rounds. the countries with the largest changes in these measures can be seen in table 1 and table 2 respectively. hoff-elimari, bardi, matti and östman ● collective action problems 31 table 1: largest changes in net self-transcendence between ess1 and ess5 table 2: largest changes in net self-transcendence between consecutive ess rounds positive negative sweden 0.27 ess5 czech republic -0.45 ess2 greece 0.26 ess5 portugal -0.35 ess2 portugal 0.26 ess3 ireland -0.33 ess5 greece -0.32 ess4 as a result, the czech republic, spain, portugal, sweden, ireland, and greece were selected for further study. figures 2a-f show the evolution of selftranscendence and self-enhancement for these countries over time, mirroring each other. figures 2a-f: public opinion self-transcendence and self-enhancement positive negative spain 0.55 czech republic -0.73 sweden 0.34 ireland -0.26 norway 0.15 ukraine -0.24 european journal of government and economics 3(1) 32 in figure 3, the unpredictable evolution of the public opinion net selftranscendence score in all six countries is illustrated. for instance, sweden has increased its score from a level that was already at the top. spain started from a medium position and moved to the top of the list. conversely, the czech republic started at the top only to end up with the lowest score of all six at the time of ess5. figure 3: public opinion net self-transcendence score for selected countries the political scientist survey (pss) the expert survey was carried out using the web service freeonlinesurveys.com which allowed the respondents to fill it in anonymously at their own leisure. emails were sent to 8-16 political scientists in each country with a link to the survey and a two week period to respond. after nine days they were all approached again with a reminder by email. the average response rate was 41% between the six countries, with an average of 4 respondents each and 22 respondents in total. with the exception of the czech republic and portugal, all respondents followed the instructions and filled in the survey for all their respective governments. the inter-judge reliability, or the degree of agreement between the different respondents for each government, was calculated using cronbach’s alpha. the range of values calculated was 0.61-0.92, except for the lower cases of the persson government in sweden (0.47), the barroso and santana lopes governments in portugal (0.52 and 0.57, respectively) and all governments in hoff-elimari, bardi, matti and östman ● collective action problems 33 greece. it was found that the unreliability for the greek case was caused mainly by one respondent. once this data set was removed, all governments except for karamanlis had acceptable cronbach’s alpha values of more than 0.6. the karamanlis government, which remained at an astonishingly low 0.09, is therefore treated with caution, as discussed under results. results and discussion to investigate feedback loops between the value-priorities expressed through governmental policies and those expressed in public opinion, graphs were plotted showing net self-transcendence both for public opinion and governments. a high net self-transcendence score shows a tendency towards self-transcendence values whereas a low score shows a tendency towards self-enhancement values. the public opinion’s net self-transcendence score is tracked on the primary vertical axis to the left and the governments’ net self-transcendence score on the secondary vertical axis to the right (figures 4, 5, 7, 8, 9 and 10). as our primary interest is to explore whether changes to the value priorities reflected in a government’s policies are associated with later changes to values of the public, we focus on the extent to which the graphs show that a change to the governments’ scores precedes changes to that of the public’s values, and we discuss each of the six countries based on the graphs with ess and pss data. to further our analysis, we also complement our results with election outcomes and opinion polls as well as data from the ess on the political preferences of the respondents. spain figure 4: net self-transcendence for spain spain has experienced the highest increase in the ess net self-transcendence score. it appeared gradually and continuously between ess1 and ess5. the three spanish governments in office 2002-2011 also show the clearest differences in their value priorities: the conservative popular party (pp) governments under josé maría aznar (1996-2004) and mariano rajoy (2011-) both receive a net selftranscendence score about three points lower than the socialist psoe government european journal of government and economics 3(1) 34 led by josé luis rodríguez zapatero (2004-2011). there is large consensus amongst pss respondents in this assessment, with cronbach’s alpha of around 0.9 for all three governments. as the zapatero government entered office between ess1 and ess2 and stayed in office until after ess5, it means the interesting juncture to study is the change of government from aznar to zapatero, which resulted from the march 2004 general elections. the run-up to the elections shows that zapatero became prime minister by accident. the pp consistently led the opinion polls ahead of the 14 march elections (libertad digital, 2004). this changed with the 11 march 2004 madrid train bombings, which left 191 dead and 1,800 injured. the government quickly blamed the basque terrorist organisation eta, despite evidence to the contrary. the government was seen to have tried to manipulate facts and the pp was consequently defeated (gordon, 2004). as mentioned above, the zapatero government is seen to have been very clearly focused on self-transcendent political goals. the ess data suggest that this may have triggered a policy feedback loop confirmed by his re-election in 2008. why, then, did the feedback loop stop at the general elections in 2011, after which the pp returned to government with a majority in parliament? we believe this may be explained by the economic crisis in which spain has been engulfed since 2008. with the recession, unemployment has soared. in none of the worst crisis-ridden euro-zone countries – greece, ireland, italy, portugal and spain – have incumbents been re-elected after 2009. in other words, just like the exogenous event of the madrid bombings may have triggered a policy feedback loop that took the spanish public opinion’s net self-transcendence score on an upward trend, another exogenous event – the economic crisis – may have put a halt to it. thus, the spanish case suggests that, in the absence of large external shocks, governments may influence the value-priorities of the mass public rather than the other way round. ireland figure 5: net self-transcendence for ireland hoff-elimari, bardi, matti and östman ● collective action problems 35 the irish public opinion net self-transcendence score remains remarkably stable over most of the time period with relatively small oscillations up and down during the tenures of bertie ahern and brian cowen (both fianna fáil, centre-right). this is matched by a stable score for the same two governments in office until 2011. this picture changes dramatically with the last ess round (2011-2012) and the change of government after the 2011 elections won by enda kenny’s fine gael party (to the right of fianna fáil). irish public opinion as well as the government turned massively towards self-enhancement. judging from figure 5, both causal directions between public opinion and government value priorities are possible: ess4 was recorded 11 september 2009-12 march 2010, the elections took place on 25 february 2011, and ess5 was recorded 20 september 2011-31 january 2012. the drop in the public opinion net self-transcendence score between ess4 and ess5 could thus have occurred before or after the february 2011 elections. in any case, the fact that public opinion in ess5 mirrors the kenny government value priorities, indicate possible policy feedback. to compare the two potential causal directions between changes to value priorities of government and those of public opinion, we look at opinion polls on voting intentions between ess4 and ess5 (figure 6). figure 6: irish opinion polls from ess4 to 2011 election source: quinn, c. (2008) the polls show that kenny’s fine gael for most of the time period, including at the time ess4 was recorded, had a score that was at a level similar to their election result in february 2011. it also had a consistent political line. if there was any feedback between value priorities of government and those of public opinion, it is more likely that the value change in public opinion occurred after the elections, which would be in line with our hypothesis. the victory of fine gael may be explained by the general pattern since 2009 that incumbents lose elections in crisis-ridden euro-zone countries. european journal of government and economics 3(1) 36 greece figure 7: net self-transcendence for greece in greece, the public opinion net self-transcendence score experienced a dip between ess2 and ess4 (greece did not take part in ess3) followed by an astonishing increase between ess4 and ess5. the government scores are stable until george papandreou’s pasok (socialist) government 2009-2011, when it increases significantly and then drops even more with the care-taking government of lucas papademos. note that the kostas karamanlis score needs to be interpreted with caution: the participants in the pss differ widely in their assessment of his government. we focus our attention on the most abrupt and significant change to the public opinion net self-transcendence score, between ess4 and ess5. ess4 took place 15 july-20 november 2009. this was after pasok had won the european elections in june 2009, confirmed by the general election on 4 october 2009 after which karamanlis was replaced by papandreou. thus the public opinion net selftranscendence score at the time of the 2009 elections was rather contrary to the value priorities of papandreou. probably as a result of the deep economic crisis, the incumbent lost. the subsequent rise of the public opinion net self-transcendence score under papandreou provides some evidence favourable to our hypothesis that the government may influence public opinion more than the other way round. hoff-elimari, bardi, matti and östman ● collective action problems 37 portugal figure 8: net self-transcendence for portugal portugal experienced a dramatic downward shift in public opinion’s net selftranscendence score between ess1 and ess2. it then bounced back up by ess3, although it remained below the ess1 level. the dip corresponds to the josé manuel barroso government 2002-2004 (social democratic party, centre-right), followed by pedro santana lopes 2004-2005 (also social democratic party) after barroso had been appointed president of the european commission. santana lopes was defeated in the general elections of 20 february 2005 and was replaced by josé sócrates of the socialist party. the bounce between ess2 and ess3 corresponds to the first two years of the sócrates government. portugal was the first country in the euro-zone to introduce fiscal austerity measures, under the barroso government, after it had broken the eu stability and growth pact ceilings on public deficits. this policy was not reversed by santana lopes. it is hardly surprising, then, that both governments appear largely selfenhancement oriented in the pss (although there was significant disagreement amongst our respondents for both these governments, with cronbach’s alpha being 0.52 and 0.57 respectively). the sócrates government from 2005 received a far higher net self-transcendence score. as in the case of greece, the coincidence of elections and ess rounds makes it straightforward to test our hypothesis for the santana lopes/sócrates shift. ess2 was carried out from 15 october 2004 until 17 march 2005. in other words, about four of the survey’s five months of interviews took place in the run-up to the elections. the ess2 results show that the portuguese public opinion had turned strongly towards self-enhancement during the barroso and santana lopes tenures. sócrates’ victory seems more the result of a “sanction vote” against santana lopes, who carried the stigma of being a substitute for barroso, had not held any national electoral office and presided over an economic slowdown and increased unemployment. the subsequent jump in the public opinion net self-transcendence score between ess2 and ess3, during sócrates’ first two years in office, is in line with our hypothesis. it leaves unexplained the slight, but gradual, decrease in the public opinion’s net self-transcendence score between ess3 and ess5, while sócrates european journal of government and economics 3(1) 38 was still in office. this decrease was, however, far smaller than the changes between the three first ess rounds. sócrates’ subsequent defeat at anticipated 2011 elections and replacement by pedro passos coelho (whose government has a strong self-enhancement orientation) could give credence to the opposite to our hypothesis, in other words that a shift in public opinion’s value priorities helped trigger the election of passos coelho. but this change of government could equally be put down to the general pattern of losing incumbents we also see in other crisisridden euro-zone countries. of the three changes of political parties in government in the period studied, the two first (antonió guterres/barroso and santana lopes/sócrates) could be interpreted in line with a causal effect of public policy on public opinion and the last one in line with the opposite effect. czech republic figure 9: net self-transcendence for the czech republic the czech republic has experienced a steady decline of the public opinion's net self-transcendence score. this started with the single biggest change between the scores of two subsequent ess rounds: between ess1 (24 november 2002-9 march 2003) and ess2 (1 october 2004-13 december 2004), it declined by almost 0.5 points. between the two ess rounds, no general elections were held but stanislav gross succeeded vladimír špidla as prime minister in august 2004. they were both from the same political party (čssd, social-democrats), but the gross government received a much lower inferred net self-transcendence score than the špidla government. the trigger of the change of government was the european elections in june 2004 (czech statistical office, 2004). the čssd lost votes both to the centre-right ods party and to the communist party (ksčm), and ended up in fifth place. it should be noted that the rate of participation was very low (28.32%). špidla stepped down and was replaced by his deputy prime minister gross, who was seen as a youthful, charismatic figure to reinvigorate the čssd. however, his premiership was quickly tainted by cronyism and corruption claims (the economist, 2005), in contrast to the dull but honest reputation of špidla. gross was forced to resign in april 2005. hoff-elimari, bardi, matti and östman ● collective action problems 39 it is difficult to find indicators that would support our hypothesis, given the large change of the public opinion’s score but short period of time between gross’ entry into office in august 2004 and the ess2 interviews (october-december 2004). it is easier to find indicators to test the opposite hypothesis: that shifts in value priorities of public opinion drove the changes to value priorities of government. an indicator would be if the rejection of the čssd were confirmed by a general move in ess2 to the political right, as the political right is associated with selfenhancement in the czech republic. ess data show that this was not the case between ess1 and ess2. it is thus possible that other factors, such as the controversies around gross and the cynicism about politics that he inspired, contributed to the change in the public opinion value-priorities between ess1 and ess2. it is also clear that prior to gross’ government, there was already a potential for cynicism building up, as the špidla government had proven to be a fractious coalition and the prime minister was unable to carry out many of his pledges. to sum up, there is no clear evidence for either causal relationship between shifts in the value priorities of governments and those of the public opinion for this country. during the rest of the time period, there was a slower but steady decline in the public opinion’s net self-transcendence score, accompanied by governments whose score remained relatively low and stable until 2010, when petr nečas took over as prime minister. its net self-transcendence score is significantly lower than the preceding ones. however, with no clear trend breaks, neither in the ess data nor in the government scores, we avoid comparing hypothetical causalities for the developments after the gross government. sweden figure 10: net self-transcendence for sweden sweden is one of the countries with the highest public opinion net selftranscendence score throughout all ess rounds. the score was stable until a significant increase appeared between ess4 and ess5. the only change of prime minister occurred in 2006, when the social-democrat government led by göran european journal of government and economics 3(1) 40 persson was defeated in the general elections by a centre-right coalition led by fredrik reinfeldt of the conservative party (moderaterna). the pss in sweden only comprised two respondents. cronbach’s alpha is satisfactory for reinfeldt (0.73) but not for persson (0.47), which means its score should be interpreted with caution. however, the large difference between the two governments’ score (more than 2.5 points) instills confidence that the reinfeldt government’s net selftranscendence score is lower than that of its predecessor. if there were any causal links between shifts in value priorities of public opinion and those of the government, a shift in public opinion should have occurred either between ess2 and ess3 (which was conducted from 21 september 2006 to 3 february 2007, immediately after the september 2006 elections) or between ess3 and ess4 (which was conducted after two years of the reinfeldt government). neither of them occurred. other factors have been at play to increase the public opinion net self-transcendence score between ess4 (2008-9) and ess5 (20102011). thus, in the swedish case we do not find any clear evidence for a public policy-public opinion effect. conclusions in three (spain, greece and ireland) of the six eu member states with the highest changes to public opinion net self-transcendence scores, findings are in line with the idea that governments’ value priorities as expressed through public policy may predict – in the same direction – self-transcendence values of public opinion. the impact of such possible influence on re-election is, however, secondary to exogenous shocks, such as economic downturns or trust-breaking revelations of government misdoings. the combination of these factors – exogenous shocks and government’s influence on public opinion – can produce contrasted trends, as exemplified by greece and ireland: an acute economic crisis left incumbents defeated in elections in 2009 and 2011 respectively; the new greek government had more self-transcendent political goals compared to its predecessor, whereas the new irish government had more self-enhancement oriented political goals. in the ess round carried out after these elections, public opinion had moved in the direction indicated by the respective government changes. in two cases (czech republic and portugal) there is support for feedback loops between value priorities expressed in government policies and public opinion, but the data do not enable us to disentangle the direction of effects in the feedback loop (czech republic), or they point to effects in both directions (portugal). the czech case is, however, interesting by the fact that it is the country with the highest increase in the purchasing power standard per inhabitant and the only one with a fall in the unemployment rate throughout the period of all the six countries studied, yet with the biggest decrease in the public opinion net self-transcendence score. it confirms the insight from the contrasting cases of ireland and greece that economic trends do not determine the direction of change in public opinion value priorities. in the last case (sweden), we have not found support for any influence between government and public opinion’s value priorities. this paper is intended merely as a preliminary attempt to investigate the existence and possible directions of effects between value priorities of the government and those of public opinion, the confirmation of which would require, inter alia: inclusion of the other ess countries (in particular to compare with countries with small variations to public opinion net self-transcendence scores – specifically, slovenia, netherlands, great britain, denmark, switzerland and belgium)  inclusion of feedback loops between governments and public opinion on the value dimension openness to change vs. conservation hoff-elimari, bardi, matti and östman ● collective action problems 41  better validation of the pss design and a higher number of respondents in each country  more objective measures of government's value priorities  quantitative analysis with control factors such as economic growth or unemployment  inclusion of ess6 data (2012-2013)  systematic consideration of socio-economic variables with these caveats, we draw the cautious conclusion that there are indeed possible feedback loops between the value priorities of governments and of public opinion, and that the values expressed by governments may influence value priorities of the public more than the other way round. this can have far-reaching implications for political strategies to solve bigger-than-self problems such as child poverty or climate change. appeals to economic growth as an end in itself may undermine voters’ willingness to finance welfare services to fight, for instance, child poverty. arguing for investments in renewable energy primarily as a competitive necessity to prevent trading partners from “catching” future jobs could reduce voters’ support for strong climate change policies in general. if our conclusions are confirmed by others, it will also give more confidence to politicians who want to focus on bigger-than-self problems. the politicians’ influence is not limited to the direct effect of policies adopted under their tenure; their rhetoric and their policies can make citizens more supportive of action to tackle such problems – even beyond the next election. references altemeyer, bob (1998) 'the other "authoritarian personality"' in mark p. zanna (ed.) advances in experimental social psychology. san diego, ca: academic, pp. 47-92. amara, roy c. (1972) 'toward a framework for national goals and policy research', policy sciences 3(1): 59-69. bardi, anat and lilach sagiv (2003) 'the eu and israel: comparison of cultures and implications' in klaus boehnke (ed.) israel and europe: a complex relationship. wiesbaden: deutscher universitäts-verlag, pp. 13-26. bardi, anat and robin goodwin, (2011) 'the dual route to value change: individual processes and cultural moderators', journal of cross-cultural psychology 42(2): 271-287. bardi, anat, kathryn e. buchanan, robin goodwin, letitia slabu and mark robinson (2014) 'value stability and change during self-chosen life transitions: self-selection versus socialization effects', journal of personality and social psychology 106(1): 131-147. barnea, marina f. and schwartz, shalom h. (1998) 'values and voting', political psychology 19(1): 17-39. brewer, paul r. and kimberly gross (2005) 'values, framing, and citizens’ thoughts about policy issues: effects on content and quantity', political psychology 26(6): 929-948. burstein, paul (2003) 'the impact of public opinion on public policy: a review and an agenda', political studies quarterly 56(1): 29-40. caprara, gian vittorio, shalom schwartz, cristina capanna, michele vecchione and claudio barbaranelli (2006) 'personality and politics: values, traits and political choice', political psychology 27(1): 1-28. european journal of government and economics 3(1) 42 cochran, charles and eloise f. malone (1995) public policy. new york: mcgrawhill. converse, philip (1964). 'the nature of belief systems in mass publics', in david apter (ed.) ideology and discontent. new york: free press, pp. 206-261. crompton, tom (2010). common cause: the case for working with our cultural values. wwf-uk. url (consulted april 2014): http://assets.wwf.org.uk/downloads/common_cause_report.pdf czech statistical office (2004). url (consulted april 2014): http://www.volby.cz/pls/ep2004/ep11?xjazyk=en dahl, robert a. (1989). democracy and its critics. new haven, ct: yale university press. davidov, eldad, peter schmidt and shalom h. schwartz (2008) 'bringing values back in. the adequacy of the european social survey to measure values in 20 countries', public opinion quarterly 72(3): 420-445. de groot, judith i. m. and linda steg (2008) 'value orientations to explain beliefs related to environmental significant behavior: how to measure egoistic, altruistic, and biospheric value orientations', environment and behavior 40(3): 330-354. dror, yehezkel (1973) public policy reexamined. london: leonard hill books. duit, andreas (2011) 'patterns of environmental collective action: some crossnational findings', political studies 59 (4): 900-920. easton, david (1953) the political system. new york: knopf. the economist (2005, 3-9 march) gross misdemeanours. url (consulted april 2014): http://www.economist.com/node/3723535 european social survey round 5 data (2010) data file edition 2.0. norwegian social science data services, norway data archive and distributor of ess data. european social survey cumulative file, ess 1-4 (2011) data file edition 1.0. norwegian social science data services, norway data archive and distributor of ess data. feldman, stanley (1988) 'structure and consistency in public opinion: the role of core beliefs and values', american journal of political science 32(2): 416-440. fiorina, morris p. (1981). retrospective voting in american national elections. new haven, ct: yale university press. gilbert, daniel t., susan t. fiske and gardner lindzey (1998) the handbook of social psychology. oxford: oxford university press. glynn, carroll j., susan herbst, garrett j. o’keefe and robert y. shapiro (1999) public opinion. boulder, co: westview press. gordon, philip h. (2004) madrid bombings and u.s. policy. testimony to the senate committee on foreign relations. url (consulted april 2014): http://www.brookings.edu/research/testimony/2004/03/31europe-gordon hayek, friedrich a. (1960) the constitution of liberty. chicago, il: university of chicago press. hjern, benny (1987) 'policy analysis: an implementation approach', paper presented at the annual meeting of the american political science association, 3-6 september 1987, chicago. hodge, ian and sandra mcnally (2000) 'wetland restoration, collective action and the role of water management institutions', ecological economics 35(1): 107-118. hoff-elimari, bardi, matti and östman ● collective action problems 43 holmes, tim, elena blackmore, richard hawkins and tom wakeford (2011) the common cause handbook. public interest research centre. url (consulted april 2014): http://valuesandframes.org hofstede, geert (1980) culture’s consequences: international differences in workrelated values. beverly hills, ca: sage. hurwitz, jon and mark peffley (1987) 'how are foreign policy attitudes structured? a hierarchical model', american political science review 81(4): 1099-1120. jacobs, lawrence r. and robert y. shapiro (2000) politicians don't pander: political manipulation and the loss of democratic responsiveness. chicago, il: university of chicago press. jacoby, william (2006) 'value choice and american public opinion', american journal of political science 50(3): 706-723. jenkins-smith, hank and paul sabatier (1993) 'methodological appendix: measuring longitudinal change in elite beliefs using content analysis of public documents', in paul sabatier and hank jenkins-smith (eds) policy change and learning: an advocacy coalition approach. boulder, co: westview press, pp. 237-276. kingdon, john w. (1995) agendas, alternatives and public policies (2nd edition). new york: harpercollins. libertad digital (2004) encuestas y sondeos. url (consulted april 2014): http://www.libertaddigital.com/suplementos/elecciones2004/encuestas.html mackuen, michael b., robert s. erikson and james a. stimson (1992). 'peasants or bankers? the american electorate and the us economy', american political science review 86(3): 597-611. ostrom, elinor (2000) 'crowding out citizenship', scandinavian political studies 23(1): 3-16. page, benjamin i. and robert y. shapiro (1983) 'effects of public opinion on policy', american political science review 77(1): 175-190. pierson, paul (1993) 'when effect becomes cause: policy feedback and political change', world politics 45(4): 595-628. premfors, rune (1989) policyanalys: kunskap, praktik och etik i offentlig verksamhet. lund: studentlitteratur. quinn, ciaran (2008) na pobalbhreitheanna is deireanaí. url (consulted july 2013): http://www.guthanphobail.net/pobalbhreitheanna.htm rasinski, kenneth r. (1987) 'what’s fair is fair—or is it? public values and conflicting views about justice', journal of personality and social psychology 53(1): 201–211. rein, martin and donald schön (1993) 'reframing policy discourse', in frank fischer and john forester (eds) the argumentative turn in policy analysis and planning. durham, nc: duke university press, pp. 145-166. rohan, meg j. and mark p. zanna (1996) 'value transmissions in families', in clive seligman, james m. olson and mark p. zanna (eds) the psychology of values: the ontario symposium, vol. 8. new jersey: lawrence erlbaum associates publishers, pp. 253-76. rohan, meg j. and mark p. zanna (1998) 'the "products of socialization": a discussion of self-regulatory strategies and value systems', in joel cooper and john m. darley (eds) attribution and social interaction: the legacy of edward e. jones. washington dc: american psychological association, pp. 279-292. european journal of government and economics 3(1) 44 rokeach, milton (1973) the nature of human values. new york, ny: the free press. sabatier, paul a. and hank c. jenkins-smith (1999) 'the advocacy coalition framework: an assessment', in paul a. sabatier (ed), theories of the policy process. boulder, co: westview press, pp. 117-166. sagiv, lilach and shalom h. schwartz (1995) 'value priorities and readiness for out-group social contact', journal of personality and social psychology 69(3): 437448. schattschneider, elmer e. (1935) politics, pressures and the tariff: a study of free private enterprise in pressure politics, as shown in the 1929-1930 revision of the tariff. new york: prentice-hall. schultz p. wesley, valdiney v. gouveia, linda d. cameron, geetika tankha, peter schmuck and marek franěk (2005) 'values and their relationship to environmental concern and conservation behavior', journal of cross-cultural psychology 36(4): 457-475. schwartz, shalom h. (1992) 'universals in the content and structure of values: theory and empirical tests in 20 countries', in mark p. zanna (ed) advances in experimental social psychology (vol. 25). new york: academic press, pp. 1–65. schwartz, shalom h. (1994) 'beyond individualism-collectivism: new cultural dimensions of values', in uichol kim, harry c. triandis, cigdem kagitcibasi, sangchin choi and gene yoon (eds) individualism and collectivism: theory, method, and applications. london: sage, pp. 85-119. schwartz, shalom h. (2006) basic human values: an overview. jerusalem: the hebrew university of jerusalem. schwartz, shalom h. (2007) 'a theory of cultural value orientations: explication and applications', in yilmaz r. esmer and thorleif petterson (eds) measuring and mapping cultures: 25 years of comparative value surveys. leiden: brill, pp. 3378. schwartz, shalom h. (2011) 'values: individual and cultural', in fons j. r. van de vijver, athanasios chasiotis and seger m. breugelmans (eds) fundamental questions in cross-cultural psychology. cambridge, ma: cambridge university press, pp. 463-493. schwartz, shalom h. and anat bardi (1997) 'influences of adaptation to communist rule on value priorities in eastern europe', political psychology 18(2): 385-410. skocpol, theda (1992) protecting soldiers and mothers: the political origins of social policy in the united states. cambridge, ma: harvard university press. soroka, stuart and christopher wlezien (2004) 'opinion representation and policy feedback: canada in comparative perspective', canadian journal of political science 37(3): 531-559. soroka, stuart and christopher wlezien (2010) degrees of democracy. cambridge: cambridge university press. soss, joe (1999) 'lessons of welfare: policy design, political learning, and political action', american political science review 93(2): 363-380. soss, joe and sanford f. schram (2007) 'a public transformed? welfare reform as policy feedback', american political science review 101(1): 111-127. stern, paul c. and thomas dietz (1994) 'the value basis of environmental concern', journal of social issues 50(3): 65–84. stern, paul c., thomas dietz, linda kalof and gregory a. guagnano (1995) 'values, beliefs, and proenvironmental action: attitude formation toward hoff-elimari, bardi, matti and östman ● collective action problems 45 emergent attitude objects', journal of applied social psychology 25(18): 16111636. stewart, jenny (2009) public policy values. basingstoke: palgrave macmillan. stimson, james a., michael b. mackuen and robert s. eriksson (1995) 'dynamic representation', american political science review 89(3): 543-565. svallfors, stefan (2010) 'policy feedback, generational replacement, and attitudes to state intervention: eastern and western germany, 1990–2006', european political science review 2(1): 119-135. tetlock, philip e., randall s. peterson and jennifer s. lerner (1996) 'revising the value pluralism model: incorporating social content and context postulates', in clive seligman, james m. olson and mark p. zanna (eds) the ontario symposium of values, vol. 8. new jersey: lawrence erlbaum associates inc., pp. 25-47. thacher, david and martin rein (2004) 'managing value conflict in public policy', governance: an international journal of policy, administration, and institutions 17(4): 457-486. wallner, jennifer (2008) 'legitimacy and public policy: seeing beyond effectiveness, efficiency, and performance', the policy studies journal 36(3): 421443. wlezien, christopher (1995) 'the public as thermostat: dynamics of preferences for spending', american journal of political science 39(4) 981-1000. wlezien, christopher (1996) 'dynamics of representation: the case of us spending on defence', british journal of political science 26(1): 81-103. wlezien, christopher (2004) 'patterns of representation: dynamics of public preferences and policy', journal of politics 66(1): 1-24. zaller, john r. (1992) the nature and origins of mass opinion. new york: cambridge university press. european journal of government and economics 3(1) 46 appendix items of the political scientists’ survey (pss) and correspondence with schwartz value types government priority corresponding pvq items in ess inferred corresponding schwartz value type 1 economic growth it is important to her/him to be rich. she/he wants to have a lot of money and expensive things. achievement 2 publicly funded services (e.g. health, pensions, education and transport funded through taxes and compulsory social security contributions) she/he thinks it is important that every person in the world should be treated equally. she/he believes everyone should have equal opportunities in life. universalism 3 making domestic companies more competitive against foreign ones being very successful is important to her/him. she/he hopes people will recognise her/his achievements. achievement 4 respect for authority it is important to her/him to get respect from others. she/he wants people to do what she/he says. power 5 fight corruption it is important to her/him to be loyal to her/his friends. she/he wants to devote herself/himself to people close to her/him. benevolence 6 liberalisation of regulated sectors being very successful is important to her/him. she/he hopes people will recognise her/his achievements. achievement 7 redistribution (from richer to poorer households) it's very important to her/him to help the people around her/him. she/he wants to care for their well-being. benevolence 8 tolerance (for different lifestyles, appearances and beliefs) it is important to her/him to listen to people who are different from her/him. even when she/he disagrees with them, she/he still wants to understand them. universalism 9 reducing the top marginal income tax rate it's important to her/him to show her/his abilities. she/he wants people to admire what she/he does. achievement 10 helping other countries in need and showing solidarity with them she/he thinks it is important that every person in the world should be treated equally. she/he believes everyone should have equal opportunities in life. universalism 11 prestige of the country it is important to her/him to get respect from others. she/he wants people to do what she/he says. power 12 protection of the environment she/he strongly believes that people should care for nature. looking after the environment is important to her/him. universalism processing of responses responses were coded numerically as follows: option value contrary to goals: -2 indifferent: 0 goal unless public opposition: 1 one of priority goals: 2 the most important goal: 3 more options with positive than negative values were given as we expected more answers with positive values and hence it was reasonable to allow for more nuances to express these. this was confirmed by the distribution of answers. this is also why the only option with a negative value was set at -2 to reflect the fact that the options with positive values ranged from 1 to 3. european journal of government and economics volume 2, number 1 (june 2013) issn: 2254-7088 5 how long can austerity persist? the factors that sustain fiscal consolidations david lodge, european central bank, germany marta rodríguez-vives, european central bank, germany abstract to put public debt on a sustainable path, many governments face the task of enacting large fiscal consolidation followed by years of sustained primary surpluses. by estimating hazard functions for the duration of consolidations, we analyse the features of past consolidation efforts across a panel of advanced economies. our contribution is to identify the factors that help to start and sustain consolidations, separately discussing governments’ “commitment” to the cause as well as their “capacity” for action. our analysis suggests that longer consolidations are initiated when public debt is high, fiscal deficits are large, the interest burden heavy and long-term sovereign bond yields elevated. however, we also find that a countries’ “capacity” to change course is important. higher initial private sector savings, a stronger external balance, a competitive position and stable financial conditions appear to provide more scope for governments to sustain longer-lasting consolidations. once we have controlled for the initial macroeconomic conditions, there is a lesser role for governments’ commitment as reflected in factors such as the composition and the pace of the fiscal adjustment or the political cycle in explaining the duration of consolidation. however, commitment to permanent, rather than temporary, fiscal adjustment is key. jel classification c23; e20; e62; h62 keywords government debt; government deficit; fiscal consolidation; panel data; survival function acknowledgements the opinions expressed herein are those of the authors, and do not necessarily reflect the views of the ecb or the eurosystem. we thank phillip rother, ad van riet, the editor and two anonymous referees for their useful comments and suggestions. european journal of government and economics 2(1) 6 introduction one legacy of the recent financial and economic crisis has been a substantial fiscal burden for most developed countries. heavy financial sector support, deep economic contraction, and counter-cyclical fiscal policies have all sharply raised government debt-to-gdp ratios since 2008. the legacy effects are potentially severe: increased government borrowing is likely to raise governments’ financing costs, impact on private financing conditions, crowd-out private investment and lower potential economic growth (hartwig lojsch, rodríguez-vives and slavik, 2011). restoring fiscal sustainability is thus a top policy priority across much of the developed world. to ensure that government debt is on a sustainable path many governments inside and outside the euro area must enact large fiscal consolidation and then sustain substantial primary budget surpluses for many years. that is especially true for the euro area countries currently under eu/imf programmes, where countries commit to rein in deficits and reform their economies. the challenge is big but not unknown. several analyses, from the ecb and other international institutions, have highlighted the past experience of countries that managed to sustain fiscal consolidations and/or large primary budget surpluses over some years (ecb, 2011). however, to understand properly the relevance of fiscal adjustment processes of the past, we need to analyse how different factors supporting the length of periods of government action to improve their fiscal balances. in this vein, there is a rich stream of empirical papers that have tried to understand the possible factors supporting “successful” fiscal consolidations. the academic literature systematically refers to the expression “success” and provides a number of definitions to it. however, the term “successful” is controversial. the aims and effects of a period of consolidation would be perceived differently by different actors which may have conflicting preferences about choices for achieving fiscal sustainability.1 while one stream of the literature defines “success” as the ability to significantly reduce government debt over the consolidation period (e.g. heylen and everaert, 2000), our work, in line with another stream of existing studies, measures the persistence (length or longevity) of the fiscal consolidation effort over time. in this article, we draw from the literature that analyses the factors affecting the length of consolidations, using duration analysis to estimating hazard functions (e.g. von hagen, hughes, and strauch (2002); gupta, clements, baldacci and mulas-granados (2004); maroto illera and mulas-granados (2008); agnello, castro and sousa (2012)). the bulk of previous empirical research on the determinants affecting consolidation makes use of case studies, descriptive statistics and econometric techniques – mainly regressions with “successful” fiscal consolidations as dummy variable (e.g. alesina and ardagna (1998); guichard, kennedy, wurzel and andre (2007)).2 the results of existing research differ, not only because of methodological differences, but also due to the different definitions of “success”. the benefit of duration analysis for our purpose is that, as gupta et al. (2004) argue, it enables us to treat the duration of fiscal consolidation as endogenous. in contrast to empirical studies which use a descriptive approach to assess the determinants of past adjustment episodes, duration analysis makes use 1 moreover, another interesting angle to explore would be whether policy makers with their choices in the fiscal consolidation mix are successful in limiting the pressure exerted by organized interest groups towards increasing quasi-private expenditure, such as the recipients of current and capital grants (e.g. buchanan and tullock (1962)), or in reducing the perception of fiscal illusion (buchanan and wagner (1977)). 2 other approaches to overcome shortcomings of regression analysis, include the use of model averaging techniques by hernández de cos and moral-benito (2011) which implies considering and estimating all the possible regressions and subsequently construct a weighted average as the estimate of the variables (determinant of consolidations). lodge and rodríguez-vives ● how long can austerity persist? 7 of all the information available in the data.3 the distinction is important: to understand what helps sustain a long-lasting consolidation effort, we also need to analyse the failures, those that stopped after only a short period of time. in doing so, we aim at addressing one of the major challenges facing several euro area countries: how likely is it that a government will be able to enact large fiscal consolidations over a number of years? and what factors are likely to support their efforts? no doubt it has been a key question posed by capital markets as sovereign bond spreads have widened since the onset of financial crisis in 2007. but this is just one approach to analysing past experiences with fiscal consolidation and it has quite a narrow focus: it asks how long governments have managed to sustain uninterrupted periods of consolidation in the past and what factors affected their ability to do so? it does not directly comment either on debt sustainability, the success of fiscal consolidation in achieving a lasting reduction in government debt ratios nor does it deal with the consequences of consolidation – either for growth or welfare. candidate factors sustaining the length of fiscal consolidations our contribution is to provide a systematic and comprehensive analysis of candidate determinants of duration of fiscal consolidation applied to 20 advanced economies between 1970 and 2010. we group the potential main factors – mostly justified by the literature as likely to sustaining consolidations – into three types of variables (see overview table 1). first, we refer to the “push factors”, which act as circumstances where adjustment is likely to start. lengthy consolidations are frequently ‘forced’ upon governments when fiscal deficits and government debt are large, but also when the interest burden is heavy and long-term sovereign bond yields are elevated. second, once consolidation has started we identify the “pull” factors that sustain the adjustment over time. we separate these variables into two sets: governments’ “capacity” for action, or in other words, a wide range of macroeconomic conditions that might enable and facilitate lengthy fiscal consolidation; and governments’ “commitment”, or the fiscal policy and political factors that might support governments’ willingness to pursue painful consolidations. other papers have already analysed, mostly using regressions but also duration analyses,4 simultaneously several factors sustaining consolidation in a comprehensive way. however, as explained in the remainder of this section, the emphasis has been largely on the fiscal policy and political factors that affect consolidations. papers that also consider the macroeconomic conditions are typically limited to analysing the evolution of the growth rate of gdp or output gap. some studies have extended the analysis of economic conditions to other variables – competitiveness, the international economic environment, or monetary conditions – but there has been a clear gap in empirical research on the role of financial or private sector balance sheets in sustaining fiscal consolidations. 3 the survival analysis comes originally from medical research, but it started to be applied in socioeconomic research in the 1970s and 1980s to plausibly model the behaviour of complex issues– also denominated duration analysis. for instance, in labour economics, empirical questions about employment can be specified in the language of hazard functions such as the longer a job is held, the less likely it is to be lost (kiefer, 1988) or, in housing markets to analyse the duration of house price upturns and downturns (bracke, 2011). 4 although duration analysis techniques have already been applied to study the persistence of fiscal consolidations, some related studies apply different sample period and countries (e.g. eu-15 countries during 1960-2004 by maroto illera and mulas-granados (2008); or 25 emerging countries during 19802001 by gupta et. al (2004)). european journal of government and economics 2(1) 8 table 1: candidate determinants explaining the length of fiscal consolidation episodes variables indicators budget balance position initial primary budget balance government indebtedness initial stock government debt-to-gdp ratio p u s h f a c t o r s p re -c on di tio ns cost of financing government interest payments short-term interest rate long-term bond yields growth gdp monetary conditions inflation rate interest rate external positions current account balance international economic conditions world gdp competitiveness real exchange rate unit labour cost private sector balance sheets stock of private debt private sector saving private sector net lending c ap ac ity financial conditions and financial imbalances asset price developments credit growth financial crisis dummy variable composition of consolidation consolidation on the revenue side versus consolidation on the expenditure side pace of adjustment size of the adjustment effort degree of frontloading fiscal rules expenditure rules budget balance rules strength of the fiscal rules durability of the adjustment structural consolidation versus temporary and one-offs p u ll f a c t o r s c om m itm en t political factors elections, as dummy variable source: authors’ classification based on european commission (2007), and the references quoted therein. although there is no consensus or theoretical model of reference5 on the determinants driving “successful” consolidations a central focus of the literature has been the fiscal factors. most papers note the role of initial fiscal conditions, in particular the influence of high initial debt or large fiscal deficits (e.g. european commission (2007); devries, guajardo, leigh and pescatori (2011)). also guichard et al. (2007) conclude that fiscal consolidations tend to occur when large budget deficits threaten sustainability. this is reflected in the “pre-conditions” in table 1. under “commitment” we include the composition of the fiscal consolidation as its role has been strongly highlighted by the literature on past consolidations. in particular, there is a general preference for the expenditure-based adjustment over tax hikes. von hagen and strauch (2001), for example, conclude that the ‘good quality’ of fiscal adjustments – i.e. the emphasis on expenditure reduction rather than tax increases – has an important effect on the persistence of consolidations. the literature appears to emphasise the effectiveness of reducing politically sensitive items such as government consumption (mainly government wages and employment), subsidies and social transfers (e.g. alesina and perotti (1995, 1997)). heylen and everaert (2000) confirmed these results with multivariate analysis, suggesting that, surprisingly, raising taxes on business proves also to be effective6. another pull factor of fiscal consolidation typically covered by the 5 as pointed by hernández de cos and moral-benito (2011), the empirical research in this field is mostly based on selecting a single regression and dependant on the decision of which factors are relevant (based on its associated t-ratio). 6 however, molnár (2012) after examining 20 different revenue and expenditure items finds that the composition of the consolidation does not seem to explain its length. lodge and rodríguez-vives ● how long can austerity persist? 9 existing literature is the size and the speed of the fiscal adjustment. the general idea is that large and persistent fiscal adjustments contribute to success7 (e.g. briotti (2004); giavazzi and pagano (1996)), which is linked to the credibility of governments in their consolidation strategies and expectations of future tax reductions. in relation to the speed, there is a preference for gradual consolidations, but more intense efforts (e.g. “cold showers”) seem to be more effective than longer consolidations at high and rising debt levels and in low growth scenarios (european commission (2007), barrios et al., 2010). moreover, we also consider under the “commitment”, the possible role of fiscal governance factors as a framework for sustaining consolidation, as suggested by the literature8 (e.g. briotti (2004), and guichard, kennedy, wurzel and andre (2007)). although not explicitly contemplated by the relevant literature, we also test the durability of the adjustment undertaken by governments: whether it is based on structural consolidation over temporary and one-off measures.9 finally, we also include under “commitment” the potential explanatory effect of political factors, via elections, in line with the growing literature – although the evidence is not conclusive (e.g. alesina and perotti (1995), alesina et al. (1998), larch and turrini (2008)). another stream of the literature relates to the importance of the prevailing macroeconomic factors. where papers assess the role of initial macroeconomic conditions the broad conclusions are that governments are ‘pushed’ into fiscal consolidations, in particular referring to weak public finances, but also when the initial macroeconomic indicators suggest a crisis situation (ahrend, catte and price, 2006). moreover, von hagen et al. (2001, 2002) 10, and guichard et al. (2007) found that weak domestic economic pre-conditions also play a role in explaining the likelihood of starting “successful” fiscal consolidations. once consolidation has started, there is also a need to analyse the wider macroeconomic factors that ‘allow’, or make room for adjustment. the leading indicator that seems to justify the role of the macroeconomic environment is output growth, and to a lesser extent monetary conditions, competitiveness and the international economic environment. in general, existing empirical findings tend to suggest that favourable economic growth (in terms of gdp or output gap) contributes to the success of consolidation strategies (e.g. alesina and perotti (1995), ardagna (2004), hernández de cos and moral-benito (2011), agnello et al. (2012)). the findings on monetary conditions are mixed: while the monetary stance could positively contribute by accommodating consolidation (e.g. ardagna (2004)), other authors found no evidence (e.g. von hagen and strauch (2001)). the role of competitiveness through exchange rate policy is not conclusive either, but evidence tends to suggest a positive effect of exchange rate depreciation on the success of consolidations (e.g. molnár, 2012). alesina and perotti (1997) found that floating exchange rates better support successful consolidations than fixed 7 however, the results are not consistent across existing studies, also due to the different definitions of success. for instance, heylen and evaraert (2000) find only a weak evidence to support the view of the importance of the size of fiscal adjustment. 8 however, some studies do not find a significant impact of the expenditure rules (european commission, 2007) or on the budget deficit rules (barrios et al. 2010) on their sustaining role in successful fiscal consolidations. 9 one-offs are temporary or non-recurrent fiscal operations that may substantially impact the government cyclically adjusted primary balance positively or negatively, e.g. capital revenues as result of pension transfers from the private sector to the social security system; sales of mobile phone licences; debt assumptions and debt cancellations; or tax amnesties (joumard et al. 2008). 10 von hagen and co-authors observe that initial conditions (in terms of output gap) were less favourable for “successful” consolidations than for “unsuccessful” ones. however, they also find that a) positive large output gaps increases the likelihood of starting fiscal consolidations, although they reduce the chances of success, and b) that fiscal consolidation is likely to start when the domestic economy is doing well compared to other economies. european journal of government and economics 2(1) 10 exchange rates as currency devaluations increase the likelihood of success. heylen and everaert (2000) suggest that currency devaluations prior to consolidation episodes may be beneficial to their success, but only under the right composition of fiscal adjustment – otherwise it is counterproductive. also a favourable international economic environment (high economic growth) has generally been assessed as sustaining fiscal consolidations (alesina and perotti (1995) and heylen and everaert (2000)). there is a lack of empirical evidence, to the best of our knowledge, on the impact of private sector balance sheets and wider financial conditions in sustaining consolidations. for some euro area members, with on-going private sector balance-sheet adjustment and financial sector restructuring, weak growth prospects and large current account deficits, the ability to change course – and not simply the government’s willingness to make the adjustment – will matter. the literature suggests that consolidations are more difficult after financial crises and depend crucially on the resolution of problems in the banking sector (baldacci, gupta and mulas-granados (2010); and barrios et al. (2010)). in identifying financial factors, we follow the imf (2008 and 2009), which identified fifteen recessions that were associated with (i.e. followed) financial crises. in doing so they relied on the narrative analysis of reinhart and rogoff in a series of recent papers (2008, 2009) describing banking crises over the past four decades. that analysis built on work identifying banking crisis episodes from the world bank – see caprio, klingebiel, laeven and noguera (2003). the article proceeds as follows. section 3 outlines a definition of consolidation episodes and presents some stylised facts. section 4 describes the analysis method, and section 5 presents estimates of survival functions for consolidation episodes. section 6 elaborates on the robustness of the results. section 7 illustrates an example of the role of push and pull factors in selected euro area countries. finally, section 8 concludes. a definition of consolidation episodes and some initial stylized facts to start the analysis we need to define a sample of consolidation episodes and their duration. the literature takes two approaches. one angle is to measure fiscal consolidations based on reductions of the general government debt-to-gdp ratios accomplished during the consolidation period. in those studies, the duration of a consolidation episode is set by the period of time required to reduce debt by a given amount. the second approach, which we take here, is to assess consolidation periods according to changes in the government deficit, with most papers choosing to look at developments in the cyclically-adjusted primary budget balance (capb) which excludes the influence of interest rate developments and the cyclical fluctuations in economic activity on the government deficit. the duration of a consolidation period depends broadly on the length of period in which the government deficit (measured by capb) improves. while the use of the capb indicator is fairly common in the literature, it is subject to some criticism regarding its limitations to capture the macroeconomic effects of fiscal consolidations (devries et al., 2011). while a number of studies use the capb approach to defining consolidations, each study tends to set different thresholds for what constitutes the start of a consolidation and what represents an ongoing improvement. perhaps the simplest definition, used by maroto illera and mulas-granados (2008) and von hagen et al. (2002), is for the duration of the consolidation episode to represent the number of years in which the capb improves. one criticism of that approach, however, is that it captures a lot of short, small changes in the government fiscal stance, which lodge and rodríguez-vives ● how long can austerity persist? 11 might not be seen as periods of active consolidation.11 thus others (e.g. guichard et al. (2007)) use a stricter definition of the beginning of a consolidation period, requiring a substantial initial consolidation effort (e.g. around 1% or 1.5% over two years). other authors define larger changes in capb to qualify for fiscal consolidation episode (e.g. two percentage points of gdp in one year by alesina and ardagna (1998)) or more refined definitions of changes in the capb (e.g. afonso (2010) uses changes in capb at least 1.5 times the panel’s standard deviation in one year, or when the change in the capb is at least one standard deviation on average in the last two years). we broadly follow the oecd approach (guichard et al., 2007) by defining a consolidation period as: starting if the capb improves by at least one percentage point of gdp in one year or at least one half percentage point in two consecutive years; and continuing as long as the capb improves (although an interruption is allowed without terminating the episode as long as the deterioration of the capb is small – less than 0.3% of gdp – and is more than offset the following year). however, we add two features to this definition. first, we rule out any consolidation episodes that lasted only one year – we aim to assess the factors affecting periods of sustained improvements in the capb and are not interested in short-lived consolidation efforts. second, we judge that a small decline in an already better fiscal position should not represent an interruption, as a sizeable primary surplus is likely to help a country reduce its government debt. thus for instances during a consolidation period in which countries have reached a sizeable positive capb – defined here as 3% of gdp – we ignore any deteriorations in the capb that did not move the deficit below that threshold. inevitably, the initial decision on what constitutes a consolidation episode sets the basis for the subsequent analysis. section 5, therefore, analyses the implications of changing our definition of consolidation episodes. before proceeding, however, it is worth noting that changes in the capb are just one way of viewing a consolidation effort. in particular, the fiscal literature tends to distinguish between consolidations that are based on structural, permanent measures and those that arise from reliance on temporary or special factors. the importance of those effects is discussed by joumard, minegishi, andré, nicq and price (2008) with a comprehensive overview on the nature and amount of one-off operations and their role in undermining the accuracy of capbs. 11 based on that definition, von hagen et al. (2002), for example, find 122 consolidation episodes from a sample of oecd countries between 1960 and 1998. illera and mulas-granados (2007) find 277 episodes of consolidation for eu countries between 1960 and 2004. european journal of government and economics 2(1) 12 figure 1: the duration of fiscal consolidations (number of consolidations lasting for x years) 0 5 10 15 20 25 30 35 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 source: oecd and author calculations. our study is based on an annual dataset covering 20 advanced economies over the period from 1970 to 2010, namely 14 eu countries (austria, belgium, denmark, finland, france, germany, greece, ireland, italy, the netherlands, portugal, spain, sweden, the u.k.), and 6 non-eu countries (australia, canada, japan, new zealand, switzerland, and the u.s.). based on our criteria and using this panel we find 65 complete episodes of consolidation as shown in table 2. past experience suggests consolidations have been short: nearly half of the episodes in the sample last only two years and most are over after four years. only a few well-known cases lasted much longer: denmark (1983-1991), belgium (1993-2005), ireland (1986-2001) and japan (1979-1990).12 overall, the average duration of fiscal adjustment in our study is just under four years, which contrasts with the findings of other approaches (for instance, two years by giavazzi and pagano (1996)). 12 the dates for these countries are the result of using the definition described above. alternative definitions, of course, would yield somewhat different dates for the consolidation episodes. lodge and rodríguez-vives ● how long can austerity persist? 13 table 2: dates of consolidation episodes country start end country start end australia 1979 1981 ireland 1986 2001 australia 1986 1989 ireland 2003 2007 australia 1995 1999 italy 1976 1978 australia 2002 2004 italy 1982 1984 austria 1996 1998 italy 1990 1994 belgium 1974 1976 italy 1995 2001 belgium 1984 1988 japan 1979 1990 belgium 1993 2005 netherlands 1972 1974 belgium 2006 2008 netherlands 1976 1978 canada 1986 1990 netherlands 1982 1986 canada 1994 2002 netherlands 2004 2006 denmark 1983 1991 new zealand 1991 1996 denmark 1999 2001 new zealand 2002 2004 denmark 2004 2008 portugal 1982 1985 finland 1981 1983 portugal 2002 2004 finland 1984 1987 portugal 2006 2008 finland 1988 1990 spain 1986 1988 finland 1996 2004 spain 1994 1998 france 1979 1981 spain 2005 2007 france 1983 1985 sweden 1975 1977 france 1994 2000 sweden 1983 1985 germany 1976 1978 sweden 1986 1989 germany 1981 1986 sweden 1994 1999 germany 1992 1995 sweden 2004 2006 germany 1996 2000 switzerland 1994 1997 germany 2005 2008 switzerland 2005 2009 greece 1979 1981 united kingdom 1979 1983 greece 1982 1984 united kingdom 1994 2001 greece 1986 1988 united states 1976 1980 greece 1990 1995 united states 1987 1990 greece 1998 2001 united states 1993 1999 greece 2005 2007 united states 2005 2007 ireland 1982 1985 note: oecd economic outlook database, variable code nlgxqa. note that there are a further 10 episodes that begin in 2010 which have not been concluded and are excluded from our analysis. duration analyses in the subsequent analysis, we study the factors that appear to have affected the length of expansions in our sample of advanced economies, using estimates of hazard functions (e.g. maroto illera and mulas-granados, 2008). duration analysis provides estimates of the probabilities of a consolidation coming to an end after a certain length of time. we estimate hazard functions for the duration of consolidations in the sample of advanced economies, where the duration represents the number of years of uninterrupted fiscal consolidation (according to the definition above). the hazard function measures the conditional rate at which a consolidation will come to an end. relative to other models such as those that focus on the probability of an event taking place (e.g. a probit or logit model), the focus of duration analysis is somewhat different. the central concept in duration analysis is not the unconditional probability of an event but its conditional probability (e.g. the probability of a consolidation ending in a given year given that the expansion has lasted up to that point). we define the length of a consolidation (t), measuring the number of years from the start to the end, as a random variable with a cumulative distribution function f(t) and probability density function f(t). the survivor function is s(t) = 1 – f(t), which gives the probability that the duration of a consolidation is greater than or equal to t. the hazard function h(t) = f(t) / s(t) measures the rate at which consolidation spells will be completed at duration t – i.e. the probability of a budget surplus or deficit deteriorating at time t, conditional on the consolidation having lasted until that moment. broadly speaking, there are two main approaches in the literature to estimating the hazard function. one is to use a non-parametric or semi-parametric approach, most commonly associated with the cox proportional hazards model. although the nonparametric approach is more flexible, it can lead to less precise estimates of the hazard function than a correctly specified parametric form. the other approach, and the one taken here, is to specify and estimate a functional form for the hazard european journal of government and economics 2(1) 14 function. various forms are possible, but a common approach – and the one used here – is the weibull model: h(t,x) 1 t exp )'( x where exp(.) is the exponential function and x is a vector of covariates which influence the hazard rate. the hazard rate (the conditional rate at which consolidations end) either rises with time ( >1) – which is termed “positive duration dependence”, falls with time ( <1) or is constant. this suggests that the conditional probability of a consolidation ending increases as time passes, either as ‘consolidation fatigue’ sets in, or perhaps because as a consolidation continues, policymakers conclude that the ‘job is done’ – governments have pushed through sufficient consolidation to put finances on a sustainable path. in the following sections we present parametric estimates, with hazard functions using the weibull model. to check the robustness of the results, we also considered other functional forms and the non-parametric approach of the cox proportional hazards model. in order to discriminate between models, the literature suggests a graphical analysis of the so-called cox-snell residuals (e.g. arjas, 1988). that analysis suggested that, indeed, the weibull model presents a relatively good fit for our estimated model. another key issue is how to handle the possible presence of country level heterogeneity. various approaches exist. one is to assume that countries have differing propensities for sustaining consolidations. these propensities – or frailties – are analogous to random effects in panel data models. an alternative is to include fixed effects or country dummies in the set of covariates. as an initial step in our estimations, we tested for the significance of fixed and random effects. however, in both instances we found the effects to be insignificant. parametric analyses: factors sustaining the length of fiscal consolidations tables 4 to 7 present estimations of hazard models for the duration of consolidation periods for our panel of advanced economies, using the weibull model. the variables included in the models are described in table 3. the coefficients are reported as hazard ratios – a coefficient greater than zero indicates that an increase in the explanatory variable increases the hazard rate (the conditional probability of a consolidation coming to an end). we also report a measure of explained variation, an adjusted r-squared employed for survival models (royston (2006)). lodge and rodríguez-vives ● how long can austerity persist? 15 table 3: definitions of variables used in duration analyses variable definition capb government cyclically adjusted primary balance (oecd definition) government debt gross government debt as percent gdp government debt >90% dummy variable indicating government debt-to-gdp ratio grater than 90% govt. interest payments (% gdp) gross government interest payments as percent gdp (expressed as negative number) real long-term interest rates yields on 10 year government bonds deflated by current cpi yearon-year growth real short-term interest rates yields on short-term government bonds deflated by current cpi year-on-year growth real gdp growth rate annual gdp growth in three years prior to consolidation real potential output growth rate potential output growth (oecd estimate) output gap output as percent of potential (oecd estimate) world gdp growth annual growth in world gdp (imf) current account balance current account balance as percent of gdp private sector net lending private sector surplus / deficit as percent of gdp private sector credit-to-gdp private sector credit to gdp ratio (source: bis) private sector credit growth change in private sector credit-to-gdp ratio in three years prior to consolidation real equity price growth annual growth in equity prices (deflated by cpi) in three years prior to consolidation financial crisis dummy for occurrence of financial crisis in three years prior to consolidation (source reinhart and rogoff (2008)). real effective exchange rate real effective exchange rate deflated by cpi inflation annual cpi inflation in three years prior to consolidation share of consolidation expenditure expenditure adjustment: share of deficit change during consolidation episode share of consolidation revenue revenue adjustment: share of deficit change during consolidation episode share of consolidation temporary measures share of deficit change during consolidation episode accounted for by one-off measures (oecd definition) intensity of consolidation average change in government deficit during consolidation episode (above minimum set by consolidation definition) frontloading (consolidation in first two years) deficit change in first two years of consolidation change in government debt over consolidation percentage point change in debt-to-gdp ratio during consolidation track record of consolidation number of years of consolidation in five years prior to consolidation episode fiscal rules fiscal rules strength index (source: european commission) source: oecd databases (unless otherwise stated). table 4 investigates the role of initial fiscal and financial conditions. we start by estimating the model without any covariates (column 1) and the results suggest positive duration dependence: the estimates of alpha are greater than one. indeed, throughout the various specifications that remains the case. that means that in general, in our sample, consolidations are more likely to end as they become older. the next set of variables analyse the initial conditions before fiscal consolidations got underway. as far as possible, we would like to isolate factors driving the duration of consolidations. we therefore focus the analysis on variables in the year before the consolidation began, rather than the first year of consolidation when the impact of changes to the fiscal deficit will already be apparent. this should help ensure that the variables are exogenous – drivers of the consolidation length rather than reflections of ongoing consolidation – although we cannot control for possible expectation effects that might have arisen from the anticipation of planned consolidation efforts. we find a strong role for the government’s starting position: a larger deficit tends to presage a longer consolidation (column 2). high government debt also matters but european journal of government and economics 2(1) 16 is no longer statistically significant when instead we also include the government interest burden (column 3). interestingly, we found no role for non-linearities: for example, we found no statistical evidence that government debt above a 90% (a threshold identified reinhart and rogoff (2012)) affected the length of consolidations. nonetheless, market pressures would appear to play a role: higher long-term real bond yields also push governments to sustain consolidations for longer (columns 6 and 7). table 4: determinants of duration of consolidations – the role of initial conditions (estimates of hazard function using weibull model, coefficients shown as hazard ratios) (1) (2) (3) (4) (5) (6) (7) duration of the episode 0.517*** 0.676*** 0.703*** 0.706*** 0.706*** 0.765*** 0.759*** cyclically adjusted primary balance 0.104* 0.116* 0.094 0.094 0.108** 0.110* government debt -0.021*** -0.003 -0.002 -0.002 government interest burden 0.197*** 0.200** 0.201*** 0.126*** 0.122*** high government debt (>90%) 0.151 0.150 government debt (60% to 90%) -0.004 real long-term interest rates -0.181*** -0.163*** real short-term interest rates -0.020 constant -2.441*** -1.380*** -1.646*** -1.758*** -1.757*** -1.729*** -1.726*** number observations 253 248 248 248 248 242 240 number of countries 20 20 20 20 20 20 20 number of consolidations 65 63 63 63 63 61 60 adjusted-r-squared 0.149 0.177 0.172 0.163 0.265 0.247 notes: ***, ** and * indicate significance at the 1%. 5% and 10% level respectively. each variable is measured in the year before the consolidation begins. the “duration of the episode” reports the value of the weibull parameter that governs the shape of the hazard function. the coefficients are reported as hazard ratios – a coefficient greater than zero indicates that an increase in the explanatory variable increases the hazard rate (the conditional probability of a consolidation coming to an end). we also report a measure of explained variation, an adjusted r-squared employed for survival models (royston (2006)). in table 5 we turn to investigating the role of initial macroeconomic conditions. we retain the significant ‘push’ factors from table 1 – the capb, government interest burden and long-term bond yields – and add other macro variables. we find that a larger initial current account surplus or higher private sector surplus tends to help sustain consolidations (columns 2 and 3), even once we have accounted for the role of the government deficit. higher private sector saving (as a percent of gdp) also appears to reduce the probability of consolidations ending (column 4). there is more mixed evidence on the role of financial conditions. a high stock of private sector debt is not a significant explanatory factor but strong credit growth in the period before the consolidation does appear to affect the likelihood of a sustained consolidation (columns 5 and 6). perhaps because strong credit growth may subsequently require retrenchment from the private non-financial and financial sectors, we find that higher private sector credit growth in the years prior to the consolidation tends to reduce the duration of the consolidation. a similar role for boom-bust dynamics may explain why strong growth in equity prices in the years before a consolidation episode also tends to reduce the duration (column 7). that finding might reflect the suggestion from mian, sufi and trebbi (2012) that in a post-boom phase countries become highly politically polarized following financial crises, and as a result, sustained consolidation efforts are more difficult. we do not find the dummy for financial crises to be significant (column 8), which might be explained by two factors. first, the impact of the financial crisis on the duration of subsequent consolidation efforts may simply be seen most through the deleterious effect on the fiscal position which then push governments towards consolidation. lodge and rodríguez-vives ● how long can austerity persist? 17 second, it may be that following a financial crisis, the state of private sector balance sheets (captured in other variables in our model) is a more important determinant of how effectively governments can tackle the subsequent need for retrenchment. table 5: determinants of duration of consolidations – the role of macroeconomic conditions (estimates of hazard function using weibull model, coefficients shown as hazard ratios) (1) (2) (3) (4) (5) (6) (7) (8) duration of the episode 0.765*** 0.832*** 0.831*** 0.856*** 0.770*** 0.780*** 0.768*** 0.759*** capb 0.108** 0.164*** 0.080* 0.080** 0.117** 0.073 0.096* 0.107** government interest burden 0.126*** 0.175*** 0.092 0.126** 0.156*** 0.127** 0.113** 0.126*** real long-term interest rates -0.181*** -0.178*** -0.177*** -0.222*** -0.173*** -0.168*** -0.230*** -0.187*** current account balance -0.081*** private sector net lending -0.081*** private saving ratio -0.080* credit to gdp ratio -0.005 change in credit to gdp ratio 0.063* real equity price growth 0.020* financial crisis 0.324 constant -1.729*** -1.742*** -1.827*** -0.239 -1.254*** -1.956*** -1.789*** -1.717*** number observations 242 228 228 232 234 232 237 232 number of countries 20 19 19 20 19 19 20 20 number of consolidations 61 57 57 57 58 57 59 61 adjusted-r-squared 0.265 0.296 0.302 0.321 0.275 0.275 0.272 0.252 (9) (10) (11) (12) (13) (14) (15) (16) duration of the episode 0.794*** 0.774*** 0.761*** 0.807*** 0.886*** 0.765*** 0.802*** 0.759*** capb 0.100** 0.091* 0.101* 0.065 0.076* 0.113** 0.155*** 0.110* government interest burden 0.146*** 0.128*** 0.125** 0.104* 0.220*** 0.125*** 0.178*** 0.122*** real long-term interest rates -0.187*** -0.162*** -0.170*** -0.191*** -0.155*** -0.181*** -0.108** -0.163*** gdp growth -0.101 world gdp growth 0.363 output gap 0.061 real exchange rate 0.026** relative ulc growth 0.201*** openness -0.001 inflation 0.084*** monetary policy interest rate -0.020 constant -1.488*** -3.155*** -1.713*** -4.500*** -1.770*** -1.683*** -2.253*** -1.726*** number observations 240 232 242 232 228 242 240 240 number of countries 20 20 20 20 19 20 20 20 number of consolidations 60 61 61 58 56 61 60 60 adjusted-r-squared 0.261 0.267 0.264 0.296 0.342 0.257 0.286 0.247 notes: ***, ** and * indicate significance at the 1%. 5% and 10% level respectively. each variable is measured in the year before the consolidation begins. the “duration of the episode” reports the value of the weibull parameter that governs the shape of the hazard function. the coefficients are reported as hazard ratios – a coefficient greater than zero indicates that an increase in the explanatory variable increases the hazard rate (the conditional probability of a consolidation coming to an end). the measure of explained variation is an adjusted r-squared employed for survival models (royston (2006)). we tested a number of other variables. the role of growth dynamics is unclear: a higher real gdp growth rate tends to increase the probability of a consolidation continuing but a larger output gap lowers the probability of a sustained consolidation. moreover, none of the coefficients are statistically significant (columns 9 and 10). in combination with other variables we find that stronger real gdp growth at the onset of a consolidation period is typically associated with a longer-lasting consolidation effort. the international environment – proxied by world real gdp growth – played no significant role. the important role for competitiveness is confirmed: a more competitive real exchange rate and smaller european journal of government and economics 2(1) 18 increases in unit labour costs (relative to other advanced economies) in the lead-up to a consolidation tend to increase the likelihood that the consolidation will last (columns 12 and 13). we find no additional explanatory power from the openness of an economy, measured as the ratio of exports to gdp (column 14). moreover, stronger inflation in the years preceding the start of consolidation tends to reduce the length of subsequent consolidations (column 15). but, the coefficient on the monetary policy interest rate is insignificant (column 16). this is line with the literature, which does not find a uniformly significant role for monetary policy in affecting the length of consolidation periods. table 6: determinants of duration of consolidations – fiscal policy and the fiscal framework (estimates of hazard function using weibull model, coefficients shown as hazard ratios) (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) duration of the episode 0.765*** 0.832*** 0.831*** 0.856*** 0.770*** 0.780*** 0.768*** 0.794*** 0.772*** 0.761*** capb 0.074 0.074 0.111 0.011 0.070 0.367*** 0.126 0.074 0.068 -0.008 government interest burden 0.242*** 0.267*** 0.288*** 0.297*** 0.246*** 0.304*** 0.277*** 0.236*** 0.234*** 0.116 real long-term interest rates -0.115* -0.109 -0.122* -0.149* -0.118* -0.149** -0.119* -0.128* -0.117* -0.683*** current account balance -0.080** -0.084** -0.088** -0.008 -0.079** -0.066* -0.088** -0.081** -0.081** -0.069 change in credit to gdp ratio 0.087** 0.102*** 0.095** 0.102** 0.086** 0.093*** 0.059 0.091** 0.087** 0.087 real exchange rate 0.037* 0.043** 0.046** 0.020 0.036* 0.028 0.031 0.035* 0.036* 0.122*** gdp growth -0.221* -0.228** -0.209* -0.058 -0.224* -0.212* -0.211* -0.263** -0.221* 0.002 inflation 0.104** 0.116** 0.125** 0.146*** 0.105** 0.111** 0.100** 0.087 0.100* 0.209 consolidation composition: expenditure share 0.001 revenue share -0.003 temporary measures share 0.012** track record in consolidation 0.032 intensity of consolidation 2.461*** frontloading 0.150 change in debt 0.024 election 0.172 fiscal rules 0.411* constant -6.055*** -6.585*** -6.694*** -5.234** -6.049*** -5.689*** -5.462** -5.634** -6.052*** -16.100*** number observations 210 200 200 173 210 200 210 195 210 96 number of countries 19 19 19 18 19 19 19 19 19 13 number of consolidations 51 51 51 44 51 51 51 49 51 22 adjusted-r-squared 0.394 0.393 0.404 0.472 0.385 0.476 0.395 0.384 0.387 0.627 notes: ***, ** and * indicate significance at the 1%. 5% and 10% level, respectively. each variable is measured in the year before the consolidation begins. the “duration of the episode” reports the value of the weibull parameter that governs the shape of the hazard function. the coefficients are reported as hazard ratios – a coefficient greater than zero indicates that an increase in the explanatory variable increases the hazard rate (the conditional probability of a consolidation coming to an end). we also report a measure of explained variation, an adjusted r-squared employed for survival models (royston (2006)). having analysed the macroeconomic backdrop to consolidation episodes, we turn to the role of fiscal policy choices. table 6 presents the estimations for duration of consolidation, again using the weibull model. we retain the set of initial conditions that proved to be significant and add different indicators of fiscal policy, including indicators of the composition of adjustment, the temporality of the adjustment, the size and pace of consolidation and the strength of the governance framework. a key finding in the literature is that the composition of consolidation – the split between expenditure and revenue measures – is important (e.g. maroto illera and mulas-granados (2008), von hagen et al. (2002)). in columns two and three we add variables that indicate the proportion of deficit reduction during the consolidation phase accounted for by changes in current expenditure, and current revenue. in contrast to several studies in the literature, we find the coefficients on the composition variables to be insignificant. once we have accounted for the role lodge and rodríguez-vives ● how long can austerity persist? 19 of initial macroeconomic conditions, the composition of the fiscal adjustment – the split between revenue and expenditure measures – does not appear to be a significant determinant of the duration of consolidation.13 but the dimension that does appear to play a role is the durability of the adjustment. our evidence suggests that consolidations lasted longer when one-off measures played a smaller role (column 4). one explanation is that the key to successful consolidation efforts is linked to the commitment of the government in approaching the task, seeking for a permanent rather than a temporary change to the fiscal position.14 the results for other features of fiscal policy choices are generally in line with the literature. we find that the intensity of deficit reduction – the average pace of improvement in the capb over the consolidation episode – is significant. longerlasting consolidation episodes were based on rather gradual improvements in government deficit positions. however, we find no evidence that the amount of frontloading of the consolidation effort is a signal of greater willingness or ability to sustain consolidation efforts for longer. neither do we find a significant role for a country’s track record of consolidation. elections do not appear to signal the end of a consolidation episode. examining the robustness of the results to assess the robustness of the results we checked variations to the definition of consolidation: first, we included in the sample consolidation periods lasting only one year; second we specified a higher starting hurdle – i.e. a greater initial change in the fiscal position (of at least 1.5% in the first two years); third we allowed for more gradual deficit reductions to qualify, with a lower initial threshold (only 0.5% over the first two years). these alternative definitions yielded respectively 101, 76 and 52 consolidation episodes. in general, in neither case did we see large changes to our findings. most variables that were significant using our main definition were also significant under these alternative specifications. in particular, we still find a strong role for the starting fiscal position: the initial deficit, interest burden and long-term real bond yields. most macro conditions also remained significant. in particular, the external position and competitiveness remained important indicators of sustainable consolidation. two exceptions, however, were in the roles of growth and inflation in the run up to consolidation, which were less consistently significant predictors of the duration of consolidations using these alternative definitions. finally, we also investigate whether adjustments to our sample affected the results. one concern might be that the outliers – in particular the three long consolidations in belgium, ireland and japan – have affected the results. however, rerunning the estimations excluding these observations, we found the results to be very similar. 13 to check this finding, we also used a more detailed breakdown of government revenue and spending, separating the proportion of deficit reduction in each consolidation episode by types of expenditure and revenue. following von hagen et al. (2002) we added to each equation a variable showing the contribution of a particular budget item to the total deficit reduction achieved during the consolidation episode. all but one coefficient (that on capital expenditure) is statistically insignificant suggesting that there is overall not strong support for the view that the composition of deficit reductions is an important determinant of the length of consolidation periods. we also investigated different definitions for the fiscal composition – e.g. using a dummy variable to define whether the adjustment was predominantly expenditureor revenue-based etc. in none of the cases were the variables statistically significant. 14 however, the role of the fiscal rules could not be demonstrated, as the analysis (column 10) shows a curious result that stronger fiscal rules appear to increase the likelihood of a consolidation episode ending. moreover, with the series for the fiscal rules strength index available only for eu countries from 1990 onwards, the sample of consolidation episodes is considerably smaller, and therefore the results are limited. european journal of government and economics 2(1) 20 the role of push factors and pull factors in euro area countries to illustrate the insights from our work, we use our results to illustrate some of the challenges that lie ahead for some euro area countries that have embarked on a consolidation path during 2010-11. while, for several member states the push factors are obvious, with high public debt levels, large deficits, high sovereign bond yields and rising interest burdens forcing governments to take strong action towards fiscal austerity, there are clear challenges too. this is particularly true for those countries suffering from twin deficits of large fiscal and current account imbalances and cumulated competitiveness problems. in this sense, figure 2 shows survivor curves for the duration of the current fiscal consolidation efforts in countries under eu/imf economic and financial programmes (ireland, greece and portugal) as well as spain and italy. figure 3 shows a decomposition of the factors affecting the estimated probability of a duration lasting at least five years. the probability estimates are based on the equation which includes the initial fiscal conditions – deficit, bond yields and interest payments – and macroeconomic capacity – current account, real effective exchange rate, recent credit developments and growth and inflation dynamics. to provide a common overview, the estimated starting point is 2010. the estimated probabilities that these countries will be able to maintain a sustained consolidation effort are generally quite low. that should not be a surprise: the estimates are derived from analysis of past experiences and previous examples of long-lasting consolidation are relatively rare, as figure 1 highlighted. for example, from our sample, the unconditional probability of consolidation lasting longer than five years is under a half. figure 2: estimated survivor curves for current fiscal consolidation efforts in selected euro area countries (probability of consolidation lasting to particular year) 0.0 0.2 0.4 0.6 0.8 1.0 0 5 10 average experience ireland portugal greece spain italy notes: survivor functions estimated using weibull distribution assumption for baseline hazard (see table 6, column 1). the “average experience” shows the average estimated probability assuming all explanatory variables are at their mean. other curves show estimates of the probabilities for ireland, greece and portugal based on current developments. figure 3: factors affecting current probabilities of achieving long-lasting consolidation in selected euro area countries (contribution to estimated probability of consolidations starting in 2010 lasting five years or more) -0.5 0.0 0.5 1.0 ireland portugal greece spain italy capb govt. interest payments long-term real int. current account balance real exchange rate gdp growth inflation prob> five years notes: survivor functions estimated using weibull distribution assumption for baseline hazard (see table 6, column 1). the “average experience” shows the average estimated probability assuming all explanatory variables are at their mean. the other contributions show the effect of each variable relative to the mean experience. however, there are interesting differences in the estimated probabilities for each country which highlight the potential varied pressures in each situation. for instance, according to the estimates, ireland appears considerably more likely to maintain fiscal consolidation than the other countries from the sample. some of the drivers for ireland are the ‘push’ factors that make consolidation a necessity, such lodge and rodríguez-vives ● how long can austerity persist? 21 as the high primary deficit, relatively heavy interest burden and high long-term real bond yields. yet the recent improvement in ireland’s competitive position, with relatively low inflation and unit labour cost growth, as well as the return towards current account balance, could also be expected to help ireland’s consolidation effort. these rebalancing efforts in the irish economy are further supported by the return to the capital markets during 2012. by contrast, greece suffers from continued poor competitiveness (a high real exchange rate and continued high unit labour cost growth) which, combined with a large current account surplus, underscores the apparent difficulty for the government to achieve lasting fiscal consolidation. the relatively low likelihood shown for greece in figure 2 was confirmed by the need to restructure its government debt in spring 2012. portugal’s problems are also related to lack of competitiveness and continued high current account deficit. another finding is that countries with less dominant ‘push’ factors (lower interest payments like spain or stronger capb position like italy) show also a relatively lower probability of sustaining long consolidations. as discussed above, one needs to recall that this analysis uses as the starting point the situation in 2010.15 developments during 2011 would perhaps have heightened concerns that these countries also needed to implement consolidation in the coming years. moreover, the estimates omit some important elements to the current efforts, such as the role of the institutional framework of the eu/imf programmes in sustaining consolidations.16 conclusions this article has presented analysis of the factors affecting the duration of past consolidations in advanced economies. our analysis suggests that the starting point matters – the fiscal and macroeconomic conditions appear to influence the success of governments in sustaining lengthy consolidations. in line with the literature, we find that governments often appear to have been ‘pushed’ towards sustained fiscal consolidations – the duration of consolidation is driven by the need to adjust course: longer consolidations occur when public debt is high, fiscal deficits are large, the interest burden heavy and long-term bond yields elevated. our contribution to the literature is to put considerably more emphasis on the macroeconomic backdrop to consolidation episodes. we find that other factors, what we call pull factors, affect the duration of consolidations: a countries’ capacity to change course or its room for manoeuvre is important. higher initial private sector savings and a stronger external balance appear to help governments sustain longer-lasting consolidations. the important role for competitiveness is underlined: a more competitive real exchange rate in the lead-up to a consolidation tends to increase the likelihood of a sustained fiscal adjustment. finally, in contrast to some studies in the literature, we also find that there is less prominence for factors that might indicate a government’s willingness or resolve in its approach to consolidation. in particular, once we have accounted for the role macroeconomic conditions, the composition of the fiscal adjustment – the split between expenditure and revenue measures – does not appear to be a significant determinant of the duration of consolidation. however, commitment to permanent, rather than temporary, fiscal adjustment is key. 15 the regression estimates use the situation in the year before consolidation began. in this framework, for an assessment of the consolidation prospects from 2011 onwards, it is therefore the situation in 2010 that affects the length of consolidation. 16 see, for example, gupta et al. (2004) who analyse a wide dataset for countries with imf-supported programmes in the 1990s. european journal of government and economics 2(1) 22 references afonso, antónio (2010) ‘expansionary fiscal consolidations in europe: new evidence’, applied economics letters 17(2): 105-109. agnello, luca, vitor castro and ricardo sousa (2012) ‘what determines the duration of a fiscal consolidation program?’, nipe working paper no 17. ahrend, rudiger, pietro catte and robert price (2006) ‘interactions between monetary and fiscal policy: how monetary conditions affect fiscal consolidation’, oecd working paper, no 521. alesina, alberto and silvia ardagna (1998) ‘tales of fiscal adjustment’, economic policy 13(27): 487-545. alesina, alberto and roberto perotti (1995) ‘fiscal expansions and fiscal adjustments in oecd countries’, economic policy 21(2): 205-248. alesina, alberto and roberto perotti (1997) ‘fiscal adjustments in oecd countries: composition and macroeconomic effects’, imf staff papers 44(2):210248. alesina, alberto, roberto perotti and josé tavares (1998) ‘the political economy of fiscal adjustments’, brookings papers on economic activity 29(1): 197-266. ardagna, silvia (2004) ‘fiscal stabilisations: when do they work and why’, european economic review 48(5): 1047-1074. arjas, e. (1988) ‘a graphical method for assessing goodness of fit in cox’s proportional hazards model’, journal of the american statistical association 83(401): 204–212. baldacci, emanuele, sanjeev gupta and carlos mulas-granados (2010) ‘restoring debt sustainability after crises: implications for the fiscal mix’, imf working paper no 232. barrios salvador, sven langedijk and lucio pench (2010) ‘eu fiscal consolidation after the financial crisis: lessons from past experiences’, european commission economic papers no 418. bracke, philippe (2011) ‘how long do housing cycles last? a duration analysis for 19 oecd countries’, imf working paper no 231. briotti, maria gabriella (2004) ‘fiscal adjustment between 1991 and 2002: stylised facts and policy implications’, ecb occasional paper series, no 9. buchanan, james m. and gordon tullock (1962) the calculus of consent: logical foundations of constitutional democracy, ann arbor, mi: university of michigan press. buchanan, james m. and richard e. wagner (1977) democracy in deficit: the political legacy of lord keynes. new york: academic press. caprio, gerard, daniela klingebiel, luc laeven and guillermo noguera (2003) ‘banking crisis database: an update of the caprio-klingebiel database (1996, 1999), october 2003’, url (consulted june 2013): www1.worldbank.org/finance/html/database_sfd.html. devries, pete, jaime guajardo, daniel leigh and andrea pescatori (2011) ‘a new action-based dataset of fiscal consolidation’, imf working paper no. 128. european central bank (2011) ‘past experience with sustaining large primary budget surpluses’, monthly bulletin, april, 94-95. european commission (2007) ‘public finances in emu 2007’, european economy 3/2007, directorate-general for economic and financial affairs. lodge and rodríguez-vives ● how long can austerity persist? 23 european commission (2011) ‘public finances in emu 2011’, european economy 3/2011, directorate-general for economic and financial affairs. giavazzi, francesco and marco pagano (1996) ‘non-keynesian effects of fiscal policies changes: international evidence and the swedish evidence’, swedish economic policy review 3(1): 67-112. guichard, stéphanie, mike kennedy, eckhard wurzel and christophe andré (2007) ‘what promotes fiscal consolidation: oecd country experience’, oecd working papers no 553. gupta, sanjeev, benedict clements, emanuele baldacci and carlos mulasgranados (2004) ‘the persistence of fiscal adjustments in developing countries’, applied economic letters 11(4): 209-212. hartwig lojsch, dagmar, marta rodríguez-vives and michal slavik (2011) ‘the size and composition of government debt in the euro area’, ecb occasional paper series no. 132. heylen, freddy and gerdie everaert (2000) ‘success and failure of fiscal consolidation in the oecd: a multivariate analysis’, public choice 105(1-2): 103124. hernández de cos, pablo and enrique moral-benito (2011) ‘endogenous fiscal consolidations’, documentos de trabajo del banco de espana no. 1102. international monetary fund (2008) ‘what happens during recessions, crunches and busts?’, imf working paper no 274. international monetary fund (2009) world economic outlook april 2009: crisis and recovery. washington, dc: international monetary fund. joumard, isabelle, makoto minegishi, christophe andré, chantal nicqand and robert price (2008) ‘accounting for one-off operations when assessing underlying fiscal positions’, oecd working paper no 50. kiefer, nicholas m. (1988) ‘economic duration data and hazard functions’, journal of economic literature 26(2): 646-679. larch, martin and alessandro turrini (2008) ‘received wisdom and beyond: lessons from fiscal consolidation in the eu’, european commission, economic papers no 320. maroto illera, r. and c. mulas-granados (2008) ‘what makes fiscal consolidations last? a survival analysis of budget cuts in europe (1960-2004)’, public choice 134(3): 147-161. mian, atif r., amir sufi and francesco trebbi (2012) ‘resolving debt overhang: political constraints in the aftermath of financial crises’, nber working paper no 17831. molnár, margit (2012) ‘fiscal consolidation: what factors determine the success of consolidation efforts?’, oecd economics department working papers no 936. reinhart, carmen and kennet rogoff (2008a) ‘is the 2007 us sub-prime crisis so different?’, nber working paper no 13761. reinhart, carmen and kennet rogoff (2008b) ‘banking crises: an equal opportunity menace’, nber working paper no 14587. reinhart, carmen and kennet rogoff (2009) ‘the aftermath of financial crises’, american economic review 99(2): 466-472. reinhart, carmen and kennet rogoff (2012) ‘debt overhangs: past and present’, nber working paper no 18015. european journal of government and economics 2(1) 24 royston, patrick (2006) ‘explained variation for survival models’, the stata journal 6(1): 83-96. von hagen, jürgen, andrew hughes hallett and rolf strauch (2002) ‘budgetary consolidation in europe: quality, economic conditions, and persistence’, journal of the japanese and international economies 16(4):512-535. von hagen, jürgen and rolf strauch (2001) ‘fiscal consolidation: quality, economic conditions and success’, public choice 109(3-4): 327-346. microsoft word ejge_04_2015-010.docx european journal of government and economics volume 4, number 2 (december 2015) issn: 2254-7088 134 the scope of public organisations with productive functions: insights from the inefficiency of italian local public transport ugo arrigo, university of milano-bicocca, italy giacomo di foggia, university of milano-bicocca, italy abstract it is a well-known fact that reforms necessary to enhance competition and make the economy more attractive play a major role in the government activity in europe. considering the italian state of affairs, this paper focuses on the inefficiency of public producers in the light of certain market and legal impediments. this paper describes key barriers that undermine the healthy functioning of an important industrial and service sector of the italian economy: specifically local public transport. this paper also sheds some light on this problem analysing the impact that a controversial regulation framework may have on an industry and suggests that the degree of liberalisation affects the return on investments. jel classification h11; h40; d78. keywords state-owned enterprises; efficiency; italy; local public transport; competition. european journal of government and economics 4(2) 135 1. introduction the economic literature on public economics has often analysed relationships between the government, society and the market, recurrently to evaluate the role of public organisations and the impact of public policies on the functioning of modern markets. indeed, over the past decades, there has been heightened interest in the organisation and efficiency of public organisations with vast consensus of essential key roles of modern governments that comprehend the protection of individuals’ freedom, the redistribution of resources and the mitigation of market failures (pollitt, 2004), accordingly, many scholars agree on the good reason for the existence of government organisations with productive functions. a relatively young area of research is the analysis of european integration on local governments alongside their sublevels. in this regard, it is generally acknowledged that public budgets in many european countries are under significant constraints. schmidt (2006) argues that the institutional structure has contributed to alter the balance of power among national institutions and weakened national decision autonomy. the efficiency of the state-owned enterprises of eu countries varies widely, however available information on their performance is fragile. nevertheless, significant even if minute steps forward are being made in italy, in fact, the publication of the database on soes (especially key accounts variables) has constituted a considerable advance over the previous situation, allowing the knowledge of the main characteristics of such firms, and facilitating the evaluation of performance according to the different industries. with this in mind, it is underlined that the performance can be strongly affected by a multitude of exogenous factors that can’t be observed from our analyses. at any rate, putnam et al. (1994) provide measures to evaluate institutional performance given that the quality of the institution is strongly correlated with growth (arnold et al. 2011). prior research provides useful inputs using different approaches and shows that new challenges arise for public organisations as they have to deal with rapid and accelerating economic change, increasing complexity of regulatory issues, new technologies and services, calls for openness, transparency and citizen participation (pitlik et al. 2012). reforms must be implemented and sustained during times when fiscal constraints and budget consolidation pressures become increasingly tighter since good legislative frameworks can promote the healthy functioning of the market (arnold et al. 2011). however, regulatory frameworks (i) may be flawed by various factors that undermine their public interest aims, (ii) may involve costs that exceed their expected benefits; (iii) may suddenly become obsolete because of technical developments or the evolution of demand. many economists tend to be inclined to foster policies aimed at increasing competition in order to increase efficiency (amaral et al., 2009; jonsson, 1984; niskanen, 1973). such competition can also be generated with the aid of continuous productivity and efficiency comparisons between public bodies which provide similar output. likewise, other authors have been stressing the importance of collaboration (entwistle and martin, 2005; hartley et al. 2013; osborne and strokosch, 2013; warner, 2011). in this paper the ownership of enterprises is also discussed, there is in fact renewed interest in the role that soes can play in promoting economic development, the relationship between soes and the government as a shareholder, and the criteria used in assessing options for investing in or disposing of state-owned enterprises. given the importance of a well-functioning sector and the presence of certain indications that competition in the italian local public transport struggles (boitani et al. 2009; ibl, 2013), we try to look at the reasons for these inefficiencies. arrigo and di foggia ● the scope of public organisations with productive functions 136 in an effort to provide these concepts with more details, we focus on the performance of the selected enterprises operating in it. as the industry is strongly regulated and the behaviour of companies needs to be assessed in the context of the existing regulatory framework, we also look in broad terms at aspects of the regulatory framework. we then champion some conditions to be achieved or defended, such requirements lies in the restriction of the spoil system, choice of management made from specific lists of candidates with a certified level of expertise and based on free-access comparative selection procedures and implementation of training schools and human resources development. appropriate policies could provide citizens with efficient services while at the same time may contribute to the creation of an open business environment that supports the competitiveness of the industry. that said, policies will ensure that services entrusted by open, transparent and non-discriminatory tenders, make available the public service obligations and the procedures for determining compensation, and define precisely the subject of the tender and the amount of investments. the results of our analysis suggest a positive relation between open business environment and productivity. nevertheless this relation is only tested using the return on equity (roe) as dependent variable consistently with the italian ministry of economics. we find that soes operating in a highly liberalised market (energy) outperform those operating in local public transportation. the remainder of this paper is structured as follows: in the first section, we provide a review of related literature used to introduce and support our statements. after that, the background of the italian situation is presented; this is followed by the introduction of the italian local public transport market. next, the empirical analysis is performed. the discussion of the results and conclusion follow. 2. review of relevant literature the evolution of the role of government in modern states has been well documented and arguably represents one of the current most thought-provoking topics (heady, 2001). a keystone in the classification of roles is the work of musgrave (1959); providing for adjustments in the distribution of income and the function of contributing to stabilisation, see habermas (1994), dahl (1947) and kettl (2002). similarly, bator (1958), newman (1974), brown, potoski and van slyke (2006), haque, (2001) describe the triple role of government: security of the individuals’ freedom, the redistribution of resources, and the intervention in market failures. while the first two roles – redistribution of resources and the intervention in market failures – derive from the functioning of the market, as explained in normative economic theory (buchanan, 1968; martimort and pouyet, 2008), the security of the individuals’ freedom is a condition sine qua non of modern states (brennan, 1977; nozick, 1974). it might be objected thought that despite many different approaches have been proposed in an effort to shed some light on this issue the identification of scope and functions is tough (hood, 2000). this is corroborated by raadschelders and lee (2011) that present a historical review of research and infer that more recent research has tended to focus on different methods, mainly regarding organisation, metrics and performance (frederickson et al 2003; grandy, 2009; ostrom and ostrom, 1971). a critical aspect however of many studies aimed at assessing the performance of public entities in complying with their mandate is the comparability of methodology and results characterised by remarkable heterogeneity. in this sense, afonso et al. (2005) compute public sector performance (psp) and public sector efficiency (pse) indicators and find differences that lead the authors to suggest paths to public savings. to the same token dabla-norris et al. (2012) provide a public investment efficiency index that captures the institutional environment underpinning public investment management across four different stages: project appraisal, selection, implementation, and evaluation. these studies suggest that european journal of government and economics 4(2) 137 governments besides the public sector play an important role in the economic life of the state through the revenue and expenditure measures. in the same fashion of the general roles outlines above the essential economic purposes orbit around are taxation and expenditure measures, stabilisation policies (required by the government to affect the levels of aggregate demand) and allocation, i.e. the provision of public goods that cannot be provided through market mechanism. in certain cases, e.g. if there is market failure, government may be able to provide the good at a more optimal level and price if compared with the market equilibrium. as a matter of fact the economic mandate justifies a productive function, eventually, in the form of public producers, both providing market and non-market goods and services. nonetheless, other models of provision of market and non-market goods and services exist, e.g. public-private partnerships are growing in popularity as a governing model (forrer et al., 2010). as a producer of goods and services, public administration plays a significant role for the well-being of citizens and the competitiveness of countries. hence, the questions of which factors improve performance and raise capacities of a public administration to provide goods and services efficiently are at the heart of all discussions about public sector reforms (pitlik et al., 2012). to this regard a great deal has been written about the politics-related factors that impact on performance of firms (al-obaidan and scully, 1992; bottasso and sembenelli, 2004; sapienza, 2004; shleifer and vishny, 1994; zheng et al. 2003). amaral (2008) analyses the determinants of the performance differential between the private and public management of urban public transport in europe. according to the author, private management is associated with lower operating costs; however, the differential can depend on the interaction of the local authorities' capacity for expertise and the private operators' autonomy margin. in terms of total costs, no systematic advantages should be expected from private management. despite significant regional variation both between and within different regions, there is widespread consensus on the scope of soes, for example, to correct market failures, provide public goods, and foster economic development (kowalski, büge, sztajerowska and egeland 2013). the previously mentioned productive functions justify certain public producers (arrigo and di foggia, 2013; dewenter and malatesta, 2001; toninelli, 2000). although the government as a regulator and owner of assets opens a possibility of favourable treatment granted to soes (kowalski, büge, sztajerowska and egeland, 2013), it may be argued that there is, however, still little reliable information on their efficiency and on the advantages they may be enjoying because of their ownership since soes increasingly compete with private firms. firms frequently interact with public administration in many different ways and in a variety of different circumstances. in addition, firms indirectly rely on public administration outputs. thus, an efficient public administration that efficiently provides public goods and services and that absorbs relatively few public resources affects productivity and competitiveness of firms (afonso et al. 2005; allum, 1974; calandra, 1978; fratianni and spinelli,1982; ongaro, 2004; torres, 2004). other authors provide pieces of evidence on the causes of inefficiency (capano, 2003; cassese, 1993; del monte and papagni, 2001; tanzi, 1998). bandiera et al. (2008) estimate the cost of active (its presence entails direct or indirect benefit for the public decision-maker) and passive (its presence does not advantage the public decision-maker) waste from observed costs of public services. the authors assess the impact on how efficiently a certain public service is provided. their model focused on generic goods purchase which account for 40 per cent of public spending for goods and services. their findings suggest that the difference in prices across public bodies is principally due to variation in passive arrigo and di foggia ● the scope of public organisations with productive functions 138 (83 per cent of total) rather than active waste. based on average total cost (frank, bernanke and johnston, 2007) show that productive inefficiency implies an excess of production costs. other noteworthy and widely investigate elements that concur to determine the average low performance of italy are both clientelism (caciagli, 2006; della porta and vannucci, 1997; garcía, 2013; guzzini, 1995) and corruption. also results of a recent study (fiorino et al. 2012) confirm evidence of a negative correlation between corruption and growth. at the time of writing a controversial topic is the underperformance of local public transport or lpt sector. it is generally acknowledged that public budgets, including those dedicated to cover public transport, are under significant constraints. recent austerity measures have affected this sector too. between 2010 and 2012, the industry witnessed an overall decline in resources (600 million) or -12 per cent of public contributions. the reduction of public transfers has triggered a general tendency to raise the price of tickets. according to a recent fact-finding investigation of the italian competition authority (agcom, 2014), the ongoing situation of general inefficiency and lack of financial resources is largely attributable to the ownership, the management objectives and the regulatory framework that is layered and not always consistent. as highlighted by asquer (2011), local public transport has traditionally been operated by local state-owned enterprises – soes. boitani et al. (2009) investigate how the ownership and the procedure for the selection of firms operating in the local public transport sector affect their productivity. in order to compare different institutional regimes, they carry out a comparative analysis of companies operating in large european cities. the authors find that when firms are totally or partially in public hands their productivity is lower. moreover, firms selected through competitive tendering display higher total factor productivity. however there are cases of soes success worldwide and some soes are closing the gap with their private-sector competitors. as far as we know, there are, however, few studies that specifically analyse soes based on profitability measures. bovaird and löffler (2009) provide some highlights on key concepts of management and performance measurement that include profitability ratios. in fact the profit made by organizations can be effectively evaluated when compared to the amount of resources and activity required to generate it. return on investment (roi) is a widely used measure of overall performance and there are two key approaches to roi (walsh, 2003). the first approach is based on the return on total assets (rota) or return on capital employed (roce) which focuses on operating efficiency of the total enterprise. the second approach, employed in this paper, is based on the return on equity (roe), which concentrates on performance efficiency as translated into return to shareholders. it is anticipated that this approach opens to several interpretations. 3. background 3.1. unsuitability of regulation prevents the healthy functioning of local markets a breakthrough in regulatory policy during the last decades has been the awareness that government objectives for the utility industries can also take benefit from facilitating competition (arrigo and di foggia, 2013). the european commission enforces competition rules towards a single market by ensuring that all companies compete equally and fairly (european commission, 2014a). this benefits consumers, businesses and the economy as a whole. the competitive market is primarily a selection mechanism of economic behaviour, evaluated in terms of efficiency, able to protect consumers and ultimately the general public from the effects of poor decisions made by producers. european journal of government and economics 4(2) 139 still, if competition does not seem possible with regards to public organisations that produce administrative acts, it may be introduced for soes that compete with private companies under specific rules and in viable sectors. this can introduce adequate competition for services that will be offered by many organizations using the same infrastructure, as it is not economically convenient to duplicate it (oecd, 2004). to this extent, arnold et al. (2011) underscore that regulation generally addresses public-interest concerns about market failures, including monopoly conditions, externalities and asymmetric information. in this context, a good regulation can promote competition in certain industries by ensuring that market power in natural-monopoly segments is not used abusively and by providing the correct incentives to business participants. however, regulatory frameworks may be flawed by various factors, thus (i) some regulations may drift away from their original public interest aims, (ii) regulations sometimes involve costs that exceed their expected benefits and (iii) technical development, the evolution of demand and progress in regulatory techniques can make the design of regulations obsolete. even if it does not create effective competition, the separation of production and distribution would make the introduction of forms of comparative competition between suppliers operating in different geographical areas possible. the same method may also be introduced in order to evaluate the performance of similar offices in the typically public area of the production of administrative acts. 3.2. degree of liberalisation the world economic forum defines competitiveness as the set of institutions, policies, and factors that determine the level of productivity of a country. at the roots of competitiveness are the institutional and microeconomic policy arrangements that create the right conditions for a wealthy business environment under which businesses can grow (european commission, 2014b). by analysing eight sectors, asquer (2011) suggests that the difficulty to implement liberalisation and regulatory reforms in italy may be explained by various concurrent forces, which have to do with: the rent-seeking behaviour, the rise of barriers to entry against competitors, and the risk of collusive practices between regulators and regulated. it becomes therefore important to shed some light on factors that impact the efficiency of italian public administration.table 1 gives an idea about the rank of liberalisation in different sectors in italy. for each sector a number of criteria are employed in order to assess the existence of barriers to entry in, organization within, and exit from the market. the existence of such barriers is inferred from a number of quantitative as well as qualitative indicators. among the former, concentration indices, price dynamics, the public vs. private ownership of the incumbents, and other structural indicators are considered. among the latter, the index of liberalisation looks at such indicators as the degree of unbundling for essential facilities in network industries, the existence of fiscal or legal discriminations against the newcomers, the regulatory quality, etc. arrigo and di foggia ● the scope of public organisations with productive functions 140 table 1: index of liberalisation (italy) sectors 2007 2008 2009 2010 2011 2012 2012/07* natural gas 48 54 52 55 62 64 +16 post 37 38 38 41 47 52 +15 electricity 63 65 69 71 72 77 +14 labour market 50 35 55 60 60 60 +10 motorway 32 32 29 29 28 40 +8 financial services 59 62 63 64 69 66 +7 professional orders 46 46 47 47 47 52 +6 italy (total) 47 47 49 48 49 52 +5 telecommunicatio n 40 35 39 41 42 45 +5 art 55 55 56 56 56 58 +3 inland revenue 44 43 52 48 48 47 +3 public administration 40 37 35 38 39 42 +2 local public transport 45 46 43 43 44 45 0 airways 66 70 68 60 62 65 -1 water 27 27 32 17 19 19 -8 tv 70 68 67 65 62 61 -9 railways 49 49 49 41 36 36 -13 *change over the period 2007 – 2012. source: ibl (2012), liberalisation index, page 41. table 1 also shows the overall rankings of the ibl liberalisation index, comparing this year’s rankings with those from the previous edition of the index, showing all sectors ranked together. as graph 1 shows the ibl index of liberalisation appears to be correlated with the world economic forum’s competitiveness index while it is negatively correlated with the oecd state control indicator (the indicator covers formal regulations in the area of state control of business enterprises). these relationships support the idea that the selected index captures factors that are important for developing the lpt industry. on the same basis we see the ibl liberalisation index is taken as a reference for measuring the liberalisation degree of two italian sectors, namely lpt and energy to be used in our analysis. european journal of government and economics 4(2) 141 graph 1: correlation between ibl's index of liberalisations, wef's competitiveness index and oecd state control indicator source: own elaboration based on ibl liberalisation index (ibl_lib), oecs state control indicator (oecd_stc), wef competitiveness index (wef; gci). reference year: 2013. 3.3. the ownership of organisations there are well-known differences between governance of public and private companies (rainey et al. 1976) and the concept of governance is a growing body of european literature (peters and pierre, 1998). although state-owned enterprises as government assets open a possibility of favourable treatment these organisations may cope with multiple, unclear, or conflicting financial and social objectives. political interference can prompt decisions that threaten a company’s financial goals; therefore, managers can find it thorny to match the private sector’s performance standards. the defining characteristics of such organisations are that they have a distinct legal form and they are established to operate in markets even in they may also have public policy objectives. to better comprehend the struggle of public management and decision-taking process one must remember that administrative action within the sphere of the state that produces administrative acts, is governed by rigid written procedures that are consistent with the justice/equality objectives that administrative action pursues. garrone et al. (2013), underline that management discretion is a significant source of inefficiency in municipal enterprises. public sector organisations are similar to private companies since they have hierarchical structures that impose mandatory allocations of productive factors. however, while in private companies personal authority as defined by arrow (1974) is dominant, in public organisations impersonal authority prevails, consisting of written rules and laws and regulations that are relatively rigid and stable over time (arrow, 1974). in this way, public sector organisations do not benefit from the advantage that private companies have, represented by non-rigidity and the open and only partially predetermined content of work contracts. in private companies, the rules and procedures are subject to the leadership, on the contrary, in public organisations, leadership is subject to rules. therefore, in the case of poor legal arrigo and di foggia ● the scope of public organisations with productive functions 142 systems, when producing services for citizens, managers slog to achieve dynamically efficient solutions. dwyer and edwards (2009) digest this concept: dealing with change in the external environment creates considerable challenges for managers. given the complexity of strategy formulation it is difficult for managers to analyse all aspects of their environment or establish precise objectives, consequently business strategies tend to be characterised by small strategic adjustments or “incrementalism”. however, these marginal adjustments of strategy within an organisation’s existing culture may lead to strategic drift reflecting strategies that are inconsistent with changes taking place in the external environment. besides, andrews et al. (2006) show that the organisational failure is to some extent attributable to difficult circumstances and management features such as weak leadership and poor performance management. ongaro and valotti (2008) identify some factors affecting implementation of public management reform: characteristics of the administrative tradition and features of the politico-administrative context; the design of the reform package and the coalition of interests sustaining the reform; behavioural triggers; and the building over the time of management capacity at the level of individual public sector organisations. according to the financial police, in 2013, fraud and waste amounted to €5 billion and (over 19,000 managers identified by the military). specifically, the complaints to the judicial authorities, concerning crimes against the public administration, were more than 4,300 among others: bribery, extortion, embezzlement, misappropriation of funds, and abuse of office. the offences were put in place by directors, officers and civil servants alike. additionally, it is worth noting that, results of 25,000 inspections performed in 2013 show that among 1,704 public employees (total public employees amounted to 3.3 million) were sued for incompatibility with other assignments or jobs (gdf, 2013). state-owned enterprises may be overmanned to please employees that are voters too (boycko et al. 1996). non-commercial purposes can also make a firm vulnerable in the wage-bargaining process (haskel and szymanski, 1992). willner (2001) presents conditions under which political interference yields higher welfare than under commercial purposes, and vice versa. this supports a further efficiency gap of public organisations linked to the control of the organisation by politicians, resulting in resources for institutional duties instead being used by the political class, above all to bestow private benefits to constituencies (boubakri et al. 2008). the data presented in table 2 relate to 6,151 companies out of a total of 7,065 participated. in particular, for about 700 companies no data were available and so they were excluded. nevertheless, according to spicer (2010), scholars have too often been condescending towards the linkages between politics and governance, viewing political activities and interests as opportunities for corruption, mismanagement, and skewed priorities. table 2: participated companies (by local administrations) performance number of companies percentage operative result (billion €) average share (participation) profit 2879 47% 1.413 29% neutral* 1249 20% 60% loss 2023 33% -2.214 15% total 6151 100% -0.801 * companies with a loss or a gain up to 0.5% of net income are neutral: -0.5% < net income/asset < 0.5%. source: mef (2013) 3.4. the lpt industry at the time of writing a controversial issue is the apparent underperformance of local public transport sector. according to a recent fact-finding investigation of the european journal of government and economics 4(2) 143 italian competition authority (agcom, 2014), the ongoing situation of general inefficiency and lack of financial resources is largely attributable to the ownership and the management objectives. the report highlights several critical issues, among others: the organisation and procedures to select service operators (in many cases, the transport services are entrusted to state-owned inefficient companies or are managed on the basis of repeated extensions); the lack of transparency, efficiency and regulation of contracts, the overlap of activities of entrusted operators: monopoly and competition and problems that arise with respect to the determination of compensation, often awarded on the basis of criteria not aimed at efficient costs. the inefficiency of soes can generate losses other than the operating ones (€1.2 billion in 2012 according to the italian competition authority). even if almost all industries in which soes operate underperform, remarkable losses refer to the lpt (€300 million in 2012), other nonoperating losses are funded indeed by service contracts and government transfers (to current or capital account), or by citizens through tariffs based on the service total cost, e.g. the waste management sector. as highlighted by asquer (2011), local public transport has traditionally been operated by local soes. in addition, the regulatory framework is layered and not always consistent. as per the market structure one may note that the lpt sector is highly regulated, and differently from other european countries, on the supply side, there are plenty of smes. a european comparison shows that in italy the market share of the top three operators is equal to 26 per cent compared to 56 per cent in the uk, 77 per cent in france and 40 per cent in sweden, and that 88 per cent of enterprises have fewer than 100 employees. other data are shown in table 3. amaral (2008) analyses the determinants of the performance differential between the private and public managements of urban public transport in europe. according to the author, private management is associated with lower operating costs; however, the differential can depend on the interaction of the local authorities’ capacity for expertise and the private operators’ autonomy margin. in terms of total costs, no systematic advantages should be expected from private management. boitani et al. (2009) investigate how the ownership and the procedure for the selection of firms operating in the local public transport sector affect their productivity. in order to compare different institutional regimes, they carry out a comparative analysis of companies operating in large european cities. the authors find that when firms are totally or partially in public hands their productivity is lower. moreover, firms selected through competitive tendering display higher total factor productivity. however, there are cases of soe success worldwide and some soes are closing the gap with their private-sector competitors. table 3: key data of the lpt industry tpl industry values (≈) unit of measure companies 1150 number employees 116500 number turnover 10 billion € kilometres year (billion) 5.4 billion km fleet (n°) 50000 number buses 92 share (%) other means 8 share (%) source: own elaboration it is worth noting that 82 per cent of the production value comes from soes, which run 95 per cent of the urban transport services and 75 per cent of suburban public transport services. in 2013, most of the companies generated losses, with worse outcomes compared to those of other european countries (uk, germany, france, sweden, the netherlands and belgium). in the segment of road transport, the average ebit (-1.1 per cent) is poorer than the european average (3.5 per cent). the comparison with other countries is also negative considering the revenue per km of traffic (€1.08 against a european average of €1.34) and the public contributions €2.2/km against a €1.4/km european average of (agcom, 2014). arrigo and di foggia ● the scope of public organisations with productive functions 144 table 1 shows that while the italian situation increased from 2007 to 2012, local public transport did not. it is important to emphasise the referendum held in 2011 that has affected the existing rules of local public services and in particular on local public transport (dpr, 2011). the mentioned referendum has limited the floor for fair competition in the field of local public transport. furthermore, the structure of the italian lpt industry is characterised by fragmentation as several firms operate in relatively small user basin areas. asquer (2011) advises that, considering the italian environment, lpt has been traditionally run by local state-owned enterprises. graph 2: ibl liberalisation index – focus on energy and lpt source: own elaboration on ibl (2012). energy = electricity + gas 4. empirical analysis our primary source of data is the database of companies wholly or partially owned by the government (at different levels) published in 2014 by the ministry of economy and finance (mef). this choice was aimed at getting a general picture of the performance of the firms listed in the database. for our analysis, we managed to retrieve additional financial information and contextual variables from reliable sources: istat as per contextual variables and amadeus, a database of comparable financial information for public and private companies. we decided to omit variables related to industrial output and to take into consideration industrial activities and location to group the information. the statistical technique employed was ols regression. one of the principal challenges inherent in this approach – besides the assumption that the data are independent – was the lack of a definitive and universally accepted variable to be used as an object for the evaluation or dependent variable; thus we decided to employ the same index of the mef report; the return on equity roe. serious precautions were taken accurately and thoroughly to screen, clean and model the data in an effort to enhance results: 1,554 firms out of 3,500 used. the main limitation of the variables omission is the amount of variance accounted for in the response variable by the predictors. rsquared in fact indicates that the model accounts for a very low variance. however, provided the purpose of this paper, major benefits of models predominate. our work clearly has some limitations: sample, methodology, and absence of a time element to foster causal relations. 4.1. variables the key variable for the performance appraisal of the firms is roe, which is the most valuable indicator. it includes both dividends paid to shareholders and the retained profit which technically belongs to shareholders and which generates future capital growth (pizam, 2010). since the roe is useful for comparing the -5 0 5 10 15 20 2007 2008 2009 2010 2011 2012 a bs ol ut e ch an ge italy (total) local public transport energy european journal of government and economics 4(2) 145 profitability of a company to that of other firms in the same industry, we add some dummy variables to define both industrial activities and geographical distributions. as follows, the used variables accompanied the meaning.  roe: return on equity (retrieved for 2011, 2012 and 2013)  ep: (operating revenue/cost of employees)*100  empl: log of employees 2012  prco12: log of total production cost  prva12: log of total production value  derd: debt equity ratio 2012 – debt equity ratio 2011  av12: added value per capita  cl12: cost per employee  gdp12: gross domestic product at nuts3 level (italian provinces)  pop: population of the municipality  alt: log of altitude (metres)  net: dummy variable, it takes a value of 1 for companies that operate in a typical network public service (energy, gas, water, lpt, waste) 0 otherwise.  energy: dummy variable, it takes a value of 1 for companies that operate in the energy sector “energy” , 0 otherwise  tpl: dummy variable, it takes a value of 1 for companies that operate in the lpt sector, 0 otherwise.  ncs: 1= north, 2=centre, 3=south and islands (sicily and sardinia). 4.2. evidence table 4 presents correlations (2012 and 2013) for return of equity and financial output per employee, pe. weak even if interesting features emerge from a reading of this table. profitability of the analysed companies slightly decreases in southern regions, sicily and sardinia. table 4: correlation of some variable with geography variables roe13 roe12 ep13 ep12 ncs roe13 1 roe12 0.44 1 ep13 0.13 0.10 1 ep12 0.12 0.14 0.88 1 ncs -0.02 -0.06 -0.08 -0.06 1 source: own elaboration. similarly, correlations for independent variables used in the models are shown in table 5. unsurprisingly prco12 and prva12 are highly correlated, however this feature does not undermine the effectiveness. all other variables are pretty much uncorrelated. arrigo and di foggia ● the scope of public organisations with productive functions 146 table 5: correlation of dependent variables empl prco1 2 prva12 derd av12 cl12 gdp12 pop alt empl 1 prco12 0.86 1 prva12 0.85 1.00 1 derd -0.07 -0.02 -0.02 1 av12 0.03 0.37 0.40 -0.02 1 cl12 0.25 0.43 0.43 -0.01 0.54 1 gdp12 -0.04 0.03 0.04 0.05 0.13 0.03 1 pop 0.26 0.27 0.27 -0.01 0.08 0.18 0.00 1 alt -0.15 -0.15 -0.16 -0.01 -0.05 -0.10 0.12 -0.46 1 source: own elaboration our first estimation results are shown in table 6, which presents several alternative specifications of our reference model in column (1) that represents the starting point. in model (2) the variable net identifies the network public services, within model (3) the energy & gas sector is defined by the variable energy, while the model (4) is aimed at characterise the lpt sector. the model in column (1) provides broad preliminary support for our further assumptions regarding the determinants of roe. first, with respect to production cost prco12 and value prco12, we find that coefficients are as expected negative in the case of prco12 and positive considering prova12, both at the 1 per cent significance level. in line with what anticipated we find a negative coefficient associated with the number of employees empl, also (p<0.001). moreover, the gross domestic product per capita has a positive impact on the return of equity (p<0.001). a morphological characteristic, the altitude alt, though (p<0.1) negatively influences the performance too. it must be stressed that factors that do not show up as significant are by no means unimportant. european journal of government and economics 4(2) 147 table 6: model outputs (1) (2) (3) (4) variables roe roe roe roe empl -2.876*** -2.933*** -2.537*** -2.558*** (0.788) (0.789) (0.800) (0.795) prco12 -54.01*** -54.11*** -54.29*** -53.23*** (4.460) (4.462) (4.457) (4.461) prva12 57.06*** 57.06*** 56.92*** 56.11*** (4.746) (4.747) (4.741) (4.750) derd -0.00244*** -0.00245*** -0.00250*** -0.00241*** (0.000709) (0.000709) (0.000708) (0.000708) av -1.940 (1.308) cl12 0.136 (1.304) gdp12 6.351*** 6.578*** 6.425*** 6.103*** (2.202) (2.212) (2.202) (2.202) pop -0.113 (0.338) alt -0.659* -0.687** -0.718** -0.613* (0.340) (0.340) (0.340) (0.339) net 1.421 (1.223) energy 4.794** (2.056) tpl -6.390*** (2.328) constant -68.84*** -70.55*** -67.41*** -66.18*** (22.51) (22.57) (22.51) (22.51) observations 1,554 1,553 1,553 1,553 r-squared 0.152 0.153 0.155 0.156 standard errors in parentheses. non-significant variables other than dummies are only kept in column (1) *** p<0.01, ** p<0.05, * p<0.1 the results in model alternatives in table 6 are generally consistent with those in the model (1) and in particular reinforce our arguments. all of the variables that were significant in column (1) remain significant. overall, the results of the three regression analyses, columns (1), (2) and (3) with roe as dependent variable reveal a number of insights into what how the industrial activity (net stands for network public service, lpt because of its low liberalisation rank and energy since one of the most liberalised sectors) contribute to the measure of performance. considering the model in column (2) one may note the coefficient is positive though not significant. indeed, coefficients associated with the dummy variables energy, positive (p<0.05), and lpt, negative (p<0.01) are both significant. 5. discussion and conclusion examining our results, we speculate that while not all of the results were significant, the overall direction of results showed trends that could be helpful to learning about profitability of local public transport. indices that measure the efficiency of the participated companies can be an important stimulus to their performance improvement. the database published by the ministry of economics and finance reports a well-known index of efficiency, namely the return on equity roe. return on equity measures a corporation's profitability by revealing how much profit a company generates with the money shareholders have invested. noticeably profitability measures alone fail to capture in full the performance of soes because of the scope that these organisations pursue. in addition, although state-owned enterprises as government assets open a possibility of favourable treatment, managers of these organisations may cope with multiple, unclear, or conflicting financial and social objectives, therefore, they can find it thorny to match the private sector’s performance standards. as anticipated in the literature section arrigo and di foggia ● the scope of public organisations with productive functions 148 there are contrasting studies focused on the eventual political interference and on its effective ability to prompt decisions that threaten a company’s financial goals. our findings share a number of similarities with those reported in some of the papers above mentioned and in terms of the models presented in table 6, results support the theoretical assumption that different levels of liberalisation associated with different industrial activities affect the roe. apparently, a competitive environment with a sound regulatory framework may enable improvements in the profitability of firms, specifically the performance efficiency as translated into return to shareholders. in addition coefficients associated with the dummy variables energy, positive (p<0.05), and lpt, negative (p<0.01) are both significant, therefore as put forward, our results show different profitability of firms across different markets, regulatory frameworks and competition levels. given that our findings are based on a limited number of firms, the results from such analyses should consequently be treated with the utmost caution. important limitations might have influenced the results obtained, allowing alternative explanations. in particular, as previously indicated, it is possible that inefficiency may be driven more by omitted forces than by the variables highlighted in our theoretical and empirical arguments. as per the regulation of local public services, the current situation is at a standstill. although the goal is to move from a system of attribution of exclusive rights by issuing administrative orders to the liberalisation of the service, which entails the consequent free delivery, some critical aspects must be considered, among others the public service obligations to be observed by the operating entity, the definition of economic compensation and the issue of minimum standards (efficiency and quality) of services to be provided. we underline the need for high quality services granted in efficient and affordable procedures and providing all stakeholders with the required legal certainty. in this sense appropriate policies would provide people with efficient services while at the same time contribute to the creation of an open business environment that supports the competitiveness of the industry. as a premise, the simplification of the market access procedure, in terms of time required for authorisation is needed. at the same time the widely recognised lack of transparency in the current system impedes the efficient development of this market. we claim that the completion of the liberalisation and simultaneously the privatisation of a number of enterprises would pilot a better allocation of resources and consequently to an increase of the demand (apparently under the level of the mentioned european countries). as a matter of fact, according to oecd (2006), soes face explicit difficulties regarding their governance since are often effectively protected from two major threats i.e. threat of takeover and bankruptcy. data in table 7 seem to confirm that problems arise when control of public entities exceed the 50 per cent. we also agree with findings of the competition authority agcom (2014) who advises that the most important aspects for the market development are the proper definition of the areas of public service, both in terms of territory as regards the nature of the services offered and the definition of minimum services and the related question of the correct quantification of the compensation of the public service obligations. table 7: net income per employee and ebitda margin (italy eu 15) country public ownership ebitda net income per employee it >50.01% 8.05 -2.69 ≤ 50% 10.11 43.80 eu-15 >50.01% 11.18 18.39 ≤ 50% 13.24 74.39 source: own elaboration our results contribute to the literature on the governance of italian public transport by focusing on how different market structures influence the profitability of firms. results support leaders of public sector organisations as well as policy makers. european journal of government and economics 4(2) 149 our research suggests that, as per the profitability of firms within the lpt sector, soes could boost their productivity. however, companies and other organisations and policy makers need to address considerable challenges if they are to capture the full potential of such markets. 5.1. conclusion in this paper, we have highlighted that public organisations with productive responsibilities – public producers – may face multiple problems linked to the ownership, the management, the regulatory framework and the degree of competition of the markets in which they operate. this is done in conformity with the findings of the european commission (2014b) report on competition as it shows that competition enforcement helps promoting growth and competitiveness of states, prevents dominant companies from expelling competitors from the market, keeps markets open and efficient, and creates the conditions for lower input prices for the industry. as per the government expenditure related to soes (also in the form of state aid), policies aimed at enforcing competition help to steer public resources towards growth-enhancing objectives. indeed, on the one hand we have stressed that like unlawful government subsidies, unproductive government expenditure in soes can distort the level playing field, erect superfluous barriers and undermine the growth potential. on the other hand, we have also pointed out that government support and intervention can also have a positive impact when it is well-targeted. we advocate that efficiency and profitability problems arise in markets characterised by biases in some of the above mentioned aspects which threaten competition. this is particularly true in those markets where governments tend to intervene to protect incumbents. focusing on local public transport, we have highlighted critical aspects that have a negative impact on market development. specifically, the evidence of this study and the related literature points towards the idea that both exogenous factors like the regulatory framework and the level of competition coexist with endogenous factors (ownership and size) and mutually concur to the inefficiency of lpt. given these points, we have underlined that the results of our estimates reinforce our arguments. our results endow preliminary support for our assumptions. overall, the results of the three regression analyses reveal a number of signals on how the market structure impacts on the employed measure of performance, the roe. indeed, coefficients associated with the variables aimed at identifying the energy and lpt markets, confirm the better performance of firms operating in the energy market. the findings are not representative of the entire soes universe firstly because of the lack of consistent and robust information and secondly because in line with the opening decision this paper does not address in detail the potential additional affecting forces. the results however will be trusted considering the uniformity with most of the dedicated literature, though the findings cannot be generalised to other markets, countries or even measures of performance. our research suggests that policies aimed at containing government expenditure and in the meantime increasing the efficiency of the services provided will (i) ensure that service operators are entrusted by open, transparent and nondiscriminatory tenders and procedures, (ii) put at the disposal of all potential operators clear and sound rules to comply with the public service obligations and simplify procedures for determining compensation, and (iii) define precisely the subject of the tenders in order not to hamper considerably the participation of new entrants. the findings of this research have managerial implications too. governments should be able to select top-level management on soes who shall then have autonomy. after that, the selection of personnel (both at top and middle levels) should only be made from specific lists detailing candidates with certified expertise based on free-access comparative selection procedures. to do this, advanced and arrigo and di foggia ● the scope of public organisations with productive functions 150 professional schools for public management should be enforced, creating incentive schemes to encourage young people who are motivated and prepared for a public career. future work will concentrate on an international comparison of controlled companies and soes’ governance, regulation and performance in specific sectors by share of ownership. references afonso, antonio, ludger schuknecht and vito tanzi (2005) ‘public sector efficiency: an international comparison’, public choice 123(3-4):321-347. agcom (2014) audizione presso la ix commissione della camera dei deputati del presidente dell’autorità garante della concorrenza e del mercato professor giovanni pitruzzella. indagine conoscitiva sul settore del trasporto pubblico locale. allum, percy a. (1974) ?italy—republic without government?the world today’ 30(11): 448-459. al-obaidan, abdullah m. and gerald w. scully (1992) ‘efficiency differences between private and state-owned enterprises in the international petroleum industry’, applied economics 24(2): 237-246. amaral, miguel (2008) ‘public vs private management of public utilities–the case of urban public transport in europe’, research in transportation economics 22(1): 8590. amaral, miguel, stéphane saussier and anne yvrande-billon (2009) ‘auction procedures and competition in public services: the case of urban public transport in france and london’, utilities policy 17(2):166-175. andrews, rhys, george a. boyne and gareth enticott (2006) ‘performance failure in the public sector: misfortune or mismanagement?’, public management review 8(2): 273-296. arnold, jens m., giuseppe nicoletti and stefano scarpetta (2011) ‘regulation, resource reallocation and productivity growth’, european investment bank papers 16(1): 90-115. arrigo, ugo and giacomo di foggia (2013) ‘schemes and levels of state aid to rail indutry in europe: evidences from a cross-country comparison’, european journal of business and economics 8(3):14-20. arrow, kenneth j. (1974) the limits of organization. fels lectures on public policy analysis. london: ww norton & co. asquer, alberto (2011) ‘liberalization and regulatory reform of network industries: a comparative analysis of italian public utilities’, utilities policy 19(3):172-184. bandiera, oriana, andrea prat and tomasso m. valletti (2008) active and passive waste in government spending: evidence from a policy experiment. ceis working paper (115). retrieved from: http://ssrn.com/abstract=1115339. bator, francis m. (1958) ‘the anatomy of market failure. the quarterly journal of economics’ 72(3): 351-379. boitani, andrea, marcelo nicolini and carlo scarpa (2009) do competition and ownership matter? evidence from local public transport in europe. working papers siet 2009 issn 1973-3208. bottasso, anna and alessandro sembenelli (2004) ‘does ownership affect firms’ efficiency? panel data evidence on italy’, empirical economics 29(4):769-786. boubakri, narjess, jean claude cosset and walid saffar (2008) ‘political connections of newly privatized firms’ journal of corporate finance 14(5): 654-673. european journal of government and economics 4(2) 151 bovaird, tony and elke löffler (2009) public management and governance. taylor & francis: abingdon. boycko, maxim, andrei schleifer and robert w. vishny (1996) ‘a theory of privatisation’, economic journal 106(435): 309–319. brennan, willian j. jr.(1977) ‘state constitutions and the protection of individual rights’, harvard law review 90(3):489-504. brown, trevor l., mattew potoski and david m. van slyke (2006) ‘managing public service contracts: aligning values, institutions, and markets’, public administration review, 66(3):323-331. buchanan, james m. (1968) the demand and supply of public products (vol. 6). chicago: rand mcnally. caciagli, mario (2006) ‘the long life of clientelism in southern italy’, in kawata, j. (ed.) comparing political corruption and clientelism, 157-170, burlington: ashgate. calandra, piero (1978) storia dell'amministrazione pubblica in italia [history of italian public administration] (vol. 180), bologna: il mulino. capano, giliberto (2003) ‘administrative traditions and policy change: when policy paradigms matter. the case of italian administrative reform during the 1990s’, public administration, 81(4):781-801. cassese, sabino (1993) ‘hypotheses on the italian administrative system’, west european politics, 16(3): 316-328. dabla-norris, era, jim brumby, annette kyobe, zac mills and chris papageorgiou (2012) ‘investing in public investment: an index of public investment efficiency’, journal of economic growth, 17(3): 235-266. dahl, robert a. (1947) ‘the science of public administration: three problems’, public administration review, 7(1): 1-11. del monte, alfredo and erasmo papagni (2001) ‘public expenditure, corruption, and economic growth: the case of italy’, european journal of political economy, 17(1): 1-16. della porta, donatella and alberto vannucci (1997) ‘the ‘perverse effects’ of political corruption’, political studies 45(3): 516-538. dewenter, kathryn l.and paul h. malatesta (2001) ‘state-owned and privately owned firms: an empirical analysis of profitability, leverage, and labor intensity’, the american economic review, 91(1): 320-334. dwyer, larry and deborah edwards (2009) ‘tourism product and service innovation to avoid “strategic drift” 1, international journal of tourism research, 11(4): 321–335. entwistle, tom and steve martin (2005) ‘from competition to collaboration in public service delivery: a new agenda for research’, public administration 83(1): 233-242. esping-andersen, gosta (1990) the three worlds of welfare capitalism (vol. 6). cambridge: polity press. european commission (2014a) directorate-general for competition, competition policy in the european union. retrieved from http://ec.europa.eu/dgs/competition/index_en.htm. european commission (2014b. report on competition policy 2013 {swd(2014) 148 final}. fiorino, nadia, emma galli and ilaria petrarca (2012) ‘corruption and growth: evidence from the italian regions’, european journal of government and economics 1(2): 126-144. arrigo and di foggia ● the scope of public organisations with productive functions 152 forrer, john, james e. kee, kathryn e. newcomer and eric boyer (2010) ‘public– private partnerships and the public accountability question’, public administration review, 70(3): 475-484. frank, robert h., ben bernanke and louis d. johnston (2007) principles of economics. new york: mcgraw-hill/irwin. fratianni, michele and franco spinelli (1982) ‘the growth of government in italy: evidence from 1861 to 1979’, public choice 39(2): 221-243. frederickson, h. george, kevin b. smith and christopher w. larimer (2003) the public administration theory primer. boulder, co: westview press. garcía, césar (2013) ‘clientelism, economic structure, and public relations in southern europe: an example of diversity in the western world’,. public relations journal 7(2): 214-241. garrone, paola, luca grilli and xavier rousseau (2013) ‘management discretion and political interference in municipal enterprises. evidence from italian utilities’, local government studies 39(4): 514540. guardia di finanza (2014) rapporto annuale 2013 [annual report 2013]. grandy, christopher (2009) ‘the “efficient” public administrator: pareto and a well‐rounded approach to public administration’, public administration review, 69(6):1115-1123. guzzini, stefano (1995) ‘the ‘long night of the first republic’: years of clientelistic implosion in italy’, review of international political economy 2(1): 27-61. habermas, jürgen (1994) ‘three normative models of democracy’, constellations 1(1): 1-10. haque, m. shamsul (2001) ‘the diminishing publicness of public service under the current mode of governance’, public administration review, 61(1): 65-82. hartley, jean, eva sørensen and jacob torfing (2013) ‘collaborative innovation: a viable alternative to market competition and organizational entrepreneurship’, public administration review 73(6): 821-830. haskel, jonathan and stefan szymanski (1992) ‘a bargaining theory of privatization’, annals of public and cooperative economics, 63(2): 207–228. heady, ferrel (2001) public administration, a comparative perspective. new york: crc press. hood, christopher (2000) ‘paradoxes of public-sector managerialism, old public management and public service bargains’, international public management journal 3(1): 1-22. istituto bruno leoni (2012) indice delle liberalizzazioni 2012 [index of liberalisation 2012]. torino: ibl libri. jonsson, ernst (1984) ‘recurrent cost and performance comparisons within public administration: a method of generating efficiency-promoting competition’ public choice 42(2): 185-192. kettl, donald f. (2002). the transformation of governance: public administration for twenty-first century america. baltimore: johns hopkins university press. kowalski, przemyslaw, max büge, monica sztajerowska and matias egeland (2013) state-owned enterprises: trade effects and policy implications, oecd trade policy paper, no. 147, oecd publishing. european journal of government and economics 4(2) 153 martimort, david and jerome pouyet (2008) ‘to build or not to build: normative and positive theories of public–private partnerships’, international journal of industrial organization 26(2):393-411. ministero dell'economia e delle finanze (2013) rapporto sulle partecipazioni detenute dalle amministrazioni pubbliche al 31 dicembre 2011 [report of participation of public administrations by december 31 2011] musgrave, richard a. (1959) the theory of multi-level public finance. in proceedings of the annual conference on taxation under the auspices of the national tax association (pp. 266-278). national tax association. newman, meredith a. (1994) ‘gender and lowi's thesis: implications for career advancement’, public administration review, 54 (3): 277-284. niskanen, willian a. (1973) bureaucracy: servant or master? london: the institute of economic affairs. nozick, robert (1974). anarchy, state and utopia. new york: basic books. organisation for economic co-operation and development (2004) access pricing in telecommunication. paris: oecd publishing. organisation for economic co-operation and development (2006) comparative report on corporate governance of state-owven enterprise. paris: oecd publishing ongaro, edoardo (2008) ‘introduction: the reform of public management in france, greece, italy, portugal and spain’, international journal of public sector management 21(2): 101-117. ongaro, edoardo and giovanni valotti (2008) ‘public management reform in italy: explaining the implementation gap’, international journal of public sector management 21(2): 174-204. osborne, stephen p. and kirsty strokosch (2013) ‘it takes two to tango? understanding the co‐production of public services by integrating the services management and public administration perspectives’, british journal of management, 24(s1): s31-s47. ostrom, vincent and elinor ostrom (1971) public choice: a different approach to the study of public administration. public administration review 31(2): 203-216. peters, b. guy and john pierre (1998) ‘governance without government? rethinking public administration’, journal of public administration research and theory 8(2): 223-243. piattoni, simona (ed.) (2001) clientelism, interests, and democratic representation: the european experience in historical and comparative perspective. new york: cambridge university press. pitlik, hans, werner hölzl, christof brandtner, friedrich heinemann, florian misch, mustafa yeter, geert steurs, simon gagnage and kristof mertens (2012) excellence in public administration for competitiveness in eu member states. report for the european commission, dg enterprise and industry. pizam, abraham (ed.) (2010) international encyclopedia of hospitality management (second ed.). oxford: elsevier. pollitt, christopher. a. (2004) public management reform: a comparative analysis. new york: oxford university press. putnam, robert d., robert leonardi and raffaella y. nanetti (1994) making democracy work: civic traditions in modern italy. princenton: princeton university press. arrigo and di foggia ● the scope of public organisations with productive functions 154 raadschelders, jos c.n. and kwang-hoonlee (2011) ‘trends in the study of public administration: empirical and qualitative observations from public administration review, 2000–2009’, public administration review 71(1):19-33. rainey, hal g., robert w. backoff, and charles h. levine (1976) ‘comparing public and private organizations’, public administration review, 36(2):233-244. sapienza, paola (2004) ‘the effects of government ownership on bank lending’, journal of financial economics 72(2): 357-384. schmidt, v. a. (2006) ‘procedural democracy in the eu: the europeanization of national and sectoral policy-making processes’, journal of european public policy 13(5): 670-691. shleifer, andrei and robert w. vishny (1994) ‘politicians and firms’, the quarterly journal of economics, 109(4): 995-1025. spicer, michael w. (2010) in defense of politics in public administration: a value pluralist perspective.tuscaloosa: university of alabama press. tanzi, vito (1998) corruption around the world: causes, consequences, scope, and cures. international monetary fund working paper wp/98/63. toninelli, pier a. (ed.) (2000) the rise and fall of state-owned enterprise in the western world (vol. 1). new york: cambridge university press. torres, lourdes (2004) ‘trajectories in public administration reforms in european continental countries’, australian journal of public administration, 63(3): 99-112. walsh, ciaran (2003) key management ratios. master the management metrics that drive and control your business (3rd ed.). harlow: financial times prentice hall publishing. warner, mildred e. (2011) ‘competition or cooperation in urban service delivery?’, annals of public and cooperative economics 82(4): 421-435. willner, johan (2001) ‘ownership, efficiency, and political interference’, european journal of political economy,17(4): 723-748. zheng, jinghai, xiaoxuan liu and arnbe bigsten (2003) ‘efficiency, technical progress, and best practice in chinese state enterprises (1980–1994)’, journal of comparative economics 31(1):134-152. external economies of scale, government purchasing commitment and welfare improvements in the vaccines industry european journal of government and economics volume 1, number 2 (december 2012) issn: 2254-7088 163 external economies of scale, government purchasing commitment and welfare improvements in the vaccines industry pedro garcia-del-barrio, universitat internacional de catalunya, spain* abstract some industries exhibit external economies of scale. in these cases, government intervention may generate increases of welfare through tax-subsidy programmes and advanced purchasing commitments. the issue is initially examined for the case of competitive markets in a long-run equilibrium. then, taking the vaccines industry as an illustrative example, the paper proposes a suitable framework that is tailored to account for the most relevant characteristics of the real situation. the paper concludes advocating for intervention in competitive markets whenever economies of scale exist and implementing the policy is inexpensive. jel classification d31; d64; h2; l13 keywords external economies of scale; social welfare; tax-subsidy; income transfer; government intervention acknowledgements the author gratefully acknowledges financial support from the ministerio de ciencia y tecnolo gia (sej2007-67295/econ, spain). the study owes much to francisco galera and two anonymous referees. at the writing stage, i benefited from comments by zubin mistry. all remaining errors lie with the author alone. * address for correspondence : universitat internacional de catalunya, school of economics and social sciences, immaculada 22, 08017 barcelona, spain. tel.: +34 932541800. e-mail: pgarcia@uic.es. european journal of government and economics 1(2) 164 introduction this paper illustrates that social welfare can be improved by introducing taxsubsidy programs in markets with external economies of scale. in addition to this, the paper examines the vaccine industry as an illustrative example in which the issue can be tested in practice. the aim of the paper is thus twofold: first, to develop a theoretical model for proving that, under certain circumstances, intervention of competitive markets through redistributive income policies always lead to pareto-improving situations; second, to present an ad hoc theoretical framework with which to analyse the modern industry of vaccines. this will permit to apply some of the theoretical ideas of the paper to understand how the modern vaccines industry operates at the present. before we start the analysis, it is useful to record some basic theoretical principles. generally, to achieve the highest possible level of welfare, economic theory recommends avoiding intervention in competitive markets. however, in the presence of externalities, correcting this type of market failure is prescribed for the market to deliver the best outcome in terms of welfare.1 we are going to focus here on the external economies of scale affecting the supply-side of the market, which implies that the aggregate supply presents a negative slope in the long-run.2 in particular, our attempt is to show how public policies may affect the market in a way that the price diminishes as the size of the industry grows, thereby permitting that all the economic agents expand their welfare level. external economies of scale occur when a fall in unit costs arises from an expansion of an industry, without necessarily increasing the size of individual firms. of course, this is possible because any number of firms may enter the market in the long run. from a theoretical point of view, economists largely acknowledge the compatibility of external economies of scale with competitive markets: chipman (1970) provides interesting comments for the theoretical debate, while meade (1952, p. 33) makes it clear that perfect competition can prevail under conditions of increasing returns insofar as the economies are external to individual firms. this paper stresses the role played by external economies of scale in industries that, though being small in number, are nonetheless significant. among the examples mentioned in the literature, some affect prominent markets.3 the contribution of our analysis is the demonstration that, when discriminating among buyers is something possible, it is always realisable to design self-financing taxsubsidy schemes that expand the social welfare. in particular, the paper shows that government intervention in competitive markets may be welfare improving if external economies of scale are present. as regards the related literature, some authors, like thépot (2003), have tackled the issue within the context of imperfect competition. our approach deviates from 1 externalities are identified as one of the causes of market failures, whose distortions should be corrected through regulation. previous studies have granted support to the view that government intervention is justified in front of externalities; see for instance: chandra, franck and naqvi (2002), helpman and krugman (1985), and panagariya (1981). 2 another possible approach adopts a demand-side perspective: the so-called network economies of scale result in individuals enjoying better market conditions as new consumers enter the industry. although this is an interesting phenomenon, its analysis is left aside for the moment. notice that, even if the type of economies considered here are basically linked to the supply, our approach involves both sides of the market, since tax-subsidy schemes do actually alter the elasticity of demand. 3 trade theorists, dating back to graham (1923), had recognized the beneficial effects of real-locating resources to increasing returns to scale industries. see also ethier (1982) in this regard. for a description of external economies in markets for cultural products, such as books, motion pictures, newspapers, etc., see marvasti (1994). this paper explains how economies of scale derived from the population size lead to comparative advantage in trade. besides, the presence of external economies of scale in the cotton industry is documented by broadberry and marrison (2002). european journal of government and economics 1(2) 165 these studies in the fact that we examine perfectly competitive situations. thépot (2003) showed that government intervention (through income transfers, for instance) is able to correct the market failures associated with imperfect competition and to restore the pareto efficiency. yet, pareto-improving policies are easily explained when some degree of market power exists. this is due to the fact that regulations can generate greater efficiency by bringing the market equilibrium closer to the competitive one. in contrast with that, our results do not derive from correcting market failures associated with market power, but stem from the greater efficiency associated to the external economies of scale existing in certain industries. two seminal papers ought to be mentioned in this context. aoki (1971) shows first that a tax-subsidy system makes the competitive market mechanism work efficiently in various economic environments with externalities. then, osana (1977) goes beyond and proves that, in a wide class of economies with marshallian externalities, a pareto optimum can be sustained by a competitive equilibrium with the aid of a tax-subsidy system. the conclusions reached by these two papers are beyond the scope of our study. nonetheless, the model presented here deals with the topic in an original way and, more interestingly, permits concluding that selffinancing tax subsidy policies can always be implemented so that they lead to pareto-improving situations. the relevance of the theoretical points described so far depends on the possibility of them to occur in real industries. note first that our theoretical idea may grant support to the rationality of existing policies, such as granting home ownership vouchers in india or low-cost access to mobile phone technology in african countries. in particular, the applied section of the paper focuses on the case of the vaccines market, a peculiar part within the pharmaceutical industry. the rest of the paper is organized as follows. having motivated the topic in the introduction, section 2 proposes a basic theoretical model for approaching the issue in a competitive framework in the long-run. then, section 3 accommodates the analysis to the vaccines industry. this new framework, tailored to account for the chosen example, is apposite to illustrate how our theoretical idea applies to existing industries. finally, section 4 summarizes the main conclusions of the paper. the model the analysis of this section is carried out in a partial equilibrium context at the long run. we examine the case of a linear demand, even if other functional specifications may of course be also valid. consider a competitive industry with external economies of scale and identical firms. the total cost function of each firm is ct(x) = c(x )x, where x represents the production of the individual firm and x accounts for the total quantity traded in the industry. the market demand is given by p = d − dx, in which the quantity demanded at zero price has been normalized to one. notice that neither linearity or normalization of the demand function implies loss of generality. following thépot (2003), the analysis could be applied to a one-unit goods market. (this approach is particularly conclusive for highly-esteemed or indispensable goods; such as vaccinations, housing, or access to communications or education). we assume that a continuum of consumers with identical preferences exists whose disposable income for this good differs among them but is uniformly distributed. thus, the consumers’ reservation prices for one unit of good directly stems from european journal of government and economics 1(2) 166 their income, leading to a linear demand in which d represents the income of the richest consumer.4 in this context, the implementation of redistributive policies by the government can lead to an equilibrium with a greater quantity and a diminishing price. moreover, tax-subsidy schemes can be always devised in such a way that they preserve the welfare level of each individual under taxation, while improving the welfare of subsidized consumers. notice that competitive industries bring forth no extraordinary profits to producers in the long-run, since individual firms have no market power. accordingly, the total social welfare is composed of the surplus of consumers and the government solely. figure 1 illustrates this situation, where the initial equilibrium is found at (p, x1). figure 1.welfare improvements of self-financing tax subsidy programmes the logic of the argument can readily be seen with the aid of the graphical representation. the government could be able to reduce the average production cost if it can provoke an expansion of the industry output, thereby lowering the costs of individual firms operating in a larger industry. because of average cost pricing, the expansion of the industry leads to a lower consumer price at equilibrium. but, how does the government expand the demand for the good while meeting a self-financing requirement? first, the government taxes high-reservation-price buyers, keeping the tax-inclusive price for these consumers at the initial price. because the price faced by these consumers does not change, they do not gain or lose consumer surplus. the revenue generated through the taxes imposed on the first group of consumers is then transferred to low-consumer-surplus buyers in the form of a consumption subsidy, inducing new buyers to enter the market for the first time. overall, subsidized buyers gain consumer surplus because of the lower market price, the subsidy, and the additional surplus accruing to new buyers. the 4 the point is clear in the case of consumers who buy, at most, one single unit of the commodity. but the analysis is still valid as long as the tax charges consumption and the subsidy takes the form of a voucher. besides, it is important to be aware now that section 3 will adopt a different interpretation of the demand by representing the social willingness to pay involving both the private and public sector. european journal of government and economics 1(2) 167 tax-subsidy scheme is more than a transfer from one group of buyers to another; it improves welfare because the average cost of the product has been lowered.5 a more formal description of the issue follows. if the final aggregate production is set at the level x, the total amount of the subsidy is defined by s = c(x) − (d − dx), and the tax by t = p − c(x). the first step is then calculating the number of individuals, α, who must be taxed to ensure that the programme is self-financing. (in terms of the areas of figure 1, the self-financing constraint implies t = c + d + f). then, we solve (p − c(x)) α = (x − α)s, and obtain: x dx)(dp dx)(dc(x) αorx ts s α      (1) the meaning of the previous result is straightforward. for the programme to be always affordable, the required number of individuals paying the tax is a fraction of the total number of final consumers, which is determined by the relative weight of the subsidy as shown in (1). besides, note that the drop in prices implies that none of the α contributing consumers experience welfare losses once the programme has been implemented. moreover, as far as x > x1, any feasible self-financing transfer entails positive increases in social welfare, represented as a+b+c+d or t+a+b−f in figure 1, and defined by the equation: w (x) = (s + t)(x1 − α) + 2 (s + t)(x − x1) (2) expression (1) has established the condition under which the transfer programme is costless for society and, hence, immediately affordable. then, the second step is exploring how the size of the market x can be determined, which of course will depend on the target pursued by the policy makers. maximizing the social welfare initially, we consider the case in which the government aims to establish the level of x that maximizes total welfare. in our model, the value that society assigns to consuming amount x is calculated as v(x) = d (x − 0.5 x2). defining the cost of producing x as ct(x) = c(x)x, and given that the government seeks maximizing v (x) − ct(x), we get the first order condition for a maximum: *)x('cd *)x(cd *xor *x s *)x('c    (3) the optimal level x* is such that it entails the greater increase in social welfare. this optimal value is only congruent for x* < 1, and defines a maximum if the second order condition is verified, which in our model requires: d + 2c’ (x*) + c’’(x*)x* > 0 (4) 5 in other words, to illustrate our point while avoiding unnecessary complications, the setting of the model adds a couple of assumptions: on the one hand, it assumes that the tax charges consumption and that the subsidy is granted in the form of a voucher. in this way, we prevent the effects of redistributive policies to affect other demands of the markets. on the other hand, we account for government intervention as the mechanism through which to exclude, from claiming and receiving subsidies, the individuals who can initially afford the good. there are other situations in which similar features may apply. for instance, it is surely the case if these two circumstances occur together: (i) uncertainty about the actual amount of the good that people will eventually consume; and (ii) the necessity to take a decision on the amount needed upfront, before and regardless of the willingness to pay that private consumers have in reality. the analysis of that type of situation is made in section 3, where we examine the overall social demand for vaccines. european journal of government and economics 1(2) 168 note, however, that the fulfilment of the above inequality is not necessarily warranted. among other things, the value of α associated to x* might be greater that x1, implying that the solution is not feasible. but even then, the important fact remains that the government is always in the position to ensure welfare gains. this may be done by simply establishing the level of x associated with the maximum level of coverage in the industry; that is to say, by setting α = x1. maximizing the size of the industry consider now that the goal of the government (rather than maximizing welfare) is providing access to the industry to as many individuals as possible. in accordance with this aim, the x1 initial consumers must be contributing consumers, which permits collecting the largest amount of funds to afford the greatest possible subsidy scheme. the problem is solved by equalizing α to x1, which determines the level of maximum coverage: 1x s ts x    (5) the last expression is only consistent for values of  x that are always greater than x1, since (s+t) > s. but again, it might also be the case that the full coverage of the market is reached for a number of contributing consumers smaller than α = x1. in other words, the size of the market imposes the feasibility constraint:  x ≤ 1. hence, if the value delivered in (5) does not satisfy the feasibility constraint, the optimal outcome instead has to be the corner solution defined by  x = 1 and α = s/(s+t). note anyway that the government always achieves its purpose of maximizing the coverage level. whether the maximum feasible self-financing coverage is the one established in (5) or it is  x = 1 is an incidental matter. the vaccine industry: an illustrative example some troubles may arise regarding the feasibility of implementing the tax-subsidy scheme and, more importantly, concerning the actual relevance of our theoretical argument in real markets. to come across these difficulties, this section adapts the theoretical elements to fit with the organisational aspects of the vaccines industry, which provides us with a suitable example. the activities and features that characterise the vaccine industry are described, for instance, in gordon and samant (2008, p. 37): “the vaccine industry is composed of companies that are engaged in any of the following activities: research, development, manufacture, or sales, marketing and distribution of vaccines (...). vaccine development is difficult, complex, highly risky, and costly and includes clinical development, process development and assay development.” in accordance with that, to judge the existence of external economies one should examine the above mentioned activities one by one. on one hand, these authors stress that, at the manufacturing stage, some production processes in the vaccine industry are scalable. this implies that a greater size of the manufacturing laboratory (or a larger market) would permit the unit cost per vaccine to be reduced. nonetheless, it seems that other manufacturing activities are not scalable, thereby implying that the unit manufacturing cost of vaccines does not significantly decrease as the production volume expands. on the other hand, unlike in the manufacturing processes, the existence of external economies of scale has to be recognised at the research and development stage. precisely, gordon and samant (2008: 41) point out that the size of the market is a european journal of government and economics 1(2) 169 crucial factor in developing new vaccines: ‘a typical vaccine company will have several vaccine candidates in basic research... those that are most promising in terms of technical feasibility, strong patent protection, and potential market size will be taken forward into development ... thus, go/no go decisions must be made and market size is a major determinant of the choice between two candidate vaccines ... this system has worked extremely well for vaccines with large potential markets in the developed world when technical feasibility is demonstrated. it does not work for vaccines for diseases which exist predominantly in the poorer regions of the world ...; it works imperfectly for diseases of the developed world that affect only a relatively few persons because of geographic restriction.’ in any case, the existence of economies of scale in this type of industries should not be a surprise, insofar as the success in this business depends on research activities. indeed, once the companies have assumed heavy investments and capital risks associated to research, a larger potential market makes more likely that the monetary return compensates for the initial expenses.6 in conclusion, we should accept that the cost of developing (and to some extent the cost of producing and commercialising) this type of medicines decreases along with the size of the market: the more units produced, the cheaper the average costs. hence, the vaccines industry seems an appropriate context in which to apply the idea of this paper.7 in particular, we examine here the provision of vaccines – in 2009 and 2010 – for preventing influenza h1n1 to spread out. the large potential size of the market (1 billion doses of vaccinations were initially agreed, only in 2009, between the governments of the major economies and the industry) allowed for reducing the average production cost of the vaccine to about 2 € per unit. besides, as table 1 shows, a significant number of vaccines were donated to the less developed countries. of course, should the aggregate demand of the market have a smaller size, the cost per unit of the vaccines would have been much larger. this is typically the case in markets where a large amount of investments (for the research and development of new patents) are needed and must be repaid by the consumers who eventually buy the product. 6 gordon, jerald and vijay (2008, p. 42) explicitly explain that: ’large companies believe that vaccines should be priced according to value to society; reduction in health care and related costs, relief from pain and suffering, and/or prevention of death, and that they should be rewarded for taking the enormous risks inherent in early vaccine development. such prices far exceed manufacturing costs, but are essential to produce the revenue streams that allow vaccines to be competitive for research and development and manufacturing resources within large pharmaceutical companies, or that make biotech companies attractive investment opportunities.’ a similar argument is given by plotkin, orenstein and offit (2008, p. xxi) in the preface of their book: “this high cost [to bring vaccines form initial ideas to licensed biologicals] is a major disincentive for companies to develop new and improved vaccines. the requirement by major pharmaceutical companies for large markets to justify the expense of development means that many needed vaccines for geographically localized infections are not available.” 7 further information on the matter can be found again in gordon, jerald and vijay (2008 : 40-1): ‘the role of large, full-service vaccine companies is predominantly in development. they engage in some limited basic research, significant amounts of targeted research regarding specific organisms, but the preponderance of activity is in clinical and process development. expertise and sufficient personnel in process development and chemical engineering reside almost exclusively in such companies; there is no other resource for such development. clinical development that will satisfy fda standards is also done mostly by the large companies, funnelled through academia and contract research organizations... many smaller organizations, often referred to as biotechnology companies, are engaged in vaccine research. they are often started by university scientists, supported by venture capitalists, and are capable of basic research on a vaccine idea. at this early state, they usually have limited capacity in process development and manufacturing, and none in distribution, sales, or marketing... because of the large cost of adding new capacities and expertise, many biotech companies in advanced product development will opt to partner with large, full-scale companies. although 60 or so small companies claim engagement in vaccine research and development, only about a dozen or so consider it a major activity.’ european journal of government and economics 1(2) 170 table 1: influenza h1n1: vaccine sales by company and donations company 2009 sales (mill euro) donors donations* (mill doses) sanofi aventis 580 sanofi aventis 100 glaxo smith kline 580 glaxo smith kline 50 novartis 580 8 rich countries 50 baxter 290 medimmune 290 sino vac 116 csl 116 solvay 58 other companies 690 total 3300 total 200 *the donations were made to who (world health organization). sources: international meeting on the inuenza h1n1 vaccine market.\asociación mémoire des luttes (www.medelu.org) and knol unit of knowledge (http://knol.google.com/k/inuenza-2009-h1n1-vaccine-market). in summary, the vaccine industry is considered an illustrative example with which to analyze the theoretical ideas of this paper. yet, given that the pharmaceutical companies must plan in advance the amount of vaccines they will produce, implementing self-financing programmes requires in practice that the governments approve the quantity before the private willingness to pay is revealed. this peculiarity, as well as the lack of information on some crucial aspects of the industry, has resulted in choosing the ad hoc framework of section 3, which has been designed to account for the specific features of the case. a theoretical framework for the vaccines industry there is one critical issue that must be considered before applying our theoretical framework to the vaccines industry: to what extent does the vaccines industry behave as a competitive market? according to gordon and samant (2008), about one-half of vaccines for children in the us are exchanged in the private market. it means that there is a significant share of the market where the price-setting takes place under a free market regimen. the rest is then delivered through organizations of public character at a reduced price. and the share of free market tends to be smaller in japan or in west european countries. moreover, the mentioned authors explicitly state that: ‘in addition to the burden of partial price controls, the vaccine industry is subject to intense regulation. it cannot sell products until the vaccine and the facility in which it is manufactured are approved by the fda [food and drug administration] or other regulatory authorities; each batch must be released by the appropriate regulatory agency; and the usage, and therefore market size, is largely determined in the united states by the cdc [centers for disease control and prevention] and in europe by national regulatory authorities. thus, the vaccine industry does not operate in a free-market environment, and its behaviour reflects these constraints.’ (cf gordon and samant, 2008: 42). but, even if the last quote seems to be indisputably asserted, the debate must remain still open if considering the sub-industry of research and development activities. indeed, on the bases that little control is exerted on the initial research, and given the large number of small laboratories and other institutions involved in the matter, denying absolutely the competitive behaviour (at least in activities of this market) seems misleading. anyway, in front of these theoretical concerns, and due to lack of the necessary information, the applied section of this paper deviates from the preceding section. first, to deal with the theoretical analysis – while reflecting the manner in which the market operates in reality – in section 3 we have taken into account the commitment of the governments to purchase certain amounts of vaccines. typically, this agreement has to be made upfront, which is especially critical in order for the poor countries to enjoy cheaper prices. european journal of government and economics 1(2) 171 the following text by gordon and samant (2008: 40-1) is very conclusive in this regard: ‘to involve large companies in development and manufacturing of vaccines to meet needs such as bio-defense or health needs of poorer countries, incentives must be established to convince these companies that they should develop and manufacture such products. such incentives might take the form of guaranteed purchase of certain volumes of a vaccine if specified standards are met, direct contracting by a government agency, or some other publicly funded mechanism. the use of advanced market commitments to create a funding mechanism for vaccines needed in the developing world has been endorsed... companies may be willing to engage in such work. indeed, they may already have donated or sold vaccines at very low prices to poorer countries. without special incentives, it is unrealistic to expect companies to engage in research and development on diseases that only, or predominantly, affect the poorer regions of the world.’ in other words, advanced government commitments are essential to make it possible that new consumer gain access to the market, thereby increasing the total size of the industry. second, the issue cannot be treated with a single aggregate demand, as the subject seemed to require initially. but this difficulty is not such that it hinders other ways to address the topic in an appropriate manner. in particular, the domestic demand – that captures the overall willingness to pay of society – is going to be separately defined in each country by its respective government. actually, governments are asked to commit themselves to buy vaccines at a price that is previously settled by the world health organization (who). the range of prices had previously been established according to the standard of living of the different countries.8 another relevant aspect of the vaccines are the positive externalities attached to health goods and services. to account for this feature, the relevant demand in this case has to be defined as the overall willingness to pay of society, including the private consumers as well as the public sector. in summary, and with the aim of exemplifying some theoretical ideas of this paper, the model is tailored to account for the following characteristics: (i) uncertainty on the private consumers’ willingness to pay; (ii) public interest of the good under examination; (iii) commitment of the government to buy a certain amount of goods, which has to be pre-established upfront; and (iv) self-financing programmes in which the full-coverage (up to the quantity chosen by the government) is warranted by design. a suitable model for the vaccines domestic market let us describe now the way in which our theoretical analysis can be modified to explain how the vaccine industry operates in reality. the new theoretical elements are proposed so that the framework became suitable for dealing with the available information displayed in table 2. 8 according to the estimations of who, the price of the h1n1 vaccine would range between 2.5 and 20 us✩, depending on the economic status of the buyer country. the 2009 vaccine could be sold at around 10 to 20 us$ in western economies, in the range of 5 to 10 in mid-level developing countries and in the range of 2.5 to 5 in the poor economies. the latter countries were extendedly funded by charities, global agencies, the world bank, etc. many policy makers agree that charging patients and countries by their ability to pay is better than providing free treatments. european journal of government and economics 1(2) 172 table 2: pandemic influenza h1n1: vaccine orders and purchases (2009-2010) country price per unit (€) ordered doses (mill. units) ordered market value (mill. €) purchased doses (mill.) purchased market value (mill. €) used doses (mill.) used market value (mill. €) spain 6.93 37 280.0 13 90.0 3 20.8 france 9.25 94 869.0 5 46.3 germany 10.44 50 522.0 6 62.6 netherlands 9.21 34 313.2 15 138.2 italy 9.24 48 443.7 belgium 8.75 13 110.2 uk 8.70 132 1148.4 48 417.6 usa 8.67 251 2175.0 110 953.2 portugal 9.11 28 143.2 greece 9.18 24 220.4 australia 9.39 21 197.2 japan 9.19 53 487.2 canada 9.28 50 464.0 hungary 9.28 20 185.6 sweden 9.34 18 168.2 switzerland 9.37 13 121.8 south korea 9.37 13 121.8 china 5.80 65 377.0 sources: international meeting on the inuenza h1n1 vaccine market.\asociación mémoire des luttes" (www.medelu.org) and knol unit of knowledge (http://knol.google.com/k/inuenza-2009-h1n1-vaccinemarket). in this regard, two important elements are going to be treated differently, as compared to section 2: 1. the long-run supply of the industry is now replaced with a function capturing the cost that society abides, due to the government commitment to purchase a fixed amount of vaccines. this function is then defined, for each country, as: px = d, for )/,0( pdx  (6) where p is the final price (per unit of vaccine) actually paid. 2. regarding the domestic demand, we focus here on the public sector, because it is the government who makes the purchasing decision. the relevant demand in each country is therefore defined as the government willingness to pay for the vaccines. besides, we assume that the government plays a subsidiary role, thereby accounting for the fact that some individuals are willing and able to afford themselves the cost of the vaccination. in accordance with that, we consider the government net willingness to pay as the result of taking into account the following two elements: (a) the social willingness to pay for the vaccines provision, as revealed by the sum paid upfront by the government (to account for the number of citizens that the government considers should be surely protected). notice that the population is free to buy or not the vaccine at the price per unit shown, for 18 countries, in table 2. but, these people who initially decide not to buy the vaccine will eventually get free access to it, up to the level established by the ex ante decision of the government. the issue can thus be considered a subsidized consumption programme that ensures that a certain proportion of population joins the vaccination campaign. (b) the private willingness to pay for vaccinations, which implies alleviating the net amount of money spent by the government in this matter. of course, a large private consumption manifests a big private willingness to pay. and hence, given the subsidiary role granted to the government, the resulting public willingness to pay becomes smaller. european journal of government and economics 1(2) 173 the conjunction of these two elements, by subtracting the latter one (b) from the former (a), yields the effective public demand function, which we treat as the relevant domestic demand in each country (derived from the decision problem of the government): ,xpdp  or p pd x   (7) the interpretation of this demand is straightforward. the government is ready to pay upfront a sum so high as to ensure sufficient vaccines freely available (for the proportion of the country’s population that was previously chosen). however, given its subsidiary role, the government’s willingness to pay decreases along with that of private consumers. but of course, the fact is that the private willingness to pay is known only after the payment for the vaccines has been made. obviously, the above ad hoc description deviates from the framework developed in section 2, implying that the graphical representation is also different.9 to understand the particularities of this new approach, figure 2 illustrates the situation. figure 2: a suitable framework for the vaccines domestic market among other features, the new framework implies that the full-coverage level is internally warranted by the model, regardless of the willingness to pay of private consumers. it is always the case, provided that the intervention of the government makes that any value given to the “old” α (as characterised in section 2) is compatible with a self-financing programme. 9 in fact, it may be argued that – within this framework – little difference exists between the costs and the willingness to pay. it is actually true that the specificities of this peculiar approach introduce some difficulties to rightly interpret the resulting demand and supply functions. european journal of government and economics 1(2) 174 to avoid misunderstandings with the notation, in section 3 we use β to express the amount of vaccines that private consumers are willing to buy. note further that, in figure 2, the areas no longer represent the social welfare associated to consumer surplus (as it was the case in section 2), since the functions here have a different meaning. similarly, table 3 shows the values of the relevant parameters and functions representing the situations of the countries for which sufficient information was available. table 3: a model for influenza h1n1 vaccine market by countries country d x p cost function demand function france 869.0 94 9.2447 p x = 869 p = 869 – 94 869 · x germany 522.0 50 10.440 p x = 522 p = 522 – 50 522 · x netherlands 313.2 34 9.2118 p x = 313.2 p = 313.2 – 34 2.313 · x spain 90.0 13 6.9231 p x = 90 p = 90 – 13 90 · x perhaps the most relevant feature of the new characterisation – of the model in section 3 – is the fact that full-coverage is always achieved. for this to be under stood, consider that the size of the vaccines market – as far as the public interest is concerned – is given by the fixed amount determined ex ante by the government. in this context, the actual amount of vaccines that private consumers purchase is denoted by β. but, of course, the amount purchased by private consumers, even if consistent with a full-coverage situation (as is the case for any other value of β in this framework), is not necessarily the optimal choice. in addition to full-coverage, it is easy to see that the model – in its current version – always delivers self-financing outcomes, insofar as public resources give support to the programme for it to be always affordable. a simple look at table 4 permits easy verification of this feature, by cross checking the values of the relevant areas in figure 2. of course, the first identity that holds is: d = e + t + a’ = e + c + d + f. table 4: self-fnancing programmes. value of the areas in figure 2 by countries country t + a d + f t + a’ c + d + f e + t + a’ e + c + d + f france 9.3452 9.3452 822.7766 822.7766 869.0 869.0 germany 10.6576 10.6576 459.3600 459.3600 522.0 522.0 netherlands 9.4999 9.4999 175.0235 175.0235 313.2 313.2 spain 7.5577 7.5577 69.2308 69.2308 90.0 90.0 then, in table 5, we summarize the main results from sorting out some relevant values of the model. the table reports the pair of values associated with various relevant values of the consumption levels of vaccines. for instance, if we consider the situation actually chosen by private consumers, β = x0, it comes out that the effective cost per unit of the vaccines eventually used is as high as p0. given the small demand for vaccinations that table 4 reports (as deduced from the amount chosen by final private consumers in each country), p0 takes extremely high values. of course, if we consider the so called full-coverage situation x , the price per unit is equal to the price per unit actually paid, p . european journal of government and economics 1(2) 175 table 5: quantities and prices. influenza h1n1 vaccine market country β = x0 p = p0 x = x̄/2 p = 2 p x = x1 p = p1 x = x p = p france 5 173.8 47.0 18.49 92.9891 9.3452 94 9.2447 germany 6 87.0 25.0 20.88 48.9792 10.6576 50 10.4400 netherlands 15 20.9 17.0 18.42 32.9687 9.4999 34 9.2118 spain 3 30.0 6.5 13.85 11.9083 7.5577 13 6.9231 notice that, according to our assumptions, the payments eventually made by final consumers alleviate the government expenses in vaccines. if considering the case of β = x0, which corresponds to the number of vaccinations actually consumed, we get that the consumers pay the amount p , and the government subsidizes the rest: )pp( 0  . naturally, the total amount of aggregating the public and public payment is equal to βp0. but the government must set upfront the number of vaccines, and it chooses x . at this larger amount, the effective price per unit becomes cheaper (the price drops from p0 to p ), which allows new consumers to gain access into the vaccine market. again, the total expenses are afforded thanks to both the willingness to pay of private consumers: p ; and of the government: p)x(  . hence, the overall payment in this case is given by px , which, by design of the model, is equal to βp0. finally, some comments may help to understand the logic of the model in its ad hoc version. to this aim table 6 reports the calculations of estimating the effective cost per unit that the government affords at different possible values of β. the results of this exercise are computed with respect to both the number of vaccines eventually purchased by the consumers and with respect to the total available units agreed upfront by the government. european journal of government and economics 1(2) 176 table 6: effective government cost per unit (dependent on the chosen  ) wrt purchased units by consumers wrt total available units β france (€/unit) germany (€/unit) netherland s (€/unit) spain (€/unit ) france (€/unit) germany (€/unit) netherland s (€/unit) spain (€/unit ) 0 ∞ ∞ ∞ ∞ 9.24 10.44 9.21 6.92 1 859.76 511.56 303.99 83.08 9.15 10.23 8.94 6.39 2 425.26 250.56 147.39 38.08 9.05 10.02 8.67 5.86 3 280.42 163.56 95.19 23.08 8.95 9.81 8.40 5.33 4 208.01 120.06 69.09 15.58 8.85 9.60 8.13 4.79 5 164.56 93.96 53.43 11.08 8.75 9.40 7.86 4.26 6 135.59 76.56 42.99 8.08 8.65 9.19 7.59 3.73 7 114.90 64.13 35.53 5.93 8.56 8.98 7.32 3.20 8 99.38 54.81 29.94 4.33 8.46 8.77 7.04 2.66 9 87.31 47.56 25.59 3.08 8.36 8.56 6.77 2.13 10 77.66 41.76 22.11 2.08 8.26 8.35 6.50 1.60 11 69.76 37.01 19.26 1.26 8.16 8.14 6.23 1.07 12 63.17 33.06 16.89 0.58 8.06 7.93 5.96 0.53 13 57.60 29.71 14.88 0.00 7.97 7.73 5.69 0.00 14 52.83 26.85 13.16 7.87 7.52 5.42 15 48.69 24.36 11.67 7.77 7.31 5.15 16 45.07 22.19 10.36 7.67 7.10 4.88 17 41.87 20.27 9.21 7.57 6.89 4.61 18 39.03 18.56 8.19 7.47 6.68 4.33 19 36.49 17.03 7.27 7.38 6.47 4.06 20 34.21 15.66 6.45 7.28 6.26 3.79 21 32.14 14.42 5.70 7.18 6.06 3.52 22 30.26 13.29 5.02 7.08 5.85 3.25 23 28.54 12.26 4.41 6.98 5.64 2.98 24 26.96 11.31 3.84 6.88 5.43 2.71 25 25.52 10.44 3.32 6.79 5.22 2.44 26 24.18 9.64 2.83 6.69 5.01 2.17 27 22.94 8.89 2.39 6.59 4.80 1.90 28 21.79 8.20 1.97 6.49 4.59 1.63 29 20.72 7.56 1.59 6.39 4.38 1.35 30 19.72 6.96 1.23 6.29 4.18 1.08 31 18.79 6.40 0.89 6.20 3.97 0.81 32 17.91 5.87 0.58 6.10 3.76 0.54 33 17.09 5.38 0.28 6.00 3.55 0.27 34 16.31 4.91 0.00 5.90 3.34 0.00 35 15.58 4.47 5.80 3.13 36 14.89 4.06 5.70 2.92 37 14.24 3.67 5.61 2.71 38 13.62 3.30 5.51 2.51 39 13.04 2.94 5.41 2.30 40 12.48 2.61 5.31 2.09 41 11.95 2.29 5.21 1.88 42 11.45 1.99 5.11 1.67 43 10.96 1.70 5.02 1.46 44 10.51 1.42 4.92 1.25 45 10.07 1.16 4.82 1.04 46 9.65 0.91 4.72 0.84 47 9.24 0.67 4.62 0.63 48 8.86 0.44 4.52 0.42 49 8.49 0.21 4.43 0.21 50 8.14 0.00 4.33 0.00 51 7.79 4.23 … … … … … … 94 0.00 0.00 the “effective government cost per unit” function is dependent on the values of β that the consumers choose in the end. besides, its definition differs depending on the reference with respect to which the calculation is expressed. we denote as gp the cost that the government pays per unit of the vaccines that were effectively used by the consumers; and by gp the government cost per unit of available european journal of government and economics 1(2) 177 vaccines. the respective definitions of these functions are given by the following equations: )pd( 1 pg    and )pd( x 1 pg x  (8) figure 3: government cost per unit (wrt available and purchased vaccines) the outcomes resulting for different values of β are reported in table 6, and the same findings are also illustrated by means of two diagrams in figure 3. a simple inspection of the data confirms that the minimum possible payment made by the government corresponds to the case in which β =. x it cannot be different because, at this level, the β chosen by the consumers is equal to the total amount that the government purchased through the advanced market commitment. european journal of government and economics 1(2) 178 an interesting matter is, perhaps, to calculate the threshold at which the value of β permits the government subsidy to be smaller than the price actually paid by the consumers. for this purpose, we propose a transformation of the first function in expression (8), which leads to: p x )pd( 1 pg      (9) notice that by definition β < x . by looking at the new form of function gp , it is obvious that β = x leads to the minimum value of the function: 0pg  . but the question now is to find the value of β at which it starts to hold that ppg  . it is very easy to prove that it happens for any β > x /2. this result, even if trivial, is highlighted in table 6 by using bold and italic characters. conclusion the main idea of this paper refers to competitive industries with external economies of scale. in this context, we have shown that, unless the policy involves implementation costs that are too heavy, adopting well-planned tax subsidy programmes can always have pareto-improving effects. first, the paper has examined the impact – in terms of social welfare – of introducing tax-subsidy programs in markets characterized by the presence of external economies scale. then, the analysis has focused on studying the vaccine industry as one appropriate example for illustrating the theoretical aspects. from the theoretical analysis, the presence of external economies of scale permits that tax-subsidy schemes may always be designed in such a way that they improve the welfare status of all the economic agents involved in the matter. besides, the setting of the model has been designed for the redistributive policy to be selffinancing. moreover, if the transfer is properly arranged, it can always lead to pareto-improving situations. we venture that this theoretical possibility may help explaining the success of existing programs that provide vouchers to low-income individuals. this is because none of the incumbent consumers experience losses in welfare, whereas a number of new individuals gain access to the market by purchasing the good at a subsidized price. our approach involves both the demand and the supply side of the market, since the external economies affect the shape of the aggregate supply, while the tax subsidy programme influences the effective demand function. in this context, we have proved that intervention is capable of increasing the total amount of trade in the industry while the equilibrium price declines, thereby having pareto-improving effects. naturally, this is the case unless the cost of implementing the policy was too high. our result reinforces the idea that intervention could be preferred in front of externalities, even under the assumption of perfect competition. to illustrate a real example, this paper has examined the case of the vaccine for the pandemic influenza h1n1. even if the market of vaccinations was perfectly apt to examine the relevance of external economies of scales, due to the limited available information, we have adopted in section 3 an approach that deviates from the original framework of section 2. in the latter approach, two main features have been accounted for: (i) uncertainty about the amount of good that is demanded and (ii) the requirement that the government should take a decision (on the quantity to be purchased) ex ante, regardless of the actual private demand. after having described the behaviour and regulation of the modern industry of vaccines in reality, the paper shows that it is possible to implement better public practices to make the provision of vaccines more efficient and feasible for broader geographic areas. european journal of government and economics 1(2) 179 in summary, the main idea of this paper is taking advantage of the external economies of scale to improve the welfare of all the agents of a particular market. our study ultimately implies advocating intervention in competitive markets with external economies of scale, by means of adopting self-financing tax subsidy schemes. the originality of the analysis consists of stressing that these policies can always be implemented in one or another way. further examination of the issue is, however, needed to evaluate the practical relevance of this idea and to examine to what extent certain specific industries are characterised by external economies of scale. references aoki, masahiko (1971) ‘marshallian external economies and optimal tax-subsidy structure’, econometrica 39(1): 35-53. broadberry, stephen and andrew marrison (2002) ‘external economies of scale in the lancashire cotton industry, 1900-1950’, economic history review 55(1): 51-77. chandra, vandana, david franck and nadeem naqvi (2002) ‘world increasing returns and production subsidies’, economica 69: 223-227. j. s. chipman (1970) ‘external economies of scale and competitive equilibrium’, the quarterly journal of economics 84(3): 347-85. ethier, wilfred j. (1982) ‘decreasing costs in international trade and frank graham’s argument for protection’, econometrica 50, 1243-1268. gordon, r. douglas and vijay b. samant (2008) ‘the vaccine industry’, in stanley a. plotkin, walter orenstein and paul a. offit (eds) vaccines saunders elsevier, 5th edition. graham, frank d. (1923) ‘some aspects of protection further considered’, quarterly journal of economics 37, 199-227. helpman, elhanan and paul r. krugman (1985) market structure and foreign trade. london england: mit press. marvasti, akbar (1994) ‘international trade in cultural goods: a cross-sectional analysis’, journal of cultural economics 18(2): 135-48. meade, james e. (1952) a geometry of international trade. london: george allen and unwin ltd. osana, hiroaki (1977) ‘optimal tax-subsidy system for an economy with marshallian externalities’, econometrica 45(2): 329-40. panagariya, arvind (1981) ‘variable returns to scale in production and patterns of specialization, american economic review 71: 221-230. plotkin, stanley, a., walter orenstein and paul a. offit (2008) vaccines, 5th edition. philadelphia: saunders elsevier. thépot, jacques (2003) ‘pareto-improving redistribution in a monopoly’, japanese economic review 54(2): 165-177. microsoft word ejge_03_01_010.doc european journal of government and economics volume 3, number 1 (june 2014) issn: 2254-7088 60 determinants of tax morale in spain and turkey: an empirical analysis cevat bilgin, cukurova university, turkey abstract tax morale is defined as the intrinsic motivation to pay taxes, and is closely related to tax compliance. determinants of tax morale need to be investigated for a more comprehensive understanding of tax compliance. in this paper, determinants of tax morale in turkey and spain are analysed on the basis of world values survey data. firstly, descriptive statistics of the variables used in the models are provided. since tax morale is an ordered categorical dependent variable, ordered probit models are estimated separately for turkey and spain to derive the relations between tax morale and relevant variables. marginal effects are computed since the coefficients of the models cannot be interpreted because of the nonlinearity of the estimated models. the marginal effects related to the top level of tax morale category are presented. the independent variables are combined by demographic factors, employment categories, economic status of the respondents and social capital variables. the findings from the estimated model suggest that social capital variables and some of the demographic factors have important effects on tax morale in turkey. confidence variables have positive effects; if taxpayers feel confidence in political entities they are willing to pay taxes. religion and national pride affect tax morale positively. on the other hand, the results are different for spain; social capital variables do not have effects on tax morale. specifically, confidence variables are found to be statistically insignificant. age, education level and the income level have significant effects on tax morale in spain. jel classification h26; c51; c52. keywords tax morale; tax evasion; ordered probit. bilgin ● determinants of tax morale in spain and turkey 61 introduction tax evasion is an important issue in almost all countries. while some taxpayers seek the ways of evasion, the others are not eager to evade. tax morale is, to large extent, to do with the physiological background of these behaviours. arguably, tax morale is concerned with why people do not evade. there has been much in common between tax morale and tax compliance. tax compliance is an observable action; that is most people pay their taxes. tax compliance is not only a function of tax ratios and probability of detection, but also a function of an individual’s willingness to comply with and to evade. as torgler (2007) pointed out, level of tax compliance is relatively high when tax morale is high; therefore tax morale is needed for providing on an account of the puzzle of tax compliance. determinants of tax morale need to be investigated for a much more comprehensive understanding of tax compliance. nevertheless, there is a limited work on the issue in the existing literature (torgler, 2004, 239). as feld and frey (2002, 88) suggest: most studies treat ‘tax morale’ as a black box without discussing or even considering how it might arise or how it might be maintained. it is usually perceived as being part of the metapreferences of taxpayers and used as the residuum in the analysis capturing unknown influences to tax evasion. the more interesting question then is which factors shape the emergence and maintenance of tax morale. indeed, few scholars have conducted research on tax morale in detail. by definition, tax morale is an intrinsic motivation to pay taxes (torgler, 2007, 4). morality could be defined as an individual’s internalisation of such concepts and legal rules to perform his or her social duties in a proper way. hence, individuals’ general attitude (or tax mentality) towards taxation and specific liabilities imposed by the related legislation is substantial to the debate. from this perspective, tax morale is largely framed in the general concept of tax mentality (schmölders, 1976: 107). for torgler (2004), tax morale measures taxpayers’ attitudes while tax evasion measures taxpayers’ behaviours. in other words, tax morale is not an outcome variable, like tax evasion. tax morale can, hence, be defined as a moral obligation of paying taxes; it is, in turn a belief of contributing to society by paying taxes. tax compliance and tax morale are affected by social and physiological factors. therefore increasing the level of tax morale and tax compliance depends heavily on these factors. legal infrastructure and state enforcement power, to some extent, raise the level of tax compliance. social and physiological factors are still decisive factors to change tax morale. social capital evokes individuals’ economic and cultural capitals; perceptions on income distribution, on fairness in society, trust in others and institutions, and confidence in government and in government policies are some of examples of social capital variables. socio-demographic variables affect tax morale substantially. the first of them is the variable of age. relatively older people are more vulnerable to threats of sanctions, suggested by the relevant studies. as torgler (2007) points out, the reason is that people are attaining new social characteristics, such as property, status, dependency on others’ behaviours as they get older. therefore the potential costs of penal sanctions for older people seem to be relatively greater. in consequence, there might be a significant relation between tax morale and age. furthermore, socio-physiological studies reveal that females are more compliant and less selfish than males. nevertheless, it should be noted that the traditional role of females is substantially different from the role of females in a modern society. moreover, females have more risk-averse behaviours than males have. marital status is another individualistic variable that might affect tax morale. the more social ties the individual has, the more restrictions imposed on her or his behaviour. thus married individuals are more prone to exhibit legal behaviours. european journal of government and economics 3(1) 62 for torgler (2007), employment status is another important factor for tax morale. it might be argued that self-employed persons have lower tax morale than full-time and part-time employees. this might be explained by the fact that the selfemployed have more opportunities to evade tax. the relation between education level and tax morale is not clear. according to torgler and schneider (2006), educated taxpayers are thought to have more information about tax regulations and fiscal relations. besides, they might be aware of civil services provided by state, and so they might have high level of tax compliance. on the other hand, they seemingly have knowledge of public corruption and thereby they might have critical perception on how tax revenues are used by government. moreover, they know a lot about the opportunities to evade tax. as a result of these factors, it is assumed that they have a low level of tax morale. consequently, there have been different kinds of relations between education level and tax morale. the relation between level of income and tax morale is a bit complicated and thus depends on some conditions. when people are not satisfied with their financial situation, they might tend to be evader. if a taxpayer feels a gap between his current and desired financial situation, he would be unwilling to pay taxes. moreover, they would be more reluctant to pay taxes if they feel they have less when they compare their income and their wealth with others’. on the contrary, duch, palmer and anderson (2000) claim that people who have low level of income are willing to pay taxes since they think they have benefit from public goods and services more than people having high level of income have benefit from. wealthy people, on the other hand, would be less willingly pay taxes by the perception of having benefited from public goods and services less than others. taxpayer confidence in parliament, government, justice system, tax administration has a valuable relation with tax morale. if the level of this confidence is high enough, individuals’ loyalty to public administration would increase and so they would be more willingly to pay taxes. in most of the empirical studies, these variables turn out to have significant relation with tax morale. montero and torcal, (2006) defines political disaffection as the subjective feeling of powerlessness, cynicism, lack of confidence in the political process, in politicians and in democratic institutions. in this context, political disaffection is expected to have negative effect on tax morale. in addition, importance of politics and perception on democracy are important factors for explaining tax morale. people seem to be complying more in democratic countries, as they have the ability to affect tax and expenditure policies in these countries. national pride is another interesting factor that should be taken into account on the analyses about tax morale. when people are happy with their national identity, they would be more loyal to their countries, and therefore tax morale would be high. religiosity is also affecting tax morale; indeed, positive relations between tax morale and religiosity have been found in some research. as frank (1996) points out; tax compliance is more common among the people whose perception depends on moral and ethic codes. religious people are so interested in what is right and what is wrong; therefore they believe that people have some duties to be performed: paying taxes is one of these duties. the aim of this paper is to derive a comprehensive framework of the determinants of tax morale in two countries. as outlined here, various socio-economic factors are related to the concept of tax morale. it is thought that they affect tax morale in different degrees. the previous empirical studies prove the existence of these effects on tax morale; yet the way and the degree of impacts differentiate across the studies. in this paper, the effects of these socio-economic factors on tax morale are evaluated for turkey and spain individually; then the findings are compared to picture the differences and similarities for each country. there is much that turkey and spain have common. they share a number of historic similarities which coupled with contemporary considerations (chislett, 2008). according to chislett (2008), both of the countries are at opposite ends of the mediterranean on the periphery of europe, both have a long islamic past, both have had large empires, both were significantly agricultural economies, both have undergone massive bilgin ● determinants of tax morale in spain and turkey 63 internal migration from rural areas to cities and towns, both have ‘exported’ hundreds of thousands of workers to elsewhere in europe, both countries have had strong statist economic policies until their economies began to be opened up, both countries were enlisted for geostrategic reasons during the cold war years by the us, both joined the organisation for economic co-operation and development (oecd) in 1961, both have problems in containing minority nationalisms, both spain and turkey have also suffered from real or imagined ‘black legends’ that even today affect the countries’ images abroad. this paper searches whether the effects of the independent variables on tax morale differentiates between these similar countries. the next section reviews the previous empirical literature about the determinants of tax morale. the third section is about the methodology and data; it gives essential information about the estimation method and defines the variable used in the estimation. the fourth section gives the estimation results; it starts with the descriptive statistics and ordered probit estimation results for the case of spain. the statistics and estimated parameters are explained in details. the same results and the interpretations are also presented for the case of turkey. finally, concluding remarks relating to the estimated results and the comparison of the countries are outlined. previous empirical literature studies on tax morale can be evaluated in two categories. the studies in the first group have focused on only one country and have analysed regional discrepancies and changes in time (torgler 2005; martinez-vazquez and torgler 2009; prieto et al 2006). a second group of studies have been conducted in a comprehensive perspective with more countries included (torgler and schneider 2006; alm and torgler 2006; cummings et al 2006), while torgler (2006) and alm and torgler (2006) have used a wider set of data with many countries. all studies on tax morale and its determinants used the international databases such as international social survey programme (issp), world values survey (wvs), european values survey (evs) and african opinion survey (afrobarometer). alm and torgler (2005) compared tax morale levels of usa with spain by using 1990 and 1995 wvs data that concluded that tax morale level in usa is higher. an explanation of their result might be that “compliance” as a social norm is better in usa when compared in spain. a further comparative study conducted by alm and torgler (2006) analysed tax morale in 14 european countries and the usa by using wvs data in 1990-1993; they still found that the highest tax morale is in the usa. in both of these articles, weighted ordered probit model was estimated to get the results. torgler and schneider (2006) have a similar research analysing spain, switzerland and belgium, by using the 1995-1997 wvs and 1999-2000 evs data. they estimated a weighted ordered probit model in which the variables of gender, age, marital status, education, employment status, social class, income level, attendance at church, direct democracy, national pride, confidence in political institutions and government, participation in democracy, income tax rate, fine rate and probability of detection are used as independent variables. the regional discrepancies in spain and switzerland affect tax morale, females have higher tax morale than males. they concluded that confidence in justice system, confidence in government, confidence in parliament, national pride, attitudes supporting democracy have positive effect on tax morale. torgler (2006) looked at the determinants of tax morale in 32 countries (including spain) by using the wvs data of 1995-1997. the included variables in the weighted ordered probit model are gender, age, marital status, education, employment status, social class, financial situation, risk averness, religiosity, corruption and credibility. the author found out that religiosity in particular for european journal of government and economics 3(1) 64 catholics, hindus, and buddhists increases tax morale. risk aversion and financial satisfaction positively affect tax morale. tax morale is low for the high-class, and it is high for the retired, housewives, and the part-time employed. in addition, females and married people have high tax morale. on the contrary, there is a negative relation between education and tax morale; perception of level of corruption decreases tax morale. similar research was conducted by cummings et al (2006) analysing botswana and south africa by using afronarometer data of 1999 and 2000, by gokbunar et al (2007) on turkey by using evs data of 2002, and by martínez-vázquez and torgler (2009) on spain by using wvs and evs data of 1981, 1990, 1995 and 1999-2000. these scholars provided similar results about the determinants of tax morale, whose levels heavily depend on socio-demographic and social capital factors. kaynar-bilgin (2011) searched for the determinants of tax morale in turkey by using the wvs data of 2005-2008. the included variables for probit model are gender, age, marital status, education, employment status, income level, financial satisfaction, religiosity, national pride, confidence in government, justice system and parliament, importance of politics and religiosity. the author found out that religiosity, importance of religiosity, importance of politics, national pride, and confidence in government, all positively affect tax morale. tax morale is low for the unemployed. methodology and data the methods used by the empirical studies are fairly similar since tax morale is a categorical variable. in general, ordered probit models were preferred for use in determining the relations and the interactions between tax morale and personal, socio-economic factors. ordered probit models are very useful to analyse dependent variables of tax morale containing ordering information. therefore, the same estimation method is used in this paper. the magnitudes of estimation are not interpreted, only the signs are analysed because equation in ordered probit model is in the nonlinear form. thus marginal effects should be derived to determine the effect of each variable on tax morale. marginal effect indicates the probality of specific tax morale category when an independent variable increases by one unit. in practice, only marginal effects related to the top level of tax morale category have been evaluated. for this paper, wvs database is used to analyse the level of tax morale and its determinants in turkey and spain. the fifth wave of the wvs is used for this aim. wvs is a survey which researched socio-cultural and political changes in a global base. the fourth and fifth waves have more than 50 countries representing approximately 80 percent of the world population. in this survey, individuals are asked to respond to the following question to evaluate level of tax morale: please tell me for the following statement whether you think it can always be justified, never be justified, or something in between: … ‘”cheating on tax if you have the chance”. the question leads to a ten-scale index of tax morale with the two extreme points “never justified” and “always justified”.the responses close to “1” can be thought as respectively high level of tax morale while the ones close to “10” indicate low level of tax morale. the ten-scale index of tax morale is transformed into four-scale index (0,1,2,3) by the same method used in the related studies. in the transformed scale “0” implies “always justified” and “3” implies “never justified”. the responses in the interval 4-10 in the original scale become “0” in the new scale since they imply that people justify tax cheating anyhow. besides “never justified” option “1” in the original scale now is “3”. the independent variables are classified as four categories: social capital variables; demographic factors; employment status; and economic status. bilgin ● determinants of tax morale in spain and turkey 65 variables of social capital are importance of politics, importance of religion, confidence in the justice system, confidence in government, confidence in parliament, confidence in civil services, religiosity, national pride and education level. since religiosity, national pride and education level are mostly defined as social capital variables in the empirical literature, they are evaluated as parts of the social capital group. gender, age and marital status are among the demographic factors that are included in the model. for the employment status, we have full-time employed, part-time employed, retired, housewife, student, unemployed, employed in public sector, or employed in the private sector. income level and financial satisfaction level are the indicators of economic status. the detailed definitions of the variables are presented in the appendix. estimation results first of all, the descriptive statistics of all variables used in the model for spain are derived. table 1 shows these statistics. the percentage rates of persons having specific properties in the whole population are as follows; female 50, married 58, the 30-49 age group 36, the 50+ age group 40, the 16-29 age group 24. the percentage rates for education level are as follows: level 2 is 20 percent, level 3 is 16 percent, and level 4 is 8 percent. of the respondents 3.9 percent are part-time employed, 7.7 percent are self-employed, 19 percent are retired, 17 percent are housewives, 7 percent are students, and 4.6 percent are unemployed. when it comes to the variable of tax morale, 63 percent of the individuals have the highest tax morale level. these individuals respond that cheating on tax is never justified. the rates of the levels 3 and 2 are 10 percent and 6 percent respectively. the rate of people saying cheating is always justifiable is 19 percent. the rates of income levels are 34 percent for level 1, 43 percent for level 2, 21 percent for level 3, 3 percent for level 4. the rates of financial satisfaction for levels 1, 2, 3 and 4 are 10, 30, 43 and 17 percent respectively. the variables of religiosity and national pride give the relatively high values of 44 and 92 percent. besides 14 percent of them thinks that religion is important. the ratio of respondents considering politics is an important issue is relatively low, 7.7 percent. other figures are as follows: confidence in justice system, 54 percent; confidence in government, 44 percent; confidence in parliament, 49 percent; and confidence in the civil service, 39 percent. among the social capital variables, national pride has the highest figure with 92 percent of respondents proud of their national identity. confidence in entities of the political system is relatively high from 39 to 54 percent. european journal of government and economics 3(1) 66 table 1: descriptive statistics of the variables for spain variables mean std. deviation minimum maximum tax morale (level 1) 0.195833 0.397006 0 1 tax morale (level 2) 0.064416 0.245152 0 1 tax morale (level 3) 0.102500 0.303431 0 1 tax morale (level 4) 0.637500 0.480923 0 1 1) social capital importance of politics 0.077500 0.267495 0 1 importance of religion 0.147500 0.354751 0 1 confidence in justice system 0.540000 0.498605 0 1 confidence in government 0.440833 0.496694 0 1 confidence in parliament 0.491667 0.500139 0 1 confidence in civil services 0.390000 0.487953 0 1 religiosity 0.440833 0.496694 0 1 national pride 0.920833 0.270111 0 1 education (level 1) 0.539167 0.498671 0 1 education (level 2) 0.206667 0.405083 0 1 education (level 3) 0.160833 0.367530 0 1 education (level 4) 0.076666 0.266173 0 1 2) demographic factors gender (male) 0.500000 0.500208 0 1 gender (female) 0.500000 0.500208 0 1 16-29 age interval 0.229167 0.420472 0 1 30-49 age interval 0.362500 0.480923 0 1 50-98 age interval 0.408333 0.491730 0 1 marital status (unmarried) 0.415000 0.492927 0 1 marital status (married) 0.585000 0.492927 0 1 3) employment status full-time 0.390000 0.487953 0 1 part-time 0.039166 0.194072 0 1 self-employed 0.077500 0.267495 0 1 retired 0.195833 0.397006 0 1 housewife 0.170833 0.376520 0 1 student 0.074166 0.262151 0 1 unemployed 0.046666 0.211012 0 1 private 0.565833 0.495854 0 1 4) economic status income level (level 1) 0.254167 0.435573 0 1 income level (level 2) 0.431668 0.495515 0 1 income level (level 3) 0.211668 0.408660 0 1 income level (level 4) 0.029167 0.168344 0 1 financial satisfaction (level 1) 0.091667 0.288675 0 1 financial satisfaction (level 2) 0.300833 0.458812 0 1 financial satisfaction (level 3) 0.430833 0.495399 0 1 financial satisfaction (level 4) 0.170000 0.375789 0 1 data source: world values survey database; http://www.worldvaluessurvey.org/ bilgin ● determinants of tax morale in spain and turkey 67 table 2 gives the results of the estimated ordered probit model for spain. in this estimation, the dependent variable is the tax morale variable as defined earlier. the independent variables are drawn from social capital, demography, employment status, and the economic situation. the second column indicates the estimated coefficient for each independent variable. the coefficients cannot be directly interpreted since the model is non-linear. yet the signs of each coefficient can be evaluated. in the third and fourth columns, there are associated standard errors and z-statistics for each variable respectively. the statistical significances of the variables are determined by using the z-statistics. in the last column, the marginal effects of the independent variables on the dependent variable are given. these effects are the ones related to the top level of tax morale category (y = 3). moreover, the specification test statistics are given at the end; the estimated model seems to be statistically significant. most of the social capital variables, importance of politics, importance of religion, confidence in justice system, confidence in government, confidence in parliament, and confidence in the civil service, appear to be statistically insignificant and do not have effect on tax morale in this sample level. the estimated coefficients of the variables of national pride and education level 4 are statistically significant at the 1 percent level while the one of religiosity is significant at the 5 percent level; besides they are all positive. these findings imply that national pride and religiosity have positive effects on tax morale. moreover, the most educated people have higher tax morale. the estimated coefficient of the highest level of education has the largest marginal effect among the significant social capital variables at 0.18 according to the estimated model, gender and marital status seem to have nothing to do with tax morale. age interval coefficients, on the other hand, are fairly significant and their signs are positive implying that tax morale is higher for elderly persons. the marginal effect belonging to the variable of the 30-49 age group is 0.12. being in the highest education category increases the probability of being at the top level of tax morale by 0.12 units comparing to the lowest level of education. when it comes to employment status, the results are not remarkable all of the variables in this category except students are not significant. students have higher tax morale than full-time employees have. interestingly, respondents who are at the top level of income turn out to have lower tax morale than the lowest group. in addition to this, the financial satisfaction variable has a negative significant effect on tax morale. the highest level of this category has a statistically significant estimated coefficient. european journal of government and economics 3(1) 68 table 2: ordered probit model estimation for spain independent variables coefficient standard error z-value marginal effects 1) social capital importance of politics importance of religion confidence in justice system confidence in government confidence in parliament confidence in civil service religiosity national pride education (level 2) education (level 3) education (level 4) 2) demographic factors female 30-49 age interval 50-98 age interval married 3) employment status part-time self-employed retired housewife student unemployed private sector 4) economic status income level (level 2) income level (level 3) income level (level 4) financial satisfaction (level 2) financial satisfaction (level 3) financial satisfaction (level 4) number of observations prob (chisqd>value) information criterion (aic) pseudo r2 -0.193061 0.176889 0.058499 -0.051011 -0.020400 0.076616 0.135098* 0.379701*** 0.060485 0.150077 0.552271*** -0.088738 0.349768*** 0.294453** 0.039771 0.151088 -0.229823 0.137709 0.181493 0.418436** 0.246105 0.111949 0.007230 -0.043904 -0.399906* -0.171399 -0.191946 -0.328749** 1200 0.000000 2.02991 0.026466 0.135398 0.115744 0.079132 0.084690 0.087950 0.081206 0.080430 0.133833 0.104411 0.120894 0.158983 0.082358 0.112433 0.133045 0.087424 0.194842 0.156199 0.131056 0.138894 0.175769 0.190259 0.091400 0.090271 0.111705 0.231603 0.141282 0.140196 0.157227 -1.426 1.528 0.739 -0.602 -0.232 0.943 1.680 2.837 0.579 1.241 3.474 -1.077 3.111 2.213 0.455 0.775 -1.471 1.051 1.307 2.381 1.294 1.225 0.080 -0.393 -1.727 -1.213 -1.369 -2.091 -0.0739 0.0644 0.0219 -0.0191 -0.0076 0.0285 0.0503 0.1475 0.0224 0.0549 0.1828 -0.0331 0.1276 0.1085 0.0149 0.0548 -0.0883 0.0506 0.0661 0.1434 0.0875 0.0419 0.0027 -0.0165 -0.1561 -0.0647 -0.0719 -0.1264 notes: dependent variable is tax morale on a four-point scale (0 to 3). the reference groups: the 16–29 age interval for the age variable, males, unmarried for marital status, full-time employed for employment status, public sector for the institution of occupation, education level 1 for education, income level 1 for income, financial satisfaction level 1 for the financial satisfaction variable. *, ** and *** refer to statistically significant parameter estimates at the 10, 5 and 1 percent levels respectively. data source: world values survey database; http://www.worldvaluessurvey.org/ table 3 shows descriptive statistics of the variables for the case of turkey. of the respondents 49 percent are female, 66 percent are married, 41 percent are in the 30-49 age interval, 17 percent are in the 50+ age interval and 42 percent are in the 16-29 age interval. rates of education level 1, 2, 3 and 4 are 52, 10, 27 and 11 percent respectively. the shares of the full-time, part-time and self-employed persons in all are 30, 2 and 13 percent respectively. the figures for the retired, students and the unemployed are 9, 8 and 6 percent. religiosity and national pride figures are relatively high at 80 and 89 percent. confidence in the entities of political system is lower compared to that in spain. income level rates indicate that the largest part of the population is in the lowest income level. 74 percent of the respondents think religion is important and 13 percent of them think politics is important in their life. when it comes to tax morale level, level 4 gives the highest rate which is that 80 percent of respondents find that cheating on taxes is never justifiable. besides, only 3 percent of them say that it is always justifiable. bilgin ● determinants of tax morale in spain and turkey 69 the ordered probit model is indicated by table 4. the dependent variable is the tax morale level, the independent variables are related to social capital, demographic factors, employment status and the economic situation. the model is statistically significant since the probability value of chi-square is much less than 0.01. the important part of the social capital variables appear to be statistically significant. importance of politics, importance of religion, religiosity, and national pride are all statistically significant at the 1 percent level. the estimated coefficients for the variables of confidence in government and confidence in the justice system are significant at the 5 and 10 percent levels. marginal effect of confidence in justice system on tax morale is 0.04 and is positive. those trusting the justice system have high tax morale. the same applies to confidence in government, which implies higher tax morale. taxpayers’ confidence in government means that they approve government’s taxation policies and its related decisions. the marginal effects of the importance of religion and religiosity are 0.12 and 0.08 respectively. in this case, it might be said that religion affects tax morale positively. as an explanation of this, religion extensively covers the beliefs about the right behaviour. according to hull (2000) religion has the ability to prevent illegal behaviours since it legitimises and reinforces social values. paying taxes is seen as a social norm and it is motivated by religion. the marginal effect of national pride is 0.13, so it is positive and among the largest marginal effects. it is clear that national pride supports tax morale. when people are satisfied with their national identities, they are usually devoted to their states and governments. none of the demographic factors are statistically significant. among the employment status factors, only unemployed and private sector have effect on tax morale. while the unemployed persons have less tax morale, private sector employees have higher tax morale than public sector employees. when it comes to the category of economic status, none of these factors can be evaluated as determinants of tax morale since the related estimations are not statistically significant at any level. the estimated model points out that the most effective factors on tax morale in turkey are the factors of social capital. when compared with the individualistic variables, the variables related to social capital have stronger relations with tax morale. the feelings of confidence, religiosity and national pride seem to be very important to understand the concept of tax morale in turkey. in other words what persons feel about themselves and how much they trust social entities are strongly related the level of tax morale. european journal of government and economics 3(1) 70 table 3: descriptive statistics of the variables for turkey variables mean std.deviation minimum maximum tax morale (level 1) 0.034201 0.181812 0 1 tax morale (level 2) 0.037175 0.189260 0 1 tax morale (level 3) 0.119703 0.324734 0 1 tax morale (level 4) 0.809822 0.393297 0 1 1) social capital importance of politics 0.131599 0.338180 0 1 importance of religion 0.748699 0.433923 0 1 confidence in justice system 0.353903 0.478358 0 1 confidence in government 0.260967 0.439325 0 1 confidence in parliament 0.203717 0.402911 0 1 confidence in civil service 0.099628 0.299615 0 1 religiosity 0.805204 0.396191 0 1 national pride 0.899628 0.300607 0 1 education (level 1) 0.504089 0.500169 0 1 education (level 2) 0.105576 0.307409 0 1 education (level 3) 0.278811 0.448581 0 1 education (level 4) 0.111524 0.314897 0 1 2)demographic factors gender (male) 0.501859 0.500183 0 1 gender (female) 0.498141 0.500183 0 1 16-29 age interval 0.402974 0.490678 0 1 30-49 age interval 0.418587 0.493511 0 1 50-98 age interval 0.178439 0.383024 0 1 marital status (unmarried) 0.337546 0.473049 0 1 marital status (married) 0.662454 0.473049 0 1 3) employment status full-time 0.228996 0.420343 0 1 part-time 0.027509 0.163623 0 1 self-employed 0.133829 0.340595 0 1 retired 0.094423 0.292526 0 1 housewife 0.343494 0.475051 0 1 student 0.088475 0.284091 0 1 unemployed 0.060964 0.239358 0 1 private 0.446840 0.497351 0 1 4) economic status income level (level 1) 0.609665 0.488007 0 1 income level (level 2) 0.180669 0.384887 0 1 income level (level 3) 0.088478 0.284091 0 1 income level (level 4) 0.101118 0.301593 0 1 financial satisfaction (level 1) 0.129368 0.335732 0 1 financial satisfaction (level 2) 0.231226 0.421774 0 1 financial satisfaction (level 3) 0.406691 0.491399 0 1 financial satisfaction (level 4) 0.228995 0.420343 0 1 data source: world values survey database; http://www.worldvaluessurvey.org/ bilgin ● determinants of tax morale in spain and turkey 71 table 4: ordered probit model estimation for turkey independent variables coefficient standard error z-value marginal effects 1) social capital importance of politics importance of religion confidence in justice system confidence in government confidence in parliament confidence in civil service religiosity national pride education (level 2) education (level 3) education (level 4) 2) demographic factors female 30-49 age interval 50-98 age interval married 3) employment status part-time self-employed retired housewife student unemployed private sector 4) economic status income level (level 2) income level (level 3) income level (level 4) financial satisfaction (level 2) financial satisfaction (level 3) financial satisfaction (level 4) number of observations prob (chisqd>value) information criterion (aic) pseudo r2 0.369703*** 0.435832*** 0.176320* 0.296404** -0.022028 -0.059884 0.299047*** 0.447546*** 0.040732 0.068301 0.271146 0.045841 -0.041882 0.038892 0.114222 -0.124043 -0.087781 0.040662 -0.121057 0.123917 -0.417013** 0.186067* -0.163385 -0.076305 0.083103 0.131735 -0.002190 0.008572 1345 0.0000000 1.27499 0.073878 0.133793 0.097435 0.105660 0.125291 0.134359 0.161653 0.105325 0.125907 0.142562 0.118135 0.167764 0.123619 0.106269 0.157216 0.110406 0.242470 0.142623 0.190237 0.166322 0.182094 0.171534 0.109411 0.108557 0.150657 0.158473 0.140117 0.131878 0.143839 2.763 4.473 1.669 2.366 -0.164 -0.370 2.839 3.555 0.286 0.578 1.616 0.371 -0.394 0.247 1.035 -0.512 -0.615 0.214 -0.728 0.681 -2.431 1.701 -1.505 -0.506 0.524 0.940 -0.017 0.060 0.0815 0.1213 0.0435 0.0699 -0.0056 -0.0155 0.0821 0.1318 0.0101 0.0170 0.0617 0.0116 -0.0106 0.0097 0.0294 -0.0331 -0.0229 0.0101 -0.0312 0.0298 -0.1232 0.0466 -0.0434 -0.0199 0.0204 0.0322 -0.0006 0.0022 notes: dependent variable is tax morale on a four-point scale (0 to 3). the reference groups: 16–29 age interval for the age variable, males, unmarried for marital status, full-time employed for employment status, public sector for the institution of occupation, education level 1 for education, income level 1 for income, financial satisfaction level 1 for the financial satisfaction variable. *, ** and *** refer statistically significant parameter estimates at the 10, 5 and 1 percent levels respectively. data source: world values survey database; http://www.worldvaluessurvey.org/ comparing these findings with the ones derived from spain’s model indicates that the determinants of tax morale in the two countries are not so similar. contrary to turkey case, confidence in political entities does not explain tax morale in spain. religion does not have such an important role as in turkey. under the social capital title, only national pride and education level 4 have a real effect on tax morale. tax morale in spain is more related to the individualistic factors while it is a social confidence phenomenon in turkey. age, education, income level and financial satisfaction are significant in spain. conclusion the determinants of tax morale were analysed by estimating ordered probit models for spain and turkey. for this aim, we estimated two models for each country. in the first model for spain, the variables of religiosity, national pride, education level 4, 30-49 age interval, 50+ age interval, student, income level 4, and financial satisfaction level 4 were found to be statistically significant at different significance levels. that is, all these variables have effects on tax morale in different degrees. in spain, mostly individualistic factors are explaining tax morale. as social capital european journal of government and economics 3(1) 72 variables, only national pride and one part of the education level have strong effects on tax morale. religiosity is significant at only the 10 percent level. the second model for turkey indicates that the variables of the importance of religion, importance of politics, confidence in the justice system, confidence in government, religiosity, national pride, unemployment, or private sector employment are statistically significant. the perceptions of religion and politics, confidence in the justice system, confidence in government, being religious, having national pride, and working for the private sector have positive effects on tax morale. the unemployed have lower tax morale. the findings in this paper suggest that the determinants tax morale in spain and turkey seem to be different. in other words, the first and second models give different results that suggest the different structures of tax morale in spain and turkey. the estimations based on the wvs data for spain and turkey indicates that tax morale in turkey can be seen as a social phenomenon where social capital variables have positive effects. personal factors such as gender, age, being selfemployed or retired, income level and financial satisfaction seem to be irrelevant. on the contrary, age, income level, financial satisfaction and education level have significant effects on tax morale in spain. it can be said that tax morale in spain is highly related to individualistic conditions when compared to turkey. from this point of view, increasing institutional credibility as a public policy might raise the level of tax morale in turkey leading to more tax compliance by taxpayers. on the other hand, the willingness of paying tax has nothing to do with the public institutional structure in spain. it should be found that the ways to increase the level of tax morale of the wealthy taxpayers by raising their perception of having benefited from public goods and services might give improving results. besides, increasing the general level of education and making younger people conscious of the benefits of public goods and services might increase the level of tax morale in spain. references alm, james and benno torgler (2005) ‘estimating the determinants of tax morale’, national tax association papers and proceedings, minneapolis, pp. 269274. alm, james and benno torgler (2006) ‘culture differences and tax morale in the united states and europe’, journal of economic psychology, 27(2): 224-246. chislett, william (2008) spanish trajectory: a source of inspiration for turkey, open society institute: istanbul. cummings, ronald g., jorge martinez-vazquez, michael mckee and benno torgler (2006) ‘effects of tax morale on tax compliance: experimental and survey evidence’, berkeley program in law and economics working paper series. duch, raymond m., harvey d. palmer and christopher j. anderson (2000) ‘heterogeneity in perceptions of national economic conditions’, american journal of political science, 44(4): 635-652 . feld, lars p. and bruno s. frey (2002) ‘trust breeds trust: how taxpayers are treated’, economics of govarnance, 3(3): 87-99. frank, robert h. (1996) ‘what price the morale high ground?’, southern economic journal, 63(1): 1-17. gökbunar, ali r., sibel selim and halit yanıkkaya (2007) ‘türkiye’de vergi ahlakını belirleyen faktörler üzerine bir araştırma’, ekonomik yaklaşım, 18(63): 69-94 . hull, brooks b. (2000) ‘religion still matters’, the journal of economics, 26(2): 3548. bilgin ● determinants of tax morale in spain and turkey 73 kaynar-bilgin, handan (2011) ‘türkiye'de vergi ahlakının belirleyicileri’, odtü gelişme, 38(2):167-190. martinez-vazquez, jorge and benno torgler (2009) ‘the evolution of tax morale in modern spain’, journal of economic issues, 43(1): 1-26. montero, jose r. and mariano torcal (2006) ‘political disaffection in comparative perspective’, in torcal, mariano and josé ramón montero (eds) political disaffection in contemporary democracies: social capital, institutions, and politics. london: routledge prieto, juan.; maría j. sanzo, and javier suárez-pandiello (2006) ‘análisis económico de la actitud hacia el fraude fiscal en españa’, hacienda pública española, 177:107-128. schmölders, günter (1976) genel vergi teorisi, i̇stanbul: fakülteler matbaası. torgler, benno (2004) ‘tax morale in asian countries’, journal of asian economics, 15(2): 237-266. torgler, benno (2005) ‘tax morale and direct democracy’, european journal of political economy, 21(2): 525-531. torgler, benno (2006) ‘the importance of faith: tax morale and religiosity’, journal of economic behavior & organization, 61(1): 81-109. torgler, benno (2007) tax compliance and tax morale: a theoretical and empirical analysis, cheltenham: edward elgar. torgler, benno and friedrich schneider (2006) ‘what shapes attitudes toward paying taxes? evidence from multicultural european countries’, berkeley program in law & economics, working paper series, no. 190. european journal of government and economics 3(1) 74 appendix definition of variables variable definition tax morale please tell me for the following statement whether you think it can always be justified, never be justified, or something in between: … ‘cheating on tax if you have the chance’ (3=never justified 0=always justified) 0 4,5,6,7,8,9,10 1 3 2 2 3 1 gender 0 male 1 female age how old are you? 18-29 30-49 50-98 marital status 0 unmarried (living together as married, divorced, separated, widowed, single/never married) 1 married education level 1 (no formal; inadequately completed elementary; completed elementary; incomplete secondary) level 2 (intermediate vocational; intermediate general) level 3 (maturity level; higher education) level 4 (university with degree) employment status are you employed now or not? 1 full-time 2 part-time 3 self-employed 4 retired 5 housewife 6 student 7 unemployed importance of religion how important is religion in your life? 0 not at all important; not very important 1 very important; rather mportant religiosity independently whether you go to church or not, would you say you are? 0 not a religious; a convinced atheist, other 1 religious importance of politics how important is politics in your life? 0 not at all important; not very important 1 very important; rather important national pride how proud are you to be (nationality)? 0 not at all; not very 1 very; quite confidence in justice system how much confidence do you have in justice system? 0 none at all; not very much; quite a lot 1 a great deal confidence in government how much confidence do you have in government? 0 none at all; not very much; quite a lot 1 a great deal confidence in parliament how much confidence do you have in parliament? 0 none at all; not very much; quite a lot 1 a great deal confidence in civil services how much confidence do you have in civil services? 0 none at all; not very much; quite a lot 1 a great deal scale of incomes scale of incomes? level 1 (lower step; second step, third step) level 2 (fourth; fifth) level 3 (sixth; seventh) level 4 (nineth; tenth) financial satisfaction how satisfied are you with the financial situation of your household? 0 1,2,3 1 4,5 2 6,7 3 8,9,10 institution of occupation 0 public sector 1 private sector microsoft word ejge_05_2016-008.docx european journal of government and economics volume 5, number 1 (june 2016) issn: 2254-7088 5 the role of market power in economic growth: an analysis of the differences between eu and us competition policy theory, practice and outcomes stephane ciriani, orange, france marc lebourges, orange, france abstract the european union has experienced weak economic performance over the past 15 years, compared to the united states. in order to restore investment, innovation, and therefore growth, the european commission seeks to raise the level of static competition in all markets. the commission’s economic policy is largely determined by its competition policy. this policy is derived from its doctrine on competition law, which regards the exercise of market power as a source of inefficiency and advocates that its effects should be banned. by contrast, the united states competition authorities, under the influence of the chicago school, consider that market power is a necessary incentive to invest and a fair return on investment. recent findings in economic growth theory, which state that increased competition intensity may harm endogenous innovation, provide a theoretical basis to support the united states approach and call for a review of european doctrine. keywords antitrust; competition; endogenous growth; innovation; market power; market structure. jel classification l10; l40; o30; o40. ciriani and lebourges ● the role of market power in economic growth 6 economic growth in the european union (eu) over the past fifteen years has been weak by international standards, and the pace of recovery from the financial crisis is slow. the economy of the eu has started to lose ground against the united states (us) since the mid-1990s. the european commission has acknowledged that the weak economic performance of the eu is notably due to low investment in new technologies, and pointed at the need to increase the innovation effort in order to foster economic growth in the eu and to compete on a global scale. the economic policy of the european commission is largely influenced by the principles and actions of the european competition authorities. the commission’s economic doctrine posits that the ever-increasing level of competition intensity invariably leads to higher investment in innovation, and that competition policy is a major instrument to foster economic growth. the core principle of the european competition policy is that market power, although not considered unlawful, is the main source of economic inefficiency, and its effects should therefore be removed. an analysis of the principles and the effects of this economic doctrine is proposed, in order to examine whether the commission’s policy is the most appropriate to reverse the structural trends of investment and productivity in the eu. the first section describes european commission’s economic doctrine and its views on the role and the effects of competition policy, on the basis of reports and communications published by the european authorities. this section shows that the european commission applies its competition policy in order to increase the level of static competition intensity in all markets (i.e. to eliminate all the effects of the exercise of market power), with the objective to promote aggregate investment in equipment, technology and innovation. the second section provides a comparative analysis of the principles and the practices of the competition courts and agencies in the us and the eu. it highlights the differences between the us and the european authorities over the purpose of competition law and over the way authorities address market power. the european competition doctrine attempts to prevent the emergence and to ban the exercise of market power, market power being considered as a factor of inefficiency per se. by contrast, the us authorities tend to encourage the effects of market power, after their doctrine experienced an in-depth review in the 1970s, led by the chicago school of economics. the third section presents recent findings in the fields of economic growth theory and innovation economics, which show that market power is both a necessary incentive and a fair return on investment and that a systematic increase in the level of static competition might end up discouraging endogenous innovation. a fourth section provides an assessment of the consistency of the european and the us doctrine on competition law and competition policy with these recent results. it concludes that the european commission should review and update its doctrine in order to promote robust and sustainable economic growth in the eu. 1. the european commission’s doctrine on the role and effects of competition policy this first section describes the european commission’s economic doctrine on the nature of competition, the concept of market power, and the role of competition policy as the main engine of economic growth. this economic doctrine is made visible and explicit through a set of quotes extracted from a body of documents published by various european commission entities over the last ten years. the analysis of these quotes helps formalise the core principles of the commission’s economic doctrine, which underlies its competition policy. this analysis also demonstrates how the commission applies these principles in order to tackle european weaknesses in terms of investment in technology and innovation and to meet the objectives of the europe 2020 strategy, whereby the commission notably european journal of government and economics 5(1) 7 commits to ‘develop an economy based on knowledge and innovation’, and to promote a ‘more competitive economy’. this section provides evidence that in order to stimulate aggregate investment, the european commission aims to increase the level of static competition in all markets, based on the doctrine that more intense competition invariably leads to higher rates of innovation and increased productivity. in the rest of this article, the term ‘market power’ refers to ‘the ability to set price above marginal cost’ as defined by landes and posner (1981), who state that if ‘a firm’s price is above its marginal cost, the implication is that the firm does not face perfect competition, i.e., that it has at least some market power’. 1.1. competition policy is the core of the european commission’s economic doctrine in its most recent report on competition policy, the european commission (2015a) outlines a primary objective of its competition policy, which consists in being ‘vigilant that manufacturers of important input products do not acquire the power to raise prices above competitive levels through mergers’. the european commission (2012) advocates competition as the main driver of economic growth and regards competition policy as the main instrument to support productivity and competitiveness. competition is viewed as ‘the most important single driver of innovation, competitiveness and therefore growth’, while ‘competition policy can create the conditions for growth and employment across europe’. moreover, ‘promoting strong competition enforcement in europe -at every levelis a powerful energiser for the economy because its effects can be immediate’. the european parliament (2015) recently recalled that competition policy remains at the core of the economic policy of the eu-28. it stated that ‘for a long time, competition policy has been providing a prevailing conceptual model influencing policy developments at eu level, which is still at work nowadays’, and that ‘there is actually little call for a radical overhaul in this respect’. the european commission (2012a) considers that competition always drives economic growth and maximises social welfare, as ‘competition enforcement and advocacy serve wider long-term objectives such as enhancing consumer welfare, supporting the eu’s growth, jobs and competitiveness in line with the europe 2020 strategy for smart, sustainable and inclusive growth’. the commission (2004a) has identified competition as the major policy instrument to foster economic growth, stating that ‘vigorous competition is a key driver for competitiveness and economic growth’, which ‘generates pressures on firms to innovate and (…) improve their cost structure and reap productivity gains.’ competition is regarded as a ‘continuous structural adaptation process’ and ‘a vital market process which rewards firms offering lower prices, better quality, new products, and greater choice’. for the commission, competition is an evolutionary process that selects the most efficient firms: ‘competition leads to the introduction of improved products and processes, weeding out inefficient firms and reallocating productive resources from retreating firms to new entrants or more efficient competitors’. competition separates efficient firms from inefficient ones and reallocates inputs and financial resources to the most efficient activities. by doing so, the most productive sectors are endowed with greater resources at a lower price, allowing them to invest in private fixed assets. a crucial argument raised by the european commission in favour of its competition policy is that it reallocates resources from inefficient sectors to the most productive ones. this process generates productivity gains and spreads them across sectors. the commission (2011a) aims to support ‘structural change towards dynamic sectors and high-productivity activities’, through the reallocation of production factors and financial capital from ‘slow-growing to dynamic firms and sectors’, and from ‘non-tradable to tradable sector’, hence, from services to industrial sectors. ciriani and lebourges ● the role of market power in economic growth 8 this reallocation depends on the ‘strict implementation of competition policies at both eu and national level’. the cross-sector reallocation of these productive resources depends on markets’ ability to give investors ‘adequate price signals’. since price signals are fully accurate only under perfect competition, the productive inputs will move towards the most efficient and productive sectors provided that the markets actually tend towards perfect competition. a fall in the price of intermediate inputs which are produced by service sectors should result from a convergence of markets towards their perfectly competitive frame. 1.2. for the european commission, macroeconomic performance depends on the control of markets by the competition authorities according to the european parliament (2013), the instruments of competition policy (antitrust, merger control, state aid and ex-ante regulation of specific sectors) promote ‘growth drivers such as productivity, innovation, investment and low prices’. following this rationale, the competition policy enforced by the commission’s agencies is assumed to achieve static efficiency through price competition that leads to equating prices and marginal costs, simultaneously generating dynamic efficiencies by fostering innovation through an evolutionary process of rivalry. the european authorities believe that in order for competition to fully deliver its growth-enhancing effect, markets have to be organised by competition authorities. the european commission (2011a) recommends to strengthen ‘the administrative capacity of the competition authority.’ the policy priority to ‘create a businessfriendly environment conducive to improving the economy’s competitiveness’ has to be achieved through the ‘efficient functioning of competition authorities, market regulators and judicial authorities’. the european commission (2015a) outlines that technological progress stems directly from innovation and that a strict enforcement of antitrust law is the primary condition to stimulate innovation: ‘vibrant competition is essential to stimulate innovation and spread the benefits of technological development among europe’s citizens’, and ‘effective enforcement of antitrust and merger policy makes it easier for small businesses to thrive and gain access to markets in sectors dominated by network effects’. 1.3. the commission acknowledges that the european union has been losing ground in terms of competitiveness because of low investment in innovation even prior to the financial crisis the european commission (2015b) acknowledges that ‘productivity growth remains slow’ in the eu, ‘affecting competitiveness and living standards’, and that the recovery from the post-crisis recession remains moderate. according to van ark and erumban (2015), the eu has had a slower productivity growth rate than the us since the early 1990s, which was largely due to a slower pace of innovation and technology adoption, mainly in the services sector. the commission acknowledges that the economic weaknesses of the eu might be related to structural inefficiencies rather than being just the outcome of the financial crisis which broke out in 2008. the european commission (2010a) indicates that ‘even before the crisis, there were areas where europe was not progressing fast enough relative to the rest of the world’, and that ‘europe’s average growth rate has been structurally lower than [its] main economic partners, largely due to a productivity gap that has widened over the last decade’. this productivity gap is notably due to ‘lower levels of investment in r&d and innovation’, and ‘insufficient use of information and communications technologies’. european journal of government and economics 5(1) 9 the european commission (2013a) has more recently pointed at the structural weaknesses of the european economy, notably, slow productivity growth and a lack of contribution of information and communications technologies (ict) to aggregate productivity growth. the commission (2004b) also expressed concerns about the labour productivity slowdown in the union and the widening gap with the us over 1995-2004, especially in high-technology sectors. the commission (2004b) showed that growth in european high-technology sectors’ productivity had been stronger than that of the overall economy of the eu, but lower than that of their us counterparts. the eu has experienced a slowdown in the productivity of its manufacturing industries and also of its high-technology sectors such as the digital industries. moreover, gorning and schiersch (2014) find that the eu in general and the euro area in particular had a lower level of investment than the organisation for economic co-operation and development (oecd) average during 1999-2007. the commission (2009a) warned that the ‘potential gdp of the european union could fall to a permanently lower trajectory’, as ‘the stock of equipment and infrastructure will decrease and become obsolete due to lower investment’. 1.4. the european commission acknowledges the crucial role of investment in economic growth the european commission (2012b) acknowledges that investment is the driving force behind economic growth. it points to the need to keep up with the pace of technological progress in order for european firms to compete on a global scale: ‘new investment is now urgently needed to stimulate economic recovery and bring innovation and new technologies back onto factory floors. if europe does not keep up with investment in the adoption and diffusion of these technologies, its future competitiveness will be seriously compromised’. the commission (2014a) insists on the need to increase investment in all sectors of the economy, especially in equipment, in order to improve the competitiveness of the eu. the commission (2014b) reaffirms that investment to restore potential growth is a key policy priority and that the slow accumulation of fixed capital has hampered productivity and growth in the union over 2007-2014. the commission (2014c) also acknowledges that strong contraction of investment since 2008 is one of the primary reasons for the weakness of the recovery and that persistently low investment deteriorates potential growth. 1.5. the commission acknowledges that private investment can be harmed by weak profitability and that expected profitability influences investment the european commission acknowledges that firms have to restore profitability to sustain investment. the commission (2013b) indicates that ‘the german case in the 2000s illustrates how restoration of profit margins may also have contributed to fostering investment and the possible link between profit margins and investment and innovation’. the commission (2012c) acknowledged that sufficient profitability is required to finance investment: a ‘reduction in the profitability of french firms, which reached historically low levels in 2011, weighs on their investment potential as well as their innovation capacity to the detriment of their non-cost competitiveness’. the commission (2013b) indicates that ‘the disappointing evolution of firms’ profitability is particularly alarming because it may prevent firms from raising their investment in equipment, r&d, marketing, brand’. the commission also acknowledges that expected profitability drives investment forward. it relies upon jorgenson (1963) who links investment to expected profitability and tobin (1969) who notes that investment decisions depend on the expected value of future profits. moreover, herbet (2001) shows that restoring profit margins might have encouraged investment in the main developed countries in the 1990s. the commission (2011a) acknowledges the link between investment ciriani and lebourges ● the role of market power in economic growth 10 decisions and expected profitability: ‘expectations of higher growth will contribute to restoring confidence and stability on financial markets. with improved prospects, businesses will start to invest again’. the commission recognises both the need for profit margins to finance current investments and the need for sufficient expected returns to plan future investments. 1.6. to support private investment, the commission recommends fostering competition in order to tackle low profitability by reducing input costs in 2008, the european commission expressed its commitment to a strong application of competition rules during the financial crisis, based on the view that ‘competition enforcement strengthens the economy, even perhaps particularly in times of difficulty’. more recently, the commission (2015b) indicated that competition has to be strengthened in order to encourage investment and efficient allocation of resources, through eased market entry and lower prices of services. for the commission (2015c), a trade-off between competition and innovation is not empirically relevant, and competition succeeds ‘in fostering productivity thanks to higher investments, better managerial organisation and innovation’. in addition, the commission (2013c) recommends to increase competition in the markets of intermediate inputs in order to raise industrial investment. business services are essential inputs for the manufacturing sector and represent an important share of production costs. for the commission, business services (including network industries) are sheltered from international competition while industrial sectors are exposed to foreign competitors. opening sheltered business services to competition would then lower the cost of services used as inputs by exporting industry sectors. the european commission (2014d) affirms that increased competition intensity in these sheltered intermediary markets could contribute to lowering services’ costs, hence improving competitiveness of the economy. it posits that ‘low levels of competition in services translate into higher intermediary costs for firms and less innovative services’, and as a result, ‘further stimulating competition both in the product and services sectors would benefit exporting firms and contribute to their competitiveness’. the commission (2013c) posits that with lower factor costs, firms will restore their profit margins and their investment capacities. lower production costs due to greater competition also ‘strengthen entrepreneurs’ expectations due to improved competitiveness that has translated into the recovery in investment’. the other priority to restore investment is to overcome firms’ insufficient access of to external sources of funding, through the capital increase of the european investment bank, the allocation of structural funds and the implementation of the european commission’s action plan on access to finance for small and medium businesses (2011b). in the opinion of the commission, firms with insufficient profit margins are supposed to overcome the shrinkage of their internal resources by turning to external (and notably public) funding sources to finance their investments. 1.7. for the european commission, dynamic efficiencies arise directly from static efficiency the european commission promotes static competition, and aims to achieve static efficiency, a state where firms are not able to set their price above their marginal costs of production and make no profit. in this perfectly competitive frame, pricecost margins are eliminated, and all the effects of market power are removed. the european competition policy strives to eliminate the effects of market power at the level of upstream markets by lowering the intermediate inputs’ and capital goods’ prices, which in turn reduces production costs and leads to lower prices in the european journal of government and economics 5(1) 11 downstream markets. in addition, the commission affirms that investment depends upon the wide availability of low-cost production factors, and expects investment to arise from intense static competition. the commission considers that dynamic efficiencies arise directly from static efficiency. in the commission’s view, the resources needed to finance investment do not accrue from expected profit margins as fair return on investment, but from a greater availability of low-cost inputs. this greater availability stems from increased static competition in the intermediate input markets. competition is either viewed as a steady state of static efficiency (where prices equal marginal costs) or an evolutionary process generating dynamic efficiencies. the commission considers that investment arises exogenously from high levels of static competition. the commission does not address the fact that the producers of intermediate inputs and capital goods might be discouraged to invest because of either insufficient funding resources or insufficient profit prospects. 2. the differences between the european and the us doctrine on competition law and practice of competition policy this section describes the differences between the us and the european practices of competition policy, and explains how these differences can influence economic outcomes. the differences in the practices of competition policy relate to different doctrines on the objectives of competition law, the effects of market power on economic performance and the autonomy of dominant firms in setting their commercial and industrial strategies. the european competition policy strives to tackle all the effects of market power. by contrast, the us competition authorities, under the influence of the chicago school, regard market power as a necessary condition to foster private investment, and do not consider the exercise of market power as inefficient. in practice, the us authorities intend to maintain the incentives of firms to invest in order to gain market power, whereas in the eu, the competition authorities consider that the effects of market power need to be eliminated. 2.1. the european competition authorities monitor markets in order to ban market power effects the european competition authorities are empowered to monitor markets to ensure they reach their highest level of static competition. the european commission (2009b) considers that a firm which is able to ‘profitably increase prices above the competitive level for a significant period of time’ does not ‘face sufficiently effective competitive constraints and can thus generally be regarded as dominant’. the commission considers that ‘persistently high market shares may be indicative of the existence of barriers to entry and expansion’. consequently, it considers that ‘the stronger the dominant position, the higher the likelihood that conducts protecting that position leads to anticompetitive foreclosure’. the practical elements of competition policy relating to concentrations are defined in the guidelines on the assessment of mergers. in the guidelines on the assessment of horizontal mergers, the european commission (2004) explains that it must assess ‘whether or not a concentration would significantly impede effective competition, in particular as a result of the creation or strengthening of a dominant position’. according to these guidelines, the commission must ‘take into account any significant impediment to effective competition likely to be caused by a concentration. the creation or the strengthening of a dominant position is a primary form of such competitive harm’. market shares are the indicator of static competition intensity: the european commission (2004) indicates that ‘the larger the market share, the more likely a firm is to possess market power. and the larger the addition of market share, the more likely it is that a merger will lead to a significant increase in market power’. the commission notes that threats on ciriani and lebourges ● the role of market power in economic growth 12 effective competition do arise when ‘the merged entity has a significant degree of market power (which does not necessarily amount to dominance) in at least one of the markets concerned’. 2.2. the appraisal of market power by the european competition authorities might hamper the incentives of private firms to invest and innovate as explained by fox (1997), the european commission and the european court of justice ‘readily presume dominance and increases in dominance without the kind of factual records that might be required in the united states.’ the european competition law considers that a dominant firm is responsible for the competitive structure of the market. according to marty and pillot (2010), european competition law assigns the dominant firm a ‘responsibility for competition’. the authors argue that the application of the doctrine of essential facilities to intangible assets like intellectual property has limited the strategic options of dominant firms. according to lang (1979), in the case of united brands in 1978, the european court of justice defined dominant position as the ability of a firm to prevent effective competition being maintained ‘even if that power has not been exercised and so effective competition has not been ended’. for marty and pillot (2010), the 1979 hoffman-laroche judgment shows that the european competition authorities act to limit the market dominance of a firm because market power is regarded as a distortion of the competitive market structure. therefore, competition authorities can sanction an abuse of dominant position on the basis of anticompetitive behaviour rather than on the basis of actual anticompetitive effects. marty (2012) indicates that the 2003 decision of the court of justice (michelin ii) further confirmed that a dominant firm can be sanctioned because its behaviour might undermine competition, and not necessarily because of its actual anti-competitive effects. marty (2012) also argues that, in the case of teliasonera’s abuse of dominant position in 2011, the european court of justice reaffirmed the liability of the dominant firm to maintain ‘effective’ competition. marty (2012) argues that the 2011 judgment provides further confirmation of the european court of justice’s 2007 statement on france telecom: ‘showing an anticompetitive object and an anticompetitive effect may, in some cases, be one and the same thing. if it is shown that the object pursued by the conduct of an undertaking in a dominant position is to restrict competition, that conduct will also be liable to have such an effect’. if a dominant firm is shown to intend to distort competition, the european competition authorities can sanction it as an abuse of dominant position, even in the absence of anti-competitive effects. 2.3. the influence of the chicago school of antitrust has reoriented the us practice of competition law towards the promotion of economic efficiency the doctrine underlying the practice of competition law in the us experienced a major revision in the early 1970s, which, under the influence of the chicago school of antitrust, led to a more flexible approach to market concentration and the behaviour of dominant firms. prior to this revision, between 1940 and 1970, us competition law was influenced by the ‘structuralist’ approach developed by the harvard school of economics, which considered that concentrated market structures led to anticompetitive behaviour and harmed economic welfare. during this period, the us competition authorities widely applied per se rules, according to which numerous practices of dominant firms where considered illegal per se, independently of their actual effects on the markets. this emphasis on market structures and per se rules led the us courts and competition agencies to broaden the range of conducts likely to be sanctioned, and to apply a strict antitrust policy. european journal of government and economics 5(1) 13 for kovacic (2007), ‘the courts defined the concept of wrongful behaviour so broadly that a wide range of conducts sufficed to create liability for dominant firms’, and ‘judicial decisions adopted an exceptionally expansive view of abuse’. the revision of the us doctrine on antitrust law which occurred by the early 1970s was inspired by the chicago school’s critical views regarding per se rules and their advocacy for economic efficiency ‘as the exclusive basis for the design and application of antitrust rules’, as noted by kovacic (2007). in addition, marty (2010) mentions that in the opinion of the chicago school, the aims of firms cannot be regarded as a proof of anticompetitive conduct, and that anticompetitive behaviour needs to be sanctioned on the basis of its actual effects on the market and on the basis of its impact on economic efficiency, rather than on the basis of its potential effects. for piraino (2007), the approach of the chicago school was adopted by the us courts and agencies in order to ‘redress the harsh approach of the harvard school, which often invalidated conduct that had the potential to enhance firms' efficiency and thereby benefit consumers’. for kovacic and shapiro (2000), this adoption was motivated by the observations that us firms were losing competitiveness both on foreign markets and in the domestic market, and the idea that their declining performances might be induced by an excessive enforcement of antitrust policy. they explain that the chicago school claimed that conduct such as vertical restraints were ‘often benign or procompetitive’ and that ‘many phenomena, including industrial concentration, mergers, and contractual restraints’ were sources of economic efficiency and should be allowed by the us courts and competition agencies. as a result, the us courts gave ‘dominant firms considerable freedom to choose pricing, product development and promotional strategies’. for kovacic (2007) the trend of us antitrust doctrine over the past thirty years ‘has been to give dominant firms greater freedom to select pricing, product development, and distribution strategies’, and ‘the progression toward greater doctrinal permissiveness has not been unbroken’. 2.4. the us doctrine on competition law regards market power as a source of economic performance and vertical integration as a potential source of efficiency two key contributions of the chicago school to the practice of us competition law relate to the appreciation of market power and the assessment of vertical relationships. the chicago school considers that market power reflects the ability of firms to improve their performance and that sanctioning firms which acquired monopoly power might discourage them from competing intensively in order to provide consumers with higher quality products. according to marty (2010), the us competition authorities have considered since the 1970s that the exploitation of a dominant position is lawful and justified because it allows invested capital to be recouped and is a fair return on a risky investment. moreover, market power is an incentive for competitors to enter the market, and as such should not be sanctioned by the authorities. for marty (2010), the chicago school posits that a firm which acquire market power and extract an economic rent will increase social welfare in the short run. in the long run, its dominance will be challenged by new competitors, based on a self-correcting process inherent in the market. as mentioned by porter (2002), market power (prices above the marginal costs of production) allows for improvements in the quality of products, and ‘high-value products provide the consumer with superior performance and features, and therefore justify higher prices’. porter (2002) argues that higher prices should only be a concern for competition authorities if they are not justified by increased value for consumers, and claims that ‘firm profitability is a good thing if it reflects truly superior products or significant advantages in process technology or operating efficiency’, and that ‘it is a bad thing if it occurs in the absence of a healthy rate of dynamic improvement’. therefore, antitrust authorities that strive to ‘limit short term ciriani and lebourges ● the role of market power in economic growth 14 price/cost margins or profitability’ might underestimate the value brought to the consumer by dynamic competition and the related efficiency improvement. the measure of consumer welfare would then be affected by a downward bias and lead to decisions that are detrimental to consumers. according to riordan (2005) the chicago school posits that vertical integration increases economic efficiency and rejects the structuralist view that it induces anticompetitive behaviour such as exclusionary practices and the ‘leverage of monopoly power from one market to another’. rey and tirole (2007) explain that the chicago school, led by the works of bork (1978) and posner (1976), considered that the ‘leverage concept resulted from confusion about the exercise of market power’. bork and posner opposed the doctrine that vertical integration led to foreclosure. they claimed that a monopolist in the upstream market which can extract the entire monopoly profit from a competitive downstream market cannot obtain additional profits by extending its monopoly power. in the absence of efficiency gains, a vertical merger between an upstream monopolist and a downstream firm ‘cannot increase the profitability of the merging firms’. there is only a single source of monopoly profit, which is unrelated to the extension of market power to the downstream market. the only possible rationale of vertical mergers relates to efficiency grounds (the possibility of reducing production costs) and not to conduct related to the foreclosure of competitors on the downstream market. 2.5. the european and the us doctrine and the practice of competition agencies differ on the effects of market power and on the evidence of anticompetitive conduct contrary to the us, the doctrine underlying the practice of competition law in europe has not yet experienced an in-depth critical review. for marty and pillot (2009), the european competition authorities are still significantly influenced by the ordo-liberal school, or freiburg school founded in the 1930s. this school of thought regards unrestricted competition as a founding principle of public action, and posits that the fundamental role of public authorities should be to preserve competition in itself. dominance is viewed as a stable and permanent situation which cannot be tackled by the market and which requires public intervention to be eliminated. the primary role of competition policy is to promote and preserve the access of dominant firms’ competitors to the market. hence, the authorities’ assessment of market structures should prevail over their assessment of the economic effects of dominant firms’ behaviour. marty (2010) asserts that the ordoliberal doctrine bans the concentration of economic power because such concentration prevents competitors’ entry. the purpose of european authorities, under the ordo-liberal influence is to preserve the competitive market structure rather than to promote a market structure that maximises efficiency. for marty (2010), the european court of justice’s interpretation of the ordo-liberal doctrine entails a degree of similarity with the structuralist approach of the us authorities before 1970 as both can lead public authorities to attempt to prevent dominant positions per se. marty (2010) highlights that in the structuralist as well as in the european court of justice’s views, the public authorities’ duty is to prevent all dominant firms’ behaviour that could potentially foreclose competitors and limit their ability to grow. for cooper et al. (2005), ‘ordoliberalism tends to focus on the form, rather than the competitive effect of business relationships’. the european doctrine on competition law regards market power as self-perpetuating, and regards vertical relationships as prone to anticompetitive conduct. for example, vertical restraints such as exclusive dealing and exclusive supply contracts ‘necessarily foreclose competitors from contracting opportunities, and thus may be seen as perpetuating dominance’. in the context of persistent ordo-liberal influence, the european doctrine on vertical restrictions has recently shifted to an approach closer to the post-chicago analysis, european journal of government and economics 5(1) 15 as argued by marty (2007) and kovacic (2008a). the post-chicago analysis, which developed in the early 1990s, has criticised the chicago school’s view that vertical mergers are pro-competitive and that vertical integration generates efficiencies. according to riordan and salop (1995) the ‘post-chicago’ approach considers that vertical mergers can have anticompetitive effects by ‘creating, enhancing, or facilitating the exercise of market power’, and that vertical mergers ‘can be motivated by monopoly power, economic efficiency concerns or both’. the postchicago analysis on vertical integration differs from the chicago school’s view because it considers that the upstream monopolist has a rationale for exercising its market power in the downstream market, otherwise downstream competition may limit its ability to exercise its upstream market power. therefore, contrary to the chicago school’s view, the rationale for a vertical merger might not always be related to the prospect of increasing efficiency. the european commission (2015d) agrees with the view that dominance in the upstream market is likely to extend to the downstream market, reducing the incentives of downstream firms to innovate because ‘the rents which downstream firms can expect from efficiency improvements are likely to be partially captured by the suppliers of the intermediate inputs upstream’ and that ‘a lack of competition in the upstream market can generate entry barriers limiting competition downstream if access to downstream markets requires using intermediate inputs produced upstream’. for kovacic (2008b), the european competition authorities have tended to ‘create a wider zone of liability for dominant firms than the decisions of the us courts’ and that ‘courts and enforcement agencies commit greater errors by intervening too much rather than too little. this perspective does not appear in eu jurisprudence or in speeches by eu enforcement officials’. this quote implies that, in the us perspective, overly restrictive antitrust enforcement would likely harm the economy because it could prevent efficiency gains from occurring. by contrast, in the european perspective, overly restrictive enforcement is to be preferred to an overly flexible approach because it is better to prevent anticompetitive behaviour from arising than to allow for potential efficiency gains to occur. in its guidelines on vertical restraints, the commission (2010b) outlines that the softening of competition (along with foreclosure and collusion) at the wholesale level (in the upstream market) harms consumers ‘by increasing the wholesale prices of the products, limiting the choice of products, lowering their quality or reducing the level of product innovation’. it might also harm consumers at the retail level ‘by increasing the retail prices of the products, limiting the choice of priceservice combinations and distribution formats, lowering the availability and quality of retail services and reducing the level of innovation of distribution’. by contrast, according to marty and pillot (2012), the us supreme court stated that for industries where there is an ex-ante regulation, such as network industries, the enforcement of antitrust to tackle market power of an upstream firm is not necessary: ‘when there exists a regulatory structure designed to deter and remedy anticompetitive harm, the additional benefit to competition provided by antitrust enforcement will tend to be small, and it will be less plausible that the antitrust laws contemplate such additional scrutiny’. this difference in antitrust enforcement relates to the different doctrines adopted in the us and in the eu, where the post-chicago approach recommends eliminating the effects of market power in upstream markets in order to prevent dominance being exercised in downstream markets, advocating a strict control of vertical restraints and vertical mergers. by contrast, the chicago school does not recommend such strict antitrust enforcement, because market power is not viewed as a source of inefficiency and because vertical integration is viewed as a source of efficiency gains. ciriani and lebourges ● the role of market power in economic growth 16 2.6. market power (mark-ups on competitive price) is not regarded as economically inefficient by the us competition authorities for shapiro (2011), the us antitrust law has ‘understood for a very long time that the market power resulting from successful innovation is an important and inevitable part of the competitive process’, and that the us merger policy does not strive for perfectly competitive market structures. shapiro (2011) does not find any conflict between ‘competition policy principles and the schumpeterian observation that successful innovators often are able to price well above marginal cost and often gain substantial market shares.’ according to marty and pillot (2009), a core principle of us competition policy is the preservation of firms’ incentives to acquire market dominance. according to the us competition law, the exercise of market power is in the interest of the global economy. the authors argue that, contrary to the european approach, the us competition law does not strive to limit the strategic autonomy of the dominant firm. it considers that the purpose of competition law is not to affirm the specific responsibility of the dominant firm towards its competitors and the market structure. in this view, the dominant firm is not responsible for maintaining its competitors on the market. the dominant position is therefore a fair return on the capital invested. market dominance provides incentives to invest in order to challenge the market power of the dominant firm. as noted by newman (2014), the antitrust division of the us department of justice acknowledges that the social cost of false positive (i.e. wrongly sanctioning nonanticompetitive behaviour) is higher than the false negative (i.e. failing to condemn real anticompetitive behaviour). this practice is justified by the belief that markets can efficiently correct ‘judicially unchecked anticompetitive behaviour’ while they are not able to ‘correct a judicial ruling that wrongly condemns benign behaviour’. in verizon vs. trinko (2004), the us supreme court stated that ‘the charging of monopoly prices was not a breach of the sherman act’ unless ‘it is accompanied by an element of anticompetitive conduct’, and that forcing the dominant firm to share the source of its competitive advantage with its competitors contradicts the purpose of antitrust policy. in the case of blue cross (1995), the court of appeals stated that a lawful monopolist ‘may charge any price that he wants, for the antitrust laws are not a price-control statue or a public-utility or a common-carrier rateregulation statue’. the case of microsoft (2004) evidenced the differences between the us and the eu in the appraisal of market power. the €497 million fine imposed by the european commission on microsoft for abusing its dominant position has been criticised by us competition law practitioners and theorists alike. e.-m. fox (2006) claimed that the decision limited the strategic autonomy of the dominant firm, and reflected ‘europe’s principle of special responsibility of the dominant firm and the prohibition of using leverage and power to obtain advantages over rivals. these principles overshadow any freedom of the dominant firm to choose with whom and how to deal’. moreover, t.o. barnett (2007) opposed the european commission’s decision, stating that ‘in the absence of demonstrable consumer harm, all firms, including dominant firms, are encouraged to compete vigorously’. barnett (2007) outlines that the us competition authorities favour the strategic autonomy of dominant firms over maintaining competitive market structures: ‘us courts recognize the potential benefits to consumers when a firm, including a dominant firm, makes unilateral business decisions, for example to add features to its popular products or license its intellectual property to rivals, or to refuse to do so.’ european journal of government and economics 5(1) 17 2.7. the purpose of us competition law is to promote incentives to gain market power, whereas the purpose of the european competition law is to maintain ‘effective’ competition the us competition law does not ban the effects of market power and concentrated market structures as long as they do not result from unfair anticompetitive behaviour. market power is not viewed as a barrier to entry as long as it results from a firm’s efforts to acquire and develop a competitive edge. the european competition authorities ban the effects of market power on the basis of the ordo-liberal doctrine. the influence of the post-chicago approach induces european authorities to prevent market power leverage (i.e. the extension of upstream market power to the downstream market). the european competition authorities prevent dominant firms from deriving the full return on their investments because they consider that exercising market power is economically inefficient. the appraisal of market power by the european competition authorities might as a result hamper private incentives to invest. the european practice of competition law tends to limit the strategic autonomy of the dominant firms, because its objective is to promote the access of its competitors to the market rather than to maximise efficiency. 3. the endogenous nature of innovation and its relationship with competition this third section builds on recent results from theoretical and empirical economics which call the european commission’s doctrine into question. these results explain why a systematic increase in the level of static competition might discourage investment. technological progress only occurs provided that firms respond to incentives to invest in new technologies and products, and these incentives depend on existing and expected levels of competition. there is a tradeoff between the level of competition and endogenous innovation, so that dynamic efficiencies hardly arise from static efficiency, especially in industries with high innovation rates. 3.1. technological progress is created by economic agents who respond to market incentives in the neoclassical framework, technological change is exogenous to the market. in the seminal model of robert solow (1956), technological change is a public input provided exogenously to the market. it is a non-rival and non-excludable good. solow (1957) posits that growth is sustained in the long run by technological change, which only depends on the evolution of time. paul romer (1986) renders technological change endogenous: it is the outcome of private investment. longterm growth is sustained by the accumulation of knowledge, a process which occurs through the investment of private firms. romer (1990) posits that technological change results in large part from ‘intentional actions taken by people who respond to market incentives’. technological change occurs only if ‘selfinterested individuals’ can capture a sufficient benefit from their investment in the improvement of technology, whose benefits are ‘at least partially excludable’. market incentives are crucial to technological progress because they drive the transformation of new knowledge into goods with practical value. romer (1990) explains that ‘initial understanding of electromagnetism arose from research conducted in academic institutions, but magnetic tape and home videocassette recorders resulted from attempts by private firms to earn a profit’. to increase productivity, firms have to invest in capital goods which incorporate technological innovations. the driver of this investment is the prospect of earning a profit. this assertion implies that unless there is private investment in fixed assets, available innovations per se will not have any effect on productivity. ciriani and lebourges ● the role of market power in economic growth 18 3.2. firms invest in the production of technological innovation provided there exist sufficient expected reward, hence a sufficiently concentrated market structure for romer (1990), the creation of innovations is equivalent to incurring a fixed cost, and once they are produced, they can be used at no additional cost. technological innovations are partially excludable and non-rival goods. the costs incurred to produce new goods are recovered provided they are sold at a price higher than their marginal cost of production. therefore, endogenous innovation cannot occur under price-taking competition. the investment in capital goods that embody innovations depends on the ability of investors to recover their investment. a certain degree of market power is needed for investments to occur, thus for technological progress to spread through society. romer (1994) posits that endogenous growth theory has allowed economists to take account of two evident facts on innovation: the endogenous nature of technological progress and the existence of market power and monopoly rents stemming from innovations. as noted by shapiro (2011), ‘we would not expect to see atomistic market structures in industries that have experienced significant technological progress, and we may see high levels of concentration in markets that have recently experienced significant innovation’. a notable policy implication of endogenous growth models is that imperfect competition is needed to support the accumulation of knowledge through discoveries and the spread of technological progress through investment in capital goods. firms are incentivised to invest provided they expect temporary monopoly rents in the form of mark-ups over competitive prices or higher market shares. for these reasons, a policy only aiming to increase static efficiency would hardly succeed in providing firms with incentives to invest in new technologies and products. 3.3. the relationship between competition and investment is non-linear in empirical literature ‘are we so sure that competition always favours innovation in developed countries?’ with this question, aghion and griffith (2005) underline the uncertain relationship between competition and economic growth. there is no clear evidence that greater competition necessarily fosters innovation. the relationship between competition and growth has remained an open question so far. to the question ‘can one turn to economists for clear and definite views on this debate?’, the authors answer ‘no’ and claim ‘it is fair to say that economists still have a limited and sometimes contradictory understanding of (competition’s) economic effects and, in particular, of the relationship between competition and growth’. empirical studies on cross-industry panels provide evidence of a non-linear relationship between competition and innovation. aghion et al. (2005) show that the schumpeterian effect dominates the escape from competition effect once competition exceeds its optimal level. they provide empirical evidence of an inverted u-shaped relationship between competition and innovation. instead of indefinitely increasing with competition, the rate of technological progress decreases after competition exceeds an optimal threshold. above this optimal threshold, competition hinders the reward from investment. competition reduces the incentives for firms to invest in innovation ‘by reducing the rents that can be captured by a follower who succeeds in catching up with its rival by innovating’. an increase in the level of competition can foster innovation when firms have a same level of technology, according to the escape from competition effect. however, when the level of technology is unevenly distributed across firms, an increase in competition intensity will discourage laggard firms from investing in order to catch up with the leaders because it reduces the post-innovation rents. following aghion et al. (2005), askenazy et al. (2008) evidence an inverted u-shape relationship between competition and r&d investment, finding that competition has weak european journal of government and economics 5(1) 19 effects on investment in the production of innovations when their costs are relatively high. in sectors where innovation becomes too costly, policies to increase competition will unlikely induce firms to invest. this suggests ‘modulating competitive policies according to the state of the industry’. 3.4. recent results in microeconomics invalidate the single monotonic increasing relationship between competition and innovation theoretical microeconomics provides insights that the relationship between competition and investment is ambiguous and depends largely on models and assumptions. for belleflamme and vergari (2010), ‘an intermediate form of competition may provide a higher incentive to innovate than the traditional polar cases (either monopoly or perfect competition)’. they show that optimal level of competition varies across sectors, in accordance with their specific characteristics, and find that ‘different industries are affected in qualitatively different ways by an increase in competition’. the profit incentive that motivates investment is maximised ‘depending on the characteristics of the industry of interest’. the profit incentives are maximised either ‘by a competitive firm (arrow’s claim), by a monopoly (schumpeter’s claim) or by an intermediate form of competition’. schmutzler (2013) provides a general theoretical framework to analyse the effects of increasing competition on investment, and concludes that these effects are ambiguous and vary largely according to the initial levels of efficiency (the marginal costs) and the initial level of competition. the theoretical framework accounts for all types of relationships. sacco and schmutzler (2011) show that the relationship between competition and r&d investment depends on the theoretical framework. however, ongoing microeconomic research tends to show that the non-monotonic relationship between competition and investment in technology might be related to its innovation intensity. houngbonon and jeanjean (2014) develop a theoretical model of the relationship between competition and investment in technology, where technological progress is captured by the impact of innovation on the marginal cost of production. they show that the inverted u-shape relationship between the level of competition intensity and the investment in technology is more likely to occur when the rate of technological progress is high. moreover, the higher the innovation potential, the higher the level of price-cost margin which maximises investment in innovation. 3.5. capital-intensive industries with high rate of innovation require a concentrated market structure (i.e. the existence of market power) to sustain their investment effort a recent literature has explored the link between market concentration, the level of competition and the rate of innovation. it suggests that too restrictive merger policies can hinder the rate of innovation. ivaldi & mccullough (2010) examine the welfare effects of mergers that have occurred in the us railroad freight market, where consolidations have increased both market power and the efficiency of service providers. during 1978-2006, mergers reduced the number of providers, increased market concentration and brought efficiency gains. productivity and efficiency gains have been largely transferred to consumers through lower prices and higher volumes. in the long run, ‘despite a dramatic degree of consolidation in the industry’, the initial decline in consumer welfare has been followed by a sharp continuous increase. mergers have not harmed social welfare and have brought large efficiency gains. for tilford (2008), in high-technology sectors like pharmaceuticals or ict, large market shares do not always mean less competition as few firms often bear the bulk of investment costs. market power is often temporary because firms are competing through investments in technology, which sustains a permanent process ciriani and lebourges ● the role of market power in economic growth 20 of transition between dominant technologies. tilford (2008) argues that if firms ‘are forced to share their intellectual property with competitors, or prevented from controlling the price at which their products and services are sold, there is a risk that they will innovate less’. for shapiro (2011), ‘a highly successful innovator may come to dominate a market, in which case observing a high level of ex post concentration would hardly imply a lack of ex ante competition, or a lack of innovation’. shapiro (2011) indicates that atomistic market structures should not be expected in industries where significant technological progress has occurred. on the contrary, highly concentrated market structures should normally occur in industries that have experienced significant innovations. gans (2015) outlines that a permissive antitrust policy which makes mergers more likely to occur can promote competition in industries which have high innovation rate and where the returns on investment in technology are short-lived. the prospect of merging with highly innovative firms encourages entry and increases competition in innovation. gans (2015) suggests that merger policies can have long-lasting effects on competition and the rate of innovation. indeed, under a stronger antitrust enforcement, more potential mergers are likely to be blocked, which causes firms that ‘might otherwise have stayed in longer to find a merger partner’ to exit the industry. the market is thus deprived of r&d investment and innovations that those firms would have produced. this effect is investigated empirically by igami and uetake (2015) who show that the strengthening of merger policy in a highly innovative industry would not necessarily result in higher competition, innovation and welfare. they study the hard-disk drive global industry between 1976 and 2014, and provide evidence that a tighter merger policy can decrease firms’ value and increase the number of exits by liquidation. when the opportunities to merge with an innovative firm decrease, firms expect that productivity improvements and market power will decrease and thus choose to exit. a high rate of merger blocking in highly innovative industries could hinder competition and innovation by discouraging entry and encouraging exit. according to the oecd (2012), ‘there can be sectors or industries where innovation is greater when firms have the possibility to acquire a monopolist position, at least temporarily, and thus charge a price which is above marginal cost during such a period’. the oecd (2012) explains that in sectors with high rates of innovation such as the pharmaceutical industry, competition occurs mainly through ‘races to innovate’ rather than through price setting, and ‘during the period of patent protection, firms are usually able to charge a supra-competitive price, which allows them to recoup the high costs incurred during the research and development stages’. dynamic efficiencies, which are related to ‘the ability of a firm and its incentives to introduce new products or processes of production (or to improve existing ones)’, will hardly arise from a market where static efficiency is achieved, because in such market, prices would equal marginal costs of production. the oecd (2007) concludes that for this reason, the ‘textbook perfect competition may be inconsistent with dynamic efficiencies’. the capacity to set prices above the marginal costs of production is necessary for innovation to occur. for this reason, investment in new technologies and new products would not arise in markets where static efficiency is achieved. overall, the results presented in this subsection support the view that investments in technologies and capital goods that incorporate technologies are likely to be discouraged when static competition is too high, and need either concentrated market structure or market power to occur. european journal of government and economics 5(1) 21 3.6. cross-country differences in growth rates stem from differences in the levels of investment in fixed assets which incorporate technologies from romer (1990), it appears that productivity gains stem from the diffusion of capital goods which have incorporated new technologies through fixed assets investment. cross-country differences in the rates of innovation diffusion through investment should then explain a large part of the cross-country differences in rates of productivity growth. comin and mestieri (2010) provide evidence that the diffusion of new technologies has a major role in explaining long-run international differences in per-capita income, and that these differences are mostly due to the cross-country differences in investment in the fixed assets that incorporate the new technologies. the authors study the contribution of technological change to the growth of aggregate productivity of 166 countries over the period 1820-2003. they posit that the adoption of new technologies by a country has two distinct components: the extensive margin of adoption, which captures the adoption lag, (how long it takes to adopt a new technology) and the intensive margin of adoption, which measures how many intermediary goods embodying the new technology are being used to produce the final goods. they estimate that the international differences in the adoption of technologies explain nearly 70 percent of the differences in countries’ per-capita income in the long run, and that the larger part of these differences can be attributed to differences in intensive margins of adoption, hence, in investment in intermediary goods that incorporate the new technologies. van ark et al. (2008) find that the contribution of ict capital to the growth of aggregate labour productivity has been relatively higher than the contribution of labour and of non-ict capital in either the eu (eu-10) or the us over 1980-2004. moreover, the growth of labour productivity has slowed down since 1995 in the eu while it has significantly accelerated in the us, due to both higher levels of ict investment and a faster diffusion of ict assets in the market services sector. the authors underline that the main driving force of the growth of labour productivity in the us in the mid-1990s related to both a rise in the productivity of industries producing ict equipment and a capital-deepening effect fuelled by investment in ict assets by the other sectors of the economy. van welsum et al. (2013) confirm that the weaker labour productivity growth of the eu compared to the us has persisted after the mid-2000s and can be largely explained by a weaker impact of investment in ict capital goods on labour and total factor productivity. the differences in aggregate labour productivity growth between the eu and the us are significantly related to differences in the levels of aggregate investment in ict capital goods. these results show that the level of aggregate investment in fixed assets which incorporate digital technologies has a major impact on the rate of productivity growth of a country, and a large share of cross-country differences in the rates of productivity growth can be attributed to cross-country differences in the levels of such investment. 4. the european and us doctrines in light of recent economic findings on competition and innovation evidence presented in the three preceding sections outlines that the us competition policy appear more consistent with the recent advances in theoretical and empirical economics than the european doctrine and practice. the european competition authorities still regard market power as a source of inefficiency and consider that the effects of market power have to be removed because they harm the competitive process. this can lead the european authorities to sanction strategic behaviours that are in reality pro-competitive, hence taking the risk of depriving the economy from potential additional efficiency gains. ciriani and lebourges ● the role of market power in economic growth 22 european competition policy rules out any possible trade-off between static efficiency and dynamic efficiencies. it strives to enhance static competition and to remove the effects of market power. the views of the european competition authorities on market power are still strongly influenced by the ‘ordo-liberal’ doctrine, which considers that the main purpose of economic policy is to promote entry and to limit the possibility of the dominant firm to exert its market power. however, literature examined in the third section shows that investment in innovation will not arise from a market where static efficiency is achieved. firms which cannot set prices above marginal costs will have neither the incentives nor the capabilities to invest in technology. a market where the level of static competition exceeds its optimal threshold will tend to decrease the rate of endogenous innovation. the european competition authorities, under the influence of the post-chicago synthesis, tend to prevent efficiencies due to vertical integration from occurring. indeed, they consider that vertical restraints are potentially anticompetitive and that vertical integration is more likely to allow the exercise of upstream market power than to generate efficiencies. by contrast, the u.s courts and competition agencies work to preserve market power and consider that upstream market power unlikely extends itself to the downstream markets. they consider that market power is a necessary incentive for firms to engage in risky investment and that supra-competitive profit is a normal return on investment. the us courts (since the adoption of the chicago school’s views) ceased to consider competitive market structure as an objective per se, and focused on the promotion of efficiency. the higher rates of productivity growth achieved by the us during the last twenty years can to a large extent be attributed to higher levels of investment in ict during this period. on the basis of the analysis of the u.s competition principles provided in the second section, it can be inferred that these levels of investment could have been favoured by the us doctrine on competition law. under us competition law, the dominant firm has more strategic autonomy and the search for market power is promoted rather than discouraged by the us courts. in particular, the higher level of investment of ict industries (which are network industries under an ex-ante regulation) in the us appears to be at least in part favoured by the doctrine of us competition agencies. indeed, the us competition authorities promote market power, do not ban concentrated market structure per se, and are sceptical of the incentives of integrated firms to leverage market power. conclusion the european commission’s economic policy, which regards competition policy as the main engine of economic growth, builds upon its doctrine on competition law. this doctrine is influenced by a structuralist approach of market power and a postchicagoan view on the assessment of vertical restraints. these approaches lead european authorities to favour non-concentrated market structures independently of economic efficiency. besides, european policy strives to increase the level of static competition (i.e. lowering prices down to the marginal costs) because it considers that dynamic efficiencies (investment in technology) arise from static efficiency. this policy favours static over dynamic competition, and ignores the trade-off between static and dynamic efficiencies. the european economic policy tends to limit the incentives and capacities to invest in new technologies and in fixed assets that incorporate those technologies. such a policy hampers dynamic efficiencies by impeding expected profit margins which are needed to sustain current and future investments. on the contrary, us doctrine, under the influence of the chicago school, has led us authorities to favour economic performance over maintaining non-concentrated market structure, to acknowledge market power has a legitimate european journal of government and economics 5(1) 23 incentive and investment reward, and to avoid enforcing robust antitrust in regulated industries where vertical integration generates efficiencies. this doctrine might have led to a competition policy more likely to sustain investment in the most productive and technologically advanced industries. indeed, the higher levels of us ict investment have contributed to higher growth, and the gap with the eu has begun to widen when ict were being heavily adopted through high investment efforts. in light of these results, the european commission’s economic doctrine needs a review in order to restore profitability, technological leadership and growth. references aghion, philippe and rachel griffith (2005) competition and growth. reconciling theory and evidence. cambridge, ma: mit press, 2005. aghion, philippe, nick bloom, richard blundell, rachel griffith and peter howitt (2005) ‘competition and innovation: an inverted-u relationship’, quarterly journal of economics 120(2): 701-728. askenazy, philippe, christophe cahn and delphine irac (2008) ‘competition, r&d, and the cost of innovation’, banque de france, notes d’etudes et de recherche, ner r n°197. barnett, thomas o. (2007) ‘issue statement on european microsoft decision’, us department of justice, antitrust division. press release 202-514-2007, september 17th. belleflamme, paul and cecilia vergari (2010) ‘incentives to innovate in oligopolies’, the manchester school 79(1): 6-28. bork robert, h. (1978) the antitrust paradox: a policy at war with itself, free press, new york. cette, gilbert (2015) ‘which role for icts as a productivity driver over the last years and the next future?’, digiworld economic journal 100 , 65-83. comin, diego and marti mestieri (2010) ‘the intensive margin of technology adoption’, harvard business school working paper 11-026. cooper, james c., luke m.froeb, daniel p. o’brien, and michael g. vitta (2005) ‘a comparative study of united states and european union approaches to vertical policy’ george mason law review (13)2: 289-308. fox, eleanor m. (1997) ‘us and eu competition law: a comparison’, in e.m graham & d.j. richardson d.j. (eds), global competition policy. washington, dc: peterson institute for international economics pp. 339-354. fox, eleanor m. (2006) ‘abuse of dominant position: the americanization of european competition law: the critical role of perspective: comment’, in h. ullrich (eds), the evolution of european competition law: whose regulation, which competition?, cheltenham: edward elgar publishing, pp. 233-237. gans, joshua, s. (2015) ‘mergers and disruptive innovation.’ hogan levels global media and communications quarterly autumn 2015. gorning, martin and alexander schniersch (2014) ‘weak investment in the eu: a long-term cross-sectoral phenomenon’, diw economic bulletin s.23-30. herbet, jean-baptiste (2001) ‘peut-on expliquer l’investissement à partir de ses déterminants traditionnels au cours de la décennie 90 ?’, économie et statistique n° 341-342, 85-106. houngbonon, georges v. and francois jeanjean, (2014) ‘is there a level of competition intensity that maximises investment in the mobile telecommunications industry?’. paper presented at the 25th european regional conference of the ciriani and lebourges ● the role of market power in economic growth 24 international telecommunications society (its), brussels, belgium, 22-25 june 2014. igami, mitsuru, and kosuke uetake, (2015) ‘mergers, innovation and entry-exit dynamics: the consolidation of the hard disk drive industry’ (1996-2015), mimeo, yale university. ivaldi, marc and gerard mccullough (2010) ‘welfare tradeoffs in us rail mergers’, working paper n°344, idei. jorgenson, dale w. (1963) ‘capital theory and investment behaviour’, american economic review 53(2):247-259. kovacic, william e. (2007) ‘the intellectual dna of modern us competition law for dominant firm conduct: the chicago/harvard double helix’, columbia business law review 2007 (1): 1-80. kovacic, william e. (2008a) ‘competition policy in the european union and the united states: convergence or divergence in the future treatment of dominant firms?’, competition law international october 2008, 8-18. kovacic, william e. (2008b) ‘competition policy in the european union and the united states: convergence or divergence?’ bates white fifth annual antitrust conference washington, d.c. june 2, 2008 kovacic, william e. and carl shapiro (2000) ‘antitrust policy: a century of economic and legal thinking’, journal of economic perspectives14(1): 43-60. landes, william m. and richard a. posner, (1981) ‘market power in antitrust cases’, harvard law review 94 (5): 937-996. lang, john t. (1979) ‘some aspects of abuse of dominant positions in european community antitrust law’, fordham international law journal 3(1): article 1. marty, frédéric (2007) ‘le contrôle des concentrations en europe et aux etatsunis. critères économiques et sécurité juridique’, revue de l’ofce 2007/1,100. marty, frédéric (2010) ‘politiques européennes de concurrence et économie sociale de marché’, document de travail de l’ofce no 2010-30. marty, frédéric (2012) ‘de la notion de responsabilité particulière de l’opérateur dominant dans la politique de concurrence européenne : quelles conséquences sur les libertés économiques ?’, in l. potvin-solis & h. ueda (eds), économie de marché, droits et libertés et valeurs communes en europe et en asie. editions de la chaire jean monnet université de lorraine, pp. 181-204. marty, frédéric and pillot julien (2009) ‘les politiques de concurrence européenne et américaine face aux remises de fidélité accordées par une entreprise dominante : l’affaire intel’, document de travail de l’ofce no 2009-26. marty, frédéric and pillot julien (2010) ‘divergences transatlantiques en matière d’application de la théorie des facilités essentielles aux actifs immatériels’, revue d’économie industrielle 2010/1 (no 129-130), 277-300. marty, frédéric and pillot julien (2012) ‘pratiques de boycott ou refus d'accès à une facilité essentielle’. de terminal railroad à l'irm d'arcachon’. revue lamy de la concurrence 2011, 17-25. newman, john m. (2014) ‘antitrust in zero-price markets: foundations’, university of pennsylvania law review 164, 149-206 oecd (2007) ‘dynamic efficiencies in merger analysis’, oecd policy roundtables. directorate for financial and enterprise affairs competition committee. european journal of government and economics 5(1) 25 oecd (2012) ‘the role of efficiency claims in antitrust proceedings’, oecd policy roundtables. directorate for financial and enterprise affairs competition committee. piraino, thomas a. (2007) ‘reconciling the harvard and chicago schools: a new antitrust approach for the 21st century’, indiana law journal 82(2): 345-409. porter, michael, e. (2002) ‘competition and antitrust: a productivity-based approach’, mimeo, harvard business school. posner richard a. (1976) antitrust law: an economic perspective, chicago: university of chicago press,. rey, patrick and jean tirole (2007) ‘a primer on foreclosure’, in mark armstrong and rob porter (eds), handbook of industrial organization 3, oxford:north holland, pp. 2145-2215. riordan, michael, r. (2005) ‘competitive effects of vertical integration’, discussion paper no.0506-11, department of economics, columbia university. riordan, michael h. and steven c. salop (1995), ‘evaluating vertical mergers: a post-chicago approach’, antitrust law journal 63, 513-568. romer, paul m. (1986) ‘increasing returns and long-run growth’, the journal of political economy 94(5):1002-1037. romer, paul m. (1990) ‘endogenous technological change’, the journal of political economy 98(5): 71-102. romer, paul m. (1994) ‘the origins of endogenous growth’, the journal of economic perspectives 8(1): 3-22. sacco, dario, & armin schmutzler, (2011) ‘is there a u-shaped relation between competition and investment?’ international journal of industrial organization 29(1): 65-73. shapiro, carl (2011) ‘competition and innovation: did arrow hit the bull's eye?’, in j. lerner & s. stern (eds), the rate and direction of inventive activity revisited. nber, pp. 361-404. solow, robert m. (1956) ‘‘a contribution to the theory of economic growth’, the quarterly journal of economics 70(1): 65-94. solow, robert m. (1957) ‘technical change and the aggregate production function’, the review of economics and statistics 39(3): 312-320. schmutzler, armin (2013) ‘competition and investment – a unified approach.’ international journal of industrial organization 31(5): 477–487. tilford, simon (2008) ‘is eu competition policy an obstacle to innovation and growth? centre for european reform essays. tobin, james (1969) ‘a general equilibrium approach to monetary theory’, journal of money, credit and banking 1(1): 15-29. van ark, bart, mary o' mahoney, and marcel p. timmer (2008) ‘the productivity gap between europe and the united states: trends and causes’, journal of economic perspectives 22(1): 25-44. van ark, bart and abdul erumban (2015) ‘productivity brief 2015. global productivity stuck in the slow lane with no signs of recovery in sight’, the conference board productivity brief 2015, the conference board. van welsum, desiré, overmeer, willem and van ark, bart (2013) ‘unlocking the ict growth potential in europe: enabling people and businesses’, the conference board, report for the european commission gd communications networks, content & technology. ciriani and lebourges ● the role of market power in economic growth 26 documents from the european commission and european parliament european commission (2004a). a pro-active competition policy for a competitive europe. brussels, 20.4.2004, com(2004) 293 http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=com:2004:0293:fin:en:pdf european commission (2004b). fostering structural change: an industrial policy for an enlarged europe. brussels, 20.4.2004, com(2004) 274 final. http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=com:2004:0274:fin:en:pdf european commission (2004). guidelines on the assessment of horizontal mergers under the council regulation on the control of concentrations between undertakings. official journal of the european union. (2004/c 31/03). http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=oj:c:2004:031:0005:0018:en:pdf european commission (2008). press release speech/08/658. ‘achieving selfsustaining competition in telecommunications’. neelie kroes, 9th annual ecta regulatory conference. brussels, 27 november 2008. http://europa.eu/rapid/press-release_speech-08-658_en.htm european commission (2008). guidelines on the assessment of non-horizontal mergers under the council regulation on the control of concentrations between undertakings. official journal of the european union (2008/c 265/07). http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=oj:c:2008:265:0006:0025:en:pdf european commission (2009a). economic crisis in europe: causes, consequences and responses. directorate-general for economic and financial affairs 7/2009. http://ec.europa.eu/economy_finance/publications/publication15887_en.pdf european commission (2009b). guidance on the commission’s enforcement priorities in applying article 82 of the ec treaty to abusive exclusionary conduct by dominant undertakings. official journal of the european union (2009/c 45/02). http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=oj:c:2009:045:0007:0020:en:pdf european commission (2010a). ‘europe 2020. a european strategy for smart, sustainable and inclusive growth’. http://ec.europa.eu/eu2020/pdf/complet%20en%20barroso%20%20%20007 %20-%20europe%202020%20-%20en%20version.pdf european commission (2010b). guidelines on vertical restraints. http://ec.europa.eu/competition/antitrust/legislation/guidelines_vertical_en.pdf european commission (2011a). annex to the annual growth survey 2012. http://ec.europa.eu/archives/commission_20102014/president/news/documents/pdf/annex_2_en.pdf european commission (2011b). an action plan to improve access to finance for smes. brussels, 7.12.2011. european journal of government and economics 5(1) 27 http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=com:2011:0870:fin:en:pdf european commission (2012). press release speech/12/172. ‘competition policy for innovation and growth: keeping markets open and efficient’. joaquin almunia. brussels, 8 march 2012. http://europa.eu/rapid/press-release_speech12-172_en.htm?locale=en european commission (2012a). report on competition policy 2011. brussels, 30.5.2012. http://ec.europa.eu/competition/publications/annual_report/2011/part1_en.p df european commission (2012b). a stronger european industry for growth and economic recovery. industrial policy communication update brussels, 10.10.2012. http://eurlex.europa.eu/lexuriserv/lexuriserv.do?uri=com:2012:0582:fin:en:pdf european commission (2012c). alert mechanism report 2013. http://ec.europa.eu/economy_finance/economic_governance/documents/alert_mec hanism_report_2012-11_en.pdf european commission (2013a) ‘recent changes in europe’s competitive landscape and medium-term perspectives: how the sources of demand and supply are shaping up’. fellowship initiative the future of emu. economic papers 485 april 2013. http://ec.europa.eu/economy_finance/publications/economic_paper/2013/pdf/ecp4 85_en.pdf european commission (2013b). macroeconomic imbalances france 2013. http://ec.europa.eu/economy_finance/publications/occasional_paper/2013/pdf/ocp1 36_en.pdf european commission (2013c). in depth review for france. brussels, 10.4.2013. http://ec.europa.eu/europe2020/pdf/nd/idr2013_france_en.pdf european commission (2014a). press release. progress in industrial competitiveness per eu country. brussels, 11 september 2014. http://europa.eu/rapid/press-release_memo-14-526_en.htm european commission (2014b). alert mechanism report 2015. brussels, 28.11.2014. http://ec.europa.eu/europe2020/pdf/2015/amr2015_en.pdf european commission (2014c). drivers and implications of the weakness of investment in the eu. directorate-general for economic and financial affairs. economic forecasts autumn 2014. http://ec.europa.eu/economy_finance/eu/forecasts/2014_autumn/box1_en.pdf european commission (2014d). macroeconomic imbalances france 2014. brussels, 5.3.2014. http://eur-lex.europa.eu/legalcontent/en/txt/pdf/?uri=celex:52014sc0081&from=en european commission (2015a). report on competition policy 2014. brussels, 4.6.2015. http://ec.europa.eu/competition/publications/annual_report/2014/part1_en.pdf european commission (2015b). annual growth survey 2015. strengthening the recovery and fostering convergence. brussels, 26.11.2015. ciriani and lebourges ● the role of market power in economic growth 28 http://ec.europa.eu/europe2020/pdf/2016/ags2016_annual_growth_survey.pdf european commission (2015c). ‘competition and innovation’. key note speech of johannes laitenberger (director general dg competition), annual charles river associates conference. brussels, 9th december 2015. http://ec.europa.eu/competition/speeches/text/sp2015_04_en.pdf european commission (2015d). ‘ex post evaluation of competition policy enforcement. what can we learn from the literature and from the experience of competition authorities?’ european commission competition policy brief 2015-04, june 2015. http://ec.europa.eu/competition/publications/cpb/2015/004_en.pdf european parliament (2013). the contribution of competition policy to growth and the eu 2020 strategy. http://www.europarl.europa.eu/regdata/etudes/etudes/join/2013/492479/ipolecon_et(2013)492479_en.pdf european parliament (2015). eu industrial policy: assessment of recent developments and recommendations for future policies. directorate-general for internal policies. http://www.europarl.europa.eu/regdata/etudes/stud/2015/536320/ipol_stu(20 15)536320_en.pdf. microsoft word ejge_03_02_015.doc european journal of government and economics volume 3, number 2 (december 2014) issn: 2254-7088 100 editorial statement: lessons from goodhart’s law for the management of the journal diego varela, universidade da coruña, spain giacomo benedetto, royal holloway, university of london, united kingdom jose manuel sanchez-santos, universidade da coruña, spain abstract in this editorial statement we summarise some of the discussions we have had in the last months regarding the risks associated with the use of indicators for the measurement of research outputs, and how these risks should affect the management of the european journal of government and economics. in particular, we focus on the consequences of the so-called goodhart’s law, which states that when a measure becomes a target, it ceases to be a good measure. we also explain the latest developments in the journal in the light of our previous editorial statements, and present our strategy for the upcoming years. jel classification a12; a30; p27. keywords academic publishing; academic journal; goodhart’s law; indicators; research assessment. varela, benedetto and sanchez-santos ● lessons from goodhart’s law 101 introduction in the past year, our journal has continued its expansion along the lines set out in our last editorial statement of december 2013 (varela et al., 2013). since then, we have expanded our journal’s base by including new authors from diverse disciplines and geographical locations. this year we have published papers from authors affiliated to institutions in sweden, italy, turkey, germany, new caledonia, and spain. we have also been accepted by one new index in this period, namely latindex, which is probably the most prominent index for journals based in spain, portugal and latin america. some spanish universities place latindex on a par with scopus in their bureaucratic assessment methods for open competitions. joining latindex has also meant indexing by the isoc database of the spanish higher council of scientific research (cesic), which is the largest public research institution in spain and the third largest in europe. we do not have news from scopus yet but we expect to have some advances in the coming year. another index we expect to join soon is econlit, which is an abstracting database service published by the american economic association and dating back to 1969. the service focuses on literature in the field of economics and uses jel classification codes for classifying papers by subject. we have been using this classification system since the start of the journal, and now that we are about to enter our fourth year it is time to apply for inclusion in this database. in spite of the relative youth of the journal, many of our papers have started to get cited in other journals and working papers. some papers are in fact harvesting a great number of citations, such as a paper on corruption and growth with evidence from the italian regions (fiorino et al., 2012) or one about cyclical synchronization in the emu along the financial crisis (cancelo, 2012). but there are many others with a great potential impact. indexing by certain prestigeous databases is a form of recognition for a journal, and it can even affect the number of citations received by the papers it publishes (varela, 2012). however, both joining a given index and being cited should not be seen as ends in themselves but as indicators of quality in academic performance. we shall explain why. the risks of goodhart’s law the problem arises when an indicator is used by policy makers to try to influence the performance of their institutions or academics. quite often, university funding authorities try to base their decisions on some objective measure of research productivity, and it is not uncommon to see that this productivity is measured by some rough indicator, such as the number of publications in journals belonging to a given index or with a certain impact factor. university authorities, in their selection and promotion decisions, also make use of such objective indicators. if publishing in a given journal or having a lot of citations will affect a department’s funding or academic promotion opportunities, indicators become ends in themselves, therefore creating a set of incentives to “trick” the system, for example by concentrating publications in the easiest journal of a given index and boosting the number of self-citations. this may eventually end up breaking the original relationship between the indicator and the quality it intended to measure, an instance of goodhart’s law. although originally applied to monetary policy (see goodhart, 1975), this is now a well established regularity in many fields of economic policy that can be summarised by saying that when a measure becomes a target, it ceases to be a good measure. as a consequence, policy makers are forced to change their indicators periodically. european journal of government and economics 3(2) 102 the problem of goodhart’s law for research assessment is especially serious in centralised bureaucratic systems that, on the one hand, rely heavily on such objective measures of performance and, on the other hand, are very slow to react to attempts to trick the system. it is much easier for an autonomous selection panel to detect how a candidate has concentrated his/her publications in lesser quality journals belonging to a given index than for the minister of education that approves performance indicators every number of years. but even in centralised bureaucratic systems indicators are updated from time to time, and it is not uncommon for some of those who play by their rules to feel that such updates are unfair change in the rules in the middle of the game. but that is not necessarily the case. others might argue that those who complain were just too slow to trick the system. lessons for the management of the journal the main implication of goodhart’s law for the management of the journal is that it is risky to rely on a set of politically-sanctioned indicators such as being indexed by a given service or having a lot of citations. the key to reduce this risk is to treat such indicators as imperfect measures of quality and not as ends in themselves. does this mean that indicators are not important? not at all. it just means that efforts should not be blindly directed to joining a given index or boosting a journal’s impact factor, but to increasing the underlying quality of the journal, which will bring about indexing and citations as a side effect. focusing on underlying quality implies an effort to induce what the current journal quality indicators are really trying to measure without trying to influence them directly. this requires monitoring indicators regularly but, since they vary across institutions and countries, and throughout time, we should look at a wide array of them instead of focusing on the ones prevailing in a particular institutional and temporal setting. we believe that such a policy is much safer in the long run because we cannot know for sure what the prevailing indicators will be in the future, but at least they should be positively correlated with the underlying quality they aim to measure. if we focus on underlying quality based on a wide array of different indicators (not just the prevailing ones in a particular institutional setting), we shall be better insured against changes in the ones adopted by policy makers. in order to continue improving the journal’s quality, it is important to continue benefiting from the cooperation of the members of the editorial board, the authors and the reviewers, who represent our strongest asset. but if we are to continue growing it is also important to be able to strengthen the journal’s infrastructure. this will require improving its institutional support and funding so that will make it possible to continue improving the journal’s technological and human resources. a way to achieve this would be by partnering with some commercial publisher, as we have suggested before, but that is not the only option. an interesting venue to explore would involve pooling resources from different institutions through some sort of formal cooperation agreement that would allow to share the burdens (mostly in terms of human resources) and the benefits of the journal (mostly in terms of visibility). this will be one of the challenges for the upcoming years. references cancelo, jose r. (2012) 'cyclical synchronization in the emu along the financial crisis: an interpretation of the conflicting signals', european journal of government and economics 1(1): 86-100. fiorino, nadia, emma galli and ilaria petrarca (2012) 'corruption and growth: evidence from the italian regions', european journal of government and economics 1(2): 126-44. varela, benedetto and sanchez-santos ● lessons from goodhart’s law 103 goodhart, charles a.e. (1975) ‘problems of monetary management: the u.k. experience’, papers in monetary economics (reserve bank of australia). varela, diego (2013) 'the contribution of isi indexing to a paper's citations: results of a natural experiment', european political science 12(2): 245-53. varela, diego, giacomo benedetto and jose m. sanchez-santos (2013) 'editorial statement: the first two years of ejge', european journal of government and economics 2(2): 95-9. microsoft word ejge_04_2015-010.docx european journal of government and economics volume 4, number 2 (december 2015) issn: 2254-7088 104 a first formal approach to animal spirits beyond uncertainty gerasimos t. soldatos, american university of athens, greece erotokritos varelas, university of macedonia, greece abstract standard macroeconomics treats animal spirits as a source of uncertainty disturbing otherwise rational expectations. but, keynesian animal spirits ensue from suboptimal emotional responses to socioeconomic status change beyond matters of uncertainty. this paper identifies such spirits with the disturbance from the optimal decision-making implied by an emotional well-being utility function. the introduction of a policy-maker, holding its own view of private welfare in a society of emotional individuals, generates by itself, i.e. in the absence of animal spirits, uniform business fluctuations. this is the result of the income redistribution needed to reconcile the policy-maker’s with the emotional individual’s view of private welfare. consequently, if animal-spirits induced fluctuations are already present when a policy-maker is introduced in the economy, the aim of policy intervention should be the design of that income redistribution that would not aggravate the business cycle but that would end up in uniform only cycles, with the aid perhaps of discretionary interest rate policy. nevertheless, if animal spirits do not exist when the policy-maker enters the system, the income-redistribution induced cycles may incite such spirits by themselves in which case the cycles will not be of the uniform type. all comes down to “income and emotion”, to an ageless and ecumenical fact of life, complicated purposefully or not by authority. jel classification d63; e32; e43; h11; z13. keywords animal spirits; emotions; socio-economic status; individual vs. social perception of private welfare; business cycles. acknowledgement we are grateful to two anonymous reviewers for useful comments and suggestions. european journal of government and economics 4(2) 105 1. introduction ‘most, probably, of our decisions to do something positive, the full consequences of which will be drawn out over many days to come, can only be taken as the result of animal spirits – a spontaneous urge to action rather than inaction, and not as the outcome of a weighted average of quantitative benefits multiplied by quantitative probabilities’: this is precisely what keynes (1936, 162) had said about animal spirits in his general theory. since then macroeconomics tried successfully to distance the field from the notion of animal spirits, because they are not susceptible to data and reason (or control by government); uncertainty and risk came to replace it – rightly from one point of view as we shall soon see – rationalizing concepts such as business confidence accordingly (roskam 2014). indeed, many, like frantz (2005) and shiller (2015) see in animal spirits sheer irrationality. but, the truth of the matter is that keynes himself did not identify animal spirits with uncertainty contrary to what mainstream macroeconomics maintains. akerlof and shiller’s (2009) widely appreciated book on the subject points to the possibility of ‘predictable irrationality’. and, others, like pech and milan (2009) and schwartz (2010) try to appreciate animal spirits in the light of cognitive psychology towards some behavioural macroeconomics. it is this last approach to the matter, which is promising, simply because irrationality is not predictable and much more rationalizable by definition. the matter is really one of rationalizing emotions. according to pham (2007) and cohen et al. (2007), for example, emotions arise from a cognitive appraisal of the emotional object or situation in terms of its meaning for one’s well-being: this is the case with the so-called “integral emotional responses”, otherwise emotion states are defined to be “incidental” ones. and, according to scherer (2011), for instance, three conceptions of rationality need to be distinguished in discussing the relation between emotion and rationality. the first is rational in the sense of intellectual/inferential, emphasizing reasoning, consistency, and logic à la kahneman (1994). the second is rational in the sense of functional/purposeful, pertaining to the optimality of the means-end relationship à la aristotle’s notion of practical reason (nicomachean ethics, ross 1908) or weber’s (1964) zweckrationalität. and the third type of rationality is rational in the sense of reasonable/consensual, associated with behaviours and actions that are “rational” not because they are logically consistent or serve one’s self-interest but because they fulfill broader societal goals, comply with superior moral values, or back greater evolutionary purposes even against one’s material self-interest, much à la weber’s (1964) wertrationalität, (clore 2005). these definitions of emotion types and rationality types in connection with emotion are standard in the relevant literature. could it be that the strict separation between ratio and passio, maintained by mainstream economics, derives methodologically from plato’s notion of a tripartite soul that categorically distinguishes between cognition, emotion and motivational urges (republic iv, jowett 1892)? the answer is negative, because plato, the inventor of the notion of ‘rationality’ (moss 2008), also points out that virtue is a matter of ‘harmonizing the three [parts of soul], just like the three notes in a musical scale, lowest and highest and middle’ (republic iv, 443d, jowett 1892). so, not only realism but also methodology dictates the incorporation of emotion to economic theory towards the investigation of the influence of animal spirits on microeconomic behaviour and macroeconomic performance. it is straightforward logically to associate incidental emotion states with uncertainty and risk, which need not bother us here because it has already received much attention by the literature. therefore, a first approximation to the notion of animal spirits is what the present paper tries to offer by confining analytically attention to integral emotional responses in connection with functional rationality. animal spirits may be emanating from other rationality types and from incidental emotion states with soldatos and varelas ● a first formal approach to animal spirits 106 implications far more complex than “simple” uncertainty. but, outside the scope of uncertainty, the emotion-rationality nexus studied here is the simplest one analytically, because a utility function may be employed and because there does exist in the literature parada-daza and parada-contzen’s (2013) utility function incorporating emotional well-being. the next section offers a formal definition of animal spirits based on a version of this utility function. it is followed by a section clarifying this definition from a macroeconomic point of view in a kaldor-type environment. the paper concludes with still another section pointing to the complexity of a holistic approach to animal spirits, encompassing not only integral emotional responses but incidental emotion states too, the latter in the form of uncertainty. such an approach becomes much more effortful when one attempts to incorporate in the analysis consensual rationality and inferential rationality; the former in the form, for instance, of staveren’s (2007) ethics in economics, and the latter by considering explicitly framing effects and the like. 2. the formal definition of animal spirits to model the genuinely keynesian notion of animal spirits, the one which is dissociated from uncertainty, their presence is introduced into an otherwise certain environment. following parada-daza and parada-contzen (2013), the version of individual emotional utility function considered here is: , 1 2 ∙ 1 where , and , are constants, with 1, showing the proportion utility derives from pure rationality and emotion, respectively, under the particular historical period being experienced by the individual in its lifetime. ∈ 0,1 captures emotion regulation by constantly adjusting the emotional response either itself qualitatively (fine-tuning) or its intensity (volume adjustment) given the intent of the emotion. alternatively, in terms of kahneman and deaton’s (2010, 16489) distinction between emotional well-being (‘the frequency and intensity of experiences of joy, stress, sadness, anger, and affection that make one's life pleasant or unpleasant’) and life evaluation (‘thoughts that people have about their life when they think about it’), and might be connected with life evaluation and emotional well-being, respectively. and, ∈ 0,1 is the share of individual income, , to total steady-state income, , and is a choice variable too, from the point of view of how much one wants to work to increase one’s share to total income: ⁄ ⁄ , where is labor, is the wage rate, is some positive coefficient, and the constant 1. the focus is on rather than on , because it hinges upon the broader matter of income distribution, which is of concern herein. emotions are exogenous to the economy and give rise to instability once is not the one matching : 0 ⇒ 1 4 2 that is, there is an optimal feedback loop between the dynamic unfolding of emotion and the evolution of socio-economic status, which once ruptured, animal spirits and instability will crop out. put differently, it is one thing to react emotionally when it is required (emotion elicitation), another to react with the appropriate emotion (emotion differentiation). one reason, for instance, why 2 may be violated is cognitive biases like those of prospect theory. animal spirits and the subsequent instability are the outcome of inappropriate emotional response; inappropriate from the viewpoint of disturbing 2 . figure 1a, with on the horizontal axis and , on the vertical, illustrates this utility function under 0.6, and 0.4. the blue line assumes that 0.2 while the red line is based on stronger emotions described by a value of 0.6. the destabilizing influence of is clear once 2 is violated; much more so when emotion is strong european journal of government and economics 4(2) 107 0.6 . 0.2 and national income is higher as is the case with the left-hand of the figure where 2 whereas in the right-hand part, 1. now, animal spirits are defined to be the outcome of the violation of 2 , with subsequent destabilizing influence as depicted by figure 1a. increased income share is not always consonant with increased utility unless emotion adjusts as indicated by 2 , leaving the area below the blue or red line, leaving that is, welfare unchanged. ideally, lines such as these, lines, should be parallel to the horizontal axis, but may lag behind , welfare may not be stable over the lifetime, and what really matters is that instability does not influence total welfare once the individual passes away: the areas above and below the parallel line cancel each other out. yet, comparing areas below blue and red lines in figure 1a, emotions cause a big loss in welfare in the right-hand part, which is not the case in the left-hand part with the lower . so, animal spirits do take their toll on welfare, and welfare instability may be attributed to the effort of the individual to deal with the welfare changes coming out of the violation of 2 . much more so when one considers emotion to be the main source of utility as the comparison between the black lines 0.4, 0.6 and the violet lines 0.6, 0.4 in figure 1b illustrates under 1. this awareness of the individual regarding the role of emotion, prompting action and reaction on its part, is what prevents a characterization of it as irrational. there does exist rationality, which although it may not be the emotion-free, 0 ⇒ sin 2 ∙ 0, is nevertheless some sort of emotional rationality. as such, it tries to solve the optimization problem: max , 1 2 ∙ figure 1a blue: , 0.2 0.6 1 0.4 2 0.2 red: , 0.6 0.6 1 0.4 2 0.6 blue: , 0.2 0.6 1 2 0.4 2 0.2 4 red: , 0.6 0.6 1 2 0.4 2 0.6 4 soldatos and varelas ● a first formal approach to animal spirits 108 figure 1b velvet: , 0.2 0.6 1 0.4 2 0.2 black: , 0.2 0.4 1 0.6 2 0.2 velvet: , 0.6 0.6 1 0.4 2 0.6 black: , 0.6 0.4 1 0.6 2 0.6 where: is the income constraint. the emotional rationalization first-order conditions include 2 and: 2 ∙ 2 ∙ 1 0 3 or, in short: 3′ is the marginal utility of individual income. the marginal private welfare from individual income given by the first two terms (and divided by ), has to be equated with this marginal utility. that is, emotion may be inducing private welfare fluctuations over the life span, but once it is rationalized in the sense of endeavouring to minimize unpleasant outcomes out of 2 ’s disturbance, 3′ will come to characterize the individual over the lifetime. according to this condition, a person who is emotionally rational too, thinks that if for some reason 2 were violated, it would be optimal on its part if the temporary welfare ups and downs had counterbalanced each other in the end of its life, having been locked up by . such a result is in line with the hypotheses regarding the self-fulfilling socioeconomic-status beliefs in the long-run (piketty 1998). nevertheless, we have seen that once 2 is disturbed, 3′ is unlikely to be satisfied as well: animal spirits have consequences, which the policy-maker has to address. suppose that the incumbent has a world view reflected through the following bergson-samuelson social welfare function: ∑ ∑ 1 2 ∙ where the ’s are weights attached to individual ’s so that 1, and the summation operator, ∑, applies to the entire population. figure 1 depicts the way the policy-maker views individual utility when 1 . figures 2 and 3 show this under 0.5 and 1.5, respectively, continuing for comparison but also simplicity purposes to assume that 1. these diagrams verify the comments made on the basis of figure 1, and suggest in addition the three of them that the increase in amplifies utility volatility. (and ) reflect policy-maker’s perception of society, which clearly has a powerful effect on the course of the economy. in this paper, it is the ineffective policy-maker who allows exogenous emotion to become european journal of government and economics 4(2) 109 endogenous to the economy, provoking ultra-emotional spirits that reinforce instability. the reaction of citizenry in response to erroneous policy-making is labelled ultra-emotional because adds to pre-existent exogenous emotion. figure 2 blue: , 0.2 0.6 1 0.4 2 0.2 red: , 0.6 0.6 1 0.4 2 0.6 blue: , 0.2 0.6 1 2 0.4 2 0.2 4 red: , 0.6 0.6 1 2 0.4 2 0.6 4 figure 3 blue: , 0.2 0.6 1 0.4 2 0.2 red: , 0.6 0.6 1 0.4 2 0.6 blue: , 0.2 0.6 1 2 0.4 2 0.2 4 red: , 0.6 0.6 1 2 0.4 2 0.6 4 now, the policy-maker’s optimization problem is: max , ∑ 1 2 ∙ ∑ where: ∑ is the relevant income constraint. that is, authorities are interested not only in income distribution, i.e. in manipulating so that the average, so to speak, is matched, but also in keeping at its steady state. the first-order conditions are: 2 ∙ 2 ∙ 1 1 2 ∙ 0 or: soldatos and varelas ● a first formal approach to animal spirits 110 0 4 and ∑ 1 ∑ 0 ∑ 0 ⇒ 0 5 is the marginal utility of national income to the policy-maker. being equal to zero to satisfy 5 , it would in turn imply from 4 that ⁄ 0 and hence, from 3′ that 0, which indeed might be the case at a steady state of universal bliss. but, once such an eventuality is unrealistic, 5 explains that this can be so because 1 ∑ 0 ⇒ 1 ∑ ⁄ ⁄ 0 ⇒ ∑ ⁄ 0 6 that is because more or less than is expected by the policy-maker to be distributed among the individuals, and expectations are frustrated. so, although individuals would be content if 2 and 3′ were satisfied, the policy-maker will be pleased only if what it considers to be universal bliss ( 0 ⁄ ) is attained. there is clearly a public-private sector discrepancy in the valuation of individual welfare even if there were no animal spirits: from 3 and 4 , ; the two theta are not the same. and, the residual income implied by 6 goes, as we shall see shortly in the next section, as rate of return to physical capital. steady-state general equilibrium requires the conversion of 6 into equality: ∑ ⁄ 6 in which case, ⁄ ⁄ ⁄ ∑ ⁄⁄ . according to the policy-maker this is the condition describing prudent policy-making when 2 and 3′ are violated, or the same, in the presence of animal spirits. improper is the policy-making when the marginal private welfare from individual income differs from the marginal social welfare of the same income as seen by the policy-maker. but, we just saw that the presence and only of a policy-maker causes such a discrepancy even if all in the economy runs smoothly. this discrepancy, engendering loss of confidence and trust towards the policy-maker, is the source of emotions, of ultra-emotional spirits, destabilizing the economy (a) by itself in the form of uniform cycles as we shall see below, if 2 and 3 are satisfied, (b) even further than exogenous emotion does if 2 and 3 are violated. if 1, condition 6′ is necessary but not sufficient to prevent the emergence of ultra-emotional spirits as may be seen through the following, for example, second-order derivative: 4 ∙ 2 ∙ 1 1 0 this derivative is always negative if 1 in which case, condition 6′ is both necessary and sufficient to ensure absence of animal spirits. but, if 1, the second derivative will be negative only if its positive term is absolutely smaller than its negative one. in this case, additional conditions beyond the satisfaction of 6′ are needed to have a negative cross-partial and prevent the appearance of animal spirits when 2 is breached. 3. the macroeconomics perspective of animal spirits additional insight into the meaning of 6 and 6′ may be gained by considering it from a macroeconomic point of view à la dana and malgrange (1993) as follows. emotion frequency, , changes with time according to some function, and any disturbance of 6′ and change thereby in implies disequilibrium in the goods market and discrepancy between investment, european journal of government and economics 4(2) 111 7 and saving, 8 altering according to 9 or inserting 7 and 8 in 9 , 1 10 where: is the accelerator, is the propensity to save, is the speed of adjustment of supply to demand, and the derivative with respect to , is used in the place of the time derivative entering customarily dynamic analysis. in view of, ∑ ⁄ 10 may be rewritten as follows, ∑ ⁄ 10′ which in turn implies that ∑ ⁄ 1 11 that is, in view of 6′ , at steady-state general equilibrium, 0, which prompts another perspective of what prudent policy-making means as follows: one might have sensed from the diagrams above that the area below the blue lines tends to be greater than that under the red lines, suggesting a higher welfare level for the less emotional individuals given the presence of animal spirits. proper policymaking according to the policy-maker means income redistribution aiming at compensating for welfare differences owing to differences in animal-spirits related emotion, ceteris paribus; policy-making should be neutralizing “exogenous”, nonpolicy induced animal spirits if it does not want to become part of the problem. this is very important, because it tends to take care of the destabilizing influence caused by the introduction and only of a policy-maker in the economy, too. animal spirits come up and have critical consequences for the stability of output once policy does not accomplish this task. one can see this through what stability means under 0, by noting that outside the stead-state, conditions 3 and 4 imply that ⇒ 1 2 ⇒ 2 1 1 12 which means that 0 along a uniform cycle unless 0, in which case: 1 1 ⇒ exp 1 1 stability in the form of uniform cycles is what macroeconomic stability and steadystate mean for the policy-maker when society consists of emotional individuals. if there are no animal spirits, such cycles are the result of the income redistribution soldatos and varelas ● a first formal approach to animal spirits 112 pursued by the policy-maker to reconcile the social (its own) view of private welfare with the individual’s view of its own welfare. and, uniform oscillations are consequently what the policy-maker would have in mind as the goal of its intervention to subdue animal spirits when they do exist. if they do not exist, the income-redistribution induced cycles may trigger animal spirits by themselves in which case they will not be of the uniform type. the course of physical capital, , would be shaped by these developments if which is a kaldor-like assumption, and which in view of 7 gives: the path of is a scaled down version of that of . so, once condition 2 is violated for some reason, and is not plausible to expect to be restored through 6′ , what becomes important as second best policy-making is to ensure that 1 0 i.e. that the cycles are prevented from obtaining a downward trend, as prescribed by the relationship: ⋚ 0 ⋚ income redistribution is one policy means towards that direction, but the rate of interest, , can become a second one, because, letting 1, 2 2 2 and, circumventing the matter of cycles, one can postulate that 1 ⇒ 1 ⁄ so that 1 2 1 1 ⁄ as figure 4 shows in the space for 0.1, 0.01, and 0.001. hence, 2 1 1 2 obtaining in turn from 12 : 1 1 1 2 1 13 the cross-partial derivative ⁄ in the appendix is negative, capturing the important role interest rate policy can play in influencing animal spirits. european journal of government and economics 4(2) 113 figure 4 blue: 0.1 1 ⁄ red: 0.01 1 ⁄ black: 0.001 1 ⁄ the intuitive reason is pretty clear: in the presence of animal spirits, capital markets are inefficient, the loanable funds doctrine simply does not hold, and there can be no natural rate of interest (pilkington 2014). yet, note that the relevant policy prescription is not exactly moore’s (1988) ‘horizontalism’ that the central bank set a certain rate of interest as a policy target and then let the quantity of money float. this would be the case if the matter of cycles was assumed away as is the case behind figure 4; an assumption made for algebraic tractableness. but, animal spirits are synonymous to fluctuations, and interest rate policy should be discretionary, taking on a flavour of “verticalist” interest rate float. as arestis (2011) points out, a taylor rule presumes the existence of a natural rate of interest à la wicksell (2007 [1898]), which is not the case here unless a fluctuating natural rate is somehow contemplated… what is for sure is that: ‘…the short-term output responses are more sensitive to animal spirits in the fixed interest rate regime than in the variable one. thus the interest rate response under the variable interest rate regime tend to reduce the impact of animal spirits on the transmission mechanism, thereby reducing the volatility in this transmission’ (de grauwe 2012, 53). moreover, the animal spirits in this paper are first and foremost exogenous, unrelated to economic uncertainty, which only the ineffective policy-maker can trigger and add to the exogenous emotion. the viewpoint about a liquidity preference as a buffer against economic uncertainties should be appreciated accordingly. 4. concluding remarks to sum up: animal spirits ensue from suboptimal lifetime integral emotional response to socioeconomic status change under functional rationality. the introduction of a policy-maker, holding its own view of private welfare in a society of emotional individuals, generates by itself, i.e. in the absence of animal spirits, uniform business fluctuations. this is the result of the income redistribution needed to reconcile the policy-maker’s with the emotional individual’s view of private welfare. consequently, stability restoration in the presence of animal spirits involves policy-making to restore the uniform character of cycles through income redistribution and perhaps discretionary interest rate policy. nevertheless, if animal spirits do not exist, the income-redistribution induced cycles may incite such spirits by themselves in which case the cycles will not be of the uniform type. the situation becomes much more perplexed when the uncertainty originating in incidental emotional states is added to this picture. incidental states may be taking, soldatos and varelas ● a first formal approach to animal spirits 114 for example, the form of a random disturbance term, , with mean, ,̅ and variance, , in an environment in which is uncertain too, with mean, , and variance, . under such circumstances, one has to choose and so as to keep as close to 0 on average as possible: min ≡ 1 1 2 0 1 1 2 1 1 2 2 1 1 2 14 where is the expectations operator and is the correlation coefficient between and . it is clear that introducing the simplest holistic approach to animal spirits in economics complicates the discipline considerably, and this may be one reason of confining attention only to the implications from uncertainty. much more so when such an approach should also be encompassing not so much the explicit treatment inferential rationality’s cognitive biases that invalidate (2) as the matter of consensual rationality à la say staveren’s (2007) ethics. the simplest way to incorporate ethics in 1 is to add a term indicating how the rest of society affects the individual, with the positive/negative sign if the influence is positive or negative: , 1 2 1 1 1 2 1 15 the bracketed term after the “ ” sign reflects society’s utility function, which is independent of how the individual feels: 1; society does not care about a given individual. the first-order conditions are now: 2 2 1 1 2 1 0 and 2 2 1 ∓ 1 1 1 ∓ 2 1 2 1 0 it is clear that the coordination between integral emotional response and socioeconomic status change becomes much more difficult than under 2 . the same difficulty characterizes individual effort to even out the temporary welfare ups and downs over the lifetime vis-à-vis what 3′ dictates. and, of course, even more likely becomes the destabilizing “capacity” of the policy-maker. now, this scenario behind 15 has to be merged with the one behind 14 to have a truly holistic approach to animal spirits. what such an intricate venture will produce in practice is certainly increased alertness in policy-making, improved finetuning. of course, a piece of the story will always be missing in the real world, but fine-tuning minimizes the fears that something which is theoretically wrong has been done, that the theoretical background of policy intervention was not a sound one, in the sense of falsifying observations. in this connection, it is methodologically improper to identify animal spirits with an exogenous random shock of a stationary rational-expectations equilibrium (howitt 1992) because the source of the shock may be anything, and a result like ‘bayesian updating induces convergence to the equilibrium with positive probability even if people start with no definite belief that animal spirits affect the profitability of hiring’ (howitt 1992, 493) may be quite misleading in so far as policy-making is concerned. the same holds at least policy-wise when animal spirits are tried to be approached empirically as ‘interrelations between waves of optimism and pessimism’ (chauvet and guo 2003, 140) because any such approach is subject to lucas critique. european journal of government and economics 4(2) 115 in any case, the profession does already know that when crises break up, a piece of the story is missing because of animal spirits beyond those associated with uncertainty. so, the least the profession can start doing about it from now on, analytically, of course, and in so far as it purports to be empirically relevant, is to cease confusing “nature” with smoothness. akerlof and shiller (2009) report a great many reasons attesting to this call; and cognitive psychology, “twice” as many: all come down to “income and emotion”, to an ageless and ecumenical fact of life, complicated purposefully or not by authority… it is this fact of life that made keynes (1923, 80) emphasise that the: ‘…long run is a misleading guide to current affairs. in the long run we are all dead. economists set themselves too easy, too useless a task, if in tempestuous seasons they can only tell us, that when the storm is long past, the ocean is flat again.’ indeed, ‘psychological ideas… are more important in keynes’s economic theory than is sometimes recognized’ (barnett 2015, 307), which perfectly justifies his support for discretionary policy-making as this paper does, too. does a version of okun’s law 10′ not favour active employment policy-making beyond discretionary monetary policy? appendix let 1 and θ ⁄ . from 13 , 1 2 θ 2 2 and consequently, 2 θ 2 λ 4 where λ 2 2 2 2 2 references akerlof, george a. and robert j. shiller (2009) animal spirits: how human psychology drives the economy, and why it matters for global capitalism. princeton, nj: princeton university press. arestis, philip (2011) ‘keynesian economics and the new consensus in macroeconomics’, in e. hein and e. stockhammer (eds) a modern guide to keynesian macroeconomics and economic policies, 88-111. cheltenham, uk: elgar. barnett, vinvent (2015) ‘keynes and the psychology of economic behavior: from stout and sully to the general theory’, history of political economy 47(2): 307333. chauvet, marcelle and jang-ting guo (2003) ‘sunspots, animal spirits, and economic fluctuations’, macroeconomic dynamics 7(1): 140-169. clore, gerald l. (2005) ‘for love or money: some emotional foundations of rationality’, chicago kent law review 80(3): 1151-1165. cohen, joel b., michel tuan pham, and eduardo b. andrade (2007) ‘the nature and role of affect in consumer judgment and decision making’, in c. p. haugtvedt, p. m. herr, and f. r. kardes (eds) handbook of consumer psychology, 297-438. mahwah, nj: erlbaum. dana, rose-anne and pierre malgrange (1993) ‘the dynamics of a discrete version of a growth cycle model’, in m. jarsulic (ed) non-linear dynamics in economic theory, 497-524: aldershot: elgar. soldatos and varelas ● a first formal approach to animal spirits 116 de grauwe, paul (2012) lectures on behavioral macroeconomics. princeton, nj: princeton university press. frantz, roger s. (2005) two minds: intuition and analysis in the history of economic thought. new york: springer. howitt, peter (1992) ‘animal spirits’, american economic review 82(3): 493-507. jowett, benjamin (trans. and ed) (1892) ‘republic’, in the dialogues of plato, vol. 3. oxford: oxford university press: http://classics.mit.edu/plato/republic.html kahneman, daniel (1994) ‘new challenges to the rationality assumption’, journal of institutional and theoretical economics zeitschrift fur die gesamte staatswissenschaft 150(1): 18-36. kahneman, daniel and angus deaton (2010) ‘high income improves evaluation of life but not emotional well-being’, proceedings of the national academy of the usa 107(38): 16489-16493: http://www.pnas.org/content/107/38/16489.long keynes, john maynard (1923) a tract on monetary reform. london: macmillan. keynes, john maynard (1936) the general theory of employment, interest and money. london: macmillan. kraut, richard (2014) ‘aristotle’s ethics’, in e.n. zalta (ed) the stanford encyclopedia of philosophy, (summer 2014 edition): http://plato.stanford.edu/archives/sum2014/entries/aristotle-ethics/>. moore, basil j. (1988) horizontalists and verticalists: the macroeconomics of credit money. cambridge, ma: cambridge university press. moss, jessica (2008) ‘appearances and calculations: plato’s division of the soul’, oxford studies in ancient philosophy 34: 35-68. parada-daza, jose rigoberto and miguel ignacio parada-contzen (2013) ‘economic and emotional rationality: an application to wealth concentration’, theoretical economics letters 3(4): 233-244. pech, wesley and marcelo milan (2009) ‘behavioral economics and the economics of keynes’, journal of socio-economics 38(6): 891-902. pham, michel tuan (2007) ‘emotion and rationality: a critical review and interpretation of empirical evidence’, review of general psychology 11(2): 155178. piketty, thomas (1998) ‘self-fulfilling beliefs about social status’, journal of public economics 70: 115-132. pilkington, philip (2014) ‘endogenous money and the natural rate of interest: the reemergence of liquidity preference and animal spirits in the post-keynesian theory of capital markets’, levy economics institute working paper no. 817: http://www.levyinstitute.org/pubs/wp_817.pdf roskam, john (2014) ‘interest rate tweaks won’t manufacture animal spirits’, financial review oct 10, 2014: http://www.afr.com/news/economy/monetarypolicy/interest-rate-tweaks-wont-manufacture-animal-spirits-20141009-11cnpf ross, william david (trans. and ed) (1908) nicomachean ethics. oxford: clarendon press: http://classics.mit.edu/aristotle/nicomachaen.html scherer, klaus r. (2011) ‘on the rationality of emotions: or, when are emotions rational?’, social science information 50(3-4): 330-350. schwartz, hugh (2010) ‘does akerlof and shiller’s animal spirits provide a helpful new approach for macroeconomics?’, journal of socio-economics 39(2): 150-154. european journal of government and economics 4(2) 117 shiller, robert j. (2015) irrational exuberance , 3rd ed. princeton, nj: princeton university press. staveren, irene van (2007) ‘beyond utilitarianism and deontology: ethics in economics’, review of political economy 19(1): 21-35. weber, max (1964) ‘wirtschaft und gesellschaft’, in j. winckelmann (ed) grundriss der verstehenden soziologie. cologne: kiepenheuer & witsch. wicksell, knut (2007 [1898]) interest and prices: a study of the causes regulating the value of money. digitally by mises institute: https://mises.org/library/interestand-prices. microsoft word ejge_02_02_043_corr2.doc european journal of government and economics volume 2, number 2 (december 2013) issn: 2254-7088 100 the optimal top marginal tax rate: application to hungary áron kiss, magyar nemzeti bank (central bank of hungary), hungary abstract the paper applies recent developments in the theory of optimal income taxation to the hungarian personal income tax system. the main conclusion is that the optimal top marginal tax rate in hungary is likely to be higher, perhaps substantially, than the actual rate. it is discussed how this result depends on the parameters describing labour-supply behaviour, the income distribution, and the redistributive preferences of society. jel classification h21; h24 keywords optimal income taxation; top income tax rate; hungary; emerging markets acknowledgements this research had been conducted before the author started to work at the european commission. views and opinions expressed in this paper are those of the author and do not necessarily represent those of his past or present institutions. the author would like to thank péter benczúr, iván csaba, gábor kátay, seminar participants at the central bank of hungary, participants of the 'symposium on taxes, transfers and the labour market' organised by mta-krtkkti and mnb in june 2012 and the mta-krtk-kti conference on 'wages, taxes and transfers', held in szirák, november 2012, as well as two anonymous referees and the editor for useful comments and suggestions. any remaining error or omission is the responsibility of the author. current affiliation: european commission, dg ecfin. kiss ● the optimal top marginal tax rate 101 introduction there was hardly a year in the last decade that the hungarian personal income tax (pit) remained unchanged. the last three years saw radical changes, including the introduction in 2011 of a flat tax rate. the current wave of reforms is expected to be complete in 2013 when the single pit rate is 16 percent. the reforms benefited different groups in different years but altogether they brought a radical cut in the marginal (and average) tax rates of high-income earners. at the same time, the elimination of the employee tax credit (adójóváírás) in 2012 brought an increase in the average tax rate of low and middle income earners without children. amid such frequent and radical changes it is important to ask what, if anything, economic theory can say about the desired characteristics of the income tax system. this paper builds on recent developments of the theory of optimal income taxation and applies one of its main results to the hungarian tax system.1 the foundational work of the theory of optimal income taxation is by mirrlees (1971). at the core of the theory is the insight that while society (or a government making policy to implement the preferences of society2) would like to redistribute income from high earners to low earners, redistribution dampens the work incentives of both high and low earners. the optimal tax policy thus reflects a balance between redistribution and incentives or, in other words, between the principles of ‘equity’ and ‘efficiency’. while mirrlees’ work was very influential in economic theory, results derived from optimal income tax theory were not seen as being of much practical guidance to tax policy. this changed with the work of saez (2001). saez, building on work by diamond (1998), was able to express the results of the theory as functions of estimable parameters. the relevant parameters include those that describe the shape of the income distribution and those that describe how sensitively people’s earnings react to changes in the tax rates. even after these developments it should not be expected that the theory gives unequivocal answers to all tax policy questions. one reason is that the optimal income tax schedule depends on the revenue to be achieved, which in turn depends on the desired level of government expenditures and on how distortionary alternative taxes are relative to the pit. the other important reason is that the optimal tax system depends on the strength of the redistributive preferences of society. an applied optimal taxation model needs those preferences as inputs to derive optimal tax rates. the theory of optimal income taxation can provide useful qualitative guidance for tax policy because some of its results do not depend on these important (but uncertain) factors. it is even possible, under relatively unrestrictive assumptions, to derive even quantitative guidance with respect to the top marginal tax rate. this paper follows two recent studies that provide examples of how to use the theory of optimal taxation to derive recommendations for tax policy. the first of these studies, by brewer et al. (2010) was prepared in the framework of the mirrlees review, a detailed review of the uk tax system by an international group of researchers coordinated by the institute for fiscal studies. the second study, by 1 an earlier version of the paper appeared as a discussion paper (kiss 2013). that version had a somewhat broader focus, also surveying theoretical results about the optimal lower tax rates. some results of the analysis have been discussed in the survey article of benczúr, kiss and mosberger (2013). 2 in the paper ‘society’ and ‘government’ is used interchangeably. the analysis abstracts from the mechanisms of policy formation and the political process and adopts the approach of classical public economics in which society (or the government) is searching for the best solution for a given problem. clearly, the political economy aspect is important in the real world but it is not in the focus of this paper. european journal of government and economics 2(2) 102 diamond and saez (2011), describes three broad principles that can be derived from the theory and discusses them in the context of current us tax policy.3 this paper applies one result of the literature to the hungarian context, suggesting that high marginal tax rates for top earners can be optimal. the result is based on the observation, made by saez (2001), that the optimal top marginal tax rate can be expressed, under general assumptions, by just three parameters: a parameter describing the shape of the top of the income distribution, a behavioural elasticity expressing how sensitively high-income taxpayers react to changes in the marginal tax rate and, finally, a parameter that expresses the social value of an additional dollar kept by a top earner, expressed in terms of public funds. the first two of these parameters can be estimated, while the third parameter is the function of the preferences of society. brewer et al. (2010) and diamond and saez (2011) argue that the social marginal value of an additional dollar kept by a top earner (an individual belonging to the top 1 percent of earners in their definition) is close to zero. this can be valid on a utilitarian basis (where the welfare of every citizen is equally important to the government) if an extra dollar of income adds to the welfare of a high earner much less than to the welfare of a low earner (or, more precisely, if the marginal utility of consumption declines to zero at very high incomes). in this case, the optimal top marginal tax rate is equal to the top rate that maximises government revenue. the utilitarian interpretation is appealing because then the question of optimality is viewed from the hypothetical perspective of a not-yet-born individual, ‘from behind the veil of ignorance’, who expects her earning capacity to be a random draw from the empirical income distribution. this individual prefers a tax system that maximises her expected utility over the possible levels of earning capacity she could be assigned in the ‘birth-lottery’. the present paper derives the formula for the optimal top marginal tax rate and evaluates it for hungary for alternative values of the social-marginal-value parameter (zero as well as non-zero). the calculations suggest that the revenuemaximising top marginal tax rate is higher than the current top marginal tax rate: the revenue-maximising top marginal pit rate is estimated to be about 30 percent (or about 40 percent if pension contributions were capped). it is discussed how the optimal top marginal tax rate may depend on the definition of top incomes and on the social-marginal-value parameter. the optimal top marginal tax rate is shown to be lower, although not dramatically, than the revenue-maximising rate if plausible non-zero values are chosen for the parameter expressing the social marginal value of a dollar kept by a top earner. fiscal effects of hypothetical tax reforms that are consistent with these results are simulated. the rest of the paper is organised as follows. the next section derives the theoretical formulae for the revenue-maximising and optimal top marginal tax rates. section 3 evaluates the theoretical benchmarks in the hungarian case. section 4 compares these benchmarks to recent actual top marginal tax rates in hungary, derives conclusions for policy, and surveys alternative considerations not taken into account in the baseline analysis. section 5 concludes. the optimal top marginal tax rate: theory the surprising result that optimal tax theory can be used to give quantitative guidance about the top marginal tax rate was first derived by saez (2001). the methodology has since been used to analyse whether actual top marginal tax rates in the us (diamond and saez, 2011) and the uk (brewer et al., 2010) are in line 3 in recent papers, similar approaches have been applied to germany by bach et al. (2012) and finland by riihelä et al. (2013). kiss ● the optimal top marginal tax rate 103 with the optimal tax rates derived from theory. this section follows these studies to derive the simple formula for the optimal top marginal tax rate, evaluates the formula for the hungarian tax system, and compares the resulting tax rate with actual top marginal tax rates of recent years. the presentation of the theoretical background follows brewer et al. (2010). before setting out to derive the formula for the optimal top marginal tax rate, it should be clarified what is meant by ‘top’ incomes. diamond and saez (2011) defines top incomes as the top 1 percent of the income distribution. in this paper it is be explored how conclusions depend on whether top incomes are defined as the top 1 percent (starting at annual income level of huf 10.6 million in hungary in 2008, or about eur 35,000);4 or the top 5 percent (starting at huf 5.3 million, or about eur 18,000); or the top 10 percent (starting at about huf 3.8 million, or about eur 13,000). consider an economy where individuals have different earning capacities. individual i earns a gross income zi. of this gross income the individual pays t(zi) in taxes and consumes the rest, ci = zi – t(zi). individuals value consumption and leisure and can adjust their hours worked (or, more generally, any aspect of their labour effort) as a response to the rate of exchange between the two. how much the individual can consume in exchange for an additional hour worked depends on the marginal tax rate τ(z) = t’(z). the government sets tax rates and transfers in a way as to maximise social welfare while reaching an exogenously given level of net revenue required for the provision of public goods. (as a simplification it is often assumed that the sum of tax revenue has to be equal to the sum of transfers paid, that is, the exogenous level of other public spending is zero.) the marginal weight of an individual in the social welfare function is g(z): this expresses the value society attaches to an additional dollar consumed by an individual with gross income z, expressed in terms of public funds. if the state has redistributive preferences, then g > 1 for low earners (a one-dollar increase in their consumption is worth more than one dollar for the government) while g < 1 for high earners (a one-dollar increase in their consumption is worth less than a dollar for the government). the trade off facing the government is this: it values redistribution but redistribution dampens individuals’ incentives to work. if individuals work less because of the tax system, revenues decrease and less redistribution can be achieved. this trade off between equity and efficiency is at the centre of the theory of optimal income taxation. to see this trade off in specific, consider an increase dτ in the marginal tax rate facing top income earners, i.e., the n individuals earning an income higher than ż. this will affect social welfare in three ways: (1) tax revenue increases mechanically and that is a social gain; (2) the increased tax burden makes those affected worse off and that is a social loss; (3) those affected will reduce their work effort and with that their taxes payable and that is again a social loss. the first effect is thus the mechanical effect on tax revenue: all individuals earning more than ż will pay more taxes than before. in the present example there are n taxpayers earning more than ż. let their average gross income be zm. the tax increase affects their income above the threshold ż. thus, the mechanical effect on tax revenue is: dm = n[zm – ż]dτ > 0. 4 a conversion rate of 1 eur = 300 huf is used throughout the paper. this rate prevailed during most of 2012 and 2013. the rate was about 270 to 280 from 2008 to 2011, and about 250 in the years before 2008. european journal of government and economics 2(2) 104 note that this effect is defined as the change in tax revenue before any behavioural change occurs on the part of the individuals affected. the second effect is the direct welfare loss of those who have to pay more taxes. in the present example the welfare effect is given by: dw = – g·n[zm – ż]dτ < 0. where g is the average social marginal value of consumption of individuals earning more than ż. the last is the behavioural effect: the increase of the marginal tax rate induces high earners to decrease their work effort which results in a fall in tax revenues. the decrease in tax revenues is db = dz·τ·n where dz is the average change of income for individuals affected by the tax increase. the empirical studies estimating behavioural responses to tax changes estimate a parameter given by the following expression: e = (dz/z) · [(1 – τ)/d(1 – τ)]. as can be seen from this expression, parameter e is an elasticity: it measures the percentage change of reported income as a response to a 1 percent change of the marginal net-of-tax rate (1 – τ), which is the share of the last unit of gross income that the individual can take home as net income. expressing dz from this formula and substituting into the definition of db, we get db = – n · e · zm · dτ · τ/(1 – τ) < 0. at the optimal marginal tax rate τ* the sum of these effects must be equal to zero. if the welfare effect of a small tax increase were positive (negative), the government would want to increase (cut) the tax rate further; thus the initial tax rate could not have been optimal. from this argument it follows that the equation dm + dw + db = 0 implicitly determines the optimal top tax rate. introducing the parameter a = zm/(zm – ż) we can solve the equation to reach a simple formula: τ* = (1 – g)/(1 – g + e·a). the optimal top marginal tax rate thus depends on three parameters, two of which can be estimated: parameter e is the elasticity of taxable income with respect to the net-of-tax rate, while parameter a characterises the shape of the income distribution. since the third parameter g is a function of society’s preferences, it is less straightforward to assess its plausible values. an upper bound to the top marginal tax rate can be obtained by considering the case when g = 0. in this case the value society attaches to an additional dollar kept by a top earner is negligible compared to the value society attaches to an additional dollar kept by the average earner (or to an additional dollar of government revenue). then, the only force keeping the marginal tax rate of top earners from rising is the behavioural effect. in this case the optimal top marginal rate is equal to the revenue-maximising rate, with the formula simplifying to: τ* = 1/(1 + e·a). brewer et al. (2010) and diamond and saez (2011) argue that g = 0 is plausible for top earners (the top 1 percent of the income distribution in their definition). for most social welfare functions with redistributive preferences it will be the case that g decreases with income, and the zero-marginal-weight result will hold asymptotically for social welfare functions that satisfy the property lim g(z) = 0 as z goes to infinity. this is the case, for example, in a utilitarian framework (where the welfare of every individual is equally important for the government) where the marginal utility from consumption declines to zero. in all these cases it will be true that the narrower the top income bracket is defined, the higher the optimal top kiss ● the optimal top marginal tax rate 105 marginal tax rate is and the closer it is going to be to the revenue-maximising tax rate. but parameter g does not have to converge to zero for the revenue-maximising tax rate to be approximately optimal. note that g enters (with the same sign) both the numerator and the denominator of the general formula. this means that the effect of g will be of second order as long as g is not too large. a sensitivity analysis to the value of g is presented in the next subsection. evaluating the theoretical benchmarks for hungary the revenue-maximising top marginal tax rate in hungary the previous subsection established that the revenue-maximising top marginal tax rate depends on only two parameters: parameter a describing how thin the income distribution is at the top and parameter e describing how sensitively top earners react to changes in the marginal tax rate. in this subsection, estimates for parameters a and e are presented. with these parameters it is possible to calculate the revenue-maximising top marginal tax rate. as seen above, the revenuemaximising top marginal tax rate is optimal when the value of parameter g is zero. the next subsection contains the more general case, calculating the value of the optimal top marginal tax rate as a function of parameter g. the income-distribution parameter first, the empirical value of parameter a is estimated for hungary. recall that parameter a is defined as a = zm/(zm – ż), that is, for any income limit ż, a is equal to the average income of individuals above the income limit divided by the difference of that average income and the income limit. a ten-percent sample of the 2008 pit returns is used for this exercise, compiled by the hungarian tax authority (the population excludes the full-time self-employed). figure 1 shows the value of a for annual gross income levels between huf 0.1 million (about eur 330) and huf 40 million (about eur 130,000). for easier interpretation of the figure it can be noted that the average annual gross income in 2008 was about huf 1.9 million (about eur 6,300) while an individual belonged to the top 1 percent of tax filers with an annual gross income of about huf 10.6 million (equivalent to about eur 35,000). european journal of government and economics 2(2) 106 figure 1. the value of parameter a as a function of the income limit chosen note: author’s calculation based on 2008 tax return data provided by the hungarian tax authority. figure 1 shows that the value of parameter a is very stable for income limits above huf 5 million (about eur 17,000). it is around 2.35 for incomes between huf 6 and 23 million and is very close to 2.5 for income levels above that. (note that individuals above income of huf 5 million represent about the top 5 percent of the income distribution, while the top 1 percent starts at about huf 10.6 million.) that this parameter is stable for the upper part of the income distribution is a general result first noted by the italian economist and statistician vilfredo pareto and confirmed for many countries and time periods ever since. in more intuitive terms it is equivalent with the statement that the average income of individuals earning an income higher than ż is a constant multiple of ż. in the hungarian tax data the average income of individuals earning more than ż is about 1.7 · ż. based on figure 1, a central estimate of a = 2.5 is adopted but the optimal tax rate formula is also evaluated for somewhat lower and higher values of parameter a (also called the pareto parameter). the taxable-income elasticity the second parameter needed to calculate the optimal top marginal tax rate is e, the elasticity of taxable income with respect to the marginal rate, net-of-tax. this parameter first came into the focus of economic research with the estimations of feldstein (1995). before his work, labour economists estimated the effect of tax changes on hours worked and found low elasticities. it is since feldstein’s work that economists estimate the effects of tax changes on reported taxable income. clearly, this measure encompasses more than the change in hours worked: it can reflect changes in work intensity, change of jobs, moonlighting, but also tax avoidance and evasion. this list of possible factors behind the elasticity makes clear that it does not necessarily reflect changes of real economic activity. nevertheless, it is the welfare-relevant measure of the behavioural effect. this is because, unless there are significant externalities between various tax bases, the elasticity is a direct measure of how government revenue changes in response to a change in the tax rate. kiss ● the optimal top marginal tax rate 107 since the work of feldstein (1995) panel data with more years and individual observations became available which allowed more robust statistical methods to be applied. the elasticity of e = 0.4 estimated by gruber and saez (2002) is considered as representative of the newer literature for the us (see also recent surveys of the literature by giertz [2004] and saez et al. [2012]). there are two studies that estimated the taxable income elasticity for hungary. in the first such study, bakos et al. (2008, henceforth bbb), used the elimination of the middle tax bracket in 2005 as the policy experiment to identify the elasticity. since most of the tax changes in their data affected middle-income earners, most results of bbb reflect the taxable-income elasticity of middle-income individuals. updated and re-estimated results of bbb have been surveyed, for a non-technical audience, by benczúr et al. (2013). the re-estimation resulted, in general, in lower estimated elasticities. one income range in which the results were robust to the reestimation is the range between huf 1.5 to 1.95 million (about eur 5,000 to 6,500 at 2013 conversion rates, which was slightly above the average wage in 2004; see benczúr et al., 2013, table 2.4). in this income range bbb find an elasticity of about e = 0.12, together with a significant, although moderate income effect. (how optimal tax rates depend on the income effect is discussed in subsection 2.4.) the second study of the taxable-income elasticity for hungary was conducted by kiss and mosberger (2011). they use the introduction of an extraordinary tax on high-income individuals in 2007 as their policy episode, thus they focus on the group of individuals most relevant for the present purposes (their main specification includes individuals earning huf 5 to 8 million in 2005 or about eur 17 to 27 thousand at 2013 conversion rates). the main estimates for the elasticity fall between 0.15 and 0.2. the present calculation of the optimal top marginal tax rate uses this estimate (e = 0.2) as a benchmark but the formula is evaluated at somewhat lower and higher values as well. the income effect estimated by kiss and mosberger (2011) is relatively large in magnitude, but not statistically significant in the main specification. this central estimate is in line with most estimates for countries outside the us.5 some observers attributed the higher estimated us elasticities at least partly to tax optimisation strategies (timing and form of compensation, among others) that are available to executives in the us (goolsbee 2000). this consideration shows that the elasticity of taxable income is not an immutable parameter but can be influenced by tax policy. the revenue-maximising top marginal tax rate based on these parameter values one can evaluate the formula for the revenuemaximising top marginal tax rate. table 1 shows the value of the tax rate as a function of parameters a and e, with g = 0. the table shows that the revenuemaximising top marginal tax rate of high earners is 67 percent at the central parameter estimates (a = 2.5 and e = 0.2). of course, this tax rate cannot be directly compared to the actual pit rates, since taxpayers also pay social security contributions (ssc) and consumption taxes. comparable actual top marginal tax rates for the last few years in hungary are calculated in the next subsection. 5 for references to elasticity estimates in other countries, see benczúr et al. (2013). european journal of government and economics 2(2) 108 table 1: the revenue-maximising top marginal tax rate as a function of a and e values of a values of e 0.1 0.2 0.3 0.4 2 83% 71% 63% 56% 2.5 80% 67% 57% 50% 3 77% 63% 53% 45% note: author’s calculations. table 1 also shows how, within the plausible range of parameter estimates, the revenue-maximising top marginal tax rate depends on the parameters. if the taxable-income elasticity were 0.1 or 0.3 rather than 0.2, this would change the optimal top marginal tax rate by about 10 percentage points. at the same time, if the value of the pareto parameter were equal to 2 or 3 instead of 2.5, this would change the optimal top marginal tax rate by 4 percentage points. the intuition behind the effect of this parameter is the following. a smaller a implies a fatter tail of the income distribution, which in turn implies that the behavioural effect will be more significant relative to the behavioural effect (the revenue effect is proportional to zm – ż, while the behavioural effect is proportional to zm). the optimal top marginal tax rate but how sensitive is the optimal top marginal tax rate to the value of parameter g? to be able to answer this question, first let us take a simple theoretical benchmark for guidance as to what are sensible values of g. a popular benchmark, and one that is also used by diamond and saez (2011), supposes that the social marginal value of consumption is inversely proportional to income, i.e., g(z) = 1/z. such a social marginal welfare function could be a result of social preferences that favour redistribution per se, or it could be a result of utilitarian preferences if the welfare of the individual is a logarithmic function of income. in the utilitarian framework the social marginal welfare weight of highincome individuals is low not because their welfare is less important to the government (or to society) but because an additional dollar is worth less for them than for lower-income individuals. tax return data from 2008 are used to calculate the average income of the top 1 percent, the top 5 percent, and the top 10 percent of the income distribution before calculating the value of parameter g for these groups according to the formula g(z) = 1/z. table 2 shows the results of this exercise. table 2: average income of top earners and calculated values of parameter g. income range income limit (huf million) average income (huf million) value of g in example top 1% 10.6 18.3 0.11 top 5% 5.3 9.3 0.21 top 10% 3.8 6.8 0.28 note: author’s calculations. in the last column, g is calculated as the inverse of the ratio of the average income of the respective top income group to the overall average income (huf 1.91 million) in 2008. table 2 shows that in 2008 an individual belonged to the top 1 percent of tax filers if her income exceeded huf 10.6 million (about eur 35,000). the average income of the top 1 percent was about huf 18.3 million (about eur 61,000). in the same year the average gross annual income was about huf 1.9 million (about eur 6,300). if the average social marginal value of income is inversely proportional to kiss ● the optimal top marginal tax rate 109 income and the value of g is about 1 for the average earner then the value of g is about 0.1 for the top 1 percent. similarly, the lower limit of the top 5 percent was, in 2008, at about huf 5.3 million (about eur 18,000) while the calculated g is about 0.2. the lower limit of the top 10 percent was, in the same year, at about 3.8 million (about eur 15,000) while the calculated g is about 0.3. using these simple calculations as orientation the optimal top marginal tax rate can be calculated as a function of parameter g. table 3 shows the results of such an exercise, with the value of parameter a held fixed at the central estimate of 2.5. table 3: the optimal top marginal tax rate as a function of g and e, with a = 2.5 values of g values of e 0.1 0.2 0.3 0.4 0 80% 67% 57% 50% 0.1 78% 64% 55% 47% 0.2 76% 62% 52% 44% 0.3 74% 58% 48% 41% 0.4 71% 55% 44% 38% note: author’s calculations. table 3 shows that small changes of g have little effect on the optimal top rate. to take an example, choosing g = 0.1 instead of zero affects the optimal top marginal tax rate very little: for the central parameter values a = 2.5 and e = 0.2 the optimal top marginal tax rate becomes 64 percent instead of 67 percent. as it was shown above, if g is inversely proportional to income, then the value of g should be about 0.1 for the top 1 percent in hungary. thus this result means that it matters little whether, for the top 1 percent of earners, we set g = 0 or use g(z) = 1/z as a reasonable approximation. it follows from the calculations of table 3 that the broader the top income bracket is defined, the lower the optimal top rate will be. in the numerical example above the marginal welfare weight of the top income group was calculated to be about 0.1 for the top 1 percent, about 0.2 for the top 5 percent, and about 0.3 for the top 10 percent. in this example the optimal top marginal tax rate is approximately 64 percent if it affects only the top 1 percent (applying to income above huf 10.6 million at 2008 prices) but it is only 58 percent if it affects the top 10 percent (applying to income above huf 3.8 million at 2008 prices). comparing the theoretical benchmarks to actual tax rates the actual top marginal rate in hungary, 2005—2013 the actual tax rate that can be compared to the theoretical benchmark is not simply the top pit rate. if there is a weak link between social security contributions (ssc) and future benefits received, then ssc can be assumed to have the same disincentive effect as the pit. ideally, the tax rate corresponding to the theoretical benchmark answers the following question: by how much can an individual increase her consumption if her total labour cost is increased by one unit? thus, consumption taxes must also be taken into account. based on these considerations, the effective marginal tax rate of top earners for hungary is evaluated for recent years according to the following formula: τtop = 1 – (1 – τcons) · [(1 – τpit – τee)/(1 + τer)]. european journal of government and economics 2(2) 110 here, τpit is the top pit rate, τee is the rate of employee contributions, τer is the rate of employer contributions and τcons is the effective tax rate on consumption. (the last term of the r.h.s. expression shows the ratio of net wage to total wage cost.) two issues regarding the tax rates merit further discussion.6 the first issue is related to the employee contributions. the actual top marginal tax rate is calculated here for a hypothetical person with income above the pension contribution ceiling. until 2013, individuals with income above the ceiling did not have to pay the employee pension contribution, which reduced their marginal effective tax rate by about 10 percent. the ceiling has in recent years been between the 95th and the 99th percentile of the income distribution, which means that individuals at the 99th percentile faced a lower marginal tax rate than individuals at the 95th percentile. in 2013 the pension contribution ceiling was abolished. for our purposes it complicates matters somewhat that the abolition of the contribution ceiling will mean higher future pension benefits for current high-income individuals. if the relationship between additional contributions and additional future benefits is close, this measure shouldn’t be seen as a tax increase at all. the issue of the relationship between contributions and benefits is discussed in subsection 4.3. the second issue is the effective consumption tax rate. ideally, we should be able to measure the tax share of the consumption basket of high-earning households. this is not the case in practice, however. thus an approximation of the effective rate of consumption taxes has to be used. similarly to brewer et al. (2010, see online appendix, p. 3), the effective consumption tax rate is calculated here as government revenue from consumption taxes divided by total consumption from the national accounts. consumption taxes include excise duties on tobacco, alcohol products and gasoline, and further smaller items, besides vat. table a1 of the appendix shows, for selected years, the tax rates taken into account in the calculation. the table shows that the effective consumption tax rate is between about 15 to 16 percent in the period 2005 to 2010 (for the years after that the 2010 value was used for lack of data). alternative methods to calculate the effective consumption tax rate would most likely result in similar values. the actual top marginal tax rates that result from these calculations are shown in figure 2, along with the optimal top marginal tax rate at different values of g. until 2013 the actual top marginal tax rate is depicted both below and above the pension contribution ceiling. in 2013 the difference ceases to exist due to the abolition of the ceiling. 6 additionally, it should be noted that this approach abstracts from general-equilibrium effects and assumes that labour bears the burden of all labour taxation. kiss ● the optimal top marginal tax rate 111 figure 2. actual effective top tax rates in hungary, 2005-2013 note: calculations of the author based on statutory tax rates, budget reports and the national accounts. the effective consumption tax rate for the years 2011 to 2013 is, for lack of data, taken to be equal to its 2010 value. it is apparent from figure 2 that the gradual introduction of a flat tax of 16 percent between 2011 and 2013 has cut the top marginal tax rate significantly: from about 65 percent (above the pension contribution ceiling) and about 72 percent (below the ceiling) to about 56.6 percent in 2013 (see also table a1 of the appendix for the exact figures).7 policy conclusions and simulations figure 2 suggests two conclusions about top tax rates in recent years. first, for individuals with income above the pension contribution ceiling, the marginal tax rate before 2010 was around the revenue-maximising tax rate, while the 2011 tax cut resulted in marginal tax rates that were much lower. second, individuals just below the pension contribution ceiling faced marginal tax rates in the period 2005 to 2010 that were above the revenue-maximising top marginal tax rate. the first conclusion means that, if the marginal social weight of the highest earners is close to zero, the tax revenue foregone by the post-2010 tax cuts for top earners cannot be justified by the behavioural effect: the estimated taxable income elasticity is not large enough to motivate a pit rate of 16 percent. the calculated optimal top marginal tax rate is consistent, after the abolition of the pensioncontribution ceiling, with a top pit rate between 24 percent (about optimal under g = 0.2) and 31 percent (about optimal under g = 0). if the pension contribution ceiling had not been abolished, the calculated optimal top marginal tax rate would be consistent with a top pit rate between 34 percent (about optimal under g = 0.2) and 41 percent (about optimal under g = 0). for the top pit rate of 16 percent to be optimal, g has to be around 0.35 if the taxable income elasticity is 0.2. alternatively, if g = 0, a top pit rate of 16 percent is optimal if the elasticity parameter e = 0.3, much higher than our central estimate 7 because of the lack of contemporaneous data, the calculations of figure 2 do not take into account minor increases in consumption taxes that came into effect in 2011 and 2012 (the most significant among these is the increase from 25% to 27% of the base vat rate in 2012). these would probably not add more than 1 or 2 percentage points to the top marginal rate, however. thus they would not affect the results qualitatively. european journal of government and economics 2(2) 112 (this pit rate corresponds to an effective top marginal rate of 57 percent in table 3, row 1). the second conclusion (that individuals with income just below the pension contribution ceiling faced a marginal tax rate above the revenue-maximising top marginal tax rate) shows that, at this point of the analysis, the definition of the ‘top income range’ becomes important. even though the theoretical analysis suggests that at high incomes the marginal tax rate should be increasing, individuals below the pension contribution ceiling were facing a higher marginal tax rate than those above the ceiling in recent years (the marginal tax rate schedules for incomes above huf 3 million [or about eur 10,000] for selected years between 2005 and 2013 are shown in figure a1 of the appendix). very high marginal tax rates are justifiable at income levels where the taxable-income elasticity is thought to be close to zero (as it might be the case around the average wage), but this is not the case, according to the estimations of kiss and mosberger (2011), in the income range just below the pension contribution ceiling. in the following, two policy simulations are presented to show examples of tax policy measures that are consistent with the theoretical results of this paper. the examples were chosen to be as simple as possible and clearly they are not the only conceivable policy measures that are consistent with the theoretical results. thus it is not implied here that these examples constitute ‘the optimum’. it is also not attempted here to design a whole new pit system: the example measures affect only the top part of the income distribution. the simulations were conducted in the microsimulation model described by benczúr et al. (2011, 2012). the model takes into account the behavioural responses to taxation by incorporating the taxable-income elasticity of 0.2 estimated by kiss and mosberger (2011), and it also takes into account generalequilibrium effects through a simple neo-classical macroeconomic model. this means that if individuals change their labour supply as a response to changes in the tax system, the general-equilibrium model calculates how this affects wages and the stock of capital. while the labour-supply response of the individuals is thought to occur relatively quickly, the dynamic macroeconomic effects are longrun effects since the adjustment of the capital stock takes time. generally, full longrun macroeconomic adjustment takes place within a decade. table 4 summarises the results of the simulations. for each simulation, a ‘static fiscal effect’ is shown first: this says by how much government revenue would increase from the pit and ssc if the given measure were adopted, absent behavioural change. the ‘dynamic fiscal effect,’ in turn, takes into account both the behavioural response to the tax change by the taxpayers and the long-run macroeconomic adjustments. additionally, the ratio of the dynamic and static effects is displayed. the two simple measures analysed are: (1) the introduction of a pit rate of 31 percent above huf 10 million (about eur 33,000 or roughly 1 percent of taxpayers); (2) the introduction of a pit rate of 24 percent above huf 5 million (about eur 17,000 or roughly 5 percent of taxpayers). all tax policy measures are compared to a baseline of a flat pit of 16 percent, as effective in 2013, after the abolition of the pension contribution ceiling. kiss ● the optimal top marginal tax rate 113 table 4: hypothetical tax policy scenarios static fiscal effect (huf billion) dynamic fiscal effect (huf billion) ratio of dynamic to static (1) pit rate of 31% above huf 10 million 51 23 45% (2) pit rate of 24% above huf 5 million 72 39 54% note: the 2013 tax system was used as benchmark. revenue is estimated at 2013 prices. fiscal estimates include effects on the pit and ssc. the dynamic simulations are based on the central taxable-income-elasticity estimates by kiss and mosberger (2011). table 4 shows that the introduction of a 31 percent tax bracket for the top 1 percent would increase government revenue by about huf 51 billion under no behavioural change, and by about huf 23 billion after behavioural effects and macroeconomic feedback effects are taken into account. the effects of a lower top rate levied on a broader base (24 percent on income exceeding huf 5 million) are larger by about one-half. while the modelling strategy of the microsimulation model does not conform fully to the assumptions made in the approach followed in this paper, the simulations seem to confirm that the higher the top marginal tax rate is increased, the higher the behavioural effect (and with that the income lost) as a proportion of the static effect. this illustrates the conflict between equity and efficiency that is in the core of the optimal taxation problem. alternative considerations this subsection looks at some considerations that might qualify or modify the conclusions based on the benchmark results. the shape of the income distribution the optimal top marginal rates depend on the shape of the income distribution. mankiw et al. (2009) find that while the optimal marginal tax rate at $150,000 is between 60 and 70 percent if a pareto distribution is assumed, it is only about 40 percent with a lognormal distribution. diamond and saez (2011), in contrast, give detailed arguments about why calculations should be based on a pareto distribution. the argument empirically turns on the constancy of parameter a. in the case of our data it can be confirmed that the empirical value of this parameter remains very stable even at the highest income levels where only dozens of individuals are observed in the 10 percent sample of hungarian taxpayers. this is a strong empirical argument for the pareto distribution. alternative welfare paradigms this paper has reviewed arguments for the plausibility of a utilitarian approach which implies that the social marginal weight of high earners is close to zero if the marginal utility of consumption approaches zero at very high consumption levels. alternatives to this approach exist. a ‘rawlsian’ approach would perhaps assign a positive social marginal welfare weight to a group of low-earning (or low-welfare) individuals and a zero weight to the rest of society. alternatively, the approach dubbed ‘charitable conservatism’ by atkinson (1990) values the reduction of poverty, but does not value redistribution above the poverty line.8 this approach is consistent with assigning a weight greater than 1 to ‘the poor’ and a positive weight smaller than 1 to the rest of society. the result that the current top marginal rate in 8 riihelä et al. (2013) show example calculations for this type of social welfare function besides the approach used in this paper. european journal of government and economics 2(2) 114 hungary is consistent with a social marginal weight of 0.35 for top earners could potentially be interpreted in this light. how would the income effect change the results? throughout the analysis it was assumed that the marginal tax rate is the only aspect of the tax system that affects the behaviour of high-income earners. it is possible, however, that their behaviour is affected by the so-called income effect, too. the income effect is operative when an individual decreases her work effort after receiving a transfer that does not affect her marginal incentives. if such an effect were indeed operative, then the gain from the behavioural effect would be dampened in response to a cut in the top marginal tax rate. to see this, suppose that the marginal tax rate is cut by ∆τ for those with an income above ż. this tax change would affect an individual with income z’ > ż in two separate ways. first, the cut in the marginal tax rate would have a positive incentive effect. but also, the individual would experience a ‘quasi lump-sum’ increase in her income of the magnitude ∆τ · (z’ – ż). if the income effect is operative, this is a disincentive to work. since the optimal tax rate is calculated as a function of the behavioural effect (the greater the behavioural effect, the lower the optimal tax rate), the presence of the income effect, by dampening the behavioural effect, increases the optimal top marginal tax rate. most studies that estimate the taxpayer response to tax changes have not found a significant income effect. the study by bakos et al. (2008), is an exception both regarding its original results and as re-estimated (the re-estimated bbb results are surveyed by benczúr et al. [2013]). kiss and mosberger (2011) also found a sizeable income effect which, however, was not precisely estimated. in a robustness analysis of microsimulation results, benczúr et al. (2012, table 8) report that the point estimate of the income effect found by kiss and mosberger wipes out most of the estimated behavioural effect of the tax changes between 2010 and 2013. if an income effect of such magnitude were established, this would imply much higher optimal tax rates than the benchmark results in this paper. these considerations show that while the income effect is potentially important, its magnitude cannot be precisely assessed based on the available information. for this reason, among others, the measurable effects of the large recent tax reforms will be of great importance. if the behavioural effect appears to be large (or small) based on incoming data in 2014, this may give some guidance as to which end of the results presented in this paper may be more relevant. actuarial considerations the baseline results have been derived under the assumption that ssc have the same effects as taxes. (this is the approach taken by brewer et al. [2010] and diamond and saez [2011].) this assumption is easy to justify if the link between present contributions and future benefits is weak, but not if they are strong. arguably, the relationship is weak in the case of healthcare benefits but stronger in the case of pension benefits. if individuals view part of their ssc as savings which they receive back in old age (at least in expectation), then these contributions may not have the same distortionary effect on their working decisions than taxes have. in this case actual marginal tax rates should be calculated with the exclusion of one part of contributions, and the optimal pit rates will be higher. in the case of hungary, it is not clear that the abolition of the pension contribution ceiling in 2013 has the same effect as increasing the top marginal pit rate would have. high-income individuals previously above the pension contribution ceiling now have to pay employee pension contributions, but their promised pension benefits also increase. in this case, the relationship between additional kiss ● the optimal top marginal tax rate 115 contributions and additional benefits may be quite salient and hard to undo by future policy-makers. for this reason, an example has been calculated in which the marginal link between pension contributions and benefits is perceived by taxpayers to be 50 percent. this is the middle way between two extreme theoretical cases. in the first theoretical extreme, a fixed-sum basic pension is financed from income-dependent tax revenue: here the marginal link between ‘contributions’ and benefits is zero. in the other theoretical extreme, pension contributions are paid to actual private accounts which form the basis of the future pension benefit annuity. in this case, the marginal link between contributions and benefits is 100 percent. the reality of the hungarian paygo pension system is arguably somewhere in-between. figure 3 shows the actual top marginal tax rates in recent years under the assumption that taxpayers can consider half of their pension contributions as ‘their own money’ (while another part of their future pension benefits are assumed to be lump-sum). figure 3. actual effective top tax rates in hungary under the assumption that the marginal link between pension contributions and future benefits is 50 percent note: see notes to figure 2. while the example is necessarily somewhat arbitrary, figure 3 shows that actual top marginal tax rates depend greatly about the perceived status of ssc. in this example, if only half of pension contributions are perceived as a tax, the main conclusions of this analysis are changed significantly. first, in this case top tax rates in hungary have never reached the revenue-maximising tax rate. second, the top marginal tax rate in 2013 is below 50 percent, even further from theoretical benchmarks than under the baseline assumptions. tax evasion, tax avoidance, long-term elasticities and international tax competition the logic of the results presented in this paper can accommodate considerations related to tax evasion, tax avoidance, long-term effects of taxes on labour supply, or international tax competition. tax evasion. if high earners hide some of their income as a response to a tax increase, this will appear to the state as response at the intensive margin and will be taken into account just like a real economic response. tax avoidance. if taxpayers ‘relabel’ some of their income after a tax change, then this means that the loss in one tax revenue will be partly made up by another tax revenue. this means that the total tax elasticity might be lower than the european journal of government and economics 2(2) 116 elasticity measured by estimations that focus on a single source of revenue. in hungary there is no strong evidence for such cross-tax effects although they may exist. long-term elasticities. possibly, some types of adjustment by taxpayers take time. lower top marginal tax rates may imply different career decisions by some individuals or even different educational decisions. there is no strong evidence about these effects either way, but they could lead one to assume that long-term elasticities are greater than short-term ones researchers have estimated so far. international tax competition. if low taxes attract the business of some high-addedvalue activity (e.g., finance), this will appear to the government as a response of the tax base to changes in the tax rate. in theoretical terms this should be considered as adjustment at the extensive margin. strong international competition effects would imply an increase in the relevant elasticity to be taken into account in the analysis. for hungary (or even internationally) evidence for this kind of tax competition is scarce (but see, e.g., recent examples in somewhat special contexts by kleven et al. [2010 and 2013]). conclusion the theory of optimal income taxation was long considered an esoteric field with little to say about actual tax policy. this view changed with the developments of the last 15 years. although it is not to be expected that the theory enables us to derive ‘the optimal tax system’, we can arrive at some robust qualitative results and in some cases even quantitative guidance for tax policy. this paper applies one recent result of the optimal theory of income taxation to the hungarian case: it suggests that the optimal top marginal tax rate is likely to be higher than the current actual rate. the theory requires surprisingly few assumptions, and only two estimable parameters, to derive a value for the revenue-maximising top marginal tax rate. this tax rate is optimal if the value society attaches to an additional dollar kept by a top earner is negligible. if this ‘social marginal value of income’ for high earners (parameter g) is not negligible, it can be considered as a parameter while the optimal top marginal tax rates can be calculated as a function of this parameter. according to simulations in this paper, the revenue-maximising top marginal tax rate is about 67 percent in hungary (including ssc and consumption taxes), consistent with a pit of about 30 percent (or 40 percent if pension contributions were capped). the actual top marginal pit rate is 16 percent in 2013. the actual rate is lower than the optimal top marginal tax rate under plausible values of parameter g. a number of factors are discussed that could qualify or modify the baseline results. notably, the optimal top pit rate is higher than in the baseline results if the income effect is significant or if pension contributions have a strong link to future pension benefits. in contrast, the optimal top pit rate is lower than in the baseline results if the long-run elasticity of taxable income is greater than the short-run elasticity about which evidence is available, or if international competition for talent is a binding constraint for the social planner. references atkinson. anthony b. (1990) ‘public economics and the economic public’, european economic review 34 (7): 225-248. bach, stefan; corneo, giacomo and viktor steiner (2012) ‘optimal top marginal tax rates under income splitting for couples’, european economic review 56(6): 10551069. kiss ● the optimal top marginal tax rate 117 bakos, péter; benczúr, péter and dóra benedek (2008) ‘the elasticity of taxable income: estimates and flat tax predictions using the hungarian tax changes in 2005’, mnb working paper 2008 no.7. benczúr, péter; kátay, gábor; kiss, áron; reizer, balázs and szoboszlai mihály (2011) ‘analysis of changes in the tax and transfer system with a behavioral microsimulation model’, mnb bulletin, october 2011. benczúr, péter; kátay, gábor and áron kiss (2012) ‘assessing changes of the hungarian tax and transfer system: a general equilibrium microsimulation approach’, mnb working paper 2012 no.7. benczúr, péter; kiss, áron and pálma mosberger (2013) ‘the elasticity of taxable income’, in k. fazekas, p. benczúr, and á. telegdy (eds) the hungarian labour market 2013, budapest: hungarian academy of science. brewer, mike; saez, emmanuel and andrew shephard (2010) ‘means-testing and tax rates on earnings’, in j. mirrlees, s. adam, t. besley, r. blundell, s. bond, r. chote, m. gammie, p. johnson, g. myles and j. poterba (eds) dimensions of tax design: the mirrlees review, oxford university press, pp. 90-173. diamond, peter (1998) ‘optimal income taxation: an example with a u-shaped pattern of optimal marginal tax rates’, american economic review 61(1): 8-27. diamond, peter and emmanuel saez (2011) ‘the case for a progressive tax: from basic research to policy recommendations’, journal of economic perspectives 25(4): 165-190. feldstein, martin (1995) ‘the effect of marginal tax rates on taxable income: a panel study of the 1986 tax reform act’, journal of political economy 103(3): 551-572. giertz, seth h. (2004) ‘recent literature on taxable-income elasticities’, technical paper series, congressional budget office, washington d.c. goolsbee, austan (2000) ‘what happens when you tax the rich? evidence from executive compensation’, journal of political economy 208(2): 352-378. gruber, jonathan and emmanuel saez (2002) ‘the elasticity of taxable income: evidence and implications’, journal of public economics 84: 1-32. kiss, áron (2013) ‘developments in the theory of optimal income taxation with applications to the hungarian tax system’, discussion papers of the institute of economics, centre for economic and regional studies, hungarian academy of sciences, budapest. kiss, áron and pálma mosberger (2011) ‘the elasticity of taxable income of high earners: evidence from hungary’, mnb working paper 2011 no. 11. kleven, henrik j.; landais, camille and emanuel saez (2010) ‘taxation and international migration of superstars: evidence from the european football market’, nber working paper no. 16545. kleven, henrik j.; landais, camille; saez, emanuel and esben a. schultz (2013) ‘taxation and international migration of top earners: evidence from the foreigner tax scheme in denmark’, nber working paper no. 18885. mankiw, n. gregory; weinzierl, matthew c. and danny yagan (2009) ‘optimal taxation in theory and practice’, journal of economic perspectives 23(4): 147-74. mirrlees, james a. (1971) ‘an exploration in the theory of optimal income taxation’, review of economic studies 38(2): 175-208. riihelä, marja; sullström, risto and matti tuomala (2013) ‘top incomes and top tax rates: implications for optimal taxation of top incomes in finland’, tampere economic working paper series no. 88. european journal of government and economics 2(2) 118 saez, emmanuel (2001) ‘using elasticities to derive optimal income tax rates’, review of economic studies 68(1): 205-229. saez, emmanuel; slemrod, joel b. and seth h. giertz (2012) ‘the elasticity of taxable income with respect to marginal tax rates: a critical review’, journal of economic literature 50(1): 3-50. appendix table a1. actual top tax rates in hungary, 2005-2013 2013* 2012* 2011* 2010 2008 2005 top pit rate 16.0% 20.3% 20.3% 40.6% 40.0% 36.0% employee pension contributions 10.0% 10.0% 10.0% 9.5% 9.5% 8.5% employee health contributions 8.5% 8.5% 7.5% 7.5% 6.0% 4.0% employer contributions 27.0% 27.0% 27.0% 27.0% 32.0% 32.0% effective consumption tax ratea 15.9% 15.9% 15.9% 15.9% 15.1% 14.8% actual top tax rate (above pension contrib. ceiling)b 56.6% 52.9% 52.2% 65.7% 65.3% 61.3% actual top tax rate (below pension contrib. ceiling) 56.6% 59.5% 58.8% 72.0% 68.8% 66.8% theoretical optimal top tax rate (g = 0) 66.7% 66.7% 66.7% 66.7% 66.7% 66.7% a the effective tax rate on consumption was calculated as tax revenue from consumption taxes (vat, excise taxes and other, minor taxes) divided by total household consumption. for the years 2011 to 2013 the empirical effective consumption tax was not known at the time of writing. thus for these years the value for 2010 was used. b until 2013, high-income individuals did not pay employee pension contributions on income exceeding the pension contribution ceiling (employer contributions have not been capped). thus here only the health-care contributions were accounted on the employee side until 2013. the employee pension contribution ceiling was abolished effective 2013. figure a1. marginal effective tax rates (metr) at high incomes in hungary, 2005— 2013 note: marginal effective tax rates in this figure are consistent with the calculations of the rest of the paper: they take into account social security contributions as well as the effective consumption tax rate. european journal of government and economics volume 2, number 1 (june 2013) issn: 2254-7088 79 social capital in spain: are there gender inequalities? celia muñoz-goy, university of a coruña, spain abstract social capital is an asset for individuals because it grants access to important resources embedded in their social networks. but social capital is not evenly distributed among different groups. gender groups are analysed in this paper in order to examine if there are differences in diverse indicators of social capital and whether these possible differences remain when control variables are considered. the data used in this paper come from a representative sample of 3,400 people in spain. the main results show gender differences in the access, mobilisation and type of social networks, as well as in the extent and type of social participation. however, these differences are mostly reduced for the groups in more advantageous social positions, which have the possibility to contact with greater and more varied groups, or which have been educated in less traditional gender roles. in general, gender inequalities in social capital remain for the other groups. jel classification z13; j16 keywords social capital; social networks; civic participation; gender inequalities muñoz-goy ● social capital in spain 80 introduction over the last few years, there has been a growing interest in social capital in the fields of sociology and economics. some of the more salient authors have considered social capital either as potential or actual resources linked to the presence in networks of relationships (bourdieu, 1985) or as aspects of the social structure that facilitate certain actions of the individuals within that structure (coleman, 1988) or as features of the social organisation, that is to say networks, norms and social trust, which facilitate connections among individuals as well as coordination and cooperation for mutual benefits (putnam, 1995). even if there still is an open debate about the definition and measurement of this wide concept (dasgupta and serageldin, 2001), social capital can be understood as a resource available to individuals derived from their location on a structure of social relations (adler and kwon, 2002: 18). the importance of this resource is that it improves the chances of obtaining a very wide variety of benefits, ranging from instrumental rewards, such as faster promotions or better jobs, to expressive rewards such as better mental health and even an increased subjective feeling of well-being or life satisfaction (adler and kwon, 2002; lin, 2000; kroll, 2011). but access to social capital and the benefits derived therein is not evenly distributed among all individuals or social groups. as lin (2000: 786-787) explains, depending on the historical processes and institutional construction of societies, some groups defined by race, gender, religion, caste, and other ascribed or constructed characteristics, are provided with unequal opportunities compared with the members of other social groups. and this phenomenon is more prevalent when combined with another principle, named homophily, which points out that individuals tend to interact more frequently with members in their own group, or in groups with similar socioeconomic characteristics. as a consequence of this tendency, members of disadvantaged social groups will share a lower variety of information and influence than members of resource-rich networks. it is precisely women who make up a social group with a disadvantaged socioeconomic status compared to men, and who tend to establish links with other peers or similar groups. as follows from the work of gwen moore (1990), there are differences in the composition of male and female social networks in the united states in the 1980’s. although many of these differences decrease when controlling variables related to employment status, family and age, some variations still remain, especially regarding the type of social networks, because women tend to connect mainly with the family, while men extend their networks of relationships beyond parental ties. maxine molyneux (2008: 68) has also found gender differences, shown by the fact that women’s social capital is located more so at home rather than in the public environment of the workplace, and involves exchanges of time and money as opposed to capabilities. it includes a significant proportion of voluntary work and undertaking the role of a carer and usually involves affective or ethical issues, a degree of altruism and often mobilises feelings associated with motherhood. so, this social capital creates strong ties (bonding social capital) more so than weak links (bridging social capital). these differences imply that women's networks are located in areas different to those of men’s capital. while men’s networks are traditionally associated with the public world of the workplace and politics, women’s networks focus more on solving domestic problems and less on economic and employment issues. this situation reflects the fact that, despite the changes that have occurred in recent decades, women are still in an unequal power situation, and thus continue to carry out the majority of family duties, having fewer resources and less power than men. other studies note the gender inequality in relation to the mobilisation of social capital, or the potential use of the resources embedded in the network. as pointed muñoz-goy ● social capital in spain 81 out by lin (2000: 791), in the specific area of the labour market, men have greater advantages than women, either because women do not use or do not mobilise the adequate social capital, or because they mobilise the adequate social links but these links are reluctant to invest in their favour, or even because there are different responses from the labour market itself depending on whether candidates sought for a job or promotion are male or female. timberlake (2005: 43) also concludes in his analysis that women are less able to mobilise social capital to get better jobs and seek promotion in organisations where they work. furthermore, pippa norris and ronald inglehart (2006) analyse the participation in community life, as an indicator of social capital, since this participation facilitates the establishment of networks and the achievement of individual benefits, such as opportunities in the workplace and the availability of support networks that encourage collaboration channels to solve local problems. these authors distinguish two types of inequality on civic engagement: horizontal segmentation and vertical segmentation. vertical segmentation refers to the differences in the intensity of organisational participation of men and women, while horizontal segmentation takes into account the differences in the type of organisations to which men and women belong. it is in this latter type of horizontal segmentation that norris and inglehart (2006) have found differences according to gender, in the u.s. and in most countries they have analysed. on the one hand, male participation is clearly superior in recreational associations, unions, political parties and professional associations, while, on the other hand, women tend to participate more in voluntary associations related with education, religious organisations and groups devoted to help the disabled or the elderly, as well as women's groups. regarding vertical segmentation, men participate in organisations to a higher degree than women, even when controlling variables related to structural factors (age, education) and cultural values (ideology, religion), but vanish when controlling agency factors, or the impact of relationships with others who can encourage participation (time spent with family, friends or co-workers). as shown by the previous studies, there are social capital inequalities according to gender, although some differences may be reduced or even disappear if certain control variables are taken into account. this research topic has a theoretical interest since it elucidates the relationships between social capital and structural and socioeconomic factors. it also has a social interest because knowledge in this area can contribute to the decision-making in policies to reduce gender inequalities. in this paper, i analyse individual social capital inequalities according to gender in spain, to test if significant differences occur and whether these differences are maintained or not in the presence of several control variables. the control variables -education level, age, occupational status, employment status, marital status and having or not dependantsreflect several structural aspects that could affect civic engagement and the establishment of social networks, promoting or restraining the availability of a range of resources such as time, money, knowledge and skills. to carry out this analysis i consider whether there are gender differences in several social capital indicators: network variety, possible use or mobilisation of resources, type of network and social participation. in a more detailed manner, i test the following hypotheses: h1. the more advantaged gender group (men) will have more social capital than the less advantaged gender group (women): h1.1. women are expected to have less network variety than men. h1.2. women are expected to have fewer possibilities of use or mobilisation of resources from their networks than men. muñoz-goy ● social capital in spain 82 h2. gender groups are expected to differ on the type of networks: women are expected to have more internal or family link while men are expected to connect with more external groups. h3. gender groups are expected to have different patterns of social participation: h3.1. women are expected to have lower levels of social participation than men (vertical segmentation). h3.2. women and men are expected to participate in distinctive types of social organisations (horizontal segmentation). after testing these hypotheses for the whole sample under analysis, i create separate subgroups, using the aforementioned control variables, to confirm whether or not the expected relationships between gender and social capital indicators remain when considering the effect of structural factors. data and measures in this paper i use the data base “social capital and inequality in spain”, from the research group osim (social organisations, institutions and markets) of the university of a coruña, spain. data collection took place between november and december of 2011, with telephone interviews (cati) to 3,400 people aged 18 or over, residing in spain. the total error estimated for this sample size is ± 1.7% for global data, assuming a confidence level of 95.5% and p=q. the sample was stratified according to autonomous communities (the spanish regions) using a mixed procedure: half of the sample size was assigned proportionally according to population size of the autonomous community, and the other half of the sample size was divided disproportionally to ensure a minimum of 100 interviews in each one of the 17 autonomous communities. in each autonomous community, there was a proportional distribution of the sample according to the size of municipalities. weighting based on the national distribution of gender, age groups and size of autonomous communities, was introduced to adjust sampling data. measures of social capital not only is there a plethora of definitions of social capital, but there are also many different ways of measuring the concept. in this paper i analyse various indicators of individual social capital, accounting for diverse aspects that may facilitate a person attaining the resources embedded in its net of social relations. as stated in the hypotheses, our interest lies in the variety and potential use of networks, the type of networks and the social participation. the variety of network, or accessibility, considered as the diversity of contacts that a person can access, is an indicator of social capital, since the larger the network size the greater the likelihood of achieving various resources. accessibility or the variety of network was measured using the data of a question with 14 items. respondents were asked to report if among their family, friends or acquaintances, there was someone with certain characteristics included in a list of the survey as shown in table 1. should the respondent answer affirmatively that he or she has a relative with one of the mentioned characteristics, the response is collected not recording whether the person also has friends and/or acquaintances with those characteristics. to summarise the information of this question we added up the number of different characteristics each respondent knows, regardless of whether they are relatives, friends or acquaintances, and expressed such amount in percentages. muñoz-goy ● social capital in spain 83 table 1. list of items included in the measures of variety of network (colum a) and the network mobilisation (colum b) variety of network: is there anyone, among your relatives, friends or acquaintances, with the following characteristics? network mobilisation: is there anyone, among your relatives, friends or acquaintances, who could help you out in the following situations? an university professor finding a job for a family member a person working in the administration of the education system giving advice in a labour dispute a person working in the legal system helping to move a person working in the tax office giving advice or help in educational matters a person working in local government (but not as a politician) helping with the shopping when is sick a person working in the banking system giving medical advice when the respondent is dissatisfied with a doctor a doctor lending a large amount of money (e.g. € 5,000) a person working in the health service (but not a doctor) lending a place to live while the respondent cannot use his or her home a policeman/policewoman giving advice on tax issues a person holding a political office giving advice on financial issues a journalist or someone having influence on the media giving advice on legal matters a person having the opportunity to hire people providing good references when looking for a job a person earning over € 3,000 a month helping to take care of children a person having more than one million euros, or that the respondent thinks is a very rich person helping to take care of the disabled or the elderly source: osim survey on social capital in spain, 2011 another indicator of social capital is network mobilisation, understood as the possibility of effectively using the social contacts to obtain resources embedded in the net. mobilisation was measured using the data of a question with 14 items. respondents were asked to report if among their family, friends or acquaintances, there was someone that might help him or her in different situations as shown in table 1. the situations of the list are hypothetical ones, meaning that there is no need for the respondent to be actually in the situation, but must answer as if the situation could happen, so, for example, people without children or elderly to their charge are asked to answer as if he or she would be in such case. to summarise the information of this question, we proceeded in the same way as explained for the variety of network or accessibility. analyzing social capital, as aforementioned, different contributions (moore, 1990; molyneux, 2008) found differences according to gender in the type of social networks, in the sense that men’s networks tend to be extra parental or placed in a public environment (bridging social capital) while women’s networks tend to be more focused on parental or domestic areas (bonding social capital). to substantiate these claims, i analyse the information from the question about the variety of social network, creating three variables to express in percentages the type of network composition, taking into account whether the contacts mentioned by the respondents are mainly focused on family, friends or acquaintances. finally, participation in community life is often used as an indicator of social capital, since such participation facilitates the establishment of networks. specifically, robert putnam (2000) uses the downward trend of civic participation in the united states as an indicator of the decline of social capital in that country. to measure social participation, i use the information from the question based on the voluntary organisations to which respondents belong, from a total of eight different categories: religious organisations; sports organisations; art, music or educational organisations; trade unions; political parties; associations related to ecology and environmental associations; professional associations; ngos (social welfare, health, human rights, humanitarian). this information was treated from a double perspective, to analyse the vertical and horizontal segmentation. as a measure of vertical segmentation, we added up the total number of organisations to which muñoz-goy ● social capital in spain 84 each respondent belongs, and so as to explore horizontal segmentation i compare the specific type of organisations to which men and women adhere. to test the hypotheses outlined in the previous section, i compare the summary measures of the different indicators of social capital by gender, using an analysis of variance (anova) or a chi-square test, depending on whether the summary is a mean or a percentage. after this first comparison, i shall explore whether or not gender differences are still significant among groups formed by introducing the control variables education level, age, occupational status, employment status, marital status and having or not dependants. some of these variables have been recoded to avoid comparisons among an excessive number of categories and low sampling sizes. the distributions of sample sizes for independent variables are shown in table 2. as for education, the group named as low level corresponds to primary education or less, the group named as moderate level corresponds to secondary education and vocational training, and the group named as high level corresponds to university-level education. a special comment deserves the grouping of age; we have decided to use three groups in line with other research strategies and the youngest group (under 35) corresponds to the grouping of the original groups of 18-24 and 25-34. regarding the variable about having dependants or not, there is no distinction in the original wording of the questionnaire among those having children and those taking care of elderly or other dependent adults. table 2. description of the sample by independent variables women men all 1736 1663 high 548 556 moderate 772 814 education low 399 282 under 35 479 502 35-64 855 864 age groups 65 or more 401 297 manager/professional 123 197 middle range supervisors 196 202 skilled working 198 387 occupational status unskilled working 204 123 in paid work 698 894 work status not in paid work 1038 769 married or cohabiting 1021 1026 marital status single 384 508 yes 581 543 dependants no 1155 1120 source: author, based on data from osim survey on social capital in spain, 2011. data analysis accessibility and gender results for the network variety show gender differences as expected and in line with previous analyses (table 3). there are significant differences in the mean of accessibility by gender, supporting the hypothesis that states that men have more diversified social networks than women. however, the network variety differences by gender are modified in some cases when controlling other structural factors. the level of education has an evident impact on network variety, since the higher the level of education, the greater the accessibility to a more diversified social network for both men and women, in accordance with previous analysis that established the crucial role of education in social capital generation (huanga, et al. 2009). but when controlling this variable, we found that among people with a higher level of education, gender differences are not statistically significant, while at the other levels of education, gender inequity in social capital persists. muñoz-goy ● social capital in spain 85 in terms of age, there is no such direct relationship with network accessibility as in the case of the level of education, but we can see that such accessibility is lower among the elderly than among the younger and middle age group. we also observe that among younger people there are no gender differences in the size of the social network. table 3. mean of accessibility according to gender and control variables women men difference sig.1 all 32.93 36.79 -3.86 *** high 45.43 47.21 -1.79 n.s. moderate 31.21 34.86 -3.64 ** education low 19.57 22.93 -3.36 * under 35 36.36 37.50 -1.14 n.s. 35-64 36.29 39.29 -3.00 * age groups 65 or more 21.71 28.50 -6.79 *** manager/profe ssional 49.21 49.29 -0.07 n.s. middle range supervisors 44.43 46.07 -1.64 n.s. skilled working 33.07 30.36 2.71 n.s. occupational status unskilled working 20.14 25.86 -5.71 * in paid work 41.21 41.07 0.14 n.s. work status not in paid work 27.43 31.86 -4.43 *** married or cohabiting 33.71 38.64 -4.93 *** marital status single 37.14 34.00 3.14 n.s. yes 38.21 41.64 -3.43 * dependants no 30.29 34.43 -4.14 *** source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance of anova: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 higher occupational status seems related to more network variety, but it is also important to note that in most of the groups defined by occupation no gender differences were found. the only exception appears in the group of unskilled workers, where men have significantly higher network accessibility than women. with reference to the labour market, people having paid work show higher accessibility means than unemployed people. it is important to highlight the fact that the average social network variability is almost equal between men and women who are in paid employment, suggesting that either being employed helps people to expand their network and makes gender differences vanish, or it is precisely those women with high social capital who are part of the labour market and for that reason gender inequality in network accessibility disappears. in terms of marital status, married men have higher accessibility than married women, while among those who are single there is a slight difference in favour of women although such difference is not statistically significant. in other words, accessibility differences by gender vanish among single people. moreover, people who have children or dependants have higher network accessibility than people with no dependants, but gender differences continue to be significant in both groups, showing more variety of social networks for men than for women. in summary, the network accessibility differences according to gender are significant in all groups except among people with a higher level of education and those who are young, single and with a medium or high level of occupational status. the disappearance of gender differences in these cases could be due to the fact that it is precisely the younger women who have higher levels of education and are therefore more likely to join the labour market -even at intermediate or higher level occupationsthat gives them access to a public sphere where they can muñoz-goy ● social capital in spain 86 establish a wider range of contacts. in this way, they overcome the phenomenon of homophily described by lin (2000), and so they participate in the conditions in which men have traditionally established their social networks. mobilisation of social network and gender as shown in table 4, men have a higher average of resource mobilisation than women, but such differences are no longer significant for some groups formed when controlling structural factors. table 4. mean of network mobilization according to gender and control variables women men difference sig.1 all 58.71 63.21 -4.50 *** high 71.86 73.50 -1.64 n.s. moderate 59.64 63.21 -3.57 * education low 39.71 44.21 -4.50 * under 35 71.50 70.64 0.86 n.s. 35-64 60.57 65.21 -4.64 ** age groups 65 or more 39.57 44.79 -5.21 * manager/profe ssional 74.43 74.00 0.43 n.s. middle range supervisors 68.21 68.93 -0.71 n.s. skilled working 61.50 56.07 5.43 * occupational status unskilled working 47.36 54.07 -6.71 * in paid work 67.86 69.21 -1.36 n.s. work status not in paid work 52.57 56.21 -3.64 ** married or cohabiting 58.93 62.93 -4.00 ** marital status single 69.00 66.36 2.64 n.s. yes 64.64 65.29 -0.64 n.s. dependants no 55.79 62.21 -6.43 *** source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance of anova: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 the higher the level of education, the greater is the mobilisation of resources. and it is precisely in the group of people with a high level of education where gender differences vanish. the same happens with age; the mobilisation of resources in the network is higher when we consider younger groups, and among the youngest people there are no differences by gender in this indicator of individual social capital. the mobilisation of resources in the network is higher as we move to upper positions in the occupational status scale. and for the groups of managers/professionals and middle range supervisors the differences of mobilisation between men and women are not statistically significant. people in paid work have more possibilities to use their network resources in comparison to unemployed people. among the first group there are no clear differences according to gender, showing that access to the labour market opens up the possibility to establish contacts that will expand the available span of resources. the mobilisation of resources in the network is lower for married people than for single people and among the latter there are no significant gender differences on mobilisation. the group of people having dependants show a greater resource mobilisation than the group without dependants, which may reflect that the need to take care of others leads to generating a stronger network where to extract resources from. among those with dependants, no differences by gender were found in the mobilisation of the network. muñoz-goy ● social capital in spain 87 in summary, in general terms, the findings on the relationship of the mobilisation of resources in the network and gender are similar and consistent with those for network accessibility: gender differences are not significant among those with higher levels of education, those who are young, single and with high or medium occupational status, yet it remains significant in the other cases. type of network and gender the analysis of type of network (table 5) demonstrates that links with family members are more common for women than for men, but the difference is not significant. furthermore, considering the groups formed in accordance with the control variables, we also observe that, in most of the cases, women have higher rates of family networks than men, but there are practically no cases with statistically significant differences. the only exception is the group of single people. distinct results are found for the network of friends as notable differences are appreciated. women have a lower percentage of network composition based on friends than men. and this difference in favour of men holds and remains significant in the presence of most of the control variables. however, gender inequality vanishes in some cases that coincide with those that were mentioned when analyzing network accessibility, i.e. those groups with a higher level of education, youngest in age, in a high or medium occupational status, who have paid work and who are single people. in addition, there are also no differences in the network of friends among people who have to take care of children or dependants. muñoz-goy ● social capital in spain 88 table 5. mean of type of network according to gender and control variables family network friends network acquaintance network w om en m en d if. si g1 w om en m en d if. si g1 w om en m en d if. si g1 al l 15.22 14.73 0.49 n.s. 11.40 14.15 -2.75 *** 6.33 7.93 -1.60 *** h ig h 19.15 19.28 -0.14 n.s. 18.68 18.74 -0.05 n.s. 7.63 9.23 -1.60 n.s. m od er at e 14.75 13.60 1.16 n.s. 10.29 13.91 -3.62 *** 6.20 7.35 -1.15 n.s. ed uc at io n lo w 11.01 9.28 1.73 n.s. 3.80 6.38 -2.59 ** 4.77 7.30 -2.53 ** u nd er 3 5 15.28 14.67 0.61 n.s. 14.48 15.01 -0.53 n.s. 6.59 7.82 -1.23 n.s. 35 -6 4 16.03 15.35 0.68 n.s. 13.30 15.66 -2.36 ** 6.99 8.26 -1.28 n.s. ag e gr ou ps 65 o r m or e 13.41 13.03 0.38 n.s. 3.68 8.28 -4.61 *** 4.63 7.17 -2.54 ** m an ag er / pr of es si on al 18.03 19.80 -1.77 n.s. 24.00 21.11 2.88 n.s. 7.20 8.34 -1.13 n.s. m id dl e ra ng e su pe rv is or s 20.32 19.08 1.24 n.s. 17.77 16.99 0.78 n.s. 6.32 10.02 -3.70 ** o cc up at io na l s ta tu s sk ille d w or ki ng 14.60 12.60 2.00 n.s. 10.76 10.97 -0.20 n.s. 7.70 6.75 0.95 n.s. muñoz-goy ● social capital in spain 89 u ns ki lle d w or ki ng 9.82 10.49 -0.67 n.s. 5.92 10.29 -4.37 ** 4.43 5.09 -0.66 n.s. in p ai d w or k 16.49 15.91 0.58 n.s. 17.27 17.31 -0.03 n.s. 7.43 7.85 -0.42 n.s. w or k st at us n ot in p ai d w or k 14.36 13.35 1.00 n.s. 7.45 10.48 -3.03 *** 5.59 8.03 -2.44 *** m ar rie d or co ha bi tin g 15.93 15.81 0.13 n.s. 11.50 14.15 -2.65 *** 6.29 8.68 -2.39 *** m ar ita l s ta tu s si ng le 15.74 12.95 2.80 * 14.09 14.49 -0.40 n.s. 7.28 6.58 0.69 n.s. ye s 16.51 16.21 0.31 n.s. 14.82 16.65 -1.84 n.s. 6.90 8.81 -1.92 * d ep en da nt s n o 14.56 14.01 0.55 n.s. 9.68 12.93 -3.25 *** 6.05 7.51 -1.46 ** source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance of anova: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 regarding the last type of networks, women have significantly fewer links with acquaintances than men. however, this difference is somehow modified by taking into account the control variables in a similar way as with the network of friends. in other words, in most of the cases it holds true that men have a network of acquaintances significantly greater than women but there are some exceptions. specifically, no significant differences appear in the network of acquaintances among people with higher or medium levels of education, those younger or middle aged, employed, and single. in relation with occupational status a slight variation was found, since in this case there are no significant differences in any of the groups defined by this variable, except for those who work as middle range supervisors. in summary, our data partially support the claim of moore (1990) that there are gender differences in the type of social network: the network of the family type is more common among women than among men, although in our data this difference is not statistically significant, and the networks based on friends and acquaintances are most common among men. furthermore, for these last two types of networks, although some differences in favour of men hold true in the presence of the control variables, in other cases similar to those mentioned when analyzing network accessibility or network mobility, these differences disappear. muñoz-goy ● social capital in spain 90 social participation and gender the first salient result for this indicator of social capital is that, according to the data obtained in spain, participation levels are low (table 6) in comparison with the results obtained from the world values survey, 2001 (norris and inglehart, 2006: 86) for post-industrial societies. on average the number of organisations to which both men and women belong in spain is inferior to one, because more than half of interviewees claimed not to be members of any of the organisations included in the survey list. regarding our hypothesis, data show the existence of vertical segmentation in spain, since there is a significant difference in the mean of membership to voluntary organisations in favour of men. moreover, once the various control variables are taken into account, in almost all cases women have a lower average of civic participation than men and most of these differences are significant, with the exceptions of the groups with low education, an older age, intermediate and lower occupational status, which are those with lower levels of participation. that is, men generally have higher levels of civic participation than women and, unlike the data observed with regard to network accessibility or network mobility, the effect of structural factors do not follow the same direction and have less power in diminishing gender differences. table 6. mean of organisational membership according to gender and control variables women men difference sig.1 all 0.75 0.94 -0.19 *** high 1.12 1.29 -0.17 * moderate 0.61 0.87 -0.26 *** education low 0.49 0.47 0.02 n.s under 35 0.61 0.89 -0.28 *** 35-64 0.87 1.04 -0.17 ** age groups 65 or more 0.64 0.74 -0.10 n.s manager/profe ssional 1.21 1.61 -0.40 * middle range supervisors 1.25 1.32 -0.07 n.s skilled working 0.64 0.78 -0.14 n.s occupational status unskilled working 0.48 0.63 -0.15 n.s in paid work 0.96 1.10 -0.14 * work status not in paid work 0.60 0.76 -0.16 ** married or cohabiting 0.81 1.02 -0.21 *** marital status single 0.65 0.80 -0.15 * yes 0.91 1.09 -0.18 ** dependants no 0.67 0.87 -0.20 *** source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance of anova: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 one possible explanation for differences in the level of participation could be found in the type of organisations included in the list because of the so called horizontal segmentation that is reflected in the gender inequality of the membership of different kinds of organisations. as shown in table 7, in most organisations there are significant differences between the percentages of male and female membership. on the one hand, there is a negative balance for women in trade unions, professional associations, associations related to ecology and the environment and, especially, in sports organisations. on the other hand, the balance for women is positive in religious muñoz-goy ● social capital in spain 91 organisations and humanitarian or charitable ngos. there are also two types of organisations, political parties and cultural and educational associations, in which, although the balance is negative for women, the difference is not significant. table 7. membership percentage of voluntary organisations according to gender women men difference sig.1 religious organisations 10.00 8.00 2.00 * sports organisations 8.30 19.80 -11.50 *** art. music or educational organisations 13.10 15.10 -2.00 n.s. trade unions 8.50 13.30 -4.80 *** political parties 3.90 4.10 -0.20 n.s. associations related to ecology and environmental associations 4.00 6.00 -2.00 ** professional associations 5.60 9.30 -3.70 *** ngos (social welfare. health. human rights. humanitarian) 19.90 17.30 2.60 * source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance chi-square: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 our results are consistent with those obtained by mcpherson and smith-lovin (1982), norris and inglehart (2006), lowndes (2006) and caiazza and gault (2006), which found that predominantly male organisations are those related with economic matters that offer opportunities for advancement, and also those related with sports. meanwhile women tend to adhere to organisations focused on the household, on community problems and helping others, as well as religious organisations, generally related to traditional feminine roles. nevertheless, we have to clarify some of the above differences. since certain types of organisations, in which the balance is negative for women, are clearly related to the labour market, employment status has to be taken into account as a control variable. in fact, it has been found that when considering only the people who are in paid employment, although there is still a slight negative balance for women’s membership in trade unions and professional organisations, such balance is no longer significant (table 8). regarding membership of associations related to the environment, ecology, animal rights and non-governmental organisations, gender differences diminish in intensity when considering control variables, for example, among younger people where these differences are no longer significant. gender differences in terms of membership of religious organisations among young people are neither significant, and the balance is even reversed, as the percentage of men belonging to these organisations is slightly higher than the percentage of women. these results seem to indicate that the changes in the last decades in the gender role expectations are being reflected in the new generations. table 8. membership percentage of voluntary organisations according to gender, among selected categories of several control variables women men difference sig.1 trade union membership among those with paid work 15.2 16.9 -1.7 n.s. professional associations membership among those with paid work 10.6 13.3 -2.7 n.s. membership of associations related to ecology and environmental associations among young people (under 35) 3.1 5.6 -2.5 n.s. religious organisations membership among young people (under 35) 4.6 6.6 -2.0 n.s. source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance chi-square: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 finally, in relation to sports organisations, it is noteworthy that gender differences in favour of men remain and are significant even when considering all the control muñoz-goy ● social capital in spain 92 variables (table 9) since men have, for all groups, higher membership percentages to these organisations than women. neither the fact of participating or not in the labour market nor the level of education affects the different membership of men and women of these organisations; variables that in other cases have reduced gender inequality. table 9. membership percentage of sports organisations according to gender and control variables women men difference sig.1 all 8.30 19.80 -11.5 *** high 4.3 12.4 -8.1 *** moderate 7.5 19.4 -11.9 *** education low 12.2 23.9 -11.7 *** under 35 7.1 25.7 -18.6 *** 35-64 10.4 19.6 -9.2 *** age groups 65 or more 5.0 10.4 -5.4 *** manager/profession al 9.0 27.9 -18.9 *** middle range supervisors 13.7 26.6 -12.9 *** skilled working 6.6 15.2 -8.6 *** occupationa l status unskilled working 4.4 20.3 -15.9 *** in paid work 10.3 22.7 -12.4 *** work status not in paid work 6.9 16.4 -9.5 *** married or cohabiting 9.6 19.0 -9.4 *** marital status single 7.6 21.5 -13.9 *** yes 12.2 20.3 -8.1 *** dependants no 6.3 19.6 -13.3 *** source: author, based on data from osim survey on social capital in spain, 2011. 1 level of statistical significance chi-square: n.s. = not significant, *: p ≤ 0.05; **: p ≤ 0.01; ***: p ≤ 0.001 in summary, we have found that there is a horizontal segmentation in terms of civic participation in line with other studies, since the rates of participation of men and women vary according to the type of organisations under consideration. however, it is true that the segmentation is reduced in most cases by introducing control variables, except when considering membership of sport organisations. conclusions in this analysis of social capital in spain, we have found clear gender differences in diverse measures of social capital, supporting our hypotheses. as for the extension and use of social capital, there are gender inequalities, in favour of men on the accessibility and possible use or mobilisation of resources embedded in social networks. regarding the type of social capital, the family network density is slightly more common among women and networks based on friends and acquaintances are more common among men. civic participation is also higher for men than for women, and the differential types of organisations to which men and women belong show the existence of a gendered horizontal segmentation. however, the differences in accessibility, mobilisation, type of network and horizontal segmentation are affected by the introduction of structural factors. in this way, inequality between men and women disappears when we consider the groups in more advantageous social positions: that is the highly educated and those with higher occupational status; those in situations that allow contact with greater and more varied groups, which are related to the fact of having paid employment; or the younger age groups that have been educated in less traditional gender roles. in general, gender inequalities remain in all the other groups. it is only with regards to civic participation that control variables do not reduce gender differences in the same direction as in the above cases. one possible explanation for the persistence of this vertical segmentation could be related with the time restriction derived from the fact that women still continue to do most of the housework, a variable not muñoz-goy ● social capital in spain 93 included in the dataset analysed, since time is an unavoidable factor to participate in social organisations. the impact of the time distribution of housework by gender in the inequalities in social capital could open new lines of inquiry. future research may also focus on an in-depth multivariate analysis of the relationships found between social capital and various structural factors, as well as the extension of the analysis to the international comparison to confirm if results in other countries are similar to those found in spain. finally, we might wonder whether our findings of a lower level of social capital among women could also be due to the fact that part of women’s social capital is invisible to the standard measurements. perhaps we need to explore other forms of political participation and civic engagement, as noted by dieltlind stolle and michele micheletti (2006) when analyzing the behaviour, motivations and consumer decisions for political reasons, and explore other areas of relationships, as proposed by maxine molyneux (2008), since women tend to develop their links and networks in fields different to those of men. bibliography adler, paul s. and seok-woo kwon (2002) ‘social capital: prospects for a new concept’, academy of management review 27(1): 17-40. bourdieu, pierre (1985) ‘the forms of capital’, in john g. richardson (ed.) handbook of theory and research for the sociology of education. new york, ny: greenwood press, pp. 241-258. caiazza, amy and barbara gault (2006) ‘acting from the heart: values, social capital and women’s involvement in interfaith and environmental organizations’, in brenda o’neill and elisabeth gidengil (eds) gender and social capital. new york, ny: routledge, pp. 99-126. coleman, james (1988) ‘social capital in the creation of human capital’, american journal of sociology 94(supplement): 95-120. dasgupta, partha and ismail serageldin (eds.) (2001) social capital, a multifaceted approach. washington, d.c.: the world bank. huanga, jian, henriëtte maassen van den brinka and wim groota (2009) ‘a metaanalysis of the effect of education on social capital’, economics of education review 28(4): 454-464. kroll, christian (2011) ‘different things make different people happy: examining social capital and subjective well-being by gender and parental status’, social indicators research 104(1) 157-177. lin, nan (2000) ‘inequality in social capital’, contemporary sociology 29 (6): 785795. lowndes, vivien (2006) ‘it´s not what you’ve got, but what you do with it: women, social capital and political participation’, in brenda o’neill and elisabeth gidengil (eds) gender and social capital. new york, ny: routledge, pp. 213-240. mcpherson, j. miller and lynn smith-lovin (1982) ‘women and weak ties: differences by sex in the size of voluntary organizations’, american journal of sociology 87 (4): 883-904. molyneux, maxine (2008) ‘la política de desarrollo y la dimensión de género del capital social’, papeles 101: 64-79. moore, gwen (1990) ‘structural determinants of men´s and women’s personal networks’, american sociological review 55(5): 726-35. muñoz-goy ● social capital in spain 94 norris, pippa and inglehart, ronald (2006) ‘gendering social capital. bowling in women’s leagues?’, in brenda o’neill and elisabeth gidengil (eds) gender and social capital. new york, ny: routledge, pp. 73-98. putnam, robert d. (1995) ‘bowling alone: america’s declining social capital’, journal of democracy 6(1): 65-78. putnam, robert d. (2000) bowling alone: the collapse and revival of american community. new york, ny: simon and schuster. stolle, dietlind and michele micheletti (2006) ‘the gender gap reversed: political consumerism as a women-friendly form of civic and political engagement’ in brenda o’neill and elisabeth gidengil (eds) gender and social capital. new york, ny: routledge, pp. 45-72. timberlake, sharon (2005) ‘social capital and gender in the workplace’, journal of management development 25(1): 34-44. cyclical synchronization in the emu along the financial crisis: an interpretation of the conflicting signals european journal of government and economics volume 1, number 1 (june 2012) issn: 2254-7088 86 cyclical synchronization in the emu along the financial crisis: an interpretation of the conflicting signals josé ramón cancelo, university of a coruña, spain* abstract we analyze how cyclical synchronization in the emu evolved since the onset of the current financial crisis. the standard measures of cyclical correlation suggest that while the cycle of the euro area became more aligned with the cycles of other developed economies, the emu itself apparently entered into a phase of cyclical divergence. we show that as a matter of fact the bulk of the member states remained closely aligned, and the seeming decline in synchronization is due to a few countries decoupling from the euro area. next we present empirical evidence that the foundations that explain the evolution of the national cycles against the emu aggregate through the crisis were already latent in 2007. greece and ireland deviate from the general pattern, the former because of its loose fiscal policy all along the period 2000-2007, and the latter due to the flexibility of its labor market. jel classification e32; f42; f44 keywords international business cycles; euro area; cyclical convergence; asymmetric responses * address for correspondence: josé ramón cancelo, university of a coruña, faculty of economics and business, campus de elviña, 15071 a coruña, spain. e-mail : jramon@udc.es. doi: https://doi.org/10.17979/ejge.2012.1.1.4278 european journal of government and economics 1(1) 87 introduction cyclical synchronization has always been regarded as a primary indicator to evaluate the running of the european monetary union (emu). since its very formation, the euro area congregated a bunch of countries with marked structural differences, heterogeneous institutional settings, and high risks of experiencing asymmetric shocks or displaying asymmetric responses to common shocks. in spite of that, they agreed to surrender monetary and exchange rate policies to a central authority, and hence to losing control of key instruments to stabilize their economies. it was thought that such loss would be compensated by the endogenous mechanisms of convergence of the currency area, the power of national fiscal policies to offset idiosyncratic shocks, and the adoption of structural reforms that would make the weaker economies more competitive. the first years of the emu witnessed some fluctuations in cyclical synchronization, but the crisis that originated in 2007 casts serious doubts on whether the emu will manage to survive in its current form. for that reason, in this paper we examine how cyclical synchronization in the euro area evolved since the onset of the crisis. more specifically, we address three issues. first, we compare output comovements within the euro area to international developments, to gauge whether the emu displayed some particular behavior in the period. second, we investigate to what extent the observed decline of a broad indicator of intra-euro-area synchronization may be due to a small group of countries decoupling from the aggregate cycle, while the bulk of member states remains closely aligned, or even is increasingly in phase. finally, we seek to identify the determinants of the observed differences across member states, in order to figure out why each country performed the way it did. the paper is organized as follows. after this introduction, section 2 reviews the literature on currency areas, to make clear why cyclical synchronization is so important for a monetary union, and to highlight the factors that strengthen the convergence of the national cycles of its members. section three sketches the methodology for extracting cyclical components from observed data, and to measuring cyclical synchronization. section four reports the empirical findings and discusses their implications for the euro area performance in recent years. section 5 concludes. literature review the theoretical foundations of the emu lie on the theory of optimal currency areas (mundell, 1961). the benefits of joining a currency area are the lower costs for international transactions and the explicit commitment to building long-term relationships with the other member states, while the major drawback is the loss of monetary and exchange rate policies independence. the final balance is casespecific, and must be assessed in an empirical way. the aftermaths of giving up policy autonomy depend on the degree of asymmetry of the shocks and of the transmission channels. if the shocks are common and propagate in a symmetric way, all the members will be at the same cyclical phase and the centralized management of monetary policy will be fully efficient to stabilize the national economies. if, on the contrary, there are idiosyncratic shocks, or common shocks that propagate in an asymmetric way, national cycles will diverge and the common monetary policy will be rather different from the one that an independent, national authority would have adopted. therefore, cyclical synchronization is a central issue to gauge the success of a currency area. even though some authors expressed their concerns on the success of the euro area because of the low level of ex ante synchronization (eichengreen, 1990 and 1993), more recent research pointed out that the formation of a currency area european journal of government and economics 1(1) 88 fosters output comovements across its members (de grauwe and mongelli, 2005). according to this endogeneity hypothesis, cyclical profiles need not be identical when the area is launched, as its operation will promote synchronization. however, some minimum level of output comovements seems to be required at the beginning, to avoid the costs of imposing a common monetary policy to countries that are not in the same phase of the cycle at the first stages of the union. the literature has analyzed in full detail the factors that strengthen cyclical comovements, both from the theoretical perspective and its relevance for the euro area. the most mentioned are trade intensity, financial integration and convergence of macroeconomic policies. price rigidities and labor market regulations are also pointed out, as more flexible economies are better prepared to cope with asymmetric shocks. trade integration is usually considered to be the most prominent factor leading to increased cyclical comovements. nevertheless, the theoretical effect of an increase in trade on the comovement of output is ambiguous, as it depends on the type of goods that are traded. frenkel and rose (1998), see also rose (2000), claims that higher bilateral trade leads to more correlated cycles, on the assumption that the increase in trade is mainly intra-industry trade. if this is the case, any demand or industry shock will have similar effects on the cycles of both countries, and they will move together. but krugman (1991) observed that openness to trade induces higher specialization, and in the medium term the increase in trade is mainly of the inter-industry type. in that event, the same demand or sectoral shock will have asymmetric effects, and trade integration will result in lower correlations across countries. as the final effect is an empirical matter, there has been a lot of research to determine the net contribution of the observed growth in trade within the euro area. most studies find that the association is positive and significant, but there is still an open debate on the actual impact on trade intensity on cyclical synchronization, see the survey by de haan et al (2008). currency unions promote financial integration across members, but its outcome on cyclical synchronization is theoretically undetermined. demand effects arising from savings or investment decisions in one country will affect the real economy in other member states if the links across financial markets are strong. on the other hand, increased bilateral financial flows allow for more efficient allocation of capital, and stimulate specialization of production according to each country’s comparative advantages. if demand effects prevail, financial linkages will propagate any given shock throughout, but if the currency area results in more specialized economies, asymmetric effects will dominate and output comovements will be lower. in the case of the euro area, there is no empirical evidence of a negative impact of financial integration on cyclical convergence, but it is also difficult to find a strong positive relationship between financial linkages and output correlation (de haan et al 2008, dees and zorelli 2011). convergence in macroeconomic policies is seen as another major force to inducing higher synchronization. but, once again, its actual contribution to cyclical convergence must be assessed in each situation, and in the euro area it seems that the outcomes were not as satisfactory as expected. in the 2000s the common monetary policy allowed for inflation differences amongst the member states that translated into differences in real interest rates, in a self-reinforcing process that was one of the major sources of decoupling of some faster growing economies. fiscal policies, which were managed independently by national authorities, were far from optimal, as they tended to be pro-cyclical and fueled divergence between overheating countries and the rest (european commission, 2004 and 2006). finally, the role of flexible product and labor markets has been stressed in recent years. if prices and wages are flexible, changes in the cyclical conditions translate european journal of government and economics 1(1) 89 immediately into changes in competitiveness, reducing the need of large variations in the policy instruments, the duration of the adjustment, and the risk of overshooting. on the contrary, when the markets are rigid and prices are sticky, large drops in quantities are needed to compensate idiosyncratic shocks and the adjustment is much more costly (european commission 2004 and 2005, artis et al 2008). methodological framework in the empirical analysis we apply well established procedures to facilitate the comparison of our results to those previously reported in the literature. we adopt the deviation-cycle approach, where the cyclical component is defined as the deviation of the observed gdp from its long-run trend. it assumes a structural model of the form )1(cpy ttt  where yt is logged gdp at t, pt its long-run trend, and ct the cycle. to carry out such decomposition we use the hodrick-prescott (hp) filter, which estimates the trend as   )2()p()py(minargp t 3t 2 t 2 t 1t 2 tt t 1tt           where t is the sample size,  is the difference operator and  is a penalty parameter that determines the degree of smoothness of the trend; with quarterly data, as we shall consider in section 4, the usual practice is to set =1600. once pt is known, the cyclical component is derived from (1); by construction, it is expressed as a percentage deviation from trend gdp. the properties of the hp filter have been extensively documented, see kaiser and maravall (2001) for a thorough discussion. from the perspective of signal extraction theory, the hp procedure relies on a lowpass filter that separates a low frequency component (the trend) from mediumand short-frequency components (the cycle); the cutoff point is set by . recent research opts for extracting the cycle by using bandpass filters, that assume a structural model of the form )3(ucpy tttt  where ut captures oscillations with periods too short to be considered part of the business cycle. to estimate ct the cyclical band should be defined ex ante, and it is typically assumed to lie between 1.5 and 8 years. some examples of this approach are the bk filter (baxter and king, 1999), the cf filter (christiano and fitzgerald, 2003), and the bandpass variant of the hp filter (artis, 2003). the theoretical properties of the hp filter and the bandpass alternatives are quite different, but there is empirical evidence that the qualitative conclusions on cyclical synchronization are invariant to the specific procedure used to extract the cyclical component. although different methods lead to different cyclical signals for a given time series, the measures of comovement are similar as long as the same procedure to estimate the cycle is used for all the series (altavilla 2004, de haan et al 2008). once the cyclical components are known, the next step is to measure to what extent the cycles move in phase. a variety of approaches have been suggested: dispersion statistics, correlation coefficients, concordance indices, factor analysis, european journal of government and economics 1(1) 90 spectral analysis, etc. the most popular procedure consists of computing bivariate correlations to quantify pairwise relationships, and calculating their mean to get a summary measure. yet, to execute this procedure some practical aspects have to be addressed. the first issue is whether there exists a reference cycle that national developments should be compared to. there has been a long debate on the existence of a euro area cycle, see for instance the review by konstantakopoulou and tsionas (2010), and for the moment no firm conclusion has been achieved. in view of that, some authors do not consider any reference and measure global synchronization by taking the unweighted mean of all the possible pairwise correlations between the member states (gayer 2007, artis et al 2008). another approach consists of selecting a reference cycle in an empirical way, by defining its value at t as the median of all the national cycles at that moment (mink et al, 2007). in this paper we opted for taking the cycle of the euro area as the benchmark, because this is the reference that the european central bank tracks to set its policy instruments. another decision has to be made concerning the time span of observations used to computing the correlation coefficients, usually referred to as the size of the window. this is an extremely relevant issue, as the conclusions may be sensitive to the final choice (european commission, 2006). long windows are to be avoided when the focus is on the changes in the very final part of the sample, as they miss the most recent developments. short windows, however, are known to be more volatile, and to produce artificial drops in the correlations near the turning points of the cycles if the size of the window is less than the duration of the cycle. the empirical literature has considered window sizes ranging from three to ten years (papageorgiou et al, 2010). in particular, the european commission monitors cyclical convergence in the euro area by using windows of 4, 6 and 8 years (european commission 2004, 2006 and 2010). we opted for a window size of four years, which appears to be the shortest span free from spurious drops in correlations. once the coefficient is computed, the result is assigned to the last t of the time interval that was used in its calculation, so the correlation at t measures comovement in the four years up to that point. the bivariate correlations between each member state and the euro area were calculated over a series of rolling windows with a fixed length of four years, to track the change in output comovements over time. synchronization for the area as a whole at each t is measured by the mean value of all bivariate correlations for that moment. empirical results and discussion we use data on quarterly real gdp for the aggregate of the euro area, eleven member states (belgium, germany, ireland, greece, spain, france, italy, the netherlands, austria, portugal and finland), three members of the european union that did not join the currency area (denmark, sweden and the united kingdom), and three external countries (switzerland, japan and the usa). all the data are obtained from eurostat. following the usual practice in empirical studies, the emu aggregate consists of the 12 countries that formed the currency union up to december 2006. quarterly gdp for greece is only available from 2000:1 onwards, so the working sample starts at this date. previous data on the other countries were used to improve the estimates of the cyclical component at the beginning of the sample. the sample information ends in 2011:4, with the exceptions of ireland (2011:3), greece (2011:1) and portugal (2011:3). in these countries the missing quarters of 2011 were forecasted with the restriction that the annual figure should be equal to the european commission forecast for the whole year 2011. figure 1 displays some raw measures of synchronization from 2004:1 to 2011:4. the line for the euro area represents the mean values of the bivariate correlations european journal of government and economics 1(1) 91 between each member state and the area as whole, and captures the degree of intra-euro-area synchronization. the line for the countries of the european union that do not belong to the euro area is computed as the mean of the bivariate correlations of the three countries within this group (denmark, sweden and the united kingdom) with the emu aggregate, and gives information about how the cyclical synchronization within the european union has evolved. the third line reports the mean correlation of switzerland, japan and the usa with the euro area, and is aimed to reflect global trends in economic integration. taken together, the three lines give insight into whether the changes in output comovements within the emu are a particular development of the currency union, or part of a more general european or world pattern. the figure shows that all types of synchronization fell until 2005, and then started to increase up to 2008. the values for 2005 were computed with data for the period 2002-2005, which included the recession of the beginning of that decade and the early stages of the subsequent recovery. national cyclical developments were rather heterogeneous in that period, but became more homogeneous as the expansion consolidated. as expected, cyclical synchronization was higher within the euro area than in the european union or the developed world. the onset of the crisis brought about substantial changes. aside from some transitory, idiosyncratic responses that lowered the correlations dated at the second half of 2008, the contraction in international trade and the financial contagion caused international synchronization to be especially high in 2009 and 2010, to the extent that the mean correlations of the euro area with the rest of the european union and with external countries attained their highest values in the sample period. figure 1. mean cyclical correlations with the euro area for a 4-year rolling window 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 intra-euro-area with the rest of the eu with third countries at the same time, the emu looked as if it had entered into a phase of cyclical divergence. the mean correlation within the area fell and became smaller than the correlations of the euro area with the rest of the european union and the rest of the world. to confirm that the fall in the mean correlation of the euro area is not a spurious drop due to using a window that is too short, figure 2 plots the results for a 6-year rolling window. despite some changes attributable to the different time span, the figure confirms that the mean correlation for the members of the euro area is falling since 2008, and that in the final part of the sample it is below the two other lines. european journal of government and economics 1(1) 92 figure 2. mean cyclical correlations with the euro area for a 6-year rolling window 0.0 0.2 0.4 0.6 0.8 1.0 2006 2007 2008 2009 2010 2011 intra-euro-area with the rest of the eu with third countries figure 3 corroborates the apparent euro area divergence. at each t it depicts the standard deviation of the eleven bivariate correlations between the cycle of each member state and the emu cycle. an increase of the spread of the correlations is a sign of divergence (gayer, 2007), and it can be seen that the dispersion has been increasing since 2009. figure 3. standard deviation of the bivariate correlations of the cycles of the member states with the euro area cycle, for a 4-year rolling window 0.00 0.05 0.10 0.15 0.20 0.25 0.30 0.35 0.40 2004 2005 2006 2007 2008 2009 2010 2011 actually, the features sketched in figure 1 are surprising. first, because the european and international upsurges in synchronization suggest that the current crisis is driven by common shocks that propagate worldwide through symmetric transmission channels, and the intra-euro-area increasing divergence does not match in this context. second, because figure 1 indicates that cyclical synchronization is higher between the emu as a whole and other currency areas european journal of government and economics 1(1) 93 than inside the euro area, an empirical finding that is in opposition to the primary theoretical foundations of monetary unions. a likely explanation to bring together the opposed evolutions of the correlation coefficients in figure 1 is that the decline within the euro area is not a general feature of the emu as a whole, but a particular trait of a small group of member states. if this is the case, one would find large, stable correlations with the euro area aggregate for most members, together will a few countries that are decoupling quite abruptly from the euro area cycle. figure 4 depicts the sequence of rolling correlations of the cycle of each member state with the euro area aggregate from 2004:1 to 2011:4. figure 5 highlights the changes in the correlations through the crisis. for each country, it concentrates on the correlations with the emu dated at 2007:3 and 2011:4; the first set was computed with data for the period 2004:4 to 2007:3, while the second set merges the observations from 2008:1 to 2011:4. to complete the picture, it includes the six countries in the sample that do not belong to the emu, to give an account of the change in the correlations of the euro area with the rest of the european union and the rest of the world. figures 4 and 5 confirm the existence of different cyclical patterns amongst the emu members. the correlations for belgium, germany, france, italy, the netherlands, austria and finland are high and do not exhibit any trend. these countries were closely aligned in 2007 and, except for some minor, transitory decoupling in 2008, have remained approximately in the same situation all through the crisis. the other four member states exhibit remarkable changes. spain and portugal entered into a phase of growing desynchronization, to the extent that in 2011:4 the six correlations of the non-member countries with the euro area were higher than the portuguese, and four were higher than the spanish. greece has totally decoupled from the euro area. ireland evolved in the opposite direction, as its correlation picked up and has been growing steadily since 2009. european journal of government and economics 1(1) 94 figure 4. bivariate cyclical correlations between the member states and the emu, for a 4-year rolling window: whole sequence from the working sample belgium 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 germany 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 greece -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 ireland 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 spain 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 france 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 italy 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 netherlands 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 austria 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 portugal 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 finland 0.0 0.2 0.4 0.6 0.8 1.0 2004 2005 2006 2007 2008 2009 2010 2011 european journal of government and economics 1(1) 95 figure 5.bivariate correlations between the member states and the emu, for a 4year rolling window: summary for relevant dates 0.0 0.2 0.4 0.6 0.8 1.0 bel ger ire gre spa fra ita net aus por fin den swe uk swi usa jap 2007:3 2011:4 these results offer a tentative explanation to harmonize the conflicting signals in figure 1, namely increased international synchronization and growing intra-euroarea divergence. on the whole, large symmetric shocks, worldwide economic integration and international spillovers led to relative highs in international synchronization. but in a few members of the euro area the common shocks launched powerful asymmetric effects, that induced quite idiosyncratic fluctuations. it can be shown that the decline in the intra-emu mean correlation reported in figure 1 is explained for the most part by the contribution of the bivariate correlations between greece and the euro area. when the sequence of mean correlations is computed without the greek data, it is found that the output comovements within the emu have not reduced during the crisis, and that the intra-euro-area synchronization has been higher than the synchronization with the rest of the european union or the rest of the world. the decline in the spanish and portuguese correlations has some influence on the average, but their contribution is not big enough to induce a false impression of divergence for the area as a whole. as a final point, we investigate the factors that may explain why member states behave differently. for doing so we analyze the relationships between the change in the cyclical correlation of the members of the euro area between 2011:4 and 2007:3, and a variety of indicators. the influence of each indicator is assessed individually, as there are only eleven observations and some of them are outliers, as we discuss below, so there are not enough degrees of freedom for a joint appraisal. the variables were selected on the basis of the literature review in section 2, including some that capture latent imbalances and policy management to account for the convergence of macroeconomic policies. all in all, nine indicators are considered: one is for trade intensity with the euro area (12 countries). for each member state, trade intensity in goods is obtained as the sum of exports and imports for 2007 with that partner, expressed in % of gdp. one for financial openness, the net international investment position in % of gdp for 2007. one for labor market rigidity, version 2 of the oecd’s employment protection index for 2007. european journal of government and economics 1(1) 96 two for discretionary fiscal policy: the cyclically adjusted net lending of general government in % of gdp for 2007, to capture the fiscal stance just before the beginning of the crisis; and its sum over the period 2008-2010, as an indication on how national authorities managed fiscal policy in the recent past. one for the loss of competitiveness between 2000 and 2007, the real exchange effective rate computed with the gdp deflator. two for the private credit dynamics, the 3-year average for 2005-2007 of the private credit flow in % of gdp, and the non-consolidated debt of the private sector in % of gdp for 2007. one for excess national demand, the 3-year average for 2005-2007 of the current account balance in % of gdp. all the data were taken from the eurostat web with the exception of the employment protection index, that was obtained from the oecd. most indicators refer to 2007 or the period 2005-2007, as we are interested in determining whether the factors that led to changes in the output comovements for the period of the crisis were already latent before it started. besides, the use of lagged values of the indicators reduces the endogeneity bias (imbs 2004, de haan et al 2008), and simplifies testing whether the relationship is statistically significant. to illustrate the type of association we focus on, figure 6 plots the change in the cyclical correlations between 2011:4 and 2007:3 against the indicator of trade intensity with the euro area. to measure the direction and strength of this relationship we compute the correlation coefficient between the two variables and report its value in table 1, as well as the p-value that tests its significance. figure 6.change in the bivariate cyclical correlation between 2011:4 and 2007:3 against trade intensity with the euro area for 2007 aus bel ita spa gerfra net fin gre ire por -1.0 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0 20 40 60 80 100 120 trade intensity with the euro area in % of gdp c ha ng e of th e cy cl ic al c or re la ti on . simple inspection of figure 6 reveals that greece is an outlier that distorts the underlying relationship, and a more detailed analysis shows that ireland is another outlier. this is not a special trait of the trade indicator, but a general characteristic of the set of potential explanatory variables that will help to assess the idiosyncratic behavior of these two countries. because of that, for each indicator reported in table 1 three variants of the correlation coefficient were computed: with the data of all the member states in the sample (eleven countries), without greece (ten countries) and without greece and ireland (nine). european journal of government and economics 1(1) 97 table 1. correlation coefficients between the change in output comovements and its potential determinants indicator trade intensity with the euro area 0.32 (0.331) 0.21 (0.561) 0.39 (0.296) net international investment position 0.62 (0.043) 0.70 (0.025) 0.86 (0.003) employment protection -0.34 (0.305) -0.84 (0.002) -0.74 (0.024) cyclically adjusted net lending of general government (2007) 0.77 (0.006) -0.02 (0.963) 0.19 (0.620) cyclically adjusted net lending of general government (2008-2010) 0.45 (0.161) -0.12 (0.740) 0.84 (0.004) real effective exchange rate with gdp deflator -0.10 (0.776) -0.21 (0.563) -0.71 (0.031) private credit flow -0.06 (0.855) -0.18 (0.616) -0.73 (0.024) private debt 0.42 (0.204) -0.20 (0.575) -0.53 (0.143) current account balance 0.56 (0.074) 0.48 (0.156) 0.79 (0.012) notes: see the main text for the definition of the variables; the figures in parentheses are the p-values to test the null hypothesis of no linear dependence all the members without greece without greece and (11 countries) (10 countries) ireland (9 countries) when greece and ireland are filtered out most relationships are statistically significant at the 5% level, and the coefficients have the expected signs. the results in table 1 show that the foundations that explain the evolution of most national cycles vis-à-vis the emu aggregate in the crisis were laid in 2007, in the form of the net international investment position, the level of employment protection, competitiveness developments, the growth of the credit for the private sector, and the current account balance. besides, the fiscal stance in the period 2008-2010 is also relevant, as loose discretionary fiscal policies in the first years of the crisis tended to lower national comovements with the rest of the emu. the only indicators that are not statistically significant are trade intensity and the private sector debt. both point estimates are high and have the expected sign, and it is quite likely that the null hypothesis of no linear relationship has not been rejected because of the low power of the tests due to the small size of the sample. table 1 also gives some hints on why greece and ireland deviate from the general pattern. the greek performance is the consequence of its fiscal stance and the ensuing public debt burden, as the yearly cyclically adjusted net borrowing of the greek government averaged 6.3% of gdp in the period 2000-2007. the current level of the public debt is a major structural weakness, its correction will take a long period where the greek cycle will be out of sync with the euro area, and all through that time greece will continue to be an outlier within the emu. in the irish case the imbalances of the economy in 2007 suggested that ireland would decouple in the course of the crisis, as spain and portugal did. quite the opposite, the irish cycle became more aligned with the euro area cycle, and that explains why the correlation coefficients with the excess demand and credit indicators are not statistically significant when ireland is included. in contrast, the correlation with the employment protection is higher in absolute value and much more significant than when ireland does not enter in its calculation. taken together, these results point at the irish recoupling being mainly explained by the high flexibility of its labor market. such flexibility may have allowed market forces to play a leading role in the adjustment process, reducing the asymmetric impacts of the common shocks that hit the euro area, and compensating the negative starting conditions induced by the overheating of the irish economy in 2007. conclusions cyclical synchronization within the emu has traditionally been a source of concern, but in recent years there are serious doubts whether the euro area will be able to cope with the powerful asymmetric responses to the shocks that launched the financial crisis in 2007. though the crisis is still on the run, it is essential to evaluate its potential consequences on the emu, and this paper aimed to give some tentative empirical evidence on this issue. european journal of government and economics 1(1) 98 as we centered on a short, particular period of time, the data were scarce, and such paucity limited the statistical techniques that could be applied. therefore, the empirical analysis has a number of limitations, and its conclusions should be interpreted with prudence. even so, the methodology developed in the previous sections provided some appealing results on the evolution of cyclical convergence, and on the factors that determined such evolution. in particular, we focused on three major issues. first, we compared synchronization in the euro area to international developments. we found that, on average, the cycle of the euro area became more aligned with the cycles of the rest of the european union and of other developed economies. this result supports the conventional view that the crisis was activated by common financial shocks that propagated worldwide along symmetric transmission channels, leading to growing international synchronization amongst the developed economies in 2009-2010. within the euro area, however, member states gave the impression of entering into a phase of cyclical divergence. furthermore, since 2009 the mean correlation within the emu is lower than the mean correlations of the area with the rest of the european union and the rest of the world, a result that challenges the theoretical foundations of a currency area. as a consequence, we concentrated on investigating the causes of the intra-euroarea performance. we found that the cycles of most member states remained closely aligned to the emu aggregate, while a few members, namely ireland, greece, spain and portugal, reported remarkable changes in their comovements. actually, the apparent decline in the intra-euro-area synchronization is mainly due to the contribution of the greek data. when the sequence of mean correlations is computed only with the data of the ten other member states, it displays no trend along the crisis, and its values are higher than the mean correlations with the rest of the european union and the rest of the world. this finding confirms that large symmetric shocks, economic integration and international spillovers resulted in higher synchronization, both inside the emu and across the developed economies all through the world. in some countries, however, the common shocks unleashed forceful asymmetric effects that induced quite distinctive fluctuations, and such fluctuations were influential enough to distort the general measures of comovement for the euro as a whole. finally, we examined the possible causes of the different cyclical performances of the members of the euro area. we compared the correlations between each national cycle and the emu aggregate in 2011:4 and 2007:3, and appraised to what extent the observed variation could be explained by a variety of potential determinants. most indicators referred to 2007 or the period 2005-2007, to assess whether the factors that led to changes in the output comovements in the course of the crisis were already latent before its inception. for the largest part of members, the foundations of the change in the cyclical correlations can be traced back to 2007: the position of each country in terms of its net international investment position, the level of employment protection, competitiveness developments, private credit dynamics, and current account balance, all of them are statistically related to its cyclical performance in the period 2007-2011. trade intensity and the private sector debt in 2007 were not found to be significant, although the point estimates were quite high and had the expected sign. we also investigated why greece and ireland deviated from the general pattern. the greek performance is the consequence of its loose fiscal policy all along the period 2000-2007, that resulted in a sharp increase of the public debt. the irish recoupling appears to be explained by the high flexibility of its labor market, that allowed market forces to play a leading role in the adjustment process. their contribution compensated the adverse starting conditions induced by the excessive european journal of government and economics 1(1) 99 private demand in 2007, and helped to offset more rapidly the shocks that hit the euro area. the major policy implication stemming from our work is the need that national authorities maintain policy discipline, not only in the current particular circumstances, but also as a general rule. whatever indicator is considered, we showed that most overheating countries in 2007 decoupled from the euro area cycle in the following years, with the consequence that the single monetary and exchange rate policies became exceedingly costly for the member states that reported large positive output gaps before the crisis. endogenous synchronization was not powerful enough to coping with the accumulated national disequilibria. this task has to be tackled by national macroeconomic policies, that should avoid discretionary pro-cyclical biases that fuel internal imbalances. references altavilla, carlo (2004) ‘do emu members share the same business cycle?’, journal of common market studies 42(5): 869-896. artis, michael. (2003) ‘is there a european business cycle?’, cesifo working paper no. 1053. artis, michael., jarko fidrmuc and johann scharler (2008) ‘the transmission of business cycles. implications for emu enlargement’, economics of transition 16(3): 559-582. baxter, marianne and robert g. king (1999) ‘measuring business cycles: approximate band-pass filters for economic time series’, review of economics and statistics 81(4): 575-593. christiano, lawrence and terry j. fitzgerald (2003) ‘the band pass filter’, international economic review 44(2): 435-465. de grauwe, paul and francesco p. mongelli (2005) ‘endogeneities of optimum currency areas: what brings countries sharing a single currency closer together?’, working paper series no. 468, european central bank. de haan, jakob, robert inklaar and richard jong-a-pin (2008): ‘will business cycles in the euro area converge? a critical survey of empirical research’, journal of economic surveys 22(2): 234-273. dees, stéphane and nico zorelli (2011) ‘business cycle synchronisation. disentangling trade and financial linkages’, working paper series no. 1322, european central bank. einchengreen, barry (1990) ‘one money for europe? lessons from the us currency union’, economic policy 5(1): 117-187. eichengreen, barry (1993) ‘european monetary unification’, journal of economic literature 31(3): 1321-1357. european commission (2004) ‘cyclical convergence in the euro area: recent developments and policy implications’, quarterly report on the euro area 3(2): 2738. european commission (2005) ‘growth differences in the euro area’, quarterly report on the euro area 4(2): 19-29. european commission (2006) ‘cyclical synchronisation within the euro area: what do recent data tell us?’, quarterly report on the euro area 5(2): 19-24. european commission (2010) ‘how has the financial crisis affected cyclical differences within the euro area?’, quarterly report on the euro area 9(2): 25-28. european journal of government and economics 1(1) 100 frankel, jeffrey a. and andrew k. rose (1998) ‘the endogeneity of the optimum currency area criteria’, economic journal 108(449): 1009–1025. gayer, christian (2007) ‘a fresh look at business cycle synchronisation in the euro area’, european economy – economic papers number 287. imbs, jean (2004) ‘trade, finance, specialization, and synchronization’, review of economics and statistics 86(3): 723–734. konstantakopoulou, ioanna and efthymios tsionas (2011)‘the business cycle in eurozone economies (1960 to 2009)’, applied financial economics 21(20): 14951513. kaiser, regina and agustín maravall (2001) measuring business cycles in economic time series, new york: springer-verlag. krugman, paul r. (1991) geography and trade, cambridge, ma: mit press. mink, mark, jan jacobs and jakob de haan, j. (2007) ‘measuring synchronicity and co-movement of business cycles with an application to the euro area’, cesifo working paper no. 2112. mundell, robert a. (1961) a theory of optimum currency areas, american economic review 51(4): 657–665. papageorgiou, theofanis, panayotis g. michaelides and john g. milios (2010) ‘business cycle synchronization and clustering in europe (1960-2009)’, journal of economics and business 62(5): 419-470. rose, andrew k. (2000) ‘one money, one market: the effect of common currencies on trade’, economic policy 15(30): 9–45. regulating labour platforms, the data deficit european journal of government and economics 7(1), june 2018, 5-23. european journal of government and economics issn: 2254-7088 regulating labour platforms, the data deficit adrian john hawleya, * a royal holloway, university of london, united kingdom * corresponding author at: school of management, royal holloway, university of london, egham tw20 0ex, united kingdom. email: adrian.hawley.2015@live.rhul.ac.uk article history. received 24 january 2018; first revision required 23 april 2018; accepted 22 may 2018. abstract. it is widely reported that there is a data deficit regarding working conditions in the gig economy. it is known, however, that workers are disadvantaged because they are not classed as employees with the result that they lack work-related entitlements and may not be protected by the social welfare safety net. nor is this compatible with the social market economy enshrined in the european union treaties. two obstacles are that labour law and social policy are mainly a national competence and that platforms are reluctant to share data with regulators. in this paper i take the specific case of offline labour platforms intermediated by app and smart phone such as driving and delivering and look for new pathways between access to data and the shaping of public policy in member states with potentially legal certainty.. keywords. data; gig economy; social market; labour platforms; public policy; soft power 1. introduction 'a paradox in the digital and internet economy is that never before has so much data been collected, and never before has it been so difficult to access. the value of this data is likely to be much higher for social and public policy purposes than it is for private purposes of the platform operators' (codagnone, biagi and abadie 2016:60,61) the labour platform model of the gig economy offers popular services but depends for its competitive advantage over established service providers on low wages, precariousness and lack of inor -out of work entitlements and benefits. this raises the question of its compatibility with the concept of the social market enshrined in the european union (eu) treaties, specifically with regard to employment practices. there are different models of the social market throughout the eu but it has been broadly defined as a 'fundamental social model that ensured people’s rights, inclusion to social protection, a good wage model that shared productivity through collective bargaining and social dialogue' (burrow 2018) and, specifically, 'the creation of a world of work that is humane, socially balanced and sustainable' (jürgens, hoffmann and doi. https://doi.org/10.17979/ejge.2018.7.1.4330 mailto:adrian.hawley.2015@live.rhul.ac.uk a.j. hawley / european journal of government and economics 7(1), 5-23. 6 schildmann 2017:222). this is reflected in the concept of 'good work' in the gig economy by taylor (2017) who measures it against a number of criteria such as, 'employment quality, working conditions, consultative participation and collective representation in addition to wages' and also 'good gigs' by balaram, warden and wallace-stephens (2017) and as 'gute arbeit' in germany (german federal ministry of labour and social affairs 2017). with the rise of the gig economy is there a need therefore for new regulation or are existing rules being breached or evaded? how do we know? this is a question of data, its availability and quality. two important considerations are the role of the eu, since labour law is mainly a national competence, and the perennial cleavage of opinion about when and whether to regulate any economic activity. it is, however, the subject of data in relation to working conditions in the gig economy, and specifically to labour platforms, which is principally addressed in this paper. labour platforms do not provide the entitlements generally expected for their workers because they do not accept that they are employees. there is little if any corroborated, independent data, however, on how much they are paid, the hours they work, discrimination of any kind, the transparency of rating systems and opportunities for worker training and representation. we do not know much about their conditions except anecdotally and what platforms are willing to tell us through the medium of a few privileged researchers. by contrast, independent survey data presents a per country aggregate of demographics, people’s motivations and total hours worked on platforms. except in one case that i have found (balaram, warden and wallace-stephens 2017) they do not reveal data on conditions on specific platforms. nor, why should the latter reveal them for competitive or any other reasons? since their workers are generally treated as self-employed, even though they may be largely or even wholly dependent upon a single platform, their conditions presently do not fall under national labour law regulations, for example, the minimum national wage (or minimum living wage in the united kingdom) or transposed eu legislation in the case, for example, of the working time or written statement directives. extending benefits to workers on labour platforms in the gig economy will require changes in public policy which is a political decision. in an exceptional case, a tribunal (employment tribunals 2016; employment appeal tribunal 2017) has ruled in favour of the claim by some london-based drivers of uber, a ride-sharing service, to limited entitlements. the decision, however, does not have universal effect in the united kingdom (uk). both in the formulation and implementation of new rules or the extension of existing ones, data will need to be accessible from platform operators. generally no such data are shared directly by them with any national authority with the very recent exception of france in the case of payment of taxes. if new rules apply, does that mean that, for compliance, platforms will be mandated to supply certain data? they are unlikely to do so on a voluntary basis. and how might this be achieved without revealing commercially sensitive data or failing to protect the privacy of personal data as required by the general data protection regulation which became effective throughout the eu in may 2018? in the first half of this paper i start with a definition of what is meant by the catch-all term of a.j. hawley / european journal of government and economics 7(1), 5-23. 7 the 'gig' economy and the part described as 'labour platforms' with which this paper is concerned. there follows an overview of the literatures of organisation and management, paradigms of political economy and cultural responses towards rapid technological change that may have some explanatory value for the rise of the phenomenon and also differing attitudes towards it. these differences, i show, are being played out in an eu context as well as a national one. the second half of the paper is devoted to how data might be acquired and put to work systemically to improve the conditions of gig workers. i start with a brief review of the data from surveys and studies carried out in europe and the united states. if there is a change in public policy, the “elephant in the room” however, will be data compliance by platforms, an issue which looks as though it will present formidable obstacles and one on which little research has been done with the notable exception of arun sundararajan (2017). i sketch out his approach generally known as 'shared regulation' and the few others with the similar objective. since labour law and social policy are mainly national competences i consider the eu's use of soft power to achieve its objectives in this field. specifically i look for a pathway between access to data on working conditions and legal certainty in establishing minimum standards for workers. finally, i conclude by suggesting further areas of research for a deeper evaluation of opportunities for change in this field following the recommendations and principles of the recently proclaimed european pillar of social rights (ec 2017). 2. defining the ‘gig economy’ there is no precise definition in the literature which contributes to the difficulty in measuring it. the eu still refers to it by the loose, catch-all term 'collaborative economy' which could include both for-profit and not for-profit models and those where only the expenses of providing the service are recovered plus a small percentage (e.g. blablacar, a car-sharing service in france). it is clear, however that the european commission (ec) is referring mainly to the for-profit model in its agenda for the collaborative economy (ec 2016) and this is the one to which i refer in this paper. the literature, particularly when the collaborative economy first started to attract attention, referred to 'peer-to-peer' to signify individuals exploiting their under-used cars, bicycles, spare rooms, tools, time and know-how, to provide services to other individuals for a little extra money on an occasional basis. from this the term 'gigs' came to be adopted. this peer-to-peer model, however, has proved not to be the case with uber and other labour platforms in the driving and delivering sectors. many of its drivers in europe are engaged in the so-called uberx model (superseding the earlier uberpop version) and are licenced professionals (sometimes with financial support from the platform for vehicle purchase or lease). a similar development is true of airbnb, an accommodation-sharing service, where a large proportion of properties have been bought solely for renting out. 'gigs', however, is more widely understood as referring to activities where labour is performed rather than the exploitation of a.j. hawley / european journal of government and economics 7(1), 5-23. 8 assets such as accommodation. 'gigs or 'gig work' therefore best describes services performed by individual workers dependent upon platform intermediaries, sometimes several, to which they are connected by a smartphone or computer app, typically poorly paid (sometimes less than the national minimum wage), capable of being 'deactivated' at any time, without a collective voice, and lacking in either any of the work-related entitlements enjoyed by workers classified as 'employees' or full access to statutory state benefits when they need them. by contrast, surveys show that the flexibility to work when and where they want is liked by gig workers. equally, however, many say have little other choice. work in the gig economy can be online 'mind' services ranging from simple, repetitive tasks (as in clickworker) to those requiring higher skill levels (as in upwork) or offline physical services (such as ride-hailing or delivering). it is with the offline labour platform model with which this paper is concerned and how it fits into the interlocking circles of the overall 'collaborative economy' is shown in figure1. figure 1. the interlocking circles of collaborative economy. 'gigs' are very often neither 'peer-to-peer' as has been said nor in the case of uber occasional. according to the european court (ecj 2017:§47) 'it has become apparent that most trips are carried out by drivers for whom uber is their only or main professional activity'. although i am chiefly concerned with the physical rather than mind services, the social dimension of poorly paid, precarious work without entitlements or benefits can equally apply to both. a.j. hawley / european journal of government and economics 7(1), 5-23. 9 3. literature and theory the gig economy is a recent development brought to popular attention under the title ‘collaborative consumption’ by botsman and roo (2010) followed by rifkin (2014) whose theme was ‘the collaborative commons’. it has since become the object of wider scholarly research. bardhi and eckhardt (2012), for example, found that a car-sharing model (zipcar) set up primarily for access rather than profit (costs of operation covered) with benefits for sustainability failed to gain brand recognition loyalty. martin (2016) has drawn attention to the co-option of the ‘sharing’ ideal by ‘the market’, rosenblat and stark (2015, 2016) have studied how uber exercises ‘continuous soft surveillance’ and ‘remote control’ (2015:6) over its drivers. raval and dourish (2016) drawing on the concept of ‘body work and affective labor’ (2016: 99, 101) have surveyed ways in which its drivers are obliged to ‘earn’ their ratings under a reputation system to which most of them object, including, having also to perform 'emotional labour' for example, tolerating rude behaviour and providing (at their own expense) small comforts for passengers or having them sit in the front passenger seat. a plethora of reports have emerged from the eu institutions, national governments and independent research centres. 3.1 organisational aspects bad for workers, good for consumers (and potentially investors) the rise of the gig economy can be situated within the context of the information society described by nye (2014) and nagirnaya (2014) while its salient organisational characteristics are identified as information asymmetry (rosenblat and stark 2015:2016), ruthless, flexible and evasive entrepreneurialism (mejia 2016; sennett 2006; elert and henrekson 2016). traits such as disruptiveness and lack of corporate social responsibility are pre-figured in earlier texts by christensen (1997), christensen, raynor and mcdonald (2015) and crouch (2006), while it remains unclear whether schumpeter's (1934, 1939, 1942) theory of creative destruction is applicable as i have found no data which confirms that there has been a net increase in new jobs. in terms of wages, evidence is of the opposite. consumers, however, may benefit from lower costs and greater convenience. 3.2 management aspects back to the 19th century management is about how workers are treated and the organisational model described above represents, i contend, a regression from the human resources (hr) school of management to earlier forms inspired by f.w taylor and h. fayol's theories of scientific management, if not to even earlier forms of the division of labour inherent in the first industrial revolution and the experience of the tolpuddle martyrs. gone is the employee-centric orientation of the hr school a.j. hawley / european journal of government and economics 7(1), 5-23. 10 and the practices of human resource management foreshadowed by durkheim, mary parker follet and chester barnard and pioneered by elton mayo and the hawthorne studies. uber, for example, shares the principle of scientific management with its algorithmic control and panoptic surveillance of every task undertaken by its workers. its aim is the same to achieve the greatest possible return for the minimum outlay using new developments in technology and in this respect and others it goes well beyond taylor. in his theory maximum output by workers equated with highest wages and self-satisfaction, for which management was responsible for providing all the resources and skills training required. labour platforms such as uber or deliveroo, a goods-delivery service, treat their workers as instrumental, merely a commodity, a factor of production to be acquired (and disposed of) at the lowest possible cost in terms of wages, entitlements and representation, exemplifying marx's conception of 'the reserve army of unemployed labour' (braverman 1974 [1998]:265-276) which is made available for low paid, unskilled service jobs first by mechanisation, then automation, and now (latterly) digitalisation. by contrast, the peer-to-peer nature of gig work has been portrayed by platforms as a democratisation of labour by eliminating the traditional hierarchies of both taylorism and the hr school and fostering flexible work, wherever and whenever desired. offline labour platforms, however, have substituted an often mythic peer-to-peer claim for bogus self-employment. moreover, there is evidence (boston consulting group in uber, 2016; reuters, 2017; codagnone and martens, 2016:18) that net rates of pay quoted by uber and other platforms are dubious and very long hours (virtually limitless) are required to make a living or top-up another source of income. the data deficit, however, means that there are few empirical studies of the impact of this type of work, notably within the eu. a more truly peer-to-peer relationship could only be established by offline labour platforms operating as cooperatives but so far these have not made headway in europe. a de-centralised system such as blockchain promises its complete fulfilment if it were to materialise. 4. paradigms of political economy and cultural attitudes opinion on the extent to which the gig economy should be regulated varies throughout the eu and within the eu institutions themselves. the broadest division which animates policy makers and is already evident in current regulation is between pro-laissez faire, pro-choice ('leaving it to the market') and pro-social justice ('pro-values'). there are also those who do not unreservedly accept the application of new digital technology seeing it as 'the tyranny of rational choice solutions, alienated from concrete social practices' (strong and sposito 1995: 268) or 'the discrepancy between the growth of technically exploitable knowledge, on the one hand, and the absence of any worthwhile form of social life, on the other' (adorno and horkheimer quoted in finlayson 2005: 67). are the technologies exploited by uber leading towards a dystopia of contingent employees permanently on call and winner-takes-all monopolists, or economies of scale permitting cooperative enterprises? (adler 2016). habermas (2001:46) has warned us of a.j. hawley / european journal of government and economics 7(1), 5-23. 11 alienation resulting from the monetization of 'value orientations, binding norms and processes of understanding’, while we should not ignore weber's 'iron cage' (weber [1904/5], 1930: 123, 128) in which individuals are trapped by teleological efficiency, rational calculation and control the bureaucratisation of social order, the 'disenchantment' of the world (greisman and ritzer 1981:35). michael sandel (2012) draws attention to the 'market society' in which we now live rather than simply the 'market economy', and in which non-market values are 'crowded out'. 4.1 a european dimension these cleavages manifests themselves in several ways. firstly, they align closely with a varieties of capitalism typology, where uber, for example, and its drivers, are free to operate in the liberal market economies of uk, ireland, poland, estonia and lithuania, subject to relatively light touch conditions. by contrast, in most of the other member states of the eu which follow, in various ways, the social market model, it is either banned or its drivers are subject to considerably more onerous licencing requirement. secondly, they are reflected in the stance taken by lawmakers for example in the european parliament (ep), where an analysis which i conducted of parliamentary questions on the subject between november 2014 and may 2017 revealed that members of the ep were divided almost equally between those who called for further regulation at eu level and those who either did not or whose view was that it was expressly a matter for member states. thirdly they are demonstrated in the communications of the european commission which show that priority in policy making with regard to the gig economy has been placed more on the economic aspects namely growth and competitiveness than with social aspects. it is noteworthy that the commission has consistently declined to bring forward any new legislation on the gig economy since publication of its set of non-binding guidelines contained in its agenda for the collaborative economy (ec2016). the stance of the commission is seen by some, principally by trade unionists and others on the left, as consistent with its liberalising, deregulatory, pro-business tendency. by contrast, there is a push back by industry and market economists who hold it responsible for largely perverse effects on business of social measures such as the working time directive (wtd) and others relating to agency, temporary and part time workers, the prevention of discrimination against minority groups and health and safety. none of these, however, materially affect the conditions of gig workers because they only apply to employees. the proclamation of the european pillar of social rights (epsr) can be seen as an acknowledgement that 'social europe' has been somewhat neglected, particularly during the period of recovery from the great recession following the banking collapse of 2007 and 2008. president junker has said as much with his call for a 'social triple a for europe' (ec 2017a). like the agenda, referred to above, the epsr's twenty principles are non-binding. among them are explicit references to the elimination of precariousness, abusive contracts and in-work poverty and making statutory work-related entitlements, social welfare benefits and collective bargaining available to all a.j. hawley / european journal of government and economics 7(1), 5-23. 12 workers regardless of the type and duration of the employment relationship. the epsr concretely addresses social policy, and in particular, labour aspects of the gig economy, in a way that the agenda does not but it is still, essentially, an instrument of the eu's soft power. for the first time it treats decent wages and work-life balance as rights but the question is how these might gain a legal basis when labour law is, for the most part, a national competence and the commission does not intend to bring forward new legislation except for a proposal for a new directive on contracts (written statement directive) and clarification of the existing working time directive, both of which were already 'in the pipeline'. this is where access to data on working conditions has a critical role to play in shaping public policy and how this might be approached is discussed in the second half of this paper. 5. the data deficit the lack of empirical data is widely acknowledged (codagnone and martens 2016; huws, spencer and joyce 2016; oecd 2016; taylor 2017; cipd 2017; german federal ministry for labour and social affairs 2017; ec 2018). it would be more accurate and constructive, however, to talk about a lack of access as never before has so much data been collected as it has by platforms (codagnone, biagi and abadie 2016;60,61). a review i conducted of the relatively small number of independent surveys that have been carried out in europe reveals a focus on overall numbers, motivations and demographics of both service providers and consumers using offline labour platforms but very little data on specific working conditions and employment status of the former. further investigation is required into these, notably pay, leave, health and safety, working hours, tax, insurance, collective bargaining and quality of working life (huws et al. 2016:51). more detail has come from surveys commissioned by, or facilitated in the united states and europe by uber, but they have not been independently corroborated. the most noteworthy of these are the analyses by hall and krueger (2015, 2016). these have attracted criticism in the press and among other researchers being described as 'mostly descriptive and inconclusive' and whose findings on earnings as 'an utter misrepresentation of the reality' (codagnone and martens 2016:17,18). the operative data that would make a difference in formulating public policy, however, is not from surveys, whatever their credibility. the data, in the era of big data are already there. they are collected by platforms twenty four hours a day, justified in some cased for security reasons. uber, for example, collects licence, insurance and background checks (medical and criminal records) on each of it drivers as well as credit card details and mobile phone numbers for every rider. in effect it 'holds the key to the digital identities of its millions of contractors and customers' (lashinsky (2017:140). the task is making it available to those who legitimately need it (and not, by oversight, to those who do not as has recently been reported), namely policy makers, regulators and tax authorities. this raises questions of commercial sensitivity and data privacy protection. several approaches will be considered in the next section. a.j. hawley / european journal of government and economics 7(1), 5-23. 13 5.1 deriving data cooperation or coercion? if the objective is to ensure that workers on offline labour platforms enjoy the statutory entitlements and social welfare benefits available to those classed as employees, as envisaged by the epsr, then platforms need to be monitored. this suggests mandatory access to the data they hold which would require new regulations at either national or eu level. it is also a question of what data. treasuries are losing tax revenue and social security contributions from workers on platforms and at least two member states have made or will shortly make changes in the reporting of their liabilities, one of which is voluntary, the other mandatory. an opt-in system in place in estonia since 2016 allows drivers working on ride hailing platforms uber and taxify to register their willingness for their earnings to be transmitted to the tax office which is then added to their tax return. in france the problem of tax losses was raised by the sénat, the upper house of the french parliament, in 2015, alluding to platforms' 'exhaustive and instantaneous knowledge'. a new law will now require platforms to report their users' earnings to the relevant authorities over a secure link from 2019. the data will include identification, email address, private individual or professional status, total gross income received by the user during the calendar year in return for activities on the online platform, or paid through the platform (ep 2018:21). already, from 2018, a new right under the comte personnel d'activités (personal activity account) gives workers access to all their payslips. organisations are required to provide these in a familiar electronic form at any time requested. having oversight of their pay and hours over a period in an easily communicable way would help gig workers to determine the hourly rate for the work they have done. this is not necessarily clear given their irregular hours and may even be less than the national minimum or living wage. these are examples of one-way transmission of data by the platform, for one specific purpose, and only the french system is designed to catch all platform workers, who may, moreover, be working on several platforms. a more ambitious development would be the twoway question and reply procedure proposed by sundararajan (2017). here the platform's database would be interrogated by any legitimate state authority for, presumably, any data which it is authorised to access. this would be achieved by the use of application programming interfaces (that is, computers talking to computers). in this decentralised system data remains inside platforms, no commercially sensitive data is released to third parties and only queried for the attainment of public policy objectives. the process is described as 'digital audit' which might suggest either an obligation on the part of the platform to comply with minimum employment entitlements or only, at least at first, a data gathering exercise for policy formulation within an ethos of best practice. uber has on more than one occasion declared itself willing to make data on the earnings of its drivers available for tax purposes. travis kalanick, its co-founder and former ceo, offered to do so in france (le monde 2015) if an agreement was reached with government, and most recently in its white paper on work and social protection in europe (uber 2018:21) featured a screen shot on its app of pay out and hours worked over a period a.j. hawley / european journal of government and economics 7(1), 5-23. 14 which could be shared directly with the tax authorities. in voluntarily making this data offer, uber strenuously denies acting as an employer this runs wholly counter to its business model and its drivers can choose whether to use the option or not. the reception of such data might be useful if also made available to social policy makers, but this raises the issue of data protection under the eu's general data protection regulation (gdpr) of 2018, which is supreme over national law and has immediate effect. nor could they depend on it for a complete picture as uber drivers might also be working on other platforms without a similar option. in sundararajan's model which he calls 'data-driven regulation', regulatory responsibility is 'delegated to the party that naturally has the data' (sundararajan 2017:24). he compares this favourably with two other forms of self-regulation such as rating systems ('peer-regulation'), and transportation network companies ('tncs', or 'self-regulatory collectives') where government sets standards with which drivers working on a platform registered as tnc must comply. rating systems, however, have been much criticised for being faked and lacking transparency while the standards with which tncs such as uber need to comply are primarily concerned with consumer protection. in both cases the aim is promoting trust. this is undeniably important but neither of these forms of regulation address the central issue of worker entitlements and social benefits. on the contrary, they are aimed at 'market' issues. where the profit interests of platforms do not coincide with the wellbeing of their workers and they evidently may not as they deny responsibility for them as employees then a regulatory authority with the power of enforcement needs to step in. as has been said, platforms already hold all the data that regulators might need to know and data analytics can present a picture of demographics (age, ethnicity, gender, family status), motivations and any other aspects that might be of interest for social cohesion in addition to pay, hours, leave and representation. here, there are several operative issues. firstly, each government, particularly those member states within the eurozone to which epsr is primarily addressed, needs to decide on the objectives of its employment and social policies, secondly the data it needs to achieve them, thirdly whether platforms are to be mandated or consensual on the basis of best practice and finally, what are the privacy issues? on the first two, it is intended that member states will be guided by accompanying actions prescribed by the epsr which will be discussed later in this paper. on the third, a dialogue between government and an industry association such as seuk (sharing economy uk) might be constructive. privacy is of concern to both platforms and to their workers. the former need to be assured that there is no possibility of commercially sensitive data being leaked to competitors and the latter that the privacy of their personal data is protected in conformity with the gdpr. digital audit using apis as proposed by sundararajan would reassure platforms on the first score. the concept of the national hourly minimum wage is difficult to apply in the case of labour platform workers who can log-on and log-off whenever they want and who may work on several platforms with different pay rates and ways of determining them (for example by time, journey, delivery). the more important objective for public policy in ensuring that platform workers earn a living wage, defined as 'as a wage above the legal minimum that reflects the cost of living a.j. hawley / european journal of government and economics 7(1), 5-23. 15 families face . . the living wage should be based on an up-to-date basket of goods and services, reflecting social consensus as to what constitutes a decent standard of living' (resolution foundation 2016). in the uk, the national minimum wage (nmw) has been 're-branded' by government as the national living wage (nlw) but may not reflect what is required to meet this definition. moreover, paying the nlw is voluntary for employers. the rise of the platform economy presents a further complication and that is why access to current data held by platforms in an era of non-standard employment is now required in setting a rate for a living wage. 5.2 making data work it is, so far, not clear how a how a conflict of interests between labour platforms and the social market can be resolved either through voluntary or coercive means. a first step is access to all the data necessary for the attainment of the employment and social policy objectives outlined in the epsr. in this section therefore, i consider the accompanying actions prescribed by the epsr to visualise how the use of data might create a pathway from its rhetoric to making a difference to the lives of gig workers some of whom are 'just about managing' to make a living and some who are not. the epsr is a soft power instrument relying for eventual effect on a raft of other soft power procedures and measures. the most prominent of these are the social scoreboard and the country specific recommendations (csrs) whose progress towards implementation are to be to be monitored on the european semester. legal certainty may also arise from treaty clauses (hard law) relating to social protection. firstly, social policies might be interpreted according to the 'social clause' of the treaty (namely art. 9 tfeu) in specific cases by the european court of justice (ecj), secondly the competence of the eu 'to assist and complement' member states with regard to social security and the social protection of workers is established by art.153, and finally its right under art.352 to take action to attain these objectives where the treaty clause itself does not provide powers to give effect an outcome characterised as 'spillovers' in the neo-functionalist hypothesis of eu integration theory (hass, lindberg, schmitter 1955-75; sandholtz, stone sweet 1998; rosamund 2005). moreover it has been noted that what starts out in the eu as an aspiration may eventually become primary law, that is, a treaty commitment, as happened with the incorporation of the charter of fundamental human rights in tfeu. there seems little likelihood however of a treaty change and even less of a new treaty in the foreseeable future. in any case, i am more concerned, with the scoreboard and crs because they depend on access to data which is available if it can be accessed and could lead to changes at member state level. the eu's scoreboard is a comparative country by country yearly analysis of a number of areas of public policy. social policy, which covers employment and social services, was added in 2017. performance is rated, either on a numerical scale with base 2010 or a simple ordinal a.j. hawley / european journal of government and economics 7(1), 5-23. 16 scale, according to the twenty principles of the epsr. twelve indicators are grouped under three headings, the most relevant of which for the gig economy, is 'dynamic labour markets and fair working conditions'. most of the latest indicators for the social scoreboard, however, do not go beyond 2016. the criteria on which they should be based has been criticised (piasna 2017:5) for its emphasis on the quantity rather than quality of work which tends towards the economic rather than the most trenchant social impact. revisions are due for 2019. sources of data are both quantitative and qualitative. the former are compiled from a range of sources by eurostat and the latter, which concern labour market policy, by the european commission's directorate-general for employment, social affairs and inclusion. the results of the social scoreboard will from now on contribute to the country reports which are developed by the commission in collaboration with member states, the social partners and other stakeholders for the european semester on policy coordination. trade unions advocate for mandatory social protection for workers, provided it does not lower existing national provisions, whereas employers are split with some favouring voluntary measures. such differences go back to the unresolved question raised earlier as to the auditing of data held by platforms. country reports conclude with country specific recommendations which will also now reflect progress on implementing the principles pf the epsr. opinions, however, differ on the effectiveness of csrs since they were introduced in 2011 (costello 2017). what is clear is that little direct evidence of gig workers' conditions, and specifically on offline labour platforms, finds its way onto the social scoreboard and hence influences social policy at least so far. both this type of work, often still referred to as 'a-typical' or 'non-standard' despite its growing prevalence, and a union system intended for measuring its impact are, after all, recent phenomena. few member states have yet adjusted their employment or social security systems. the lack of an adequate statistical base prevents a comprehensive quantitative picture of those affected. the commission (ec 2018) has decided that a way forward is for another soft law instrument, namely a recommendation by the council, to be approved by the co-legislators, on access to social protection for workers and the self-employed. it asks member states to commit themselves 'to collect and publish reliable statistics on access to social protection broken down by labour market status (self-employed/employee), type of employment relationship (temporary/permanent, part-time/full-time, new forms of work/standard employment), gender, age and citizenship'. new forms of work are understood to mean ondemand and platform work which are currently excluded from social protection. in this way and cooperating with eurostat, suitable indicators should become available. in the end it comes down to member states to decide on their policies towards the social protection of workers but they are likely to be influenced by the data that is exposed on the scoreboard and which become the subject of country specific recommendations. a.j. hawley / european journal of government and economics 7(1), 5-23. 17 6. conclusions the argument of this paper is that the timely and comprehensive provision of data is required in adjusting labour law and social welfare polices to protect workers on offline labour platforms which i started by situating within the overall terminology of the collaborative, gig, sharing or peer-to-peer economy. my argument is based on two assertions that may not be obvious. firstly, it is not so much about looking for data they are already there in abundance but about access to them as observed at the outset (codagnone et al. 2016:60,61). this led to a number of unresolved questions that will require further research. one is about the conduct of platforms, access to whose data on working conditions is required. is this to be on a voluntary, cooperative basis as they would be expected to prefer? are the offers made by uber in its recent white paper to be taken at face value? digital audit was suggested as a 'safe' way of yielding data but does not answer the question of doing so freely or being obliged to. if the former, does this mean adhering to an industry code of best practice? if so, who decides on the rules? if it is the latter, might this mean a change in labour law, for example, the reclassification of employment status, the re-alignment of work-related entitlements and statutory welfare benefits? or, alternatively, the establishment of a regulatory authority for the sector with legal powers to insist on compliance, for example like the gangmasters and labour abuse authority? secondly, i regard this as an eu issue as much as a national one because work on such platforms does not appear to me to be compatible in its present form with the european social market model, diverse as it is. this, as i have demonstrated, is amply borne out by the proclamation of the epsr and subsequent measures taken at union level using existing institutional forums. i showed that rules were changing in two member states for yielding data on tax liability resulting from work done on their platforms. if this were extended in scope and made available to other state actors would it be permitted under the new general data protection regulation whose effects would be a subject of future research? such data i suggested could be instrumental, for example, in setting a minimum living wage and by that i mean a truly 'liveable' minimum wage, not a re-branded national minimum wage. i referred to the epsr but it is far from clear whether its rhetoric will result in concrete action. my analysis reveals a pivot towards the use of soft power versus hard power (or hard law) by the eu institutions and a highly relevant area for further research would be the effectiveness of the eu's soft versus hard power in social, employment and other policy areas for comparison. in the past we have become accustomed to the use of hard power to reduce barriers to trade within the internal market such as the services directive (2006) and the mutual recognition regulation (2009) or to extend the social acquis with directives to eliminate discrimination, clarify the treatment of posted, agency and part-time workers and promote workplace safety and work-life balance. in the present case, the european commission has consistently rejected further legislation and instead we are presented with a plethora of agendas, scoreboards, dialogues, principles and recommendations. the object is to 'nudge' member states to taking a.j. hawley / european journal of government and economics 7(1), 5-23. 18 action themselves and to build a consensus for doing so, primarily within the eurozone. by exception, a proposal for a new measure on contracts (the written statements directive) could be of significance for platform workers but this was already planned before the proclamation of the epsr and will only benefit them if it is applied to all workers regardless of the employment relationship. since the european economic community (the 'common market') became the eu at maastricht, the main thrust has been to make free movement in its various forms a reality. this has met with varying degrees of success, of which services, except financial, have scarcely been one despite the services directive of 2006. a major exponent of the labour platform model, uber, has recently been ruled by the ecj to be outside the acquis. trade is an exclusive eu competence; employment and social policy are not. attempts to harmonise the latter have accordingly been limited but they remain a union objective of closer integration within the social market model laid down in the treaty. we have also witnessed a push-back against extension of the social acquis since the great recession following the banking collapse of 2007/2008 and the ensuing sovereign debt and euro crises. germany has been regarded as an epitome of the social market yet in 2015 the then federal finance minister wolfgang schäuble commented (varoufakis 2017:211,212) that the 'over-generous' european social model was no longer sustainable and had to be ditched. comparing welfare states in europe with india and china where no social safety net exists at all, he argued that 'europe was losing competitiveness and would stagnate unless social benefits were curtailed en masse'. moreover, some do not believe that controlling the gig economy is a matter for jurisdictions and statutory regulation. beyond protecting consumer safety, this should be left to market forces. this position is expressed by niemietz and zuluaga (2016) in their discussion paper on the uk's taxi industry for the institute of economic affairs, a neo-liberal think tank. for them competitive pressures and trial and error will eventually determine what is tolerated through 'private' regulation and the emergence of 'regulatory brands' related to reputation, services offered and the suitability of their regulatory mechanisms (2016:37). as platforms are still an evolving phenomenon, this is an area for further research as is the sustainability of their business model should greater restraints be imposed upon them such as treating their workers as employees. there is no reference, however, in the 'market' view to the quality of employment. hans kundnani (2018), to whom i am indebted for drawing my attention to varoufakis' book argues that a 'competitive europe' has now become the model for the european commission and of pro-europeans led by angela merkel, which 'bears little resemblance to the one of the “proeuropean” imagination with its emphasis on the “social market economy”. in becoming more “competitive”, suggests kundnani, 'the eu may be hollowing out the model for which it once stood'. hence, for the present, only the exercise of soft power in relation to bringing social responsibility to labour platforms seems to be available to it. it may not be enough. in a globalised labour market which is rapidly changing due to technological developments the overarching question is whether there is room for an economic model that is both 'competitive' a.j. hawley / european journal of government and economics 7(1), 5-23. 19 and 'social,' to which the eu aspires and which bears heavily on the integration project. references adler, p.s. (2016) 'alternative economic futures: a research agenda for progressive management scholarship', academy of management perspective, 30(2): 123-128. https://journals.aom.org/toc/amp/30/2 balaram, b., warden, j.r. and wallace-stephens, f. (2017) 'good gigs, a fairer future for the uk's gig economy', rsa action and research centre, pp.3-64. . https://www.thersa.org/globalassets/pdfs/reports/rsa_good-gigs-fairer-gig-economyreport.pdf bardhi, f. and eckhardt, g.m. (2012) ‘access-based consumption: the case of car sharing’, journal of consumer research, vol. 39, dec 2012. http://www.jstor.org/stable/pdf/10.1086/666376.pdf?refreqid=excelsior%3a226fc37df9e6f054 2a55cfadad7a8644 botsman, r. and roo, r. (2010) what’s mine is yours, how collaborative consumption is changing the way we live, collins braverman, h. (1974[1998]) labour and monopoly capital, the degradation of work in the twentieth century, monthly review press, ny burrow, s. (2018) 'wef co-chair: greed is still the economic engine' https://www.euractiv.com/section/economy-jobs/interview/wef-co-chair-greed-is-still-theeconomic-engine-peace-and-democracy-the-collateral-damage/ christensen, c.m. (1997) the innovator's dilemma : when new technologies cause great firms to fail, boston, mass : harvard business school press christensen, c.m., raynor, m.e. and mcdonald, r. (2015) 'what is disruptive innovation?', harvard business review, dec 2015. accessed via “what is disruptive innovation?” cipd (2017) 'to gig or not to gig? stories from the modern economy', survey report by the cipd, pp.2-54. https://www.cipd.co.uk/images/to-gig-or-not-to-gig_2017-stories-from-themodern-economy_tcm18-18955.pdf codagnone, c., abadie, f. and biagi, f. (2016) 'the passions and the interests: unpacking the ‘sharing economy’, ec, jrc science for policy report, pp. 3-156. http://publications.jrc.ec.europa.eu/repository/bitstream/jrc101279/jrc101279.pdf codagnone, c. and martens, b. (2016) 'scoping the sharing economy: origins, definitions, impact and regulatory issues', institute for prospective technological studiesdigital economy working paper 2016/01jrc technical reports, 3–25. http://publications.jrc.ec.europa.eu/repository/bitstream/jrc101279/jrc101279.pdf costello, t. (2017) 'six years on, assessing the impact of the country specific recommendations. http://www.iiea.com/ftp/publications/2017/6%20years%20on%201st%20june.pdf https://journals.aom.org/toc/amp/30/2 https://www.thersa.org/globalassets/pdfs/reports/rsa_good-gigs-fairer-gig-economy-report.pdf https://www.thersa.org/globalassets/pdfs/reports/rsa_good-gigs-fairer-gig-economy-report.pdf http://www.jstor.org/stable/pdf/10.1086/666376.pdf?refreqid=excelsior%3a226fc37df9e6f0542a55cfadad7a8644 http://www.jstor.org/stable/pdf/10.1086/666376.pdf?refreqid=excelsior%3a226fc37df9e6f0542a55cfadad7a8644 https://www.euractiv.com/section/economy-jobs/interview/wef-co-chair-greed-is-still-the-economic-engine-peace-and-democracy-the-collateral-damage/ https://www.euractiv.com/section/economy-jobs/interview/wef-co-chair-greed-is-still-the-economic-engine-peace-and-democracy-the-collateral-damage/ https://hbr.org/search?term=michael+e.+raynor https://hbr.org/2015/12/what-is-disruptive-innovation/ https://www.cipd.co.uk/images/to-gig-or-not-to-gig_2017-stories-from-the-modern-economy_tcm18-18955.pdf https://www.cipd.co.uk/images/to-gig-or-not-to-gig_2017-stories-from-the-modern-economy_tcm18-18955.pdf http://publications.jrc.ec.europa.eu/repository/bitstream/jrc101279/jrc101279.pdf http://publications.jrc.ec.europa.eu/repository/bitstream/jrc101279/jrc101279.pdf http://www.iiea.com/ftp/publications/2017/6%20years%20on%201st%20june.pdf a.j. hawley / european journal of government and economics 7(1), 5-23. 20 crouch, c. (2006) 'modelling the firm in its market and organizational environment: methodologies for studying corporate social responsibility', organization studies, 27(10):1533-1551. http://journals.sagepub.com/doi/pdf/10.1177/0170840606068255 ec (2016) 'a european agenda for the collaborative economy', ec com (2016) 184 final, https://ec.europa.eu/transparency/regdoc/rep/1/2016/en/1-2016-356-en-f1-1.pdf ec (2017) 'establishing a european pillar of social rights', com(2017) 250 final, 26 april, 2017 pp 1-10. https://ec.europa.eu/transparency/regdoc/rep/1/2017/en/com-2017-250-f1-enmain-part-1.pdf ec (2017a) 'president juncker at the social summit for fair jobs and growth', 17 nov 2017. https://ec.europa.eu/commission/news/president-juncker-social-summit-fair-jobs-and-growth2017-nov-17_en ecj (2017) opinion of advocate general szpunar delivered on 11 may 20171 case c-434/15 asociación profesional elite taxi v uber systems spain sl. https://eur-lex.europa.eu/legalcontent/en/txt/pdf/?uri=celex:62015cc0434&from=en ec (2018) proposal for a council recommendation on access to social protection for workers and the self-employed, com, 132. https://ec.europa.eu/transparency/regdoc/rep/1/2018/en/com-2018-132-f1-en-mainpart-1.pdf elert, n. and henrekson, m. (2016), ‘evasive entrepreneurship’, small business economics, 47:95–113. https://link.springer.com/content/pdf/10.1007%2fs11187-016-9725-x.pdf employment tribunals (2016) 'reason for the reserved judgement on preliminary hearing sent to the parties' 28 oct 2016. https://www.judiciary.gov.uk/wpcontent/uploads/2016/10/aslam-and-farrar-v-uber-reasons-20161028.pdf employment appeal tribunal (2017) 'at the tribunal on 27 & 28 september 2017 judgment handed down on 10 november 20'. https://assets.publishing.service.gov.uk/media/5a046b06e5274a0ee5a1f171/uber_b.v._and _others_v_mr_y_aslam_and_others_ukeat_0056_17_da.pdf ep (2018b) 'the collaborative economy and taxation: taxing the value created in the collaborative economy', in-depth analysis, european parliamentary research service, pp.328. http://www.europarl.europa.eu/regdata/etudes/idan/2018/614718/eprs_ida(2018)61471 8_en.pdf finlayson, j.g. (2005) habermas, a very short introduction, oxford german federal ministry of labour and social affairs (2017), 're-imagining work. white paper work 4.0 (arbeiten 4.0), pp. 4-232. https://www.bmas.de/shareddocs/downloads/en/pdfpublikationen/a883-white-paper.pdf?__blob=publicationfile&v=3 greisman, h.c. and ritzer, g. (1981) ‘max weber, critical theory, and the administered world’, qualitative sociology 40(1):34-55. https://link.springer.com/article/10.1007/bf00987043 habermas, j. (2001) postnational constellation, political essays, trans. and ed. pensky, m. mit press. http://journals.sagepub.com/doi/pdf/10.1177/0170840606068255 https://ec.europa.eu/transparency/regdoc/rep/1/2016/en/1-2016-356-en-f1-1.pdf https://ec.europa.eu/transparency/regdoc/rep/1/2017/en/com-2017-250-f1-en-main-part-1.pdf https://ec.europa.eu/transparency/regdoc/rep/1/2017/en/com-2017-250-f1-en-main-part-1.pdf https://ec.europa.eu/commission/news/president-juncker-social-summit-fair-jobs-and-growth-2017-nov-17_en https://ec.europa.eu/commission/news/president-juncker-social-summit-fair-jobs-and-growth-2017-nov-17_en https://eur-lex.europa.eu/legal-content/en/txt/pdf/?uri=celex:62015cc0434&from=en https://eur-lex.europa.eu/legal-content/en/txt/pdf/?uri=celex:62015cc0434&from=en https://ec.europa.eu/transparency/regdoc/rep/1/2018/en/com-2018-132-f1-en-main-part-1.pdf https://ec.europa.eu/transparency/regdoc/rep/1/2018/en/com-2018-132-f1-en-main-part-1.pdf https://link.springer.com/content/pdf/10.1007%2fs11187-016-9725-x.pdf https://www.judiciary.gov.uk/wp-content/uploads/2016/10/aslam-and-farrar-v-uber-reasons-20161028.pdf https://www.judiciary.gov.uk/wp-content/uploads/2016/10/aslam-and-farrar-v-uber-reasons-20161028.pdf https://assets.publishing.service.gov.uk/media/5a046b06e5274a0ee5a1f171/uber_b.v._and_others_v_mr_y_aslam_and_others_ukeat_0056_17_da.pdf https://assets.publishing.service.gov.uk/media/5a046b06e5274a0ee5a1f171/uber_b.v._and_others_v_mr_y_aslam_and_others_ukeat_0056_17_da.pdf http://www.europarl.europa.eu/regdata/etudes/idan/2018/614718/eprs_ida(2018)614718_en.pdf http://www.europarl.europa.eu/regdata/etudes/idan/2018/614718/eprs_ida(2018)614718_en.pdf https://www.bmas.de/shareddocs/downloads/en/pdf-publikationen/a883-white-paper.pdf?__blob=publicationfile&v=3 https://www.bmas.de/shareddocs/downloads/en/pdf-publikationen/a883-white-paper.pdf?__blob=publicationfile&v=3 https://link.springer.com/article/10.1007/bf00987043 a.j. hawley / european journal of government and economics 7(1), 5-23. 21 hall, j.v. and krueger, a. b. (2015) 'an analysis of the labor market for uber’s driverpartners in the united states', working paper #587 princeton university industrial relations section, pp.1-27. http://arks.princeton.edu/ark:/88435/dsp010z708z67d hall, j.v. and krueger, a. b. (2016) 'an analysis of the labor market for uber’s driverpartners in the united states', update, working paper 22843 national bureau of economic research pp 1-35. http://www.nber.org/papers/w22843.pdf huws, u. spencer, n., h. and joyce, s. (2016) 'crowdworking survey for feps and uni europa', in conjunction with ipsos mori pp. 1-50. http://www.fepseurope.eu/assets/39aad271-85ff-457c-8b23-b30d82bb808f/crowd-work-in-europe-draftreport-last-versionpdf.pdf jürgens, k., hoffmann, r. and schildmann, c. (2017) 'let’s transform work! recommendations and proposals from the commission on the work of the future', hansböckler stiftung, trans. from the german by andrew wilson pp. 8-261. https://www.boeckler.de/pdf/p_study_hbs_376.pdf kundnani, h.(2018) 'the troubling transformation of the eu', webpost 06 april 2018 https://www.socialeurope.eu/the-troubling-transformation-of-the-eu lashinsky, a (2017) wild ride, inside uber's quest for world domination, portfolio penguin le monde (2015) 'uber, "les batailles juridiques seront sans fin", 21 jan 2015 http://www.lemonde.fr/economie/article/2015/01/21/uber-les-batailles-juridiques-seront-sansfin_4560361_3234.html martin, c.j. (2016) ‘the sharing economy: a pathway to sustainability or a nightmarish form of neoliberal capitalism’, ecological economics 121 (2016): 149–159. http://dx.doi.org/10.1016/j.ecolecon.2015.11.027 mejia, r. (2015) 'a pressure chamber of innovation: google fiber and flexible capital', communication and critical/cultural studies, 12 (3): 289-308. http://dx.doi.org/10.1080/14791420.2015.1027240 nagirnaya, a.v. (2014) ‘the information revolution and some issues in communication geography’, geography and natural resources, vol. 35, no. 1: 1-6. https://link.springer.com/content/pdf/10.1134%2fs1875372814010016.pdf niemietz, k. and zuluaga, d. (2016) 'hire authority. turning statutory regulation into private regulation for the uk’s taxi industry', iea discussion paper no.76, pp3-49 https://iea.org.uk/wp-content/uploads/2016/11/hire-authority-pdf.pdf nye, j.s. (2014) ‘the information revolution and soft power’, current history 113(759): 1922. http://nrs.harvard.edu/urn-3:hul.instrepos:11738398 oecd (2016) new forms of work in the digital economy', working party on measurement and analysis of the digital economy, (dsti/iccp/iis(2015)13/final http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=dsti/iccp/iis(2015 )13/final&doclanguage=en http://arks.princeton.edu/ark:/88435/dsp010z708z67d http://www.nber.org/papers/w22843.pdf http://www.feps-europe.eu/assets/39aad271-85ff-457c-8b23-b30d82bb808f/crowd-work-in-europe-draft-report-last-versionpdf.pdf http://www.feps-europe.eu/assets/39aad271-85ff-457c-8b23-b30d82bb808f/crowd-work-in-europe-draft-report-last-versionpdf.pdf http://www.feps-europe.eu/assets/39aad271-85ff-457c-8b23-b30d82bb808f/crowd-work-in-europe-draft-report-last-versionpdf.pdf https://www.boeckler.de/pdf/p_study_hbs_376.pdf https://www.socialeurope.eu/the-troubling-transformation-of-the-eu http://www.lemonde.fr/economie/article/2015/01/21/uber-les-batailles-juridiques-seront-sans-fin_4560361_3234.html http://www.lemonde.fr/economie/article/2015/01/21/uber-les-batailles-juridiques-seront-sans-fin_4560361_3234.html http://dx.doi.org/10.1016/j.ecolecon.2015.11.027 http://dx.doi.org/10.1080/14791420.2015.1027240 https://link.springer.com/content/pdf/10.1134%2fs1875372814010016.pdf https://iea.org.uk/wp-content/uploads/2016/11/hire-authority-pdf.pdf http://nrs.harvard.edu/urn-3:hul.instrepos:11738398 http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=dsti/iccp/iis(2015)13/final&doclanguage=en http://www.oecd.org/officialdocuments/publicdisplaydocumentpdf/?cote=dsti/iccp/iis(2015)13/final&doclanguage=en a.j. hawley / european journal of government and economics 7(1), 5-23. 22 piasna, a. (2017) ''' bad jobs’' recovery? european job quality index 2005-2015. working paper 2017-06, etui, pp. 4-43 https://www.etui.org/publications2/working-papers/bad-jobs-recovery-european-job-qualityindex-2005-2015 raval, n. and dourish, p. (2016) ' standing out from the crowd: emotional labor, body labor, and temporal labor in ridesharing', proceedings of the 19th acm conference on computer-supported cooperative work & social computing 27 feb to 2 mar, 2016, san francisco, ca. http://wtf.tw/ref/raval.pdf resolution foundation (2016) 'making the living wage', pp.2-39. https://www.resolutionfoundation.org/app/uploads/2016/07/living-wage-review.pdf reuters (2017) 'france's gig economy creates hopes and tensions as election looms', webpost 11 april 2017. https://www.reuters.com/article/us-france-election-gigeconomyanalysis/frances-gig-economy-creates-hope-and-tension-as-election-looms-iduskbn17d0iu rosenblat, a. and stark, l. (2015) ‘uber’s drivers: information asymmetries and control in dynamic work’. https://www.valuewalk.com/2015/12/ubers-drivers-information-asymmetriesand-control-in-dynamic-work/ rosenblat, a. and stark, l. (2016) ‘algorithmic labor and information asymmetries: a case study of uber’s drivers’, international journal of communication, 10: 3758–3784. http://ijoc.org/index.php/ijoc/article/view/4892 rifkin, j. (2014) the zero marginal cost society, st. martin’s press sandel, m. (2012) what money can't buy, the moral limits of markets, penguin sennett, r. (2006) the culture of the new capitalism, new haven, ct: yale university press strong. t. b and sposito, f. a. (1995) 'habermas's significant other', in the cambridge companion to habermas, ed. white, s.k. cambridge, 'part v. the defence of modernity', pp.263-286 sundararajan, a. (2017) 'the collaborative economy: socioeconomic, regulatory and labor issues', in-depth analysis for directorate general for internal policy, ep, ip/a/imco/2016-12, pp.6-33. http://www.europarl.europa.eu/regdata/etudes/idan/2017/595360/ipol_ida(2017)595360 _en.pdf taylor, m. (2017) 'good work, the taylor review of modern working practices', commissioned by uk government, pp.4-112. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/627671/goodwork-taylor-review-modern-working-practices-rg.pdf uber (2016) 'new data reveals uber’s economic impact in france', web post 28 dec 2016. https://medium.com/uber-under-the-hood/new-data-reveals-ubers-economic-impact-infrance-47debf888ef6 uber (2018) 'white paper on work and social protection in europe'. https://newsroomadmin.uberinternal.com/wp-content/uploads/2018/02/uber-white-paper-onwork-and-social-protections-in-europe.pdf https://www.etui.org/publications2/working-papers/bad-jobs-recovery-european-job-quality-index-2005-2015 https://www.etui.org/publications2/working-papers/bad-jobs-recovery-european-job-quality-index-2005-2015 http://wtf.tw/ref/raval.pdf https://www.resolutionfoundation.org/app/uploads/2016/07/living-wage-review.pdf https://www.reuters.com/article/us-france-election-gigeconomy-analysis/frances-gig-economy-creates-hope-and-tension-as-election-looms-iduskbn17d0iu https://www.reuters.com/article/us-france-election-gigeconomy-analysis/frances-gig-economy-creates-hope-and-tension-as-election-looms-iduskbn17d0iu https://www.valuewalk.com/2015/12/ubers-drivers-information-asymmetries-and-control-in-dynamic-work/ https://www.valuewalk.com/2015/12/ubers-drivers-information-asymmetries-and-control-in-dynamic-work/ http://ijoc.org/index.php/ijoc/article/view/4892 http://www.europarl.europa.eu/regdata/etudes/idan/2017/595360/ipol_ida(2017)595360_en.pdf http://www.europarl.europa.eu/regdata/etudes/idan/2017/595360/ipol_ida(2017)595360_en.pdf https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/627671/good-work-taylor-review-modern-working-practices-rg.pdf https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/627671/good-work-taylor-review-modern-working-practices-rg.pdf https://medium.com/uber-under-the-hood/new-data-reveals-ubers-economic-impact-in-france-47debf888ef6 https://medium.com/uber-under-the-hood/new-data-reveals-ubers-economic-impact-in-france-47debf888ef6 https://newsroomadmin.uberinternal.com/wp-content/uploads/2018/02/uber-white-paper-on-work-and-social-protections-in-europe.pdf https://newsroomadmin.uberinternal.com/wp-content/uploads/2018/02/uber-white-paper-on-work-and-social-protections-in-europe.pdf a.j. hawley / european journal of government and economics 7(1), 5-23. 23 varoufakis, y. (2017) adults in the room. my battle with europe's deep establishment, the bodley head, london. weber, m. ([1904/5], 1930, 1946, 1958) the protestant ethic and the spirit of capitalism, trans. parsons, t., allen & unwin abstract. it is widely reported that there is a data deficit regarding working conditions in the gig economy. it is known, however, that workers are disadvantaged because they are not classed as employees with the result that they lack work-related en... keywords. data; gig economy; social market; labour platforms; public policy; soft power references european journal of government and economics volume 4, number 1 (june 2015) issn: 2254-7088 the role of political competition in the link between electoral systems and corruption: an extension maria rosaria alfano, seconda università degli studi di napoli, italy anna laura baraldi, seconda università degli studi di napoli, italy abstract this work provides an extension to an international context of the analysis made by alfano, baraldi and cantabene (2013) on the role of political competition as a channel through which electoral systems affect corruption. our result conflicts with that found by empirical literature on that topic that makes plurality rules the most virtuous in terms of corruption. political scientists must be cautious in designing the degree of proportionality of electoral rules without take into account the variation in political competition that follows. keywords political competition; electoral systems; corruption. jel classification d72; c23; k42. 5 alfano and baraldi ● political competition, electoral systems and corruption introduction the choice of an electoral system is one of the most important institutional decisions for any democracy. a country’s electoral system is the method used to calculate the number of elected positions in government that individuals and parties are awarded after elections. in other words, it is the way that votes are translated into seats in parliament or in other areas of government. there are many different types of electoral systems in use around the world, and even within individual countries, different electoral systems may be found in different regions and at different levels of government (e.g., for elections to school boards, city councils, state legislatures, governorships, etc.). the choice of a particular electoral system has a profound effect on the future political life of the country;the electoral system guarantees the representation of voters’ desires and, once chosen, often remains fairly constant. voting systems are generally divided into majoritarian/plurality rule and proportional representation (pr) with a number of variations and methods. in a democratic system, the mechanism of representation of political parties is also characterized by the degree of political competition among political parties and by the conflict between voters and candidates (the political corruption). this work deals the effect of electoral systems on the level of corruption of a country. the wide literature studying the causes of corruption (tanzi, 1998; rose-ackerman, 1999), considers the electoral system a most important political determinant of corruption. indeed, in the public sector, corruption arises and persist when bureaucrats and politicians possess discretionary power which allows to extract economic rents. this happens when institutions (political, bureaucratic, juridical and economic) are weak (aidt, 2011). electoral rules characterize those institutions, as to the degree of political competitiveness among political parties. the theoretical literature investigating the relationship between the electoral system and corruption seems to conclude that the way in which electoral rules affect corruption depends on contrasting forces while the empirical literature suggests that countries with proportional systems have much more widespread corruption than countries with majoritarian systems (see subsection 2.1). that literature, theoretical and empirical, seems to completely neglect the role played by political competition in the link between the electoral system and corruption. political competition (defined as the competition among political parties to collect votes at elections, that is, as the competition for political power (bardhan and yang, 2004)) may be an important channel through which the electoral system affects corruption. both the electoral system and political competition use the same mechanism to affect corruption: the accountability of incumbent politicians but, as for electoral rules, political competition may drive corruption in opposite directions. moreover, we may believe that the electoral system has some effect on the degree of political competitiveness among political parties. here we are interested in underlining that the literature, in particular the empirical literature, did not consider that political competition might affect corruption in combination with electoral systems. alfano, baraldi and cantabene (2013) (hereafter abc) were the first to attempt this unexplored issue. the deep analysis of the complex web wrapping electoral systems, political competition and corruption motivated the hypothesis formulated by abc that the electoral system may affect corruption directly and indirectly, via political competitions.the two effects may drive corruption in the same direction or in the opposite one; the total effect of the degree of proportionality of the electoral formula on corruption is the sum of the two effects. abc use the suitable italian scenario and the gallagher disproportionality index as a measure of the degree of proportionality of an electoral system to test their hypothesis. they find that the way in which corruption is affected by the proportionality degree of an electoral system (that is, the total effect) depends on how the degree of political competition reacts to changes in the degree of proportionality of the electoral rule. the last issue is still unexplored. this powerful result underlines that it can be 6 european journal of government and economics 4(1) misleading to analyze the impact of electoral rules on corruption regardless of the role of political competition, and further investigations are encouraged. we pick up the invitation of the authors. we extend the testing hypothesis of abc to an international context, on a cross-country panel data and we use different measures of corruption, available at the cross-country level, in order to check the robustness of results. our finding confirmed that of abc. firstly, the direct effect of the degree of proportionality on corruption is positive: an increase in proportionality of the electoral rule is beneficial for corruption. this result conflicts with that of previous empirical literature on a cross-country basis. the interpretation of this contrasting result is linked to the measure of the proportionality degree of the electoral system we used which allowed us properly to consider all electoral systems variants in an empirical setting. secondly, as in abc, we find that the indirect effects matter: political competitiveness is a channel through which the electoral system affects corruption and the direction of its effect depends on the degree of proportionality of the electoral system. the paper is organized as follows. section 2 summarizes the abc framework and shows the extension we made to their model. section 3 describes the empirical model and variables we used. section 4 explores the empirical strategy and shows the results. section 5 presents the concluding remarks. the framework and the extension before describing our work, we briefly summarize the general framework of the existing literature on the link between electoral systems and corruption which the abc analysis refers to. the abc framework the theoretical literature has explored the impact of electoral rules on corruption according to two dimensions: the district size (i.e. the number of seats in a district) and the electoral formula (i.e. how votes are translated into seats). regarding the district size, pr promotes competition among politicians reducing the possibility of rent for incumbents (myerson, 1993; ferejohn, 1986); instead, looking at the electoral formula, the greater accountability of politicians induced by majoritarian representations provides a lower incentive for corruption than in pr (persson and tabellini, 1999a,b; 2000). therefore, from the theoretical point of view, the effect of the electoral system on corruption goes in opposite directions. the empirical literature suggests that countries with proportional systems have much more widespread corruption than countries with majoritarian systems (persson et al., 2003; gagliarducci et al., 2011; kunicova and rose-ackerman, 2005). the study of effects of political competition on the economic variables is limited. political competition may affect economic performance via the quality of politicians (besley et al., 2010; padovano and ricciuti, 2009; alfano and baraldi, 2012). in the political economy literature, the concept of political competition seems close to that of accountability for incumbents (persson et al., 1997): if political competition is intense, the incumbent politician is more accountable for his actions in office and has an incentive for good performances because, otherwise, he can be easily removed and replaced by the public, with challengers. therefore, according to this concept, an intense political competition leads to less corruption (mulligan and tsui, 2006).otherwise, when political competition is intense, the electoral base of each party tends to be smaller, the probability of re-election reduces and politicians have an incentive to adopt myopic behavior maximizing rents during their remaining time in office, and corruption increase (stigler, 1972). also for political competition, its effect on corruption is difficult to define. 7 alfano and baraldi ● political competition, electoral systems and corruption while it is widely documented that the party system is largely determined by the choice of the electoral system (duverger, 1954; cox, 1997; lijphart, 1994; 1999; sartori, 1976; taagepera and shugart, 1989), there is no evidence about the relationship between the latter and political competitiveness among parties. the number of political parties competing at the elections does not measure the degree of political competitiveness among them. as sartori (1976) pointed out, for political competition, it is important, indeed, to consider the relative size of political parties (this will justify the choice of the herfindahl index). the electoral system and political competition use the same mechanism in order to affect corruption and it is not hard to assume that the electoral system has some effect on the degree of political competition. this last consideration, within the framework of political determinants of corruption, motivated the abc analysis. abc advance in the empirical literature on the link between electoral systems and corruption in two ways. firstly, they were the first that consider the role of political competition in the relationship between the electoral system and corruption. they argued that electoral systems, political competition and corruption are wrapped in a complex web and formulate the hypothesis that there is a direct and an indirect effect of electoral rules on corruption, the latter via political competition. secondly, they distinguished electoral systems by using a continuous measure of their degree of proportionality (the gallagher disproportionality index), differently from previous works which did that by using dummy variables (persson et al., 2003). indeed, the electoral rules a country decides to adopt defines the way in which votes obtained by political parties are translated into seats in parliament; it defines the degree of proportionality of the electoral system. therefore, in order to properly consider electoral systems in an empirical setting, a measure of its degree of proportionality is the correct way. a further advantage of such a measure of proportionality is the possibility to consider mixed electoral rule, beside the pr and majoritarian. indeed, mixed electoral systems, combining pr and majoritarian elements, are more likely to be characterized by intermediate degrees of proportionality. abc exploited the gallagher disproportionality index (gallagher, 1991) in order to treat italian mixed systems, that is, to differentiate mixed rules that alternate according to their degree of proportionality during the time span they consider for the analysis. they computed the gallagher index using the electoral outcomes of the senate elections from 1979 to 2006 for the 20 italian regions. abc tested their hypothesis on a sample of the 20 italian regions since 1979 to 2005 arguing that it is a suitable scenario because of the particular characteristics of corruption and the electoral system. the hypothesis of the abc analysis is that direct and indirect effects of electoral rules on corruption may drive corruption in the same direction or in the opposite direction depending on how the degree of political competition reacts to variations in the degree of proportionality of the electoral rule; therefore, the total effect of the electoral system on corruption is the sum of the two described effects. abc measure the degree of political competitiveness among political partiesthrough the normalized herfindahl index over the votes of each political party at elections from 1979 to 2006. the indirect effect of the electoral system on corruption has been caught by an interaction variable constructed by multiplying the two political indices just above described. as a dependent variable, abc used the number of crimes against public administration and estimated a distributed lag model, where corruption is regressed on the past values of regressors. the reason of the choice of a distributed lag model relies on the kind of dependent variable they used. indeed, corruption crimes perpetuated in a given year may be actually detected contemporaneously or with lags; crimes committed at different times can be detected contemporaneously; there is a lag between the beginning of the investigation and the conclusion of the penal action. thus, the empirical model should allow for lags between the year the crime 8 european journal of government and economics 4(1) is committed and that of the sentence, that justifies the estimation of a distributed lag model, where corruption is regressed on the past values of regressors, with lags from 2 to 4 years. abc found the following results: • the direct effect of the degree of disproportionality of an electoral rule on corruption is positive: the more the degree of proportionality of the electoral system the less the level of corruption. • the indirect effect shows that the way in which political competition affects corruption depends on the degree of proportionality of the electoral rule: there is a threshold of the degree of proportionality that allows us to separate an increase from a decrease of corruption due to an increase in the concentration of votes in the hands of political parties. • the total effect depends on how political competition reacts to changes in the degree of proportionality. if the gallagher index is below the threshold and if it is assumed that political competition moves in the same direction as the degree of proportionality of the electoral rule, the beneficial (negative) effect on corruption of an increase (decrease) in the degree of proportionality of the electoral system is reinforced by an increase (decrease) in political competition; otherwise,if the two indexed move in opposite direction, the two effects go in opposite direction and the total effect on corruption is indeterminate. when, instead, the value of the gallagher index goes up to that threshold and an increase (decrease) in political competition follows an increase (decrease) in proportionality, the direct and indirect effects drive corruption in opposite directions; if the variations of the two political variables are reversed, the overall effect on corruption is the same. table 1 below offers a graphical illustration of the abc results. table 1. effects of the degree of proportionality of the electoral system on corruption effect on the level of corruption below the threshold above the threshold direct (δgdi) + (-) + (-) + (-) + (-) indirect (δnhi) + (-) (+) + (-) -(+) total + (-) +/+/+ (-) authors’ elaboration the authors conclude, firstly, that this “within country” result contrasts with the previous cross-country one that, instead, stated that countries with pr have much more widespread corruption thancountries with majoritarian representations (persson et al., 2003); secondly, that, assuming a variation in the degree of proportionality of the electoral system only, is not sufficient to establish what is the consequent trend of corruption; how political competition reacts to some changes in the degree of proportionality should allow us to draw conclusions. the relationship between the electoral system and political competition still remains unexplored. the extension the hypothesis that abc test and the results they found are very interesting within the political determinants of corruption framework. but, as they suggest, further investigations in that topic are needed. the present work tries to fill some weaknesses of the abc analysis; in detail, we are referring to: 1) the “within-country” analysis. italy certainly provides a suitable scenario, but results (as the authors stressed) could be unresponsive to other realities; 2) the gallagher index, although it is on the italian regional basis, was computed under the same electoral rule over time. that is, changes in the italian electoral 9 alfano and baraldi ● political competition, electoral systems and corruption system during the period under exam concerned all the regions at the same time. this is the reason why the gallagher index showed little variability across regions (indeed, it ranges from 0.02 to 0.52); 3) the cross-sectional dimension of the italian panel data is short (n = 20 italian regions); this may create some problem in estimations using the system gmm technique which, instead, is designed for panel with short t and long n. thus, we test the same hypothesis as abc to a cross-country scenario. we conduct the empirical analysis on a panel of 85 countries over 28 years (from 1984 to 2010). this extension to abc’s work allows us to compare our results with those previously found by the literature. it seems very interesting to the light of the more suitable measure we used to distinguish electoral systems and their degrees of proportionality that were never done before. moreover, the panel dimensions are perfect to exploit the system gmm estimation technique. the empirical model and variables the estimated equation of country i at time t is 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡 = ∑𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡−𝑗𝑗 + 𝛽𝛽1𝐺𝐺𝐺𝐺𝐺𝐺 + 𝛽𝛽2𝐻𝐻𝐺𝐺 + 𝛽𝛽3(𝐺𝐺𝐺𝐺𝐺𝐺 ∗ 𝐻𝐻𝐺𝐺) + ∑𝛿𝛿 𝐶𝐶𝑟𝑟𝑟𝑟𝐶𝐶𝑟𝑟𝑟𝑟𝑟𝑟𝐶𝐶𝐶𝐶𝑟𝑟𝑖𝑖,𝑡𝑡 + +𝛼𝛼𝑖𝑖 + 𝜇𝜇𝑡𝑡 + 𝜀𝜀𝑖𝑖,𝑡𝑡 (1) where αi is a country specific effect, µt is a time-specific effect. the dependent variable is the corruption index (thereafter corr) provided by the international country risk guide icrg. (even though the icrg database includes a collection of records for about 150 countries, our analysis is cut off from some countries which showed few observations. therefore we reduced the dataset to 85 countries. table a1, appendix, provides a full description of the variable.) it is a measure of “perceived” corruption and is one of the three most popular indices based on corruption perception. the other two are the corruption perception index (by transparency international) and the control of corruption index (by the world bank). it summarizes the valuation of corruption within the political system; in particular, it deals with the warning for foreign investments. the corruption index ranges in an interval [0 to 6] where 0 represents the highest risk of corruption and 6 is the lowest risk; it provides the longer time series of corruption data, from 1984 to 2010. tables a2 and a3, appendix, show respectively the descriptive statistics of corr and partial correlation. figure 1 below shows an overview of the corruption index distribution for different countries. for each country in the figure we calculated the mean over years (19842010). to the left with a high index value (meaning low corruption risk) we find the scandinavian countries and the three countries of oceania (australia, new zealand and papua new guinea). european countries in the dataset show low/medium levels of corruption while countries in asia, africa and south america have the highest value. figure 1. mean of corruption index over the years 10 european journal of government and economics 4(1) the dynamic panel data model (as equation (1)) can be identified only for stationary time series cross-sectional data; therefore, a panel test for unit roots on the dependent variable corr is needed. we perform the fisher-type test developed by maddala and wu (1999). a fisher-type test combines the p-values from n independent unit root tests. it is based on the p-values of individual unit root test; the null hypothesis is that all series are non-stationary against the alternative that at least one series in the panel is stationary. we chose this test because it does not require a balanced panel. at 1 percent we reject the null hypothesis of non-stationary series, thus our dependent variable does not show unit root. we perform the fisher-type test including drift and 1, 2 and 3 lags in the individual adf regressions. we always reject the null. the test is not shown. two lags of the dependent variable are introduced in the estimated equation because of the dynamic of corruption (aidt, 2003). estimations of equation (1) without lags of corr showed autocorrelation of residuals. in order to solve this problem, we introduced two lags of the dependent variable in the right-side of equation (1). the two regressors of our interest are political indices. the first one, as mentioned above, is the gallagher disproportionality index. as in abc, we use it as a measure of the degree of proportionality of the electoral system. as for the within-country analysis, this is a novelty also in the cross-country empirical literature on the effects of electoral systems on corruption. the gallagher index (or least squares index) is a representation index of political parties within a parliament; it may be considered as a very good proxy for the degree of proportionality of an electoral system because the electoral system that guarantees a greater representation of political parties is a more proportional one while the less representative one is less proportional. moreover, empirical studies have shown that a majoritarian system produces a higher level of disproportionality than a proportional representation system (lijphart, 1994; anckar, 2001), whereas a mixedelectoral system produces an intermediate level (powell and vanberg, 2000; anckar, 2001). the gallagher index (thereafter gdi) is constructed as 𝐺𝐺𝐺𝐺𝐺𝐺 = � 1 2 � (𝑣𝑣𝑖𝑖 − 𝑟𝑟𝑖𝑖)2 𝑖𝑖 where vi and si are respectively the share of votes and of seats of a single political party (i=1,....,n political parties) at elections in each country in the time span under consideration. the index may range from 0 to 100 with 100 indicating perfect proportionality between seats and votes and 0 meaning that the only seat at stake goes to the winner (in which case the index equals the percentage of votes obtained by the defeated candidate). clearly the bounds of the gdi (0 and 100) are only theoretical values. among the investigated countries, the gdi ranges from 0.26 to about 33 (see table a2, appendix, for the descriptive statistics of the gdi). the dataset comprises countries that experienced majoritarian, proportional and mixed systems. the upper bound of the gdi (33.25) is very far from the theoretical value of 100 of perfect disproportionality. this means that also countries under plurality rules have a relatively strong proportionality. therefore, all the three systems have a certain degree of proportionality; moving from pr to majoritarian systems, such degree of proportionality decreases. the other political index measures the political competitiveness among political parties at elections. it is the herfindahl index (hi) calculated as: 𝐻𝐻𝐺𝐺 = �𝑣𝑣𝑖𝑖2 𝑛𝑛 𝑖𝑖=1 11 alfano and baraldi ● political competition, electoral systems and corruption where vi is the vote share of a single political party at elections in each country from 1984 to 2010 and n is the number of political parties at each election. it ranges from 0 (theoretically perfect competition with n equally sized parties) to 1 (monopoly) and it is open on the lower bound. the herfindahl index is usually used to measure the size of firms in relation to an industry; therefore, it is an indicator of the amount of competition among them. following stigler (1972), which interprets competition in the market for votes as competition in the goods’ market (the more competitive the parties, the more responsive the political system will be to the desires of the majority), an index of goods’ market power seems the correct way to measure the market for votes’ power. abc employed the normalized herfindahl index because of its desirable properties for that kind of analysis. we are forced to use the standard herfindahl index because we are unable to collect the number of political parties at elections in each country, required to construct the normalized herfindahl index. table a2, appendix, summarizes the descriptive statistics of this index. the direct effect of the degree of proportionality on corruption is caught by the coefficient β1 in eqaution (1). the indirect effect is caught by the coefficient β3 of the interaction term gdi*hi. de haan and seldadyo (2005) in their survey on the causes of corruption, detect tens of such determinants. among them, we chose, as control variables, those we believed the more suitable for the analysis we will perform. therefore, control variables are: • per capita gdp, in natural log (lngdp); it controls for structural differences in economic development (de haan and seldadyo, 2005). we expect its positive correlation with perceived corruption (ades and di tella, 1999; la porta et al., 1999; treisman 2000). hall and jones (1999) and kaufmann et al. (1999) question the causal relationship between corruption and income: the per capita gdp is high because of low corruption. for this reason we treat lngdp as endogenous. • population (pop); it controls for country size. empirical literature found contrasting evidence (knack and azfar, 2003; tavares, 2003). • government stability (gov_stab); it controls for quality of government. the higher the quality of government, the lower the probability of corruption (de haan and seldadyo, 2005). • democratic accountability (dem); it controls for the level of democracy of a country. there is a general consensus that democracy reduces corruption (de haan and seldadyo, 2005). • freedom of press (press); it controls for democratic governance. informed voters are better able to hold elected officials accountable for their policy decisions; the greatest part of people get their information via the media (snyder and stromberg, 2008). this variable is found to be negatively correlated with corruption (brunetti and weder, 2003). • law and order (law_order); it controls for the rule of law as a measure of the confidence that agents have in the rules of society, the effectiveness of judiciary and the enforceability of contracts (de haan and seldadyo, 2005). a stronger rule of law reduces the likelihood of corruption to take place. also in this regard, an issue of causality may emerge: agents may have trust in the rule of law because corruption is low. in order to take this problem into account, in some estimations we treat law_order as endogenous. • woman (wom); it is the proportion of seats held by women in national parliaments (percentage); it controls for the gender dimension of corruption meaning that conventional wisdom states that women in public life can be an 12 european journal of government and economics 4(1) effective anticorruption strategy because women are less corruptible than men (dollar et al., 1999; goetz, 2004; sung, 2003). • general government consumption expenditure (g); it controls for government size. there is no consensus among authors on the theoretical relationship between government size and corruption (abdiweli and hodan, 2003; bonaglia et al., 2001; fisman and gatti, 2002). we normalize general government consumption expenditure in percentage of gdp and per capita. • net enrollment primary rate, in natural log (lnschool); it controls for the human capital development. empirical literature found contrasting evidence (ali and isse, 2003; frechette, 2001). tables a.1, a.2 and a.3, appendix, show respectively the detailed description of all the variables, the statisticsand the correlation matrix. • we follow standard practice of counting a country as democratic according to its rate of polity iv political freedom score; we define as a democracy a country which scores a polity iv index greater than +3 in the year of the election (gleditsch and hegre, 1997). see table a.1 for a detailed description of this index and table a.2 for its descriptive statistics. in order to generalize the estimation results, we will provide robustness checks by using other measures of corruption available at cross-country level. empirics empirical strategy equation (1) is a dynamic panel data model which has been estimated using arellano-bover (1995)/blundell-bond (1998) system gmm estimators;1 estimation results are shown in table 2.the empirical analysis has been conducted on a panel of 85 countries over 28 years (from 1984 to 2010). in order to control for heteroskedasticity, every estimated equation has cluster-robust standard errors. the second-last raw of table 2 (see sub-section 4.1) shows the chi-squared (and the pvalue in parentheses) of the hansen test whose null hypothesis is that the overidentification restrictions are valid; we do not reject the null and the model is correctly specified. (we also compute, but we do not show, the difference-in-hansen test in order to test the joint validity of the full instrument set; we do not reject the null.) the last raw of table 2 displays the p-value of the arellano-bond test for second-order autocorrelation in the first differenced residuals; the null hypothesis is the absence of autocorrelation of residuals that we always accept.in order to control for common shocks in a given year, calendar year dummies are included. (in table 2 of estimation results we do not display such dummies.) every specification in table 2 is estimated by the two-step options with the windmeijer (2005) correction. windmeijer (2005) finds that the two-step efficient gmm performs somewhat better than one-step in estimating coefficients, with lower bias and standard errors, and that the two-step estimation with corrected errors is superior to robust one-step. we start estimating equation (1) including the two typical controls in cross-country analysis, the (log of) per capita gdp and the population size; in order to test the robustness of results, in the following specifications we add, step by step, all the control variables described above. all the regressors in equation (1) are introduced contemporaneously: given that our dependent variable is an index of perceived corruption, we have no reason to believe that this perception by citizens is affected 1 we used the stata command xtabond2 provided by david roodman (2009) 13 alfano and baraldi ● political competition, electoral systems and corruption by past values of the variables. this is the reason why we do not introduce lag structure in the estimated model. an important issue here is to deal with the possibility of endogeneity of the gallagher index. first of all, the theoretical literature analyzing the link between electoral rules and corruption considers the first as a determinant of corruption and not the reverse. second, it seems unlike to think that the perception of corruption (as a menace for foreign investments) may affect the way in which electoral system is designed by politicians; third, it seems also unlike to believe that a more or less corrupt system may affect the way in which votes are translated in seats, as the electoral system does. however, an endogeneity problem may arise when dealing with political institutions, that is, there may be some omitted factors that influence electoral systems and simultaneously influence corruption.in order to verify the exogeneity of gdi we perform the c test on the gdi variable. under the null, the hansen statistic tests the validity of a subset of orthogonality conditions. to perform the c test we have to estimate two models, one where gi is exogenous and another where the gi is endogenous. the estimation of the first model gives us a hansen statistic (called h1) and the estimation of the second model gives us another hansen statistic (called h2). we need to use the same set of exogenous instruments for both estimations that is we have to assume that all the other orthogonality conditions hold, i.e. all the other included and excluded instruments remain exogenous. h1 and h2 are both distributed as a chi-squared with the dof of h2 smaller than the dof of h1. the c test on gi is simply a test of h1h2. the test statistic h1-h2 is distributed as chi-squared with dof equal to the number of regressors being tested for endogeneity (in our case 1, gdi). if it is endogenous, then h1-h2 will be high because h1 is high while h2 is not. in order to deal with the general endogeneity issue, system gmm treats the model as a system of equations—one for each time period—where the predetermined and endogenous variables in first differences are instrumented with suitable lags of their own levels (see table a4, appendix). columns (a) and (a’) display the estimates of equation (1) where gdi is treated respectively as exogenous and endogenous (only with lngdp and pop as control variables). this allows us to calculate the statistic (h1-h2). it is distributed as a chi-squared with dof=1 and it is equal to 0.03. looking at the critical value of the chi-squared distribution with 1dof, the test says that at 1 percent we do not reject the null: gdi is exogenous. moreover, the coefficients of gdi, hi and gdi*hi are significant when gdi is endogenous. as mentioned above, the per capita gdp is treated as endogenous and it is instrumented with its own lags. we treat the low and order variable as exogenous and endogenous; where endogenous, it is instrumented with its own lags. 14 european journal of government and economics 4(1) results table 2. estimation results. dependent variable: corruption index (a) (b) (c) (c’) (d) (d’) (e) (f) (f’) (g) corr (-1) 1.08*** (20.8) 1.03*** (16) 1.01*** (14.4) 1.08*** (12.5) 1.04*** (15.7) 1.04*** (14.6) 0.99*** (12) 1.05*** (15) 1.04*** (15) 0.98*** (8.03) corr (-2) 0.23*** (-5.7) 0.21*** (-5.3) 0.21*** (-4) 0.23*** (-3.3) 0.21*** (-4.4) 0.22*** (-5) 0.21*** (-4.2) 0.21*** (-4.3) 0.21*** (-4.5) -0.17** (-2.2) gdi -0.01** (-2.23) -0.01** (-2.28) -0.01* (-3) -0.01* (-1.77) 0.01*** (-3.18) 0.01*** (-2.7) -0.01** (-2.5) 0.01*** (-2.88) 0.01*** (-3.04) -0.01** (-2.1) hi -0.48** (-1.92) -0.6** (-2.03) -0.8* (-2.6) -0.36* (-1.72) -0.6** (-2.10) -0.67* (-1.86) -0.41* (-1.62) -0.54* (-1.8) -0.59** (-2.01) -0.43* (-1.8) gdi *hi 0.02*** (2.6) 0.02** (2.49) 0.03* (3.2) 0.02** (2.23) 0.03*** (2.8) 0.03*** (3.02) 0.02** (1.92) 0.03** (2.5) 0.03*** (2.63) 0.02* (1.8) lngdp 0.02 (0.7) 0.02 (0.5) -0.05 (-0.7) 0.03 (1) -0.09 (-1.1) -0.06 (0.9) -0.02 (-0.26) -0.08 (-1.09) -0.09 (-1.1) 0.03 (0.4) pop -5.77e11 (-0.5) -8.90e11 (-0.7) -2.05e10 (-1) 5.63e11 (0.8) -1.08e10 (-0.7) -1.45e10 (-0.8) 3.30e11 (0.27) -1.14e10 (-0.7) -7.88e11 (-0.4) 1.84e11 (0.1) gov_stab 0.05*** (3.1) 0.05* (2.95) 0.03** (1.96) 0.03** (2.4) 0.04 (1.54) 0.03*** (2.7) 0.03** (2.17) 0.03** (2.2) 0.03 (1.48) dem 0.05** (1.96) 0.003 (0.2) 0.003 (2.39) 0.02 (0.5) 0.004 (0.35) 0.006 (0.5) -0.01* (-0.7) press -0.004 (-1.1) law_order 0.15*** (2.73) 0.06 (0.5) 0.12*** (2.11) 0.13*** (2.6) 0.14*** (2.7) 0.10 (1.3) women 0.006** (2.14) g/gdp 0.003 (0.8) g/pop 0.08* (0.8) lnschool 0.6 (0.8) n. obs. (n. groups) 1303 (70) 1298 (70) 1191 (69) 1151 (70) 1191 (69) 1191 (69) 1155 (69) 1182 (69) 1182 (69) 697 (65) n. instrum. 36 37 47 41 47 48 47 48 48 49 chi2 (pvalue) hansen test 1.1 (0.89) 1.9 (0.7) 12.4 (0.5) 6.6 (0.35) 10.4 (0.58) 11.9 (0.53) 9.36 (0.7) 11.6 (0.47) 10.2 (0.6) 17.3 (0.18) p-value 2nd order autocor 0.17 0.13 0.12 0.12 0.18 0.14 0.25 0.17 0.2 0.9 notes. all specifications contain calendar year dummies (results not reported); the time span is 19842010. the dependent variable is corr. standardized normal z-test values are in parentheses; clusterrobust standard errors. lngdp is endogenous everywhere and it is instrumented with its own lags; law_order is endogenous in (d’) and it is instrumented with its own lags. significant coefficients are indicated by * (10% level), ** (5% level) and *** (1% level).two-step estimations with windmeijer (2005) correction. specification (a) contains only the per capita gdp (in natural log) and population; the specifications which follow contain all the control variables we described above. the signs and significance of the political indices of interest do not change in every specification. at the cross-country level, our findings exactly confirm those in abc. the coefficient of the gdi is negative everywhere; recalling that it is a disproportionality index, the more the degree of proportionality of the electoral system, the less the level of corruption (remember also that the higher the value of the corruption index, the lower the level of corruption). this result states that the direct effect of the proportionality of the electoral system on corruption is positive also at the cross-country level and contradicts the previous empirical findings of persson et al. (2003). this may be due to the fact that we distinguish electoral 15 alfano and baraldi ● political competition, electoral systems and corruption systems according to their degree of proportionality instead of dummy variables; in our opinion, as stated above, the measure of proportionality is the correct way to identify electoral rules in an empirical setting. the magnitude of β1 is 0.01 meaning that if the proportionality increases, for example, by 0.1, the level of corruption decreases by 0.001. the hi shows a negative and significant coefficient equal, on average, to 0.55: a decrease in this index means an increase in political competition which is beneficial for corruption. finally, the novelty of the present work (as that of abc) is the interpretation of the interaction term gdi*hi, capturing the indirect effect of the degree of proportionality of the electoral system on corruption. looking at the specification (a), we can write (2) that is, the effect of political concentration on corruption depends positively on the degree of disproportionality of the electoral system. in equation (2), there is a threshold value of the gdi (gdi=24) such that if gdi> 24, hi corr ∂ ∂ > 0, if gdi<24, hi corr ∂ ∂ < 0. (the threshold value of gdi is the value such that hi corr ∂ ∂ = 0 in equation (2). this is equal to 24.) therefore, an increase in political competition may have a positive or negative impact on corruption depending on whether the variable gdi is above or below that threshold value. the threshold values slightly change according to the estimated coefficients of specifications in table 2. we can now deal with the total effect of the degree of proportionality of the electoral system on corruption. it depends on how political competition reacts to changes in the degree of proportionality of the electoral system. where electoral systems are characterized by high degrees of proportionality, an increase (decrease) in their degrees of proportionality followed by an increase (decrease) in political competition, is beneficial (is not beneficial) for corruption because the direct and the indirect effects go in the same direction. otherwise, under less proportional electoral systems (gdi above 24), an increase (decrease) in political competition which follows an increase (decrease) in proportionality, leads to the direct and the indirect effects in opposite directions. in this situation, in order to reduce corruption, an increase in the degree of proportionality should be followed by a reduction in political competitiveness. therefore, if the change in political competition has the same sign as the change in the degree of proportionality of the electoral formula, the indirect effect strengthens the direct effect under more proportional rules and it mitigates the direct effect on corruption under less proportional regimes. in terms of the data, only few countries in few years of elections experienced a degree of proportionality greater than 24. (those countries are albania, bahamas, france, jamaica, mongolia, papua new guinea, philippines, sri lanka, trinidad & tobago and turkey.) this means that, according to our estimation, the relevant case to be considered is hi corr ∂ ∂ < 0. now look at the other variables. past values of corr are always highly significant and the long-run effect is positive, as expected. the introduction of all the control variables does not change the results.the lngdp and pop are never significant. the government stability (gov_stab) is positive and significant, as expected, almost everywhere with magnitude, on average, of 0.03: an increase of 0.1 in government stability index leads to an increase in the corruption index (that is, a decrease in corruption) of 0.003. in columns (c) and (c’) we introduce demand press: they are gdi= hi corr ∗+− ∂ ∂ 0.0248.0 16 european journal of government and economics 4(1) both variables related to the democratization of countries. [d]emis positive, as expected, and significant in (c); when we introduce press in (c’), dem loses significance and press is not significant. we drop this latter from the following estimations. [l]ow_order is positive everywhere and significant only where it is exogenous (in (d’) it is treated as endogenous): a stronger rule of law is confirmed to deter corrupt behaviour. a positive sign is confirmed also for the presence of women in the country’s parliament. as a proxy for the government size, we control for both public consumption spending/gdp and public consumption spending/population. only the second one is significant with positive sign. finally, the rate of schooling seems to have no relevance in the explanation of corruption. robustness analysis we perform a further robustness check of the analysis, concerning the dependent variable. we test the same model with the two other most widely used indicators of corruption worldwide: the corruption perceived index (hereafter cpi) and the control of corruption index (hereafter c_c). the cpi measures the perceived levels of public sector corruption (cpi, 2012). based on expert opinion, countries are scored from 0 (highly corrupt) to 10 (very clean). c_c index reflects perceptions of the extent to which public power is exercized for private gain capturing all the forms of corruption by which elites and private interests take advantage from public sector. it ranges from approximately -2.5 (high corruption) to 2.5 (low corruption). (see tables a1 and a2 for the description and statistics of those variables.) the estimation results are in table 3 where we introduce, as controls, only lngdp and pop.2 table 3. estimations. dependent variables: cpi and c_c (h) dep. var.: cpi (i) dep. var.: c_c cpi (-1) 0.05 (0.9) cpi (-2) 0.46** (2.1) c_c (-1) -0.08 (0.5) gdi -0.08* (-1.9) -0.05** (-2) hi -1.81* (-1.9) -1.78* (-1.82) gdi *hi 0.25** (2) 0.11* (1.65) lngdp 0.4 (1.2) 0.33* (1.66) pop -2.15e-10 (-0.5) -6.63e-10* (-1.21) n. obs. (n. groups) 734 (67) 540 (69) n. instrum. 39 35 chi2 (p-value) hansen test 21.4 (0.2) 25.3 (0.2) p-value 2nd order autocorrelation 0.2 0.7 notes. cpi is the corruption perception index; c_c is the control of corruption index. all specifications contain calendar year dummies (results not reported); in (h) the time span is 1995-2011; in (i) the time span is 2002-2011. standardized normal z-test values are in parentheses; cluster-robust standard errors. lngdp is endogenous everywhere. significant coefficients are indicated by * (10% level), ** (5% level) and *** (1% level).two-step estimations with windmeijer (2005) correction. as we can notice, both direct and indirect effects of the degree of proportionality of the electoral system on corruption remain robust to different measures of corruption. the number of observations drastically decreases because the time span is 19952 in the estimation with c_c as dependent variable, in column (i) table 3, we introduce only one lag of the dependent variable because it is enough to remove the autocorrelation of residuals. 17 alfano and baraldi ● political competition, electoral systems and corruption 2011 for the cpi and 2002-2011 for the c_c. the estimated coefficients for gdi, hi and the interaction gdi*hi show that the threshold values of the gdi fall into the allowable range: it is equal to 7.24 when the dependent variable is the cpi, and it is equal to 16.2 when the dependent variable is the c_c. this robustness check seems very important; it indicates that the threshold value of the disproportionality index is not easily determined. the threshold value would allow to identify the direction of the indirect effect of the degree of proportionality on corruption, and therefore, to deal with the total effect. in our estimations, it depends on the corruption index we use, but, probably, it may widely vary among countries depending on their specific characteristics. concluding remarks this work extends the abc paper to an international context. the very interesting results they found, within the italian scenario, on the role played by political competition in the link between the electoral system and corruption required further investigations in order to be generalized. therefore, we test the same hypothesis as in abc on a cross-country panel data, from 1984 to 2010. our findings, firstly, contradicts previous empirical literature on cross-country data that makes plurality systems those most virtuous in terms of corruption. the interpretation of this reversed result can be due to the use of a more suitable measure of electoral rules, as a degree of proportionality index, instead of a dummy variable, that allowed us to properly consider all electoral systems in an empirical setting. secondly, they confirm that one cannot draw conclusions about the way the electoral system could be designed as a tool in fighting corruption without considering political competition. again we stress the fact that how political competition reacts to changes in the degree of proportionality of the electoral system, both theoretically and empirically, remains, until now, still unexplored. references abdiweli, ali m. and isse s. hodan (2003) ‘determinants of economic corruption: a cross-country comparison’, cato journal 22(3): 449-466. ades, alberto and rafael di tella (1999) ‘rents, competition, and corruption’, american economic review 89(4): 982-92. aidt, toke s. (2003) ‘economic analysis of corruption: a survey’, the economic journal, 113: 632-652. aidt, toke s. (2011) ‘the causes of corruption’, cesifo dice report, ifo institute for economic research at the university of munich 9(2): 15-19. alfano maria r. and anna l. baraldi (2012) ‘is there an optimal level of political competition in terms of economic growth? evidence from italy’, european journal of law and economics 39(2): 263-285. alfano, maria r., anna l baraldi and claudia cantabene (2013) ‘the role of political in the link between electoral systems and corruption: the italian case’, the journal of socio-economics 47(1): 1-10. anckar, carsten (2001) ‘effects of electoral systems. a study of 80 countries’, working paper presented at the sns seminar in stockholm. arellano, manuel and stephen bond (1991) ‘some tests of specification for panel data: monte carlo evidence and an application to employment equation’, the review of economic studies 58(2): 277-297. bardhan, pranab and tsung-tao yang (2004) ‘political competition in economic perspective’, bureau for research and economic analysis of development, working paper 78. 18 european journal of government and economics 4(1) besley, timothy, torsten persson and daniel m. sturm (2010) ‘political competition and economic performance: theory and evidence from the united states’, review of economic studies 77: 1329–1352. blundell, richard and stephen bond (1998) ‘gmm estimation with persistent panel data: an application to production functions’, institute for fiscal studies, working paper series, n. 99/4. bonaglia, federico, jorge braga de macedo and maurizio bussolo (2001) ‘how globalization improves governance’, discussion paper n. 2992, centre for economic policy research, organisation for economic co-operation and development, paris, france. brunetti, aymo and beatrice weder (2003) ‘a free press is bad news for corruption’, journal of public economics 87(7-8): 1801-1824. cox, gary w. (1997) making votes count: strategic coordination in the world’s electoral systems. cambridge: cambridge university press. cpi. (2012) url (consulted june 2015): http://www.transparency.org/cpi2012/in_detail de haan, jakob and harry seldadyo (2005) ‘the determinants of corruption: a reinvestigation’, unpublished paper prepared for the epcs-2005 conference durham, england, 31 march–3 april 2005. dollar david, raymon fisman and roberta gatti (1999) ‘are women really the fairer sex? corruption and women in government’, world bank working paper series 4. duverger, maurice (1954) political parties: their organization and activity in the modern state. london, wiley. new york: methuen. fisman, raymon and roberta gatti (2002) ‘decentralization and corruption: evidence across countries’, journal of public economics 83(3): 325-345. forejohn, john (1986) ‘incumbent performance and electoral control’, public choice 50: 5-25. frechette, guillaume r. (2001) ‘a panel data analysis of the time-varying determinants of corruption’, paper presented at the epcs. gagliarducci, stefano, tommaso nannicini and paolo naticchioni (2011) ‘electoral rules and politicians’ behavior: a micro test’, american economic journal: economic policy 3(3): 144-174. gallagher, michael (1991) ‘proportionality, disproportionality and electoral systems’, electoral studies 10(1): 33-51. gleditsch, nils p. and havard hegre (1997) ‘peace and democracy: three levels of analysis’, journal of conflict resolution 41(2): 283-310. goetz, anna m. (2004) political cleaners: how women are the new anti-corruption force. does the evidence wash? url (consulted may 2015): http://www.u4.no/document/showdoc. cfm?id=124. hall, robert e. and charles i. jones (1999) ‘why do some countries produce so much more output per worker than others’, quarterly journal of economics 114: 83-116. kaufmann, daniel, aart kraay and massimo mastruzzi (1999) ‘governance matters’, world bank policy research working paper 2196. knack, stephen and omar azfar (2003) ‘trade intensity, country size and corruption’, economics of governance 4: 1-18. 19 alfano and baraldi ● political competition, electoral systems and corruption kunicova, jana and susan rose-ackerman (2005). ‘electoral rules and constitutional structures as constraints on corruption’, british journal of political science 35(4): 573–606. la porta, rafael, florencio lopez-de-silanes, andrei shleifer and robert w. vishny(1999) ‘the quality of government’, journal of law, economics, and organization’ 15(1): 222-79. lijphart, arend (1994) electoral system and party systems: a study of ttwentyseven democracies, 1945-1990. new york: oxford university press. maddala, gangadharrao s. and shaowen wu (1999) ‘a comparative study of unit root tests with panel data and new simple test’, oxford bullertin of economics and statistics, pp. 631-652. mulligan, casey b., tsui, kevin t. (2006). ‘political competitiveness’, nber working paper 12653. myerson, roger b. (1993) ‘effectiveness of electoral systems for reducing government corruption: a game-theoretic approach’, games and economic behavior 5(1): 118–32. padovano, fabio and roberto ricciuti (2009) ‘political competition and economic performance: evidence from the italian regions’, public choice 138(3-4): 263–277. persson, torsten and guido e. tabellini (1999a) ‘the size and scope of government: comparative politics with rational politicians’, european economic review 43(4-6): 699–735. persson, torsten and guido e. tabellini (1999b) ‘political economics and public finance’, in auerbach, alan j., feldstein, martin. (eds.) handbook of public economics. vol. 3. amsterdam: northholland. persson torsten and guido e. tabellini (2000) political economics: explaining economic policy. cambridge: the mit press. persson, torsten, guido e. tabellini and francesco trebbi (2003) ‘electoral rules and corruption’, journal of the european economic association 1(4):958-989. persson, torsten, gerard roland and guido e. tabellini (1997) ‘separation of powers and political accountability’, quarterly journal of economics 112(4): 1163– 1202. powell, g. bingham and georg s. vanberg (2000) ‘election laws, dis-proportionality and median correspondence: implications for two visions of democracy’, british journal of political science 30(3): 383–411. roodman, david. (2009) ‘how to do xtabond2: an introduction to difference and system gmm in stata’, stata journal 9: 86-136. rose-ackerman, susan (1999) corruption and government, causes, consequences and reform. cambridge: cambridge university press. sartori, giovanni (1976) parties and party systems: a framework for analysis. cambridge: cambridge university press. snyder, james m., stromberg, david (2008) ‘press coverage and political accountability’, journal of political economy 118(2): 355–408. stigler, george j. (1972) ‘economic competition and political competition’, public choice 13(1): 91–106. sung, hyung jin (2003) ‘fairer sex or fairer system? gender and corruption revisited’, social forces 82(2): 703-723. taagepera, rein and matthew shugart (1989) seats and votes: the effects and determinants of electoral systems. new haven: yale university press. 20 european journal of government and economics 4(1) tanzi, vito (1998) ‘corruption around the world: causes, consequences, scope and cures’, imf staff papers 45(4): 559–94. tavares, jóse (2003) ‘does foreign aid corrupt?’, economic letters 79: 99-106. treisman, daniel (2000) ‘the causes of corruption: a cross-national study’, journal of public economics 76(3): 399–57. windmeijer, frank (2005) ‘a finite sample correction for the variance of linear efficient two-step gmm estimators’, journal of econometrics 126(1): 25–51. 21 alfano and baraldi ● political competition, electoral systems and corruption appendix table a1: variables description corr corruption index. it summarizes the valuation of corruption within the political system; in particular, the presence of corruption is a threat to foreign investment because it ‘distorts the economic and financial environment; reduces the efficiency of government and business by enabling people to assume positions of power through patronage rather than ability, and introduces an inherent instability into the political process’ (http://www.prsgroup.com/icrg_methodology.aspx). the result is that corruption makes it difficult to conduct business and, in some cases, it may force the withdrawal of investments. source: icrg, 1984-2010. cpi corruption perceptions index. the corruption perceptions index (cpi) ranks countries and territories based on how corrupt their public sector is perceived to be. it is a composite index – a combination of polls – drawing on corruptionrelated data collected by a variety of reputable institutions. the cpi reflects the views of observers from around the world, including experts living and working in the countries and territories evaluated see more at: http://cpi.transparency.org/cpi2012/in_detail/#sthash.bqb3zgzv.dpuf source: transparency international, 1995 – 2012. c_c control of corruption index. it reflects perceptions of the extent to which public power is exercized for private gain capturing all the forms of corruption by which elites and private interests take advantage from public sector. it is based on 30 underlying data sources reporting the perceptions of governance of a large number of survey respondents and expert assessments worldwide. the used data are selected from the worldwide governance indicators (wgi) research dataset which estimate the quality of governance. the estimated data of governance ranges from approximately -2.5 (weak) to 2.5 (strong) governance performance. source: worldwide governance indicators by world bank, 1996 -2011 (with missing in 1997, 1999, 2001). gdi gallagher disproportionality index. source: gallagher electoral disproportionality data,1945-2011 http://www.tcd.ie/political_science/staff/michael_gallagher/elsystems/docts/electionindices.pdf. hi herfindahl index. source: gallagher electoral disproportionality data,1945-2011 http://www.tcd.ie/political_science/staff/michael_gallagher/elsystems/docts/electionindices.pdf. lngdp natural logarithm of gross domestic product at constant price 2000 us. source: world bank, 1980-2011. pop urban population refers to people living in urban areas as defined by national statistical offices. source: world bank population estimates and urban ratios from the united nations world urbanization prospects, 1980-2011. gov_stab government stability. it is an assessment both of the government’s ability to carry out its declared program(s), and its ability to stay in office. the risk rating assigned is the sum of three subcomponents (government unity, legislative strength, popular support), each with a maximum score of four points and a minimum score of 0 points. a score of 4 points equates to very low risk and a score of 0 points to very high risk. this index ranges into the interval (0, 12). source: icrg, 1984-2010. dem democratic accountability. measure of how responsive government is to its people, on the basis that the less responsive it is, the more likely it is that the government will fall, peacefully in a democratic society, but possibly violently in a nondemocratic one. the points in this component are awarded on the basis of the type of governance enjoyed by the country in question. this index ranges into the interval (0, 6). source: icrg, 1984-2010. press freedom of press. freedom house has been at the forefront in monitoring threats to media independence which plays a key role in sustaining and monitoring a healthy democracy, as well as in contributing to greater accountability, good government, and economic development. the annual index contains the most comprehensive data set available on global media freedom; it provides numerical rankings and rates each country's media as "free," "partly free," or "not free". index score ranges from 0 to 100 cover the period 1993-2011; 0 states that press freedom is maximum and 100 means press is absolutely not free. during the period 1980-1992 the statistics published by freedom house are expressed using the status free, partly free, not free. the freedom house specifies that these status could be scored in different intervals, in particular the status free press range in the interval [0-30]; the status partly free press range in the interval [31-60] and the status not free press range in the interval [61-100]. taking these instructions, we assign the mean score of each interval of status only in the case the mean value of interval was very close to index score. source: freedom house’s annual press freedom, 1980-2014. https://www.freedomhouse.org/report-types/freedom-press#.vk_reu0u_iu law_order law and order is composed by sub-component comprising zero to three points. the law sub-component is an assessment of the strength and impartiality of the legal system, while the order sub-component is an assessment of popular observance of the law. thus, a country can enjoy a high rating – 3 – in terms of its judicial system, but a low rating – 1 – if it suffers from a very high crime rate of if the law is routinely ignored without effective sanction (for example, widespread illegal strikes). this index ranges into the interval (0, 6). source: icrg, 1984-2010. wom proportion of seats held by women in national parliaments (%). the data referred to unicameral assembly or lower chamber of bicameral assembly. these data are comparable with united nations women's indicators and statistics database – wistat published by world bank. source: parlia database, 1980-2011. http://www.ipu.org/wmn-e/classif-arc.htm, http://www.ipu.org/parline-e/parlinesearch.asp, http://databank.worldbank.org/data/views/reports/tableview.aspx g general government final consumption expenditure (% of gdp). it includes all government current expenditures for purchases of goods and services, most expenditures on national defense and security, but excludes government military expenditures that are part of government capital formation. source: world development indicators, 19802011 http://data.worldbank.org/indicator/ne.con.govt.zs. lnschool natural log of the net enrolment primary rate. it is the ratio between who are enrolled in primary school and the total population of the official primary school age. source: world development indicators http://data.worldbank.org/indicator/se.prm.nenr, 1980-2011. polity iv the polity iv index is a combined polity score ranging from -10 (strongly autocratic) to +10 (strongly democratic), reached by subtracting the autocracy score from the democracy score. the democracy and autocracy indices were originally constructed additively based on the following indicators: competitiveness of executive recruitment, openness of executive recruitment, constraints on chief executive, regulation of participation, and competitiveness of participation. scholars have reduced the index to a dichotomous measure of democracy and autocracy. a perfect +10 democracy, like australia, greece, and sweden, has institutionalized procedures for open and competitive political participation; chooses and replaces chief executives in open, competitive elections; and imposes substantial checks and balances on the powers of the chief executive. in a perfect -10 autocracy, by contrast, citizens’ participation is sharply restricted or suppressed; chief executives are selected according to clearly defined (usually hereditary) rules of succession from within the established political elite; and, once in office, chief executives exercise power with few or no checks from legislative, judicial, or civil society institutions. a polity score of -88 indicates economies in transition. source: polity iv individual country regime trends, 1946-2013: http://www.systemicpeace.org/polity/polity4.htm 22 european journal of government and economics 4(1) table a2: descriptive statistics of variables variable mean std. dev. min max observations corr 3.39 overall 1.41 0 6 n = 2160 between 1.18 n=85 within 0.76 t=25 cpi 5.02 overall 2.4 0.4 10 n = 1223 between 2.32 n=85 within 0.5 t=14 c_c 0.34 overall 1.08 -1.7 2.56 n = 1105 between 1.07 n=85 within 0.18 t=13 gdi 7.64 overall 6.54 0.26 33.25 n = 1975 between 5.46 n=85 within 3.67 t=23 hi 0.3 overall 0.13 0.06 0.89 n = 1772 between 0.12 n=79 within 0.06 t=22 lngdp 8.25 overall 1.46 4.9 10.9 n = 2566 between 1.44 n=83 within 0.22 t=31 pop 3.97e+07 overall 1.14e+08 210600 1.24e+09 n = 2688 between 1.13e+08 n=84 within 1.89e+07 t=32 gov_stab 7.63 overall 2.01 1 11.5 n = 2153 between 0.88 n=85 within 1.82 t=25 dem 4.92 overall 1.79 0 11.5 n = 2153 between 1.43 n=85 within 1.05 t=25 press 33.4 overall 19.2 5 100 n = 1762 between 18.2 n=85 within 6.2 t=21 law_order 3.93 overall 1.53 0 6 n = 2153 between 1.32 n=85 within 0.75 t=25 wom 14.4 overall 10.1 0 47.3 n = 2347 between 7.5 n=84 within 6.8 t=28 g/gdp 16.5 overall 5.98 2.9 43.4 n = 2526 between 5.16 n=83 within 3.04 t=30 g/pop 0.08 overall 0.32 3.33e-06 2.99 n = 2510 between 0.31 n=83 within 0.06 t=30 lnschool 4.48 overall 0.2 2.9 4.6 n = 1494 between 0.17 n=81 within 0.08 t=18 polity iv 8.4 overall 1.79 4 10 n=1995 between 1.73 n=79 within 0.73 t=25 23 alfano and baraldi ● political competition, electoral systems and corruption table a3. correlations cor r gd i hi lngd p po p gov_st ab de m pres s law_ord er wo m g/gd p g/po p lnscho ol corr 1 gdi 0.2 1 hi 0.2 4 0.1 7 1 lngdp 0.6 3 0.2 3 0.3 4 1 pop 0.0 9 0.0 2 0.0 8 0.05 1 gov_sta b 0.0 8 0.0 8 0.1 1 0.11 0.0 4 1 dem 0.3 5 0.2 8 0.2 6 0.38 0.0 2 0.08 1 press 0.6 0.1 9 0.1 5 -0.8 0.0 9 -0.12 0.3 1 law_ord er 0.7 0.1 8 0.2 8 0.66 0.0 6 0.12 0.3 9 -0.6 1 wom 0.3 9 0.3 8 0.1 6 0.33 0.1 3 0.06 0.2 1 -0.5 0.28 1 g/gdp 0.4 3 0.1 8 0.1 8 0.45 0.1 6 -0.01 0.1 4 0.44 0.51 0.3 7 1 g/pop 0.2 7 0.1 4 0.0 1 0.28 0.1 0.17 0.0 4 0.26 0.25 0.1 4 0.14 1 lnschool 0.2 2 0.1 8 0.3 0.6 0.0 4 -0.01 0.3 7 0.38 0.33 0.1 5 0.32 0.1 1 table a4: c test. dependent variable: corruption index (a) (a’) corr (-1) 1.83*** (20.8) 1.08*** (20.8) corr (-2) -0.23*** (-5.7) -0.23*** (-5.7) gdi -0.01** (-2.23) -0.007 (-0.17) hi -0.48** (-1.92) -0.42 (-0.7) gdi *hi 0.02*** (2.6) 0.13 (0.17) lngdp 0.02 (0.7) 0.02 (0.4) pop -5.77e-11 (-0.5) -6.23e-11 (-0.5) n. obs. (n. groups) 1303 (70) 1303 (70) n. instrum. 36 35 chi2hansen test (dof) 1.1 (4) 1.07 (3) p-value 2nd order autocorrelation 0.17 0.7 notes. all specifications contain calendar year dummies (results not reported); the time span is 19842010. the dependent variable is corr. standardized normal z-test values are in parentheses; clusterrobust standard errors. lngdp is treated as endogenous everywhere. in (a) gdi is treated as exogenous while in (a’) it is treated as endogenous and it is instrumented with its own lags. significant coefficients are indicated by * (10% level), ** (5% level) and *** (1% level). two-step estimations with windmeijer (2005) correction. 24 sustainability of italian budgetary policies: a time series analysis (1862-2013) european journal of government and economics 6(2), december 2017, 126-145 european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 sustainability of italian budgetary policies: a time series analysis (1862-2013) gordon l. brady a, cosimo magazzino b, * a department of economics, bryan school of business and economics, university of north carolina, greensboro, nc, usa b department of political sciences, roma tre university, italy * corresponding author at: department of political sciences, roma tre university, via gabriello chiabrera, 199, 00145 roma, italy. cosimo.magazzino@uniroma3.it article history. received 11 june 2017; first revision required 3 august 2017; accepted 5 september 2017. abstract. the aim of this paper is to empirically investigate the sustainability of italian national accounts in the years 1862-2013. the focus of the paper concerns applied tests related to the solvency and sustainability of fiscal policies. in fact, the growth rate of public debt should in the limit be smaller than the asymptotic rate of interest. moreover, the debtto-gdp ratio must eventually stabilize at a steady-state level. unit root and stationarity tests show that the variables are first-difference stationary. the results of structural breaks tests evidence the presence of some breaks, due to internal and external crises. thus, the applied analysis covers the entire period, but it also considers two different sub‐periods (1862‐1913 and 1947‐2013). furthermore, several cointegration tests highlight the evidence of a long-run relationship between public revenue and expenditure only for the first sub-period (1862-1913). our econometric insight reveals that italy faced sustainability problems in the republican age. keywords. fiscal policy; sustainability; time series; cointegration; italy jel classification. c22; h11; h60; o52 1. introduction the sustainability of fiscal policies is a central topic with regard to both economics and public policy. the rise of public indebtedness of many industrial countries during the last decades of the twentieth century has caused increasing concern about its potentially unfavourable effects. theoretically, equilibrium growth paths ought to be supported by adequate fiscal policy. moreover, the european union’s treaties impose the practical necessity of sustainable public accounts, keeping the public debt/gdp ratio below 60%, and the public deficit/gdp below 3%. a major question emerging from the global economic and financial crisis of 2008 is how to restore a country’s economic growth while restoring fiscal health. this is relevant to the euro area due to its dismal economic growth prospects coupled with high levels of public debt. government debt and slow growth underscore the importance of understanding the potential effects for fiscal sustainability and economic growth and the trade-offs these often conflicting goals entail. http://www.ejge.org/ 127 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 fiscal consolidation programs involve actions on public sector spending and tax rates to accomplish the goal of stimulating the economy (oecd, 2011). when longer-term structural changes are required to increase aggregate supply, governments must address: • impediments such as market structure; • how prices are set; • how public sector finance is conducted; • the borrowing and growth of government-owned enterprises; • financial sector regulation both domestically and from international agreements, and the functioning of labour markets and the rules and regulations that govern them. in addition, governments must be cognizant of the effects on the social safety net and institutions that affect social capital. fiscal structural reforms hold the potential to enhance the prospects for growth and debt reduction through use of automatic stabilizers, labour market reforms, which reduce labour taxes and social security contributions, and the effects on specific groups such as the elderly and youth, which might be adversely affected by policies. traditionally, the italian economy, the third largest economy in the eurozone, has had a high debt-to-gdp ratio. italy has been hampered by weak productivity growth and low economic growth. the italian budget has expanded continually since 1947, which we define as the republican period due to increasing economic activity and the pressures of inflation. currently at 135%, italy has the second largest debt to gdp ratio in the currency union after greece. the size of the italian economy remains smaller than in 2008, and approximately unchanged from the early years of the 21st century. the maastricht treaty (1992) required italy and other eu nations to undertake a well-focused fiscal consolidation in order to meet the maastricht reference values. at that time, italy’s debt level exceeded its gdp and the fiscal deficit was 10% of gdp. from 1986, italy had pursued fiscal consolidation policies with relatively moderate success although small setbacks had occurred periodically. during the period 1985-1990, the primary structural deficit was reduced by 1.7% of gdp. the fiscal consolidation process gained force after 1990 when external conditions were not conducive for initiating fiscal consolidations and the predicted survival rate was low (von hagen et al., 2001). the usual way pursued in literature to analyze the sustainability of fiscal policies implies stationarity and unit root tests for public debt and deficit, as well as cointegration tests between public expenditures and revenues. however, a common criticism to most of the available literature is that the econometric procedures used require a large number of observations, which is not usually the case in most tests of the intertemporal budget constraint. we try to overcome this problem by using an extended dataset covering 152 years, for the case of italy. the italian case is of interest because of the difficulties in reordering the public accounts to meet fiscal consolidation goals. this paper examines the sustainability of italian public finance policies by applying unit root and cointegration tests to the data over the period 1862-2013. moreover, we 128 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 also test for the existence of structural breaks in the sample time period. our study addresses a gap in the literature by applying autoregressive distributive lag (ardl) technique to examine this relevant issue for italy. the ardl approach is an important tool in modelling non-stationary time series data and the effect of structural breaks. our main contribution to the debate is that we work on deep time series for a single country rather than rely on panel analysis over a shorter time-span. besides the introduction, the outline of this paper proceeds as follows. section 2 provides a survey of the literature. section 3 contains an overview of the applied empirical methodology and a brief discussion of the data used. section 4 discusses our empirical results. finally, section 5 presents some concluding remarks and policy implications. 2. theoretical framework and empirical literature review the sustainability of the fiscal policies of europe and the united states is in the headlines from the early 1990s. a number of empirical studies have found that successful fiscal consolidation programs focus on cutting government spending as a percentage of gdp. many successful fiscal consolidations also reformed tax systems to lower marginal income tax rates and reduce the aftertax cost for business investment in productive assets while eliminating “special interest” tax preferences for specific firms, industries, and locations. lilico et al. (2009) found that successful fiscal consolidation programs were comprised of at least 80% government spending reductions and no more than 20% tax increases. the basic framework of the theoretical analysis on fiscal sustainability draws on recent contributions, such as hamilton and flavin (1986), macdonald and speight (1986), spaventa (1987), trehan and walsh (1988), bohn (1991a, 1991b, 1995, 1998), hakkio and rush (1991), corsetti (1991), kremers (1988, 1989), macdonald (1992), de haan and siermann (1993), vanhorebeek and van rompuy (1995), payne (1997), bravo and silvestre (2002), afonso (2005), and mendoza and ostry (2008), to name a few. recalling the intertemporal budget constraint (ibc), it is possible to present analytically two definitions of sustainability suitable for empirical testing (hamilton and flavin, 1986): (i) the value of current public debt equals the sum of future primary surpluses. (ii) the present value of public debt approaches zero in infinity. to test the absence of ponzi games, we inspect the stationarity of the first difference of the stock of public debt δggcgd t, and cointegration between primary balance, ggnpl, and the (lagged) stock of the public debt, ggcgdt-1 (bohn, 2007; afonso and jalles, 2015): ggnplt = α + βggcgd t-1 + ut [1] this ‘backward-looking’ approach implies that past increases in the level of public debt would imply larger primary balances today. according to the transversality condition, the ibc implies that the current value of the outstanding public debt is equal to the present value of the expected 129 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 future (primary) surpluses. thus, this condition constrains the public debt to growth no faster than the real interest rate. in a study with a descriptive nature, balassone et al. (2002) concluded that the consolidation of italian public finances in the 1990s has been highly successful in putting an end to endemic high deficits and preventing the country from sliding into debt default. however, while fiscal consolidation has avoided major economic and social shocks, it has not been a panacea for italian fiscal problems. in some areas of public spending it has reduced waste, but it has also induced governments to neglect allocative, distributive and stabilization issues. one of the first applied study to the solvency of italy’s public finance is baglioni and cherubini (1993), where has been analyzed the sustainability of the italian fiscal policy in the 1979-1991 period, using monthly data. the principal findings show that primary surplus is stationary, while public debt is not; permanent shocks explain about 90% of forecast error variance of public debt; debt is not sustainable even if the discount rates are considered. the paper applies some stationarity tests, but does not account for cointegration. paesani et al. (2006), focusing on the usa, germany and italy over the 1983-2003 period, studied how the accumulation of government debt affects long-term interest rates, both nationally and across borders. empirical evidence shows that in all cases a more sustained debt accumulation leads at least temporarily to higher long-term interest rates. this transitory impact also spills-over into other countries, mainly from the us to the two european countries. this paper uses a multivariate econometric model. a different empirical approach, semi-parametric estimations using penalized spline smoothing, is in greiner and kauermann (2008) tested how the primary surplus in two countries of the euro area, germany and italy, reacts to changes of public debt. italian public debt does not seem to be sustainable although consolidation efforts in the nineties have stabilized italian debt. in a similar empirical context, piergallini and postigliola (2013) investigated the sustainability of italy’s public finances from 1862 to 2012 adopting a non-linear perspective. they used a smooth transition regression approach to explore the scope for nonlinear fiscal adjustments of primary surpluses in response to the accumulation of debt. the empirical results show the occurrence of a significantly positive reaction of primary surpluses to debt when the debt/gdp ratio exceeded the trigger value of 110 percent. piergallini and postigliola (2012) examined the historical dynamics of government debt in postunification italy (1861-2009). they found that, controlling for fiscal feedback policies, the debtgdp ratio is mean-reverting. moreover, policymakers reacted to the debt accumulation taking corrective measures to avoid potential long-run sustainability problems. balassone et al. (2011) investigated the link between government debt-to-gdp ratio and real per capita income growth in italy over 1861-2009. the empirical findings support the hypotheses of a negative relation between public debt and growth, and of a stronger effect of foreign debt compared to domestic debt before world war i. the effect of public debt on growth appears to work mainly through reduced investment. they model the test on a standard production function, using a two stages least squares estimator (2sls) to take account for endogeneity, and 130 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 performed the johansen and juselius cointegration test. magazzino and intraligi (2015) studied the relationships between government debt/gdp and its macroeconomic determinants (such as primary balance/gdp, real gdp, the inflation rate and the average interest rate on treasury bills) in the period 1958-2013 in italy. consistent with the theory, the results reveal a significant causal relationship moving from the primary balance to the real growth rate, as well as a clear influence of the inflation on the interest rate. in contrast, the influence of public debt on growth rate emerges only marginally. the empirical strategy applies some causality tests (granger, toda and yamamoto), several structural breaks tests, and a vector autoregressive (var) model. alternative studies present a prospective nature. marattin and marzo (2009) investigated the consequences of the adoption of a fiscal policy rule responding to past real debt/gdp ratio on the main public finance aggregates. according their estimates, a significant and sustainable reduction of debt/gdp ratio can be achieved over the next years if policymakers raise (up to 0.30) fiscal pressure’s elasticity to public debt evolution, and/or reduce primary government expenditure by four percentage points over the next 4 years. based on the simple arithmetic of public finance, they performed a number of simulation regarding the evolution of public finance aggregates. casadio et al. (2012) analyzed possible targets for the italian debt-to-gdp ratio with a small macroeconomic model. they found that external conditions play a fundamental role for the italian fiscal consolidation. to reach a target of 100% of debt-to-gdp ratio by 2020, a further growthsustaining policy has to be implemented. spaventa (2013) underlined how, with a large stock of outstanding debt, a precondition to avoid unstable and potentially explosive outcomes is that fiscal policy aims at a reasonably low. a desirable fiscal policy program may however prove socially and politically unfeasible unless monetary policy lends some help. the sustainability of italian public accounts has been also investigated through the analysis of the relationship between government expenditure and revenues. dalena and magazzino (2012) examined the long-run equilibrium relationship between government expenditure and revenues in italy from 1862 to 1993, using cointegration and causality techniques in the long as well as in the short-run. empirical findings show that, for each sub-period, the policy adopted reflects the prevailing paradigm of public finance. in fact, the ‘tax-and-spend’ argument received empirical support from the liberal period data. in contrast, the interwar years are in line with the ‘spendand-tax’ hypothesis. finally, the ‘fiscal synchronization’ hypothesis emerges in the republican ages (magazzino, 2012a). this paper uses an error correction model (ecm), performing both the standard granger causality test and a granger non-causality test (due to toda and yamamoto). trachanas and katrakilidis (2013) evaluated the sustainability of the fiscal deficit as well as the long-run macroeconomic relationship between government spending and revenues for italy, greece and spain in the years 1970-2010. the evidence for all three countries suggests that, allowing for structural break, the fiscal deficits are weakly sustainable in the long-run, the ‘spend-and-tax’ hypothesis is supported, and the budgetary adjustment process is asymmetric in italy and spain. this paper focuses on cointegration between government spending and revenues, applying several tests: engle-granger, johansen trace, autoregressive distributed 131 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 lags (ardl) bounds, and gregory and hansen tests. buiatti et al. (2014) reconstructed the macro regional government deficits of italy. they found that the incredibly large and persistent fiscal imbalances of poorer southern regions are the ultimate cause of the national public debt of italy. they suggest the introduction of a tight set of hard budget rules and fiscal responsibility that must substitute the current set of norms and discretionary budget procedures. the empirical strategy uses a generalized method of moments (gmm) estimator, some structural breaks test, and the j-test approach. the effects of the recent economic-financial crisis are inspected in di mascio and natalini (2014), where has been analyzed the italian government’s response to the sovereign debt crisis. the findings reveal that the current crisis has been managed with straight cutback management, as public administration has been considered by policy makers just as a source of public expenditure to be squeezed rather than as a provider of public services in need of modernization so as to sustain economic growth. some related issues are discussed in magazzino (2012b), who assessed the empirical evidence of wagner’s law in italy for the period 1960-2008 at a disaggregated level, using a time series approach. the causality results show evidence in favour of wagner’s law only for passive interests spending in the long-run, and for dependent labor income spending in the short-run. while forte and magazzino (2016) empirically assessed the relationship between government size and economic growth in italy (1861-2008). the results show the presence of a non-linear relationship between the size of public sector (measured by the share of government expenditure over gdp) and the economic growth rate. in general, the presence of an inverted “u-shape” curve, which emerges for the last two decades, suggests that expenditure cuts might foster the gdp dynamic (magazzino, 2013; 2014). 3. data and methodology the first step of our empirical strategy concerns stationarity and unit root tests. according to engle and granger (1987), a linear combination of two non-stationary series can be stationary, and if such a stationarity exists, the series are considered to be cointegrated. this requires, however, that the series have the same order of integration. in order to examine the stationarity properties of the time series, we apply different unit root tests: the augmented dickey and fuller (adf, 1979) test, determining the number of lags using the hannan‐quinn criterion; the elliott, rothenberg, and stock (ers, 1996) test, setting the maximum lag according to the method proposed by schwert (1989); the phillips and perron (pp, 1988) test, selecting the bandwidth automatically in accordance to the newey‐west procedure using bartlett kernel; and the kwiatkowski, phillips, schmidt, and shin (kpss, 1992) test, with equivalent band‐width selection procedures. the tests differ with respect to their null hypotheses: the null hypothesis of the adf, ers and the pp tests is the existence of a unit root in the time series, whereas the null hypothesis of the kpss test is trend stationarity of the time series. moreover, we also checked if any of the variables have structural breaks. to this extent, the zivot and andrews (za, 1992) and the clemente, montañés, 132 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 and reyes (cmr, 1998) tests were performed (see the appendix for details). once we found that the variables are non-stationary at their levels and are in the same order of the integration, we can apply the cointegration test. as in trachanas and katrakilidis (2013), we performed several cointegration tests, to check the robustness of the results.the ardl bounds testing approach of cointegration is developed by pesaran and shin (1999) and pesaran et al. (2001). this approach has several advantages over the traditional cointegration approaches of engle and granger (1987), and johansen and juselius (1992). this takes care of small sample properties and simultaneity biasness in relationship among variables. the main constraint in the application of the conventional cointegration techniques is that they require all the variables included in the model to be non-stationary at levels but should be integrated of the same order. the present ardl approach to cointegration method surmounts this problem as it is applicable irrespective of order of integration of regressors whether i(0) or i(1) or mixture of both. apart from that, the ardl model also has advantages in selecting sufficient numbers of lags to capture the data generating process in a general-to-specific modeling framework. these meritorious features justify the use of ardl model to obtain robust estimates.the bounds testing procedure is based on the joint f-statistics or wald statistics that is tested the null of no cointegration, h0: δr=0, against the alternative of h1: δr≠0, r=1, 2, …, 4. if the calculated f-statistics lies above the upper level of the band, the null is rejected, indicating cointegration. if the calculated f-statistics is below the upper critical value, we cannot reject the null hypothesis of no cointegration. finally, if it lies between the bounds, a conclusive inference cannot be made without knowing the order of integration of the underlying regressors. the next step is to test for stability of the long-run coefficients as well as the dynamics of the short-run ones following pesaran (1997). cointegration analysis considered also the gregory and hansen (1996) test for cointegration with regime shifts. the null hypothesis (h0) is no cointegration, against the alternative (h1) of cointegration with a single shift at an unknown point in time. the gregory and hansen approach is an extension of similar tests for unit root tests with structural breaks, for example, by zivot and andrews (1992). gregory and hansen propose the cointegration tests, which accommodates a single endogenous break in an underlying cointegrating relationship. the null hypothesis of no cointegration with structural breaks is tested against the alternative of cointegration. the single break date in these models is endogenously determined. in our analysis, the log transformations of the variables have been derived. the empirical analysis uses the time series data of public expenditure (% of gdp, g) and revenue (% of gdp, r) for i t a l y i n the 1862-2013 years. we used the data recently reconstructed by forte (2011) for italy. the choice of the sub-periods is in line with studies that cover a similar time span (balassone et al., 2011; burret et al., 2013; brady and magazzino, 2017). moreover, the sample used in the regression analysis excludes the years 1915-1946 to prevent distortions from the extreme values recorded for most variables over that period because of the two world wars. the partition of the sample is also driven by major facts in italian history (early unification; world wars and fascism; republican period). moreover, structural break analysis confirm this choice. figure 1 shows the dynamic of our series. in the right-side panel, the first-differences series 133 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 are graphed. a visual inspection of the series in logarithmic form shows that there was a clear upward trend for both series after the wwii. some descriptive statistics are summarized in table 1 as a preliminary analysis. both variables have negative value of skewness in the sub-period 1947-2013, indicating that the distributions are skewed to the left. correlation analysis show that public expenditure and revenue are highly correlated in each period. figure 1. public expenditure and revenue in italy (1862-2013, (% of gdp, log-scal). source: forte (2011) and own elaboration. 4. empirical analysis the focus of the present section is the analysis of the fiscal policy sustainability for each time period in italy. in order to test for fiscal sustainability, we test whether the transversality condition is met (burret et al., 2013) by conducting various stationarity tests on public expenditure and revenues. first, the solvency of the fiscal policies pursued by italy will be tested according to the trehan-walsh procedure, i.e. by analysing the statistical properties of the deficit inclusive of interest payments (trehan and walsh, 1988). more specifically, the stationarity of the deficit-togdp time series will be the maintained hypothesis. 1. 5 2 2. 5 3 3. 5 4 1860 1880 1900 1920 1940 1960 1980 2000 2020 year log(g) log(r) -.6 -.4 -.2 0 .2 .4 1860 1880 1900 1920 1940 1960 1980 2000 2020 year d1.log(g) d1.log(r) 134 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 table 1. exploratory data analysis. 1862-2013 variable mean median s.d. skewness kurtosi s range iqr 10-trim g 3.1726 3.1760 0.5446 0.0066 1.5150 1.6608 1.1027 3.174 r 2.9109 2.8380 0.5533 0.3431 2.0341 2.1788 0.7949 2.892 1862-1914 variable mean median s.d. skewness kurtosi s range iqr 10-trim g 2.5494 2.5461 0.1197 0.0080 2.3130 0.4784 0.1856 2.549 r 2.3870 2.4352 0.1586 -1.7616 8.0278 0.8862 0.1800 2.407 1947-2013 variable mean median s.d. skewness kurtosi s range iqr 10-trim g 3.6275 3.7578 0.3055 -0.6052 1.8945 1.0379 0.5642 3.655 r 3.4065 3.4400 0.4033 -0.6542 2.8381 1.7175 0.6542 3.444 sources: our calculations on forte (2011) data. the presence of a unit root in these time series clearly reflects fiscal insolvency, implying that the solvency condition is violated (vanhorebeek and van rompuy, 1995). we have also split our time span into two periods: 1862-1913 and 1947‐2013. an additional reason to conduct such an analysis rests on the fact that longer sample periods may actually “hide” unsustainability periods in the series. several unit root and stationarity tests are used in an attempt to verify the stability conditions. since unit roots in fiscal data imply that economic shocks have a sustaining effect on the data over time, the identification of a unit root denotes a non‐stationary (unsustainable) time series. in order to take a possible distortion of structural breaks into account, we follow a twofold approach. first, we conduct the unit root and stationarity tests on the entire sample (1862‐2013) and on the two sub‐samples (1862-1913 and 1947‐2013). second, we additionally apply two tests on unit root and structural break. the stationarity tests on the 1862-2013 years are only indicative, since the whole period is characterized by large-scale events and structural breaks. if we allow for a constant, the tests indicate that both variables are non-stationary in the levels, but stationary in first-differences (integrated of order 1). if we also include a trend in the estimation, the results for their levels are ambiguous; in fact, expenditures tend to be stationary in levels with trend, as the ers and pp test statistics allow for rejecting the hypothesis of a unit root on the five percent significance level. in contrast, all proposed tests do not reject the hypothesis of stationarity for the differenced series. given these mixed results, we conclude that the analysis of the whole time series could not be meaningful. moreover, the power of standard unit root tests decreases substantially if there are significant structural breaks in the time series. therefore, we divide the sample into two sub‐ periods as discussed above. with regard to the first sub-period (1862-1913), expenditures and revenues tend to be firstdifferences stationary, in both specifications. thus, in this respect, evidence in favour of solvency is found. finally, in the last period (1947-2013), as for expenditures, if we allow for a constant, the tests 135 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 indicate that they are non-stationary in the levels, while the differenced series is stationary. if we also include a trend, expenditures continue to be i(1). on the other hand, the results for revenues are ambiguous, since, allowing for a constant, the adf (at 10 percent) and pp tests (1 percent) reject the unit root hypothesis; while including a trend in the deterministic component, the ers and pp test statistics allow for rejecting the non-stationary hypothesis. nevertheless, all proposed tests clearly indicate the absence of a unit root when the differenced series of revenues is analyzed. these findings clearly indicate that italian fiscal policies have undoubtedly been insolvent in the last sub-period. in summary, the results allow the rejection of the non-stationarity hypothesis for the entire period as well as for the two selected sub-periods. to further explore unit root properties of the variables, we supplement a za unit root test that is sensitive to structural breaks a) in the intercept, b) in the intercept and trend (table 2). table 2. results for unit roots and stationarity tests. 1862-2013 variable unit root and stationarity tests deterministic component adf ers pp kpss g constant -1.4290 -0.1598 -1.4290 1.3706*** r constant -1.0549 0.3597 -1.5968 1.3243*** g constant, trend -3.4063* -3.4309** -3.8367*** 0.0368 r constant, trend -3.0884 -2.4993 -3.4300* 0.2219*** δg constant -6.6434*** -2.6640*** -11.2386*** 0.0217 δr constant -10.8174*** -0.9545 -11.0261*** 0.0563 δg constant, trend -6.6207*** -6.4510*** -11.2009*** 0.0211 δr constant, trend -10.7720*** -1.5720 -10.9828*** 0.0562 1862-1913 variable unit root and stationarity tests deterministic component adf ers pp kpss g constant -2.2051 -1.1322 -3.3304** 0.4682** r constant -2.4543 -1.0413 -5.2232*** 0.6861** g constant, trend -1.2706 -1.1772 -3.6332** 0.1683** r constant, trend -0.6916 -1.1933 -5.4742*** 0.2742*** δg constant -9.3828*** -1.6348** -14.3553*** 0.1988 δr constant -10.1177*** -1.6606* -18.7841*** 0.3286 δg constant, trend -9.2972*** -9.3446*** -17.4027*** 0.1901** δr constant, trend -9.9661*** -2.9705* -35.8019*** 0.0948 1947-2013 variable unit root and stationarity tests deterministic component adf ers pp kpss g constant -1.8269 0.3349 -2.1533 0.9158*** r constant -2.8422* -0.1615 -4.2098*** 0.9768*** g constant, trend -0.3792 -1.0534 -0.4320 0.2219*** r constant, trend -1.7730 -3.1633** -5.2981*** 0.1320* δg constant -9.1832*** -9.3629*** -12.8693*** 0.4858** δr constant -4.6307*** -1.7669* -4.5872*** 0.4596* δg constant, trend -9.2348*** -8.6189*** -14.6196*** 0.1107 δr constant, trend -5.1860*** -4.3358*** -5.2588*** 0.1243* notes. tests are performed on the log-levels of the variables. adf, ers, pp, and kpss refers respectively to the augmented dickey-fuller test, the elliot, rothenberg, and stock point optimal test, the phillips-perron test, and the kwiatkowski, phillips, schmidt, and shin test. when it is required, the lag length is chosen according to the hqic. ***p<0.01, **p<0.05, *p<0.10. lag length based on modified sbic for adf and ers, bartlett kernel for pp and kpss. for the whole period, we find ambiguous results: in the first case, we cannot reject the hypothesis that revenues have a unit root. yet, if we also allow for a structural break in the trend, the hypothesis is rejected. the first specification test indicates a break point in 1975, the first year in which there were the effects of the italian tax reform; while the second specification isolates a 136 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 break at the end of the wwii, with the need to finance the reconstruction. while for the expenditures, both tests reject the null hypothesis of unit root only at 10 percent significance level, with a break at the outbreak of the first world war (1914). for this reason, previous ambiguous results are confirmed. comparing expenditure and revenue break points, the za results suggest that fiscal policy of the 19th century is significantly different from that of the 20th century. the significant breakpoint in 1914 is due to the sharp increase of expenditure growth to finance wwi. the cmr test shows that both for revenues and expenditures, despite the structural breaks, we are unable to reject the null hypothesis of a unit root in these series. notwithstanding, the rejection of the stationarity hypothesis does not mean that public accounts are not sustainable, as observed by trehan and walsh (1991): stationarity rejection does not necessarily imply the absence of sustainability of the government accounts. for the pre-wwi years, if we allow for a structural break in the intercept, we can reject the null hypothesis for both expenditures and revenues; and we obtain similar results when we include also the trend in the model. curiously, all tests indicate as a break point the year 1881, when the iii cairoli’s government (historical left) abolished the fiat of the lira (forte and magazzino, 2016). as regards the republican age, for expenditures and revenues we retain the null hypothesis both allowing for a structural break in the intercept and for a break in the trend. both tests indicate for revenues a break point located in the first eighties, related with the effects of the so-called “divorce” between the bank of italy and the italian ministry of treasury as well as the effects of the second oil shock. if we allow for a break only in the intercept, the break corresponds to the initial phase of the second republic. while, including a break also in the trend, the date coincides with those found for revenues (1982). we therefore can conclude that both our series are integrated of order one, or i(1) only in the second sub-period, while inconclusive results are reached for the whole sample period. the lagorder selection has been chosen according to the akaike’s information criterion (aic), the schwarz’s bayesian information criterion (sbic), and the hannan-quinn information criterion (hqic). now we can proceed to investigate fiscal sustainability in italy by testing for the existence of cointegration between public expenditure and revenues. figure 1 can supply a visual inspection of the time series and a preliminary idea. one can suspects that italy in the more recent years might not pass the sustainability tests. in table 4, we show the results of the ardl bounds cointegration tests. the empirical findings allow the rejection of the cointegration hypothesis for both equations of the whole period, and only for the equation with public expenditure as a dependent variable in the second sub-period. on the other hand, for the years 1862-1913 a cointegration relation is found for both equations. therefore, considering the results of the entire sample time period (18622013), one can conclude that fiscal policy may not been sustainable for italy since unification. 137 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 table 3: results for unit root tests with structural breaks and for additive outlier unit root tests (single structural break). 1862-2013 za tests variable (a) (b) tb k tmin tb k tmin g 1914 3 -4.777* (-4.80) 1914 3 -4.984* (-5.08) r 1975 3 -4.266 (-4.80) 1943 3 -5.539** (-5.08) cmr tests variable optimal break point k t-stat 5% critical value g 1917 6 -1.791 -3.560 r 1974 1 -3.403* -3.560 1862-1913 za tests variable (a) (b) tb k tmin tb k tmin g 1881 1 -5.706*** (-4.80) 1881 1 -5.647*** (-5.08) r 1881 1 -8.548*** (-4.80) 1881 1 -8.762*** (-5.08) cmr tests variable optimal break point k t-stat 5% critical value g 1878 6 -4.416*** -3.560 r 1878 6 -1.869 -3.560 1947-2013 za tests variable (a) (b) tb k tmin tb k tmin g 1994 2 -3.561 (-4.80) 1982 2 -3.605 (-5.08) r 1981 2 -4.311 (-4.80) 1982 2 -5.058* (-5.08) cmr tests variable optimal break point k t-stat 5% critical value g 1982 2 -3.505* -3.560 r 1982 0 -5.335*** -3.560 notes. (a) refers to the model allowing for break in intercept and (b) the model allowing for break in intercept and trend. tb is the break date endogenously selected. tmin is the minimum t-statistic. k denotes the lag length. 5% critical values are given in parentheses. ***p<0.01, **p<0.05, *p<0.10. table 4. ardl bounds test estimation results. 1862-2013 model for estimation lag length f-statistics significance level critical bound f-statistics i(0) i(1) 𝐹𝐹𝑅𝑅𝐺𝐺 1 4.355 1 6.84 7.84 2.5 5.77 6.68 𝐹𝐹𝐺𝐺𝑅𝑅 1 1.432 5 4.94 5.73 10 4.04 4.78 1862-1913 model for estimation lag length f-statistics significance level critical bound f-statistics i(0) i(1) 𝐹𝐹𝑅𝑅𝐺𝐺 1 7.306** 1 6.84 7.84 2.5 5.77 6.68 𝐹𝐹𝐺𝐺𝑅𝑅 1 6.456** 5 4.94 5.73 10 4.04 4.78 1947-2013 model for estimation lag length f-statistics significance level critical bound f-statistics i(0) i(1) 𝐹𝐹𝑅𝑅𝐺𝐺 1 2.715 1 6.84 7.84 2.5 5.77 6.68 𝐹𝐹𝐺𝐺𝑅𝑅 1 10.467*** 5 4.94 5.73 10 4.04 4.78 notes. asymptotic critical value bounds are obtained from table f‑ statistic in pesaran et al. (2001). ***p<0.01, **p<0.05, *p<0.10. 138 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 table 5. gregory and hansen cointegration tests. country constant constant and trend constant and slope constant, slope and trend dependent variable: g 1862-2013 -4.24 (1920) -4.29 (1908) -5.15** (1937) -5.66** (1937) 1862-1913 -6.85*** (1869) -6.82*** (1869) -6.28*** (1872) -6.37*** (1869) 1947-2013 -4.12 (1908) -4.01 (1884) -4.14 (1908) -6.25*** (1909) dependent variable: r 1862-2013 -3.56 (1979) -4.01 (1979) -4.36 (1943) -4.84 (1938) 1862-1913 -10.47*** (1874) -10.39*** (1874) -10.36*** (1874) -10.57*** (1898) 1947-2013 -5.58*** (1992) -5.86*** (1992) -5.74*** (1990) -5.58** (1973) notes. adf statistics are reported. 5% critical values: -4.61, -4.99, -4.95, -5.50 respectively. however, to allow for the possibility of structural breaks in the long-run cointegrating relationship, we applied the gregory and hansen (1996) cointegration test with breaks. briefly, under this procedure, a dummy variable is included to account for a shift in the cointegrating regression. the minimum adf statistic endogenously determines the breakpoint and is compared to critical values supplied by gregory and hansen (1996). the procedure offers four different models corresponding to the four different assumptions concerning the nature of the shift in the cointegrating vector. table 5 clearly confirms previous ardl bounds tests results, showing the existence of cointegration with a break for the first sub-period. table 6. cointegration of government revenues and expenditures. time period dependent variable engle-granger test ardl bound tests vector p-value vector p-value 1862-1913 r [1-0.9740]*** 0.000 [1-0.9343]*** 0.001 g [1-1.2117]*** 0.000 [1-1.3639]*** 0.000 1947-2013 r [1-0.6276]*** 0.000 [1-0.8818]*** 0.000 notes. asymptotic critical value bounds are obtained from table f‑ statistic in pesaran et al. (2001). ***p<0.01, **p<0.05, *p<0.10. only cointegrating vectors with at least a 10% significance level are reported. for the first sub-period, in the revenues equation, the estimated coefficient for public expenditure is less than one. for each percentage point of gdp increase in public expenditure, in italy during the years 1862-1913 public revenues only increase by 0.93-0.97. however, this estimated coefficient is very close to one, suggesting that in the above mentioned period of time public expenditure exhibited a slightly higher growth rate than public revenues, do not thus challenging the hypothesis of fiscal sustainability. on the other side, for the more recent period (1947-2013), the public expenditure’s coefficient in the equation where revenues are the dependent variable is less than one. here, we can state that for each percentage point of gdp increase in public expenditure, in italy in the 1947-2013 period public revenues only increase by 0.6276. in this case, public expenditure exhibited a clearly higher growth rate than public revenues, suggesting that fiscal sustainability problems emerge. as can be seen from table 7, the null hypothesis of no cointegrating relationship against alternative of at most one cointegrating relationship cannot be rejected in any of the models at a 5% level of significance, suggesting that there is no cointegrating relationship among variables. although the 5% critical values were adjusted (lifted up) in order to account for a small sample bias, the null hypothesis of no cointegration could not have been rejected even using none 139 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 adjusted 5% critical values. table 7. results for johansen and juselius cointegration tests. 1862-2013 h0 h1 modela trace eig. stat. ll sbic hqic aic none at most 1 24.5119* (25.32) 16.1212* (18.96) 284.9919 -3.5995* -3.6710* -3.7199 at most 1 at most 2 8.3907 (12.25) 8.3907 (12.52) 293.0525* -3.5733 -3.6625 -3.7740* 1862-1913 h0 h1 modelb trace eig. stat. ll sbic hqic aic none at most 1 64.6594 (25.32) 49.1299 (18.96) 103.3854 -3.9001 -3.9469 -3.9759 at most 1 at most 2 11.5296* (12.25) 11.5296* (12.52) 127.9504* -4.5551* -4.6955* -4.7824* 1947-2013 h0 h1 modelb trace eig. stat. ll sbic hqic aic none at most 1 25.4214 (25.32) 23.3952 (18.96) 259.4186 -6.8652 -7.1436 -7.3259 at most 1 at most 2 2.0262* (12.25) 2.0262* (12.52) 271.1162* -6.9634* -7.3213* -7.5557* notes. 5% critical values in parentheses. a: include a linear trend in the cointegrating equations and a quadratic trend in the undifferenced data; b: include a restricted trend in the model. furthermore, our empirical findings are in line with previous results by vanhorebeek and van rompuy (1995), who found that italian fiscal policies have undoubtedly been insolvent in the period 1970-1994. corsetti and roubini (1991) found, amongst other things, the government finances of italy to be unsustainable. caporale (1995) found that the government of italy is intertemporally insolvent. payne (1997) showed that in the case of italy the budget deficits might not be sustainable due to the lack of cointegration. moreover, cointegration is present between revenues and expenditures, although the estimated coefficient (0.63-0.88) is significantly less than one, which suggests that public expenditure was growing faster than public revenues. such a relationship between public revenues and expenditure questions the issue of sustainability. papadopoulos and sidiropoulos (1999) derived that italy may incur unsustainable deficits, so that its selection in phase 2 of the emu is questionable. uctum and wickens (2000) found that the market value of the debt-gdp for italy was not mean-reverting (1994-2000), though a general improvement in fiscal stances toward the end of the century could be noted. moreover, they concluded that fiscal policy in italy was not sustainable. bravo and silvestre (2002) found that cointegration between expenditures and revenues does not emerge in the italian case, implying that condition for sustainability does not held. in addition, afonso (2005) found that italy was one of the majority eu-15 member countries with sustainability problems. on the contrary, afonso and jalles (2014) concluded that the solvency condition would be satisfied for italy, since nonstationarity can be rejected, and, therefore, longer-run fiscal sustainability cannot. 140 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 4. concluding remarks this study has extended the research on the fiscal sustainability of the italian budgetary policies in the 1862-2013 years. unit root and stationarity tests have been conducted on the entire sample (1862‐2013), and on two sub‐ samples (1862 -1913 and 1947‐2013). the results of unit root tests allow the rejection of the non-stationarity hypothesis for the entire period as well as for the two selected sub-periods. unit root tests with structural breaks confirm previous findings. cointegration analyses reveal that for the whole period a long-run relationship does not emerge. therefore, considering the results of the entire sample time period (1862-2013), one can conclude that fiscal policy may not been sustainable for italy since unification. moreover, cointegration is present between public expenditure and revenues for the first sub-period (1862-1913), with an estimated coefficient very close to one (0.93-0.97), do not thus implying problems for fiscal sustainability. on the other hand, for the republican age (1947-2013), a long-run relationship is discovered, although the estimated coefficient (0.63-0.88) is significantly less than one, which suggests that public expenditure was growing faster than government revenues, raising some concerns about the issue of sustainability. therefore, we support a fiscal consolidation strategy and refute the perception that italian fiscal policy is on a sustainable path. the concept of practical sustainability is most relevant in the framework of the budgetary preconditions of maastricht. in other words, if italian fiscal policies were to be conducted in the future as it was in the republican age (1947-2013), there could be emerge some problems. in addition, our results are in line with empirical findings in brady and magazzino (2017). acknowledgements comments from the participants at the scottish economic society 2017 conference (perth, scotland, 24-26 april 2017) are gratefully acknowledged. we would like also to thank the anonymous referees and the editor for useful comments. all remaining errors are our own. references afonso, a. (2005). fiscal sustainability: the unpleasant european case. finanzarchiv, 61 (1), 19-44. doi: 10.1628/0015221053722532 afonso a., & jalles, j.t. (2014). a longer-run perspective on fiscal sustainability. empirica, 41(4), 821-847. doi: 10.1007/s10663-013-9240-0 afonso a., & jalles, j.t. (2015). how does fiscal policy affect investment? evidence from a large panel. international journal of finance & economics, 20 (4), 310-327. doi: 10.1002/ijfe.1518 baglioni, a., & cherubini, u. (1993). intertemporal budget constraint and public debt sustainability: 141 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 the case of italy. applied economics, 25(2), 275-283. 10.1080/00036849300000033 balassone, f., francese, m., & pace, a. (2011). public debt and economic growth in italy. quaderni di storia economica – bank of italy. balassone, f., franco, d., momigliano s., & monacelli d. (2002). italy: fiscal consolidation and its legacy. ssrn paper. baum, c. (2001). stata: the language of choice for time series analysis?. the stata journal, 1(1), 1-16. becketti, s. (2013). introduction to time series using stata. texas: stata press. bohn, h. (2007). are stationarity and cointegration restrictions really necessary for the intertemporal budget constraint?. journal of monetary economics, 54, 1837-1847. doi: 10.1016/j.jmoneco.2006.12.012 bohn, h. (1991a). budget balance through revenue or spending adjustments?. journal of monetary economics, 27(3), 333-359. doi: 10.1016/0304-3932(91)90013-e bohn, h. (1991b). the sustainability of budget deficits with lump-sum and with income-based taxation. journal of money, credit and banking, 23(3), 581-612. 10.2307/1992692 bohn, h. (1995). the sustainability of budget deficits in a stochastic economy. journal of money, credit and banking, 27(1), 257-271. doi: 10.2307/2077862 bohn, h. (1998). the behaviour of us public debt and deficits. quarterly journal of economics, 113(3) 949-963. doi: 10.1162/003355398555793 brady, g.l., & magazzino, c. (2017). the sustainability of italian public debt and deficit. international advances in economic research, 23(1), 9-20. doi: 10.1007/s11294-016-9623-7 bravo, a.b.s., & silvestre, a.l., (2002). intertemporal sustainability of fiscal policies: some tests for european countries. european journal of political economy, 18(3), 517-528. buiatti, c., carmeci, g., & mauro, l. (2014). the origins of the public debt of italy: geographically dispersed interests?. journal of policy modeling, 36(1) 43-62. doi: 10.1016/j.jpolmod.2013.10.016 burret, h.t., feld, l.p., & köhler, e.a. (2013). sustainability of german fiscal policy and public debt: historical and time series evidence for the period 1850-2010. cesifo working paper, 4135. caporale, g.m. (1995). bubble finance and debt sustainability: a test of the government’s intertemporal budget constraint. applied economics, 27(12), 1135-1143. doi: 10.1080/00036849500000096 casadio, p., paradiso, a., & rao, b.b. (2012). the dynamics of italian public debt: alternative paths for fiscal consolidation. applied economics letters, 19(7), 635-639. doi: 10.1080/13504851.2011.591726 clemente, j., montañés, a., & reyes, m. (1998). testing for a unit root in variables with a double change in the mean. economics letters, 59(2), 175-182. doi: 10.1016/s0165-1765(98)000524 corsetti, g. (1991). testing for solvency of the public sector: an application to italy. economic notes, 20(3), 581-599. 142 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 corsetti, g., & roubini, n. (1991). deficits, public debt and government solvency: evidence from oecd countries. journal of japanese and international economies, 5(4), 354-380. doi: 10.1016/0889-1583(91)90004-a dalena, m., & magazzino, c. (2012). public expenditure and revenue in italy, 1862-1993. economic notes, 41(3), 145-172. doi: 10.1111/j.1468-0300.2012.00243.x de haan, j., & siermann, c.l.j. (1993). the intertemporal government budget constraint: a review and application for the netherlands. university of groningen, research memorandum, 556. dickey, d.a., & fuller, w.a. (1979). distribution of the estimators for autoregressive time series with a unit root. journal of the american statistical association, 74, 427-431. doi: 10.2307/2286348 di mascio, f., & natalini, a. (2014). austerity and public administration: italy between modernization and spending cuts. american behavioral scientist, 58(12), 1634-1656. doi: 10.1177/0002764214534667 elliott, g., rothenberg, t.j., & stock, j.h. (1996). efficient tests for an autoregressive unit root. econometrica, 64(4), 813-836. doi: 10.2307/2171846 enders, w. (2014). applied econometric time series. chichester: wiley. engle, r.f., & granger, c.w.j. (1987). co-integration and error correction: representation, estimation, and testing. econometrica, 55(2), 251-276. franses, p.h., van dijk, d., & opschoor, a. (2014). time series models for business and economic forecasting. cambridge: cambridge university press. forte, f. (2011). l’economia italiana dal risorgimento ad oggi, 1861-2011, siena: cantagalli. forte, f. & magazzino, c. (2014). economic growth and fiscal consolidation: the optimal mix between current and investment public expenditure. the case of europe. in forte, f., mudambi, r., & navarra, p. (eds.). handbook of alternative theories of public economics. cheltenham: edward elgar. forte, f., & magazzino, c. (2016). government size and economic growth in italy: a time-series analysis. european scientific journal, 12(7), 149-169. doi: 10.19044/esj.2016.v12n7p149 forte, f., & magazzino, c. (2011). optimal size government and economic growth in eu countries. economia politica-journal of analytical and institutional economics, 28(3), 295321. gregory, a.w., & hansen, b.e. (1996). residual-based tests for cointegration in models with regime shifts. journal of econometrics, 70(1), 99-126. doi: 10.1016/0304-4076(69)41685-7 greiner, a., & kauermann, g. (2008). debt policy in euro area countries: evidence for germany and italy using penalized spline smoothing. economic modelling, 25(6), 1144-1154. doi: 10.1016/j.econmod.2008.02.006 köller, u., & semmler, w. (2007). debt sustainability in the european monetary union: theory and empirical evidence for selected countries. oxford economic papers, 59(2), 194-218. doi: 10.1093/oep/gpl035 hakkio, c.s., & rush, m. (1991). is the budget deficit “too large?”. economic inquiry, 29(3), 143 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 429-445. doi: 10.1111/j.1465-7295.1991.tb00837.x hamilton, j.d., & flavin, m.a. (1986). on the limitations of government borrowing: a framework for empirical testing. american economic review, 76(4), 808-819. johansen, s., & juselius, k. (1990). maximum likelihood estimation and inference on cointegration with applications to the demand for money. oxford bulletin of economics and statistics, 52(2), 169-210. doi: 10.1111/j.1468-0084.1990.mp52002003.x kremers, j.j.m. (1988). limits on the us federal debt. economics letters, 28(3) 259-262. doi: 10.1016/0165-1765(88)90127-9 kremers, j.j.m. (1989). u.s. federal indebtedness and the conduct of fiscal policy. journal of monetary economics, 23(2), 219-238. doi: 10.1016/0304-3932(89)90049-4 kwiatkowski, d., phillips, p.c.b., schmidt, p., & shin, y. (1992). testing the null hypothesis of stationarity against the alternative of a unit root: how sure are we that economic time series have a unit root?. journal of econometrics, 54(1-3), 159-178. doi: 10.1016/03044076(92)90104-y lilico, a., holmes, e., & sameen, h. (2009). controlling spending and government deficits: lessons from history and international experience. london: policy exchange. lütkepohl, h. (2005). new introduction to multiple time series analysis. berlin: springer. lütkepohl, h., & krätzig, m. (2004). applied time series econometrics. cambridge: cambridge university press. macdonald, r., & speight, a.e.h. (1986). the intertemporal government budget constraint in the uk, 1961-1986. the manchester school, 58(4), 329-347. doi: 10.1111/j.14679957.1990.tb00426.x macdonald, r. (1992) some tests of the government's lntertemporal budget constraint using us data. applied economics, 24(12), 1287-1292. doi: 10.1080/00036849200000087) magazzino, c. (2014). government size and economic growth in italy: an empirical analyses based on new data (1861-2008). international journal of empirical finance, 3(2), 38-54. magazzino, c. (2012a). politiche di bilancio e crescita economica. torino: giappichelli. magazzino, c. (2012b). wagner versus keynes: public spending and national income in italy at a disaggregated level. journal of policy modeling, 34(6), november-december, 890-905. doi: 10.1016/j.jpolmod.2012.05.012 magazzino, c., intraligi v. (2015). la dinamica del debito pubblico in italia: un’analisi empirica (1958-2013). rivista italiana di economia demografia e statistica, 59(3), 167-178. marattin, l., & marzo, m. (2009). a note on the (un)pleasant arithmetic of fiscal policy: the case of italian public debt. economic notes, 38(3), 169-183. oecd. (2011). fiscal consolidation: targets, plans and measures. oecd journal on budgeting, 11/2. doi: 10.1787/16812336 paesani, p., strauch, r., & kremer, r. (2006). public debt and long-term interest rates: the case of germany, italy and the usa. ecb working paper series, 656, july. papadopoulos, a.p., & sidiropoulos, g.p. (1999). the sustainability of fiscal policies in the european union. international advances in economic research, 5(3), 289-307. 144 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 perron, p., & vogelsang, t. (1992). nonstationarity and level shifts with an application to purchasing power parity. journal of business and economic statistics, 10(3), 301-320. doi: 10.1080/07350015.1992.10509907 pesaran, h.m. (1997). the role of economic theory in modelling the long run. the economic journal, 107(440), 178-191. doi: 10.1111/1468-0297.00151 pesaran, h.m., & shin, y. (1999). autoregressive distributed lag modelling approach to cointegration analysis. in storm s., (ed.). econometrics and economic theory in the 20th century: the ragnar frisch centennial symposium. cambridge: cambridge university press. pesaran, h.m., & shin, y., smith, r.j. (2001). bounds testing approaches to the analysis of level relationships. journal of applied econometrics, 16(3), 289-326. doi: 10.1002/jae.616 piergallini, a., & postigliola, m. (2012). fiscal policy and public debt dynamics in italy, 18612009. rivista italiana degli economisti, 17(3), 417-440. piergallini, a., & postigliola, m. (2013). non-linear budgetary policies: evidence from 150 years of italian public finance. economics letters, 121(3), 495-498. doi: 10.1016/j.econlet.2013.09.030 phillips, p.c.b., & perron, p. (1988). testing for a unit root in time series regression. biometrika, 75(2), 335-346. doi: 10.2307/2336182 schwert, g.w. (1989). tests for unit roots: a monte carlo investigation. journal of business and economic statistics, 7(2), 147-159. doi: 10.2307/1391432 spaventa, l. (2013). the growth of public debt in italy: past experience, perspectives and policy problems. psl quarterly review, 66(266), 291-324. spaventa, l. (1987). the growth of public debt: sustainability, fiscal rules, and monetary rules. imf staff papers, 34(2), 374-399. trachanas, e., & katrakilidis, c. (2013). fiscal deficits under financial pressure and insolvency: evidence for italy, greece and spain. journal of policy modeling, 35(5), 730-749. doi: 10.1016/j.jpolmod.2013.03.013 trehan, b., & walsh, c.e. (1988). common trends, the government budget constraint, and revenue smoothing. journal of economic dynamics and control, 12(2), 425-444. doi: 10.1016/0165-1889(88)90048-6 uctum, m., & wickens, m. (2000). debt and deficit ceilings, and sustainability of fiscal policies: an intertemporal analysis. oxford bulletin of economics and statistics, 62(2), 197-222. doi: 10.1111/1468-0084.00168 vanhorebeek, f., & van rompuy, p. (1995). solvency and sustainability of fiscal policies in the eu. de economist, 143(4), 457-473. doi: 10.1007/bf01384910 von hagen, j., hughes hallett, a., strauch ,r. (2001). budgetary consolidation in emu, centre for economic policy research – economic papers. 148, march. zivot, e., & andrews, d. (1992). further evidence on the great crash, the oil price shock, and the unit-root hypothesis. journal of business and economic statistics, 10(3), 251-270. doi: 10.1198/073500102753410372 145 g. brady and c. magazzino / european journal of government and economics 6(2), 126-145 appendix methodological details the econometric literature has proposed several unit root tests with structural breaks. yet, one obvious weakness of the zivot and andrews (1992) strategy, relating as well to similar tests proposed by perron and vogelsang (1992), is the inability to deal with more than one break in a time series. addressing this problem, clemente et al. (1998) proposed tests that would allow for two events within the observed history of a time series, either additive outliers (the ao model, which captures a sudden change in a series) or innovational outliers (the io model, allowing for a gradual shift in the mean of the series). this taxonomy of structural breaks follows from perron and vogelsang’s work (1992). however, in that paper the authors only dealt with series including a single ao or io event (enders, 2014; franses, 2014; becketti, 2013; lütkepohl, 2005; lütkepohl and krätzig, 2004; baum, 2001). the ao is the type of outliers that affects a single observation. after this disturbance, the series returns to its normal path as if nothing has happened. the io is the type of outliers that affects the subsequent observations starting from its position or an initial shock that propagates in the subsequent observations. an ao affects only the t observation, whereas an io affects all observations beyond time t through the memory of the system. the double-break additive outlier model involves the estimation of the equation yt = μ + δ1du1t + δ2du2t + ÿt [a.1] where dumt = 1 for t > tbm and 0 otherwise, for m = 1, 2. tb1 and tb2 are the breakpoints, to be located by grid search. the residuals from this regression, ÿt, are then the dependent variable in the equation to be estimated. the equivalent model for the innovational outlier leads to the formulation yt = μ + δ1du1t + δ2du2t + ɸ1dtb1,t + ɸ2dtb2,t + αy t-1 + ∑ 𝑘𝑘 𝑖𝑖=1 θ iδyt-i + et [a.2] where again an estimate of α significantly less than unity will provide evidence against the i(1) null hypothesis. abstract. the aim of this paper is to empirically investigate the sustainability of italian national accounts in the years 1862-2013. the focus of the paper concerns applied tests related to the solvency and sustainability of fiscal policies. in fact,... keywords. fiscal policy; sustainability; time series; cointegration; italy jel classification. c22; h11; h60; o52 2. theoretical framework and empirical literature review references methodological details microsoft word ejge_02_02_043.doc european journal of government and economics volume 2, number 2 (december 2013) issn: 2254-7088 119 the common agricultural policy and the eu budget: stasis or change? alan greer, university of the west of england, united kingdom abstract this article evaluates the reform of the eu’s common agricultural policy, set within the context of negotiations about the multi-annual framework for 2014-20. it traces the reform process from the proposals presented by the commission in 2010 – encapsulated as ‘convergence, capping and greening’ – to the outcomes eventually agreed in 2013 after inter-institutional bargaining. the conclusion is that the reform outcomes highlight that the cap remains resistant to substantial change, both in terms of its budget and in its main policy instruments. this incremental change is explained in institutional terms. first, the balance of institutional forces within the eu, especially the member states and the continuing power of the agricultural interest, still works to insulate the cap against pressures for radical reform. there is a continuing cleavage between advocates of substantial change and a much larger bloc of member states who favour the retention of the ‘traditional’ cap. second, the new institutional setting of co-decision, which increased the role and influence of the ep, reinforced the dominant ‘state-assisted’ conception of agricultural policy. jel classification q18; h230 keywords agricultural policy, cap reform, multi-annual financial framework, co-decision european journal of government and economics 2(2) 120 introduction agriculture has always provided a fertile field for students of politics and public policy, especially those of us fascinated by how the political power of farmers in europe has persisted despite the long gradual decline in their number and the relative economic importance of the sector. this is reflected in the fact that the common agricultural policy (cap) not only has received the lion’s share of eu level spending over the years, but that it continues to do so. the durability of the cap and farm spending in the face of pressures for its substantial reduction, and the fact that it continues to be important for many member states, gives a highly political edge to debates about its reform. it is partly for these reasons – the size of the farm budget and the relative resistance to change of the cap that the agriculture sector is of interest for scholars of the eu as well as those seeking to explain the durability of public policies generally. much of the academic literature on the cap and comparative agricultural policy generally has tended to emphasise its ‘exceptional’ nature and its relative resistance to change, underpinned by an institutional balance of forces that favours the agricultural interest, both in respect of national governments and societal groups representing producers (see for example coleman et al 1997, grant 1997, rieger 2000, roederer-rynning 2010, skogstad 1998). of course this does not mean that change is impossible, and over the long term, incremental change can have substantial cumulative effects. rather what is at issue is the speed, extent and direction of change. burrell (2009) charts a pattern of continuous reform of the cap, from a highly centralized protectionist policy focused on farm incomes to one that is more market sensitive and environmentally sustainable, and in which governments have greater scope to adapt the overall framework to their particular national conditions (see also greer 2005). however it would be a mistake to see policy change as a historically driven one-way process. daugbjerg (2009), for example, uses the notion of ‘sequencing’ to show how changes to the cap over time are often reactive but not necessarily in the same direction and that each reform event opens possibilities for further reform. cap reform episodes since the late 1980s – such as the macsharry reforms in 1992, the agenda 2000 reforms in 1999 and the fischler package in 2003 have provided insightful case studies of the speed and scale of change, the endogenous institutional forces at play, and importance of exogenous factors such as enlargement and world trade liberalisation (see ackrill 2000a, cunha 2011, keeler 1996, swinbank 1999, and swinnen 2009). typically cap reform outcomes are viewed as incremental change, the outcome of bargaining between actors in which the key factors are the preferences and institutional structures of member states, with the commission also playing a crucial role in setting the reform agenda. reflecting the wider contemporary interest in the role of ideas in the policy process, some authors emphasise the link between the changing ideational context and the shape of reform outcomes (lynggaard and nedergaard (2009). also drawing on the importance of policy ideas, daugbjerg and swinbank (2009) point to the importance of trade agreements, notably the uruguay round agreement on agriculture (uraa), in the development of the cap, especially in establishing the market liberal paradigm as the ideational underpinning of agricultural trade. yet they still caution that this sits awkwardly with the continued dominance of the stateassisted paradigm as the foundation of the cap. erjavec et al (2008) also have noted how the policy discourse of agriculture commissioners (fischler and fischer-boel) exhibited a rhetorical shift from a largely mercantilist to a neo-liberal perspective. subsequently, however, some have argued that the tenure of commissioner cioloş has witnessed something of a reversal back to the stateassisted position that defended farm subsidies as a way to ensure a fair and efficient european agriculture (rutz et al, forthcoming). greer ● the common agricultural policy and the eu budget 121 budgetary considerations have played a key role in agricultural policy change, sometimes bringing the issue onto the policy agenda and opening a policy window for cap reform, for example the crisis of over-production in the 1980s. ackrill (2000b) has argued that eu budgetary rules and expenditure limits can work to constrain cap reform. in addition, he uses ideas about historical institutionalism and path-dependence to explore the linkages between institutional reform of the budget and change in the cap, especially in the period 1988-92. while these concepts are ideally suited to explain the resilience of agricultural policy, they need to be joined with the notion of ‘institutional layering’ to explain the how largely pathdependent processes can experience reform (ackrill and kay 2006a). this relationship can be seen clearly in the negotiations on the multi-annual financial framework (mff) for 2014-20, and a parallel round of cap reform, both of which concluded in 2013. in the mff case, debates about the overall level of the eu budget, and of the place of agriculture within it, were centre-stage, framed by the context of economic austerity. on cap reform the core questions concerned how best to reshape the policy to meet contemporary concerns, for example about fair treatment for all member states and the deepening of environmental sustainability, related crucially to the overall budgetary envelope. a crucial factor that influenced both process and outcomes was the changed institutional setting. as well as the contemporaneous discussions on the budget, the cap reform debates took place within the new institutional arrangements introduced in the lisbon treaty, which extended the co-decision procedure to the cap and arguably has enhanced the influence of the european parliament (ep). indeed the 2013 cap reform dossiers were the first major test of these new institutional arrangements. after first summarising the budgetary context and the historical trends on the funding of the cap, this article sets out the outcomes of the mff for 2014-20, highlighting changes in the proposed allocations during the negotiations. it then traces the cap reform process between 2010 and 2013 in light of the initial proposals presented by the commission. it concludes that the outcomes of the reform process were less radical than originally envisaged. this relative pathdependence and incremental change is explained in institutional terms. first, the balance of institutional forces within the eu, most especially the policy preferences of the main policy actors – especially the member states and the continuing power of the agricultural interest still works to insulate the cap against pressures for radical reform. there is a continuing cleavage between member states, which maps onto a broader budgetary gainers/losers division, between advocates of radical reform and those who favour the retention of the ‘traditional’ cap. second, the new institutional setting of co-decision, which increased the role and influence of the ep, reinforced the ‘traditional’ conception of the purpose and shape of agricultural policy. the central conclusion is that the outcomes of the reform processes highlight that the cap still remains resistant to substantial change, both in terms of the level and distribution of its budget, and in its main policy mechanisms and instruments, despite demands from some quarters for a radical reconfiguration of policy to take a more focused ‘public goods’ approach. cap spending in historical perspective agriculture has always consumed a big slice of the eu budget, reflecting that the cap historically has been the most important ‘common’ policy. in the late 1970s around 75 per cent of total spending went to agriculture but since then the trajectory has been downwards. by the first year of the 1988-92 financial period the cap still took up 60 per cent of spending, despite reforms to the budget that made it ‘redistributive and less cap-oriented’ (laffan and lindner 2010, p. 217). by the late 1990s farm expenditure fell to around half, reflecting successive reforms to the cap and the expansion of the eu into areas such as cohesion, regional and social policy. by the late 2000s the cap budget (now including rural development) still european journal of government and economics 2(2) 122 represented around 40 per cent of the total but is projected to continue on a gradual downwards trajectory (for a fuller discussion see greer 2012). annual budgets for agriculture and rural development are set within the overall framework of the eu’s multi annual financial frameworks. in the mff for 2007-13, around 43 per cent of a total €976bn went to the ‘preservation and management of natural resources’, mainly the cap. although total spending did not fall by much over the mff, the agriculture budget continued its relative decline because of the expansion of competences and increased expenditure in other areas. by the final year of the mff in 2013 the proportion of the eu budget devoted to the cap had fallen to around 40 per cent (€60bn) (see figure 1). figure 1: eu budget 2013 source: http://ec.europa.eu/budget/figures/2013/2013_en.cfm the distribution of these substantial resources for agriculture and rural development has always been a matter of contention, especially around spending on environmental sustainability. in the 2007-13 mff one third of the total budget (around 80 per cent of cap spending) went to ‘traditional’ direct payments and market intervention in agriculture (pillar 1) while the remainder was for measures on the environment and rural development (pillar 2). to use the commission’s somewhat misleading language, by 2013 ‘agricultural’ spending had declined to around 30 per cent of the total budget, although this is something of an ‘accountancy trick’ because almost all of pillar 2 expenditure goes to farmers and landowners under the auspices of the cap. so despite a falling share, spending on the cap remains substantial. as the commission acknowledged, ‘agriculture is, in budgetary terms, the second most important union policy, and the only fully integrated policy at union level’ (eu commission 2011b). the multi-annual financial framework 2014-2020 discussions about cap reform took place in tandem with negotiations about the mff for 2014-20, both framed by financial crisis and the ‘austerity agenda’. the draft mff unveiled by the commission in june 2011 envisaged spending of €1025 billion, an increase of nearly five per cent. following difficult negotiations, in february 2013 the european council agreed an mff that was widely referred to as the first ever budget ‘cut’ in the history of the eu (although previous mff saw ‘cuts’ in terms of percentage gross national income (gni)(see benedetto and milio 2012, p. 180). inter-institutional negotiations between the ep, council and commission produced a compromise political agreement in june 2013. table 2 illustrates the proposals and outcomes of the negotiations at various stages prior to the final inter-institutional agreement. an overall ceiling of €960bn (commitment greer ● the common agricultural policy and the eu budget 123 appropriations) was agreed, a reduction of around three per cent in 2011 prices over previous period and of eight per cent on the commission’s proposals. explanations of the outcomes of budget negotiations usually emphasise the relative influence of core actors and institutions at both the supranational and national levels. for laffan and lindner (2010) the european council ‘is the dominant player’, with the commission trying to play the role of ‘honest broker’ (2010, pp. 213-4). as swidlicki et al. also note the eu budget has proved difficult to reform ‘because it is not primarily designed to maximise economic returns across europe, but rather to balance a range of political interests. member states are keen to protect areas of the budget from which they do well...’ (2012, p. 6). a usual approach is to classify member states in terms of whether they are ‘net contributors’ (countries who pay out more than they get back such as the uk, sweden, finland and germany) or ‘net beneficiaries’ (such as greece and portugal) in terms of the eu and cap budget (with sometimes a third category of countries who pay in more or less as much as they receive (including belgium, denmark, france and italy)(see de wilde 2012, p. 1081; swidlicki et al 2012, pp. 79). on the 2014-2020 mff, the net contributors resisted a budget increase, highlighted in a letter to commission president barroso in december 2010 in which david cameron, angela merkel and other leaders argued that eu spending ‘cannot be exempt from the considerable efforts made by the member states to bring their public spending under control.’ specifically the next mff ‘should not exceed the 2013 level with a growth rate below the rate of inflation’ (cameron et al. 2010). table 2. mff 2014-2020: comparisons (€m) commitment appropriations in million euros (2011 prices) mff 2007 2013 mff 20142020 commission proposal june 2012 mff 20142020 european council conclusions 08.02.2013 european council conclusions vs mff 2007-2013 european council conclusions vs commission proposals million euros (%) million euros (%) total commitment appropriations 994.176 1.045.282 959.988 -34.188 (-3%) -85.294 (-8%) as a percentage of gni 1,12% 1,09% 1,00% -0,12% -0,09% 1. sustainable growth 446.310 503.310 450.763 +4.453 (+1%) -52.547 (-10%) 1a. competitiveness for growth and jobs 91.495 164.316 125.614 +34.119 (+37%) -38.702 (-24%) of which: connecting europe facility 12.783 40.249 19.299 +6.516 (+51%) -20.950 (-52%) of which: galileo, iter and gmes 8.047 15.548 12.793 +4.746 (+59%) -2.755 (-18%) 1b. cohesion for growth and employment 354.815 338.994 325.149 -29.666 (-8%) -13.845 (-4%) of which: investment for growth and jobs 345.935 327.116 313.197 -32.738 (-9%) -13.919 (-4%) of which: european territorial cooperation 8.880 11.878 8.948 68 (+1%) -2.930 (-25%) 2. preservation and management of natural resources 420.682 389.972 373.179 -47.503 (-11%) -16.793 (-4%) of which: market related expenditure and direct payments 318.820 286.551 277.851 -40.969 (-13%) -8.700 (-3%) of which: rural development 95.741 91.966 84.936 -10.805 (-11%) -7.030 (-8%) 3. citizenship, freedom, security and justice 12.366 18.809 15.686 +3.320 (+27%) -3.123 (-17%) 4. eu as a global player 56.815 70.000 58.704 +1.889 (+3%) -11.296 (-16%) 5. administration 57.082 63.165 61.629 +4.547 (+8%) -1.536 (-2%) 6. compensations 920 27 27 0 (0%) 0 (0%) source: http://ec.europa.eu/budget/mff/index_en.cfm european journal of government and economics 2(2) 124 negotiations between the member states on the mff reflected core national interests and revolved around established issues such as the uk ‘rebate’ and the level of funding for the cap. here the relationship between the uk and france is again instructive. with the government under pressure from a sizeable number of euro-sceptic conservative mps, the uk wanted at least a freeze, if not a substantial cut, in the total mff but also to defend its ‘rebate’ (see house of commons library 2013, p. 7). although outgoing president sarkozy (but not president françois hollande) at times supported a freeze in spending, he disagreed with the uk on the extent of ‘cuts’ and in the areas they should fall. while there were suggestions that cameron and sarkozy had agreed informally that the uk would not push too hard for cap reform if france did not demand cuts in the rebate (the guardian, 28 may 2012), hollande questioned the justification for the british rebate and brought italy and spain into a troika of ‘refuseniks’ in an effort to fend off a substantial reduction in the mff and to protect agricultural spending. overall the consensus of opinion favoured budgetary constraint, although there was disagreement about the scale of ‘cuts’. the pivotal role was played by german chancellor angela merkel who came down in favour of a squeeze on spending. for one commentator, the negotiations created ‘an impression of earnest endeavour that paved the way for a victory for the austerity agenda of david cameron and angela merkel’ (wyles 2013). certainly the british government presented the mff outcome as a major diplomatic triumph, both in terms of protecting the rebate and securing a cut in the budget – what cameron referred to as ‘europe's seven-year credit card limit’ (although it was also acknowledged that the uk’s net contributions will still rise because the rebate does not cover increased spending in new member states). cap reform and the mff 2014-2020 reaching agreement on the mff is only the beginning of the story. perhaps more important are the debates about how a smaller cake is to be divided up, whether between countries or sectors, the cap after 2013: the mff context debates about cap reform after 2013 took place within the context of the negotiations on the draft mff. this was something of a departure on previous experience because reforms of the cap have tended to take place ‘outside the framework of the big budget bargains’ (laffan and lindner 2010, p. 224). agriculture spending for 2007-13, for example, was fixed by the chirac-shroeder deal in 2002 and independently of the main mff negotiations on the mff (see greer 2012, p. 106). during the debate on the mff, some reformist voices advocated a radical shift of the eu’s budgetary priorities to fund increased spending on sustainable growth in areas such as employment, research/development and energy, rooted in a conviction that most cap spending does not promote the attainment of public goods (greer 2012, pp. 111-2). one proposal advocated a switch in funding from the cap to better targeted spending on issues such as energy and climate change (núñez ferrer, 2009). a complete and radical overhaul of the eu also was recommended in a report by the relatively euro-sceptic open europe, whose ‘alternative budget’ proposed a reduction of spending by almost a third while focusing the smaller budget that remained ‘far more effectively on boosting jobs and growth’ (swidlicki et al, 2012, p. 3). yet arguably the agreement on the mff changed little despite the rhetoric and reality of cuts. for one commentator the negotiations produced a budget agreement ‘that helps farmers but will do little to create growth and jobs’ (wyles 2013). certainly the distribution of ‘cuts’ in relation to the commission’s original proposals fell heavily on areas such as ‘citizenship’ (17 per cent), the ‘eu as a greer ● the common agricultural policy and the eu budget 125 global player’ (16 per cent) and crucially proposed expenditure on ‘sustainable growth’ was reduced by 10 per cent (€50m) on the proposals, with the ‘competitiveness for growth and jobs’ heading reduced by 24 per cent (€38m), and funds for the ‘connecting europe’ facility within this slashed by over half, although on this still represented increases on the previous mff. that a radical reconfiguration of the budget away from the cap was unlikely had been evident from the commission’s draft mff which explicitly acknowledged that ‘a significant part of the eu budget should continue to be dedicated to agriculture, which is a common policy of strategic importance’, and that the cap share of funding be maintained at the 2013 level (2011c, p. 7). nevertheless the actual resources available for the cap will be reduced. in the initial proposals €379bn was earmarked for the cap (37 per cent of the total), compared with €415bn in the previous mff. after the council agreement, this figure was adjusted marginally downwards to €363bn, representing a reduction of around 12 per cent on the previous mff, and an extra cut of 4-5 per cent on the commission proposals (see table 2). there will be real reductions in agricultural spending, which will go down in real terms and also as a percentage of the total budget (predicted to fall to around 37 per cent by 2020). while consistent with the pattern of continued gradual decline, however, this does not represent the fundamental restructuring eu spending and its core priorities that the radical reformers have argued for. within this overall context, a the mff agreement broadly maintains the 75:25 spending ratio between pillar 1 direct payments/market measures and pillar 2 rural development/environmental sustainability. the commission proposal earmarked 76 per cent to direct payments and 24 per cent to rural development; in the council conclusions these figures were roughly maintained although the reduction for rural development was eight per cent compared with four per cent in direct payments. expressed in current prices, proposed annual expenditure from 2014 is stable, although commitment appropriations on market support and direct payments will still make up 30 per cent of the eu budget in 2014 and 27 per cent in 2020 (see table 3). table 3. mff 2014-2020: commitment appropriations (€m) source: house of commons library 2013, p. 9. cap reform after 2013: the proposals between april 2010 (when it initiated preliminary discussions) and october 2011 (the publication of detailed legislative proposals), the eu commission’s directorate general for agriculture and rural development (dg agri) formulated proposals for european journal of government and economics 2(2) 126 a reformed cap after 2013. the aim was to take the opportunity provided by the new mff to ‘refocus the cap on its core and new activities’ (european commission, 2011a, p. 15). resources would be better targeted with a ‘greener and more equitably distributed first pillar’ with a second pillar ‘more focussed on competitiveness and innovation, climate change and the environment’ (european commission, 2011a, p. 16). the goal is to promote a ‘new partnership between europe and its farmers’ through creating a cap that will ‘strengthen the competitiveness, sustainability and permanence of agriculture throughout the eu in order to secure for european citizens a healthy and high-quality source of food, preserve the environment and develop rural areas’ (europa, 2011). based on the belief that there remained widespread support for a strong cap structured around its two pillars, dg agri set out three broad policy scenarios that struck different balances between them. an ‘adjustment scenario’ envisaged ‘further gradual changes’, while a ‘refocusing scenario’ would usher in far reaching reform, with a substantially reduced budget centred on environmental and climate change objectives. for the commission however, the first option would not adequately address the challenges of the future, whereas the second would ‘come at a significant social and environmental cost’ (european commission, 2011c, pp. 5-6; 2011d, p. 38; 2010a, p. 12). consequently the preferred approach was a midpoint ‘integration scenario’ that would ‘make major overhauls of the policy in order to ensure that it becomes more sustainable, and that the balance between different policy objectives, farmers and member states is better met’ (european commission, 2010a, p. 12 original emphasis). as fleshed out in policy specifics in the commission’s legislative proposals, this approach has three main elements that are intended to alter the distribution of spending within the cap: convergence, capping, and greening (see swinbank 2012). ‘convergence’ aimed to promote fairness by progressively narrowing variations in direct payments between farmers and across countries, particularly between old and ‘new’ europe, to ensure both ‘a more equal distribution of direct support’ and ‘fairer treatment of farmers performing the same activities’ (european commission, 2011a, p. 16). a unified ‘basic payment scheme’ would be phased in and replace the existing plethora of systems by 2019. national envelopes for direct payments will be adjusted so that countries receiving less than 90 per cent of the eu average payment will see one third of the gap closed. ‘capping’ proposed the progressive reduction of payments received by large beneficiaries and holdings, with resources transferred to rural development (sometimes referred to as degressivity). reductions would start at 20 per cent for entitlements between €150,000 and €200,000, gradually increasing in three stages to a 100 per cent reduction for payments over €300,000. ‘greening’ was to be substantially enhanced by making 30 per cent of direct support (pillar 1) conditional on verifiable and legally defined ‘environmentally supportive practices’ with farmers ‘receiving payments to deliver public goods to their fellow citizens’ (european commission, 2011a, p. 16). these practices were to include practices such as crop diversification, maintenance of permanent pasture, and the preservation of ecological reserves and landscapes. countries would be able to transfer up to five per cent of direct payments between pillars to reinforce rural development, although for those whose level of direct payments is below 90 per cent of the eu average, ‘reverse’ transfer to pillar 1 will be allowed. the cap after 2013: explaining the outcomes for laffan and lindner, the three financial perspectives since early 1990s did little more than facilitate incremental changes in the levels of spending on the cap (2010, p. 218). while farm spending will go down, analysis of the mff for 20142020 supports this picture, with agriculture still accounting for over one third of the total eu budget. two main ideas usually are offered to explain this: the relative greer ● the common agricultural policy and the eu budget 127 balance of influence between member states as reflected in the outcomes of bargaining, and the nature of inter-institutional governance in the eu, both of which are overlaid by an enduring structural power enjoyed by agricultural interests. because the cap still takes up over a third of the budget, there is an overlap between ‘net recipient’ countries and a bloc of ‘cap defenders’, which swidlicki et al estimate at 24 countries. although probably an over-exaggeration, around 20 countries defended the cap at an agriculture council meeting in march 2011, greeted by french agriculture minister, bruno le maire, as not just a ‘strong political signal of support by governments for the cap’ but for ‘strengthening it in the years to come’ (euractiv.com, 2 june 2010, 18 march 2011). most member states generally favour maintaining cap funding at around its current level, although they do differ about what the money should be spent on. for swidlicki et al., this is ‘a powerful bloc which continues to block reform under the influence of the farm lobby, the commission, and many meps’ (2012, p. 9). this draws attention to the importance of eu institutions such as the ep, which has a formal role on cap under the lisbon treaty and which has tended to favour the maintenance of agricultural spending at around its current level (see greer & hind 2012). the agriculture council – historically the key actor on the cap typically has been regarded as defending the interests of farmers. as swidlicki et al. pithily summarise, ‘through the commission’s agenda-setting powers and the parliament’s veto powers, the eu institutions are also an obstacle to reforming the budget’ (2012, p. 9). this, however, understates the extent to which the commission actually has at times been an influential actor in promoting cap reform. national interests and preferences negotiating outcomes on the cap and mff reflect compromise and bargaining between different national interests but fundamental disagreements between member states remain extremely important. as agra europe noted in a report on an informal meeting of agriculture ministers in spain in june 2010, ‘calls for an end to pillar one largesse were echoed by more liberal-leaning member states…while traditional defenders of the cap budget made clear their continued support for a strong cap despite the new budgetary austerity sweeping across europe’ (no. 2415, 4 june 2010). traditional cap defenders such as france, ireland and greece support maintenance of the farm budget, favour direct payments tied to past levels of production, and remain sceptical about ‘greening’. in early 2011, for example, president sarkozy made it clear that france aimed to ‘maintain the cap's budget to the last euro’ and stated that ‘we do not have to excuse ourselves for defending community preference and the cap budget’ (euractiv.com, 19 january 2011; le monde.fr, 12 may 2011). a crucial element in the political process on cap reform, as it was in 2002, was the franco-german axis. in a joint paper in september 2010 they set out agreed core principles for ‘a strong cap beyond 2013’, rejected any ‘renationalization’, supported retention of the two pillar structure and of direct payments, and opposed substantial redistributive measures towards convergence’ (bundesministerium et al, 2010, p. 2). this was reinforced by a joint communique issued after a meeting in berlin in october 2012 by agriculture ministers, ilse aigner and stéphane le foll, which rejected calls from britain, poland and others for substantial cuts in pillar 1 expenditure. both ministers also stressed the need to maintain ‘close and trustful relations during the final stages of negotiations over cap reform and the mff’ (euractiv.com, 11 october 2012). france also joined with spain in defence of farm spending. in february 2012 their agriculture ministers issued a statement that stressed the importance of maintaining a strong agricultural sector and made clear they would ‘not accept’ any budget that ‘does not guarantee the stabilisation of the cap’ or maintain the existing level of spending (euractiv.com, 15 february 2012). european journal of government and economics 2(2) 128 reflecting their general stance on the mff, pro-reform countries such as the uk, sweden, denmark and the netherlands outlined radical reform ideas that were predicated to a degree on budgetary restraint. recognising that substantial cuts in the short term were unlikely, they pressed for a substantial reorientation of support from pillar i (direct payments) to funding for the provision of public goods under pillar 2 in relation to rural development, the environment and climate change. in september 2011 for example, the agriculture ministers of the uk and poland called for ‘deep and genuine’ reform of the cap, centred on a more competitive agriculture sector and better incentives to improve the environment, with reduced emphasis on pillar 1, convergence of direct payment rates, and a ‘step change’ in measures to increase competitiveness, including support under pillar 2. this was described in one report as london and warsaw putting themselves ‘firmly on a collision course with france’ over cap reform (euractiv.com, 22 october 2011). preferences & outcomes the extent of the influence of the ‘reformers’ can be assessed by mapping the extent to which they were successful in translating policy preferences into outcomes, using the uk as an exemplar. the uk historically has been in the vanguard of those favouring radical reform of the cap. while governments of all party complexions have not advocated the total abolition of the cap, which they see as vital to free trade within an undistorted single market, they do want a much less expensive, more consumer friendly, and more market oriented agricultural policy that is better focused on core objectives such as competitiveness and the environment. this means that reform must be radical and extensive, phasing out direct income support payments and shifting emphasis to pillar 2 funded public goods in the overall context of a smaller agricultural budget that has greater cofinancing. whether this would resemble anything like the cap as we have come to know it is open to doubt. as the department for the environment, food and rural affairs (defra) noted in its response to the commission’s reform communication in january 2011, ‘successive reforms of the cap have set a direction of travel towards greater market orientation and agricultural competitiveness and a greater focus on the delivery of public benefits in return for cap expenditure. the uk wants to see the acceleration of this process...’ (defra 2011b: 1). for the uk the commission’s proposals did not go far enough in two respects. first they insufficiently recognised the changed economic circumstances in the wake of the financial crisis. with intense pressure on public finances, the cap ‘cannot be immune to the hard choices being made elsewhere in the eu’ and consequently there must be ‘a very substantial cut to the cap budget’ for the next mff, concentrating on pillar 1 and leaving pillar 2 with ‘a greater share of a smaller cap budget’ (house of commons, 2011a, p. 5). second, by not taking the radical re-focusing’ route the commission proposals were ‘insufficiently ambitious for engendering the change in eu agriculture sector that will be needed if it is to survive in an increasingly competitive international agricultural market’ (house of commons, 2011a, p. 4). specifically the uk was concerned that proposals such as capping payments and greening pillar 1 ‘would be counterproductive to eu aims to develop a competitive agriculture sector, and risk entrenching continued reliance on subsidies instead of independence.’ rather there is ‘scope for pillar 2 to better reward farmers for the important public goods they provide’ and that we ‘need to stimulate a change in behaviour rather than entrench continued reliance on subsidy which it is increasingly difficult for the european taxpayer to afford’ (defra, 2011b: para. 7). the uk therefore argued strongly for phasing out of the remaining coupled subsidies, an end to export subsidies and market support measures such as intervention, and a radical reorientation of spending from pillar 1 to pillar 2, involving the gradual abolition of direct payments to reduce spending substantially. greer ● the common agricultural policy and the eu budget 129 so how does the outcome on cap reform map onto such policy preferences? as one report put it, the cap deal was finally reached on 26 june 2013 ‘after months of haggling over how ambitious the policy would be on overhauling direct payments, ending quotas, and making farmers more environmentally accountable’ (euractiv.com, 11 july 2013). key issues were the reform of direct payments and the pillar 2 budget, the abolition of market support measures, capping, convergence and the plans for ‘greening’ pillar 1. first, it is clear that advocates of radical reform did not succeed in obtaining the substantial reduction in the budget that they desired. agriculture council ministers broadly backed the commission’s reform ideas, noting for example in march 2011 that the cap had to remain a strong common policy with financial resources commensurate with its objectives (euractiv.com, 18 march 2011). the share of agriculture spending in the new mff will still be around a third and will decrease only slowly. so a common reaction has been that ‘agriculture was spared from major cuts made elsewhere in the eu’s spending framework’ (euractiv.com, june 2013). second, the 75:25 spending ratio between the pillars remains and will not change much, if all, by 2020. if anything, the council conclusions on the mff have resulted in a relatively greater reduction of planned spending on pillar 2 than in pillar 1. this indicates that when the budgetary circumstances are tight, governments and stakeholders resort to protecting their core interests (‘there are no votes in pillar 2’). arguably ‘net contributors’ such as the uk prefer overall budgetary constraint to a reconfigured cap, which might actually cost more in the short term. so while cameron played up his ‘success’ on the mff, this came at the expense of failure to move the cap substantially in the british government’s favoured direction. indeed on some calculations the mff outcome has reduced the uk’s resources under pillar 2 by more than originally planned. third, there was a mismatch on direct payments between the uk’s long-term vision for the cap and its specific negotiating priorities. while the government argued that income support subsidies are neither desirable nor necessary, it ‘made it clear that phasing out such payments by 2020 is unrealistic, in both practical and negotiating terms’ (house of commons, 2011a, p. 6). instead the aim was for ‘a very substantial, transitional reduction followed by ultimate abolition at a later date’ (defra, 2011b, para. 12). this chimed with the environment, food and rural affairs committee’s recommendation for ‘a more pragmatic approach’ because ending direct payments ‘is unachievable in the short-term’ and sticking rigidly to this policy ‘reduces the uk’s ability to engage constructively with other member states and could diminish the uk’s influence in this round of reform’ (2011b, p. 41). essentially this recognises that while the uk does not like direct payments, most other countries do. in fact there are important voices in the uk who challenge the policy approach, which in many respects reflects a treasury driven-agenda that is ideologically opposed to ‘subsidies’. the devolved adminstrations in wales and northern ireland defend them, and stakeholders such as the national farmers’ union (nfu) are sceptical that they can be substantially reduced without having a substantial detrimental impact on the viability of many british farms. fourth, the uk criticised the plans for ‘greening’ a substantial share of pillar 1 funding as too modest but also fundamentally flawed, and ‘unlikely to deliver significantly greater or more ambitious environmental benefits than those that are currently delivered’ (defra, 2011b, para. 27). the ‘greening’ proposals were the subject of much haggling during the negotiations on cap reform, what one source has called the subject of ‘a high-pressure lobbying campaign’ from both farmers and environmentalists (euractiv.com, july 2013). other national governments expressed concerns that the policies would be burdensome. while the french and german agriculture ministers said that they supported ‘greening’, they nonetheless wanted a ‘flexible’ system so that ‘agriculturally sustainable use of areas of ecological interest may be possible’ (euractiv.com, 11 october 2012). the european journal of government and economics 2(2) 130 statement issued by france and spain was even blunter, criticising the draft rules for ‘greening’ as ‘not adapted to the challenges facing european agriculture’ (euractiv.com, 15 february 2012). in the negotiations, both the council and meps supported plans to introduce ‘greater flexibility’ into the farm subsidies system. this reflected some core concerns of farming stakeholder groups about ‘being locked into standardised environmental rules despite the diverse landscape of european agriculture’ (euractiv.com july 2013). for others, including environmental groups, this meant watering down the proposed requirements for ‘greening’. one commentator noted in the wake of the council conclusions on the mff that commission ‘is fighting a rearguard action to force farmers to work in a way that benefits the environment’, and that the scheme ‘was weakened as part of last week's grand bargain between eu leaders at their budget summit in brussels’ (harrabin 2013). a spokesperson for birdlife europe complained that ‘europe is offered a budget that scales back investment in the environment and caters for the usual fat cats that have been milking the system until now’; for the royal society for the protection of birds (rspb) ‘wildlife across europe will pay a heavy price for this terribly regressive deal’ (as quoted in harrabin 2013). under the cap deal, 30 per cent of direct payments will still be contingent on meeting certain environmental criteria, however there will be no eu-wide performance standards and member states will have greater flexibility in implementation, for example about when to apply sanctions. in addition the political agreement introduced a range of exceptions relating to water pollution, crop diversification and environmental sustainability. farms under 15 hectares do not have to comply with new requirements to create “ecological focus areas,” which in the first instance will apply to five per cent of farmland in 2015, rather than the proposed seven per cent (this may still increase pending a review in 2017). for critics, more than one-third of all farmland and 89 per cent of farmers will be exempted from the rules. farms of less than 10 hectares (one third of the total in the eu) are exempted from crop diversification rules aimed at improving soil quality, and farmers also remain outside some eu environmental and water pollution legislation. according to a spokesperson for birdlife europe this represented ‘a major blow to those who championed a more sustainable, forwardthinking policy – one which would deliver for people and the environment as well as protecting the long-term interests of farming’ (euractiv.com, july 2013). fifth, capping direct payments, inspired by a desire to better support small farmers, also was an area of dispute. similar proposals were fiercely resisted in previous reform episodes by countries such as the uk and france who have efficient and large scale enterprises. for defra capping discourages greater competitiveness by restricting ‘natural structural processes’ such as amalgamation and consolidation (2011b, para. 23), and proposals for supporting small farms should concentrate on improving their underlying competitiveness. in the event the reform deal did stipulate that payments to those receiving more than €150,000 per year would be progressively reduced, but left unresolved the level of a limit on how much any farmer could receive. the new rules stipulate that each farmer receive at least 60 per cent of the average national or regional direct payment by 2019, a reshuffling that advocates say will help smaller landholders. measures also were introduced to help ‘new entrant’ farmers. sixth, some caveats were introduced around convergence and the forward march of ‘liberalisation’ was stalled in some respects, illustrating again the influence of the protectors of a traditional cap. german, french and spanish ministers had strong reservations about convergence, not least because any redistribution in favour of ‘new’ member states would entail reduction in the benefits received by their own farmers (a commission estimate put this as high as a 7 per cent reduction in funding for french farmers). france and spain wanted a more careful phasing in of convergence and complained that the ‘magnitude’ and ‘pace’ proposed by the greer ● the common agricultural policy and the eu budget 131 commission was ‘not acceptable’. nevertheless the cap deal does provide for the gradual harmonisation of payments between old and new countries, requiring that no single state receives less than 75 per cent of the eu average by 2019. on the other hand the force of renewed arguments about food security in a world made increasingly volatile by climate change underpinned decisions to retain some production quotas. although the last remaining export subsidies were reduced to zero, facilitated by high world food prices, sugar quotas were extended until 2017 (two years later than their planned abolition) and a new vine planting scheme after 2016 was introduced, rather than withdrawal of protection for grape growers. as julie girling, agriculture spokeswoman for the british conservative party, complained, ‘old-fashioned market intervention is back in a big way, potentially taking us back to the bad old days of butter mountains and wine lakes’ (euractiv.com, 11 july 2013). constructing a balanced scorecard on all of these elements it seems clear that the result of the negotiations on cap reform was to markedly water down the original proposals of the commission in order to secure agreement between member states, each of whom had their own ‘red lines’ and negotiating objectives. in most respects this ‘watering down’ was at the expense of ideas such as ‘greening’ that are central to the proposals of the radical reformers, and consistent with objections raised by defenders of a ‘traditional’ cap. one way this has been done is to continue the trend to give greater flexibility for member states and hence more scope for national variation (see greer 2005). one analysis remarked that while the final agreement ‘retained some’ of the original goals, the inter-institutional negotiations produced an outcome that ‘gave the 28 member states more leeway, including over new environmental performance rules’ (euractiv.com, 11 july 2013). this actually is consistent with the uk’s argument that member states should continue to be given the flexibility to allocate cap funding in a way that best suits the requirements of their own regions and farming structures, providing it is consistent with the wider objectives of the cap. a one-size fits all solution will restrict rather than facilitate, as well as adding significant complexity and costs (defra, 2011b, para. 24). the cap after 2013: the influence of co-decision the outcome of the cap reform process can be explained in essentially traditional ways, mainly by focusing on the defence of important national interests by member states (acting on their own, bilaterally or in coalitions) usually under the continued influence of a domestic farm lobby. in addition the cap deal (as with the mff) embodies side payments and trade-offs between the main policy institutions and actors (for example increased budget rebates for the netherlands and austria, and extra rural development funding for several countries including ireland). one new influence compared to previous cap reform episodes was the part played by the ep, exercising the formal role given to it by the lisbon treaty. although some relatively minor issues had already been processed using the new rules, the 2013 cap reform was the first real test of these new inter-institutional arrangements (see greer & hind, 2012). the first point to make is that co-decision in agriculture, at least on the big policy issues, is likely to prolong the policy process. the inter-institutional deal reached in june 2013 came nearly two years (21 months) after the commission published its legislative proposals. as one report put it, there was a ‘slow decision-making process, highlighted by intense lobbying’, which means that much of the agreement will not be implemented until 2015, requiring the commission to introduce transitional arrangements to cover 2014 (euractiv.com, 11 july 2013). second, as suggested in some early analyses of co-decision in agriculture, while the ep now has a greater formal role in the decision-making process, this – so far at least – has buttressed the status quo on the cap rather than facilitate radical european journal of government and economics 2(2) 132 change (greer & hind 2012). the ep’s agricultural committee, comagri, dominated by farming, landowning and rural interests, defended the cap in the negotiations and played a key part in altering the original proposals in some respects. the debates over ‘greening’ showed up the structural divisions within the ep, ‘leading to battles in parliament between the more green-focused environmental committee and its agricultural counterpart’ (euractiv.com, 11 july 2013). irish agriculture minister, simon coveney, who oversaw the final stages of negotiations during ireland’s eu presidency, accepted that there had been ‘inter-institutional frictions’ but considered that ‘we have found a balance that everyone can agree with’. others welcomed the broadening out of the decision process to a wider range of actors rather than a handful of national leaders. at the very least meps exercised their formal authority, ‘re-writing parts of the european commission’s original farming proposal and working through 40 rounds of negotiations with national ministers and the commission’. one french environmental campaigner even praised the inclusion of the ep as ‘good for food democracy…it’s not perfect, but it is better than in the past’ (euractiv.com, july 2013). on the other hand, a more obvious interpretation is that offered by a spokesperson for wwf who complained that the ep has proven that it is not ready to handle its new full co-decision powers on the common agricultural policy. at every turn the agriculture committee has tried to water down this reform. it even managed to throw out the few improvements the parliament plenary had requested of them. (euractiv.com, july 2013). conclusions the parallel negotiations on a new mff for 2014-20 and proposals for the reform of the cap after 2013 introduced greater complexity into the picture. it is arguable for example that the priorities of the uk focused more on big picture reductions (or a ‘freeze’) in overall budget expenditure rather than the specifics of cap reform. in budgetary politics, radical reform is much harder under conditions of constraint, and in this case the tight agreement on the mff made it even more difficult to secure radical changes in the cap. so the window that might have opened for the mff to usher in radical cap reform stayed well shut. from the outset it was clear that the mff process would not lead to a radical reconfiguration of the eus budget priorities. although real cuts were made in the resources allocated to agriculture, the cap continues to be a substantial beneficiary from the eu budget. within this framework, the balance between the two pillars will stay broadly the same, with no real transfer of funds from direct payments to measures for rural development and environmental sustainability. if anything, the outcome of the negotiations actually was to water down the original proposals for enhanced environmental measures, achieved in some part by reinforcing the capacity for national variation, and to row back from the liberalisation agenda by reintroducing market intervention measures and allowing the retention of some coupled subsidies. the 2013 cap reform episode again highlights many of the core themes that are present in the extensive literature on agricultural policy in the eu. although in many respects path-dependent, reform is not impossible but takes place in a largely incremental fashion that can nonetheless produce considerable change over time. there remains a tension between a state-assisted and neo-liberal paradigm that can fluctuate between reform episodes, and policy actors working within institutional frameworks negotiate on the basis of preferences, which in the case of many member states continue to be favourable to agriculture. essentially the decision rules and institutional structures around the cap, plus the balance of forces between member states, still works as a barrier against radical policy change. although some governments such as the uk continue to advocate fundamental change, these remain largely voices in the wilderness. most member states support the retention of the familiar cap and its main policy instruments and greer ● the common agricultural policy and the eu budget 133 mechanisms. this preference has been buttressed, at least in the medium term, by the inclusion of the ep in the formal procedures of cap reform. exploiting its enhanced role in the decision-making process, the ep – so far at least – has worked to support the status quo on the cap rather than facilitate radical change. references ackrill, robert (2000a) 'cap reform 1999: a crisis in the making?', journal of common market studies 38(2): 343-53. ackrill, robert (2000b) ‘the european union budget, the balanced budget rule and the development of common european policies’, journal of public policy 20(1): 1-19. ackrill, robert and adrian kay (2006a) ‘historical-institutionalist perspectives on the development of the eu budget system’, journal of european public policy 13(1): 113-33. ackrill, robert and adrian kay (2006b) ‘the eu budget and the cap: an agenda for the review?’, eurochoices 5(3): 20-25. benedetto, giacomo (2013) ‘the eu budget after lisbon: rigidity and reduced spending?’, journal of public policy 33(3): 345-69. benedetto, giacomo (2012) ‘introduction: a history of the european budget and the possibilities for reform’, in giacomo benedetto and simona milio (eds.), european union budget reform: institutions, policy and economic crisis. basingstoke: palgrave macmillan, pp. 1-20. benedetto, giacomo and simona milio (2012) ‘conclusion: budget policy, past experience and the future’ in giacomo benedetto and simona milio (eds.), european union budget reform: institutions, policy and economic crisis. basingstoke: palgrave macmillan, pp. 171-192. bundesministerium für ernährung, landwirtschaft und verbraucherschutz, ministere de l’alimentation de l’agriculture et de al peche (2010) franco german position for a strong common agricultural policy beyond 2013. burrell, alison. (2009) ‘the cap: looking back, looking ahead’, journal of european integration 31(3): 271-89 cameron, david et al. (2010) ‘a letter from prime minister david cameron and other european leaders to the president of the european commission on 18 december 2010’. http://www.number10.gov.uk/news/letter-to-president-ofeuropean-commission/ coleman, william, grace skogstad, michael (1997), ‘paradigm shifts and policy networks: cumulative change in agriculture’, journal of public policy 16(3): 273301. cunha, arlindo with alan swinbank (2011) an inside view of the cap reform process: explaining the macsharry, agenda 2000, and fischler reforms. oxford: oxford university press. daugbjerg, carsten (2009), ‘sequencing in public policy: the evolution of the cap over a decade’, journal of european public policy 16(3): 395-411. daugbjerg, carsten and alan swinbank , ‘ideational change in the wto and its impacts on eu agricultural policy institutions and the cap’, journal of european integration 31(3): 311-27. defra (2011a) cap reform post 2013: defra discussion paper on the impact in england of the eu commission regulatory proposals for the common agricultural policy reform, post 2013. december 2011. european journal of government and economics 2(2) 134 defra (2011b) uk response to the commission communication and consultation: “the cap towards 2020: meeting the food, natural resources and territorial challenges of the future.” january 2011. erjavec, karmen, emil erjavec and luka juvancic (2008) ‘new wine in old bottles: critical discourse analysis of the current common eu agricultural policy reform agenda’, sociologia ruralis 49 (1): 41-55. euractiv.com (11 july 2013) ‘cap 2014-2020: a long road to reform’. http://www.euractiv.com/cap/cap-reform-2014-2020-linksdossier-508393 euractiv.com (11 october 2012) ‘franco-german statement on cap (9 oct. 2012)’. http://www.euractiv.com/cap/france-germany-agree-keep-eu-far-news-515325 euractiv.com (15 february 2012) ‘spain joins france in bid to ring fence cap budget’, http://www.euractiv.com/cap/spain-joins-france-bid-ring-fenc-news-510824 euractiv.com (22 september 2011), ‘britain, poland join calls for 'radical' cap reform’. http://www.euractiv.com/cap/britain-poland-join-calls-radica-news-507829 euractiv.com (12 september 2011) ‘nine eu countries call for slim eu budget’. http://www.euractiv.com/euro-finance/eu-countries-call-slim-eu-budget-news507532 euractiv.com (18 march 2011) ‘reformers left isolated in eu farm policy debate’. http://www.euractiv.com/cap/reformers-left-isolated-eu-farm-policy-debate-news503245 euractiv.com (19 january 2011) ‘sarkozy: france needs no excuse to defend eu farm policy’. http://www.euractiv.com/cap/sarkozy-france-needs-excuse-defend-eufarm-policy-news-501411 europa (2011) ‘press release: the european commission proposes a new partnership between europe and its farmers’, ip/11/118, 12 october 2011. european commission dg agriculture & rural development (2010), communication, the cap towards 2020: meeting the food, natural resources and territorial challenges of the future (brussels, 18.11.2010, com (2010) 672 final). european commission (2011a) communication from the commission to the european parliament, the council, the european economic and social committee and the committee of the regions, a budget for europe 2020, com (2011) 500 final, part i (brussels, 29 june). european commission (2011b), eu budget 2012: for 500 million europeans. for growth and employment. http://ec.europa.eu/budget/library/biblio/publications/2012/budget_folder/budget_20 12_en.pdf european commission (2011c) proposal for a regulation of the european parliament and of the council on the financing, management and monitoring of the common agricultural policy, com (2011) 628/3, 2011/0288 (cod), sec (2011) 1153/4 (brussels). european commission (2011d) impact assessment: common agricultural policy towards 2020, commission staff working paper, xxx sec (2011) 1153 (brussels). european commission (2010) communication, the cap towards 2020: meeting the food, natural resources and territorial challenges of the future, com (2010) 672 final (brussels, 18 november). european parliament (2013) resolution on the european council conclusions of 7/8 february concerning the multiannual financial framework’, b7-0129/2013. greer ● the common agricultural policy and the eu budget 135 greer, alan (2012) ‘reform of the eu budget: implications for the common agricultural policy’, in giacomo benedetto and simona milio (eds.), european union budget reform: institutions, policy and economic crisis. basingstoke: palgrave macmillan, pp. 103-121. greer, alan and tom hind (2012) ‘the lisbon treaty, agricultural decision-making and the reform of the cap: an analysis of the nature and impact of ‘co-decision’, politics and society 31 (4): 331-41. greer, alan (2005) agricultural policy in europe. manchester: manchester university press. grant, wyn (1997) the common agricultural policy. basingstoke, palgrave. harrabin, roger (2013) ‘eu budget deal for farmers raises wildlife concerns’, bbc news 12 february 2013. http://www.bbc.co.uk/news/science-environment21424957. house of commons, environment, food and rural affairs committee (2011a) the common agricultural policy after 2013: government response to the committee’s fifth report of session 2010–11, fifth special report of session 2010–12, hc 1356. june 2011. house of commons, environment, food and rural affairs committee (2011b) the common agricultural policy after 2013, fifth report of session 2010–11, hc 671, vols. i & ii. april 2011. house of commons library (2013), eu multiannual financial framework (mff) 2014-2020, standard note snep 6455, march 2013. keeler, john (1996) 'agricultural power in the european community: explaining the fate of cap and gatt negotiations', comparative politics 28(2): 127-49. laffan, bridget and johannes lindner (2010) ‘the budget: who gets what, when and how’, in helen wallace, mark a pollack and alistair young (eds), policy making in the european union (6th edition). oxford, oxford university press, pp. 207-28. le monde.fr (12 may 2011) ‘sarkozy : "la france veut le maintien du budget de la pac à l'euro près"’. http://www.lemonde.fr/europe/article/2011/05/12/sarkozy-lafrance-veut-le-maintien-du-budget-de-la-pac-a-l-euro-pres_1521067_3214.html. lynggaard kennet and peter nedergaard (2009) ‘the logic of policy development: lessons learned from reform and routine within the cap 1980– 2003’, journal of european integration 31(3): 291-309. núñez ferrer, jorge (2009) ‘for a future sustainable, competitive and greener eu budget: integrating the climate change objectives’. final report of a ceps task force. brussels: centre for european policy studies. paice, jim (2011) http://www.defra.gov.uk/news/2011/08/26/cap-reform-meetings/. rieger, elmar (2000) ‘the common agricultural policy: politics against markets', in helen wallace and william wallace (eds), policy-making in the european union. oxford, oxford university press, pp. 180-210. roederer-rynning, christilla (2010) ‘the common agricultural policy: the fortress challenged’, in helen wallace, mark pollack and alastair young (eds.), policymaking in the european union (6th ed.). oxford: oxford university press, pp. 181206. roederer-rynning, christilla (2003) ‘from "talking shop" to "working parliament"? the european parliament and agricultural change’, journal of common market studies 41(1): 113-35. european journal of government and economics 2(2) 136 roederer-rynning, christilla and frank schimmelfennig (2012) ‘bringing codecision to agriculture: a hard case of parliamentarization’, journal of european public policy 19(7): 1-18. rutz, cordula, janet dwyer and jorg schramek, (forthcoming) ‘more new wine in old bottles? the evolving nature of the cap debate in europe’, sociologica ruralis. skogstad, grace (1998), ‘ideas, paradigms and institutions: agricultural exceptionalism in the european union and the united states’, governance 11(4): 463-90. swidlicki, pawel, anna kullmann and mats persson (2012) seizing the moment: aligning the eu budget with europe’s economic needs. london: open europe. http://www.openeurope.org.uk/content/documents/pdfs/2012eubudget.pdf. swinbank, alan (2012) ‘another reform? proposals for the post-2013 common agricultural policy’, world agriculture: problems and potential, 3(1): 32-37. swinbank, alan (1999) 'eu agriculture: agenda 2000 and the wto commitments', world economy 22(1): 41-54. swinnen, j. (2009)(ed.) the perfect storm: the political economy of the fischler reforms of the common agricultural policy. brussels: centre for european policy studies. swinnen, jo and louise knops (2012) ‘cap reform: will the european parliament take the bull by the horns?’ ceps commentaries. centre for european policy studies: brussels. available from www.ceps.eu de wilde, peter (2012) ‘politicisation of the eu budget: conflict and the constraining dissensus’, west european politics 35(5): 1075-94. wyles, john (2013) ‘an eu budget deal that sends the wrong signals’, european voice, 14 february 2013. http://www.europeanvoice.com/article/imported/an-eubudget-deal-that-sends-the-wrong-signals/76431.aspx. microsoft word ejge_03_01_010.doc european journal of government and economics volume 3, number 1 (june 2014) issn: 2254-7088 5 new party parliamentary entry in western europe, 1960-2010 niklas bolin, mid sweden university, sweden abstract although the number of studies of new political parties is gradually increasing, we still lack a comprehensive understanding of when new parties manage to enter national parliaments. the objective of this article is to explain the circumstances under which new parties enter national parliaments for the first time. unlike earlier research, this study does not focus on individual party families. instead, generally applicable explanations are sought. this is achieved through an examination of the importance of the political opportunity structure, that is, the external environment new parties face in their quest for parliamentary representation. using data collected for 18 western european countries for the period 1960-2010, the analysis shows that institutional factors are important to understand new party parliamentary entry. the sociological factors analysed in the study, however, do not seem to be of equal importance. jel classification d72. keywords new parties; western europe; institutions; representation; political opportunity structure. european journal of government and economics 3(1) 6 introduction although the number of studies of new political parties is gradually increasing, we still lack a comprehensive understanding of when new parties manage to enter the most important political assembly, namely national parliaments. since only very few new parties emerged during the first 70 years of the 20th century, scholars showed little interest in the area. according to the frozen party system theory of lipset and rokkan (1967), the 1960s’ party systems reflected the cleavage structures of the 20th century, with the result that the party systems of the 1960s were virtually the same as those of the 1920s. during the last couple of decades, however, new parties have more frequently succeeded in entering national parliaments. accordingly, scholarly interest has also risen. due especially to the pioneer contributions of hauss and rayside (1978), harmel and robertson (1985) and the more recent studies by hug (2001), meguid (2008) and hino (2012), a new party theory is slowly starting to emerge. nevertheless, these studies, and most others as well, focus either on new party formation or on new party electoral success. although a new party must attain some electoral success to enter parliament, parliamentary entry and electoral success are not synonymous. the objective of this article is to explain the circumstances under which new parties enter national parliaments for the first time. this is achieved through an examination of the importance of the political opportunity structure, that is, the external environment new parties face in their quest for parliamentary representation. since the entry of new parties into parliament is a subject related to research on both party formation and electoral success, i will make use of explanatory factors tested in prior studies of these specific issues. in short, two different explanatory perspectives are employed. firstly, an institutional perspective postulates that a new party’s prospects for successful parliamentary entry are influenced by how the political rules of the game are designed. primarily, attention has been given to the electoral system and its impact on new parties. however, there are also other institutions of relevance. rules on public party funding as well as regulations on parties’ access to different forms of media potentially facilitate or hamper new party possibilities. also, basic structures of the political system have been hypothesised as being significant. hence, federalism and parliamentary government have been argued to be friendlier to newcomers than a unitary presidential political system (harmel & robertson, 1985; hauss & rayside, 1978; hino, 2012; willey, 1998). secondly, from a sociological perspective it is emphasised that political parties arise as a result of conflicts between different groups in society (ware, 1996). however, as major societal changes such as industrialisation, urbanisation and economic growth have taken place, social group affiliation no longer necessarily determines voting patterns. the strict divisions into social groups were loosened when citizens’ opinions and values changed. new dimensions of conflict brought new political issues. since there is a time lag between voters and party movement, new niches open up for new political parties (rydgren, 2003). from a sociological perspective, new party formation and ultimately entry to parliament are therefore a response to social change. this article will go on to survey theories about new parties in order to find out what explanatory factors should be tested in the analysis. in the first part of the empirical section the development of new party parliamentary entry, both temporally and spatially, will be mapped out. this will be conducted using data collected for 18 western european countries for the period 1960-2010. in the second part of the empirical section the explanatory factors will be statistically tested using multivariate statistical methods. finally, the conclusions are discussed and problematised. bolin ● new party parliamentary entry in western europe 7 how can new party parliamentary entry be explained? prior research on new parties is limited in at least two respects. firstly, it focuses on explaining the formation or the varying degrees of electoral success of new parties. while party formation and party parliamentary entry are obviously separate events and thus potentially explained by different factors, new party parliamentary entry and new party electoral success are more adjacent concepts. however, while electoral success is an arbitrary concept to some extent, parliamentary entry is hard to misinterpret. entry includes representation and consequently formal influence on decision-making. secondly, most of the literature focuses on specific party families, and consequently much attention has been given to radical right parties, green parties and regional parties. treating new parties as one homogeneous group of parties has not been very common, although there are a number of important studies which, to different extents, have done just that (harmel & robertson, 1985; hauss & rayside, 1978; hino, 2012; hug, 2001; meguid, 2008). this study aims to go beyond party-family-specific explanations for new party parliamentary entry. although previous explanatory studies on new parties have primarily dealt with formation and electoral success and not parliamentary entry, this study is guided by research from these adjacent areas of research. the results of these studies, so far, are rather mixed. while some argue that the institutional setting is of minor importance to understanding why new parties develop (hauss & rayside, 1978), others argue that is does. hug (2001), for instance, finds that the costs of forming a party do matter. still others have shown that different aspects of the electoral system are of importance in order to explain the electoral success of new parties (hino, 2012). in regard to sociological explanations, harmel and robertson (1985) fail to find any significant effect of various social variables such as population size, religious and linguistic heterogeneity, income equality and levels of post-material sentiments within the population. hino (2012), on the other hand, shows how different socio-economic factors do affect different types of new parties in different ways. in essence, it is possible to identify two dominant theoretical perspectives from which new party research has been conducted: an institutional perspective and a sociological perspective. together these two perspectives contribute to the understanding of what causes new parties to emerge and enter the parliamentary arena. some use the concept political opportunity structure as an illustration of how hard or easy it is for outsiders to enter the political system. this concept originates from research into social movements and is a way to express the degree of ‘openness’ or ‘accessibility’ of a political system for political entrepreneurs (arzheimer & carter, 2006, p. 422). while some have a rather narrow definition of what should be included, others define it as a broader concept which includes a number of factors. here the concept is used to collect both institutional factors and sociological factors under the same heading in order to illustrate the conditions under which new parties attempt to enter parliament. the new party’s chances of parliamentary entry are consequently dependent on a number of factors outside the party. in extreme cases, the political opportunity structure is very unfavourable. one obvious example of such a barrier is a high electoral threshold. under other circumstances the political opportunity structure can be ‘a perfect breeding ground’ for new parties (mudde, 2007, p. 202). favourable possibilities for new parties arise when the institutional barriers are low, or when there is a widespread view that the established parties no longer represent the voters. a benign political opportunity structure can therefore be assumed to be a necessary condition for new parties to be elected to the national parliament. a word of caution is needed here. even though this article focuses on an analysis of the political opportunity structure, it is important to note that this factor alone european journal of government and economics 3(1) 8 does not determine whether a new party succeeds or fails. for a new party to succeed it needs to capitalise on a favourable political opportunity structure, an aspect that research has largely neglected so far. just as new parties are not entirely free to determine their own fate (lucardie, 2000, p. 180), friendly external conditions do not automatically translate into new party parliamentary entry. data and measures in order to better understand the extent to which the political opportunity structure influences the chances of new parties, this study aims to explain the variations present in new party parliamentary entry in elections to the national parliament in 18 western european countries between 1960 and 2010. the dependent variable, i.e. the number of new party parliamentary entries in an election, is defined as a first-time entry into the national parliament by a party that was not originally a member of the once-consolidated party system, defined here as the period prior to 1960 (cf. mair, 1999, p. 210). excluded from this population are merger parties insofar as at least one of the parties included was once a parliamentary party. the reason for this exclusion is that mergers differ in a significant way from split parties and genuinely new parties. while the latter are additional contestants in the electoral race, mergers can be seen as established parties that have simply been reorganised; in other words, old contestants in a new guise. hence, the products of mergers arguably have an easier passage than the split parties and the genuinely new parties in getting elected for the first time. institutional factors the institutional perspective highlights the rules of the political game. primarily, various aspects of the electoral system have been emphasised as being of great importance. in studies of new party success, the electoral system is also a reoccurring predictor that is included in the analyses (e.g. hino, 2012; hug, 2001; willey 1998). as early as the 1950s, duverger (1964) formulated what has since become known as duverger’s law, in which he stated that ‘the simple-majority single-ballot system favours the two-party system’ and ‘the simple-majority system with second ballot and proportional representation, favour multi-partism’ (pp. 217, 239). this would accordingly lead us to expect new parties to do better in countries with proportional electoral systems than in countries with majoritarian systems (harmel & robertson, 1985). however, since only two countries within the data set of this study, namely the uk and france, have majoritarian systems, a desirable level of variation would not be obtained if only a dichotomised variable had been used. empirically, it has also been noted that it is not necessarily the case that proportional electoral systems are more favourable than majoritarian systems from the perspective of small parties (taagepera, 2002). since the total number of votes required for a seat in a country with single-member constituencies can be rather small, a dichotomisation might be misleading. it is also conventional wisdom today that the nature of the electoral system is not solely dependent on the electoral formula (gallagher, 1991; lijphart, 1994; rae, 1967; taagepera & shugart, 1989). in particular, it has been emphasised that the district magnitude can be considered to be as important as explicit thresholds when evaluating the openness of a political system (lijphart, 1994, p. 12). there are also studies that have found that the district magnitude is positively related to the electoral success of new parties (willey, 1998). in other words we cannot, a priori, infer that a country with an explicit threshold has a more closed electoral system than one without an explicit threshold without also taking into account the district magnitude. furthermore, thresholds could be set both nationally and at constituency level. this might be of importance since, for instance, regionally based parties have better chances than parties with electoral bolin ● new party parliamentary entry in western europe 9 support evenly distributed across the country if the seat distribution is mainly based on electoral results at constituency level (bochsler, 2011). in order to circumvent these technicalities, various electoral system measurements have been developed in order to scrutinise their openness (ruiz-rufino, 2007, 2011; taagepera, 2002). in this study, taagepera’s (2002) nationwide threshold of representation, t, will be employed. this measure indicates the percentage of votes a party must obtain in order to have a 50 per cent chance of winning at least one seat. the threshold is calculated according to where m is the district magnitude and s the parliament size. since a country may have an explicit electoral threshold, this must of course be taken into account. thus, the effective threshold is set at the higher of these two values.1 a couple of additional institutional factors have been identified as potentially important for explaining new party success. firstly, it has been argued that the legislature–executive relations of a country matters (hauss & rayside, 1978; willey, 1998). in a presidential system the winner-takes-all character of the presidential election tends to spill over to the parliamentary elections. hauss and rayside (1978, p. 37) argue, for instance, that when ‘attention is focused on the single office of presidency, its zero-sum nature encourages the bipolarisation of the party system and makes it hard for weak parties … to compete effectively’, while meguid (2008, p. 8) asserts that ‘voters do not want to support a candidate whose party is perceived to have no reasonable chance of winning the presidency’. no presidential system exists in the data presented in this study; there are, however, some cases of semi-presidential systems. to test whether this influences the prospects for new parties, a dichotomous variable is included in the analysis to indicate whether the election is held in a semi-presidential system or not. there is also a reason to analyse whether the vertical separation of powers is important. willey (1998, p. 660) argues, for instance, that a new party needs less resources to succeed sub-nationally compared to nationally. in addition, local or regional electoral success may lead to a decreased psychological effect on voters. that is to say, the risk of casting a wasted vote becomes less prominent to the voters if the new party has a record of sub-national electoral success that shows that there is a real chance of parliamentary entry nationally as well (see also müller-rommel, 1998). to avoid losing too much information, the vertical separation of powers will not be measured according to the commonly applied dichotomy of unitary versus federal government. since it is generally acknowledged that both unitary states and federations may vary in the extent to which their political systems are decentralised, the level of decentralisation is measured according to the regional authority index (hooghe, marks, & schakel, 2010).2 1 in this context it is important to emphasise that electoral systems are complex systems and comparisons between countries are difficult. there are several different measurements available to estimate the properties of electoral systems. firstly, there are different measures of electoral system proportionality (gallagher, 1991; loosemore & hanby, 1971; rae, 1967). however, since these measures are indirectly dependent on whether a new party is elected or not, they are not suitable for this study. secondly, since seats are allocated based on the number of votes at the constituency level, it has also been suggested that the analysis should primarily be conducted at this level (bischoff, 2009; selb & pituctin, 2010). as the focus of this study is on national parliamentary entry, neither of these measures is suitable in this context. 2 schakel (2008) shows that the regional authority index is highly correlated with other measures of decentralisation. european journal of government and economics 3(1) 10 finally, two political system factors are also included in the analysis: turnout and voter volatility. regarding turnout, there are well-grounded reasons to expect the level to be of importance. there are, however, mixed opinions about how it might matter. some argue that small parties benefit from low turnout (kestilä & söderlund, 2007; van der eijk, franklin, & marsh, 1996), the reason being that low turnout is a consequence of widespread discontent. such discontent could be expressed either through abstention or a vote for a non-established party (kestilä & söderlund, 2007, p. 785). mathematically, low turnout also decreases the absolute number of votes needed to overcome the electoral threshold (cf. second-order elections, reif & schmitt, 1980). others argue the opposite (bernhagen & marsh, 2007; o’malley, 2008; cf. finseraas & vernby, 2014). this line of argument does, however, depart from the notion that abstention is a strategic choice exercised when the voter is not expecting his/her preferred party to win a sufficient numbers of votes (jackman, 1987). consequently, high turnout might signify that voters are expressing their political discontent (poguntke, 1996, p. 328). electoral volatility is also expected to promote new party parliamentary entry. this is perhaps best explained by a counterfactual. new parties will not win any votes if the established parties do not lose some votes. however, this expected correlation needs to be interpreted with some caution since the direction of correlation is not easily decided. in short, it might be hard to establish whether volatility causes new party parliamentary entry or whether it is a cause of the emergence of new parties. regardless of the direction of the correlation, it seems to be a factor that is of interest. to test it, pedersen’s (1979) index of net volatility, v, is employed: where pi,t is the share of votes for party p at election t and n is the total number of parties. sociological factors from a sociological perspective, political parties are regarded as a result of different groups requiring societal representation (harmel & robertson, 1985, p. 502). all other things being equal, it might be asserted that big countries therefore have a greater number of societal groups with political demands. hence, we might expect there to be a correlation between the size of the country and new party parliamentary entry (cf. tavits, 2006, p. 108). to test this claim, both population size and country area will be analysed. due to the skew distribution, both these variables were transformed using the natural logarithm (ln) (tabachnick & fidell, 2006, pp. 86-88). a second set of sociological factors deserving interest might be labelled socioeconomic or socio-cultural variables. there are, for instance, well-grounded arguments suggesting that new parties are facilitated by transformative events and societal unease. such events, it is believed, cause discontent among voters, who spawn a demand for something new. to test this assertion, the analysis also includes economic indicators such as unemployment rates, inflation rates and economic growth rates (e.g. arzheimer, 2009; müller-rommel, 1998). changes in the demographic composition might also increase the attraction of new parties. there is, for example, some support for the claim that a high level of immigration explains radical right party success (arzheimer, 2009; golder, 2003). the basic variables that were included in the statistical analyses, as well as references to the sources from which the information was taken, are listed in the appendix, in table a1. bolin ● new party parliamentary entry in western europe 11 new party parliamentary entry across space and time as this study deals with democratic politics in western europe, elections in greece, portugal and spain prior to 1980 are excluded from the analysis. in all, 18 countries are included in the study. in these countries, 238 elections were held between 1960 and 2010. the distribution of the dependent variable is shown in table 1. table 1. distribution of dependent variable, number of new party parliamentary entries in election number of new party entries number of cases share of cases, per cent 0 128 53.8 1 65 27.3 2 31 13.0 3 11 4.6 4 1 0.4 5 1 0.4 6 1 0.4 total 238 100 sources: mackie and rose (1991, 1997), political data yearbook 1993-2010 (katz et al., 1993-2010), parties and elections in europe (2011), nohlen and stöver (2010). the most common outcome is that no new party gained representation. new parties entered parliament in less than 50 per cent of the elections. in about 27 per cent of the elections, one new party entered the highest decision-making body, and there are an additional 13 per cent of cases in which two new parties passed the threshold of representation. the table also shows that it is extremely rare that more than three new parties entered the parliament in the same election. notably, six new parties entered the italian lower chamber in 1994.3 the other two cases were spain in 1986 (five new parties) and belgium in 1978 (four new parties). the number of new parties viewed in a comparative perspective is presented in table 2. overall, there were 175 new party parliamentary entries in the 238 elections held. in other words, on average, in three out of four elections just one new party was elected to parliament. considering the number of parties without previous experience of representation that were actually running for parliament, the chances for the average party must be considered slim. 3 this explosion of new parties in italy was a result of the implosion of the old party system, which was caused by serious corruption allegations in the so-called tangentopoli affair. the italian party system was more or less replaced by a new one over the course of a few elections (cotta & verzichelli, 2007, pp. 49-65). european journal of government and economics 3(1) 12 table 2. new party parliamentary entries per country, 1960-2010 country number of elections number of new party entries average number of new party entries italy 13 21 1.62 spain 8 11 1.38 belgium 16 21 1.31 netherlands 15 18 1.20 switzerland 12 14 1.17 uk 13 10 0.77 finland 13 10 0.77 greece 11 8 0.73 ireland 14 10 0.71 luxembourg 10 7 0.70 france 12 8 0.67 portugal 10 6 0.60 iceland 14 7 0.50 denmark 19 9 0.47 norway 13 6 0.46 sweden 16 4 0.25 austria 15 3 0.20 germany 14 2 0.14 total 238 175 0.74 note: correlation eta = 0.41 (p = 0.00) this result shows significant differences between the countries in this study. while there were 21 new party parliamentary entries in the lower chambers of both italy and belgium, corresponding events were extremely rare in the german bundestag. pds, together with die grünen, were the only two german parties who fought their way into the german lower house during the period 1960-2010. a measure of association was calculated in order to statistically test the relationship between country and the number of new party parliamentary entries. with an eta-value of 0.41 there is evidence of a medium-strong correlation, indicating that there is an independent country effect. there is also evidence of temporal variation (see table 3). while there was about one new party parliamentary entries in every second election in the 1960s, this increased to almost one new party per election in the 1980s. during the last decade, however, there has been a significant change in this trend and we are now almost back to the same levels as at the beginning of the period. compared to the spatial variation, the temporal variation is rather small (eta = 0.15). table 3. new party parliamentary entries per decade, 1960-2010 decade number of elections number of new party entries average number of new party entries 1960-69 38 21 0.55 1970-79 48 36 0.75 1980-89 53 49 0.92 1990-99 48 40 0.83 2000-104 51 29 0.57 total 238 175 0.74 note: correlation eta = 0.15 (p = 0.28) if the decrease in new party parliamentary entries is a trend or a coincidence remains to be seen. it should be noted, however, that the high levels in both the 1980s and the 1990s were, in part, the consequence of new party family formations. while the green parties were part of the 1980s’ wave of new parties (e.g. müller-rommel, 1998), the disparate family of anti-immigration parties won electoral success during the period 1990-2000. possibly, therefore, a completely new party family needs to be established if we are to see the same levels of new 4 since data were collected up until 2010, the last period is one year longer than the previous periods. bolin ● new party parliamentary entry in western europe 13 party parliamentary entries again. on the other hand, the green and antiimmigration parties make up only a minority of all new parties. from descriptive inferences, i now turn to the matter of explaining new party parliamentary entry. explaining new party parliamentary entry it has not been possible to collect data for all independent variables during the whole period of the study. as long as missing data are scattered randomly in the data matrix the generalisability is not affected to any great extent (tabachnick & fidell, 2006, p. 62). however, since the missing data largely pertain to specific years or countries and for a particular period of time, some measures have had to be introduced. for this reason, missing data regarding the variables for the regional authority index, unemployment, economic growth, inflation and immigration have been replaced with values for the closest available period. if two values have occurred equally close in time then the mean of these was used. to test whether these replacement values have had an impact on the results, all analyses were also conducted on the original data set. no relevant differences were observed. since no new parties reached the threshold for being admitted to parliament entries in more than 50 per cent of the elections, the dependent variable is transformed into a dichotomised variable where 0 indicates no new party elected and 1 indicates that at least one new party was elected. because of this, the analysis was carried out using multivariate logistic regression.5 since the data are time-series cross-section with a binary dependent variable, there is a risk that the outcome of the dependent variable is not independent of previous outcomes. such serial correlation might underestimate the true standard errors and thus may give rise to an overestimation of the strength of any correlation (beck, katz, & tucker, 1998, pp. 1260-1261). to capture this, a first lag of the dependent variable was included in the analysis. there seems to be no consensus about which measures best evaluate a logistic regression model (garson, 2012). however, in order to provide sufficient information to evaluate the plausibility of the interpretations made, this study follows the recommendations of peng et al. (2002). firstly, information on whether each model performs better than a model with no predictors, the so-called null model, is presented. this is done by presenting a -2 log likelihood and associated chi-square values. a significant chi-square value signals that the model is significantly better than the null model (tabachnick & fidell, 2006, p. 448). secondly, information is provided on how well the model performs using the nagelkerke r square. although this measure is not equivalent to the r-square value which is usually presented in ols regression, it is often used to provide an intuitive picture of how well the model explains the phenomenon under study. the nagelkerke r square is also supplemented with data on what proportion of the outcomes is correctly predicted. finally, the individual independent variables are evaluated through the presentation of coefficients, significant tests and odds ratios in order to give an idea of how much each factor affects the probability of new party 5 an alternative to this would be to use a count model in which the dependent variable is the number of times something has occurred (long, 1997, p. 217). since new party entrances are rather rare events, the average of the dependent variable is close to 0 and similar to a poisson distribution. however, a poisson regression model assumes that each event must be independent of other events (king, 1998, p. 50) and that the variance of the dependent variable is equal to its mean (dunteman & ho, 2006, p. 23; hoffmann, 2004, p. 104). since these requirements are not met, the negative binomial distribution is a more appropriate statistical tool (hoffmann, 2004, p. 12). in order to test the robustness of the logistic regressions, this regression model was also estimated. the result from this analysis provides no reason to change the overall conclusions drawn in this article. european journal of government and economics 3(1) 14 parliamentary entry. to illustrate the probabilities, the presentation of the analysis is also illustrated with two graphs. table 4 presents the results of three logistic regressions. in the first model, the institutional factors are tested. the second model tests the sociological factors, while the third model includes all independent variables. the first aspect that is worth noting is that while both the institutional and the full models are significantly different from the null model, the sociological model is not (chi square = 10.639, df = 7, p = 0.155). for this reason, the following presentation therefore focuses on the other two models. with nagelkerke r-square values of 0.16 and 0.18 and correctly predicted outcomes of 67.5 per cent (compared to 53.6 per cent in the null model) and 70 per cent (54.1 per cent) respectively, one might argue that the overall performance of the two models is rather moderate. to fully understand new party parliamentary entry, we might therefore conclude that other important factors exist, factors which are not included in the models. the effective threshold and volatility both have significant and independent effects on new party parliamentary entry. while this is true for both the institutional model and the full model, it is also possible to conclude that turnout is significant in the latter. with a negative coefficient, support is given to the argument that increased turnout reduces the likelihood of new party parliamentary entry. it is also important to note that none of the sociological predictors reach the level of significance. the odds ratios illustrate the importance of the individual explanatory factors for the probability of a new party entering parliament. in both the institutional and the full models the odds ratio for the effective threshold is about 0.7. this means that the odds of at least one new party entering parliament are reduced by almost 30 per cent and there is an increase in the effective threshold of one percentage point. the odds ratios for volatility and voter turnout indicate that the effect of these variables is smaller. the odds for new party parliamentary entry are expected to increase by about 8 per cent if volatility is increased by one unit. the corresponding change in the odds is less than minus 3 per cent units for a 1 per cent unit increase in turnout. small numbers, however, are not necessarily the same as marginal importance. in order to capture fully the extent to which these factors influence the chances of new party parliamentary entry, graphical presentation is helpful. bolin ● new party parliamentary entry in western europe 15 table 4. logistic regression: new parliamentary entry institutional model sociological model full model b (s.e.) odds ratio b (s.e.) odds ratio b (s.e.) odds ratio effective threshold -0.326** (0.138) 0.722 -0.307** (0.155) 0.735 executive -0.141 (0.487) 0.868 0.278 (0.541) 1.321 regional authority index 0.001 (0.017) 1.001 0.024 (0.025) 1.025 turnout -0.020 (0.013) 0.980 -0.024* (0.014) 0.976 volatility 0.076*** (0.020) 1.079 0.076*** (0.021) 1.079 population (ln) 0.086 (0.122) 1.090 -0.115 (0.173) 0.891 area (ln) -0.210 (0.135) 0.811 -0.144 (0.163) 0.866 unemployment rate 0.055 (0.034) 1.057 0.042 (0.036) 1.043 economic growth rate 0.016 (0.050) 1.016 0.040 (0.053) 1.041 inflation 0.023 (0.018) 1.023 0.022 (0.019) 1.023 net migration 0.001 (0.033) 1.001 -0.001 (0.035) 0.999 lag dependent variable 0.114 (0.292) 1.121 0.523* (0.272) 1.687 0.123 (0.299) 1.130 constant 1.208 (1.089) 1.151 (1.316) 3.336* (1.975) -2 log likelihood 296.710 312.043 286.721 chi2 30.622*** 10.639 34.735*** nagelkerke r2 0.162 0.059 0.185 correct predicted (intercept only) 67.5 (53.6) 59.4 (54.3) 70.0 (54.1) n 237 234 233 note: *** p<0.01 ** p<0.05 * p<0.10 perhaps the easiest way to grasp an understanding of the significance of odds ratios is to convert them to probabilities and illustrate them in line graphs. figure 1 and figure 2 present the predicted probabilities that at least one new party enters parliament following the election. the graphs are derived from estimates in the full model, and each variable, apart from the two analysed in each figure, is held constant at its mean value. figure 1 shows the plots for the predicted probability of new party parliamentary entry for three different levels of electoral volatility as well as for the effective threshold. the graph shows that the predicted probability of new party parliamentary entry when the electoral volatility is held at its mean (12 per cent) is about 70 per cent if no electoral threshold exists at all. with a mean effective threshold of 2.1 per cent the predicted probability of a new party parliamentary entry is about a 56 per cent chance. the figure also presents the predicted probabilities for the minimum and maximum values of electoral volatility. european journal of government and economics 3(1) 16 figure 1. predicted probabilities of new party parliamentary entry; effective threshold and volatility in figure 2, three levels of turnout are plotted against effective thresholds. the line for the mean turnout is identical to the line for mean electoral volatility in figure 1. the graph illustrates that low turnout is expected to increase the chances of new party parliamentary entry. actually, with a minimum turnout (42 per cent) and an effective threshold below 1.5 per cent, there is a predicted chance of more than 80 per cent for a new party parliamentary entry. however, these low levels of turnout are rare. since turnout is somewhat unevenly dispersed (median = 82.8 per cent) it might therefore be interesting to compare the lines of the mean and the maximum turnout. here, it can be observed that the differences between these lines are rather small. this indicates that the actual impact of turnout, albeit significant, is rather marginal. bolin ● new party parliamentary entry in western europe 17 figure 2. predicted probabilities of new party parliamentary entry; effective threshold and turnout the overall impression of the statistical analysis is that the institutional perspective can, in part, explain new party parliamentary entry, while the sociological perspective is less supported in this study. in the final section these results are discussed and problematised. concluding remarks the party systems of western europe are no longer frozen. ever since the 1960s, new parties have been making inroads into national parliaments. the situation, however, is not the same in all countries. whereas countries such as italy and belgium have experienced a veritable explosion of new parties in their national parliaments, other countries, most notably germany, almost never experience new party parliamentary entry. while there might be many reasons for why this variation exists, this article has primarily dealt with the environment in which new parties attempt to win representation. under the epithet of political opportunity structure, theoretically derived explanatory factors have been tested from both an institutional and a sociological perspective. the results are mixed. while the institutional perspective provides us with some explanatory power, the sociological perspective contributes less to our understanding. the finding that the electoral system is important is hardly surprising and is well in line with prior research on related research questions such as what accounts for new party formation and electoral success (e.g. hino, 2012; willey, 1998). this finding is also of practical importance for policymakers. however, since constitutional politics, like any other policy area, is decided by the established parties, perhaps we should not expect this knowledge to be of much help for fledgling political parties. much less so if we are to believe the abundance of literature on party cartelisation (e.g. katz & mair, 1995, 2009). as expected, there is also a strong relationship between electoral volatility and the occurrences of new party parliamentary entries. perhaps more revealing is the fact that new parties seem to prosper when there is low turnout. while this has been suggested in previous studies, there are also many that have argued the opposite. from a new party perspective this finding might open up avenues of potential success, since european journal of government and economics 3(1) 18 low expected turnout, typical in so-called second-order elections, would enthuse full mobilisation. there are several potential reasons for the poor results from the sociological perspective. firstly, there might of course be methodological issues of concern: perhaps important explanatory factors have been omitted. secondly, there is always a risk that secondary data fails to meet the strongest criteria of comparability. however, and perhaps most importantly, in the light of this study, the weak support for the sociological perspective suggests that there are few such factors that influence all new parties in the same way. while high levels of immigration might have an impact on the vote for anti-immigration parties, the same might not be true for, say, green parties. also, while many argue that far right parties would prosper during an economic downturn, recent analysis suggests that they have not experienced any uniform electoral upswing despite the major financial crisis from 2008 onwards (mudde, 2010). it can be ascertained that, to some extent anyway, one of the aims of this paper, namely to find generic explanations for new party parliamentary entry, proved to be difficult to achieve. therefore, one of the conclusions of this study is that although new parties share some similarities, we still need to differentiate between different groups in order to fully understand when and why they manage to gain representation. furthermore, and regardless of the findings in this article, the political opportunity structure is just one side of the story. for new party parliamentary entry to occur, agency is also necessary and should therefore be taken into account (bolin, 2012; bolleyer 2013). recent studies have, for instance, suggested that new party success is also dependent on both the actions taken by the established parties (e.g. meguid, 2008) and the resources available to the new party (e.g. lucardie, 2000). of course, it is also important to note that new parties differ in terms of the extent to which they actually aim for parliamentary entry. while some parties are intrinsically vote-seeking and regard parliamentary entry as crucial, others primarily aim to put new or non-politicised issues on the agenda (cf. strøm, 1990). although there seems to be little to suggest that such goals do vary systematically between different countries, it is important to acknowledge that strategic considerations on behalf of new parties are also of relevance if we are to fully grasp why specific parties do or do not attain parliamentary representation. so, while this article has furthered our understanding of the importance of the external environment, more work needs to be done in order to gain systematic knowledge of the actions taken by new parties and to what extent these parties can be masters of their own destiny. references armingeon, klaus, david weisstanner, sarah engler, panajotis potolidis, marlène gerber and philipp leimgruber (2011) ‘comparative political data set i 19602009’, url (consulted 16 november 2011): http://www.ipw.unibe.ch/content/team/klaus_armingeon/comparative_political_data _sets/index_eng.html. arzheimer, kai (2009) ‘contextual factors and the extreme right vote in western europe, 1980–2002’, american journal of political science, 53(2): 259–275. arzheimer, kai and elisabeth carter (2006) ‘political opportunity structures and right-wing extremist party success’ european journal of political research, 45(3): 419-443. beck, nathaniel, jonathan n.kat and richard tucker (1998) ‘taking time seriously: time-series-cross-section analysis with a binary dependent variable’, american journal of political science, 42(4): 1260-1288. bolin ● new party parliamentary entry in western europe 19 bernhagen, patrick, and michael marsh (2007) ‘the partisan effects of low turnout: analyzing vote abstention as a missing data problem’, electoral studies, 26(3): 548-560. bischoff, carina s. (2009) ‘national level electoral thresholds: problems and solutions’, electoral studies, 28(2): 232–239. bochsler, daniel. (2011) ‘it is not how many votes you get, but also where you get them. territorial determinants and institutional hurdles for the success of ethnic minority parties in post-communist countries’, acta politica, 46(3): 217–238. bolin, niklas (2012) målsättning riksdagen. ett aktörsperspektiv på nya partiers inträde i nationella parlament. umeå: umeå university. bolleyer, nicole (2013) new parties in old party systems. persistence and decline in seventeen democracies. oxford: oxford university press. cotta, maurizio, and luca verzichelli (2007) political institutions in italy. oxford: oxford university press. dunteman, george. h., and moon-ho r ho (2006) an introduction to generalized linear models. thousands oaks: sage. duverger, maurice (1964) political parties. their organization and activity in the modern state. london: methuen. finseraas, henning and kåre vernby (2014) ’a mixed blessing for the left? early voting, turnout and election outcomes in norway’, electoral studies, 33: 278-291. gallagher, michael (1991) ‘proportionality, disproportionality and electoral systems’, electoral studies, 10(1): 33-51. garson, g. david (2012) logistic regression: binomial and multinomial. asheboro, nc: statistical associates publishers. golder, matt (2003) ‘explaining variation in the success of extreme right parties in western europe, comparative political studies, 36(4): 432-466. harmel, robert, and john. d. robertson (1985) ‘formation and success of new parties a cross national analysis’, international political science review, 6(4): 501-523. hauss, charles, and david rayside (1978) ‘the development of new parties in western democracies since 1945’, in louis maisel and joseph cooper (eds) political parties: development and decay. beverly hills: sage. hino, airo (2012) new challenger parties in western europe. a comparative analysis. london: routledge. hoffmann, john p. (2004) generalized linear models. an applied approach. boston: pearson education. hooghe, liesbeth, gary marks and arjan. h. schakel (2010) the rise of regional authority: a comparative study of 42 democracies (1950-2006). london: routledge. hug, simon (2001) altering party systems strategic behavior & the emergence of new political parties in western democracies. ann arbor: the university of michigan press. jackman, robert w. (1987) ‘political institutions and voter turnout in the industrial democracies’. american political science review, 81(2): 405-424. katz, richard s., hans keman, ruud koole, peter mair, edward. c. page and ingrid van biezen, (eds) (1993-2010). political data yearbook: 1993-2010. european journal of government and economics 3(1) 20 katz, richard s., and peter mair (1995) ‘changing models of party organization and party democracy. the emergence of the cartel party’, party politics, 1(1): 528. katz, richard s., and peter mair (2009) “the cartel party thesis: a restatement”, perspectives on politics, 7(4): 753-766. kestilä, elina and peter söderlund (2007) ‘subnational political opportunity structures and the success of the radical right: evidence from the march 2004 regional elections in france’, european journal of political research, 46(6): 773796. king, gary (1998) unifying political methodology.the likelihood theory of statistical inference. ann arbor: university of michigan press. knutsen, oddbjørn (2011) ‘western europe: parties, voters and elections’, url (consulted 17 november 2011): http://folk.uio.no/stvok1/. lijphart, arend (1994) electoral systems and party systems. oxford: oxford university press. lipset, seymour m., and stein rokkan (1967) ‘cleavage structures, party systems, and voter alignments: an introduction’, in seymour. m. lipset and stein rokkan (eds), party systems and voter alignments. new york & london: free press, pp. 1-64. long, j. scott (1997) regression models for categorical and limited dependent variables. thousand oaks: sage. loosemore, john, and victor. j. hanby (1971) ‘the theoretical limits of maximum distortion: some analytic expressions for electoral systems’, british journal of political science, 1(4): 467-477. lucardie, paul (2000) ‘prophets, purifiers and prolocutors’, party politics, 6(2): 175-185. lundell, krister and lauri karvonen (2011) a comparative data set on political institutions. åbo: åbo akademi university. mackie, thomas t. and richard rose (1991) the international almanac of electoral history. third edition. london: macmillan. mackie, thomas t. and richard rose (1997) a decade of election results: updating the international almanac. glasgow. centre for the study of public policy, university of strathclyde. mair, peter (1999) ‘new political parties in established party systems: how successful are they?’ in erik beukel, kurt k. klausen and poul e. mouritzen (eds) elites, parties and democracy. odense: odense university press, pp. 207-224.. meguid, bonnie m. (2008) party competition between unequals. strategies and electoral fortunes in western europe. cambridge: cambridge university press. mudde, cas (2007) populist radical right parties in europe. cambridge: cambridge university press. mudde, cas (2010) ‘the populist radical right: a pathological normalcy’, west european politics, 33(6): 1167-1186. müller-rommel, ferdinand (1998) ‘explaining the electoral success of green parties: a cross-national analysis’, environmental politics, 7(4): 145-154. nohlen, dieter and philip stöver (eds) (2010) elections in europe: a data handbook. baden-baden: nomos. bolin ● new party parliamentary entry in western europe 21 o’malley, eoin (2008) ‘turn out to turf out?: effects of changes in election participation rates on election outcomes’, the open political science journal, 1(1): 31-37. oecd (2010a) ‘oecd employment and labour market statistics’, url (consulted 14 november 2011): http://www.oecd-ilibrary.org/employment/data/oecdemployment-and-labour-market-statistics_lfs-data-en. oecd (2010b) oecd factbook 2010: economic, environmental and social statistics. oecd publishing. oecd (2011) ‘employment and labour markets: key tables from oecd’, url (consulted 16 november 2011): http://www.oecdilibrary.org/employment/unemployment-rate-2011_unemp-table-2011-1-en. parties and elections in europe (2011), url (consulted november-december 2011): http://www.parties-and-elections.de/. pedersen, mogens n. (1979) ‘the dynamics of european party systems: changing patterns of electoral volatility’, european journal of political research, 7(1): 1-26. peng, chao-ying j., kuk. l. lee and ingersoll, gary m. (2002) ‘an introduction to logistic regression analysis and reporting’, journal of educational research, 96(1): 3-14. poguntke, thomas (1996) ‘anti-party sentiment conceptual thoughts and empirical evidence: explorations into a minefield’, european journal of political research, 29(3): 319-344. rae, douglas w. (1967) the political consequences of electoral laws. new haven: yale university press. reif, karlheinz and hermann schmitt (1980) ‘nine second-order national elections a conceptual framework for the analysis of european election results’, european journal of political research, 8(1): 3-44. ruiz-rufino, rubén (2007) ‘aggregated threshold functions or how to measure the performance of an electoral system’, electoral studies, 26(2): 492-502. ruiz-rufino, rubén (2011) ‘characterising electoral systems: an empirical application of aggregated threshold functions’, west european politics, 34(2): 256-281. rydgren, jens (2003) the populist challenge: political protest and ethnonationalist mobilization in france. new york: berghahn books. schakel, arjan h. (2008) ‘validation of the regional authority index’, regional & federal studies, 18(2-3): 143-166. selb, peter and sandrine pituctin, (2010) ‘methodological issues in the study of new parties’ entry and electoral success’, party politics, 16(2): 147-170. strøm, kaare (1990) ‘a behavioral theory of competitive political parties’, american journal of political science, 34(2): 565-598. taagepera, rein (2002) ‘nationwide threshold of representation’, electoral studies, 21(3): 383-401. taagepera, rein, and matthew s. shugart (1989) seats & votes. the effects & determinants of electoral systems. new haven & london: yale university press. tabachnick, barbara g. and linda s. fidell, (2006) using multivariate statistics. fifth edition. boston: allyn and bacon. tavits, margit (2006) ‘party system change. testing a model of new party entry’, party politics, 12(1): 99-119. european journal of government and economics 3(1) 22 van der eijk, cees, mark franklin and michael marsh (1996) ‘what voters teach us about europe-wide elections: what europe-wide elections teach us about voters’, electoral studies, 15(2): 149-166. ware, alan (1996) political parties and party systems. oxford: oxford university press. willey, joseph (1998) ‘institutional arrangements and the success of new parties in old democracies’, political studies, 46(3): 651-668. world bank (2011) ‘world development indicators’, url (consulted 16 november 2011): http://data.worldbank.org/data-catalog/world-development-indicators. bolin ● new party parliamentary entry in western europe 23 appendix table a1. variables and sources variable sources number of new party parliamentary entries based on election results presented in mackie and rose (1991, 1997), political data yearbook 1993-2010 (katz et al., 1993-2010), nohlen and stöver (2010), parties and elections in europe (2011) effective threshold lundell and karvonen (2011), own calculations executive lundell and karvonen (2011) regional authority index hooghe et al. (2010) turnout lundell and karvonen (2011) volatility knutsen (2011) population lundell and karvonen (2011) country area lundell and karvonen (2011) unemployment rate armingeon et al. (2011), oecd (2010a, 2011) inflation world bank (2011) economic growth world bank (2011) net migration oecd (2010b) regional demands for policy participation in the eu multilevel system european journal of government and economics volume 1, number 1 (june 2012) issn: 2254-7088 9  this paper emerged from the project “governance preferences in the european multilevel system” funded by the german research foundation (grant ba3658/2-1), whose support is gratefully acknowledged. we are indebted to diana pitschel for her collaboration in the project and for her input to earlier versions of this paper. we also thank arjan schakel and charlie jeffery for valuable comments. ** address for correspondence: prof. dr. michael w. bauer, chair of politics and public administration, humboldt-universität zu berlin, institute of social sciences, unter den linden 6, d-10099 berlin, e-mail: mw.bauer@sowi.hu-berlin.de. regional demands for policy participation in the eu multilevel system∗ philipp studinger, humboldt-universität zu berlin, germany michael w. bauer, humboldt-universität zu berlin,** germany abstract over the past 50 years, an increasing amount of political authority has been delegated to the regional government level in europe. this paper analyses regional demands for involvement in policy-making by focusing on the preferences of top-level regional civil servants (“regio-crats”). a survey (n=347) of regio-crats in 60 regions of 5 european union member states serves as the empirical basis for the analysis of regional demands for policy involvement in the multilevel system. the data reveal differential patterns of demands. by and large, regio-crats emerge as being conservative, incremental and modest in their wishes for greater policy involvement, except where the regional contexts are characterised by substantial emancipatory political ambitions or cultural distinctiveness. regional demands for policy participation in the multilevel system are pragmatic, patch-worked and incremental, and more conservative than transformative. jel classification d73; h77; r50 keywords subnational bureaucrats; european union; policy allocation; multilevel system doi: https://doi.org/10.17979/ejge.2012.1.1.4274 european journal of government and economics 1(1) 10 introduction over the past 50 years, regions – and especially those in today’s european union – have been entrusted with ever greater political authority (hooghe et al. 2010). however, notwithstanding the growing political importance of regions, we know very little about the elites who run these subnational political structures, who shape regional political decisions and implement political programmes. the aim of this paper is twofold. first, we want to help fill this research gap by focussing on the preferences of top-level regional civil servants (“regio-crats”). second, we put centre stage the question of the role of regions in the european multilevel governance context. in particular, two questions are raised: what explains regiocrats’ preferences regarding involvement in policy-making (by their respective regions) in the multilevel system and, do we need regional-level variables in order to conduct a satisfying analysis of subnational preferences regarding vertical competence allocation? the questions we raise are of interest to those who study regions as agents of ongoing transformation of states and the rise of multilevel political orders, in general, and also to those investigating the regions of the european union, in particular. we look for patterns in the regio-crats’ preference variations and suggest explanations for them. in order to answer our research questions, we conducted a survey among top-level subnational officials in five european countries and asked them about their preferences with respect to competence for their regions in twelve policy areas. our quest for explanations is based on a view on socio-economic, political and cultural aspects of regional authorities. we believe that there is a link between the beliefs and attitudes of regio-crats (especially those at the top of their organisations) and the political structures for which they work. the individual – within certain range – can choose his actions and does so strategically with view of his self-interests. but he is also shaped by and thus embodies the way his organisation interprets outside reality.1 moreover, by working in an institution an individual himself becomes the bearer of the norms, scripts and standard assessments of the collectivity he represents (egeberg 2004).2 under this assumption, we argue that the preferences of regio-crats represent a general estimate of subnational political ambitions to participate in the european multilevel governance system. the paper is structured as follows. in section 2, we present competing approaches (individual, functional, national and regional) for how best to explain regio-crats’ preferences regarding policy competence allocation in multilevel constellations. section 3 describes the survey we conducted and outlines the operationalisation of the explanatory approaches. the statistical analysis follows in section 4, while section 5 presents the conclusions drawn on the basis of our findings. as the analysis will show, we find a number of regional variables that are persistently significant across several different models. in particular, the variables concerning the cultural distinctiveness, institutional endowment and socioeconomic performance of regions stand out. in addition, the analysis shows that the personal opportunities of regio-crats influence their preferences regarding greater policy competence at the regional level. 1 the use of male pronouns throughout the text is for ease of reading only and should be taken to refer to either gender. 2 institutionalists argue that “members of an institution observe and are the guardians of its constitutive principles and standards” and that their behaviour is based on a “logic of appropriateness and a sense of obligations and rights derived from an identity, role, or membership in a political community and the ethos and practices of its institutions” (olsen 2009: 9). european journal of government and economics 1(1) 11 underlying assumptions before we elaborate on our theoretical models and begin the descriptive and analytical analyses, a word about the subnational data on which they are based seems in order. it is extremely difficult to acquire this kind of data, especially if it is to be suitable for cross-national comparisons. usually, the researchers who are interested in subnational issues will have to obtain it themselves, which, in fact, was also the case for this study (details below). more concretely, our dependent variables are based on data illustrating the attitudes of regio-crats. attitudinal data relating to elites in general have both strengths and weaknesses, and so do our subnational elite data. that is, individuals who work in the upper echelons of subnational authorities and have direct links to the political sphere probably form their preferences under different conditions than does the broader, generally less well-informed and less well-trained national public. one can thus assume that regio-crats, with their particular expertise and their routine familiarity with the regional political universe, hold views which (especially with respect to core topics related to their respective political entities) are different, more reliable and more coherent than those of the general public in their regions. our research question is the following: what explains regio-crats’ attitudes concerning competence allocation in the multilevel system? in other words, what are the factors, with respect to a range of predefined policy areas, that lead regiocrats to desire codetermination rights? the standard explanatory factors found in the literature are based on the concepts of individual utility, contextual variables and functionality (hooghe 2001; loveless and rohrschneider 2008). we test these standard programmes, focusing, in particular, on the regional context. factors that reflect the regional context are summarised under an explanatory programme based on regional social identity. we want to see whether and under which conditions such regional factors have an added value in their own right in explaining differences in preferences regarding regional participation in the european multilevel governance system (pitschel 2012). we now turn to these explanatory programmes. the individual opportunity approach the first explanatory programme is strongly related to the utility calculus. it argues that individuals calculate the respective impacts of different options on their personal opportunity structures. as a consequence, when faced with a choice, individuals will favour the option they believe will lead to a higher personal pay-off (thielemann 2004: 367). other studies have shown that individual opportunity structures – especially in the case of bureaucrats – are driving forces behind preference formation (niskanen 1971; dunleavy 1985; searing 1991, 1994).regarding the question of regional authorities’ policy competences, we argue that top-level subnational bureaucrats will desire the allocation of competences to the regional level if they expect some gain for themselves. it is difficult to define the actual utility that might be involved. on the one hand, rational-choice research refers exclusively to tangible goods that are directly available to the individual (kato 1996). accordingly, financial incentives should be the driving force behind preference formations (hooghe 2001). on the other hand, public-administration research often uses a broader conceptualisation of an individual’s utility. in this context, the commodity to be maximised does not refer exclusively to the individual but also to his social context (levi 1997). in other words, tangible as well as non-financial goods – such as the prestige or power of a region – might play a role in the formation of subnational administrators’ preferences. european journal of government and economics 1(1) 12 applying such a broad concept of utility would obviously be problematic, because all manner of motives and factors that somehow relate to a utility consideration could be included.3 we thus follow the narrow conceptualisation of utility and consider only those aspects that are directly related to the amendment of the subnational administrators’ positions. research on the motivations of publicadministration personnel has demonstrated that administrative elites are basically motivated in their career prospects and their opportunities for advancement (searing 1994: 19; hooghe 2001: 21). regio-crats compete for leverage in designing policies. the more competences the regional level of authority has, the higher this leverage. in sum, the opportunity approach argues that regio-crats will desire more regional policy competences whenever they perceive that this would enhance their individual situations. the functionality approach the second explanatory programme concerns the nature of policies themselves. recent studies find that functionality is an influential factor when the allocation of competences is examined (schakel 2009). similarly, hooghe’s study (2003) on top-level officials in the european commission and in national administrations demonstrates that political elites are guided by functional considerations when they are asked which policies should be regulated at the supranational and which at the national level. the issue of functionally efficient policy allocation is debated in the literature on fiscal federalism.4 oates’ theorem states that, in the absence of problems of redistribution and negative external effects, policies should be allocated at the lowest possible hierarchy level (oates 1999: 1122). in addition, although economies of scale might push the provision of most public goods and services towards the national (or even higher) political levels, possible heterogeneity of local preferences, which would pull allocation of competences to subnational levels, also has to be considered. especially because detailed information about local diversity (which would be needed to design efficient policy solutions) is difficult to obtain and to process centrally, efficient (i.e., functional) allocation of competences is usually not quite as central as the functionality paradigm might seem to predict at first sight (hooghe and marks 2005). concerning regio-crats, we can assume that they come with special knowledge about these characteristics of the different policies. in consequence, it is plausible to argue that they are able to roughly rate the efficiency of different options regarding policy competence allocation. in sum, if top-level subnational bureaucrats base their preferences regarding policy competences for regions on a rationale of functionality, then they should favour policy competences for subnational entities only if regulation at this level of authority is functionally efficient. the subnational social identity approach the explanatory programme we call the subnational social identity approach sees the affiliation of individuals to socially defined groups as an important factor in the formation of preferences (mclaren 2002; diez medrano 2003; scully and farrell 2003; thielemann 2004; hooghe and marks 2005; van esch 2006; de vries and van kersbergen 2007). individuals form and orient their preferences in line with the norms and values of their own social groups. it is clear that subnational administrators are part of the socially defined group of the region to which they belong and for which they work. since they hold top-level positions within their 3 for a detailed discussion of this question, and for theoretical arguments, see pitschel (2012). 4 “the main analytical task for fiscal federalism has been to define the appropriate assignment of allocative responsibilities to decentralised government levels and matching revenue sources” (bird et al. 2002: 416). european journal of government and economics 1(1) 13 regional administrations, they prepare, design and implement policies and political decisions and consequently are familiar with the political interests of their region. in consequence it is plausible to assume that the interests and preferences of regions’ institutions are a primary focus of the regional administrators.5 according to the social identity approach, top-level bureaucrats should pursue the interests of their own regions and internalise regional preferences concerning political issues. before we can analyse the desire for subnational policy competences, we first need information about regional ambitions regarding these competences. we can draw on the literature on subnational mobilisation that discusses policy-making processes (hooghe 1995; jeffery 1996, 2000; tatham 2008) to find explanatory factors for the emancipatory efforts of regions. because the endowment of regions with policy competences reflects one aspect of emancipatory aspirations, we can use these same factors in order to determine subnational preferences. generally, we argue that regions seeking to enhance their political profile will want to expand their political competences. one important factor discussed in the subnational mobilisation literature is the socioeconomic situation of regions (bookman 1992; harvie 1994; marks et al. 1996). however, there is disagreement about the actual direction of this relationship. on the one hand, socioeconomically strong regions want to gain some independence from central government and assume competence for certain policies (gourevitch 1979). on the other hand, socioeconomically weak regions are also believed to have an incentive to take matters into their own hands with a view to advancing their economic development independently of the nation state (hechter 1975; fearon and van houten 2002). another factor that influences the emancipatory potential of a region is its cultural distinctiveness. subnational authorities that deviate from the national population with respect to ethnic or religious characteristics are culturally distinct from the nation state. this regional identity induces the desire to safeguard selfdetermination (esman 1977; connor 1994; keating 1998; alesina and spolaore 2003; björklund 2006; brancati 2006). therefore, subnational entities that are culturally distinct from the nation state should be interested in holding competence with respect to many policies. third, party-political competition is believed to influence subnational demands for autonomy (van houten 2003, 2009). in particular, an incongruence in the composition of the governmental coalition between the regional and national levels might produce disagreement in policymaking and consequently stimulate the desire for policy competences among regional elites. finally, the institutional setting of subnational authorities – also discussed in the mobilisation literature – also has to be taken into account. on the one hand, regions that are already institutionally well endowed are argued to be more active promoters of policies. the less restricted subnational authorities are with respect to their policy engagement, the greater their (potential) scope for development. on the other hand, the status quo of regional competences should be a good predictor of desire for competence allocation because the current setting to some extent determines the possibility of increased allocation. in other words, the degree of constitutionally defined regional autonomy should be correlated with the amount of policy competence desired by regio-crats. 5 we take some comfort from the findings of the cans project (citizenships after the nation state). the team around charlie jeffery, ailsa henderson and roland sturm find in their regional population suvey broad support for regional authority. see http://www.institute-ofgovernance.org/major_projects/citizen_after_the_nation_state citizens’ and oberhofer et al. 2011. european journal of government and economics 1(1) 14 to sum up, the social identity approach assumes that regio-crats’ preferences are influenced by specific characteristics of the regions they work for. in short, the greater the emancipatory potential of a region, the more competences for their region the elites should favour. table 1 summarises our three candidate explanatory programs. table 1: explanatory approaches and respective hypotheses explanatory approach hypothesis dependent variable individual opportunity regio-crats should desire more regional policy competences if they expect a positive impact on their individual situation. functional criteria regio-crats should favour policy competences for the regions if regulation at this level of authority is functionally efficient. subnational social identity regio-crats should desire more regional policy competences if they belong to a region with a high emancipatory potential. regio-crats’ preferences regarding regional policy competences source: authors’ compilation. research design in order to test for the territorial effect on regio-crats’ preferences regarding policy allocation, we make use of a survey addressed to subnational administrative elites concerning their attitudes about different aspects of european integration and governance. we defined the subnational units we are interested in as political authorities which are located directly below the national level and have an elected assembly.6 the selection of interviewees – high-ranking officials in subnational administrations – was carried out in several stages. first, we had to decide which european member states should be included in the survey. our aim was to ensure that the sample would feature interviewees from states with different institutional settings (decentralised and federal states). we also wanted to include entities with differences in their experience with the european integration process. we assumed that the older member states would have already consolidated their governance structures in the context of european integration. thus, we decided to interview regio-crats in germany, spain, france, poland, and hungary comprehending old and new member states as well as decentralised and federal states. second, the selection of the subnational units was guided by the consideration of including interviewees with distinct regional backgrounds. in order to increase the variety at the regional level and to ensure that we have variance on the explanatory variables, three factors were of major importance: the socioeconomic status of the interviewees’ respective regions, and their cultural and party-political incongruence with respect to the centre (the nation state). such aspects had been theoretically conceived as potential explanatory factors. hence, we ensured that both socioeconomically poor and rich regions, regions with and without cultural specificities, and regions with the same and with different governing parties compared to the party-political constellation at the centre were all represented (marks et al. 1996; keating 2008).7 third, the final criterion for the selection of the individuals was their position within the subnational administration. we focussed exclusively on heads of unit with policy responsibilities (cf. bauer 2008) because these are assumed to be the 6 this definition is similar to that of marks et al. (1996). 7 for a full list of the subnational authorities included in the survey, see table 7 in the appendix. the numbers of enclosed regions differ by country as a consequence of the size of the regional administrative body. european journal of government and economics 1(1) 15 “backbone” of the subnational administrations. high-ranking civil servants generally started their careers within their respective subnational administrations. they hold their positions for a long time and usually demonstrate a high degree of identification with the political unit for which they work. they are equally knowledgeable with respect to the technical requirements of a dossier and with the political constellation surrounding it (bauer 2008). out of the initial sample, we randomly selected the interviewees for each region in each member state. however, due to the varying size and the diverging responsibilities of the regions represented, the numbers of interviewees per region ranged from 1 to 13. altogether, the sample consisted of 347 individuals in 60 regions (see table 2). we developed a standardised questionnaire of about 100 (mostly closed) questions. telephone interviews were conducted by specially trained native speakers in 2007 and 2008.8 table 2: sample structure country regions included in sample interviews per region interviews per country response rate germany 13 länder (of 16) 4-9 78 47% poland 12 voiwodships (of 16) 2-9 70 45% hungary 19 megyek (of 19) 2-7 84 41% france 10 régions (of 26) 1-13 66 45% spain 6 autonomous communities (of 17) 5-11 49 53% total 347 source: authors’ compilation. our dependent variable is the regio-crats’ preferences regarding the participation of the subnational level in twelve policy areas in the european multilevel system.9 we thus asked top-level subnational bureaucrats to decide whether or not regional authorities should be involved in policy-making across a range of twelve specific policy areas.10 we constructed an additive index ranging from 0 (no regional competence) to 12 (competence regarding all policies under study). an even more detailed analysis is possible if we distinguish between policies which are regulated in a functionally efficient manner at the regional level and those which are not. however, we need an objective benchmark in order to evaluate whether or not a particular policy is regulated efficiently at the subnational level. such a benchmark does not exist, however, or where researchers have developed something of this nature, it is unsuitable for application to the policy categories we chose for our study.11 we adopt a second-best solution for our problem by following fiscal federalism arguments and then deducing implications for the subnational level. in essence, we assess whether the scope and externalities of policies are decisive parameters for ascertaining whether a certain policy can be regulated efficiently at the regional level or not. we derive a yardstick which is explained in more detail in 8 for a sociological overview of this data set of administrative elites and for further information, see bauer et al. (2010). 9 multilevel governance is a complex concept comprising aspects that concern policy competences and also varying modes of coordination and interaction (benz 2007; benz and zimmer 2008; tömmel 2008). in our project, we limited our analysis exclusively to aspects concerning policy competence allocation. 10 the twelve policy areas are social affairs, asylum and immigration, foreign affairs and defence, health care and consumer protection, border police and frontier defence, culture and education, agriculture, tourism, environmental protection, monetary policy, economic development and structural policy, and research and technology. note that we do not differentiate between administrative and legislative competences. 11 schakel (2009) conducted an expert survey in order to obtain information which authoritative level efficiently regulates in regard to certain policies. however, the categorisation carried out in his study is not applicable in our context as we have different and broader policies under study. european journal of government and economics 1(1) 16 the appendix.12 based on this distinction, we can derive two other variations of the dependent variable. one concerns the preference for competence allocation at the regional level in regard to policies which are efficiently regulated by subnational authorities, while the other concerns those which are not. the operationalisation of the opportunity approach is based on individual-level variables which were also collected by means of the survey.13 on the one hand, the dummy variable “career ambitions” indicates whether top-level bureaucrats want to advance their career within the regional administration where they work.14 the dummy variable “security of employment” (as a motivation for entering the subnational administration) taps into another common aspect of individual utility. the subnational social identity programme is based on factors that are common in subnational mobilisation research. the variable “regional gdp” (gross domestic product) describes the socioeconomic situation of the regions. the dummy variable “stateless nation” indicates cultural differences between the nation state and the subnational entities. the party-political situation is captured by the opposition variables: if the subnational governmental coalition is partly in opposition at the national level, the variable “partly in opposition” is coded 1 (otherwise 0). if no regional government party is represented in the national government, the variable “completely in opposition” takes the value 1.15 finally, the institutional embeddedness of the subnational authorities is operationalised by an indicator taken from the regional authority index developed by hooghe et al. (2010). the variable “institutional depth” measures the extent to which a regional government is autonomous as opposed to deconcentrated.16 finally, we include a control variable allocated at the subnational measurement level. the variable “regional population (log)” captures how populous a subnational entity is. in the literature, it is assumed that efficient provision of public goods is determined not only by the characteristics of the policies themselves, but also by the size of the affected group. we argue that, all else being equal, the larger the regional population, the more efficient (or less inefficient) is the regulation of the policy at the subnational level. because regions differ in regard to their number of inhabitants, we control for this fact. our list of policies for which subnational administrators can articulate their preferences for codetermination is both rather broad and quite general – for instance, “culture and education” instead of more fine-grained policies, such as “primary schools”. this enables us to assume that the handling of our policy categories at the regional level will be more efficient (i.e., more functional) the greater the size of the subnational population. although this factor picks up on the functionality argument, for methodological reasons we cannot integrate a variable representing the actual functional reasoning of our 12 table 9 in the appendix provides an overview of the classification of the policies. we argue that policies that are generally characterised by high externalities should be regulated at a higher level. in other words, we believe that in such cases the participation of regional authorities is less functional. this approach might be criticised as a normative and subjective decision. however, in consideration of the fact that we need an objective benchmark which is applicable to our framework, this, in our view, is the most objective and transparent approach possible. due to problems of endogeneity, we could not justify validating the classification exclusively on the basis of the preferences of the interviewees. 13 for an overview of the independent variables, the operationalisation, the coding and sources, and the expected sign of the coefficient, see table 8 in the appendix. 14 table 2 provides an overview of the operationalisation of the variables. 15 the situation where a regional government coalition is completely represented in the central government is the reference group in the quantitative analysis. the aspect of incongruence respectively congruence between the governing coalition at regional and national levels is a categorical variable. therefore we decided to split the potential situations in three dummy variables. 16 the variable has a theoretical value range of 0 to 4. in our data set, “institutional depth” takes on the values of 3 and 4. european journal of government and economics 1(1) 17 interviewees in the statistical analyses below.17 however, we do address the functionality approach by means of a descriptive analysis in the next section. to test whether the factors characterising the different approaches explain the regiocrats’ preferences for policy allocation at the regional level, we use a multilevel analysis with random-effects. the individuals constitute the first and the regions the second level of the model.18 table 3: operationalisation of the explanatory variables explanatory approach variable operationalisation expected sign career ambitions interviewee wants to advance his career within the subnational administration.19 + individual opportunity security of employment interviewee is motivated by the security of employment within the subnational administration.20 + functionality number of individual competence preferences that match with allocation as functionally efficient for the twelve policies under study. + regional gdp gross domestic product -/ + stateless nation the region is a stateless nation. + partly in opposition regional governmental coalition is partly congruent with the national governmental coalition + completely in opposition regional governmental coalition is in opposition at national level. + subnational social identity institutional depth extent to which a subnational government is autonomous rather than deconcentrated. + control regional population (log) the logarithm of the regional population. + source: authors’ compilation. note: for an explanation as to why we do not operationalise the functionality approach, please see footnote 12. empirical analyses asked whether the subnational level should hold competence regarding twelve policy areas, the average response of regio-crats was in the mid-range, although there was a high standard deviation (see table 4). in general, the regio-crats we surveyed turned out to be surprisingly reluctant to see regions participating in the multilevel governance system. a comparison of the national mean values for competence allocation at the regional level reveals variation in the preferences for subnational participation within our sample. first, we observe differences across countries. hungarian top-level subnational bureaucrats do not see a need for 17 in our descriptive analysis, we assess with respect to how many policies the individuals’ preferences are in line with functionally efficient policy allocation. however, if we included such an independent variable in the regression analysis, we would explain the dependent variable by means of a part of a modified dependent variable. 18 a potential objection to this research strategy might be the choice of the statistical model. we do not use dummy variables for the countries in order to control for country-specific factors, rather we emphasise the individual and the regional levels. given that we integrate several (potential) explanatory factors measured at the subnational level, we risk falling into the trap of multicollinearity. we also use an indicator capturing the institutional setting in the nation states. 19 the variable is based on a survey question: ”which professional position would you like to have reached in 5-10 years? that is, other or same position in the regional administration in a national ministry, in the eu administration?” whenever the interviewee indicated that he sought to reach a higher position in the administration, the variable “career ambitions” is coded “1” otherwise “0”. 20 the information is based on a survey question: “why do you have chosen to work in the administration of a region? please tell me the most important reason.” the answer options had been secure job, proximity to residence, good salary, good career prospects, interesting working area, would like to contribute to the development of the region, i was offered this job, or decentralization or fusion. for interviewees indicated the option “secure job” the variable “security of employment” is codes “1” otherwise “0”. european journal of government and economics 1(1) 18 extensive policy participation. on average, they desire competence regarding only 1.6 of the twelve policy areas. although the desire for policy competence is stronger in the other countries, german respondents are still surprisingly modest in their preferences, desiring subnational competence for only 3.4 policies. this picture is noteworthy because the preferences actually lag behind the status quo for german länder competences. the national mean values for the polish (4.8) and french (4.3) respondents point to a mid-degree claim for codetermination rights compared to the other country scores. in contrast, spanish top-level subnational bureaucrats favour regional participation in about eight policy areas, which reflects a desire for strong policy participation within the european multilevel system.21 second, the standard deviations for the countries listed in table 4 indicate considerable within-country variation as well. in other words, we do find interesting variation within the five countries. what explains such differences among regio-crats’ preferences for policy codetermination rights? table 4: national mean values for desired policy competence22 country mean sd n germany 3.4 2.3 76 poland 4.8 2.5 65 hungary 1.6 1.8 83 france 4.3 2.0 65 spain 8.0 1.3 49 total 4.1 2.8 338 source: authors’ compilation. can the differential desires for policy codetermination be explained by functionality? looking at the preference patterns of the top-level subnational bureaucrats in regard to the twelve policy areas, we are able to assess whether the administrators form their preferences on the basis of a functionality rationale or not. by comparing the preference patterns for regional policy competences to what we deem would be the objective functional policy allocation, we are able to assess whether or not there is a correlation. if the interviewees favoured regional policy participation regardless of whether or not such codetermination is functional (according to our assessment), we have at least an indication that some other than a functional-efficiency explanation must be at work. having carried out the categorisation, we can count for how many policies the answers of our regio-crats are in line with the presumably most efficient allocation. our result is that the answers of our respondents are substantially in line with functional criteria of competence allocation. table 5 gives an overview of the share of policies for which the regio-crats’ preferences for regional participation (or nonparticipation) are in accordance with the allocation on the basis of functional efficiency. for example, the preferences of about 59 percent of the german interviewees are in line with the functionality rationale for six to eight policies. in france, about 57 percent of our respondents show preferences that correspond with the efficient participation or non-participation of regional authorities for more than eight policies. all in all, in every country we studied, the preferences regarding (non-)participation of subnational authorities of the majority of interviewees correspond with the functional-efficiency expectation. only a few respondents demonstrate codetermination preferences that clearly conflict with functionality. 21 for a detailed discussion of the preferences regarding policy competences, see pitschel (2012). 22 the table shows the mean value of the desired number of policies for regional policy competences. shown are the mean values (mean) of the preferences, the standard deviation (sd), and the number of respondents (n) for the sample and per country. european journal of government and economics 1(1) 19 table 5: correspondence of regio-crats’ preferences with functional needs23 for how many of the twelve policies do subnational administrators’ responses correspond with efficient competence allocation? germany poland hungary france spain total for less than 6 policies 7.9 9.2 41.0 4.6 12.2 16.3 for 6 to 8 policies 59.2 41.5 50.6 38.5 83.7 53.3 for more than 8 policies 32.9 49.2 8.4 56.9 4.1 30.4 source: authors’ compilation. summing up, we observe that the national mean preferences for regional participation in the twelve policies differ across the five countries, with the spanish having the highest and the hungarian the lowest scores. at the same time, however, we observe variation within the countries under study. the individuals’ preferences concerning the desired extent of regional policy competence vary considerably within the countries. individual attitudes deviate most around the respective national average in poland and germany. to find a reason for this result, we turn to the statistical analysis. quantitative analysis our quantitative analysis assesses how the factors underlying the explanatory approaches are related to the dependent variable, namely the desire for regional codetermination in different policy areas. we basically run three regressions, which differ in the conceptualisation of the dependent variable. the dependent variable of the first model refers to all twelve policies under study. in the second and third models, the dependent variables relate to policies for which (according to our yardstick) regions constitute the functionally appropriate – or non-appropriate – governmental level of execution for the policies in our sample. this procedure enables us to detect stable relationships between the independent variables and the subnational administrators’ preferences regarding regional policy codetermination. moreover, we gain knowledge about the influence of policyinherent logics on the process of preference formation. the results of the models are presented in table 6. first, the subnational administrators’ preferences regarding regional competence are analysed with respect to all twelve policies (model 1 in table 6). concerning the individual-utility variables, only the indicator for safety thinking (“security of employment”) is significant. the positive sign is in line with the theoretically expected relationship: administrators who entered subnational administrations motivated by security of employment are in favour of more subnational competences. as regards the social identity approach, the socioeconomic variable reflecting the regions’ economic performance (measured in gdp) shows a negative sign – implying that the desire for subnational competence decreases with regional wealth. in other words, our data show that regio-crats from socioeconomically weaker regions are more in favour of regional policy participation than their colleagues from socioeconomically stronger entities. cultural distinctiveness is also influential as a predictor of regional desire for emancipation from the nation state. the positive and significant coefficient of the 23 percentage of respondents per country for which the preferences for regional policy competences are in line with the functionality principle. whether participation or non-participation of subnational authorities in policies is functionally efficient was evaluated on the basis of objective criteria (see table 9 in the appendix). european journal of government and economics 1(1) 20 variable “stateless nation” indicates that subnational administrators of such regions in our sample (alsace and brittany in france, catalonia and país vasco in spain) want to have more subnational policy competences. we also find significant coefficients concerning the indicators of the party-political constellation. on the one hand, the subnational context in which the regional government is “partly in opposition” to the national government or government coalition seems to be negatively related to the desire of the top-level bureaucrats for regional policy competences. in contrast, the individuals from subnational authorities where the regional government is “completely in opposition” to the national government (or governmental coalition) are obviously more in favour of policy participation by their authority. we supposed that the institutional setting would influence the preferences as well. the significant and positive coefficient of the variable “institutional depth” is in line with the theoretical expectation that top-level bureaucrats from institutionally strong regions would desire more codetermination rights over a greater range of policy areas. what also shows a significant and positive coefficient, however, is the variable measuring regional population. this means that the larger the regional population, the more competences are desired by regio-crats. the model including all twelve policies shows a considerably higher degree of variance between the groups (regions) than within the regions. overall, model 1 explains about 24 percent of the variance. table 6: regression results for all 12 policies explanatory approach variables model 1 all 12 policies model 2 7 functional policies model 3 5 non-functional policies career ambitions 0.043 0.041 0.002 (0.279) (0.223) (0.090) security of employment 0.969* 0.381 0.386** opportunity approach (0.518) (0.413) (0.168) regional gdp -0.125** -0.069* -0.058** (0.053) (0.037) (0.024) stateless nation 2.728*** 1.548*** 1.229*** (0.833) (0.587) (0.373) partly in opposition -1.427*** -1.275*** -0.112 (0.449) (0.313) (0.204) completely in opposition 1.410* 0.807 0.646* (0.773) (0.537) (0.352) institutional depth 1.521** 0.298 1.244*** subnational social identity (0.678) (0.473) (0.308) regional population (log) 0.000** 0.000*** 0.000 control (0.000) (0.000) (0.000) constant 2.086* 3.568*** -1.521*** (1.255) (0.877) (0.567) observations 297 297 296 number of regions 57 57 57 r-squared within 0.000 0.000 0.003 betwe en 0.446 0.446 0.362 overall 0.243 0.182 0.315 source: authors’ compilation. note: standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1. second, we analysed only those policies that can be deemed efficiently regulated at the subnational level (model 2 in table 6).24 whereas the individual variables of the opportunity model seem not to be influential at all, the social identity variables show similar patterns to the first model. the “regional gdp” variable is once again negative and significant. similarly, the “stateless nation” variable is significant and 24 these policies are social affairs, health care and consumer protection, culture and education, tourism, environmental protection, research and technology, and economic development and structural policy. the reasons for this classification are explained in table 9 in the appendix. european journal of government and economics 1(1) 21 has a positive sign, as in the preceding model. with respect to the party-political situation, only the variable “partly in opposition” maintains its negative and significant coefficient in this model. in contrast, the variable “completely in opposition” is insignificant in model 2. the same is true for the regional authority variable “institutional depth”. neither “fiscal autonomy” nor “regional representation” show a significant coefficient. finally, the regional population size repeatedly demonstrates a positive and significant coefficient. comparing the variance explained by this model to the first model, we do not find any enhancement, neither in the explanation of the individual-level variance nor in the explanation of the between variance. however, the overall explained variance of the model on those policies that are efficiently regulated at the regional level is lower than in the model including all twelve policies (model 1). finally, we assess the explanatory programmes with respect to those policies in our sample that are supposedly not efficiently regulated at the regional level (see model 3 in table 6).25 this variant of the dependent variable might be an interesting case for both the opportunity and the subnational identity approaches. do such variables influence the preference for regional policy competence although such participation is not efficient? in this third model, the individual variable of the opportunity explanatory program “career ambition” again shows no significant and positive coefficient. in contrast, the variable “security of employment” regains its theoretically expected positive sign. the subnational identity approach variables perform comparably to the second model. the “regional gdp” variable shows a negative and significant coefficient. the better the socioeconomic situation concerning gdp, the less policy competences are desired by regio-crats. the variable indicating that a region represents a stateless nation is positive and significant, as expected. the same holds for the political variable “completely in opposition”, which reflects the situation where the regional governmental coalition is incongruent with the party-political constellation in national government. again, we see our expectation of a positive relationship between this variable and the desire for more regional competences confirmed. the variable “partly in opposition”, however, is not significant. concerning the aspect of regional autonomy, we find that subnational administrators from regions which are institutionally well endowed favour more policy competences. the variance between the units explained by the regression model is lower than in the other two models. model 3 explains about 36 percent of the variance between the regions. compared to model 1 and 2 this proportion is lower. in contrast, with regard to the overall variance, we find a high proportion is explained by model 3 (about 32 percent). discussing the regression results in the context of non-functional policies, we believe this is evidence that subnational administrators’ preference formation is based on an opportunity rationale. personal interest in secure employment in the regional administration is influential not only in the first model, comprising all twelve policies, but also in third model, which concentrates on those policies that are not efficiently regulated at the regional level. as regards the emancipatory ambitions of regional authorities, which are supposed to be the driving force in the social identity approach, we find some evidence to confirm the theoretical reasoning. whereas the picture is clear for those administrators from culturally distinct regions who favour more policy competences, the influence of the political situation is less evident. in contrast, we find unambiguous results for the influence of the 25 these policies are asylum and immigration, foreign affairs and defence, border police and frontier defence, agriculture, and monetary policy. the reasons for this classification are explained in table 9 in the appendix. european journal of government and economics 1(1) 22 institutional setting (“institutional depth”) on the extent of regional policy participation. summing up, the results of our quantitative analysis are consistent. the significant variables do not change their signs in the different models and we find the theoretical expectations generally confirmed. however, some points have to be reinvestigated in more detail. this applies, in particular, to the influence of the party-political constellation of the regions compared to the situation at the central state level. on the one hand, we find the expected relationship for the case when the regional government is not congruent with the party-political constellation of the central government. on the other hand, the opposite is true for the case where the regional government is partly in opposition at the national level. this contradicts the theoretical expectation and requires further investigation. conclusion two sets of conclusions can be drawn from this study. the first concerns the insights our analysis is able to generate in view of the question as to how to explain regio-crats’ preferences regarding policy participation in the multilevel system. notwithstanding bold statements in the relevant literature, the desire for subnational policy codetermination is astonishingly low throughout our sample. regio-crats cannot be seen as “competence conquerors” that fuel state transformation by demanding ever greater policy involvement. quite the contrary, regio-crats appear in this respect to be rather conservative. there is little reason to fear (or hope, depending on the perspective) that regions will shake up the existing competence distribution; the suspicion that regions will actively ask for ever greater policy involvement cannot be substantiated by our data. instead, the preferences seem to a large extent based on a rationale of functionality. the big picture is that regio-crats’ policy participation demands are in harmony with what can be conceived as objectively efficient vertical competence allocation. on a smaller scale, however, the statistical analyses show that besides individual utility aspects, variables that are related to regional emancipatory ambitions also have a positive influence on the desire for more policy competence. cultural and political distinctiveness (in comparison to the situation at the central state level) increase the demand for regional policy involvement. in other words, a large-scale and uniform transformation in the direction of a europe of the regions is unlikely. pressure for greater regional policy involvement is differential and its direction and magnitude depend on particular regional situations. on the basis of our analysis, we do not expect demands for regional policy involvement to be systemic and systematic, but disorderly, disparate and diverse. regional policy participation demands in the multilevel system are pragmatic, patch-worked and incremental – and more conservative than transformative. references alesina, alberto and enrico spolaore (2003) the size of nations, cambridge, ma: mit press. alesina, alberto, ignazio angeloni and ludger schuknecht (2001) 'what does the european union do?' national bureau of economic research working paper no. 8647, washington dc. bauer, michael w. (2008) 'diffuse anxieties, deprived entrepreneurs. commission reform and middle management', journal of european public policy, 15(5): 691707. bauer, michael w, diana pitschel and philipp studinger (2010) 'governance preferences of subnational administrative elites in the european union. an european journal of government and economics 1(1) 23 empirical analysis', lehrstuhl für politik und verwaltung working paper no. 4, 2010, hu berlin. benz, arthur. (2007) 'politischer wettbewerb', in a. benz, s. lütz, u. schimank and g. simonis (eds), handbuch governance. theoretische grundlagen und empirische anwendungsfelder. wiesbaden: vs verlag, 5467. benz, arthur. and christina zimmer (2008) 'the eu’s competences: the “vertical” perspective on the multilevel system', living reviews in european governance 3(3). url (consulted 30 nov. 2008) : http://www.livingreviews.org/lreg-2008-3. bird, richard, bernard dafflon, claude jeanrenaud and gebhard kirchgässner (2002) 'assignment of responsibilities and fiscal federalism', in raoul blindenbacher and arnold koller (eds) conference reader of the international conference on federalism 2002: federalism in a changing world learning from each other, quebec: mcgill-queen’s university press. björklund, fredrika. (2006) 'the east european ethnic nation – myth or reality?` european journal of political research, 45(1): 93-121. bookman, milika z. (1992) the economics of secession. new york: st. martin’s press. brancati, dawn. (2006) 'decentralization: fueling the fire or dampening the flames of ethnic conflict and secessionism?' international organization, 60(3): 651-85. connor, walker (1994) ethnonationalism. princeton, nj: princeton university press. de vries, catherine e. and kees van kersbergen (2007), 'interests, identity and political allegiance in the european union', acta politica, 42(2-3), 307-28. diez medrano, juan (2003) framing europe: attitudes to european integration in germany, spain and the united kingdom. princeton: princeton university press. dunleavy, patrick (1985) 'bureaucrats, budgets and the growth of state: reconstructing an instrumental model', british journal of political science, 15(3): 299-328. egeberg, morten (2004) 'an organisational approach to european integration: outline of a complementary perspective', european journal of political research, 43(2): 199-219. esman, milton j. (1977) 'perspectives on ethnic conflict in industrialized societies', in milton j. esman (ed.) ethnic conflict in the western world, pp. 371-90. ithaca: cornell university press:. fearon, james d. and pieter van houten (2002). 'the politicization of cultural and economic difference. a return to the theory of regional autonomy movements'. paper presented at the fifth meeting of the laboratory in comparative ethnic processes (licep), stanford university, 10-11 may. gourevitch, peter a. (1979) 'the re-emergence of "peripheral nationalisms": some comparative speculations on the spatial distribution of political leadership and economic growth', comparative studies in society and history, 21(3): 303-22. harvie, christopher (1994) the rise of regional europe. london: routledge. hechter, michael. (1975) internal colonialism: the celtic fringe in british national development, 1536-1966. berkeley, ca: university of california press. hoeller, peter, marie-odile louppe and patrice vergriete (1996) 'fiscal relations within the european union', oecd economic department working paper no. 163, paris: oecd publishing. european journal of government and economics 1(1) 24 hooghe, liesbet. (1995) 'subnational mobilisation in the european union', west european politics, 18(3): 175-98. hooghe, liesbet. (2001) the european commission and the integration of europe. images of governance, cambridge: cambridge university press. hooghe, liesbet (2003) 'europe divided? elites vs. public opinion on european integration', european union politics, 4(3): 281-304. hooghe, liesbet and gary marks (2005) 'calculation, community and cues: public opinion on european integration', european union politics, 12(1): 419-43. hooghe, liesbet. and gary marks (2009) 'does efficiency shape the territorial structure of government?' annual review of political science, 12(1): 225-41. hooghe, liesbet, gary marks and arjan h. schakel (2010) the rise of regional authority. a comparative study of 42 democracies, abingdon: routledge. jeffery, charlie (1996) 'conclusions: sub-national authorities and ‘european domestic policy', regional and federal studies, 6(2): 204-19. jeffery, charlie (2000) 'sub-national mobilization and european integration: does it make any difference?', journal of common market studies, 38(1): 1-23. jeffery, charlie and daniel wincott (2010) 'the challenge of territorial politics: beyond methodological nationalism', in colin hay (ed.) new directions in political science: responding to the challenges of an interdependent world, basingstoke: palgrave macmillan, 167-88. kato, junko (1996) 'the challenge of territorial politics: beyond methodological nationalism’, british journal of political science, 26(4): 553-82. keating, michael (1998) the new regionalism in western europe. territorial restructuring and political change. cheltenham: edward elgar. keating, michael (2008) 'a quarter century of the europe of the regions', regional and federal studies, 18(5): 629-35. levi, margaret (1997) 'a model, a method and a map: rational choice in comparative and historical analysis`, in mark i. lichbach and allan s. zuckerman (eds), comparative politics: rationality, culture and structure, cambridge: cambridge university press, 19-41. loveless, matthew and robert rohrschneider, r. (2008) 'public perceptions of the eu as a system of governance', living reviews in european governance, 3 (1). online. available from: http://www.livingreviews.org/lreg-2008-1, (accessed 19 december 2008). marks, gary, françois nielsen, leonard ray and jane e. salk (1996) 'competences cracks and conflicts: regional mobilization in the european union', comparative political studies, 19(2): 164-92. mclaren, lauren m. (2002) 'public support for the european union: cost/benefit analysis or perceived cultural threat?', journal of politics, 64(2): 551-66. niskanen, william .a. (1971) bureaucracy and representative government. new york: aldine-atherton. oates, wallace e. (1999) 'an essay on fiscal federalism', journal of economic literature, 37(3): 1120-49. oberhofer, julia, julia stehlin and roland sturm (2011) 'citizenship im unitarischen bundesstaat', politische vierteljahresschrift, 52(2): 163-194 olsen, johan p. (2009) 'change and continuity: an institutional approach to institutions of democratic government', european political science review, 1(1): 332. european journal of government and economics 1(1) 25 pitschel, diana (2012) europa mit den regionen? eine empirische analyse der kompetenzbeteiligung von regionen im europäischen mehrebenensystem. badenbaden: nomos. schakel, arjan h. (2009) 'explaining policy allocation across government tiers by identity and functionality', acta politica, 44(4): 385-409. schakel, arjan h. (2010) 'explaining regional and local government: an empirical test of the decentralization theorem', governance: an international journal of policy, administration and institutions, 23(2): 331–55. scully, roger and david farrell (2003) 'meps as representatives: individual and institutional roles', journal of common market studies, 41(2): 269-288. searing, donald d. (1991) 'roles, rules and rationality in the new institutionalism', american political science review, 85(4): 1239-60. searing, donald d. (1994) westminster’s world. understanding political roles. cambridge, ma: harvard university press. seeman, melvin (1993) 'a historical perspective on attitude research', in dagmar krebs and peter schmidt (eds), new directions in attitude measurement, berlin: walter de gruyter, 3-20. shah, anwar (2002) 'fiscal decentralization in transition economies and developing countries – progress, problems and the promise', paper presented at the international conference “federalism in a changing world – learning from each other”, st. gallen, 27 30 august. shah, anwar (2007) 'introduction: principles of fiscal federalism', in idem. (ed.) the practice of fiscal federalism: comparative perspectives. montreal: mcgillqueen’s university press, 3-43. sinn, hans-werner. (2003) the new systems competition. oxford: basil blackwell. smekal, christian (2001) 'zentralisation und dezentralisation öffentlicher aufgabenerfüllung aus finanzwissenschaftlicher sicht', in peter pernthaler and peter bußjäger (eds) ökonomische aspekte des föderalismus, schriftenreihe des instituts für föderalismus, pp. 69-76. vienna: braumüller. tatham, michael (2008) 'going solo: direct regional representation in the european union', regional and federal studies, 18(5): 493-515. ter-minassian, teresa (1997) 'intergovernmental fiscal relations in a macroeconomic perspective: an overview', in teresa ter-minassian (ed.) fiscal federalism in theory and practice, pp. 3-24. washington: imf. thielemann, e.r. (2004) 'dividing competences: germany’s vision(s) for europe’s federal future', comparative european politics, 2(3): 358-74. tömmel, ingeborg (2008) 'governance und policy-making im mehrebenensystem der eu', in ingeborg tömmel (ed.) die europäische union governance und policy-making, pvs-sonderheft, pp. 13-35. wiesbaden: vs verlag für sozialwissenschaften. van esch, femke (2006) 'the europeanization of central decision makers’ preferences concerning europe. a perpetual motion?', in ronald holzhacker and markus haverland (eds) european research reloaded: cooperation and integration among europeanized states, pp. 119-49. berlin: springer. van houten, pieter. (2003) 'globalization and demands for regional autonomy in europe', in miles kahler and david a. lake (eds) governance in a global economy. political authority in transition, pp. 110-35. princeton, nj: princeton university press. european journal of government and economics 1(1) 26 van houten, pieter (2009) 'multi-level relations in political parties. a delegation approach', party politics, 15(2): 137-56. european journal of government and economics 1(1) 27 appendix table 7: overview of regions represented in the sample country region number of interviewees country region number of interviewees dolnośląskie 7 alpes côte 7 kujawsko-pomorskie 9 alsace 4 lubelskie 7 aquitaine 1 lubuskie 6 bretagne 6 małopolskie 5 centre 13 mazowieckie 4 franche-comté 4 opolskie 8 île-de-france 8 podkarpackie 4 midi-pyrénées 10 pomorskie 6 poitou-charentes 6 śląskie 2 france rhône-alpes 7 świętokrzyskie 5 poland wielkopolskie 7 bács-kiskun 5 badenwürttemberg 8 baranya 3 bayern 8 békés 3 berlin 4 borsod-abaújzemplén 7 brandenburg 5 csongrád 4 hessen 5 fejér 4 niedersachsen 5 győr-moson-sopron 6 nordrheinwestfalen 9 hajdú-bihar 4 rheinland-pfalz 7 heves 7 saarland 5 jász-nagyunszolnok 3 sachsen 6 komáromesztergom 6 sachsen-anhalt 7 nógrád 3 schleswig-holstein 4 pest 7 germany thüringen 5 somogy 6 szabolcs-szatmárbereg 3 andalucía 11 tolna 3 baleares 9 vas 5 cataluña 6 veszprém 3 galicia 8 hungary zala 2 madrid 10 spain país vasco 5 table 8: independent variables variable level of measurem ent operationalisation coding expected sign career ambitions interviewee wants to advance his career within the subnational administration. 1 = yes, 0 = no + security of employment individual interviewee is motivated by the security of employment within the subnational administration. 1 = yes, 0 = no + regional gdp subnation al gross domestic product gross domestic product (purchasingpower parity) per -/ + european journal of government and economics 1(1) 28 capita in 1000 euro stateless nation the region is a stateless nation. 1 = yes, 0 = no + partly in opposition regional governmental coalition is partly congruent with the national governmental coalition. 1 = yes, 0 = no + completely in opposition regional governmental coalition is in opposition at the national level. 1 = yes, 0 = no + institutional depth extent to which a subnational government is autonomous rather than deconcentrated. 0 = no functioning general-purpose administration at the regional level; 1= deconcentrated, general-purpose administration; 2 = nondeconcentrated, general–purpose administration subject to central government veto; 3 = nondeconcentrated, general–purpose administration not subject to central government veto. + regional population (log) regional population as percentage of the national population. log (percentage) + table 9: classification of policies policy is a regional participation in this policy functional? rationale social affairs yes the standard of regulation of social affairs is already high within the eu member states, so that there is no call for central regulation (alesina et al., 2001). furthermore, some studies argue that differing levels of regulation might lead to a competitive advantage for a region in some sociopolitical areas (and that heterogeneous preferences regarding regional regulation can be traced back to this) (cf. hoeller et al. 1996; smekal 2001). at any rate, the involvement of the regions is consistent with the functional principles of competence allocation. asylum and immigration no the area of asylum and immigration is characterised by high external effects. to internalise these and to avoid free-rider effects, regulation ought to be conducted as centrally as possible (i.e., at national or supranational level) (alesina, et al., 2001; shah, 2007). foreign affairs and defence no this policy area is characterised by high external effects. moreover, regulation preferences are relatively homogeneous (at least within the eu member states). by centralising (regulation at eu level), external effects can be more fairly distributed in the community, free-rider effects can be avoided and national obstacles overcome (ter-minassian 1997). other authors claim that foreign and defence policies should be regulated at the national level because resulting costs and benefits also manifest at the national level (shah 2002). according to this view, regional participation in foreign affairs and defence is not functional. health and consumer protection yes health care and consumer protection need to be adjusted to the citizens’ requirements on site, while, at the same time, standards should be uniform and binding eu wide. the most efficient approach is for health care and consumer protection to be regulated under involvement of all levels of the hierarchy (eu, nation state and subnational entities) (alesina et al. 2001; smekal 2001; hooghe and marks 2009). thus, the regions should be involved in shaping this policy. border police no this policy also entails high externalities, which – from a functional european journal of government and economics 1(1) 29 and frontier defence perspective – need to be re-allocated to the community (hoeller et al. 1996). from a european point of view, the observance of uniform standards and the export provisions of the schengen agreement must be kept in mind. culture and education yes the policy area of culture and education has a strong identitydefining component. thus, heterogeneous regulation preferences potentially outweigh external effects. resulting benefits of centralisation are estimated as limited. hence, and in line with functional considerations, a decentralised provision of services is to be preferred here (smekal 2001; schakel 2010). agriculture no in the eu context and under efficiency considerations, regulation at the community level is not justifiable. however, excessive decentralisation would counteract the single market (alesina, et al. 2001; hoeller et al. 1996). hence, from a functional perspective, this policy can be best organised at the national level. tourism yes regional preferences and requirements vary within this policy. in addition, for many regions, tourism holds an identity-defining component. accordingly, the regions should be involved in shaping this policy (hooghe and marks 2009). environment yes due to the strong external effects of environmental problems and their transnational scope, a stipulation of regulation standards on the supranational level would fit best. however, specific environmental problems are regionally concentrated, so that an involvement of national and subnational entities in environmental politics is assumed to be functionally adequate (alesina et al. 2001; sinn 2003). monetary policy no monetary policy (at least in the eu) is among those policies that, for reasons of economic efficiency, should be regulated centrally. even if states are not members of the european monetary union (euro), a central regulation seems to be appropriate to guarantee uniform standards of quality (alesina et al. 2001; smekal 2001). economic development and structural policy yes in the area of economic development and structural policy, it is assumed to be economically wise to regulate basic aspects and conditions at the central level (eu) to avoid distortions of competition. however, subnational actors should be involved during policy implementation so as to ensure an efficient realisation on site which corresponds to regional requirements (alesina et al. 2001; hoeller et al. 1996). research and technology yes differing regional standards of regulation in the area of research and technology boost regional competition and are seen as regional economic factors. accordingly, a decentralised regulation is to be preferred from a functional perspective. however, it is also argued that some research areas, e.g., nuclear energy, require a central political organisation (alesina et al. 2001). microsoft word ejge_03_02_014.doc european journal of government and economics volume 3, number 2 (december 2014) issn: 2254-7088 104 institutional quality and private sector participation: theory and empirical findings rahel schomaker, cologne business school and german research institute for public administration speyer, germany abstract through several conduits, sound institutional quality is pivotal for economic development, as there is evidence that stable democratic institutions, rule of law and sound governance structures in the administration are highly conducive to promote growth. therefore, a high institutional quality is not only the end point, but also the starting point of a more sustainable development. in this paper we provide some theoretical considerations as well as empirical evidence, based on several regression analyses, that the quality of institutions in a wider sense, and governance, which includes not only the level of “politics” itself, but also the administrative level, is relevant not only for the macro-level of development (i.e. the increase of the national welfare and foreign investment), but also on a micro-level: a stronger participation of private enterprises in public service provision and the introduction of public-private partnerships depends to a high degree on the institutional quality. this is even more relevant as the improvement of public services and of core infrastructures can be seen as crucial multipliers for future growth. jel classification a12; a30; p27. keywords institutional quality; good governance; institutional economics; public-private partnerships; public goods. schomaker ● institutional quality and private sector participation 105 introduction in several ways, institutional quality is pivotal for economic development. there is much evidence that democratic institutions, the absence of corruption, rule of law, and sound governance structures in the country’s administration are conducive to promoting growth in terms of gross domestic product (gdp), or to attracting foreign direct investment (fdi), to mention just some of its most relevant determinants (see for the discussion of the role of institutions e.g. acemoglu and robinson, 2012). democratic governance and accountability are not the end point after a country has undergone economic and social development, but rather the start point of a more sustainable development. in our paper, we provide some theoretical insights as well as empirical evidence that the quality of institutions, which includes not only the level of “political institutions” itself, but also the administrative level, is relevant not only for achieving the standard macro-level development objectives (e.g. the increase of gdp and fdi). also on the micro-level – the provision of infrastructure and other public services – institutions matter. in other words, a stronger participation of private enterprises in service provision and the introduction of public-private partnerships depend to a high degree on the institutional quality of a country. this is even more relevant as the improvement of public services and of core infrastructures can be seen as a multiplier for further growth, as theoretical considerations and a substantial number of empirical studies have clearly demonstrated. (efficient) investment in public services and infrastructure provides positive growth impulses for the overall economy by fostering entrepreneurship, providing necessary infrastructure for businesses or sending signals to foreign investors (for an overview e.g. hartwig, 2005). therefore, as conventional wisdom has it, given that private firms may provide much-needed financial resources as well as superior management know-how and technical expertise for infrastructure provision and public services, the stronger inclusion of private enterprises may be one very promising development “tool” (e.g. grimsey and lewis, 2005; ziekow, 2003). the important role of private enterprises can play in the provision of infrastructure and services is in particular given for all countries that face severe budget constraints in the public sector. while this applies generally to less developed countries, industrialised countries in europe and the us also face increasing public budget restrictions, triggered not least by the financial crisis of 2008/2009. therefore, as the general effect of institutions on the overall economic wellbeing of a country or economic growth rates has been subject to scientific research before (e.g. acemoglu and robinson, 2012; obinger, 2001; wagener, 2004), our contribution to the existing literature is new in two dimensions. by analysing the “micro-level” of development, focusing on private sector participation, for the first time we provide empirical evidence in how far institutional quality influences the involvement of private firms in the provision of public services. additionally, by tying the link between institutional quality and the nature of private sector participation, we contribute to the ongoing debate in particular on transaction costs related to private sector participation and public-private partnerships. our paper is structured as follows: in the next section, we first apply the theory of “new institutional economics” to explain why sound governance, especially the quality of bureaucracy, and not least of regulatory oversight, and of anti-corruption policies, is a crucial factor in fostering a potentially welfare-increasing stronger inclusion of the private sector. transaction costs are of specific relevance in this context, as it is reasonable to assume higher transaction costs will, ceteris paribus, always increase the price of cooperations and partnerships (assuming that transaction costs are an integral part of the full costs of contract conclusion and contract enforcement) (hart, 2005). based on this, we discuss the economics of private sector involvement more detailed, with a specific focus again on the role of transaction costs. european journal of government and economics 3(2) 106 after setting up the theoretical framework, we will provide some empirical evidence for the role sound institutions play in the context of the establishment of publicprivate partnerships in infrastructure provision, focusing on a view to the world’s less developed countries. we will finish our analysis with a brief conclusion of our main findings and some policy recommendations. theoretical background: “new institutional economics” and the “economics of private sector participation” key insights from the “new institutional economics” the main hypothesis of institutional economics is that institutions strongly influence human behaviour and therefore also have a strong relevance for the growth and development of countries (or lack thereof) (acemoglu and robinson, 2012; obinger, 2001). while an exclusive and universally accepted definition of institutions is still missing, a rather broad consensus has emerged in the literature on what constitutes institutions and what their principal functions and effects are (for a comprehensive overview see hodgson, 1998 and williamson, 2000). following north (1991 and 1992), institutions are interpreted as ‘humanly devised constraints that structure political, economic, and social interactions’ (north, 1991). they constitute the framework of a society, which to a high degree determines the individuals’ activities by providing crucial information and therefore reducing uncertainty (voigt, 2002). this framework does not only comprise the so-called material institutions, but also all mechanisms that are able to enforce these very institutions (erlei et al., 2007). in particular, institutions that determine the rules of the game and shape governance structures are the focus of nie and the quantitative approaches that are applied in this context. institutions at these levels are often seen as vital determinants for economic wealth/welfare (or the lack thereof). at the same time it is relatively easy to identify – in a normative, theory-based approach – how these institutions should be structured to reach the overall goal of “economic growth” and the intermediate goal of “good governance” as a necessary precondition for achieving economic wealth/welfare. the role of institutions is of specific importance in a setting of incomplete information and uncertainty, as is inevitably – due to prohibitive costs of gathering and processing information – the case in every real-world economic system. incomplete information or asymmetric information between actors, as well as bounded rationality and the – by nature – limited mental capacities of human beings lead to additional costs, the so-called “transaction costs” (e.g. coase, 1937). transaction costs relate to each single market and non-market transaction, may it be with regard to goods or services, or, in what constitutes a broader application of this concept, to the definition, transfer and enforcement of property rights (richter and furubotn, 1999). these costs arise due to the fact that in order to perform any of these transactions information on potential counterparts has to be obtained, checked for correctness and assessed in terms of counterparty and other forms of economic risks, and that, eventually, contracts have to be negotiated, monitored and, if necessary, enforced or renegotiated. transaction costs therefore occur both ex ante, i.e. before a contract is created, and ex post, after the contractual arrangements have been finalised. effective and generally respected institutions can help to reduce – though they will never completely eliminate – these transaction costs in several ways by providing some crucial information and some kind of security in an overall framework which is unavoidably characterised by uncertainty about future developments in an essentially unpredictable, dynamic, evolutionary world. the greater degree of schomaker ● institutional quality and private sector participation 107 certainty that the contractual partner will heed the contract provisions will, in turn, enhance the incentives (for individuals and enterprises alike) to engage in a contract in the first place because of the much reduced risks of falling into “hidden traps”. in other words, the knowledge itself that one is in the legal position to take a case to court or mediation (e.g. in case of the counterparty’s real or alleged breach of contract or for unforeseen renegotiations within the contract duration) will tend to discourage each contract party from acting opportunistically or in outright breach of contract, as this will give rise to substantial costs (including loss of reputational capital etc.). a strong legal system which provides for effective sanctions will therefore be substantially conducive to contract compliance. it will, in turn, lower transaction costs for contractual arrangements in the first place, resulting in more transactions occurring, and, ceteris paribus, in the welfare-enhancing productivity gains which can be harnessed through from the widening and deepening of the division of labour. this is primarily due to the beneficial effect of lower individual costs of control which stem from the existence of efficient institutions which at least partly even out the information asymmetry between (potential and actual) contractual partners. to sum up, in this context, a crucial function of (good) institutions is their signaling to act as stand-ins for the contract partner’s trustworthiness and reliability. for efficiently and effectively fulfilling these functions, good institutions can be regarded as a conditio sine qua non for economic development and a sound longrun economic performance of a country (e.g. recently acemoglu and robinson, 2012; see also apolte, 2004; north, 1991; sachs and warner, 1997; wagener, 2004). as klein and luu (2003) state, “[a] key to understanding economic development, then, is institutional development”. consequently, within the last few decades, a whole new branch of literature endeavoured not only to provide for the deeper theoretical understanding of the relevance of institutions, but also to seek robust empirical evidence on the impact of the institutional framework for development. most of these empirical studies measure the impact of institutional quality on the economic wealth/welfare of a country, commonly measured in terms of gross national product (gnp) or gross domestic product (gdp) figures (per head in us$ and/or based on purchasing power parities), or the rate of change of these indicators, or use different proxies for economic wealth/welfare. “institutional quality” in these studies is defined in different ways, and not exclusively as the absence of corruption, good regulatory quality, democratic participation and high accountability of the government, political stability, the secure property rights, and/or efficient public administration. to recapitulate here the main results of an meta-assessment of several seminal empirical studies, independent of their respective definition (and the data sets) used, it can be safely concluded that there is substantial evidence for a significant positive correlation between (high) institutional quality and (positive) rates of economic growth and/or wealth. as, for instance, obinger (2001) is able to show, while the regime-type itself (democracy vs. dictatorship) does not show a significant curvilinear influence, sound institutions – in the sense of the absence of corruption or a high quality level of the administration – have a significant positive effect on economic growth. furthermore, and often linked to an overall sound institutional framework, there is some evidence that in the long run especially democratic regimes experience a high level of wealth (acemoglu and robinson, 2012). klein and luu (2003) do not focus directly on gdp/gni data, but on productivity trends in a country, and were able to show that differences in technical efficiency in a country sample can be explained by the specific institutional framework. in our own analysis, focusing not on the macro-, but on the micro-level, we assume that (good) institutions are important for economic growth as they facilitate privatesector investments in infrastructure and the private sectors contributions to public services. good institutions provide reliable signals for the private sector to invest, european journal of government and economics 3(2) 108 as contracts can be secured without prohibitively high transaction costs. if the administration works efficiently, is not corrupt, and transparency as well as reliability on existing contracts can be assumed to be given, it is much more likely that private partners will invest in joint projects, which may result in a better supply of public goods and infrastructure, and therewith in the long run higher economic growth. this is even more important as contracts between the public and the private sector in its different forms arise within the context of incomplete information and uncertainty (parker and hartley, 2005). these contracts are “necessarily incomplete“ (erlei et al., 2007). not all future developments and contingencies can be foreseen by the contract parties, let alone fixed by adequate contract provisions ̶ a fact further compounded by the bounded rationality of the prospective partners, which could result in future opportunistic behaviour from either side (coase, 1937 and williamson, 1975/1985). transaction costs are the logical result of this incompleteness and play a pivotal role for the successful or unsuccessful initiation and longer-term implementation of partnerships (klein et al., 1996). to be more specific, transaction costs in these partnerships include ex ante the costs of the bidding process, negotiations and the establishment as well as ex post costs for re-negotiations and compliance (dudkin and välilä, 2005). especially for the private partner the reduction of transaction costs may be decisive: a profit-oriented enterprise will invest in public services only if the anticipated profit will at least cover its costs (plus yield a “reasonable” rate of return on its investment). with transaction costs often being a large percentage of the full project costs, excessive transaction costs may render the whole project unprofitable as a result. a decrease of the (anticipated) transaction costs therefore will likely enhance the probability that contracts conducive to development between the public and the private partner will be realised in the first place. another important role of institutions in the context of ppps lies in the signal they send out to (international) development organisations: many projects related to development-driven investment in public services and infrastructure are assisted by these development agencies financially, technically and/or through management support (hammami et al., 2006). in detail, the assistance typically takes place in specific technical support and investment loans as well as through guarantees against political risk, currency risk and (partly) credit risk. this assistance – as well as “conventional” development aid – is usually linked to “good governance” in the meaning of existing democratic structures, the adherence to human rights and the rule of law as well as the existence of (effective) anti-corruption measures (nuscheler, 2005). by having established sound institutions, a necessary precondition is therefore fulfilled for the public partner to gain direct access to additional funding for the project from international donors. moreover, for the private partner this also increases the likelihood to obtain financial support or specific warranties from development agencies – a fact which seems to be very important due the high risks of a partnership and/or for the enterprise. political risks like expropriation or civil unrest, substantial exchange rate risks and additional risks of doing business may then be covered by the development agencies’ warranties, so the availability (or lack thereof) of some independent outside protection against said risks seems to bear potentially strong influence a private sector company’s penchant to engage in a project. additionally, the existence of sound institutions can be seen as a proxy for a government's reputation and stability and, as a result, will prove helpful in attracting more private enterprise activity in that country. this is even more important as any incomplete contract – as outlined above – is inevitably linked to high transaction costs, which, however, can be effectively lowered by mutual trust between the potential partners. every credible signal of a government’s trustworthiness will allow more such partnerships to come to fruition. summarised, “ppps cannot be found in areas where institutional failure and governance gaps are exceptionally pronounced” (schäferhoff et al., 2007: 10). schomaker ● institutional quality and private sector participation 109 the economics of private sector involvement in public goods and infrastructure provision due to shrinking tax revenues, or rising governmental expenditures (i.e. for social security), or a combination of both, more and more state and local governments face severe budgetary restraints. accordingly, there is a global trend for governments trying to improve their fiscal stance, especially on the expenditure side. at the same time, on the grounds of demographic change as well as resulting from economic growth, the pressure on public goods, particularly on existing infrastructures is raising.1 the situation is even more exigent in the less developed parts of the world. still rising rates of urbanisation, high rates of population growth and rising standards of living have for years increased the stress on public services and infrastructure to the point where demand greatly exceeds supply. private investment, or as a broader concept, private sector involvement can be assumed to be one potential solution for the problems outlined above. this investment can take place in different guises – as full or in-part privatisation, or as joint projects in the form of public-private partnerships (ppp) between the public administration and private enterprises, especially international companies. as privatisation may not be the first best solution from an economic theory perspective as far as public goods – i.e. market failure – are concerned or if the privatisation process might result in merely swapping a state-run for a private monopoly, alternative forms of public-private partnerships exist which may ensure adequate control rights for the public sector over the crucial aspects of service provision. many of these “new” partnership models are still lacking a precise definition (budäus and grüb, 2007). therefore, in this paper, we use the term public-private partnerships for a whole portfolio of different partnership models which range from full ownership by the public sector to material privatisation (grimsey and lewis, 2005). the prevalent definitions of ppp typically focus either on the players involved or on procedural aspects of these partnerships. linder (1999) broadly defines the term ppp as ‘rubric for describing cooperative ventures between the state and private business’. grimsey and lewis (2005) see ppps as ‘arrangements whereby private parties participate in, or provide support for, the provision of infrastructure, and a ppp project results in a contract for a private entity to deliver public infrastructure-based services’. ziekow (2003), by contrast, defines ppp more broadly as the junction of rationalities of actors relevant for action, and therewith refers to the, probably, most important factor in the decision whether or not to pursue a ppp: the interest of both partners. regardless of the specific form it may take in practice, any ppp-type contractual agreement between a private company and the public sector – whether at the state or at the municipal level – is typically characterised by a risk-sharing arrangement with the private partner taking over (at least some) financial responsibility. the duration of ppp projects varies between one and thirty years, with mere service and management contracts being of relatively short-term duration, while ppp designs which include the construction of an asset and which are refinanced by user fees are usually based on longer contract durations so as to ensure full cost recovery for the private partner. the commercial risk of failure is mainly borne by the public sector in (short-term) projects that do not affect the ownership of the asset (which remains with the public sector). in ppp where a direct contact between the private company and the customer/user exists (i.e. concessions and build-operate-transfer/build-own-operate/build-own-operate-transfer ppp variants), the private partner has to shoulder the main portion of the commercial 1 this paper covers – in line with the world bank’s definition of infrastructure – telecommunications, transport, energy and water supply/sanitation infrastructures. european journal of government and economics 3(2) 110 risk, e.g. deficits in payments, because of the high upfront costs he must refinance over extended periods of time. the expectations related to the socio-economic outcomes of ppp differ substantially among participants and stakeholders, but are usually very high among all participants compared to their assessment (or perception) of the outcomes the institutional status quo ante. the public partner focuses on its need to procure additional capital as well as to attract managerial competences and technical skills – which are of special relevance in the case of technologically complex infrastructures like water supply, sewage systems, energy, and telecommunications – without losing the political control over infrastructure provision. on the contrary, the private company the profit motive is the dominant one (hammami et al., 2006). in addition to maximizing its profit in the specific projects at stake, the private partner typically pursues a long-term, strategic goal, tool. to gain at least indirect access, by committing to a specific ppp in a specific country, to a market which is not open for full privatisation (yet) – an investment, in turn, in a potential first mover advantage in the eventuality of a later privatisation (provided the company has gained a reputation as a dependable and fair partner in the ppp period). moreover, the likelihood of winning future tenders for similar ppp projects in other countries will increase with the specific knowledge gained. additionally, with the related concepts of corporate social responsibility (csr) and stakeholder activism becoming more and more relevant for multinational companies, in some special cases the decision to enter into a ppp project may not exclusively follow a short-term profit maximisation motive (narrowly defined). it may rather be intended as a signal of a high(er) degree of stakeholder orientation to customers, the public, the media, governments and ngos from all countries where the company is doing business (unido, 2002). in other words, under such circumstances a ppp activity may convince the public partner of the company’s goodwill and may accordingly have influence on considering the company of being the government’s preferred partner in future (case-by-case) decisions on upcoming ppp projects or privatisation programmes – thereby giving rise to another first mover advantage by raising potential rivals’ costs of market entry. beyond the interests of the public and private partner, a third party, the (international) organisations engaged in development cooperation like the world bank have a specific self-interest in boosting ppp. the emergence of new challenges in development cooperation, e.g. “nation building” in failed states like afghanistan and somalia, rises additional tasks, and the budgets of international development agencies are not keeping up with their steadily growing expenditures requirements. a stronger involvement of the private sector, especially in expensive and technologically complex infrastructure provision, may positively contribute to narrowing or even closing that gap and may also enhance the – so far often dubious – quality of the assistance (moyo, 2009 and schäferhoff et al., 2007). the potentially pivotal role of private enterprise in promoting economic development has therefore rightly been highlighted by key institutions of development cooperation, e.g. in the united nations’ “global compact”, which has been established to involve the private sector (annan, 1999). even the declaration of the united nations’ millennium development goals emphasises the crucial role of private companies as stakeholders in development cooperation to achieve a substantial reduction of global poverty by 2015. as a consequence, more and more ppps have emerged in developing countries, and it appears from anecdotal evidence as if the peak has not been reached yet, despite comprehensive data still lacking. but from the number of ppp in infrastructure projects which is a good proxy for overall private-sector investment in public services, we can observe an increase of both the investment volume itself and the sheer numbers in all regions worldwide since the 1990s. also additional projects in countries which had not been covered before have been implemented in schomaker ● institutional quality and private sector participation 111 the last decade (world bank ppi data base).2 in many developing countries, the political and public awareness of their potential merits as a toll on development strategies is rising. political initiatives like “national strategies” and specific ministries have been established to integrate more private partners in the provision of public goods and infrastructure, and more efficiently and effectively (smith et al., 1997). model and results model and variables one of the main challenges in testing the impact of “good governance” or “institutional quality” in private sector involvement is not only to define what a “sound” or “good” institution” is, but also how to adequately operationalise the concept of “institutional quality” for quantitative research. several issues have to be addressed and solved, with the conceptual difference between the material institution itself (the content dimension) and the question whether this institution is also truly enforced (or perceived to be) (the trust dimension) being of pivotal importance (voigt, 2002). generally speaking, the assessment of public policy or public institutions as well as the assessments of the public administration is built on a limited selection of indicators. as for the political level, accountability and democratic structures are popular variables, while for the administrative level, effectiveness of the bureaucracy or control of corruption are often utilised. this approach in many cases may be misleading, as e.g. the selection/composition of indicators may be biased by the research design or data-availability, or the comparability of different administrations may not be given (see van de walle, 2006). consequently, a number of different indicators and measurement methods are used in the respective economic, administrative or political science literature. there is a number of arguments against or in favour of using a specific indicator, depending on the goal and/or range of the respective work (see for the discussion in detail e.g. bovaird and löffler (2003) or van de walle (2006). as for our study, many of these popular indicators are either too specific, as they focus on the output or outcome of the bureaucracy, too wide (focusing on the general business environment or the political system), or do not cover the necessary range of countries which are included in our empirical assessment. another problem which arises in this context is the question of how to get information on the institutional quality. in many cases the quality cannot be observed directly, so sophisticated extrapolations or estimates have to be applied. another option is measuring the perception of institutional quality among citizens, officials and companies. using this method, the construction of the indicator relies on different peoples’ evaluation of one reality, but there is no guarantee that these evaluations truly reflect reality, or are at least remotely representative. it is therefore safe to assume that there is always a bias, not least as the people who were interviewed may have an own interest in a specific outcome so that they attempt to create a systematic bias by their willfully deceptive answers. a further possibility is that the bias occurs by accident, but remains unnoticed. this is particularly true in cases where different people are asked, who are subject themselves to an unequal bias due to their deviating respective interests, their different abilities to judge and intellectual capabilities. this simple example is suitable to document the main problem in the construction of reliable indicators on 2 a ppp is “established” – according to the world bank’s definition – once the respective project has reached the stage of contractual or financial closure, irrespective of later modifications in the contract or a complete failure of the ppp. european journal of government and economics 3(2) 112 institutional quality, and could be easily extended to cover a broad range of other governance indicators as well. additionally, even where a quantitative approach to constructing indicators can be reasonably applied, as for those indicators or their components no (substantial) measurement problems on the country level exist (e.g. regarding the number of patents), the questions remain how reliable these statistics really are and whether the pertinent national data are comparable across countries. another caveat results from the fact that there may be no significant difference with respect to outcomes between those cases where the institution itself is identified as “weak” and those other cases where it is just being perceived as being weak by the actors involved, like enterprises or individuals. in this case, there may be another specific kind of bias: the sheer existence of an institution may lead to the (factually erroneous) subjective judgment that institutional quality is good without interviewees taking into consideration that in reality the institutional quality is indeed bad. an example might be the inability to properly enforce the – formally existing – institution, for instance for a chronic failure to protect (intellectual) property rights where the existence of a law does not mean that violators are actually prosecuted and/or adequately punished for their infringements. but, despite of the manifold analytical limitation outlined above, using a combination of these disaggregated indicators does, in our view, provide an acceptable – for being broad-based – approach to measure institutional quality, without losing too much information due to too much aggregation. therefore, we hold that the set of governance indicators generated by kaufmann et al (2008), which measure the perception of the quality of different institutions in all countries worldwide, are an appropriate tool to measure the quality of institutions for the purposes of our study. originally we planned to use the full set of institutional variables (voice and accountability, government effectiveness, rule of law, control of corruption, political stability, regulatory quality), as provided by kaufmann et al. additionally, variables like population size, inflation and the gross national income (gni) per capita per year, and regional dummies which we consider as exogenous and which we use to control for other effects have been tested for. the choice of the variables is in full conformity with our theoretical considerations above: the variable voice and accountability influences the involvement (financial and/or technical) of international development agencies due to political and economic conditionality, the indicators political stability and government effectiveness function as a signal for the potential private partner of how risky a commitment might become, of how high the credibility of the government and the administration is and of how efficiently the public services work. the absence of corruption or at least a strong control of corruption gives a signal on the presumable transaction costs, and the indicator regulatory quality provides information on the regulatory framework which determines (especially in naturally monopolistic infrastructures like water supply) the day-to-day operation of the ppp. testing the exogenous institutional variables for multicollinearity, we found coefficients near 1 for the indicators rule of law, control of corruption, political stability, regulatory quality and government effectiveness. owing to the high correlation between these indicators and the probable endogeneity induced by the latter we decided to remove the indicators rule of law, control of corruption, political stability and regulatory quality because some of the determinants covered by these indicators may be assumed to be also covered by the two remaining ones. to be more precise, a high quality of public services normally means that the administration is not corrupt, rule of law is given and an overall quality of political regulation can be assumed as being given. to include both relevant institutions, the quality of the administrative institutions as well as the political institutions, we use the indicator government effectiveness, which proxies for the overall quality of the public administration, while voice and accountability proxies for the existence of a stable democracy. schomaker ● institutional quality and private sector participation 113 government effectiveness (ge): this indicator measures the “perceptions of the public services’ quality, the quality of the civil service and the degree of its independence from political pressures, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies” (kaufmann et al., 2008). voice and accountability (va): this indicator measures the “perceptions of the extent to which a country’s citizens are able to participate in selecting their government, as well as freedom of expression, freedom of association, and a free media” (kaufmann et al., 2008). according to the theory of institutional economics, our hypothesis is that the incidence of private sector investment and ppp projects in a country may strongly depend on the quality of local institutions. private-sector investment in public services and infrastructure or partnership regimes between the private and the public sector on the micro-level, as outlined above, comprise several different governance regimes concerning the key elements project duration, organisational structure, and “quality” of the risk-sharing and/or the control rights for the public administration. as data on the overall volume of private investment on the microlevel is hard to obtain, we use a sound and statistically valid proxy — the number of ppp projects in infrastructure projects — instead. accordingly, we propose some regressions of the “number of ppps” variable, considered to be endogeneous, by using the ordinary least squares (ols) method as well as a poisson regression model and a negative binomial regression model, respectively, as they seem to be more suitable due to the nature of the data which appear not to completely fulfil ols requirements (long, 1997). table 1: descriptive statistics n minimum maximum mean standard deviation number of ppps 724 0.00 686.00 20.35 55.39 gni2 80.00 11150.00 1553.64 1669.47 va1 -2.19 1.21 -.4484 .77632 gove2 -2.25 1.39 -.4830 .59855 bev 71079.00 1311020000.00 40541783.31 149814932.48 inflation -24.00 5400.00 29.0127 254.19555 using the variables discussed above, we ran a cross-sectional model with data for up to six years and for 118 countries (unbalanced panel; see the annex for the complete list of countries). the descriptive statistics can be drawn from table 1. results and discussion as can be drawn from table 2, the results from the different models used show a high degree of consistency in terms of the sign of the respective coefficient and the significance level. additional tests for robustness (not displayed), using different indicators for institutional quality, validate the findings of the regression models used. in all cases, the fisher-test (f-test) and the omnibus-test indicate that the model is significant at 1 per cent level; also the goodness of fit (for the non-ols models pearson's chi-squared statistic and log-likelihood, as well as r²/r² adjusted for ols) is high for the models displayed. the indicator voice and accountability (va) displayed a significant negative correlation with the number of ppp projects in a country, as the coefficient is negative for the ols-regression and beyond 1 for the non-ols regressions. this implies that a higher degree of democratic participation inhibits partnerships with private enterprise and decreases the odd that a large number of ppps exists. this result, at first glance, seems to be inconsistent with the theoretical explanation that democratic institutions will foster the implementation of ppp. on closer inspection, however, it can be explained through the governments’ willingness to use private european journal of government and economics 3(2) 114 sector participation. it might be the case that democratic governments would prefer full privatisation because they have no fear of a loss of political control over the affected sector. in these cases, they will not resort to ppp solutions, but will rather liberalise the complete sector and open it completely to the private sector (which might result in a mixed economy setting of public suppliers competing with private firms in the same sector). on the other hand, semi-democratic or autocratic regimes (characterised by a relatively low value for voice and accountability), which are not willing to hand over control of vital infrastructure sectors, prefer to use ppp projects as opposed to full-blown privatisation. table 2: results of the regression analysis independent variable ols negbin poisson intercept 14,427 (2,967)** 3,502 (0,055)** 16,624 (0,0903)** 3,349 (0,009)** population 2,503e7 (25,551)** inflation (%) 0.000 (-0,079) real gni per capita (lagged by one year) 0,007 (6,229)** 1,000 (3,1331e-005)** government effectiveness (index) 10,945 (2,668)** 1,819 (0,104)** 4,139 (0,1072)** 1,293 (0,015)** voice and accountability (index) -10,945 (-3,916)** -0,473 (0,076)** 0,633 (0,0688)** -0,545 (0,013)** dummy east asia and pacific -7,829 (-1,466) dummy europe and central asia -4,300 (-0,797) dummy sub-saharaafrica -19,983 (-4,590)** dummy mena -37,404 (-6,070)** -1,438 (0,141)** -1,743 (0,048)** dummy south asia (-5,474)** observations 714 695 695 695 r²/r² adjusted 0,574/0,567 t-statistics in parentheses: ** significant at 1 percent sources: author’s compilation, data sources: world bank ppi-date base, kaufmann et al 2008. the indicator government effectiveness has, in all models, the expected positive sign as well as a significant impact. this result is consistent with the theoretical analysis from an institutional economics perspective. it also substantiates our assumption that the existence – or at least the perception – of a high quality of the public service and a highly credible bureaucracy foster the implementation of ppp. if the quality of the civil service and the credibility of the government’s commitment is high, the odd increases that ppps are implemented. this might be caused by the direct signalling function of government effectiveness to the private sector, and to the international development agencies alike, which are more likely to support ppps if “good” governance exists. while the overall macroeconomic development, proxied by economic wealth and the inflation rate of the respective country does not matter, the size of the country in terms of population does. further studies should include this factor, whether by standardisation of the number of ppps on population or through using other dependent variables as e.g. average investment volumes per capita. first empirical tests, standardizing the number of ppps on population, nevertheless underpin the outcomes discussed above. furthermore, there are significant differences between the regions of the world, which cannot be explained by the institutional or economic conditions tested for. these regional differences may be explained by the missing political will to include the private sector in service provision at all. also “time-lags” may be an explanation, as some regions are lagging behind in terms of only recently opening up towards private sector participation, such that private companies may lack the will to invest in politically unstable regions like the middle east and north africa. schomaker ● institutional quality and private sector participation 115 concluding remarks the results of our research provide strong evidence that a significant positive relationship exists between the indicator government effectiveness and the number of ppp projects in a country. as ppp numbers can be used as a reliable proxy for the total involvement and investment of private enterprises in public services and infrastructure, our results can be extended to assess the overall impact of institutions on this kind of private sector involvement. therefore, our findings stress the importance of a sound institutional framework for more private sector involvement, which is likely to lead to economic growth. amongst other explanations, such as the lack of political will to attract private capital, a low institutional quality may therefore provide a powerful explanation for the lack of private investment in many countries. from the perspective of prospective private sector partners, weak institutions increase uncertainty and, as a some kind of “collateral damage” also the project-related transaction costs, which may thus explain the private sector’s lack of will or its inability to engage in ppp projects in a specific country. this finding is even more important against the trivial insight that in particular large or international private companies are able to decide relatively independently where to invest their resources in a ppp because of the fast growing demand for private capital and expertise in public services and infrastructure worldwide. the creation of sound and trusted institutions, especially on the micro-level, in administration, to “cure” the incompleteness of public-private contracts and to reduce transaction costs therefore is a crucial first step to attract private enterprises whose investment may substantially contribute to economic growth and development. therewith, our work gives some hints on a “second institutional channel of development” – sound institutions do not only influence foreign direct investment in the narrow sense, but also private sector activities in the provision of public services. this seems to be important as the private enterprises included in our study come from industrialised “foreign” countries as well as from the respective country itself. improving framework conditions for both kinds of enterprises to invest in public infrastructure may be an important instrument of the political level, but also the administration itself to foster long-term growth. references acemoglu, daron and james a. robinson (2012) why nations fail. new york: crown business. apolte, thomas (2004) ‘discussion on wagener: buon governo – mal governo‘, in thomas eger (ed) erfolg und versagen von institutionen. berlin: duncker & humblot, pp. 141-145. annan, kofi (1999) ‘address at the world economic forum in davos’, un press release sg/sm/6881. http://www.un.org/news/press/docs/1999/19990201.sgsm6881.html (consulted 19.03.2013). bovaird, tony and elke löffler (2003) ‘evaluating the quality of public governance: indicators, models and methodologies’, international review of administrative sciences 69(3): 313-328. budäus, dietrich and birgit grüb (2007) ‘anhaltspunkte und hypothesenbildung für eine theorie der public private partnership’, in dietmar bräunig and doeothea greiling (eds) stand und perspektiven der öffentlichen betriebswirtschaftslehre ii. berlin: berliner verlagsgesellschaft, pp. 421-431. coase, ronald h. (1937) ‘the nature of the firm”, economica, 4(16): 386-405. european journal of government and economics 3(2) 116 dudkin, g. and välilä, t. (2005) ‘transaction costs in public private partnerships: a first look at the evidence’, european investment bank economic and financial report 2005/03. erlei, mathias, martin leschke and dirk sauerland (2007) neue institutionenökonomik, 2nd edition, stuttgart: beck. grimsey, darrin and mervyn k. lewis (2005) ‘introduction’ in darrin grimsey and mervyn k. lewis (eds) the economics of public private partnerships. cheltenham: edward elger, pp. xiii-xl. gschlößl, susanne and claudia czado (2008) ‘modelling count data with overdispersion and spatial effects’, statistical papers 49:531-552 hammami, mona, jean-françois ruhashyankiko and etinne b. yehoue (2006) ‘determinants of public private partnerships in infrastructure’, imf working paper wp/06/99. hartwig, karl-hans (2005) ‘infrastrukturpolitik in der diskussion’, in karl-hans hartwig and andreas knorr (eds) neuere entwicklungen in der infrastrukturpolitik. beiträge aus dem institut für verkehrswissenschaften an der universität münster. heft 157. göttingen, pp. 8-30. hart, oliver (2005) ‘incomplete contracts and public ownership: remarks, and an application to public-private partnerships‘, in darrin grimsey and mervyn k. lewis (eds) the economics of public private partnerships, cheltenham: edward elgar, pp. 295-314. hodgson, geoffrey m. (1998) ‘the approach of institutional economics’, journal of economic literature 36(1): 166-192. kaufmann, daniel, aart kraay and massimo mastruzzi (2008) governance matters viii: worldwide governance indicators 1996-2008’, policy research working paper 4654, world bank development research group. klein, michael, jae so and ben shin (1996) ‘transaction costs in private infrastructure: are they too high?’, worldbank private sector development department viewpoint note no. 95. klein, peter and hung luu (2003) ‘politics and productivity‘, economic inquiry 41(3): 433-447. linder, stephen h. (1999) ‘coming to terms with the public-private partnership: a grammar of multiple meanings’, american behavioral scientist 43(1): 35-51. long, scott j. (1997) regression models for categorical and limited dependent variables (advanced quantitative techniques in the social sciences), thousand oaks: sage. moyo, dambisa (2009) dead aid. why aid is not working and how there is a better way for africa, new york: farrar, straus and giroux. north, douglass c. (1991) ‘institutions‘, journal of economic perspectives, 5(1): 97-112. north, douglass c. (1992) institutionen, institutioneller wandel und wirtschaftsleistung, tübingen: mohr siebeck. nuscheler, franz (2005) entwicklungspolitik, 5th edition. bonn: bpb obinger, herbert (2001) ‘demokratie, institutionen und ökonomische entwicklung: eine replik auf christian martin und thomas plümper’, swiss political science review 7(3): 81-96. organisation for economic co-operation and development (oecd) (2007) statistical annex of the 2007 development cooperation report, paris,http://www.oecd.org/ dataoecd/52/9/1893143.xls (consulted 13/12/2013) schomaker ● institutional quality and private sector participation 117 parker, david and keith hartley (2005) ‘transaction costs, relational contracting and public private partnerships: a case study of uk defence’ in darrin grimsey and mervyn k. lewis (eds) the economics of public private partnerships, cheltenham and northampton: edward elgar, pp. 303-314. richter, rudolf and eirik. g. furubotn (1999) neue institutionenökonomik, 2nd edition, tübingen: mohr siebeck. sachs, jeffrey d. and andrew m. warner (1997) ‘natural resource abundance and economic growth’, center for international development and harvard institute for international development discussion paper, nber working paper 5398. schäferhoff, marco, sabine campe and christopher kaan (2007) ‘transnational public-private partnerships in international relations: making sense of concepts, research frameworks and results’, sfb-governance working paper series 6. smith, graham r., n. shafik, p. guislain and j. a. reichert (1997) getting connected: private participation in infrastructure in the middle east and north africa, world bank middle east and north africa economic studies, washington d.c.: world bank. united nations industrial development organization (unido) (2002) unido business partnerships for industrial development, vienna. van de walle, steven (2006) ‘the state of the world’s bureaucracies’, journal of comparative policy analysis 8 (4): 437-448. voigt, stefan (2002) institutionenökonomik, paderborn: fink. wagener, hans-jürgen (2004) ‘die bedeutung guter regierung für wohlfahrt und transformation’, in thomas eger (ed.) erfolg und versagen von institutionen, berlin: duncker & humblot, pp. 119-140. williamson, oliver e. (1975) markets and hierarchies: analysis and antitrust implications. a study in the economics of internal organization. new york: free press. williamson, oliver e. (1985) the economic institutions of capitalism, cambridge: cambridge university press. williamson, oliver e. (2000) ‘the new institutional economics: taking stock, looking ahead’, journal of economic literature 38(3): 595-613. world bank ppi data base (ongoing) private participation in infrastructure database, http://ppi.worldbank.org (consulted 26/12/2013). ziekow, jan (2003) ‘verankerung verwaltungsverfahrensrechtlicher kooperationsverhältnisse’, in jan ziekow (ed) public private partnership – projekte, probleme, perspektiven, speyer: speyer university, pp. 25-78. european journal of government and economics 3(2) 118 annex table a1: countries included in the empirical analysis albania cuba laos somalia algeria djibouti lesotho south africa albania dominica liberia sri lanka angola dominican republic madagascar sudan argentina east timor malawi surinam armenia ecuador malaysia swaziland azerbaijan egypt, arab rep. maldives syria bangladesh el salvador mali tadzhikistan belarus eq. guinea mauretania tanzania belize eritrea mauritius senegal benin fiji mexico seychelles bhutan gabon moldavia sierra leone bolivia gambia mongolia thailand botswana georgia morocco togo brazil ghana mozambique tonga burkina faso grenada namibia tunisia burundi guatemala nepal turkey cambodia guinea nicaragua turkmenistan cameroon guyana niger uganda cape verde haiti nigeria ukraine central african republic honduras oman uruguay chad india pakistan uzbekistan chile indonesia panama vanuatu china iran, islam. rep. papua new guinea venezuela colombia ivory coast paraguay vietnam comoros jamaica peru west bank and gaza congo jordan philippines yemen congo, dem. kazakhstan poland zambia costa rica kenia russian federation colombia kiribati ruanda comoros kirgizstan samoa european journal of government and economics volume 5, number 2 (december 2016) issn: 2254-7088 82 productivity in europe during the great recession: any evidence for creative destruction? gonzalo paz pardo, university college london, united kingdom abstract this article analyses the effects of the financial crisis and the great recession on productivity in europe by studying the process of labour force reallocation between companies. using micro-data on company balance sheets, a fixed-effects panel estimation of the predictors of the post-crisis evolution of the number of employees for a given company is used. identification is achieved through the use of pre-crisis values of covariates. the results are in line with the theoretical predictions derived from schumpeterian (“creative destruction”) endogenous growth models. pre-crisis productivity is a predictor of a higher number of employees, which means creative destruction is taking place to some extent. companies in financially dependent sectors perform worse in the context of the financial crisis. indebtedness has an uneven effect: positive for large companies and negative for smaller ones. keywords productivity; business cycles; financial constraints; indebtedness; company performance. jel classification o47; e32; l2. european journal of government and economics 5(2) 83 introduction the impact of the 2008 financial crisis and the great recession on productivity has been very heterogeneous across european countries. some of them, like spain, have seen significant increases in labour productivity during the period 2008-2013. in others, for example the uk or greece, there has been a decrease in labour productivity. this empirical observation provides a recent motivation for the study of the underlying mechanism that links a recession or a financial crisis to the evolution of productivity. do recessions help reorganise the economy? or are they damaging to long-term productivity growth? the answer to these questions is also interesting from a normative sense, with regards to policy in terms of recessions. if a crisis destroys the most unproductive companies, then it is inefficient to help them survive during difficult times. if, on the contrary, companies are struck because of reasons out of their control and not related to productivity, it might be positive to subsidise them during crises. many theoretical papers have developed models to analyse this issue (c.f. caballero and hammour 1994, or hall 2000, for instance), reaching diverging conclusions. in the data, labour productivity is apparently procyclical, which would support the theory that recessions are negative for the evolution of productivity. however, since the effects of reorganisation are not immediate, this observation is not conclusive. it might be that the reallocation that happens during recessions does improve long-run productivity (galí and hammour, 1991). the main contribution of this paper to this stream of literature is its new approach: i study the evolution of productivity through the use of panel micro-data on company balance sheets. this will help to understand which the best predictors of company performance during a (financial) crisis period are. i predict, following schumpeterian growth models and previous studies like caballero and hammour (1994), that more productive companies grow more during the crisis, which would imply that recessions are good for productivity. however, i also predict that more financially dependent companies will grow less during the crisis (barlevy 2003), which could have as a consequence the closure of companies that would otherwise be productive and have a negative effect on productivity. the effect of indebtedness depends on company size: it is negative for small companies but positive for larger ones. i focus on eu-15 countries. the paper is structured as follows. in part 2 i analyse in detail the schumpeterian concept of “creative destruction” and point out the main studies that had been undertaken that try to establish a connection between business cycles, recessions and the evolution of productivity (both from a theoretical and empirical point of view). then, in part 3, i present some basic stylised facts about the evolution of productivity as a result of the great recession. part 4 refers to the formal framework of schumpeterian endogenous growth models, in which i am based to derive the main theoretical predictions about the channels that can be explaining the performance of a company during a crisis and, therefore, the overall level of productivity. in part 5, i use a panel-data approach, with data on balance sheets of european companies, to test the previous hypotheses by analysing which are the main predictors of company success in the context of a crisis. part 6 presents the main results. literature review the concept of “creative destruction” is usually associated with and was made famous by schumpeter (1939; 1942). creative destruction is a natural process that happens in capitalist economies and that consists, basically, on the replacement of older, less productive activities by newer, more productive ones. paz pardo ● productivity in europe during the great recession 84 individuals or companies who discover new products, processes or techniques enjoy a period of high income derived from their innovation. but, naturally, on the other side, companies that were selling or manufacturing similar products suffer the adverse effects of this innovation, in form of a reduction in sales. if the innovation is significant and they are not able to react, they may be forced to close. however, initially creative destruction was not considered a consequence of recessions. actually, schumpeter argues that some recessions are caused by a creative destruction process: the appearance of new products or ways of manufacturing can be so striking to the already established industries that it might cause a recession (schumpeter, 1942, p. 132). it was afterwards that the hypothesis that recessions induce a creative destruction process has been widely studied. many of the papers studying this issue follow the data analysis undertaken by davis and haltiwanger (1992), that find out that job reallocation is significantly countercyclical (e.g. caballero and hammour 1994; mortensen and pissarides 1994). there are two main explanations that can support this hypothesis of creative destruction during recessions. first, during a recession there is a decrease in profitability of companies and production units. in this adverse situation, the first companies or units to be affected are the least productive ones, which are forced to reduce their number of employees (or close). an increase in the overall level of productivity naturally follows. second, during recessions opportunity costs of productive factors are lower, since unemployment is high and there are large levels of unused capital. in this situation, there are incentives to undertake riskier, potentially very productive activities, which might have not been interesting during an expansionary phase (due to production factors already being profitable in established activities). this incentive to entrepreneurship also results in increased levels of productivity. if these phenomena are actually taking place, recessions would have a “cleansing effect” (caballero and hammour, 1994) through two main channels: the creation of new production units and the destruction of existing ones. in general, an insulation effect can take place during recessions: most of the change in demand is absorbed through the reduction in the creation of new production units, and it only partially causes the destruction of existing ones. nevertheless, this insulation effect is not complete, since destruction of production units actually happens. in fact, labour data show that job destruction is much more responsive to current economic conditions than job creation. a fact that stands against this hypothesis is that labour productivity is mainly procyclical. if creative destruction actually took place during recessions, productivity would rise. caballero and hammour (1994) blame this on factors like labour hoarding or externalities, while galí and hammour (1991) conclude, through a var approach, that the effects of business cycles on productivity happen in the long run. this would imply that procyclical productivity could be compatible with creative destruction and would not necessarily imply that the theory should be rejected. hall (2000) suggests a simplified channel through which creative destruction might be happening: “reorganization” can be considered as a costly economic activity (in terms of jobs), which can be basically equated with unemployment. the reason for that is that workers cannot work and efficiently look for the best job for them at the same time. therefore, only in periods with high unemployment can significant reorganisation take place: recessions boost productivity. the policy implications drawn by the different authors are varied (schumpeter (1942), for instance, was particularly contrary to keep subsidised unproductive sectors), but in general they recognise that, even if their evidence points towards european journal of government and economics 5(2) 85 creative destruction, measures should be taken to smooth the effects of recessions. another stream of literature stands against the theory that recessions help reorganise the economy and states that economic downturns might even have an adverse effect on the allocation of labour and capital. for instance, much in the line of the labour productivity issue mentioned above, barlevy (2002) points out that two effects might be happening at the same time. during a recession less efficient jobs are destroyed, but it might be the case that it complicates the creation of more efficient ones via a “sullying effect”. companies post fewer vacancies and agents are afraid of being unemployed for a long time, so they stick to their low-productive job as much as they can. a matching model calibrated with data from the united states (us) indicates that the sullying effect has a larger magnitude than the cleansing effect, which would imply that the frictions created by recessions outweigh (in terms of productivity improvement) the benefits of destroying less productive matches, units or activities. the damage of recessions can also be analysed from the point of view of firms. frequently, young firms are seemingly unproductive but have the potential to become productive in the long run as they grow. these potentially productive firms suffer particularly the effects of recessions and can be destroyed during early stages of their development, before their potential can be fully realised (and known by the economic agents). this effect is negative in overall productivity and for economic growth, but due to its characteristics it is quite difficult to apprehend from data (“scarring” effect in ouyang 2009). from a more general perspective, caballero and hammour (2005) find out that empirical evidence seems to show that recessions decrease the level of restructuring in the economy: there is resource misallocation as a result of recessions. they point out two main channels through which this might happen: the first of them is financial (firms have a very limited access to financing during recessions and in the period that immediately follows a recession), the second one is a selection mechanism (low-productivity firms cannot be created during a recession, so they cannot be destroyed afterwards, reducing the economy’s restructuring following a recession). the relevance of the financial channel is also highlighted by barlevy (2003). the reallocation that takes place during economic downturns might not go from least productive sectors to more productive ones, but the other way around, depending on the credit situation of the company or the economic sector. during recessions, there are credit market frictions and it is difficult for companies to secure financing. in this context, activities that require large amounts of credit might fail due to the impossibility of securing the required finance, while (maybe less productive) activities that do not need as much financing might survive for a longer time. according to barlevy (2003), more efficient projects require more credit in equilibrium. therefore, the effect of recessions, in this situation, would be exactly the opposite of the one suggested by creative destruction. based on all of these observations, a substantial literature, surveyed in aghion, akcigit and howitt (2015a) and whose main contributions are summarised in aghion, akcigit and howitt (2015b), has used models that generate endogenous growth from a schumpeterian paradigm. aghion and saint paul (1998) develop a model of productivity growth under cyclical fluctuations and analyse which kinds of assumptions are necessary to generate procyclical or countercyclical productivity improvements. they find that considering productivity growth as a cost in terms of current production induces procyclical productivity, while considering that productivity growth is a cost independent of current production induces countercyclical productivity. paz pardo ● productivity in europe during the great recession 86 most related with the analysis in this paper, aghion, hemous and kharroubi (2014) find that countercyclical fiscal policy has larger effects for sectors that rely heavily on external finance. 1. stylised facts: the evolution of productivity a first step towards analysing the effects of the crisis on productivity, and later on trying to separate whether creative destruction or some other force was taking place, is the analysis of the general evolution of macro-variables. figure 1. labour productivity us-eu15 source: own elaboration from oecd data. eu15 includes austria, belgium, denmark, finland, france, germany, greece, ireland, italy, luxembourg, the netherlands, portugal, spain, sweden and the united kingdom. a simple look at productivity trends (figure 1) indicates that productivity has been growing quite steadily over the past 40 years in both western europe and the us. the us has an overall higher level of productivity than europe, though until around 2000 it seemed that the gap was slowly closing. for example, blanchard (2004) provides an analysis of this. in recent times, and particularly after the financial crisis, the us has managed to maintain a more sustained growth of productivity than the european union. a simple analysis of labour productivity growth rates in the eu15 (figure 2) confirms the procyclicality that had already been analysed by caballero and hammour (1994). in particular, labour productivity fell significantly during the mid1970s recession and the current financial crisis. less significant falls in output (early 1980s, early 1990s) did not have such a remarkable effect on productivity. this contemporaneous relationship between productivity and gdp is probably mostly due to the fact that companies adjust labour slowlier than production (labour hoarding), a fact that has been widely studied in literature, as for example in burnside, eichenbaum and rebelo (1993). 0 10 20 30 40 50 60 70 1 9 7 0 1 9 7 2 1 9 7 4 1 9 7 6 1 9 7 8 1 9 8 0 1 9 8 2 1 9 8 4 1 9 8 6 1 9 8 8 1 9 9 0 1 9 9 2 1 9 9 4 1 9 9 6 1 9 9 8 2 0 0 0 2 0 0 2 2 0 0 4 2 0 0 6 2 0 0 8 2 0 1 0 2 0 1 2 g d p  p e r  h o u r  w o rk e d  ( u sd  2 0 0 5 ) united states eu15 european journal of government and economics 5(2) 87 figure 2. gdp and labour productivity (gdp/hours worked) growth rates (eu15) source: own elaboration from oecd data. the main focus of attention of this work is the evolution of productivity as a result of the current crisis. figure 3. labour productivity (gdp per hour worked) in constant terms source: own ellaboration from oecd data. the evolution of productivity was clearly divergent among the largest european economies (figure 3). we can see mainly four big types of countries in the graph: those that kept a rising level of productivity with a drop after the onset of the crisis, but that currently show a positive tendency (france, germany, eu15 as a whole); those that showed a significant rise of productivity before the crisis and a significant decrease afterwards (uk); the opposite pattern, with a significant rise of productivity after the crisis (spain); and the particular italian case, in which productivity kept stagnant during the whole period. ‐6 ‐4 ‐2 0 2 4 6 8 1 9 7 0 1 9 7 2 1 9 7 4 1 9 7 6 1 9 7 8 1 9 8 0 1 9 8 2 1 9 8 4 1 9 8 6 1 9 8 8 1 9 9 0 1 9 9 2 1 9 9 4 1 9 9 6 1 9 9 8 2 0 0 0 2 0 0 2 2 0 0 4 2 0 0 6 2 0 0 8 2 0 1 0 2 0 1 2 r e al  a n n u al  g ro w th  r at e  ( % ) productivity gdp 34 36 38 40 42 44 46 48 50 199920002001200220032004200520062007200820092010201120122013 2 0 0 5  u s$ france germany italy spain united kingdom eu15 paz pardo ● productivity in europe during the great recession 88 figure 4. percentage change in productivity 2013/2007 source: own ellaboration from oecd data. including other eu-15 countries on the list confirms the hypothesis that the effects of this crisis in productivity have been quite diverse across europe. it is relevant to note that there is no consistent relation between being worse struck by the crisis and the evolution of productivity. some of the countries that suffered a deeper recession (spain, ireland, portugal) top the chart, while others (namely greece) are at the bottom. the us is offered as a benchmark, with an evolution which showed to be quite significantly better than most of the european countries. stylised facts and creative destruction therefore, in the most basic of analyses, countries like spain, ireland or portugal show a priori evidence that the recession has fostered creative destruction in the sense of caballero and hammour (1994), while countries like the uk or italy show basic evidence of a scarring or sullying effect in the sense of barlevy (2002) or ouyang (2009). a simple schumpeterian growth model. finance. in order to provide a formal framework to undertake the analysis of labour reallocation, i rely on schumpeterian models by aghion and howitt (1998 and 2009). these authors were the first to develop a comprehensive theory of endogenous growth based on a schumpeterian vision of business cycles. it is from simple schumpeterian models that i derive the main hypotheses that i later test empirically. in a dynamic schumpeterian model, production units that do not adapt to new technologies face closure because they reach one point in which they can no longer cover the amount of fixed costs they have to face. this implies that the workers of that production unit will be fired and sent into unemployment. in the context of a recession, the likelihood that firms that did not undertake reorganisation investments go bankrupt is higher (as can be derived from aghion ‐15 ‐10 ‐5 0 5 10 15 luxembourg greece united kingdom finland italy netherlands belgium france denmark germany eu15 sweden austria portugal ireland united states spain % change european journal of government and economics 5(2) 89 and saint-paul (1998)’s model of “opportunity costs” with fixed costs of production). therefore, we can hypothesise that the natural process of destruction of unproductive units is sped up during recessions due to the general drop in demand (what aghion and howitt (1998, p.242) call “disciplinary effect”). this would imply an increase in the aggregate level of productivity. from the combination of these models i draw hypothesis 1: (h1) i expect that, when a recession arrives, the likelihood that firms that did not update their technology recently have to close production units is higher than before. this effect has as a result an overall increase in the level of productivity. i derive hypothesis 2 from a fully-fledged endogenous growth schumpeterian model. let’s assume that companies can invest in research, which is costly, but can potentially make the company generate an innovation (the probability of this happening is increasing the amount of research). this innovation implies that the products sold by the company will have a better quality, so it will lead its sector for a certain amount of time, in which it will obtain an extraordinary monopoly rent. after that period, the rest of the companies in the sector will copy the innovation and the monopoly rent will be gone. when financial constraints are introduced in such a model, companies need to borrow from the financial system to finance this research. banks need to screen companies that demand credit in order to discern whether they will be able to repay. the higher this “screening cost” is, the lower is the probability of innovation and the overall growth rate. i expect that countries in a financial crisis will be less willing to lend funds to companies, thus implying an increase in the level of screening costs. this is because, in a crisis context, many companies are going bankrupt and have financial trouble, so banks need to screen their debtors very carefully to make sure their projects are viable. furthermore, when the interbank market or international capital markets are dried up, banks cannot easily borrow from them, so they will be very unwilling to lend money. moreover, during the recent financial crisis some capital ratio requirements were raised, which further constrained the willingness of banks to lend. all of these factors, amongst others, drive up the screening costs. in this line we derive the second main hypothesis that i test. (h2) countries in a financial crisis should have a higher level of screening costs. therefore, industries in those countries should suffer from a lower probability of innovation and lower growth rates. this is particularly true for industries that are more dependent on external finance. 3. empirical study: panel data on balance sheets 3.1. data and variables the analysis is based on data from amadeus (bureau van dijk), a database of financial information from a large number of public and private european firms. the sample chosen corresponds to the eu15 from 2004 to 2013 (this is due to data availability, but allows for sufficient observations per company both in the precrisis and post-crisis period). i only consider companies with at least 10 paz pardo ● productivity in europe during the great recession 90 employees5, to avoid issues with data reliability, and i dropped companies without enough observations of the variables that interest me.6 3.2. methodology 3.2.1. hypotheses as i have argued before, i am interested in testing two hypotheses: has the labour force grown more after the crisis in companies that were more productive before? if the crisis was a “creative destruction” recession, then we should expect that the least-productive companies suffered from the biggest job losses, i.e., the crisis should have generated some positive labour reallocation (h1). has the financial dependence of industries played any role in the evolution of the number of employees in companies? if finance had a relevant role in job destruction, we expect the most financially-dependent sectors to suffer the largest job losses (a company that doesn’t need any external finance will not be affected, at least directly, by higher screening costs) (h2). i add a third hypothesis, derived from nickell, wadhani and wall (1992), as a control variable to make sure that we are properly identifying h2 and because of its own interest: (h3) does the level of indebtedness of a company play any role on the evolution of the number of employees after the crisis strikes? in all cases, i use (log) number of employees as a dependent variable. this is due to two reasons. first, it allows to summarise, in a clearly comparable metric, how different companies evolved during the crisis period. second, it is interesting per se, as employment outcomes are of utmost importance for policy makers. 3.2.2. identification strategy to answer these questions i use a temporal identification strategy, since endogeneity implies that it is not possible to directly run a regression of growth of the number of employees in a given year on these covariates in that same year (for example, maybe indebtedness decreases due to the fact that banks refuse to lend more money). to avoid it, we use pre-crisis values, i.e., we calculate both productivity for a company, finance dependence for a sector (rz_us and rz_eur, computed respectively from us and european data) and indebtedness for a company as an average of the values these ratios had between 2004 and 2007. they are introduced in the regression in an interaction term with a crisis dummy. the basic idea, for example with regards to the financial dependence component, is how much, ceteris paribus, belonging to a financially-dependent sector (measured in the pre-crisis years for identification) conditioned the evolution of a company once the crisis struck. the identification strategy proposed is methodologically equivalent to blalock, gertler and levine (2008) or dell’ariccia, detragiache and rajan (2008). it is alike 5 this does not imply a significant reduction in sample size, as small companies are frequently not present in the database, and allows to control for possible nonrandom selection of firms into the sample. 6 details about the variable definitions and sample selection criteria are available from the author on request. european journal of government and economics 5(2) 91 to a difference in differences approach that tries to compare how different were the evolutions of companies that had different levels of productivity and finance dependence in the pre-crisis years. the exogeneity of pre-crisis levels of productivity, finance dependence and indebteness depends on several assumptions. in particular, it requires that there is no unobserved factor that is correlated both with pre-crisis productivity levels and with trends in terms of the number of employees. to test the first hypothesis (h1) i check whether pre-crisis productivity is a good predictor of post-crisis increases in the number of employees. i expect that companies with a higher level of productivity before the crisis were the ones that were the least affected by the recession and therefore their number of workers increased. to test the second hypothesis (h2) i check whether a measure of financedependence of a company is a good predictor of post-crisis decreases in the number of employees. i expect that more financially-dependent companies were more affected by the recessions and had to dismiss more workers. this goes in line with kleemann and wiegand (2013), who found that, in a database of german firms, financially constrained ones generated significantly less employment than unconstrained ones, ceteris paribus. to test the third hypothesis (h3) i check whether the fact of being more indebted in the pre-crisis years is a good predictor of post-crisis decreases in the number of employees. it is relevant to note that in a normal situation, industries that are more externally dependent have in general a higher growth than the ones who do not, in countries with better institutions to support external finance (e.g. accounting standards) (rajan and zingales, 1998). 3.2.3. panel data specification the available dataset expands, as i have described, from 2004 to 2012-2013. this availability allows us to run panel data estimations under the fixed effects assumption. using fixed effects allows to control for the many omitted variables that unavoidably arise in this kind of regression, both observables that were not conditioned on and unobservables that cannot be directly introduced in a regression (e.g., good management) but that do affect the outcome of a company. however, the fact that they stay constant through time does not allow to specifically control for possible pre-existing trends. furthermore, we allow for covariance of the errors of the same firm in different years by clustering them at firm level, using again the same methodological approach as blalock, gertler and levine (2008). we do so because it is natural to expect that standard errors of the same firm will be correlated through time via unobserved factors that are not specifically modelled in our regression. in this case, the type of equation that we estimate is (where i represents firm and j represents sector): the dependent variable is in this case the log number of employees. rz is a finance dependence measure for a given sector, calculated following rajan and zingales (1998) (whether the european version, rz_eur; the current us version, rz_us, or the original version in their paper, rz); indebt is the ratio (loans + long-term debt)/tangible fixed assets; prod refers to the pre-crisis level of labour productivity (added value/number of employees). paz pardo ● productivity in europe during the great recession 92 further, is the fixed effect for every company and is a dummy for every year (which is reported) that summarises the overall economic situation of that year. coefficients of every year are reported as differences with respect to 2004. i cannot directly include country dummies because they are correlated with the fixed effects, but i consider differential cross-country effects later on. 4. results 4.1. basic results the basic regression results are presented in table 4 and table 5. results using the original measure of rajan and zingales (1998) and thus neglecting the nonmanufacturing sectors, and also using the absolute number of employees, are presented in appendix a (table a.1). table 4. main regression, all firms dependent variable: log number of employees all companies (10+ employees) rz measure rz_eur rz_us pre-crisis productivity * crisis 0.0000912*** (0.00002) 0.0000903*** (0.00002) rz external dependence * crisis -0.0001835*** (0.00005) -0.0001716*** (0.00004) pre-crisis indebtedness * crisis -0.0000034*** (1.4e-6) -0.00000339** (1.4e-6) year 2005 0.0428574*** (0.001) 0.0429737*** (0.001) year 2006 0.0816191*** (0.002) 0.0813291*** (0.002) year 2007 0.1381333*** (0.002) 0.1371525*** (0.002) year 2008 0.1859196*** (0.002) 0.1851432*** (0.002) year 2009 0.1853602*** (0.002) 0.1863146*** (0.002) year 2010 0.2084462*** (0.002) 0.2099721*** (0.002) year 2011 0.1966012*** (0.002) 0.1984595*** (0.002) year 2012 0.172461*** (0.003) 0.1742194*** (0.002) year 2013 0.192023*** (0.003) 0.1933403*** (0.003) constant 3.614179*** (0.002) 3.599731*** (0.002) observations 1095872 (130320 firms) 1140805 (135633 firms) f 1236.74*** 1317.35*** r2 0.0464 0.0475 *** significant to 1% level, ** significant to 5% level *significant to 10% level. special caution must be taken with 2013 coefficients, since data availability for that year is limited and in some cases provisional. european journal of government and economics 5(2) 93 table 5. main regression, large and small firms dependent variable: log number of employees large companies (150+ employees) small companies (10-150 employees) rz measure rz_eur rz_us rz_eur rz_us pre-crisis productivity * crisis 0.0001012*** (0.00003) 0.000099*** (0.00003) 0.0000826*** (0.00002) 0.0000826*** (0.00002) rz external dependence * crisis -0.0003755*** (0.00008) -0.0001494* (0.00008) -0.0001417*** (0.00005) -0.000174*** (0.00004) pre-crisis indebtedness * crisis 0.0000344*** (5.07e-6) 0.000033*** (4.5e-6) -0.00000357*** (1.17e-6) -0.00000357*** (1.17e-6) year 2005 0.048464*** (0.004) 0.0472898*** (0.004) 0.0418195*** (0.002) 0.0422472*** (0.001) year 2006 0.0970335*** (0.004) 0.0977771*** (0.004) 0.0780784*** (0.002) 0.0776263*** (0.002) year 2007 0.1687916*** (0.005) 0.1682157*** (0.005) 0.1312475*** (0.002) 0.1302932*** (0.002) year 2008 0.2139485*** (0.006) 0.2131201*** (0.005) 0.1798454*** (0.003) 0.1791782*** (0.002) year 2009 0.2167734*** (0.006) 0.2177575*** (0.006) 0.1785432*** (0.003) 0.179582*** (0.003) year 2010 0.2454604*** (0.006) 0.2478897*** (0.006) 0.2005456*** (0.003) 0.2020094*** (0.003) year 2011 0.2347438*** (0.006) 0.2374719*** (0.006) 0.1882018*** (0.003) 0.1900019*** (0.003) year 2012 0.2270153*** (0.006) 0.2289912*** (0.006) 0.1604976*** (0.003) 0.1623815*** (0.003) year 2013 0.2422487*** (0.008) 0.2430696*** (0.008) 0.1815711*** (0.003) 0.1831574*** (0.003) constant 5.49183*** (0.004) 5.475769*** (0.004) 3.205318*** (0.002) 3.197701*** (0.002) observations 195345 (23799 firms) 200703 (24468 firms) 899457 ( 106390 firms) 939016 (111032 firms) f 232.11*** 245.36*** 1020.49*** 1090.45*** r2 0.0677 0.0678 0.0418 0.0431 *** significant to 1% level, ** significant to 5% level *significant to 10% level. special caution must be taken with 2013 coefficients, since data availability for that year is limited and in some cases provisional. the coefficient of pre-crisis productivity is uniformly significant and positive in all the specifications. this indicates that, in some sense, (creative) destruction has been taking place. the more productive companies are the ones that have hired the most new workers: labour has been reallocated at a company level. we can interpret the coefficient 0.0001012 on productivity of the rz_eur specification for large companies as follows. i assume two companies, a and b, which have exactly the same characteristics but only differ in their productivity level, a having a productivity level of € 100,000 (2005 constant) per worker per year higher than b in the pre-crisis period. we expect company a to have on average 1.012 per cent more employees after the crisis stuck than b. even though the dimension of this coefficient is not very large, it is relevant to take into account that much of the effect of the crisis is being captured in the year fixed effects (which represent percentages with respect to 2004). further, the significance of the positive coefficient is robust in all the specifications we have considered (considering the absolute number of employees, including other variables like sectorial competition or in general in most countries, as i analyse later), and does not depend on the sample choice with regards to the size of companies. therefore, at a company level and considering the eu15 as a whole, we do see evidence that successful innovators in a schumpeterian sense indeed did better (positive evidence towards h1). paz pardo ● productivity in europe during the great recession 94 the coefficient of finance dependence that proxies the one used by rajan and zingales has a negative significant effect. this implies that finance had indeed an effect on the evolution of companies. companies that had a structural need for more finance experienced more decreases (less increases) in their number of workers. this is coherent with a situation of difficulties in international financial markets in which banks are unwilling to lend money, regardless of the situation of the specific company. in the context of the aghion-howitt schumpeterian model with financial system, this is equivalent to the screening cost f being higher and affecting particularly these sectors, which have therefore lower growth rates and generate less employment. this conclusion agrees with aghion, hemous and kharroubi (2014), who find that sectors that strongly rely on external finance are those who can gain the most from policy interventions in recessions. furthermore, results go in the same line as those of dell’ariccia, detragiache and rajan (2008) or kroszner, laeven and klingebiel (2007). this result is robust towards calculating the rz_eur measure with current us data (2004-2007), for all three sets of companies. the use of the original rz coefficient (appendix a, table a.1) yields different results. the differences with rz_eur and rz_us are not due to the use of only manufacturing companies (results above are robust towards that selection), but probably to the outdated nature of the coefficient. the economic significance of this coefficient is not so straightforward to analyse. in the next section i consider an alternative specification with dummy variables that represent financial dependence deciles in order to get an understanding of the dimensions that this effect may have. the coefficient for the variable that proxies indebtedness does not have a consistent sign. it is negative for small companies and positive for large companies. overall, the coefficient is negative. therefore, being more indebted in the period 2004 to 2007 implied more hirings after the crisis for large companies, while for smaller companies it implied less hirings. a hypothesis that might explain why this is happening for large companies is that more indebtedness is related to having more profitable investments (barlevy, 2003). it can be also related to the fact that these companies have a better access to the financial system due to their size, their internationalisation or interconnections with banks. another, from nickell, wadhani and wall (1992), is the so-called “discipline of debt”: high debt levels force managers to be more dynamic (seek new opportunities, expand markets), thus improving the overall performance of the company. the same argument appears in aghion et al. (2005): firms with a higher debt ratio will try to innovate more to escape the possibility of bankruptcy. for the whole population of firms, the negative coefficient is possibly related to the intuitive prediction: an indebted firm suffers more during the crisis because it needs to repay its debt. furthermore, during the crisis it might not obtain a refinancing of the amounts due while in an expansion context it would. for the overall sample, an increase of 10 percentage points of the indebtedness ratio (as defined above) is correlated with a 0.36 per cent lower number of workers during the crisis (3 per cent higher for large firms). the differences between large and small companies can also be related to an argument proposed by dell’ariccia, detragiache and rajan (2008): small european journal of government and economics 5(2) 95 companies rely more on banks situated on their same country, while large companies can get financing abroad or in securities markets. 4.2. financial dependence deciles as we anticipated earlier, we have run a regression exactly identical to the one earlier, just substituting the rz component for decile dummies interacting with a crisis dummy, in order to more deeply understand the effect of belonging to a financially-dependent sector. the results of the regression (for large firms) are in appendix a (table a.2). on average, a company in the most financially dependent decile, ceteris paribus, employs 1.48 per cent fewer people during the crisis than a company in the least financially dependent quartile. in terms of the dimensions of the other coefficients that we are considering, this effect is quite relevant (although in the specification proposed the effect fails to be significant at the 10 per cent level); however, coefficients for the deciles do not show a homogeneous pattern. 4.3. other explanatory variables: sectorial competition apart from the variables in the main model specification, i have also derived the same model including a measure of sectorial competition following aghion et al. (2005), calculated for european data in the period 2004 to 2007 to avoid endogeneity issues, as earlier. results are in appendix a (table a.3). the result of this coefficient is uniformly positive, significant overall and for large firms, but not for the smaller ones. this indicates that firms in sectors with a higher level of competition, overall, have performed better since the onset of the financial crisis than sectors with a lower level of competition. this evidence is consistent with the “escape competition” effect as described by aghion et al. (2005): higher sectorial competition (in the context of the current crisis) implies that companies are more incentivised to innovate to escape competition. this implies, in terms of the framework in this paper, that they grow more on average. 4.4. cross-country interaction terms in order to perform a country-level analysis of how the crisis affected the overall level of productivity, i have repeated the panel data fixed-effect estimation with log specification, as described above, including country-variable interaction terms. there are only three eu-15 countries for which i do not report coefficients: greece, netherlands and denmark. this is due to the fact that companies in these countries did not fulfil the requirements set in terms of availability of data (particularly, of value added). for ease of interpretation of coefficients, they do not reflect country differences with respect to a benchmark country, but the effect that each of the variables had in each specific country. the results are summarised (in terms of signs, by country and company size) in table 6 below and reported in detail in appendix a (tables a.4, a.5, a.6). paz pardo ● productivity in europe during the great recession 96 table 6. cross-country effects, signs specification productivity indebtedness rz country all l s all l s all l s austria + + belgium + + + finland + + + france + + + germany + + + + + ireland + + + + italy + + + luxembourg portugal + + + + + spain + + + + sweden + + + uk + + + this table summarises the results of tables a4, a5 and a6. l are large firms and s are small firms, as defined in general for this section. + indicates positive significance and – indicates negative significance. no sign indicates no significance. these results show, first of all, that there is in general a positive evidence for h1 (“creative destruction hypothesis”) in most of the european countries, and particularly in all of the larger ones, where the coefficient is significantly positive regardless of company size. however, magnitudes are quite different. for instance, considering the results for all companies, in spain a much more productive company (€ 100,000 constant 2005 per worker per year more) is ceteris paribus related with a number of workers nearly 3 per cent higher than a less productive one (in france the value is similar). in the spanish case, as pointed out earlier, flows out of the construction sector can be a good part of the explanation for the relevance of this phenomenon. on the other hand, the percentage drops to 0.7 in the case of the uk or germany, countries in which labour productivity did not improve significantly (germany) or decreased (uk) since the beginning of the crisis. the french case is more puzzling, since it shows a strong effect of creative destruction, while labour productivity performance did not improve much during the crisis period. in ireland creative destruction was only prevalent for large firms. for the overall sample of firms, the effect fails to be significant at 10 per cent level. portugal, another country that was particularly affected by the crisis, shows a very small coefficient. in terms of magnitude, it is around 20-30 times lower than the spanish, even lower than the german. however, overall portuguese labour productivity did increase; these two observations together seem to indicate that this productivity improvement took place inside firms (possibly through reduction of the number of workers, particularly in temporal contracts, that were the main source of adjustment of employment in portugal (carneiro, portugal and varejão 2014). this effect is likely to have happened as well in other european countries. second, it is clear that the effect of finance dependence (as measured by the rz_eur coefficient) is of particular significance in dimension in southern european countries like spain or portugal, that suffered trouble in their financial systems (and thus companies had to face higher screening costs) and a sovereign debt crisis (that generated further tensions in the financial system, increasing borrowing costs). in the same line, the coefficient is also negative for large firms in italy and belgium, countries that endured similar situations. european journal of government and economics 5(2) 97 this is also the case of the uk (for both subsamples of companies) and might be one of the relevant reasons that could explain the decrease in overall labour productivity. the negative coefficient on the rz measure for the united kingdom is the largest amongst all the studied countries, reflecting a particularly negative effect of finance dependence for british companies. in all of these cases the magnitude of the rz coefficient is similar to the one analysed earlier for large firms at an eu-15 level. third, the overall negative effect of indebtedness, which i have analysed to be the result of two very different effects (positive for large companies and negative for smaller ones), is less homogeneous. it seems that in countries like france or belgium the negative effect of debt is significant, while the arguments for the positive effect of debt seem to apply particularly to portugal, germany, finland or ireland. the positive effect of debt for large companies is shown to be much more consistent across countries than the negative effect for small companies is. the only exception for this is france, where coefficients are significantly negative for all company sizes. on the other hand, germany and portugal do show a positive effect of debt for all company sizes, which is particularly striking in the portuguese case, where the dimension of the coefficient is considerable (an increase of 1 percentage point on indebtedness is related with having a 3 per cent higher number of workers after the crisis). in contrast with the significantly negative effect of finance dependence for the uk, indebtedness fails to be a relevant variable in order to explain the performance of british companies in the post-crisis period. 5. conclusions i have studied the main determinants of company performance in the context of the current financial crisis in order to understand the transmission channels through which it impacted on productivity. the results i obtain explain some of the remarkable findings in the stylised facts. spain, the country where productivity improved the most, was the one with the largest effect of “creative destruction” (even though finance also played a role). the uk, on the other hand, where productivity actually decreased, shows the largest effect of finance dependence on the evolution of companies. policy implications are, therefore, heterogeneous amongst countries and companies. on the one hand, the existence of a consistent creative destruction effect (even though limited in dimension) suggests that recessions do help to reorganise the economy. this would imply that policy intervention to support companies during a recession would be negative in the long run. on the other hand, however, financially dependent companies suffer more during a recession, independently of their indebtedness level. this provides a further justification for policy intervention to stabilise the financial system or to help financially dependent companies by offering them the loans they cannot secure in the private sector. moreover, results show that fostering sectorial competition would help companies to survive in the context of a crisis. with regards to indebtedness, the effect is heterogeneous and dependent on company size. while so-called “discipline of debt” remains a good explanation in theory, the fact that indebtedness is a positive predictor for company performance during a crisis only if the company is large might be pointing out that indebtedness is proxying other unobservable company paz pardo ● productivity in europe during the great recession 98 characteristics (good relations with financial sector, bargaining power…) that can also be very related with resistance in a crisis context. there are many future possibilities for research on this topic. following the same methodological approach, other explanatory variables could be included in the analysis (both at a sectorial and company level) of post-crisis performance determinants. this could offer a solution for some of the countries whose evolution is not well explained by this framework (france or ireland, for instance). an aspect that has not been considered is the analysis of opening and closing companies, where part of the creative destruction process is taking place. it would also be informative to repeat this kind of analysis for other recessions or financial crises, be it in a different time frame or in a different geographical area. this would help to compare the current crisis with other past recessions and to draw conclusions with more general validity. references aghion, philippe, nicholas bloom, richard blundell, rachel griffith and peter howitt (2005) ‘competition and innovation: an inverted-u relationship’, the quarterly journal of economics 120(2): 701-728. aghion, philippe and peter howitt (1998) endogenous growth theory. cambridge: mit press. aghion, philippe and peter howitt (2005) ‘growth with quality-improving innovations: an integrated framework’, in philippe aghion and steven n. durlauf (eds) handbook of economic growth, vol 1. amsterdam: elsevier, pp. 67-110. aghion, philippe and peter howitt (2009). the economics of growth. cambridge: mit press. aghion philippe, ufuk akcigit and peter howitt (2014) ‘what do we learn from schumpeterian growth theory?’, in philippe aghion and steven n. durlauf (eds) handbook of economic growth, vol 1. amsterdam: elsevier, pp. 515-563. aghion philippe, ufuk akcigit and peter howitt (2015a), ‘the schumpeterian growth paradigm’, annual review of economics 7(1): 557-575 aghion philippe, ufuk akcigit and peter howitt (2015b) ‘lessons from schumpeterian growth theory’, american economic review 105(5): 94-99. aghion, philippe, david hemous and enisse kharroubi (2014) ‘credit constraints, cyclical fiscal policy and industry growth’, journal of monetary economics 62: 4158. aghion, philippe and gilles saint-paul (1998) ‘virtues of bad times: interacction between productivity growth and economic fluctuations’, macroeconomic dynamics 2(3): 322-344. barlevy, gadi (2002) ‘the sullying effect of recessions’, the review of economic studies 69(1): 65-96. barlevy, gadi (2003) ‘credit market frictions and the allocation of resources over the business cycle’, journal of monetary economics 50(8): 1795-1818. blalock, garrick, paul j. gertler and david i. levine (2008) ‘financial constraints on investment in an emerging market crisis’, journal of monetary economics 55(3): 568-591. blanchard, olivier (2004) ‘the economic future of europe’, journal of economic perspectives 18(4): 3-26. burnside, craig, martin eichenbaum and sergio rebelo (1993) ‘labour hoarding and the business cycle’, journal of political economy 101(2): 245-273. european journal of government and economics 5(2) 99 caballero, ricardo j. and mohamad l. hammour (1994) ‘the cleansing effect of recessions’, the american economic review 84(5): 1350-1368. caballero, ricardo j. and mohamad l. hammour (2005) ‘the cost of recessions revisited: a reverse-liquidationist view’, review of economic studies 72(2): 313341. carneiro, anablea, pedro portugal and josé varejão (2014) ‘catastrophic job destruction during the portuguese economic crisis’, journal of macroeconomics 39: 444-457. davis, steve j. and john haltiwanger (1992) ‘gross job creation, gross job destruction and employment reallocation’, quarterly journal of economics 107(3): 817-863. dell’ariccia, giovanni, enrica detragiache and raghuram rajan (2008) ‘the real effect of banking crises’, journal of financial intermediation 17(1): 89-112. galí, jordi and mohamad l. hammour (1991) ‘long run effects of business cycles’, mimeo. new york: columbia university graduate school of business. hall, robert e. (2000) ‘reorganization’, in carnegie-rochester conference series on public policy 52: 1-22. kleemann, michael and manuel wiegand (2013) ‘the causal effect of restrictive bank lending on employment growth: a matching approach’, mpra paper no 43529. kroszner, randall s., luc laeven and daniela klingebiel (2007) ‘banking crises, financial dependence and growth’, journal of financial economics 84(1):187-228. mortensen, dale t. and christopher a. pissarides (1994) ‘job creation and job destruction in the theory of unemployment’, review of economic studies 61(3): 397-415 nickell, stephen, sushil wadhwani and martin wall (1992) ‘productivity growth in u.k. companies, 1975-1986’, european economic review 36(5): 1055-1091 ouyang, min (2009) ‘the scarring effect of recessions’, journal of monetary economics 56(2): 184-199. rajan, raghuram. and luigi zingales (1998) ‘financial dependence and growth’, american economic review 88(3): 559-586.. schumpeter, joseph a. (1939) business cycles: a theoretical, historical and statistical analysis of the capitalist process. new york: mcgraw hill. schumpeter, joseph a. (1942) capitalism, socialism and democracy. new york: harper and brothers. paz pardo ● productivity in europe during the great recession 100 appendix a. tables. table a.1. other functional specifications for the main regression: rz coefficient, absolute values. specification (dependent variable: log number of employees) all companies small companies larger companies all companies (considering absolute value of employees) measure rz rz rz rz_eur pre-crisis productivity * crisis 0.000787*** (0.00002) 0.000191 (0.0002) 0.0000688*** (0.00001) 0.0206754*** (0.008) rz external dependence * crisis 0.038832*** (0.005) 0.0375329*** (0.006) 0.0373786*** (0.01) -0.021964 (0.125) pre-crisis indebtedness * crisis 0.0004638*** (0.0001) 0.0003801*** (0.0001) 0.0005149*** (0.0002) -0.0004457 (0.0007) year 2005 0.233758*** (0.002) 0.0227549*** (0.002) 0.0274249*** (0.005) 5.175511** (2.61) year 2006 0.0395673*** (0.002) 0.0366141*** (0.002) 0.0522684*** (0.006) 15.6772*** (3.84) year 2007 0.0780634*** (0.002) 0.0726387*** (0.003) 0.1027773*** (0.006) 26.59072*** (4.232) year 2008 0.0908126*** (0.003) 0.0819294*** (0.01) 0.1049978*** (0.008) 34.40845*** (4.65) year 2009 0.0654195*** (0.003) 0.0556513*** (0.01) 0.0838619*** (0.008) 32.36799*** (4.95) year 2010 0.0758409*** (0.003) 0.0659838*** (0.01) 0.0948655*** (0.008) 35.18816*** (5.05) year 2011 0.0776454*** (0.003) 0.0676491*** (0.01) 0.0973285*** (0.009) 41.15221*** (5.338) year 2012 0.0661422*** (0.004) 0.0545099*** (0.01) 0.0935452*** (0.009) 45.34283*** (5.95) year 2013 0.0708986*** (0.004) 0.0606266*** (0.01) 0.0914669*** (0.01) 58.79747*** (8.65) constant 3.812305*** (0.002) 3.425448*** (0.002) 5.657646*** (0.006) 299.6515*** (4.20) observations 448156 (52684 firms) 370296 (43381 firms) 77369 (9245 firms) 1096350 (130320 firms) f 252.18*** 219.73*** 43.20*** 10.82*** r2 0.0232 0.0185 0.0471 0.0006 *** significant to 1% level, ** significant to 5% level *significant to 10% level. special caution must be taken with 2013 coefficients, since data availability for that year is limited and in some cases provisional. table a.2. results with decile dummies. specification (dependent variable: log number of employees) large companies measure rz_eur pre-crisis productivity * crisis 0.0001013*** (0.00003) pre-crisis indebtedness * crisis 0.0000345*** (5e-6) decile 2 -0.0454783*** (0.015) decile 3 -0.0278186** (0.013) decile 4 -0.0280597 (0.017) decile 5 -0.0013969 (0.014) decile 6 -0.0243156* (0.014) decile 7 0.0141897 (0.016) decile 8 0.0045195 (0.013) decile 9 -0.0032661 (0.015) decile 10 -0.0147987(0.016) constant 5.491495*** (0.004) observations 195345 (23799 firms) f 141.71*** r2 0.0681 *** significant to 1% level, ** significant to 5% level *significant to 10% level. year fixed effects are included but not reported. deciles are ordered from less financial dependence to more financial dependence, with coefficients being relative towards the lowest decile (least financially-dependent companies) european journal of government and economics 5(2) 101 table a.3. results including sectorial competition specification (dependent variable: log number of employees) all companies small companies larger companies pre-crisis productivity * crisis 0.0000912*** (0.00002) 0.0000826*** (0.00002) 0.0001012*** (0.00003) rz external dependence * crisis -0.0001844*** (0.00005) -0.0001422*** (0.0000519) -0.000379*** (0.00008) pre-crisis indebtedness * crisis -0.0000034** (1.44e-6) -0.00000357*** (1e-6) 0.0000344*** (5e-6) sectorial competition* crisis 0.0000425*** (0.00001) 0.0000233 (0.00002) 0.0001329*** (0.00003) constant 3.614075*** (0.0016) 3.20522*** (0.0018) 5.491749*** (0.004) observations 1095627 (130291 firms) 899260 (106367 firms) 195297 (23793 firms) f 1141.66*** 942.10*** 214.44*** r2 0.0464 0.0418 0.0678 *** significant to 1% level, ** significant to 5% level *significant to 10% level. year fixed effects are included but not reported table a.4. cross-country effects (all companies) specification all companies country productivity*crisis indebtedness*crisis rz_eur * crisis austria 0.0000608 (0.00006) 0.0051556 (0.003) 0.0025071 (0.0033) belgium 0.0000571** (0.00002) -0.000003*** (9.9e-7) -0.0002017 (0.0002) finland 0.0001785 (0.0001) 0.0019087** (0.0008) -0.0002199 (0.003) france 0.0002657*** (0.00009) -0.0000399*** (0.0001) -0.0004387 (0.0004) germany 0.00007*** (0.00001) 0.0001227*** (0.00004) -0.0001074 (0.00008) ireland 0.0000628 (0.00005) 0.00000676* (3.7e-6) 0.0003986 (0.0014) italy 0.0002119*** (0.00005) 0.000127 (0.00008) -0.0000855 (0.00006) luxembourg -0.0000214 (0.000307) 0.003082 (0.001) 0.008684 (0.009) portugal 0.0000115* (6.56e-6) 0.0003118*** (0.00007) -0.0002609** (0.0001) spain 0.0002921*** (0.00005) 0.00000357 (3e-6) -0.0002309** (0.00009) sweden 0.0005777*** (0.0002) 0.0000736 (0.0001) -0.0001457 (0.0002) united kingdom 0.0000659*** (0.00001) 0.0000306 (0.00003) -0.0003876*** (0.0001) *** significant to 1% level, ** significant to 5% level *significant to 10% level. year fixed effects are included but not reported. sample: all companies >10 workers. paz pardo ● productivity in europe during the great recession 102 table a.5. cross-country effects (large companies) specification large companies (>150 employees) country productivity*crisis indebtedness*crisis rz_eur * crisis austria -0.0000993* (0.00006) 0.098242*** (0.014) -0.0018293 (0.004) belgium -0.000113 (0.00015) 0.0010556*** (0.0004) -0.0007397*** (0.0002) finland -0.000084 (0.00006) 0.003328*** (0.0002) 0.0002542 (0.0005) france 0.0005744*** (0.00017) -0.000038*** (5e-6) -0.0023801 (0.002) germany 0.0000743*** (0.000017) 0.0003514** (0.0002) -0.0002428* (0.0001) ireland 0.0003226*** (0.0001) -0.0004037* (0.0002) 0.0014643* (0.0009) italy 0.0001381*** (0.00004) 0.0004487 (0.0004) -0.0003791** (0.0002) luxembourg -0.0002139 (0.0002) -0.0138419 (0.033) -0.000368 (0.0058) portugal -0.0002336 (0.0002) 0.0024675*** (0.0003) -0.0004822*** (0.0002) spain 0.0003826*** (0.00014) 0.0000326*** (2e-6) -0.0002151 (0.00024) sweden 0.0007177*** (0.0002) 0.0042386 (0.003) -0.0008262 (0.0007) united kingdom 0.0001954*** (0.00008) 0.0000221 (0.00002) -0.0003483* (0.00021) *** significant to 1% level, ** significant to 5% level *significant to 10% level. year fixed effects are included but not reported european journal of government and economics 5(2) 103 table a.6. cross-country effects (small companies) specification small companies (<150 employees) country productivity*crisis indebtedness*crisis rz_eur * crisis austria 0.0002439** (0.0001) 0.0017293 (0.0016) -0.000216** (0.000097) belgium 0.0000592** (0.00002) -0.00000322*** (1e-6) 0.0000604 (0.0002) finland 0.0004548*** (0.00015) -0.0004037 (0.0005) -0.0004428 (0.00034) france 0.000163** (0.000074) -0.0000251** (0.00001) -0.0003021 (0.0004) germany 0.0000435 (0.00003) 0.0001027** (0.00004) -0.0000287 (0.0001) ireland 0.0000237 (0.00002) 0.0000093*** (2.5e-6) -0.0064138 (0.004) italy 0.0003906*** (0.00014) 0.0000997 (0.00007) -0.0000712 (0.000062) luxembourg -0.0000349 (0.00004) 0.0004389 (0.00097) 0.0160682 (0.014) portugal 0.000012* (7e-6) 0.000268*** (0.00003) -0.0001917 (0.00013) spain 0.0002647*** (0.00005) 0.000000605 (1e-6) -0.000216** (0.0000964) sweden 0.0005633*** (0.00017) 0.0000645 (0.0001) -0.0000766 (0.000246) united kingdom 0.0000554*** (5e-6) -0.00000197 (0.00001) -0.0003854*** (0.0002) *** significant to 1% level, ** significant to 5% level *significant to 10% level. year fixed effects are included but not reported. only companies with at least 10 employees are included. microsoft word ejge_03_02_014.doc european journal of government and economics volume 3, number 2 (december 2014) issn: 2254-7088 119 what drives sub-national bioenergy development? exploring cross-level implications of environmental policy integration in eu and swedish bioenergy policy charlotta söderberg, luleå university of technology, sweden abstract what are the sub-national implications, in policy and practice, of environmental policy integration (epi) in eu and swedish bioenergy policy? focusing on the exceptional bioenergy expansion within the biofuel region in north sweden, this paper discusses cross-level implications of supranational and national policy decisions on bioenergy; whether environmental perspectives are observable also in sub-national bioenergy discussions; and explores the drivers of sub-national bioenergy development in a multi-level governance setting. the study finds that higher-level epi plays an important role for sub-national bioenergy development. the degree of sub-national epi in bioenergy and the type of renewables invested in is to a large extent set by top-down influence from the eu and national level through agenda setting, policy goals and economic mechanisms. local policy entrepreneurs play an important role for finding ‘win-win’-solutions that can help initiating local energy projects and ensure sub-national epi, but environmentaleconomic – rather than merely economic – motives for getting involved are important to ensure long-term local commitment to renewable energy projects. jel classification keywords bioenergy; sub-national; cross-level epi; regulation; sweden; multi-level governance. european journal of government and economics 3(2) 120 introduction this paper focuses on cross-level implications of supranational and national bioenergy policy for sub-national bioenergy development. bioenergy is energy (electricity and transport fuels) derived from biomass (i.e. by-products from forestry and agriculture, municipal and industrial waste streams and dedicated energy crop cultivation, ec, 2005; eea, 2006). through studying the effect of eu and swedish bioenergy policy on the developments in the biofuel region area in sweden, which is the site for a unique expansion in the biofuel industry over the last decade, this paper provides insight into what works, when and how for stimulating sub-national renewable energy development. overarching policies are formed and policy instruments are instated on the supranational and national levels. one overarching policy objective that has influenced eu and swedish bioenergy policy for the last decades is environmental policy integration (epi), which endeavours to incorporate environmental policy objectives into all sector policies with a view to promote sustainable development (wced 1987). epi is a key objective within euand member state policy (under article 6.1. of the lisbon treaty). there is a rich literature available on epi, to a large extent focusing on the european and national levels (e.g. lenschow, 2002; nilsson and eckerberg, 2007; jordan and lenschow, 2008; lafferty, 2004). however, sub-national actors are the ones who, in the end, are left with the delicate task of interpreting and realising higher-level policies. in other words, the sub-national level is the level on which policy comes to life; where policy moves from rhetoric to practice. thus, the sub-national level can be said to hold the key to the realisation of sustainable development ambitions, such as renewable energy expansion. however, as urwin and jordan (2008, p. 182, based on radaelli, 2003) argue: “the way in which the intentions of policy are expressed at the higher level, can be very different to the way in which it is actually conducted (and thus interacts with cognate policies) at lower levels”. thus, in order to fully understand the process, and results, of environmental policy integration it is highly relevant to start scrutinising cross-level implications of epi; not only on the supranational and national levels, but also on the sub-national level. therefore, the overarching aim of this article is to explore the following question: what are the sub-national implications, in policy and practice, of epi in eu and swedish bioenergy policy? how do efforts to promote epi at eu and national levels influence the development within bioenergy investments at subnational level (or is it rather the sub-national level that affects epi in eu and national policy)? apart from shedding light on cross-level epi, the particular focus on the field of sub-national bioenergy expansion provides insights into the crosslevel implications of eu and swedish bioenergy policy, and into what drives subnational bioenergy development. the theoretical and empirical focus of the study is further developed below in this section. the second section provides a theoretical background, where key concepts are discussed and defined and an analytical framework and methodology for the study of sub-national epi is outlined. the sub-national development and framing of bioenergy is explored in the third section. then, the cross-level influence of european and swedish bioenergy policy on the development within the biofuel region is discussed in the fourth section. finally, conclusions and policy implications are discussed. a cross-level analysis focusing on the biofuel region in sweden with regard to previous studies on cross-level implications of environmental policy, there are general studies on regional governance for sustainable development (e.g. eckerberg and forsberg, 1998; lafferty and narodoslawsky, 2003), and on governance for sustainable development at the inter-subnational level (happaerts söderberg ● what drives sub-national bioenergy development? 121 et al., 2010). as regards epi on the sub-national level, goria et al (2010) maps out current research with an overarching perspective on epi on different levels – including the sub-national level. however, the focus in this paper – i.e. the subnational policy-making arena and the interplay between the three different policy levels (eu, national, sub-national) in specific policy areas – are still virtually black areas within our understanding of epi. according to jordan and lenschow (2008, 2010), three main analytical dimensions of epi exist: 1) an institutional perspective, focusing on epi as a policy coordination problem; 2) a political perspective, focusing on epi as a problem of political conflict and lobbying; 3) a cognitive perspective, focusing on different actor interests, how they are embedded within ideational frames, and how cognitive learning occurs. in this paper, a cognitive, ‘reframing’ approach to epi is taken. accordingly, epi is viewed as a process where policymakers and actors in non-environmental policy sectors gain environmental knowledge, reframe their view of a certain policy area/issue and change their actions accordingly. sweden has been awarded high epi ‘scores’ with this approach to epi (nilsson and persson, 2003; swartling et al 2007; nilsson and eckerberg, 2007). sweden is also given high epi-scores in other types of epi-analyses focusing on the available environmental management systems in a country (e.g. jacob and volkery, 2004). the overall high epi scores, together with the country’s reputation of being a forerunner in environmental policy (lundqvist, 2001, 2004; kronsell, 2006) and the fact that sweden can be considered a “critical most-likely case for epi” (nilsson, 2005a, p. 16) makes sweden a relevant case for exploring cross-level implications of epi. bioenergy policy is a policy area where epi has been observed on both the eu and swedish policy levels over the last decades. according to previous studies, the framing of bioenergy in euand swedish policy today includes an environmental perspective (söderberg, 2008; 2011a; 2011b; söderberg and eckerberg, 2013). this makes the sub-national swedish level a relevant case for exploring cross-level epi in bioenergy policy. since the eu, its member states and the sub-national entities constitute a multi-level-governance system, the sub-national interpretation of-, influence by-, and contribution to nationaland eu-policies is essential to study. two subsets of questions guide the analysis. first, given that we know how bioenergy policy is framed on eu and swedish levels: how do bioenergy actors in a sub-national swedish setting frame bioenergy? through analysing the sub-national framing of bioenergy over time, the driving forces for sub-national bioenergy development can be detected. are there similarities/differences between the bioenergy frames on different levels and are environmental issues integrated also on the sub-national level? second, how can sub-national bioenergy development be explained in a multi-level setting? through comparing the framing of bioenergy on different levels, the direction of policy impulses can be indicated – what policy level reframes policy first – eu, swedish, or sub-national? furthermore, interviews with key biofuel region actors provide insight into what role eu and swedish bioenergy policy, respectively, play for sub-national bioenergy development. the case of sub-national bioenergy expansion in focus of this study is the biofuel region in sweden, which is a strategic network uniting public and private actors such as municipalities, national agencies, researchers and business interests in the four northernmost swedish counties. the biofuel region was initiated in 2003 “with a view to creating and managing the development of an entirely new industry centred on renewable fuels from cellulose-based raw materials obtained from forest, field and recycling sources” (christensen, 2005, p. 18) and was then mainly situated in two counties in the north of sweden, but has expanded geographically as the biofuel interest has grown over the past seven years. in terms of visible results, there were nine gas stations within the biofuel region offering e85 (bioethanol) in 2003, and the demand on ‘green cars’ was small (bfr, 2006). in european journal of government and economics 3(2) 122 2009, there were 75 gas stations offering e85 (bioethanol) and two gas stations providing biogas. four percent of all cars within the biofuel region in 2009 and 33 percent of the new cars sold within the biofuel region in 2009 were classified as ‘green cars’ (bfr, 2009). in 2010, four production sites for biogas, one for dme (dimethyl ether) and pine diesel and one for bioethanol were running in the region. this is a unique expansion in sweden. the biofuel region originally focused on developing production of cellulose-based biofuels (mainly bioethanol and biodiesel), but has since then broadened its scope to include other types of biofuels, such as biogas. therefore, the development within the biofuel region, besides providing insight into sub-national bioenergy development, may also have something to tell us about the driving forces behind choosing to promote certain types of energy sources on the sub-national level. the biofuel region is a pioneering case in terms of regional biofuel development in sweden; it is a famous project, which has been running for a decade. these characteristics make the region highly relevant to study in the context of subnational bioenergy development and cross-level epi. there is no other comparable case on the sub-national level in sweden. focus in the study lies on exploring the driving forces behind this development, whether environmental policy integration on the eu and swedish levels has any impact on the sub-national level development, and whether the biofuel region has any impact on the development at the eu and swedish levels. analytical framework and methodology here, the theoretical background is further developed and key concepts such as epi, reframing, policy learning and multi-level governance are discussed and defined. the methodology of the study is also addressed here. the first part of this section outlines the analytical framework and methodology for answering the first central question of the study, while the second part outlines the analytical framework and methodology for answering the second central question of the study. a policy learning approach to epi the first central question explored here is: how do bioenergy actors in a subnational setting frame bioenergy? given that we know that epi has occurred on the euand swedish level in bioenergy policy, it is relevant to explore the driving forces for sub-national bioenergy development. since the biofuel region area is the site for a unique expansion in the biofuel industry over the last decade, this analysis will provide insight into what works, when and how for stimulating subnational renewable energy development. furthermore, through exploring the framing of bioenergy it is also possible to discern whether the sub-national framing of bioenergy qualifies as epi. to discern integration of environmental aspects in this study, epi is viewed as a policy learning process (söderberg, 2011b). learning can be studied on the individual or collective level (kemp and weehuizen, 2005). collective learning, in turn, can be studied as organisational learning (e.g. argyris and schön, 1978) or, in broader contexts, as social learning (reed et al, 2010, nilsson and swartling, 2009). here, learning is studied on the collective level, with focus on a specific type of collective learning: policy learning (kemp and weehuizen, 2005), which “implies a change in thought about policy, which subsequently contributes to a change in the policy process” (swartling et al 2007, p. 50). sabatier (1993, p. 19) defines policy learning as “a relatively enduring alteration of thought or behavioural intentions that are concerned with the attainment (or revision) of the precepts of a policy belief system”. thus, epi is detected through looking for policy learning, inducing policy change. söderberg ● what drives sub-national bioenergy development? 123 the main tool for tracing policy learning here is studying how policy is framed over time. frames can be said to shape actors’ conception of reality, but when actors are provided with new knowledge (through e.g. research reports, media attention or lobbying) this can incite a learning process, which may lead to reframing. when such a reframing towards environmental sustainability occurs, environmental policy integration can be detected (rein and schön, 1993; 1994; nilsson, 2005; hall, 1993). however, the reframing must be visible in both policy rhetoric and in motives for policy measures to qualify as epi, which here is defined as in söderberg (2008), where environmental rhetoric in combination with environmental considerations underpinning policy practice (i.e., policy strategies are adopted on environmental grounds) constitute indicators for epi (see also nilsson and persson, 2003). there are also other types of learning, which may entail environmental reframing without qualifying as epi; if environmental concerns are visible only in policy rhetoric but not followed-up in political measures it qualifies as rhetorical environmental learning. if environmental concerns are visible only in policy instruments (for example, as a response to regulatory requirements) it qualifies as instrumental environmental learning (nilsson, 2005). epi, on the other hand, thus requires a combination of both types of environmental learning, which signals a deeper entrenchment of environmental issues in both policy rhetoric and instruments. the view on epi adopted here – environmental policy integration as a learning process traced through studying policy frames– has been developed and tested in previous research. for more elaborate discussions on the framing approach to epi, please see lenschow and zito (1998), nilsson and eckerberg (2007) and söderberg (2011b). but is it enough that environmental aspects are included in policy rhetoric and practice in order to qualify as epi? that depends on how rigidly we interpret the epi concept. there are on-going discussions within epi research regarding what priority has to be awarded to environmental issues in order to qualify as epi (see e.g. lafferty and hovden, 2003; baker, 2007). in an overview of different interpretations of the ‘e’ (environmental) and the ‘i’ (integration) in epi, söderberg (2011b, p. 26) concludes that “it is possible to interpret the ‘e’ either from an anthropocentric worldview, aiming for ecological modernisation/the three-tiered sustainable development concept; or from an ecocentric worldview, aiming for ecologically sustainable development through significant value and lifestyle changes. similarly, the meaning of ‘i’ can be interpreted either as “adding environment and stir”, i.e. paying consideration to the environment in sector policymaking; or as giving principled and consequential priority to environmental issues”. thus, based on whether the environment is prioritised in biofuel region policy documents (‘strong epi’), or if it is mainly considered together with other types of issues (‘weak epi’), the strength of the integration of environmental issues within biofuel region can be evaluated. the analysed materials consist of five biofuel region reports of activities (for the years 2003-2010, available on the biofuel region webpage), printouts from ten semi-structured telephone interviews conducted in 2010 with all biofuel region board members (representing small and large municipalities; industries; research; and central administrations)1, and a study from 2005 on the development of the 1 the interview questions were sent out to the interviewees before the interview. the interview questions focused on four main themes. 1) purpose: explain the purpose of bfr; why did your organisation choose to join the bfr; what explains the initial focus on ethanol; what explains the broadened area of interest and geographical areas? 2) underlying problems: what problems does bfr aim to help solving; what problems does bfr encounter; on what grounds do municipalities choose to support the bfr; on what grounds have municipalities left the bfr? 3) policy networks: what actors does your organisation cooperate with regarding bioenergy issues; how do national/eu policy/regulations affect your bioenergy work; how does the bfr affect national/eu politics? 4) political european journal of government and economics 3(2) 124 biofuel region (christensen, 2005). the bioenergy policy frame(s) present within the biofuel region is detected through idea analysis (bergström and boreus 2005), analyzing the selected material on the basis of four thematic questions: policy goals; what are the overarching policy goals for the biofuel region? underlying problems; what are the main problems that the biofuel region is to address, and what problems does the biofuel region face? policy preferences (instruments and strategies); what policy instruments is the biofuel region promoting, and how are the biofuel region’s goals to be achieved? promoted values and role of bioenergy; what values are promoted, and what is the role of bioenergy? the article examines how these four themes have been articulated in the biofuel region between the years 2003-2010 to detect the policy frames guiding the biofuel region. what are the driving forces for sub-national bioenergy development; has it changed over time; and (if so) what may explain these shifts? sub-national bioenergy development in a multi-level governance setting the second central question analysed here is: how can sub-national bioenergy development be explained in a multi-level setting? what role do eu and swedish bioenergy policies, respectively, play for bioenergy development according to subnational bioenergy actors? this question has to do with the multi-level character of the eu system, and the fact that if sustainable development is to be achieved in practice; epi must occur on all levels (e.g. stagle, 2007). the increasing interdependence between different policy levels has been analysed within multilevel governance (mlg) theory (hooghe and marks, 2004). bache and flinders (2004) derive four common understandings of mlg: increased participation in policy making by non-state actors; that distinct decision making levels are becoming more difficult to discern; a new role of the state in this new environment; and a new decision making context which makes it necessary to rethink democratic accountability. in this context, according to oberthür and gehring (2006), the effectiveness of environmental governance instruments, such as european or national environmental instruments, is affected by interinstitutional influence from other regimes/policy instruments. since realisation of bioenergy policy goals within the eu thus depends on a multi-level governance (mlg) system, this article explores the impact from european and swedish bioenergy policy on the sub-national development within the biofuel region in order to analyse whether efforts to promote epi at supra-national and national levels influence the development within a policy area at sub-national level (and vice versa). to this aim, the biofuel region actors were encouraged in the interviews (for details, please see endnote 1) to elaborate on to what extent they experience that eu bioenergy policy and swedish bioenergy influence the development of bioenergy in the biofuel region area (in order to explore the top-down influence between bioenergy policy levels). the biofuel region actors were also asked whether they deem that they have any influence on swedish and european bioenergy policy development (in order to explore the bottom-up influence between bioenergy policy levels). how can the exceptional biofuel expansion within the biofuel region be explained? in order to start analysing this question we need to dwell a little on what institutional factors are put forward as conducive to epi and local sustainable development in previous research. policy learning is put forward as an important feature of sustainable development by many authors (e.g. folke et al., 2002; gunderson and holling, 2002; nilsson and eckerberg, 2007). stagle (2007) concludes after a preferences: how does the bfr work to achieve their goals; what policy instruments are advocated; does different arguments work differently on different actors? söderberg ● what drives sub-national bioenergy development? 125 thorough review of different learning theories that “the special challenge for learning for sustainable development is…to help establish procedures, management practices and institutions for public decision-making that support learning processes on multiple scales”(p. 59). therefore, the interviewees were also asked to map out the cross-level contacts, which the biofuel region is involved in. nilsson and eckerberg (2007) identify participant’s trust in that they benefit from cooperation; development of trust and joint-problem perceptions; alignment of reframing with short-term sector interests; and strong political leadership as important factors for cognitive epi. furthermore, they maintain the importance of ‘issue-champions’ and issue networks within sectors to make sure that sustainability issues are receiving attention throughout the policy cycle. previous epi-studies have also pointed to the relevance of external factors (such as media attention), policy windows and ‘policy entrepreneurs’ (kingdon, 1995) for epi development. the role of win-win situations and political and individual leadership for local sustainable development is also supported by eckerberg and forsberg’s (1998) study of local agenda 21 processes in sweden, which shows that the combination of policy entrepreneurs and the possibility to solve local structural problems often spur successful local sustainable development strategies. to address this issue, interviewees were asked about the driving forces for joining the biofuel region. given that also structural factors are important for understanding epi, as ‘bridges’ between environmental issues and other, economic, policy areas – that is, environment-economy win-win situations – are conducive for epi (nilsson and eckerberg, 2007; söderberg, 2008), it is relevant to map out what environmenteconomy win-win situations have been important for the development of the biofuel region. however, when doing so, it is important to pay attention to the underlying motivations for the renewable energy project in the biofuel region, since previous epi studies show that renewable energy projects initiated on environmental grounds are more durable and long-term than renewable energy projects initiated for other reasons (such as economic growth or supply-security) (söderberg, 2008). in the third section, the framing of bioenergy as well as institutional and structural factors in the biofuel region are explored on basis of the analytical framework structured above, with the aim to understand the driving forces behind the biofuel region, and how the interaction between different governance levels in a multilevel governance structure impacts on sub-national bioenergy development in this particular case. the policy frames guiding the biofuel region are detected through outlining how policy goals; underlying problems; policy preferences (instruments and strategies); and promoted values and role of bioenergy have been articulated in the biofuel region between the years 2003-2010. this part of the analysis draws on both document studies and interviews and serves the purpose of detecting both the driving forces for sub-national bioenergy development over time as well as whether the subnational framing can be classified as epi. results: sub-national bioenergy development and framing of bioenergy the biofuel region is a “strategic network”, uniting public actors, business and research, and was initiated in 2003 by a few municipalities and companies in the north of sweden, mainly situated in the (forest-covered) counties of västerbotten and västernorrland. in 2013, the biofuel region involved 11 municipalities,16 business actors and received financial support from the swedish energy agency, the swedish board of agriculture, and from three (norrbotten, västerbotten, västernorrland) county and regional administrations. the biofuel region works in three main focus areas. societal transition spreads information to sub-national actors and citizens, for example through collaborations with teachers and municipal european journal of government and economics 3(2) 126 authorities. industrial and regional development supports the development of research and production of renewable fuels in the region, e.g. through coordinating biogas producers in north sweden with the aim of making biogas a large-scale biofuel in the region. raw material contains research on energy crops; on intensifying the use of forest in a sustainable manner (within the research project future forest) and on retrieving more energy from the forest (within the research project forest power). the following two subsections explore how the institutional and structural factors (in accordance with the analytical framework) behind the formation and direction of the biofuel region have evolved over time (2003-2006 and 2006-2010). through this approach, institutional and structural explanations to the biofuel interest in the region can be detected. thereafter, the framing of bioenergy over time is outlined in order to explore how relevant environmental aspects are on the sub-national level and if this can be classified as epi. the findings in this section form the basis for cross-level comparison with the national and eu-level framing of and approach to bioenergy the is provided in the next section. institutional and structural factors 2003-2006 christensen (2005, pp. 21-23) provides four main explanations for the launch of the biofuel region: 1) the long tradition of interest in energy issues in the region’s energy and forest industries; 2) the environmental interest uniting the seven key persons in the process management group; 3) the process management group’s good insights into the handling of institutional processes at the national and subnational level; and 4) the eu biofuel directive 2003/30/ec (ec, 2003), obliging member states to ensure that biofuels and other renewable fuels accounted for at least 2 percent of transport fuels by 2005, and at least 5.75 percent by 2010. the biofuel region itself points out two main aspects as initiating factors: first, the increased production of both bioethanoland flexifuel-cars, which had been initiated by baff (bioalcohol fuel foundation) during the first years of the new millennium; second, the county board of västernorrland’s recognition of a strong biofuel cluster as a potential winner of vinnväxt (see below) grants from the swedish governmental agency for innovation systems (vinnova), which supports research and development with the aim to promote “growth and prosperity throughout sweden” (bfr, 2007; vinnova, 2010). vinnova’s vinnväxt programme started in 2001 and aims to “promote sustainable growth by developing internationally competitive research and innovation environments in specific growth fields” (vinnova, 2010a). the västernorrland project, biorefinery of the future, based in biofuel region member cities örnsköldsvik and umeå, develops “biorefineries based on forest raw materials and energy crops” (vinnova 2010b) and received grants (≈€400.000 annually for ten years) in 2008. vinnova’s motivation was that the project had a good potential to create sustainable growth (vinnova, 2010c). the first steps towards forming the biofuel region were taken by a handful of people, attempting to gain national financial support for developing a biodiesel industry in the county of västernorrland in 2002. however, the break-through for the biofuel region came in 2003, when the bioethanol champion and baffchairman per carstedt became process manager. as goals were set for the organisation and municipal politicians saw the opportunity for creating regional growth, the biofuel region grew quickly with focus on bioethanol production. all interviewees explain this ethanol focus with the involvement of the bioalcohol fuel foundation (baff, a foundation aiming to develop production techniques and use of bioethanol within the transport sector) and sekab2 (a swedish producer and 2 sekab: svensk etanolkemi ab. söderberg ● what drives sub-national bioenergy development? 127 distributor of bioethanol and green chemicals, also developing 2nd generation biofuels; cellulose-based bioethanol) in the project. sekab is by far the largest actor on the swedish ethanol market and also an important actor on the european ethanol market (swedwatch, 2009), and the ethanol focus within bfr grew stronger as ethanol cars became more common and as the ethanol distribution system was improved, partly due to national swedish regulations. institutional and structural factors 2006-2010 per carstedt saw business opportunities as baff was involved in building ethanol and flexifuel cars, while other members of the biofuel region, especially the municipalities with rich forest assets, saw risks for a conflict of interest between the wood industry and the ethanol industry regarding raw material. this potential conflict of interest, along with the global ethanol debate that questioned the sustainability of (mainly cereal-based) bioethanol (e.g. giampietro and mayumi, 2009), was behind the broadened focus within the biofuel region after 2007. “biogas as a fuel has gained ground and taken ethanol’s place since biogas is a better fuel. it is less controversial and has better environmental and climate features but is more difficult to commercialise: ethanol is easier” (interview 8). the controversy over ethanol which the interviewee refers to also has to do with another factor behind the broadened focus of the biofuel region, namely sekab’s investments from 2005 and onwards in sugar caneand durra-based ethanol production in africa, mainly tanzania and mozambique, with the intention to build up large-scale (400,000 hectares of energy crop plantations) ethanol production in africa. however, the project received strong criticism early on in a report claiming that sekab’s large-scale plantation plans threatened valuable nature areas and risked crowding out the local farming population (abn, 2007). the project also received criticism regarding the appropriateness of using municipal money (sekab is partly owned by two municipal energy companies, övik energi and skellefteå kraft; and one municipality, umeå kommun) for investments abroad (sekabinvestments had also been made in poland and hungary). these discussions forced sekab to put their investment plans for expansion abroad on hold and seek other investors in 2008 (swedwatch, 2009). the debate on the sustainability of ethanol as a fuel, and sekabs controversial plans of starting up biofuels production in africa, however generated a lot of subnational discussions, and, in combination with the economic recession, the biofuel region saw turbulent times with members leaving. as expressed by one interviewee: “the biofuel region was drawn into the sekab-story” (interview 2). therefore, it became important for the biofuel region to broaden its scope and focus on becoming an alternative fuels region rather than “being just a marketing organisation for sekab” (interview 1) to regain the trust of its members. however, the member loss in the wake of the ethanoland sekab discussions and the economic recession was notable. the biofuel region had 17 municipalities as members in 2008; in autumn 2010 the biofuel region had 11 municipalities as members. it seems that the driving forces were different between those who left the biofuel region and those who endured through the crisis: “when the biofuel region started, some municipalities had unrealistic expectations of increased job growth ... those who stayed entered with more realistic expectations. these municipalities also contain individuals who are committed” (interview 3). the framing of bioenergy underlying problems in a report summarising the first three years of bfrs work, the background to the project is described in global terms. the report refers to the increasing attention to the climate issue with the stern report (stern, 2007) and the ipcc report (ipcc, european journal of government and economics 3(2) 128 2007) and to the ‘peak oil’ theory3 and the increasing oil prices as factors important for the bfr work, and as justification for the strive to reduce the dependence of fossil fuels within the biofuel region (bfr, 2007a). similar values are expressed in interviews: “the oil is gone within the heat and industrial sector, apart from coal and coke use [within the steel industry]. but the transport sector remains, and the biofuel region can help there, to capture the many opportunities available” (interview 6). many of the interviewees point out that the transport sector within the northernmost swedish counties are highly dependent on fossil fuels, and that the underlying problem that bfr can help solving is the transition from fossil fuels to renewable fuels within the transport sector, which globally is driven by the climate change debate (interviews 3, 6, 7, 10). in the initial phase, when the biofuel region was formed, focus was therefore on the north-swedish oil-dependence rather than on the climate change aspect. please note, however, that the climate change focus within bfr has grown over time as the issue has become more generally known by citizens, politicians and sub-national companies. the increasing global climate change debate and the ensuing request for renewable energy has made sub-national actors more open to these issues and to the business opportunities provided by renewable energy production (interview 2). on the regional level, though, the main focus is on growth and municipalities have hesitated to join the biofuel region due to the costs involved. therefore, interviewees hold that the most important problem that needs to be addressed by the biofuel region is the lack of knowledge regarding biofuels amongst local, regional and national politicians, citizens and companies, and to provide an arena for cooperation between municipalities, companies and researchers. within tight budgets, municipal politicians need to prioritise between environmental development issues and the core areas (education, social care and health care), and there seems to be different driving forces for joining the biofuel region project and for withdrawing from the project. for joining the project, environmental arguments are important mainly for convincing municipal politicians that investment in biofuels provides opportunities for job creation and regional growth. however, when leaving the biofuel region, economic reasons are the most important ones; when the process is too slow in creating new jobs, municipalities tend to prioritise other areas. after 2007, however, the municipal knowledge of the global climate change debate has increased and, with that, the acceptance for the necessity of longer-term horizons regarding the regional biofuel development has increased amongst municipal actors (interviews 2, 3, 4, 5, 6, 8, 9, 10). thus, initially, reduction of the oil-dependency of the region with a view to the ‘peak oil’-theory was the underlying problem for the biofuel region as an organisation. after 2007, however, the transition to renewable energy within transport for climate change mitigation is the underlying problem driving leading the biofuel region actors, and this is also emphasised within official biofuel region documents (bfr, 2007a, 2008, 2009, 2010). an example of this is the adoption of a new set of core values in 2008, which emphasises the development of the biofuel region towards focusing on environmental sustainability, which states that: “our driving force is global climate change and the necessity of quickly developing and using sustainable solutions for the oil-dependent transport sector... we have a holistic view and strive for sustainable development in our work” (bfr, 2008, p. 5). however, for municipal politicians, the increasing attention to climate change and renewable energy are merely viewed as evidence for a long-term business opportunity; when becoming aware of the global debate, municipal politicians see 3 peak oil theory predicts that the world supply of oil will peak and be unable to meet the increasing global demand around the early 2000s (see e.g. robelius, 2007). söderberg ● what drives sub-national bioenergy development? 129 the opportunities for solving unemployment problems through creating “green growth” in their region. policy goals the biofuel region’s original aim was to: 1) be a role model for the transition from fossil fuels to celloluse-based biofuels; 2) create industrial and regional development; 3) be knowledge leading also in 2020; and 4) to become selfsufficient regarding biofuels in 2030 (bfr, 2007a, p. 2). as these goals suggest, growth within the region and self-sufficiency were in focus when the project started, even though the focus on bioenergy was driven by global peak oil-, climate change-, and renewable energy debates: “climate change is the foundation of our work, but also growth; the interest in the regional industrial development has increased in the last years. the biofuel region does not accomplish the industrial development; but the biofuel region can improve the conditions for it” (interview 10). when asked to describe the purpose of the biofuel region, in their own words, most interviewees emphasise that the goal is to become a world-leading region for the transition from fossil fuels to biofuels through increasing the knowledge about, and interest in, biofuel use and production (interviews 1, 3, 4, 5, 6, 8, 9). many interviewees point out that an important aim for the biofuel region is to make sure that the regional companies discover the existing possibilities for bioenergy production within the biofuel region area and uses this opportunity for regional growth (interviews 3, 6, 9, 10). these views are confirmed within official biofuel region documents, where three policy goals, slightly different from the original ones, are emphasised in 2009: 1) to become leading in knowledge on societal transition to renewable energy; 2) to create industrial and regional development; and 3) to increase the supply of renewable raw material (bfr 2009). in 2013, the bfr official webpage stated that the overarching goal of the organisation is to replace products and services, which are bad for the environment and health with sustainable products and services (www.biofuelregion.se). thus, the focus-change is visible also in the policy goals of the biofuel region. initially, focus lay on producing specifically cellulose-based biofuels and there was an emphasis on achieving self-sufficiency for the region. this supports the suspicion that focus in the initial phase lay on reducing oil-dependency within the region. after 2007, however, the focus has changed to broad transition towards renewable energy and an increase in renewable raw material supply are emphasised within the organisation. this refocus seems to indicate a slight reframing of bioenergy within the biofuel region, from bioenergy as a way to reduce regional oil-dependency to bioenergy as a renewable energy source. policy preferences the interviewed biofuel region actors emphasise the importance of clear politically determined renewable energy goals for the development within the biofuel region, but also point out the need for initial economic support from higher policy levels; both in terms of investments in new technology and in terms of research and development of new technology. many of the interviewees however emphasise that central government funding is only needed initially – until the market feels secure enough for other investors. other types of general policy instruments such as electricity certificates and co2-tax are also emphasised as important for the development in the biofuel region, as well as the swedish law of 2006, which forces gas stations of a certain size to provide alternative fuels (interviews 1, 2, 3, 4, 5, 6, 7). the importance of economic support is also put forward within official biofuel region documents: “the money provided by municipalities, companies and the swedish transport administration during 2007 continue to be a crucial success factor for the biofuel region. without these the process would never have come as far as it now has” (bfr 2007a: 1). thus, biofuel region actors put renewable european journal of government and economics 3(2) 130 policy goals and political funding of investments in renewable energy forward as important policy instruments for sub-national bioenergy development. promoted values and role of bioenergy “we are highly fossil-dependent in this region, and are in the hands of the fossil industry. the price is increasing and this will affect citizens and companies. [biofuel region members] are aware of bioenergy as an issue with potential” (interview 5). when talking to biofuel region members, the role of bioenergy is double; it is both an energy source which can reduce regional fossil-fuel import dependence through domestic production, and a renewable energy source with regional growthgenerating potential (interview 2, 4, 5, 7, 8, 9, 10). many interviewees point out that the climate change issue has become more important for the biofuel region debate since 2007; in the initial phase the main argument was ‘peak oil’: “the climate issue was not important – we were running out of oil, we needed to adjust and we have forest” (interview 2). similarly, the international ethanol debate has changed the focus within the biofuel region since 2007. initially, ethanol and biodiesel was the main fuels within the biofuel region. today, there is a broader focus on different alternative fuels, where also the environmental consequences from different renewable energy sources are weighed in (interview 7): “we realised that there are many solutions; big challenges demand different fuels, we cannot put all our eggs in one basket. this was part of the global debate, but the biofuel region realised this early on” (interview 5). the environmental perspective is forwarded in many of the interviews, which also emphasise that it is important to make use of the possibilities in the region for creating green regional growth (interviews 2, 4, 5, 6, 7, 8, 9, 10). a summary of the developments in the biofuel region over time is provided in table 1. a discussion of how the framing of bioenergy and the institutional and structural factors outlined in this section correspond to the view on bioenergy on the eu and swedish levels is provided in the fourth section. furthermore, how the development of the biofuel region can be understood in a multi-level context is discussed based on the findings in this section and based on the results from interviews with key actors in the biofuel region. discussion eu-sweden-biofuel region: a cross-level comparison of epi in bioenergy within eu energy, agricultural and transport policy, bioenergy and biofuels have been forwarded mainly on the basis of energy security, climate change and rural development arguments during 2003-2013 and the epi in bioenergy within the eu can be classified as ‘weak’ given that “the integration seems to be a matter of finding win–win solutions through combining environmental, security and economic goals” (söderberg and eckerberg, 2013 p. 117). in swedish energy and agricultural policy, the arguments for promoting bioenergy and biofuels during the last decade have focused mainly on sustainability and entrepreneurial opportunities, with the addendum of supply security from 2007 and onwards. epi in bioenergy in sweden can be classified as ‘weak’ given that focus lies on combining different goals rather than on prioritising environmental aspects (söderberg, 2011a, 2011b). within the biofuel region, bioenergy was initially framed as an opportunity for the region to become independent from the oil-industry through domestic production of biofuel using the region’s rich assets of forest as raw material: ‘bioenergy-for-selfsufficiency’. with the global climate change debate, biofuels have been reframed within the biofuel region. after 2007, the transition to a sustainable transport söderberg ● what drives sub-national bioenergy development? 131 sector combined with the possibilities for green growth in the region is forwarded. as mentioned above, also the policy goals show signs of biofuel reframing, from framing specific biofuels as an opportunity for making the region self-sufficient of fuel to, after 2007, framing biofuels broadly as a renewable energy source that has an important role in societal transition. after 2007, the underlying problems for the biofuel region are an oil-dependent transport sector and climate change; the policy goal is to achieve societal transition to renewable energy; the policy preferences are supra-national and national renewable policy goals and political investments in renewable energy; and the promoted values and the role of bioenergy is mainly viewing biofuels as a potential green regional growth-sector: ‘bioenergy-for-green-transition-and-growth’. thus, an environmental perspective is present within the biofuel region official documents, expressed amongst key actors and is also present within policy strategies. this indicates that the integration of an environmental perspective is not just a matter of rhetorical or instrumental environmental learning: rather, there is epi also in bioenergy policy on the sub-national level in the biofuel region case. however, the environment is not the single driving force for the development within the biofuel region, as will be further addressed below. furthermore, the relevance of environmental arguments seem to differ between different actors within the biofuel region, which may explain why the environment, though given a significantly more prominent place after 2007, is still tightly tied to growth opportunities within the biofuel region. thus, similar to the epi on the eu and swedish level, also the epi within the biofuel region can be classified as weak. a cross-level overview of the view on bioenergy is provided in table 1. table 1: framing of bioenergy: a cross-level overview eu 2003-2013 bioenergy promotion based on energy security, climate change and rural development arguments. weak epi. sweden 2000-2013 bioenergy promotion based on sustainability and entrepreneurship arguments. from 2007 also energy security important. weak epi. biofuel region 2003-2006 ‘bioenergy-for-selfsufficiency’ underlying problems: oil dependent transport sector, unemployment policy goals: self-sufficiency for bfr and regional growth policy preferences: supra-national and national investments in specific biofuels role of bioenergy: potential regional growth sector biofuel region 2007-2010 ‘bioenergy-for-greentransition-and-growth’ underlying problems: oil dependent transport sector, climate change policy goals: to achieve societal transition to renewable energy policy preferences: supra-national and national renewable energy policy goals and investments role of bioenergy: potential green regional growth sector eu-sweden-biofuel region: the role and nature of multi-level interplay biofuel region actors hold that swedish policy affects the biofuel region indirectly, mainly through national energy policy and research policy: “economic factors play a much larger role than one wants to admit” (interview 3). many of the interviewees mention policy measures such as the ‘green car premium’ (a national economic premium for buyers of new cars fulfilling certain environmental criteria) and the ‘pump law’ of 2006 (which forces gas stations of a certain size to provide alternative fuels), as well as the co2-tax on petrol and economic support in the form central government funding from which organisations can apply for project support as important for the development within the biofuel region (interview 3, 4, 5, 9, 10). furthermore, national financial contributions to research projects are pointed out as crucial: “the biofuel region is dependent on the opportunity to apply for money via scientific research” (interview 1). many of the interviewees also point out the importance of national economic support in the initial state of building up renewable energy production and new technologies: “biofuels is a growing business. market forces do not work in the european journal of government and economics 3(2) 132 initial phase and therefore economic measures are crucial” (interview 8). regarding the role of the environmental perspective within the biofuel region, one interviewee argues that: “today environment, sustainability and self-sufficiency is on the agenda for all political parties”, which means that environmental issues are no longer controversial (interview 4). at the same time, it is also emphasized that both national and eu-regulations defines what is and is not environmentally friendly; what is regarded a “green car” and a “sustainable fuel”, for example, is to a large extent guided by national and eu-definitions (interview 4). as regards eu-policy, apart from defining what is and is not sustainable, it also affects the biofuel region mainly in two other ways. first of all, climate policy and environmental issues are high on the eu-agenda and eu-level policy goals provide sub-national bioenergy actors with long-term conditions for development and with good arguments when discussing with potential contributors to their project. especially eu-goals for renewable energy, such as the 20 percent renewable energy in 2020-goal (com 2006/848) and the biofuels directive are pointed out as important for the development within the biofuel region (interview 1, 3, 4, 5, 6, 8, 9, 10). second, eu-supported projects have been important for biofuel region development recently, however there is one big catch involved in this area: lack of competence regarding eu-level project applications: “the companies have difficulties in finding the time and competence…but the eu-money is important – we need to improve our skills in bringing these funds in” (interview 9). at the same time, the eu-membership opens up the opportunity to form projects together with other european regions and the biofuel region has received financial support for projects from the eu-level, such as within the eu-financed project best (bioethanol for sustainable transport) (interview 2). furthermore, eu-projects to a large extent set the agenda for sub-national bioenergy actors: “eu-supports are very steering regarding what we choose to get on. if you can get 40% support from the eu plus the county board you tend to go in that direction” (interview 4). when it comes to cross-level contacts, biofuel region actors are invited to euseminars and to national seminars, and the organisation is also a body considering proposed legislation regarding energy policy. however, most interviewees claim that the biofuel region could improve in the area of bottom-up contacts, even though there are informal and formal contacts with national agencies such as the swedish energy agency and the swedish board of agriculture. some of the interviewees also point out that cross-level interaction through personal contacts were more important in the initial phase of the biofuel region: “per carstedt was good at using political channels in örnsköldsvik, via elvy söderström [social democrat with a leading municipal position] to göran persson [social democratic prime minister 1996-2006]” (interview 9). per carstedt has been claimed to have had good relations both with leading social democrats and with ministers from the four-party coalition governing sweden 2006-2010 (e.g. miljöaktuellt, 2008). however, the main cross-level interaction, according to biofuel region actors, occurs through (top-down) political measures, energy policy regulations and through eu and swedish policy setting the agenda for long-term policy development and project applications on the sub-national level. conclusions and policy implications what drives sub-national bioenergy development, and what are the sub-national implications, in policy and practice, of higher-level epi in the area of bioenergy policy? in the previous section, we saw that epi on the eu, swedish and the subnational (the biofuel region) levels can be classified as ‘weak’ as the environmental aspects are integrated but tightly tied to (and not prioritised over) other goals such as energy security and economic growth. in the biofuel region case, the main imprints of eu and swedish epi in bioenergy policy have been söderberg ● what drives sub-national bioenergy development? 133 made through environmental agenda setting (i.e. steering what is sustainable through regulations and goals such as the eu biofuels directive and the swedish ‘green car premium’ and ‘pump law’) and provision of economic preconditions for the development of local biofuels (e.g. project/research funding). but there are also other important aspects, which provide explanations for subnational bioenergy development in this particular case. job creation and local growth have been the main driving forces for joining the biofuel region for the smaller municipalities. this may also explain the slightly less environmental framing of biofuels in the initial phase of the biofuel region’s work: environmental arguments were not efficient in gathering municipal support in 2003. as the environmental perspective has grown more important within euand swedish policy in general and within bioenergy policy in particular, so has the environmental perspective also gained a more prominent place within the biofuel region. the reframing of biofuels within the biofuel region (from ‘bioenergy-for-selfsufficiency’ in 2003-2006 to ‘bioenergy-for-green-transition-and-growth’ in 20072010) can thus be viewed as a way of building bridges between the different actors in the region and gather broad-based support for the biofuel development. in this regard, epi progress in the biofuel region is well in line with nilsson and eckerberg’s (2007, p. 158), claim that epi success can be related to the development of win-win strategies. as shown by the biofuel region case, win-win strategies are highly important also for sub-national epi. when it comes to formal cross-level interactions, these are mainly top-down structured: eu and swedish policy strategies and funding set the agenda for subnational bioenergy actors. given the importance of financial support in the initial phase of new technology development, this study points out the relevance for supra-national and national politicians (if they want to promote sub-national bioenergy development) to make sure that their environmental rhetoric is followed up by environmentally motivated policy instruments encompassing financial resources which sub-national actors can apply for. however, despite the importance of securing economic funding, one obstacle for sub-national bioenergy development within the biofuel region has been the lack of competence regarding applications for eu-funding. here, sub-national actors need to improve their skills (or euand national politicians need to simplify the application procedure) if such epi instruments are to have full-scale impact. this is especially important on the sub-national level, where different objectives are weighed against each other within tight budgets (baker and eckerberg, 2008), and therefore external economic support is crucial for development in areas such as renewable energy. thus, the development of the biofuel region can partly be explained by higherlevel epi: eu and swedish epi in bioenergy policy has set the sub-national level policy agenda and provided financial resources for sub-national actors, who have reframed their objectives in line with the changing global debate. although some bottom-up interaction occurs, and played a significant role through personal contacts in the initial phase, cross-level interactions in the biofuel region mainly take place through top-down mechanisms such as policy formulation, central government funding programmes and regulations. however, policies and economic instruments have to be implemented in practice on the sub-national level, and here the existence of ‘policy entrepreneurs’ is equally important to understand the development of the biofuel region. previous studies of epi on the european and national level show that policy entrepreneurs are important for maintaining the environmental attention in a policy area (nilsson and eckerberg, 2007), and that renewable energy projects initiated on environmental grounds are more durable than renewable energy projects initiated for other reasons (e.g. to secure economic growth or supply-security) (söderberg, 2008). according to the interviewees, those municipalities where european journal of government and economics 3(2) 134 genuinely environmentally interested persons held leading positions also entered the biofuel region with more realistic expectations and endured longer within the project. those who mainly entered with the expectation of having biofuel factories in every municipality within a few years’ time soon became disappointed. thus, the biofuel region case shows that the presence of policy entrepreneurs and environmental motives for action that are central for epi on the national and eulevel also are important factors for epi in sub-national settings. furthermore, external factors have been important for the initiation of the biofuel region project (policy windows in the form of vinnväxt grants and the eu biofuels directive provided opportunities for the region), for the municipal level reframing and for the refocus of the biofuel region towards forming a broader renewable energy cluster rather than promoting a certain type of biofuel. in the biofuel region case, the initial direction (on ethanol – within the frame ‘bioenergy-for-selfsufficiency’) was set by the regional forest assets and by the companies involved, and this direction did not change (towards promoting different types of renewable energy within the frame ‘bioenergy-for-green-transition-and-growth’) until the global, eu and national debates on climate change and biofuels made it necessary to do so. in sum, the biofuel region case shows that top-down policy instruments such as regulations and economic support drive the direction of sub-national renewable energy development (i.e. the types of energy sources that are developed and on what – environmental or non-environmental – grounds). thus, the diminishing role of the state, which is often pointed out in mlg-studies, is only partly confirmed in this study. although the eu sets the agenda and provides economic support to local projects, traditional governance from the national level still has an important steering role in the field of energy policy. eu and national policies set the agenda and preconditions for local renewable energy development and influence the degree of epi also on the sub-national level. however, the contacts between levels are mainly top-down oriented. nevertheless, the biofuel region case also provides further understanding of why sub-national renewable energy development takes place in one location but not in others. sweden, for example, is covered with forest but there is only one biofuel region. in order to initiate and maintain durable and sustainable renewable energy projects on the sub-national level, policy entrepreneurs capable of pointing towards local win-win-solutions, with extensive personal networks (to assure higher-level support for their particular local renewable energy project) and which are capable of applying for eu and national funding are crucial. thus, although top-down steering provides the setting for sub-national renewable energy development and thus is a highly important mechanism for ensuring cross-level epi, bottom-up contacts and local multi-actor networks also have important roles to play for local projects to be successful and sustainable. references abn (2007) agrofuels in africa: the impact on land, food and forests. case studies from benin, tanzania, uganda and zambia. kenya: african biodiversity network. (available at url: http://africanbiodiversity.org/abn_old/resources.html) bache, ian and matthew flinders (eds) (2004) multi-level governance. oxford: oxford universtiy press. baker, susan and katerina eckerberg (2008) ‘conclusion: combining old and new governance in pursuit of sustainable development’, in susan baker and katerina eckerberg (eds) in pursuit of sustainable development: new governance practices at the sub-national level in europe. oxon: routledge, pp. 208-228. bergström, göran and kristina boreus (2005) textens mening och makt: metodbok i samhällsvetenskaplig textanalys. lund: studentlitteratur. söderberg ● what drives sub-national bioenergy development? 135 bfr (2007) verksamhetsberättelse av biofuel regions etapp i 2003-2006. umeå: biofuel region. bfr (2007a) verksamhetsberättelse 2007. biofuel region ab and biofuel region ideell förening. umeå: biofuel region. bfr (2008) verksamhetsberättelse 2008. biofuel region ab and biofuel region ideell förening. umeå: biofuel region. bfr (2009) verksamhetsberättelse 2009. biofuel region ab and biofuel region ideell förening. umeå: biofuel region. bfr (2010) verksamhetsberättelse 2010. biofuel region ab and biofuel region ideell förening. umeå: biofuel region. christensen, lars (2005) formation for collective action: the development of biofuel region. visanu 2005:10. stockholm: nutek förlag. dryzek, john s. (2000) deliberative democracy and beyond: liberals, critics, contestations. oxford: oxford university press. ec (2003) ‘proposal for a council regulation on establishing common rules for direct support schemes under the common agricultural policy and…’, com (2003) 23 final of 21/01/03. commission of the european communities: brussels ec (2005) ‘biomass action plan’, com (2005) 628 final of 7/12/05. communication. commission of the european communities: brussels eckerberg, katarina and björn forsberg (1998) ‘implementing agenda 21 in local government: the swedish experience’, local environment, 3 (2): 333-347. eea (2006) ‘how much bioenergy can europe produce without harming the environment?’ european environment agency report 07/2006. luxembourg: office for official publications of the european communitites. esaiasson, peter, mikael gilljam, henrik oscarsson and lena wängnerud (2005) metodpraktikan: konsten att studera samhälle, individ och marknad. stockholm: norstedts juridik ab. folke, carl, johan colding and friket berkes (2002) ‘building resilience for adaptive capacity in social-ecological systems’, in friket berkes, johan colding and carl folke (eds) navigating social-ecological systems: building resilience for complexity and change. cambride: cambridge university press, pp. 352-387. giampietro, mario and kozo mayumi (2009) the biofuel delusion: the fallacy of large-scale agro-biofuel production. london: earthscan. gunderson, lance and c.s. holling (eds) (2002) panarchy: understanding transformations in human and natural systems. washington, dc: island press. hall, peter a. (1993) ‘policy paradigms, social learning, and the state’, comparative politics 25 (2): 275–296. hooghe, liesbet and gary marks (2003) ‘unraveling the central state, but how? types of multi-level governance’, american political science review 97(2): 233– 243. ipcc (2007) fourth assessment report: climate change 2007. geneva: intergovernmental panel on climate change. jacob, klaus and axel volkery (2004) ‘institutions and instruments for government self-regulation: environmental policy integration in a cross-country perspective’, journal of comparative policy analysis: research and practice 6(3): 291-309. jordan, andrew (2002) ‘efficient hardware and light green software: environmental policy integration in the uk’, in andrea lenschow (ed) environmental policy integration: greening sectoral policies in europe. london: earthscan, pp. 35-56. european journal of government and economics 3(2) 136 jordan, andrew and andrea lenschow (2008) innovation in environmental policy?: integrating the environment for sustainability. cheltenham: edward elgar. kingdon, john (1995) agendas, alternatives and public policies. second edition. new york: harper collins. lafferty, william m. (2004) ‘from environmental protection to sustainable development: the challenge of decoupling through sectoral integration’, in william m. lafferty (ed) governance for sustainable development. cheltenham: edward elgar, pp. 191-220. lasswell, harold d. (1950) politics: who gets what, when, how. new york: peter smith. lenschow, andrea (1999) ‘the greening of the eu: the common agricultural policy and the structural funds’, environment and planning c-government and policy 17 (1): 91-108. lenschow, andrea (ed) (2002) environmental policy integration: greening sectoral policies in europe. earthscan, london. lenschow, andrea and anthony r. zito(1998) ‘blurring or shifting of policy frames? institutionalization of the economic-environmental policy linkage in the european community’, governance 11 (4): 415-442. miljöaktuellt (2008) mr biofuel – mauds okända rådgivare. url(consulted at 200802-07): http://miljoaktuellt.idg.se/2.1845/1.144053 nilsson, mans (2005) ‘learning, frames and environmental policy integration: the case of swedish energy policy’, environment and planning c: government and policy 23 (2): 207-226. nilsson, mans and katarina eckerberg (eds) (2007) environmental policy integration in practice. london: earthscan. nilsson, mans, katarina eckerberg and g. finnveden (2007) ‘discussion: what enabled epi in practice?’, in mans nilsson and katarina eckerberg (eds) environmental policy integration in practice. london: earthscan, pp.137-162. nilsson, mans and åsa persson (2003) ‘framework for analysing environmental policy integration’, journal of environmental policy & planning 5 (4): 333-359. oberthür, sebastian and thomas gehring (eds) (2006) institutional interaction in global environmental governance. synergy and conflict among international and eu policies. cambridge: mit press. persson, åsa. (2007) ‘different perspectives on epi’, in mans nilsson and katarina eckerberg (eds) environmental policy integration in practice. london: earthscan, pp. 25-47. radaelli, claudio m. (2003) ‘europeanization of public policy’, in kevin featherstone and claudio m. radaelli (eds) the politics of europeanization. oxford: oxford university press, pp. 27-56. rein, martin and donald schön (1993) ‘reframing policy discourse’, in frank fischer and john forester (eds) the argumentative turn in policy analysis and planning. durham, nc: duke university press, pp. 145-166. rein, martin and donald schön (1994) frame reflection. new york: basicbooks. rhodes, rod (2000) ‘governance and public administration’, in jon pierre (ed) debating governance. oxford: oxford university press, pp 54-90. robelius, fredrik (2007) giant oil fields – the highway to oil: giant oil fields and their importance for future oil production. doctoral dissertation. uppsala: uppsala dissertations from the faculty of science and technology. söderberg ● what drives sub-national bioenergy development? 137 sabatier, paul a. and hank c. jenkins-smith (1999) ‘the advocacy coalition framework: an assessment’, in paul a. sabatier (ed) theories of the policy process. oxford: westview, pp. 117-166. stagle, sigrid (2007) ‘theoretical foundations of learning processes for sustainable development’, international journal of sustainable development and world ecology 14 (1): 52-62. stern nicholas (2007) the economics of climate change: the stern review. cambridge university press: cambridge. swartling, åsa g., mans nilsson, rebecka engström and lovisa hagberg (2007) ‘theory and methodology for epi analysis’, in mans nilsson and katarina eckerberg (eds) environmental policy integration in practice. london: earthscan, pp. 25-47. swedwatch (2009) ‘en brännande fråga. hur hållbar är den etanol som importeras till sverige?’, swedwatch rapport nr 25. stockholm: swedwatch. söderberg, charlotta. (2008) ‘much ado about nothing?’ energy forest cultivation in sweden: how intersectoral policy coordination affects outcomes from epi in multisectoral issues’, journal of environmental policy & planning 10(4): 381-403. söderberg, c. (2011a) ‘institutional conditions for multi-sector environmental policy integration in bioenergy’, environmental politics (20) 4: 528-546. söderberg, charlotta (2011b) environmental policy integration in bioenergy: policy learning across sectors and levels? doctoral thesis. department of political science research report. umeå: umeå university. söderberg, charlotta and katarina eckerberg (2013) ‘rising policy conflicts over bioenergy and forestry’, forest policy and economics 33: 112-117. urwin, kate and andrew jordan (2008) ‘does public policy support or undermine climate change adaptation? exploring policy interplay across different scales of governance’, global environmental change 18(1): 180-191. vinnova (2010) about vinnova. url: http://www.vinnova.se/sv/in-english/aboutvinnova/ (2010-02-16) vinnova (2010a) vinnväxt: regional growth through dynamic innovation systems. url(consulted 2010-02-16): http://www.vinnova.se/inenglish/activities/strong-research-and-innovation-environments/vinnvaxt/ vinnova (2010b) winners. url(consulted 2010-0216):http://www.vinnova.se/sv/in-english/activities/strong-research-and-innovationenvironments/vinnvaxt/winners-2003-2004/. vinnova (2010c) framtidens bioraffinaderi. url(consulted 2010-02-16): http://www.vinnova.se/sv/verksamhet/starka-forsknings--ochinnovationsmiljoer/vinnvaxt/vinnare/2008/framtidens-bioraffinaderi/ . wced (1987) our common future. united nations world commission on environment and development. oxford: oxford university press. the gains from early intervention in europe: fiscal surveillance and fiscal planning using cash data european journal of government and economics volume 1, number 1 (june 2012) issn: 2254-7088 44 * the views expressed in this paper are the authors’ and do not necessarily reflect those of the european central bank (ecb). we thank j. paredes for valuable inputs, and members of the ecb’s fiscal policies division for comments. ** address for correspondence: thomas warmedinger, european central bank, kaiserstr. 29, 60311 frankfurt am main, germany. e-mail : thomas.warmedinger@ecb.int. the gains from early intervention in europe: fiscal surveillance and fiscal planning using cash data* andrew hughes hallett, george mason university and university of st andrews, united states moritz kuhn, university of bonn, germany thomas warmedinger, european central bank,** germany abstract the use of real-time cash data allows us to make accurate intra-annual forecasts of an economy’s fiscal position, and to issue early warning signals for the need to correct fiscal imbalances. this paper shows how those signals should be used to design the necessary fiscal corrections, and discusses the gains that can be achieved by such interventions. examples from germany and italy show that large corrections are often necessary early on to make the later adjustments acceptable in size and to keep debt ratios from escalating. there is a credibility issue here; we find the differences between front-loaded and backloaded adjustment schemes are likely to be vital for the time consistency and hence credibility of fiscal policymaking. we also show that, without early interventions, the later deficit reductions typically double in size – meaning governments become subject to the excessive deficit procedure and significant improvement tests more often. thus the budget savings from early intervention and the use of cash data are significant. in our examples they are similar in size to the operating budget of the department of housing and urban development in germany. similar results apply in other eurozone countries. jel classification e62; h50; h68 keywords fiscal surveillance; early warning; cash data; additive vs. slope adjustments; fiscal credibility doi: https://doi.org/10.17979/ejge.2012.1.1.4276 european journal of government and economics 1(1) 45 introduction the past few years, especially since the expanded debt and fiscal deficits that followed the 2007-09 financial crisis and recession, have given rise to an emerging literature on the credibility of fiscal policies and the anchoring of (fiscal) expectations.1 it matters a great deal that expectations of future policy should be anchored; that is, fiscal decisions and their financing need to be predictable, easily understood and credible in the sense of being both plausible and likely to be sustained into the future. but it is equally important that the necessary corrections be undertaken to return fiscal policies to their intended path following a shock or slippage. only then can we regain the full effectiveness of fiscal policy that leeper and others associate with appropriately anchored expectations.2 the goal of this paper is to use early the warning signals generated by cash data on the state of government finances. to do this we extend the results of perez (2007), pedgregal and perez (2008), and onorante et al. (2008), who show how intra-annual data can be used to derive accurate forecasts for end-of-year fiscal outcomes. these forecast can then be used to issue “early warning” signals about the corrections that need to be made in order to bring public finances (both deficits and debt) back on their intended track – as may be required by the stability and growth pact (sgp) from time to time. we show that such early warning signals can be highly valuable because they allow us to start an adjustment process earlier and spread the effort over a longer time period. that makes the size of each adjustment a smaller and less contentious issue; and therefore less likely to create opposition in government or its electorate. second, we also examine whether budget corrections are better frontloaded or back-loaded – we find that different adjustment mechanisms have different size and timing profiles. these are all relevant questions from a policy perspective. it is important that the plans submitted or announced are actually time-consistent, so that governments stick to their published targets after an unplanned departure from the officially agreed stability programme (sp).3 we highlight this point because, as figure 1 shows in the case of germany and italy, there seems to be considerable evidence that the time consistency in sp plans has been violated on a regular basis. the figure highlights how sequential multi-period sp plans are set at each point in time (bottom left of each line), and that there was a deterioration of the starting position in the following year in most cases. some analysis of how these situations can be avoided is therefore necessary. 1 this literature is principally due to eric leeper in a series of papers conveniently summarized in leeper (2009). the conditions under which expectations can be anchored in such a way as to be consistent with what both the policymakers and what the private sector expect and intend to happen are set out in hughes hallett et al (2012a). 2 see hughes hallett et al (2012b). woodford (2005) makes the same point for monetary policy. 3 governments are obliged to submit a stability programme every year, which contains a timetable for the deficit to return to the defined medium term objectives (mtos) for budgetary balance. based on the sp, the european commission reviews the national fiscal policies and can issue early warnings to member states if it considers the developments in the programme to be too optimistic or too ambitious. however, the assessment of the submitted sp is done ex ante and is based only on the information available at the point of submission. contingency plans are not included (except for some mechanical scenarios); and there is no provision to use real time or cash data to make a projection of the plan’s realisation so that monitoring and subsequent adjustments can be undertaken during the plan’s life time. in this paper, we consider ways to correct those omissions – and why it is important to do so. european journal of government and economics 1(1) 46 figure 1. announced deficit plans for germany and italy intra-annual and cash data: it has long been emphasized in the fiscal forecasting literature, that data on government deficits are only available with a long time lag.4 at the time when the data becomes available, the deficit has already realized and no policy measures can be implemented to avoid drifting away from the intended target values. however pérez (2007), pedregal and pérez (2008), and onorante et al. (2008), have shown that intra-annual data can be used to derive an accurate picture of the likely year-end fiscal outcomes with much shorter time lags. we extend their results by introducing cash data into that process. we then go one 4 see, for example, pérez (2007), onorante et al. (2008) or pedregal and pérez (2008). european journal of government and economics 1(1) 47 step further and ask how much a government can actually gain by using early warning signals from this procedure and taking immediate action to correct a fiscal slippage. this question has received rather little attention in discussions on the prudential surveillance of fiscal policy. the reasons are two-fold. on the one hand, the question requires setting an objective function for the government in order to relate departures from a proposed plan to the adjustment costs inherent in making a correction. we discuss that issue in this paper and review some alternative measures. on the other hand, this question raises the problem of time consistency in fiscal policy. we highlight the potential for time inconsistency, and discuss the trade-off between adjustment costs and time inconsistency. at that stage it becomes necessary to draw out the connection between deficit and debt surveillance within the sgp rules. both sgp components need to be monitored if we are to fulfil the purpose behind the sgp – and create fiscal policies that are sound in the long run. we have set our analysis within the framework of the existing sgp system because that is still the standard framework for monitoring and control of fiscal policies in the euro-zone. however, while early intervention may have at best only mitigated the outbreak of the fiscal crises in 2009-10, our distinction between additive and slope correction schemes (ie between the frontloading and backloading of deficit corrections) has now been pushed to centre stage given that preventing further debt escalations has necessarily become the prime focus of attention. back-loading, as we shall show, allows debt ratios to rise slowly but surely during the consolidation period and have had to be ruled out for that reason. it seems likely that these results will lead to debt target rules being formally included within the sgp process (european commission 2011), but the implementation remains to be seen. these developments all serve to highlight the importance of including both early interven-tions and cash data monitoring in any realistic fiscal surveillance programme.5 it would be nice to evaluate these questions in the context of the choice of fiscal austerity vs. growth debate which has become a feature of the european debt crisis, to check that the gains from early interventions, or frontloading vs. backloading those interventions, are not larger than their costs in a wider welfare sense (we do check for political acceptability and market credibility). clearly the economic environment and timing of interventions matter, as do the fiscal multipliers and impacts on gdp growth of different implementations or choices of fiscal consolidation measures. indeed, this involves a major debate debate with opposing positions taken by delong and summers favouring not too drastic corrections in order to avoid double dip recessions and killing off greenshoots, and neumann (2012) arguing for recovery through austerity and reduced fiscal imbalances; with corsetti (2012) and alessandrini et al (2012) emphasising the more subtle tradeoffs between growth, market credibility, and competitiveness and financial sector imbalances that lie between these two positions. however, important as those issues are, they are not directly at issue here because we are concerned with the most effective and credible way to recover a previously agreed fiscal (sp, mto) path once the budget deviates from it – not whether that path was optimally chosen in the first place. we presume that path to have been chosen taking full account of the relevant fiscal multipliers, policy preferences, market discipline, and credibility/risk premia. those choices are a separate issue from the best way to recover the path that incorporates them after a deviation or shock. it is therefore 5 there is a small literature that uses real time fiscal data (hughes hallett et al 2012; bernoth et al 2012; beetsma et al 2009; cimadomo 2007; von hagen and wolf 2006; golinelli and momigliano 2008). this literature has been concerned with the ability of such data to pick up budget failures, whether fiscal policies have been counter-cyclical, and whether governments are systematically overoptimistic in their budget projections. it does not look at using real time data to help correct unforeseen deficits. nevertheless, it shows up the unreliability and revisions in such data. for that reason, we work with cash data instead: it is real time and accurate. european journal of government and economics 1(1) 48 beyond the scope of our present analysis to assess how economic recovery itself should be achieved and if the policy implementation may be hampered in the case of a severe economic crisis.6 data the data used in this paper come from three different data series. first, since we are interested in the deficit of the general government, we use the annual series of government deficits according to esa95 accounting standards. we then augment this series with data collected using the esa79 standards in order to construct a longer time series. this data comes at annual frequency and provides the official deficit figure when it comes to the assessment of compliance with the sp plans. second, starting in 1995, there is also a series at quarterly frequency of government deficit data. this also follows esa95 accounting standards. we use this data as intra-annual information to forecast the end-of-year government deficit. finally, we use the data series of the public accounts (cash series) that comes at monthly intervals. in addition, we use quarterly nominal gdp data. these data are required to normalize all other variables as ratios of gdp. thus the data we use come at three different frequencies. however, we can reduce the frequency of the cash data by aggregating it to quarterly observations if need be. timeline of events our goal is to study the gains that can be obtained from early intervention in fiscal policy by governments. to do that, we have to specify the sequence of events of the government’s action. in period t-1, before the current budget year starts, the government submits a sp plan to the european commission. then at the start of quarter q3 of year t, the first year covered by the sp plan, a deficit forecast for the entire year is made.7 if the forecast predicts a slippage of the government deficit below the sp target, the government can consider putting early intervention measures into place immediately in order to avoid a deficit realization that is too far away from their sp target. independent of whether early intervention measures have been implemented or not, the government also has to adjust its plan for the three remaining years of the plan to satisfy the debt targets of the initial sp plan by the end of the plan period. we illustrate the timing of these events in figure 2: 6 the choice of fiscal instruments is discussed in alesina and ardagna (2009). a more critical view of austerity measures in crisis times is presented in delong and summers (2012). 7 this is the earliest period to make a reasonable prediction of the year-end deficit, given the data, model, and information restrictions in the problem. european journal of government and economics 1(1) 49 figure 2. timeline of events n tg denotes the growth of nominal gdp in year t; td̂ denotes the debt-to-gdp ratio at the end of period t, and tδ is the deficit-to-gdp ratio in period t. the time line set out in figure 2 is conceptually straightforward, but perhaps difficult to implement. it puts a lot of faith in the effectiveness of fiscal activism, and in the “costless” production and pursuit of sp plans. many governments may find that hard going in practice, and that mid-year is a bit late to adjust current (and future) budget plans which may have been in preparation for some time. nevertheless, using real time data, beetsma and guiliodori (2008) show there is considerable evidence that governments do in fact adjust their budgets mid-year. we may therefore assume that the time line in figure 2 is feasible, if uncomfortable, at least for governments that are sufficiently determined to correct any budgetary failures before they run out of control. that is not to say that all governments do make fiscal corrections in good time. recent experience in the euro-zone shows very clearly that, while some do, many do not. in an extension to figure 2’s timeline, we also discuss the case where no changes to the original plan are made and calculate the deviations in the debt ratio that then appear at the end of the sp plan. correction plans when there has been a one-time deviation in fiscal policy such that the deficit exceeds the sp plan, there are at least two ways to correct the deficit and meet the debt target at the end of the sp. one method we label the constant slope plan, and the other an additive constant plan.8 the additive constant plan takes the original plan and adds a constant correction to all deficits planned under the original sp agreement. this constant d therefore describes a level shift. this plan will have large adjustments between q3 of period t, when the corrections start if there are early interventions (or from the end of period t if not), and period t+1 when the corrections can continue at the rate originally planned. either way, these adjustments will be enough to eliminate the extra debt caused by the deficit slippage. we sketch this correction process in figure 3. 8 both plans have the property that they can be described by a single parameter. in general, a correction plan is not uniquely determined because we have three degrees of freedom to define the plan but only one target to match. but once we concentrate on plans that can be described by a single variable, they are uniquely identified by the final debt target of the sp plan. european journal of government and economics 1(1) 50 figure 3. constant additive adjustments there a two correction plans: one with early interventions (dashed line), and one without early interventions (dotted line). we also plot the original sp/mto plan (solid line). these corrections are based on the original sp plan plus an additive constant chosen to meet the same debt target at the end of the plan. the constant slope plan, by contrast, does not have this front-loading property. instead, it creates a constant change in the deficit each year. the burden of adjustment in this plan is therefore spread more equally over the remaining periods of the plan. this time, the constant d describes a constant change in the deficit from year to year, whereas d in the additive case described a constant shift in the deficit compared to the original sp plan. we sketch this correction plan in figure 4: note that the sp path is identical in both cases, so that we can easily compare the correction paths (see figure 5). we see that the constant slope plan has much smaller corrections in the deficit in the periods t to t+1. the larger corrections are postponed to later periods where, compared to the additive case, the adjustments relative to the original sp plan are plainly larger. they are back-loaded therefore.9 in that sense these two correction schemes are polar opposites in terms of where the adjustments fall. the frontloaded, additive constant case represents a rapid correction (“cold shower”) strategy and risks stirring up political opposition. 9 both correction plans include early interventions, and are calculated so that the sp/mto debt target is reached at the end of the plan horizon. european journal of government and economics 1(1) 51 figure 4. constant slope adjustments there a two correction plans: one with early interventions (dashed line) and one without early interventions (dotted line). we also plot the original sp/mto plan (solid line). all plans have a constant slope and meet the same debt target at the end of plan. by contrast, the back-loaded, constant slope version is easier to implement. but it risks allowing the policymakers to backslide and will therefore lack credibility if the markets doubt that it will be carried through to the end. the potential for time inconsistency here underlines the importance of leeper’s literature on anchoring expectations (ensuring the credibility of fiscal corrections), especially if this route is taken. on the other hand, the usual fear that once a forecasting relationship is in use the underlying behaviour upon which it is based will change (a form of the lucas critique), may be less severe here than in the monetary policy arena. monetary rules often breakdown because market behaviour changes. but people have little freedom to change the taxes they owe even if deficits are forecast to change, and governments likewise have little chance to change their spending rules since most are “contractual”. in that case, private sector responses to a forecast budget deviation from plan are likely to be limited in the short term. only in extreme cases, where the long term sustainability of the debt target is threatened, would we expect to see a breakdown of our forecasting and correction rules. that could be a possibility in the second constant slope correction plan; but not in the first, additive adjustments plan – which may be a good reason to favour the latter. european journal of government and economics 1(1) 52 figure 5. the constant slope and constant additive adjustments compared the constant slope (dashed line) and constant additive adjustments (dotted line) compared in the same problem (the early intervention strategies in figures 3 and 4 respectively). the solid line shows the original sp/mto plan. example 1: the case of germany in 2002 to give concrete examples of the differences between the two correction schemes we have just described, we first consider the case of german fiscal policy in 2002. the german government submitted the following plan (see table 1) in december 2001 to the european commission: table 2. german government deficit plan, starting in 2002, as submitted in december 2001 2002 2003 2004 2005 deficit -2.0% -1.0% 0.0% 0.0% nominal gdp growth 2.77% 4.04% 4.04% 4.04% debt 60.38% 59.04% 56.75% 54.55% deficit and growth rates are taken from the plan submitted by the government. the debt figures follow from our own calculations based on a debt ratio of 60% in 2001. recall figure 1 where we plotted this plan together with the plans that followed this one. as can be seen immediately, this plan had to be revised in the following year because the realised deficit did not coincide with the planned deficit of -2.0% of gdp. hence, early interventions would indeed have had the potential to avoid further fiscal slippages. for this plan, we take the deficit numbers and the nominal growth rates as given. the debt level in 2001 is taken to be 60 % of gdp. we update the debt ratio according to the following rule: 1 ˆ ˆˆ 1 t t tn t d d g     (2) european journal of government and economics 1(1) 53 where td̂ denotes the debt ratio at the end of period t, tδ is the deficit in period t, and n tg is the nominal growth rate of gdp between periods t and t-1.10 the deficit figure, ,ˆ t contains interest payments on the accumulated stock of debt: 1 ˆ tdi .11 for the purposes of illustration, we abstract from changes in the interest rates over time in this paper. it is however straightforward to extract the interest payments component and analyse it separately from the primary deficit. using the model proposed in onorante et al. (2008), we forecast the outcomes for germany based on june/july 2002 information.12 we expect a deficit for 2002 of 3.88%.13 we can then derive the actual debt ratio of the german government using the realization for 2002 assuming that this is a one-time deviation from the sp plan (see table 2). this case is constructed to resemble a situation in which no intraannual interventions are made, and the full deficit is realized at the end of 2002. table 3. german government deficit plan after a one-time deviation in 2002 2002 2003 2004 2005 deficit -3.65% -1.0% 0.0% 0.0% gdp growth 2.77% 4.04% 4.04% 4.04% debt 62.03% 60.62% 58.27% 56.01% debt (planned) 60.38% 59.04% 56.75% 54.55% deficit and growth rates are taken from the plan submitted by the government in december 2001, and no corrections are taken with respect to these numbers. the debt follows from our own calculations based on a debt ratio of 60% for 2001. all gdp growth rates are nominal. if the german government were simply to return to the original sp plan after 2002, the deficit corrections would be rather large (2.65% and 1% in 2003 and 2004 respectively), and there would be a permanent rise in the debt burden of 1.5% for ever – increasing the interest payments by 0.075% of gdp every year – all from one deficit slippage of 1.65% of gdp in 2002. corrections of this size are unlikely to be undertaken. next we consider a situation in which early intervention measures are taken after the deficit forecasts are made. we make an arbitrary assumption that these intervention measures are effective to the extent that, at the end of the year, the deficit becomes a convex combination (with equal weights) of the sp target and the realization for that year; thus the deficit for 2002 in the case of early intervention new 2002̂ becomes: %82.2%)0.2%65.3(5.0ˆ 2002 new (3) this step is used to provide a start to our calculations for the additive and constant slope corrections with early interventions. we replace it with a less arbitrary, but rather more difficult to implement assumption for the early interventions in section 6.3. by contrast, where there are no early interventions, we start with the realised deficit at the end of period t. 10 t is negative if the government runs a budget deficit; positive if it runs a surplus. the planned debt ratios correspond to the ratios submitted by the german government, except for 2005 where the government reported a debt ratio of 55.5% but the updating procedure gives a ratio of 54.55%. 11 i denotes the nominal interest rate on debt from period t-1 to t. 12 details of these forecasts are available from the authors on request. 13 it will turn out that, due to favourable shocks, the actual deficit was 3.66% for 2002. european journal of government and economics 1(1) 54 for simplicity we abstract from any legal complications in changing the budget. the procedures will vary by country. typically tax changes must be legislated. but spending rules, especially as regards the timing and speed of that spending, mostly do not once the spending allocations have been made. thus, for the size of corrections reported here, the changes could be introduced at will. the additive correction approach for the correction of a one time deficit slippage, we look first at the approach with a level shift for the remaining years up to the end of the planning horizon, in this case to 2005. the level shift is chosen such that the government can still satisfy its originally announced debt target.14 we want to find deficits 2003 2004 ˆ ˆ,new new  and 2005 ˆ new such that we meet the sp debt target of 54.55% in 2005. since many plans can satisfy this target, we look for a simple plan with a constant additive shift: d d d spnew spnew spnew    20052005 20042004 20032003 ˆ:ˆ ˆ:ˆ ˆ:ˆ it is easy to solve for d using the law of motion for the debt ratio in (2). we get                                  3 2 1 3 3 3 2 3 2 1 3 1 3 1 )1( 1 1 1 ˆ )1( ˆ )1( ˆ )1( ˆ )1( ˆˆ t ts n sn t sp tn t sp t t ts n s sp t t ts n s t t ts n s t g g g ggg d d d where 1 ˆ td denotes the initial debt before the sp plan starts; t̂ denotes the deficit realized in the first year covered by the sp plan; and d̂ denotes the target debt ratio at the end of the sp plan. if we evaluate d for germany in 2002, we get d = 0.0025. given the policies of 2002, this means the government has to decrease its deficit in each year, for the following three years, by 0.25% of annual gdp. the new plan can be seen in table 3. notice that the deficit corrections going into 2003 and 2004 (2.1% and 1% of gdp) are smaller than in table 1, and involve no extra debt or interest payments. table 4. german government deficit plan after a one-time deviation and revision in july 2002 2002 2003 2004 2005 deficit -2.83% -0.75% 0.25% 0.25% gdp growth 2.77% 4.04% 4.04% 4.04% debt 61.21% 59.58% 57.01% 54.55% debt (planned) 60.38% 59.04% 56.75% 54.55% deficit and growth rates are taken from the plan submitted by the government in december 2001, and no corrections are taken with respect to these numbers. the debt follows from our own calculations based on a debt ratio of 60% in 2001. 14 recent literature has emphasized the importance of targeting debt, rather than deficits, as a means of restraining fiscal policy: hughes hallett (2008a,b), , hughes hallett and jensen (2011, 2012). in view of the effects of the 2007/09 recession, policymakers have also considered the possibility of adopting explicit debt targets (blanchard, 2010; european commission 2011). european journal of government and economics 1(1) 55 the alternative would be not to take early intervention measures and to let the deficit take its realized value for 2002. in this case, the debt at the end of 2002 would be 62.03%. based on this, we can again calculate the shift value, d, necessary to reach the 54.55% debt target in 2005. we get d=-0.0051. this implies that the government has to reduce its deficit for the next three years by an additional 0.51% of gdp each year, compared to 0.25% if early intervention measures had been taken. the necessary annual corrections would have doubled in size therefore. against this, it must be kept in mind that the government already had to intervene in mid-2002 in the early intervention case. it reduced the deficit by 0.83% of gdp: from 3.65% with no early interventions, to 2.83% of gdp under early interventions (thereby avoiding the need to undergo the sgp’s excessive deficit process and the significant improvement test). the difference of 0.83% is spread as net expenditure cuts over the remaining three years if there are no early interventions – matching the 0.25% per year of early interventions, plus the additional interest charges. by contrast, the 0.83% difference in the early interventions case had to be corrected through public spending cuts in 2002.15 that may be a tough call, but it does allow the german government to avoid the sgp’s excessive deficit procedure and it is the kind of correction the german government will impose upon itself in its own new balanced budget legislation. the slope correction approach the second approach for correcting the one time deficit slippage would be to apply a constant slope adjustment path for the remaining years up to the end of the planning horizon, in this case 2005. the slope has to be chosen such that the government can still satisfy its original debt target. we need to find deficits new 2003 , new 2004 and new 2005 such that we meet the debt target of 54.55% in 2005. since there are many plans that can satisfy this target, we look for a simple plan such that: dd dd d newnew newnew new 3ˆˆ:ˆ 2ˆˆ:ˆ ˆ:ˆ 200220042005 200220032004 20022003    given that, it is easy to solve for d based on the law of motion for the debt ratio in (2)                                3 2 1 3 3 2 1 3 1 1 3 1 )1( 1 2 3 1 1 1 )1()1(ˆ )1( ˆˆ t ts n sn t t ts n s n s t ts n stt ts n s t g g g gg g d d d where 1 ˆ td denotes the initial debt before the sp plan starts, t̂ denotes the deficit realized in the first year of the sp plan, and d̂ denotes the target debt ratio at the end of the plan. if we evaluate d for germany in 2002, we get d=0.0131. this 15 notice however that the total budget cuts with no early interventions are slightly smaller than those with early interventions: 1.53% vs. 1.58%, despite extra interest charges. this is because of the effect of higher gdp growth on the debt/deficit ratios later in the plan. european journal of government and economics 1(1) 56 means that the government, under early intervention, has to decrease its deficit in each year, for the following three years, by 1.31% of annual gdp. as a result, the budget contractions are 0.46% and 0.35% of gdp smaller in 2003-4, but 1% larger in 2005. the new plan can be seen in table 4. table 5. german government deficit plan after a one-time deviation and revision in july 2002 2002 2003 2004 2005 deficit 2.83% 1.46% 0.10% 1.26% gdp growth 2.77% 4.04% 4.04% 4.04% debt 61.21% 60.30% 58.06% 54.55% debt (planned) 60.38% 59.04% 56.75% 54.55% deficit and growth rates are taken from the plan submitted by the government in december 2001, and no corrections are taken with respect to these numbers. the debt follows from own calculations based on a debt ratio of 60% for 2001. if we consider the case of no early interventions measures, then the corrections under a constant slope regime needed to meet the debt target in 2005 would require a slope parameter d=0.0182 and a period by period budget contraction of 1.82% of gdp. this is a large number: 0.5% larger in each period than with early interventions. on the other hand, the corrections here reduce the net spending cuts (or any revenue increases) to only 0.83% and 0.3% of gdp in 2002 and 2003, but then increase them by 2.1% of gdp in 2004-5. this adjustment plan is set out in table 5. the extent of the softening of the early budget contractions in 2002-3, and then the back-loading of them in 2004-5, is large and clearly visible: table 6. german government deficit plan after a one-time deviation and revision in july 2002 2002 2003 2004 2005 deficit -3.66% 1.76% 0.14% 2.04% gdp growth 2.77% 4.04% 4.04% 4.04% debt 62.04% 61.39% 58.87% 54.55% debt (planned) 60.38% 59.04% 56.75% 54.55% deficit and growth rates are taken from the plan submitted by the government in december 2001, and no corrections are taken with respect to these numbers. the debt follows from own calculations based on a debt ratio of 60% for 2001. altering the impact of early intervention for our numerical example, we assume that, at the point at which the slippage is first detected, it is still possible to achieve a deficit at the end of the year that is a convex combination of the planned and the realized deficit. to overcome possible concerns regarding this particular approach, we provide a flexible sensitivity formula. instead of assuming equal weights, we tried introducing a weighting parameter  such that sp ttt  )1( ~ˆ  where t ~ denotes the realized deficit in period t without intervention. it is easy to verify that the adjustment rules for d under both rules are linear in  . the derivative with respect to  therefore provides a measure of the flexibility and advantages of this adjustment rule (relative to our bench-mark with equal weights). constant slope adjustment in the case of constant slope adjustments, the derivative is european journal of government and economics 1(1) 57 ) ~ ( )1( 1 2 3 1 1 1 )1()1( 3 2 1 3 3 2 1 3 1 1 sp ttt ts n sn t t ts n s n s t ts n s g g g gg d                              the change in d is     d where  denotes the difference in  from 0.5, our benchmark. a simple estimate of that change can be obtained by setting nominal growth rates to zero. in this case the formula reduces to ⅔ ( )sp t t   which, in the case of germany, would imply a sensitivity of 1.1 %. if we set  equal to zero, i.e. a complete correction as soon as the slippage is first detected, we would get a slope coefficient of 1.85 % which is almost exactly equal to the numerical value of 1.83 % used earlier. for the numeric example in this section, we assume that for the remaining six months of the year after the forecast has been made, the government can implement policy measures, e.g. a budget freeze, so that the deficit remains in accordance with the submitted sp plan. in the case of germany, every month of slippage from the sp plan will increase the slope coefficient by approximately 0.09.16 this shows that there are quantitative effects from altering the impact range of early intervention. but, compared to not intervening, the gains always remain substantial. additive constant adjustment in the case of the additive constant adjustment the derivative becomes ) ~ ( )1( 1 1 1 )1( 3 2 1 3 3 1 1 sp ttt ts n sn t t ts n s g g g d                   . if we set nominal growth rates to zero as before, we get ⅓ ( )sp t t   reflecting the equal spread of the corrections over the three adjustment periods. for germany, the sensitivity in our example would be 0.55 %, i.e. we would get an adjustment parameter of 0.53 % if there were a complete correction in the period in which the slippage were first detected, instead of 0.51% as in our calculations. once again there is sensitivity with respect to the impact range of early interventions; but, as in the case of the constant slope adjustment scheme, the gains of intervening early are always substantial. example 2: the case of italy in 2002 as a second case, we take italy in 2002. the italian government had submitted the following plan (see table 6) to the european commission in november 2001: 16 slippage in this case is measured relative to our assumption of six months of compliance. european journal of government and economics 1(1) 58 table 7. italian government deficit plan starting in 2002, submitted in november 2001 2002 2003 2004 2005 deficit -0.5% 0.0% 0.0% 0.2% gdp growth 4.76% 4.85% 4.54% 4.54% debt 103.12% 98.35% 94.07% 89.78% deficit and growth rates are taken from the plan submitted by the government. the debt follows from own calculations based on a debt ratio of 107.5% for 2001. again, as figure 1 showed, the plan had to be revised in the year after implementation due to additional fiscal slippage; so early interventions would again have been beneficial. for this plan, once again we take the deficit numbers and the nominal growth rates as given. the debt level in 2001 is taken to be 107.5 % of gdp, and we update the debt ratio according to the rule given in (2).17 based on the forecast for italy made in june/july 2002, we expect a deficit for 2002 of 3.88%.18 we now derive the debt ratio of the italian government using the realization for 2002 and assume that this is a one-time deviation from the sp plan (see table 7). this is intended to resemble a situation in which there are no intraannual interventions and the full deficit is realized. next, we consider the situation where, after the forecast is made, early intervention measures are taken. we make the same assumption as before; that the measures taken are effective to the point that, at the end-of-year, the deficit is the convex combination with equal weights of the sp target and the realization for that year. in that case, the deficit for 2002 new 2002̂ becomes %76.1%)5.0%01.3(5.0ˆ 2002 new table 8. italian government deficit plan after a one-time deviation in 2002 2002 2003 2004 2005 deficit -3.01% 0.0% 0.0% 0.2% gdp growth 4.76% 4.85% 4.54% 4.54% debt 105.63% 100.74% 96.37% 91.98% debt (planned) 103.12% 98.35% 94.07% 89.78% deficit and growth rates are taken from the plan submitted by the government in november 2001, and no corrections are taken with respect to these numbers. the debt follows from own calculations based on a debt ratio of 107.5% for 2001. the additive correction approach first, we derive again the correction under the level shift as in the german example above. if we solve for d we get d=0.0038. this means that the italian government would have to cut the deficit by about 0.4% of gdp for each of the remaining years. this plan can be seen in table 8. in this scenario, with early interventions the government would escape both the sgp’s excessive deficit scrutiny and the commission’s significant improvement test. however, the early deficit reductions are severe: 1.24% of gdp in 2002, and 2.14% in 2003, but very small thereafter. 17 the updated debt ratios are lower than the debt ratios from the plan submitted by the italian government. however, to get a consistent benchmark case, we allow debt to evolve according to the proposed sp rule. 18 it will turn out that the actual deficit will be 3.01% for 2002. details are available upon request. european journal of government and economics 1(1) 59 table 9. italian government deficit plan after a one-time deviation and revision in july 2002 2002 2003 2004 2005 deficit -1.76% 0.38% 0.38% 0.58% gdp growth 4.76% 4.85% 4.54% 4.54% debt 104.38% 99.16% 94.47% 89.78% debt (planned) 103.12% 98.35% 94.07% 89.78% deficit and growth rates are taken from the plan submitted by the government in november 2001, and no corrections are taken with respect to these numbers. the debt follows from own calculations based on a debt ratio of 107.5% for 2001. the alternative would be not to take early intervention measures and to let the deficit take its course for 2002. in that case the debt at the end of 2002 would be 105.63%. based on this, we can calculate the level shift, d, that would be necessary to reach the 89.78% debt target in 2005. we get d=-0.0076. this implies that the government would have to double its deficit reductions to 0.76% of gdp, each year for three years, instead of only 0.38% had early intervention measures been taken. however, it must be kept in mind that the government already had to reduce the deficit by a large margin in 2002 in the early interventions scenario. the deficit in 2002 with early interventions is 1.76%, whereas it is 3.01% without such interventions; the difference of 1.25% being spread over the remaining three years. on the other hand, the deficit reductions of 0.38% for three years in the early interventions case are less than the 1.25% of additional cuts imposed after 2002 when there are no early interventions – but not by much. those late cuts are used to pay for extra spending in 2002. thus, in this plan, the government avoids the savage cuts in the first two years of the early interventions solution by trading cuts of 1.24% in net spending saved in 2002, and 1.38% of gdp saved in 2003, for larger cuts (of 0.76% and 0.56%) in 2004-5. the difference in the cuts required under the two plans being made up by the growth in gdp during the life of the plan. the slope correction approach the second approach is the constant slope adjustment path taken to the plan’s horizon. the slope is chosen such that the government can still satisfy the original debt target in 2005. we need to find deficits new 2003̂ , new 2004̂ and new 2005̂ such that we meet the debt target of 89.78% by 2005. if we evaluate d for italy in 2002, we get d=0.0105 with early interventions. this means the government has to decrease its deficit from year to year, for the next three years, by 1.05% of annual gdp. the new plan can be seen in table 9. the back-loading is more obvious here than in the german example; and, with an extra 1% of gdp in budget cuts in 2005, and a correspondingly larger gain in frontloading relief, the temptation for time inconsistent behaviour compared to the early intervention solution in table 7 will be large. this is guaranteed to raise doubts about the credibility of this as a consolidation plan. table 10. italian government deficit plan after a one-time deviation and revision in july 2002 2002 2003 2004 2005 deficit -1.76% -0.67% 0.42% 1.51% gdp growth 4.76% 4.85% 4.54% 4.54% debt 104.38% 100.21% 95.44% 89.78% debt (planned) 103.12% 98.35% 94.07% 89.78% deficit and growth rates are taken from the plan submitted by the government in november 2001, and no corrections are taken with respect to these numbers. the debt follows from own calculations based on a debt ratio of 107.5% for 2001. european journal of government and economics 1(1) 60 if, instead, we consider the case of no early interventions, then the constant slope regime would require d=0.0183 to meet the 2005 debt target. the resulting plan can be found in table 10. this time the budget corrections are substantially larger than those with early interventions. similarly, the back-loading element is large: the budget would have to be cut by 2.2% more of gdp in 2005 than in the scheme with additive corrections, and by 2.4% more than with both additive corrections and early interventions. the frontloading relief is equally large: 1.24% in 2002 and 1.3% in 2003. it is hard to believe that time inconsistent behaviour would not appear in such a case. as a result this plan is unlikely to have much credibility in the markets. table 11. italian government deficit plan after a one-time deviation and revision in july 2002 2002 2003 2004 2005 deficit -3.01% -1.12% 0.78% 2.67% gdp growth 4.76% 4.85% 4.54% 4.54% debt 105.63% 101.86% 96.65% 89.78% debt (planned) 103.12% 98.35% 94.07% 89.78% deficit and growth rates are taken from the plan submitted by the government in november 2001, and no corrections are taken with respect to these numbers. the debt follows from our own calculations based on a debt ratio of 60% for 2001. measuring the gains from early intervention directly it is relatively easy to derive a monetary measure of the gains to be had from early interventions. here we use the difference in the debt burden at the end of the current sp plan starting with or without early interventions. that takes out any differences in the subsequent fiscal corrections which depend on the type of correction method chosen. the gains in monetary terms a) for germany: we assume that, after the slippage, the government returns to the fiscal contractions in its original sp plan, first having made an early intervention and then without. the difference in outcomes is easily read off from table 3. we see that, for germany, the gain of early intervention would have been a difference in debt ratios in 2005 of 1.46%. if we take a nominal interest rate of 5%, this would increase the government deficit for all future periods by 0.073% of gdp. for germany, this would mean additional interest payments every year of about 1.64 billion euros at 2005 prices. this is equivalent to the entire central government budget for housing, urban development, and regional planning (1.794 billion euros) being lost forever.19 b) for italy: we follow the same approach and compare the debt ratio at the end of the sp plan in 2005 in the case where early intervention measures have been taken, to the case where they have not been taken. the difference can be read off from table 8, and we see that for italy the difference in debt ratios in 2005 would have been 2.20%. again, assuming a nominal interest rate of 5%, this means a permanent raise in interest payments of 0.11% of gdp. for italy, this is equivalent to a loss in public spending of 1.57 billion euros each year for ever. the gains in terms of budget management the budgetary execution is easier when large corrections can be avoided. in the case of the additive correction approach, the gains from early intervention can be read off from the adjustment factor d that captures the value of reducing the fiscal 19 this number is taken from http://www.bundesfinanzministerium.de/bundeshaushalt2005/pdf/vorsp/zyubfkt.pdf. european journal of government and economics 1(1) 61 contractions in the following years of the sp plan. in the constant slope approach, the change in the fiscal impulse can be derived as the difference of the corrected plan and the original sp plan. we set out the net fiscal contractions or expansions implied for germany, relative to the original sp plan, in table 11. these figures show that the two approaches offer different schedules for correcting fiscal slippages. both allow quite a large amount of extra spending in the first year of the plan. but then they differ. the constant slope plan offers two additional years of extra spending, but then demands a large cut in net spending. the additive constant approach, by contrast, offers small cuts spread evenly over the three correction years. for italy, the net fiscal contractions or expansions implied, relative to her original sp plan are set out in table 12. qualitatively the results resemble those for germany. they imply small but equally spread fiscal contractions under the additive approach; and but positive impulse in the first two periods, followed by a small contraction in 2004 and a sharp contraction in the last period under the constant slope approach. table 12. change in the fiscal impulse of the german in reaction to a additive constant and constant slope revision of the original sp plan to meet the debt target in 2005 2002 2003 2004 2005 deficit planned -2.00% -1.00% 0.0% 0.0% additive constant -2.82% -0.75% 0.25% 0.25% constant slope -2.82% -1.46% -0.10% 1.26% fiscal impulse additive constant +0.82% -0.25% -0.25% -0.25% constant slope +0.82% +0.46% +0.10% -1.26% change in the fiscal impulse of the german in reaction to a additive constant and constant slope revision of the original sp plan to meet the debt target in 2005. table 13. change in the fiscal impulse of the italian in reaction to a constant slope and constant slope revision of the original sp plan to meet the debt target in 2005 2002 2003 2004 2005 deficit planned -0.5% 0.0% 0.0% 0.2% additive constant -1.76% 0.38% 0.38% 0.58% constant slope -1.76% -0.67% 0.42% 1.51% fiscal impulse additive constant +1.26% -0.38% -0.38% -0.38% constant slope +1.26% +0.67% -0.42% -1.31% change in the fiscal impulse of the italian in reaction to a constant slope and constant slope revision of the original sp plan to meet the debt target in 2005. elsewhere in the eurozone up to now, we have looked only at two countries in two specific time periods. in this section we do the same analysis for the eurozone countries in all years for which sp plans exist and are publicly available. to get a comprehensive measure, and to avoid an overload of individual country detail, we average the fiscal impulses of the different adjustment plans over all the cases back to 2000 where forecasting deficit slippages demanded early intervention. the results are presented in table 13 for both additive and constant slope approaches. these results confirm that what we saw above for germany and italy in their 2002 plans does in fact hold more generally. the differences between the left and right european journal of government and economics 1(1) 62 panels show that there are clear and substantial advantages, in terms of both lower debt levels and permanent interest savings, in every case to instituting early corrections. the changes in the overall run of fiscal policies required to make the necessary budget corrections is almost always smaller – often substantially smaller – with early interventions; most obviously in germany, italy, the netherlands and belgium. in most cases the changes are halved if early action is taken; with those changes bunched in the final year in the slope adjustment method is used, but evenly spread out if the additive method is used. table 14. the average changes in the fiscal impulse relative to the sp plan for the additive constant and the constant slope approaches with and without early interventions with early intervention without early intervention country intervention t t+1 t+2 t+3 t t+1 t+2 t+3 belgium slope 0.49 0.17 -0.17 -0.43 0.97 0.35 -0.31 -0.88 additive 0.49 -0.15 -0.15 -0.15 0.97 -0.30 -0.30 -0.30 germany slope 0.51 0.30 -0.11 -0.62 1.02 0.48 -0.26 -1.10 additive 0.51 -0.16 -0.16 -0.16 1.02 -0.31 -0.31 -0.31 spain slope 0.13 0.04 0.01 -0.15 0.26 0.08 -0.03 -0.27 additive 0.13 -0.04 -0.04 -0.04 0.26 -0.08 -0.08 -0.08 france slope 0.38 0.08 -0.11 -0.30 0.75 0.22 -0.22 -0.66 additive 0.38 -0.12 -0.12 -0.12 0.75 -0.23 -0.23 -0.23 italy slope 0.99 0.44 -0.26 -1.02 1.98 0.80 -0.54 -1.94 additive 0.99 -0.30 -0.30 -0.30 1.98 -0.60 -0.60 -0.60 netherlands slope 1.08 0.35 -0.38 -0.90 2.15 0.74 -0.68 -1.90 additive 1.08 -0.33 -0.33 -0.33 2.15 -0.65 -0.65 -0.65 austria slope 0.26 0.06 -0.09 -0.19 0.51 0.15 -0.17 -0.43 additive 0.26 -0.08 -0.08 -0.08 0.51 -0.16 -0.16 -0.16 portugal slope 0.14 0.11 0.27 -0.47 0.29 0.16 0.23 -0.60 additive 0.14 -0.04 -0.04 -0.04 0.29 -0.09 -0.09 -0.09 the average changes in the fiscal impulse relative to the sp plan for the additive constant and the constant slope approaches with and without early interventions. similarly, the required changes are almost always much smaller (even in earlier years) if the additive adjustment method is used. in fact the changes are typically 3 to 5 times smaller with additive rather than slope adjustments, most notably in spain, germany, italy and portugal (possibly belgium and the netherlands). it is important to bear in mind that, although these results are qualitatively the same as those for the actual budget corrections discussed in sections 6 and 7, the figures in table 13 show the changes in overall fiscal policy which those budget corrections cause when superimposed on the sp/mto plans. and those are the changes in policy which the public sees and judges the policymakers by. table 15. the additional interest payments when no corrections are made, using the additive constant approach country t t+1 t+2 t+3 belgium r∆d -0.0243 % -0.0307 % -0.0370 % -0.0430 % germany r∆d -0.0255 % -0.0324 % -0.0391 % -0.0455 % spain r∆d -0.0065 % -0.0081 % -0.0096 % -0.0110 % france r∆d -0.0188 % -0.0238 % -0.0287 % -0.0335 % italy r∆d -0.0494 % -0.0623 % -0.0746 % -0.0864 % netherlands r∆d -0.0538 % -0.0676 % -0.0810 % -0.0940 % austria r∆d -0.0128 % -0.0163 % -0.0196 % -0.0227 % portugal r∆d -0.0071 % -0.0089 % -0.0105 % -0.0121 % the additional interest payments when no corrections are made, using the additive constant approach. the term r�d gives the government spending, expressed as a share of gdp that would need to be devoted to extra interest payments if no adjustments at all were taken. in this case, debt targets are not met and higher interest payments would be necessary for ever. european journal of government and economics 1(1) 63 table 16. the additional interest payments when no corrections are made, using the constant slope approach country intervention t t+1 t+2 t+3 belgium r∆d -0.0243 % -0.0147 % -0.0224 % -0.0430 % germany r∆d -0.0255 % -0.0096 % -0.0149 % -0.0455 % spain r∆d -0.0065 % -0.0043 % -0.0037 % -0.0110 % france r∆d -0.0188 % -0.0139 % -0.0189 % -0.0335 % italy r∆d -0.0494 % -0.0250 % -0.0369 % -0.0864 % netherlands r∆d -0.0538 % -0.0337 % -0.0510 % -0.0940 % austria r∆d -0.0128 % -0.0095 % -0.0137 % -0.0227 % portugal r∆d -0.0071 % -0.0015 % 0.0120 % -0.0121 % the additional interest payments when no corrections are made, using the constant slope approach. the term r∆d gives the government spending, expressed as a share of gdp, that would need to be devoted to extra interest payments if no adjustments at all were taken. debt targets are not met and higher interest payments would be necessary for ever. looking at the figures in tables 14 and 15, we can see that the interest payment consequences of not making early corrections are indeed small (justifying our having left them out of consideration until this point) – just a fraction of a percent in each case, even in italy and the netherlands. moreover the extra interest payments are not different between the two correction methods. again it was reasonable to ignore them. however, they build up over time and last for a long time since the corrections are sufficient only to get us back to the planned debt levels. conclusions in this paper we have accomplished two things. firstly, we have investigated the gains to be made, in terms of monitoring and correcting excessive deficits, by using cash data rather than accruals data to monitor the current fiscal position; and, given that information, by computing the early interventions needed to head off any excess deficits as they emerge; and secondly, we have also examined and compared two different strategies for correcting excess deficits that may have already emerged, such that the debt ratio returns to some pre-specified level at the end of the designated planning period. independently of the measure used, our results show that the gains from early intervention are certainly not negligible for governments committed to reducing their debt. what is different between the different plans is the allocation of the corrections over time. both the constant slope and the additive constant plans require a lot of effort at the moment a fiscal slippage is detected in order to avoid further slippages in the remainder of the year and in subsequent years. this means that an important part of the adjustment has to take place in the first year. after that, the two approaches differ sharply in the timing of the corrections. the additive approach typically requires a second costly adjustment after the early intervention. one could however argue that this would be a less serious problem if those corrections could be implemented in a package together with the early intervention measures and therefore avoid further discussion in the political process. the advantage of the constant slope approach is that it spreads the adjustment steps equally across the horizon of the sp plan. that approach may therefore be politically easier to implement. however, it also requires an especially strong and committed government to overcome the (political) temptation to use the surplus that occurs towards the end of the plan for opportunistic spending instead of debt reductions; for example, close to elections or as debt or deficit consolidation fatigue begins to set in among the politicians. in our discussion of the different adjustment paths available to a fiscally consolidating government, we have emphasised the gains of early intervention. in that context we have stressed the importance of accounting for the interaction between deficit and debt targets, and the danger of time consistency in the european journal of government and economics 1(1) 64 associated consolidations. the time consistency problem in particular seems to be empirically relevant and strategically important. but it has received very little attention in the literature on the monitoring and control of fiscal policies. references alesina, alberto f. and silvia ardagna (2009) ‘large changes in fiscal policy: taxes versus spending’, nber working paper, 15438 [published in tax policy and the economy, 24, nber 2010]. alessandrini, pietro, michele fratianni, andrew h. hallett and andrea f. presbitero (2012) ‘external imbalances and financial fragility in the euro area’, mofir discussion paper, 66, polytechnic university delle marche, italy. beetsma, roel and massimo giuliodori (2008) ‘on the relationship between fiscal plans in the european union: an empirical analysis based on real time data’, journal of comparative economics, 36(2): 221-241. beetsma, roel, massimo giuliodori and peter wierts (2009) ‘planning to cheat: eu fiscal policy in real time’, economic policy, 24(60): 753–804. bernoth, k., andrew hughes hallett and john lewis (2012) ‘the cyclicality of automatic and discretionary fiscal policy: what can real time data tell us?’, macroeconomic dynamics (forthcoming) blanchard, olivier (2010) ‘imf survey on “rethinking macroeconomic policy’, url: www.imf.org/external/pubs/ft/survey/so/2010/int021210a.htm cimadomo, jacopo. (2008) ‘fiscal policy in real time’, ecb working paper series, 919. corsetti, giancarlo (2012), ‘has austerity gone too far?’ voxeu (voxeu.org), (consulted 2 april 2012) delong, j. bradford and lawrence h. summers (2012) ‘fiscal policy in a depressed economy’, paper presented at the spring 2012 brookings panel on economic activity, 20 march 2012. european commission (2011) ‘public finances in emu-2011’, european economy 3/2011, dgii (ecfin), european commission, brussels, september 2011. golinelli, roberto and sancro momigliano (2008) ‘the cyclical response of fiscal policy in the euro area. the cyclical reaction of fiscal policies in the euro area: the role of modelling choices and data vintages’, fiscal studies, 30(1):39–72. hughes hallett, andrew (2008a). coordination without explicit cooperation: monetary-fiscal interactions in an era of demographic change. economic papers 305, ecfin, european commission, brussels, february 2008. hughes hallett, andrew (2008b). debt targets and fiscal sustainability in an era of monetary independence. international economics and economic policy, 5, 165187. hughes hallett, andrew and svend h. jensen (2011) ‘stable and enforceable: a new fiscal framework for the euro area’, international economics and economic policy, 8, 225-245. hughes hallett, andrew, rasmus kattai and john lewis (2012) ‘how reliable are cyclically adjusted budget balances in real time?’, contemporary economic policy, 30(1): 75-92 hughes hallett andrew and svend h. jensen (2012) ‘fiscal governance in the euro area: institutions vs. rules’, journal of european public policy, 19(5): 646664. european journal of government and economics 1(1) 65 hughes hallett, andrew, giovanni di bartolomeo and nicola acocella (2012a) ‘a general theory of control-ability and expectations anchoring for small open economies’, journal of international money and finance, 31(2): 397-411. hughes hallett, andrew, nicola acocella and giovanni di bartolomeo (2012b) ‘expectations dynamics: policy announcements and limits to dynamic inconsistency’, studies in nonlinear dynamics and econometrics, 16(2). leeper, eric (2009) ‘anchoring fiscal expectations’, reserve bank of new zealand bulletin, 72(3): 7-32. morris, richard, hedwig ongena and ludger schuknecht (2006) ‘the reform and implementation of the stability and growth pact’ ecb occasional paper series, 47, june 2006. neumann, manfred (2012) ‘too early to sound the alarm’, voxeu. url (consulted 17 april 2012) : www.voxeu.org. onorante, luca, diego pedregal, javier pérez, and sara signorini (2008) ‘the usefulness of infra-annual government cash data for fiscal forecasting in the euro area’, ecb working paper series, 901, may 2008. pedregal, diego and javier pérez (2008) ‘should quarterly government finance statistics be used for fiscal surveillance in europe?’, ecb working paper series, 937, september 2008. pérez, javier (2007) ‘leading indicators for euro area government deficits’, international journal of forecasting, 23(2): 259 – 275. von hagen, jürgen and guntram wolff (2006) ‘what do deficits tell us about debt? empirical evidence on creative accounting with fiscal rules in the eu’, journal of banking and finance, 30(12): 3259–3279. woodford, michael (2005). central bank communication and policy effectiveness, in the greenspan era: lessons for the future. kansas city fed, kansas city, mo. vol.8 • no.2 2019 issn: 2254-7088 european journal of government and economics 8(2), december 2019. european journal of government and economics    issn: 2254‐7088  number 8, issue 2, december 2019 doi: https://doi.org/10.17979/ejge.2018.8.2 why regions fail (or succeed). the role of government institutions in the long-run 114-144 doi: https://doi.org/10.17979/ejge.2019.8.2.4989 filippo bonanno restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case 145-160 doi: https://10.17979/ejge.2019.8.2.5478 joão ricardo catarino and ricardo de moraes e soares fdi in selected developing countries: evidence from bundling and unbundling governance 161-188 doi: https://doi.org/10.17979/ejge.2019.8.2.4970 simplice asongu the relationship between corporate tax rate and economic growth during the global financial crisis: evidence from a panel var governanceand domestic investment in africa 189-202 doi: https://10.17979/ejge.2019.8.2.5074 gamze oz-yalaman the impact of crd iv on bank lending 203-217 doi: https://doi.org/10.17979/ejge.2019.8.2.4656 matias huhtilainen european journal of government and economics 8(2), december 2019, 203-217 european journal of government and economics issn: 2254-7088 the impact of crd iv on bank lending matias huhtilainena* a university of eastern finland business school, finland * corresponding author at: matias.huhtilainen@uef.fi article history. received 5 november 2018; first revision required 28 may 2019; accepted 12 november 2019. abstract. this study addresses the post-financial crisis eu banking regulation reform crd iv. the specific focus is on the relationship between increased capital requirements and the subsequent change in both supply and the price of bank credit. this study employs a twofold data consisting of a panel of finnish unlisted savings and cooperative banks’ key figures over the period 2002-2018 and a representative survey conducted with personnel of finnish institutions. in addition to the consistent finding in regards to the effect of bank profitability as well as fairly consistent findings in regards to the effect of bank size and gdp growth, the key finding suggests a slight decrease in loan supply under the crd iv. keywords. basel iii; capital requirements; crd iv; lending jel codes. g28; g21; k22 doi. https://doi.org/10.17979/ejge.2019.8.2.4656 1. introduction the somewhat informal group of regulators and central bankers that had been collaborating prior to the 2008 financial crisis, became more formal after the establishment of the financial stability board (fsb) in 2009. the fsb was handed the responsibility to coordinate the work at the international level and together with the basel committee on banking supervision (bcbs), they prepared the first draft of the basel iii rules. the draft was first endorsed by the group of 20 in november 2010 and then finalized later in december (howarth & quaglia, 2016; walker, 2011). basel iii effectively imposes minimum standards for capital requirements, supervision and market discipline in order to provide “a foundation for a resilient banking system that will help avoid the build-up of systemic vulnerabilities”. (bank of international settlements, 2017). the european union has participated in and advocated the regulatory reform. in consequence, the eu implemented the basel iii framework into the eu legislation through capital requirements directive 2013/36/eu and the capital requirements regulation (eu) no 575/2013. together they form a regulatory package known as crd iv, which replaced the former directives 2006/48/ec and 2006/49/ec. the provisions have applied to institutions since 1 january 2014, though the scheduled implementation phase-in period extends to 2019. as basel iii rules are, in essence, internationally-agreed industry standards, they rather provide a basis on which the eu is building its own regulatory and supervisory infrastructure. through single rulebook, the eu seeks to facilitate banking sector consolidation, increase transparency m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 204    and enhance cross-border activity in order to expand financing capabilities and reduce intermediation costs. the european commission (2016) has depicted deeper integration to be a necessary action to restore market confidence and solve those shortcomings that were first exposed by the financial crisis and then by the subsequent euro area debt problem. meanwhile, the ec (2016) has also urged that stability should not come with the price of economic stagnation. instead, the regulatory objectives need to be achieved in a way that is fully supportive of the ongoing recovery in the eu. the ec further noted that despite technological disruption within the lending business, banks continue to act as the key source of funding for businesses and households. for banks to serve this function, the regulatory environment should in a proportionate manner consider banks' complexity, size, business profile and specificities. in accordance, the purpose of this study is to analyze the impact of the renewed eu capital requirements on bank lending in the context of the finnish banking sector. the empirical strategy relies on a twofold data, and the specific aim of this study is provided by the following research problems: 1. the relationship between increased capital requirements and the corresponding change in credit supply. 2. the relationship between increased capital requirements and the corresponding change in price of credit. the paper is structured as follows. the next chapter introduces the main elements of capital requirements under the crd iv regulation. datasets and methods are described in more detail in chapter 3. results are reported and discussed in chapter 4, after which the paper is concluded in chapter 5. 2. the eu capital rules revisited crd iv imposes general requirements regarding the quality and the minimum level of capital. as reflected by table 1, the qualitative approach allocates capital into different layers in order to reflect differences in sustainability, liquidity and predictability between different asset classes. consequently, the two main components are tier 1 and tier 2. the preferred component tier 1 is further divided into common equity tier 1 (cet1) and an additional tier 1 (at1). the mandatory capital components are supplemented with macroprudential capital buffers, whose implementation depends on certain institution and market environment-related factors. the second element of the capital framework constitutes of liquidity regulation. the importance was emphasized by the financial crisis during which the acute problem at first was not the insufficient amount of capital, but the increasing uncertainty, which forced banks to become defensive and in consequence, hoard liquidity. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 205    table 1. a review of the crd iv capital rules. capital component amount, % of riskweighted assets eligible capital cet1 4,5 % cash, common shares issued by the bank, retained earnings, other comprehensive income and reserves at1 1,5 % perpetual paid-up capital instruments and their associated share premium accounts. assets shall not be purchased by the institution nor be subject to any arrangement that enhances the seniority of the claim by the institution or under the instruments in insolvency or liquidation tier 1 6 % tier 2 2 % capital instruments and subordinated loans and share premium accounts. tier 2 instruments are issued, or the subordinated loans are raised and fully paid-up, but not purchased nor funded either directly or indirectly by the institution. the claim must be wholly subordinated to the claims of all non-subordinated creditors, and should not be secured or subject to a guarantee that enhances the seniority of its claim total 8 % additional capital buffers capital conservation buffer 2,5 % cet1 capital countercyclical capital buffer* 0-2,5 % cet1 capital g-sii buffer** 1-3,5 % cet1 capital o-sii buffer*** 0-2 % cet1 capital systemic risk buffer**** 0-5 % cet1 capital notes. *implemented by national authority based on (excessive) aggregate credit growth, **mandatory buffer for banks that are identified by the relevant authority as globally systemically important institutions, ***optional buffer on “other” systemically important institutions: either domestically important or eu important institutions, ****member state may introduce a systemic risk buffer for the sector or one or more subsets of the sector, in order to prevent and mitigate systemic or macro-prudential risks. the buffer may exceed 5 % should circumstances call for it. as a result, banks found themselves unable to honor their obligations in frozen interbank markets. in order to cope with the situation, banks engaged in coincidental fire sales of nonliquid assets, which generated a steep fall in asset prices (aldasoro et al., 2017; berrospide, 2013). to address the witnessed shortfalls, the eu adopted two specific measures to correspond with the basel work. first, bscs introduced liquidity coverage ratio, which aims to provide protection against severe short-term liquidity shocks. accordingly, banks are now required to hold high quality liquidity assets1 (“hqla”) against anticipated net outgoing cash flows over a 30-day severe stress period. the ratio shall equal to or exceed 100 % in any given 1 the category hqla includes assets that enjoy sufficient trading volume, low volatility, transparent pricing and post-trade information, and are not correlated with high-risk asset classes. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 206    time. the second tool, net stable funding ratio, has a time horizon of one year. as banks are exposed to funding liquidity risk arising from their funding structures and the maturity mismatch in their balance sheet, nsfr targets to secure a more sustainable long-term maturity structure of assets and liabilities (gobat et al., 2014). the nsfr is calculated as a ratio of available stable funding (asf) against the amount of required stable funding (rsf). the asf is defined as the portion of capital and liabilities expected to be reliable over the target time horizon. the amount of rsf is a function of liquidity characteristics and residual maturities of the various assets held by the institution. thirdly, the bcbs introduced a completely new non-risk based leverage constraint to complement the remaining risk-based capital framework. bcbs (2014) concluded that the underlying cause for the meltdown had been the build-up of excessive onand off-balance sheet leverage in the banking system. following the adoption, the binding 3 % leverage ratio (lr) became fully applicable in the eu in january 2018. in accordance with the bcbs proposal, lr is calculated as the amount of cet1 capital against bank’s all non-risk weighted assets. indeed, this contradicts other capital provisions, as lr does not consider any risk weights or risk models. furthermore, banks are expected to meet the requirement in addition to their riskbased capital requirements. 3. data and methods 3.1. key financial figures we estimate the effect of several bank-specific and macroeconomic factors on banks’ annual loan growth by employing a panel data of finnish unlisted savings and cooperative banks’ key figures over the period 2002-2018. the specific interest is on the regulation dummy variable “crd iv”, which takes on the value of 0 for period 2002-2013 and 1 otherwise. although crd iv is an encompassing regulation, this study applies it solely as a proxy for “increased capital requirements”. the dependent variable “loan growth” is the annual logarithmic change of banks’ total loans to the public (excluding loans to other institutions). hence, the econometric part of this study is solely focused on the first research problem. while the macroeconomic data was retrieved from statistics finland (tilastokeskus), the bank key figure data was obtained from the finnish financial supervisory authority’s (finanssivalvonta) website. in total, 339 unique bank identifiers exist in the dataset, though only 188 of them are identified and measured each year (17 times). hence, while 38 savings banks and 279 cooperative banks existed in 2002, the corresponding numbers in 2018 totaled at “just” 23 and 182 banks, respectively. namely, the change in individual bank’s status occurs in two primary ways: either the bank is acquired by another bank, which absorbs the business and proceeds as a larger entity, or it joins a merger of equals, where a new bank is established after two or more banks come together and cease to exist as separate, independent entities. a dummy variable “consolidation” is used to control acquiring bank’s consolidated figures in order to avoid spurious jumps following the acquisition. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 207    table 2. variables and descriptive statistics. obs. mean standard deviation min max description dependent variable loan growth 4,361 .06788 .08083 -.2653 1.7481 annual change in bank lending independent variables consolidation 0 1 a dummy to capture jumps in acquiring bank’s figures after an absorption crd iv 0 1 bank regulation dummy: 0 = 20022013, 1 = 2014-2018 gdp growth rate 4,700 1.3530 3.0626 -8.3 5.2 annual percentage change interest rate 4,700 1.6300 1.5540 -.329 4.644 3 month euribor, annual average of the daily values household indebtedness 4,700 106.98 17.157 72.1 128.6 annual total assets 4,700 198494.7 421470.7 4008 1.01e+07 a proxy for size off-balance sheet items/assets 4,700 4.7969 2.1002 .31371 18.258 a proxy for off-balance sheet activity, % of total assets equity/assets 4,700 12.088 4.2965 .33770 29.742 equity capital, % of total assets return on assets 4,698 1.0753 .6413 5.1868 6.5713 operating income (pre-tax), % of total assets cost/income ratio 4,700 63.108 15.814 10.405 518.8 a proxy for efficiency note. the descriptive statistics do not take into account the panel data structure. that is, the means and standard deviations have been calculated across all banks in all years. table 2 presents the variables and descriptive statistics. we estimate the following baseline model: 𝛥𝑦 𝛼 𝑿′ , 𝛽 𝑿′ 𝛾 𝑾′ 𝜁 𝜇 [1] where 𝑿s include the once-lagged bank-specific and macroeconomic explanatory variables, 𝑾 is the set of dummy variables and 𝜇 is the error term. all bank-specific variables are in natural logarithms. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 208    3.2. survey the second part of the study tackles both research problems. the data is based on the relevant parts of a bank survey conducted in 2016 with personnel of finnish institutions. respondents included business area directors, risk officers and in one case, a chief executive officer. they represented commercial, cooperative and savings banks as well as one credit institution. three bank groups are involved in both datasets. the survey consists of four thematic parts and a total of 27 individual items. first, the respondents were asked to examine the change in loan terms for different lending scenarios. this study interprets the expression “loan term” as a proxy for loan supply. further, the second and third question sets considered changes in both supply and the price of credit. lastly, the respondents were asked to assess the importance of different factors on the recent development of their profitability. although the survey is near fully representative of the finnish banking sector in terms of aggregate capital and market value, the small n (= 7) provides the analysis is entirely qualitative. secondly, respondents’ answers shall not be considered an accurate and objective reflection of the banks’ actual strategic behavior. rather, the answers are suggested to reflect personal, subjective views on issues brought up in the survey. thus, the paper only reports descriptive statistics, including the median, mode and the range of the numerical values assigned to different answer options. results are discussed together with and analyzed against the findings from prior studies. 4. results 4.1. key financial figures at first, a naïve pooled ols estimator is employed. results are reported in table 3. since the acquisition of another bank is expected to increase the value of acquirer’s loan portfolio, it is fairly unsurprising that the dummy variable consolidation enters positive and statistically highly significant in each specification. interestingly though, the results suggest a negative relationship between bank size (proxied by total assets) and loan growth. the regulation dummy crd iv enters negative and statistically significant, suggesting a decrease in bank lending under the renewed regulatory environment. although this result does not necessarily contradict the european central bank (2016), it is nevertheless noteworthy that the ecb noted a shift among euro area banks from investment banking, wholesale lending and lending in higher-risk sectors towards retail businesses, as crd iv has made riskier business activities costlier. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 209    table 3. ols estimations. i ii iii iv v consolidation .07202*** .07342*** .08054*** .08017*** .07999*** (.01405) (.01400) (.01423) (.01413) (.01432) crd iv -.04720*** -.0355*** -.01289** -.00802 -.01415** (.00297) (.00387) (.00443) (.00458) (.00510) gdp growth rate, t-1 .00193*** .00098** .00083* .00073* (.00034) (.00034) (.00034) (.00035) interest rate, t-1 .00410*** .00176 .00221* -.00048 (.00101) (.00101) (.00101) (.00118) household indebtedness, t-1 -.0012*** -.0012*** -.00107*** (.00008) (.00008) (.00009) total assets, t-1 -.00081 -.00478** -.00633*** (.00143) (.00170) (.00161) off-balance sheet items/assets, t1 .02018*** .01792*** (.00432) (.00386) equity/assets, t-1 .00382 -.00645 (.00441) (.00445) return on assets, t-1 .01012* (.00397) cost/income ratio, t-1 -.00535 (.01053) constant .07538*** .06252*** .20321*** .20948*** .26425*** (.00124) (.00253) (.01678) (.02262) (.05533) model diagnostics no. of obs. 4361 4361 4361 4361 4308 r-squared 0.0860 0.0974 0.1382 0.1471 0.1518 f 140.41*** 100.90*** 111.01*** 86.89*** 84.88*** root mse .0773 .07683 .0751 .07473 .07332 note. *, **, *** significant at the 5 %, 1 % and 0,1 % level, respectively. clustered robust standard errors reported in the parentheses. the results capture a negative and statistically highly significant coefficient of household indebtedness. however, solely based on these results alone, it is difficult to estimate to what degree this effect is related to supply and to what degree to demand side of bank lending. for other macroeconomic factors, the lagged gdp growth rate intuitively presents a positive effect on loan growth. this result is consistent, for instance, with aiyar et al. (2016), meriläinen (2016) and berrospide and edge (2010). on the contrary, the three-month euribor interest rate does not reveal any unambiguous effect. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 210    table 4. fixed effects estimations. i ii iii iv v consolidation .09460*** .09658*** .16903*** .16248*** .15968*** (.01926) (.01899) (.02893) (.02966) (.02978) crd iv -.05094*** -.0395*** .00128 .00290 -.00238 (.00308) (.00394) (.00513) (.00563) (.00605) gdp growth rate, t-1 .00190*** .00080* .00052 .00047 (.00034) (.00032) (.00033) (.00033) interest rate, t-1 .00428*** -.00183 -.00066 -.00265* (.00104) (.00115) (.00118) (.00130) household indebtedness, t-1 .00096** .00070* .00072* (.00034) (.00035) (.00036) total assets, t-1 -.1954*** -.1836*** -.1732*** (.02988) (.03016) (.02949) off-balance sheet items/assets, t1 .02442*** .02208*** (.00532) (.00488) equity/assets, t-1 .02604 .01139 (.01384) (.01263) return on assets, t-1 .00797* (.00392) cost/income ratio, t-1 -.00416 (.01043) constant .07489*** .06183*** 2.1897*** 1.9816*** 1.9245*** (.00137) (.00273) (.30497) (.31511) (.31501) model diagnostics no. of obs. 4361 4361 4361 4361 4308 r-squared (overall) 0.0846 0.0959 0.0039 0.0052 0.0069 f 140.28*** 95.65*** 88.72*** 67.99*** 65.16*** note. *, **, *** significant at the 5 %, 1 % and 0,1 % level, respectively. clustered robust standard errors reported in the parentheses. for the rest of bank-specific determinants, we confirm the findings by bustamante et al. (2019) and berrospide and edge (2010) who find that more profitable banks tend to grant more credit. meanwhile, the equity/assets ratio or the banks’ operational efficiency (cost/income ratio) do not exhibit any consistent effects. interestingly, the off-balance sheet activity (items as a percentage of total assets) enters positive and statistically highly significant in both specifications. since the variable is once-lagged, one possible explanation is that customers draw on their lines of credit, converting the previously unused commitments into on-balance sheet loans, thus increasing the value of outstanding loans at 𝑡 . however, this study unfortunately suffers from the lack of more detailed data on individual items within the total value of off-balance sheet exposures. next, we estimate the baseline model with a panel fixed effects estimator, which sweeps out the bank-specific heterogeneity by subtracting the cross-sectional mean from each observation. results are reported in table 4. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 211    the regulation dummy variable crd iv again suggests a negative effect on loan growth but does not remain statistically significant across the board. interestingly, the household indebtedness flips the sign in fixed effects estimations. as concluded earlier, it is difficult to assess whether these findings, taken together, indicate an increase in demand or a decrease in supply since we should also expect banks to avoid granting more loans to already indebted customers. this severe problem with identifying causal impact between the internal and external balance sheet shocks and the corresponding change in credit supply is discussed in peydro (2010) and lambertini and mukherjee (2016), who point out to the endogeneity bias between credit demand and credit supply. on the contrary, the dummy variable consolidation, bank size, off-balance sheet activity or profitability do not flip signs across estimations. the captured coefficient for bank size is consistent with fungáčová et al. (2014). similarly, in accornero et al. (2017), bank size enters negatively when bank fixed effects are included in the estimation. according to their interpretation, those banks that have grown “too big” are relatively less willing to extend new credit. on the other hand, cambacorta (2005) concludes the effect of bank size to be irrelevant on loan supply. 4.2. survey a rather rich research body is devoted to the relationship between capital requirements and the supply and price of bank credit. starting with credit supply, de jonghey et al. (2016) provide evidence on higher capital requirements corresponding with lower credit supply to corporations. in a similar vein, bridges et al. (2014) capture a relationship between increased capital requirements and a corresponding decrease in lending to private non-financial corporations. aiyar et al. (2014) conclude that u.k. banks to reduce lending in response to tighter capital requirements (1 percentage point increase in capital requirements reduced credit growth by 6.57.2 percentage points). mesonnier and monks (2015) estimate a 1 percentage point increase in core tier 1 ratio to be associated with a 1.2 percentage point reduction in credit growth. similarly, the results by noss and toffano (2016) suggest a decrease in lending subsequent to an increase in aggregate bank capital requirement, but that the effect is larger in corporate lending vis-à-vis to the household lending. in consequence, the bank lending capacity is of particular concern in the case of small and medium-sized enterprises, as they are typically the main contributors for economic growth and employment (european banking authority, 2012). although capital requirements do not have the objective to affect credit supply per se, the increased capital requirements may nevertheless have indirect effects on lending, should raising capital become (too) costly (de jonghey et al., 2016). in this vein, the survey first asked the respondents to assess whether their loan terms had tightened during the current decade. secondly, respondents were asked to assess the magnitude of regulatory reform’s direct impact on their loan terms. the results are reported in m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 212    table 5. the impact of renewed regulation on loan terms. loan portfolio as a whole loans for smes loans for large enterprises loans for households / personal loans shortterm loans longterm loans loan portfolio as a whole loans for smes loans for large enterprises loans for households / personal loans shortterm loans longterm loans n valid 7 6 5 7 6 6 6 5 4 6 5 5 missing 0 1 2 0 1 1 1 2 3 1 2 2 median 3 3 3 3 3 3 3,50 3 3,50 3 3 4 mode 3 3 3 3 3 3 4 3 3a 3 3 4 range 2 2 2 2 2 3 2 1 1 3 1 1 minimum 2 2 2 2 2 2 2 3 3 2 3 3 maximum 4 4 4 4 4 5 4 4 4 5 4 4 note. a. multiple modes exist. the smallest value is shown. at first, respondents were presented the following statement: “our loan terms have tightened notably during the current decade in the following cases”. for the latter part, the statement was: “the direct impact of regulation on our loan terms has been notable in the following cases”. answers were coded as 1 = “strongly disagree”, 2 = “disagree”, 3 = “neutral”, 4 = “agree”, 5 = “strongly agree”. table 5. accordingly, in case of both sme and large enterprise lending, the respondents were collectively neutral to slightly at one with the stance according to which the renewed regulation has had a notable impact on their loan terms. identical, consistent agreement is captured for both shortand long-term loans (last two items): due to the potential endogeneity bias discussed earlier, the respondents were asked to assess the impact of different factors on the possible change in their credit supply for the past three years. the answers are rather diversified without any clear consistency. hence, the previously voiced problems in regards to isolating the true regulatory effect are emphasized by the results reported in table 6. thirdly, the respondents were asked to estimate the impact of increased capital requirements on the price of credit in different lending scenarios. for instance, kisin and manela (2016) estimated a 0.3 basis points increase in lending rates subsequent to a 1 percentage point increase in capital requirements. king (2010) concluded a 1 percentage point increase in capital requirements and a corresponding 15 basis points increase in lending rates. consistent results are found in corbae and d'erasmo (2014), de resende et al. (2010) and hanson et al. (2011), although the latter ones propose the effect to be relatively small. in a similar vein, the responses shown in table 7 are somewhat indicative of the linkage between capital requirements and the price of borrowing, though in a lesser extent in the case of personal and household loans. table 6. impact of different factors on the possible change in credit supply. changes in demand side capital requirements under the renewed regulation changes in supply side n valid 7 7 7 missing 0 0 0 median 3 3 3 mode 3a 3 2 range 4 3 3 minimum 1 2 2 maximum 5 5 5 note. a. multiple modes exist. respondents were asked if “in the last three years, our lending volume has been negatively affected by the following reasons”. answers were coded as 1 = “strongly disagree”, 2 = “disagree”, 3 = “neutral”, 4 = “agree”, 5 = “strongly agree”. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 213    thirdly, the respondents were asked to estimate the impact of increased capital requirements on the price of credit in different lending scenarios. for instance, kisin and manela (2016) estimated a 0.3 basis points increase in lending rates subsequent to a 1 percentage point increase in capital requirements. king (2010) concluded a 1 percentage point increase in capital requirements and a corresponding 15 basis points increase in lending rates. consistent results are found in corbae and d'erasmo (2014), de resende et al. (2010) and hanson et al. (2011), although the latter ones propose the effect to be relatively small. in a similar vein, the responses shown in table 7 are somewhat indicative of the linkage between capital requirements and the price of borrowing, though in a lesser extent in the case of personal and household loans. the commencement of the crd iv, as well as this survey, coincided with somewhat weak macroeconomic and financial market conditions in finland and the eu. hence, in a further attempt to isolate the impact of different factors on banks’ business environment, the respondents were asked to assess several influences from the perspective of their profitability. as reflected by table 8, the respondents were particularly consistent regarding the negative impact of the low interest rate environment. interestingly, in the case of loan supply, the earlier econometric findings suggested that the effect of the three-month euribor rate was somewhat nonexistent. table 7. the impact of renewed regulation on the price of credit in different lending scenarios. loans to smes loans to large enterprises loans to households / personal loans short-term loans long-term loans n valid 6 6 7 6 6 missing 1 1 0 1 1 median 4 4 4 4 4 mode 4 4 4 4 4 range 2 2 3 2 2 minimum 3 3 2 3 3 maximum 5 5 5 5 5 note. respondents were presented with a claim according to which “capital requirements have increased the cost of borrowing in the following cases”. 1 = “strongly disagree”, 2 = “disagree”, 3 = “neutral”, 4 = “agree”, 5 = “strongly agree”. table 8. the assessment of different factors affecting banks’ business environment. economic development of households economic development of firms economic development of public sector developed or acquired product, technology or servicerelated innovations changing regulatory environment change in competitors' market share changes in most important reference rates n valid 6 6 6 6 6 6 6 missing 0 0 0 0 0 0 0 median 4 3,50 3 3,50 2 3 2 mode 4 2a 3 3a 2 3 2 range 3 3 2 5 2 2 1 minimum 2 2 2 0 1 2 1 maximum 5 5 4 5 3 4 2 note. a. multiple modes exist. the smallest value is shown. respondents were asked to assess the impact each specified factor has had on the bank’s profitability in the current decade. answers were coded as 1 = “highly negatively”, 2 = “somewhat negatively”, 3 = “neutral”, 4 = “somewhat positively”, 5 = “highly positively”. m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 214    5. discussion this study contributed to the literature on bank capital requirements in two ways. after controlling for several bank-specific and macroeconomic factors with panel data, the key result indicated a slightly negative relationship between increased capital requirements (proxied by the crd iv regulation dummy variable) and the finnish unlisted savings and cooperative banks’ credit supply. the study also captured consistent results in regards to the positive impact of both gdp growth and bank profitability on credit supply as well as somewhat consistent results in regards to the negative relationship between bank size and loan growth. in addition, the findings suggested that an increase in off-balance sheet items corresponds to an increase in credit supply. according to one possible explanation, the customers draw on their lines of credit, which converts the previously unused commitments into on-balance sheet loans in the subsequent period. secondly, this study conducted a representative survey with personnel of the finnish institutions. the results suggested that the renewed eu banking regulation has had a slight direct impact on banks’ corporate loan terms (for both smes and large enterprises), which was used as a proxy for credit supply. further, the respondents were collectively neutral to slightly at one with a claim according to which the increased capital requirements have increased the price of credit for smes and large enterprises. however, direct conclusions are avoided in this regards. the survey, in particular, suffered from certain and in some cases, quite severe limitations. on the other hand, the survey also revealed other factors that have affected the banking business as a whole and lending in particular. clearly, neither these nor prior findings implicate a regulatory failure. as noted earlier, the direct objective of the capital rules is not to influence the supply or the price of credit per se. rather, the current regulatory ambition is to increase systemic stability and address those shortcomings that led to the 2008 meltdown. nevertheless, the regulation still has an important role in capital and risk decisions (tanda, 2015). since banks react to external shocks by adjusting their business models, the policy analysis should always focus on second-order (and potentially counterproductive) effects. while some evidence suggests that higher capital requirements reduce excessive risk-taking (repullo, 2004) and that higher capital requirements correspond with a more resilient banking sector (basten and koch, 2015), the truly relevant question is whether regulation generates unintentional incentives to raise overall risk exposure (adrian et al., 2015; adrian and shin, 2014; blum, 1999), or incentives to shift towards shadow banking or off-balance sheet activities (goodhart, 2008; martin and parigi, 2013; plantin, 2015) or incentives to engage in regulatory arbitrage, that is, exporting the risk-taking to countries where regulation is less stringent (ongena et al., 2013). m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 215    references accornero, m., alessandri, p., carpinelli, l., & sorrentino, a.m. (2017). non-performing loans and the supply of bank credit: evidence from italy (banca d’italia occasional papers no. 374). retrieved from: https://www.bancaditalia.it/pubblicazioni/qef/20170374/index.html?com.dotmarketing.htmlpage.language=1. doi: https://doi.org/10.2139/ssrn.2954995 adrian, t., friedman, e., & muir, t. (2015). the cost of capital of the financial sector. retrieved from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2706614. doi: https://doi.org/10.2139/ssrn.2706614 adrian, t., & shin, h.s. (2014). procyclical leverage and value-at-risk. the review of financial studies 27(2), 373-403. doi: https://doi.org/10.1093/rfs/hht068 aiyar, s., calomiris, c.w., & wieladek, t. (2016). how does credit supply respond to monetary policy and bank minimum capital requirements? european economic review vol. 82, 142165. doi: https://doi.org/10.1016/j.euroecorev.2015.07.021 aiyar, s., calomiris, c.w., & wieladek, t. (2014). does macro-prudential regulation leak? evidence from a uk policy experiment. journal of money, credit and banking 46(1), 181214. doi: https://doi.org/10.1111/jmcb.12086 aldasoro, i., delli gatti, d., & faia, e. (2017). bank networks: contagion, systemic risk and prudential policy. journal of economic behavior & organization vol. 142, 164-188. doi: https://doi.org/10.1016/j.jebo.2017.05.022 basel committee on banking supervision (2017). high-level summary of basel iii reforms. retrieved from https://www.bis.org/bcbs/publ/d424_hlsummary.pdf basel committee on banking supervision (2014). basel iii leverage ratio framework and disclosure requirements. retrieved from https://www.bis.org/publ/bcbs270.pdf basten, c., & koch, c. (2015). higher bank capital requirements and mortgage pricing: evidence from the countercyclical capital buffer (ccb). bis working papers, 511. retrieved from: https://www.bis.org/publ/work511.htm. doi: https://doi.org/10.2139/ssrn.2473622 berrospide, j.m. (2013). bank liquidity hoarding and the financial crisis: an empirical evaluation. feds working papers, 2013-03. retrieved from: https://www.federalreserve.gov/pubs/feds/2013/201303/201303abs.html. doi: https://doi.org/10.17016/feds.2013.03 berrospide, j.m., & edge, r.m. (2010). the effects of bank capital on lending: what do we know, and what does it mean? international journal of central banking 6(4), 5-54. doi: https://doi.org/10.2139/ssrn.1670506 blum, j. (1999). do capital adequacy requirements reduce risks in banking? journal of banking & finance 23(5), 755-771. doi: https://doi.org/10.1016/s0378-4266(98)00113-7 bridges, j., gregory, d., nielsen, m., pezzini, s., radia, a., & spaltro, m. (2014). the impact of capital requirements on bank lending. bank of england working paper, 486. retrieved from: https://www.bankofengland.co.uk/-/media/boe/files/working-paper/2014/the-impact-ofcapital-requirements-on-bank-lending. doi: https://doi.org/10.2139/ssrn.2388773 bustamante, j., cuba, w., & nivin, r. (2019). determinants of credit growth and the bank lending channel in peru: a loan level analysis. bis working paper, 803. retrieved from: https://www.bis.org/publ/work803.htm. cambacorta, l. (2005). inside the bank lending channel. european economic review, 49 (7), 1737-1759. doi: https://doi.org/10.1016/j.euroecorev.2004.05.004 corbae, p.d., & d'erasmo, p. (2014). capital requirements in a quantitative model of banking industry dynamics. frp of philadelphia working paper, 14-13. retrieved from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2430778. doi: https://doi.org/10.21799/frbp.wp.2014.13 de jonghey, o., dewachter, h., & ongenax, s. (2016). bank capital (requirements) and credit supply: evidence from pillar 2 decisions. national bank of belgium working paper research, 303. retrieved from: https://econpapers.repec.org/paper/nbbreswpp/201610303.htm de resende, c., dib, a., & perevalov, n. (2010). the macroeconomic implications of changes m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 216    in bank capital and liquidity requirements in canada: insights from the boc-gem-fin. bank of canada discussion paper, 16. retrieved from: https://www.bankofcanada.ca/wpcontent/uploads/2010/12/dp10-16.pdf european central bank (2016). recent trends in euro area banks’ business models and implications for banking sector stability. retrieved from: https://www.ecb.europa.eu/pub/pdf/other/sfcfinancialstabilityreview201605.en.pdf european banking authority (2012). assessment of sme proposals for crd iv/crr. retrieved from: https://www.eba.europa.eu/documents/10180/16148/eba-sme-report.pdf european commission (2016). frequently asked questions: capital requirements (crr/crd iv) and resolution framework (brrd/srm) amendments. retrieved from: http://europa.eu/rapid/press-release_memo-16-3840_en.htm fungáčová, z., solanko, l., & weill, l. (2014). does competition influence the bank lending channel in the euro area? journal of banking & finance 49, 356-366. doi: https://doi.org/10.1016/j.jbankfin.2014.06.018 gobat, j., yanase, m., & maloney, j. (2014). the net stable funding ratio: impact and issues for consideration. imf working paper no. 14/106. retrieved from: https://www.imf.org/external/pubs/ft/wp/2014/wp14106.pdf. doi: https://doi.org/10.5089/9781498346498.001 hanson, s.g., kashyap, a.k., & stein, j.c. (2011). a macroprudential approach to financial regulation. journal of economic perspectives 25(1), 3-28. doi: https://doi.org/10.1257/jep.25.1.3 howarth, d., & quaglia, l. (2016). the political economy of european banking union. oxford, united kingdom: oxford university press. doi: https://doi.org/10.1093/acprof:oso/9780198727927.001.0001 king, m. (2010). mapping capital and liquidity requirements to bank lending spreads. bis working paperr, 324. retrieved from: https://www.bis.org/publ/work324.htm. doi: https://doi.org/10.2139/ssrn.1716884 kisin, r., & manela, a. (2016). the shadow cost of bank capital requirements. the review of financial studies 29(7), 1780–1820. doi: https://doi.org/10.1093/rfs/hhw022 lambertini, l., & mukherjee, a. (2016). is bank capital regulation costly for firms? – evidence from syndicated loans. the école polytechnique fédérale de lausanne working paper. retrieved from https://www.google.com/url?sa=t&rct=j&q=&esrc=s&source=web&cd=1&ved=2ahukewjr87 hk9bzeahvwsoskhrypa60qfjaaegqiarac&url=https%3a%2f%2fwww.aeaweb.org%2 fconference%2f2017%2fpreliminary%2fpaper%2f5ttrkznd&usg=aovvaw0zzce25alg ukoy4m31l41 martin, a., & parigi, b.m. (2013). bank capital regulation and structured finance. journal of money, credit and banking 45(1), 87-119. doi: https://doi.org/10.1111/j.15384616.2012.00563.x meriläinen, j-m. (2016). lending growth during the financial crisis and the sovereign debt crisis: the role of bank ownership type. journal of international financial markets, institutions and money, 41, 168-182. doi: https://doi.org/10.1016/j.intfin.2015.12.011 mesonnier, j-s., & monks, a. (2015). did the eba capital exercise cause a credit crunch in the euro area? international journal of central banking 11(3), 75-117. doi: https://doi.org/10.2139/ssrn.2451019 noss, j., & toffano, p. (2016). estimating the impact of changes in aggregate bank capital requirements on lending and growth during an upswing. journal of banking & finance, 62, 15-27. doi: https://doi.org/10.1016/j.jbankfin.2015.09.020 ongena, s., popov, a., & udell, g.f. (2013). “when the cat's away the mice will play”: does regulation at home affect bank risk-taking abroad? journal of financial economics 108(3), 727-750. doi: https://doi.org/10.1016/j.jfineco.2013.01.006 peydro, j-l. (2010). discussion of the effects of bank capital on lending: what do we know, and what does it mean? international journal of central banking 6(4), 55-69. doi: https://doi.org/10.17016/feds.2010.44 plantin, g. (2015). shadow banking and bank capital regulation. the review of financial studies 28(1), 146-175. doi: https://doi.org/10.1093/rfs/hhu055 m. huhtilainen / european journal of government and economics 8(2), december 2019, 203-217 217    repullo, r. (2004). capital requirements, market power, and risk-taking in banking. journal of financial intermediation 13(2), 156-182. doi: https://doi.org/10.1016/j.jfi.2003.08.005 tanda, a. (2015). the effects of bank regulation on the relationship between capital and risk. comparative economics studies 57(1), 31-54. doi: https://doi.org/10.1057/ces.2014.35 walker, g.a. (2011). basel iii market and regulatory compromise. journal of banking regulation 12(2), 95-99. doi: https://doi.org/10.1057/jbr.2011.4 microsoft word ejge_04_2015-010.docx european journal of government and economics volume 4, number 2 (december 2015) issn: 2254-7088 155 a quick indicator of effectiveness of “capacity building” initiatives of ngos and international organizations david lempert, independent development consultant, united states abstract the article offers an easy-to-use indicator for scholars and practitioners to measure whether ngos, international organizations, and government policies and projects meet the criteria for design and implementation of “capacity building” projects that have been established by various international organizations and that are recognized by experts in the field. the indicator can be used directly to address failures that are routinely reported in this key and growing development intervention. use of this indicator on more than a dozen standard interventions funded today by international development banks, un organizations, country donors, and non-governmental organizations (ngos) reveals that while many smaller organizations are working to change institutions and society in ways that effectively build long-term capacity, most of the major actors in the field of development have failed to follow their own guidelines. many appear to be using “capacity building” as a cover for lobbying foreign governments to promote international agendas (“purchasing foreign officials”) and/or to increase the power of particular officials at the expense of democracy, with the public lacking simple accountability tools. the indicator points to specific areas for holding development actors accountable in order to promote development goals of sustainability and good governance. the breadth of the field of “capacity building” also allows this indicator to be used, with some modifications, for a large variety of development interventions. this article also offers several examples of where current capacity building projects fail, along with a sample test of the indicator using uncdf as a case study. jel classification d02; p48; l31; o17. keywords capacity building; institution building; governance; civil society; development policy. lempert ● a quick indicator of “capacity building” initiatives 156 1. introduction “capacity building” has become the tool of choice for several major international “development” organizations in recent years, including the united nations system, the world bank and other development banks, major international government bilateral donors as well as non-governmental organizations. as of 2009, some $20 billion per year of international development intervention funding went for capacity development; roughly 20 percent of total funding in this category (otoo, agapitova and behrens, 2009). the world bank itself commits more than $1 billion per year to this service in loans or grants (more than 10 percent of its portfolio of nearly $10 billion) (world bank, 2005) and identifies it in all of its formulaic country studies as a ‘core objective,’ while the u.n. system practically defines itself by capacity development as the “how” for ‘“how” undp works’ to fulfill its mission32. between 1995 and 2004, the world bank committed $9 billion in loans and $900 million in grants to capacity building; roughly $1 billion per year. in 2007, the total bank lending was $9.1 billion (world bank, 2005:9). according to recent statistics from the organization of economic cooperation and development, 12 percent of the $15 billion that went to 38 ‘fragile states’ in 2007, of a total official development assistance of $101.3 billion, went for ‘governance and civil society,’ the key area of ‘capacity building.’ (oecd, 2008:12). using the oecd figure of $101.3 billion, the $20 billion reported by otoo et al (2009) would amount to 20 percent.) according to the undp’s current promotional campaign, ‘capacity is development’ and ‘the urgency of “how”’ (whatever that means) (undp global capacity development facility, 2014)33. despite the enormous reliance on this tool to implement missions of (sustainable) ‘human development’ or the rather different goals of ‘poverty reduction; economic growth; improved services’ (world bank, 2005) or ‘poverty eradication’ (undp, 2002), the very organizations that are most committed to this tool are also the first to admit that they do not follow their own guidelines (if any) and that results are often the opposite of what they claim they are trying to achieve. a world bank review noted that ‘examples abound’ in which these initiatives ‘severely undermine public management in recipient countries and unwittingly block rather than promote progress in public sector reform and institution-building’ (world bank, 2000). despite such harsh criticisms dating back almost two decades, little has changed since a 2005 study that showed that the world bank had almost no interest in applying the lessons and clear standards that already existed in the bank for running such projects; knowingly creating a double standard for its capacity building of foreign governments. ‘the bank does not apply the same rigorous business practices to its capacity building work that it applies in other areas. its tools—notably technical assistance and training—are not effectively used... moreover, most activities lack standard quality assurance processes at the design stage, and they are not routinely tracked, monitored, and evaluated’ (world bank, 2005). a undp study of its capacity building projects in 2002 reported almost exactly the same phenomenon and the undp has also done little or nothing to correct this problem since. among the implementing practitioners, ‘there were no generally agreed standards as to what should be expected of newly created or strengthened 32undp website (2014) url (consulted 17 june 2015): http://www.undp.org/capacity/. 33 undp (2014) going for scale, going for sustainability, going for quality. url (consulted 17 june 2015): http://www.undp.org/capacity european journal of government and economics 4(2) 157 national capacity or of institution or organisation building (undp, 200234).’ despite the fact that the u.n. system and other reports describe exactly how results based measurements could be done in this field (undp 199835, 200436, 200637), with guidelines easily accessible on the internet, u.n. officials continue to offer the excuse that results ‘cannot be easily measured’ to justify why they do not follow measurement procedures and apply a double standard. as a practitioner in the field for more than 30 years, it is this author’s view that the situation has actually gotten worse rather than better. while others may disagree, there is little that they can point to in terms of oversight or measurements that create accountability in this field. the failures may be by design. according to one early observer looking at the state of the field, ‘capacity building’ is simply being used as a ‘buzz word’ by international agencies for whatever they wish to do, with or without any accountability or logic (enemark, 2003). the current state of the field among the major donors is such that it appears to have already reached a theatre of the absurd. projects (some described below) may be the equivalent of seeking to turn current government clerks into brain surgeons (or the equivalent of teaching elephants to fly rather than buying birds) in a purported attempt to meet the need for specialized professionals without having to add or fire staff or to establish performance standards. other projects take the approach of seeking to convince wolves to turn vegetarian through “consciousness-raising” in “rights” or in “learning” how not to be corrupt, rather than facing the real underlying cause of the problem, such as the needs to build walls around the henhouse or find ways to put more power on the side of the hens. many projects have no measures of intended results and lack fit with a development strategy or logic. educators and consultants who work in the field and are paid handsomely to “build capacity” also have little incentive to hold themselves to systematic standards since it might make their contributions obsolete. by definition, the more services they are hired to provide, the more “capacity” they build and the more they profit, whether or not their contribution is really the most efficient, effective or sustainable, or addresses the institutional root causes of the lack of capacity in the first place. generally, these projects are evaluated with ‘smile sheets’, asking beneficiaries if they are ‘happy’ or ‘better off’ and measuring things like ‘raised awareness’, ‘enhanced skills’, and ‘improved teamwork’ that are ‘locally driven’, rather than on whether the underlying problems are solved, and refraining from asking whether there may be hidden agendas to buy influence, subsidize elites, and continue dependency (otoo et al, 2009). the major donors, themselves, admit that they have little incentive to end the double standard and achieve consistency in this area, despite the fact that they have long ago elaborated clear and basic frameworks and standards for how to appropriately measure key components of capacity building (including strategic planning for efficiency, service delivery, or application of skills). the world bank’s 2005 evaluation noted that practitioners shaped and reshaped definitions and 34 undp (2002) capacity building for poverty eradication: analysis of and lessons from evaluations of un system support to countries’ efforts. url (consulted 17 june 2015): http://www.un.org/esa/coordination/capacity_building_for_poverty_eradication.pdf. 35 undp (1998) capacity assessment and development. in a systems and strategic management context. technical advisory paper no.3, bureau for development policy (bahman kia and richard flaman). url (consulted 17 june 2015): http://portals.wi.wur.nl/files/docs/ppme/capsystech3-98e.pdf 36 undp (2004) national human development reports (nhdr) and the use of governance indicators. url (consulted 17 june 2015): http://www.undp.org/oslocentre/docs06/nhdr.pdf. 37 undp (2006) a review of selected capacity development methodologies, bureau of development policy. url (consulted 17 june 2015): http://74.125.153.132/search?q=cache:ll3zlvoiv0gj:www.lacworkspace.undp.org.co/fileadmin/desarrollo_capacidades/documentos/resource_cataloguereview_of_selected_capacity_assessment_methodologies.doc+undp+(1998):+capacity+assessmen t+and+development.+technical+advisory+paper+no.3&cd=3&hl=en&ct=clnk. lempert ● a quick indicator of “capacity building” initiatives 158 standards to suit themselves with some development agencies using ‘a narrow definition focused on strengthening organizations and skills’, with the world bank, itself offering ‘no operational policy to guide its capacity building works’ (world bank, 2005). a undp study in 2002 similarly reported that there was ‘no systemwide framework’ and that participants’ use of the terms ‘capacity,’ ‘capacity building’ and ‘performance’ showed ‘wide disparities’ (undp, 2002). moreover, almost none of the units within undp sought to collect any baseline data and their documents from the two previous decades where such data might exist, routinely “disappeared” into “storage.” the abuse of the “tool” of “capacity building” or the substitution of this tool and its inputs for any measurable development outputs is not unique to this international development intervention (schachter, 2000). in the absence of clear public pressures and accountability measures, many development actors of good or dubious intent justify interventions on the basis of the importance of the symptoms they are treating rather than on actual measurable impact that they say is too difficult or costly to determine; throwing money to build capacity and “strengthen” whomever receives it, rather than addressing problem causes. they often use their location overseas and their direct relations with foreign officials as ways to further hide their activities from public scrutiny in either the donor or recipient countries. in the area of “capacity building,” where the “poor” and members of the “public” are easy to exclude, the transfers of money and resources are often directly from government officials in one (powerful) country to those in another (weaker) country in what can be labelled building the “capacity” of the weaker bureaucrats through a legitimized form of corruption and abuse that circumvents international law. the (perhaps intended) result is that development projects often serve the interests of those government bureaucrats channelling the funds and those who receive them rather than the public that is supposed to benefit from measurable results (lempert, 2008). although some may view statements like this as rather bold, officials in developed countries are often the first to admit that they use ‘aid’ as a form of ‘soft power’ (as opposed to military power and economic pressure) in order to manipulate the political systems of weaker countries (nye, 2004). moreover, this attempt to influence decisions by government officials in weaker countries through use of financial benefits directed to those officials, in ways that favour the interests of the businesses and peoples of the donor country over other countries and/or over the peoples of the recipient countries, easily meets the definitions of corruption that are recognized under international law and that are the very practices that governments all claim they are trying to eliminate. it would be relatively easy for lawyers to find government officials in both the donor and recipient countries in violation of various provisions of the united nations declaration against corruption38 under articles 15 and 16 (bribery), 18 (trading influence), and 19 (abuse of functions). transparency international, the international ngo that is the recognized expert on corruption and that is largely funded by major donor countries, would also classify these behaviours as corruption under their definition of ‘political corruption’ in which funds are used to influence politicians to ‘steer away from good government’ and to make decisions on the basis of factors other than the ‘public interest’ in ways that ‘divert resources’.39 in answer to these challenges in the development field, recent articles by this author have chosen not merely to expose the problems and call for “change” but 38 u.n. declaration against corruption and bribery in international commercial transactions (1996): a/res/51/191. url (consulted 17 june 2015): http://www.un.org/documents/ga/res/51/a51r191.htm. 39transparency international (2015). url (consulted 17 june 2015): http://www.transparency.org/whatwedo. european journal of government and economics 4(2) 159 have taken some of the initial steps to establish indicators and benchmarks through which the public and organizations can easily arm themselves to hold international development actors accountable to international law, to their mission statements for their interventions and to professional standards. these indicators are offered as easy-to-use tools (essentially, new public “weapons”) to create accountability and transparency in the use of public funds in development interventions; directly exposing abuses and offering specific directives for improvements. previous indicators began with tests of whether development projects met international treaty standards for promoting the agreed development objectives of ‘sustainable development’ (lempert and nguyen, 2008) as well as other basic governance objectives to promote self reliance and end colonial ‘dependency’ (lempert, 2009), to build “democracy” and protect rights as key international legal goals of good governance (lempert, 2010a; 2011) and to hold development professionals to ethics standards in order to eliminate conflicts of interest (lempert, 1997). this is also part of a larger initiative to build organizations that will monitor and challenge donors (lempert, 2008). the article offers an easy-to-use indicator for scholars and practitioners to measure whether ngos, international organizations, and government policies and projects meet the criteria for design and implementation of “capacity building” in ways that have been established by various international organizations, themselves, and that are recognized by experts in the field. this article also serves as a model for accountability in the use of other developments inputs by offering an approach can be adapted to specific interventions. indeed, “capacity building” itself is an umbrella for many specific tools – education, awareness, providing equipment, strategic planning, consulting advice, etc. – that can be measured using similar approaches. the article begins by defining “capacity building” according to basic internationally agreed principles that can be placed into an indicator, then surveys existing indicators, explains why several international “capacity building” projects now fail in the absence of an indicator or standard to hold them accountable to minimal levels of competence, then offers a new indicator and tests it on several categories of projects, including a detailed examination of how to use the indicator on an organization like the united nations capital development fund40 that claims to be doing capacity building as it central tool. 2. principles of “capacity building” the principles of “capacity building” as a standard development intervention are generally agreed upon not only by major donors but also by the community of nongovernmental organizations and by business consulting firms that do capacity building in the government, private business, and ngo sectors (undp, 199141; 1998; world bank, 2002; 2004; 2005). this makes it relatively simple to test whether organizations are actually doing what they say they have committed to do. though organizations use different wording and order, and much of the wording is confusing given jargon in the field, there are really five key concepts that are the essential basis of capacity building: three of them relating to the ability of a system to operate and perform a public function at three different levels (its legal and political authority or “enabling environment” within a larger social and political context; its managerial ability internally to perform effectively and efficiently that is often referred to as its “institutional development”; and the skills of its staff – its 40 united nations capital development fund (uncdf) (2009) website “about uncdf”. url (consulted 17 june 2015): http://www.uncdf.org/english/about_uncdf/index.php. 41 undp (1991) a strategy for water sector capacity building' in delft, the netherlands. url (consulted 17 june 2015): http://www.gdrc.org/uem/capacity-define.html lempert ● a quick indicator of “capacity building” initiatives 160 “human resources”), one relating to good governance (its public accountability), and one relating to the long term sustainability of the function. given that these are generally recognized, they do not need to be detailed here in full, but it is easy to understand the logic of the five elements. to be effective, an organization must have sufficient freedom from constraints to conduct its activities and must be able to use the resources that it has. it also must have people who are skilled, properly selected, compensated and tasked. for public organizations, there must be public oversight of the legal framework, of the organization itself, and of the skills areas (professional and technical fields and the training) and workers applying them to assure they are in line with public needs. this whole system must be sustainable and integrated, not simply driven by donors deciding to toss in resources or training to a specific institution, but working effectively over the longterm with people seeking training according to public needs, being appropriately selected and trained, then selected on merit systems, appropriately compensated and tasked. assuming a quick fix of channelling funds or training courses to workers in an institution who may be the wrong people being wrongly tasked or overseen in ineffective organizations, or to educational institutions that may be training people who will never be hired or that lack the capacity and management for appropriate training may be irrelevant. the cause of failure may be at any one of these levels or several and that is why any solutions must address the whole system. more detail for readers new to this subfield is presented in the appendix to this article (section i). 2.1. indicators in the field and the lack of an indicator for “capacity development” interventions though there have been attempts to develop governance indicators for governance projects and diagnostics to use in planning capacity development interventions in different kinds of organizations, there are no existing indicators to score “capacity building” interventions on whether or not they are meeting the basic professional requirements of capacity building. nor is there any licensing or grading system to measure basic competence or to establish other competence levels for practitioners who do “capacity building” and who claim to be “capacity building experts.” among the most recent attempts, for example, the world bank institute has developed a manual but it is filled with dozens of questions to use and impossible to apply as a quick diagnostic (otoo et al, 2009). nothing else exists in the field. though there are several international measures of “good governance” and “democracy” these do not cover the area of capacity building. other measures, with community based organizations are closer to strategic management tools than capacity building measures (mickinsey, 2001; christensen et al., 2006; gubbels and koss, 2000; lusthaus et al, 2002). a review of these and others is presented briefly in the appendix in section ii. 2.2. the problem with many “capacity building” projects and the real value of an indicator in the absence of any professional standardization for use of the tool of capacity building or any accountability indicator for this approach, there is, in effect, no public review of the billions of dollars of public funds that are being transferred from developed countries to the governments and non-governmental institutions of the rest of the world in the guise of “capacity building.” abuses in this area are in fact running rampant. at best, “capacity development” projects are just throwing money at symptoms with no logic or analysis. at worst they are disguised bribes to european journal of government and economics 4(2) 161 government officials and attempts to undermine entire government structures by setting up foreign run ministries and foreign influenced political parties or civil society to lobby for foreign interests. these problems should be familiar to everyone in the field. in the appendix, in section iii, some of the abuses are presented directly for readers, in stark terms. although many of the descriptions are frank and direct, coming out of professional observations by the author in some 30 years in this field, the donors themselves admit the problems as well as the underlying goals of promoting their national self interest through aid as “soft power”. they simply use more euphemistic language to soften the implications and potential exposure (nye, 2004; raffer and singer, 1996; mayo, 2009). the author has detailed some of these cases in greater detail in other works as well, including those referenced here. whatever the reasons for the failures, it is possible to use a simple tool to measure and expose the failures. 2.3. the indicator of “capacity building” that can measure adherence to recognized professional standards of the field as a first step towards the licensing of practitioners who claim expertise in “capacity building” and as a way for citizens to hold donors accountable in the spending of their funds or in the acceptance of funds for “capacity building” while exposing approaches with hidden agendas, the indicator below is presented with 20 simple questions as a litmus test of basic competence in the field. by asking these 20 easy “yes or no” questions and then counting up the results (possible 11 points), one can determine the relative competence and integrity of a “capacity building” project or intervention on the following scale: scale: 8 11 points comprehensive approach to “capacity building” in ways that also appear to be promoting sustainable development, self reliance, and democracy in line with the rio declaration and international conventions 4 7 points minimally competent approach to “capacity building” with several failures in procedure or safeguards 0 – 3 points narrow or weak intervention < 0 incompetent project with hidden agenda that has been corrupted either by the donor agency, stakeholders in a developing country bureaucracy, or both, in a possible attempt to purchase or influence foreign policies or to collude to misuse funds note that the indicator is not an absolute scale since it is not offered as a social science research tool (though it can be used as such) but as a project evaluation and selection tool. it is best used to show the relative value of different projects. the indicator does not measure the quality of specific capacity building, since that depends on many factors and comparisons, with benchmarks and cost-benefit analysis. it also does not measure how effectively a capacity building intervention is at promoting sustainable development, self-reliance, or democracy, since these are covered by other indicators, though a project that gets the highest score must at least be paying attention to these issues. the purpose of the indicator is not to measure “gross benefit” or “cost benefit.” it is simply to measure compliance with professional standards for this type of intervention. indeed, a capacity building initiative may be technically competent at building long term capacity but could still send a country down the wrong development path. that is why it is important to use this indicator in combination with others and why this indicator qualifies only as a litmus test for basic competence in use of this tool. (what the indicator does is lempert ● a quick indicator of “capacity building” initiatives 162 determine whether the project is actually doing capacity building according to the technical standards of the tool or is hiding another objective. it also determines whether other agendas and ideologies have taken control of the funds and whether the overall goal is really to promote development and public control or not. it is essentially a litmus test of basic competence in the field and application of appropriate safeguards to protect professionalism and the public interest.) like most indicators, answers to each question would need to be “calibrated” to assure that different observers make the exact same determinations. to do so would require a longer manual for standardized, precise answers across observers. 2.4. measures/sub-factors below, is an explanation of how anyone can apply the test to any project by asking the 20 questions and recording the scores. most of the questions are clear cut “yes” (1 point for applying a standard procedure, 0 loss of points for protecting against conflicts of interest or other corruption of standards) or “no” (0 points for failing to apply a standard procedure, -1 indicating a loss of points for facilitating conflicts of interest and corruption of standards), but in cases where there is a judgment call, you can opt for a “debatable” (0.5 points for benefits and – 0.5 points for harm). the measures of performance can be placed into two categories that look at application of basic professional requirements for project design (positive scoring) and protection of the project in implementation from conflicts of interest or negative development impacts (negative scoring): 1) proper application of the basic principles and standards of capacity building (5 of the 5 recognized principles of capacity building, including diagnostic of the three levels of analysis for capacity) and of development interventions, with the donor organization also serving as a model of accountability with a total of 11 questions broken into three sub-categories.  the first sub-category (4 of the 5 recognized capacity building principles) looks at basic professionalism in diagnostic and design (7 questions for a possible 7 points).  the second sub-category (the fifth of 5 capacity building principles), contingent on meeting the standards of the first category, addresses sustainability (2 questions for two possible points).  the third sub-category addresses whether the donor or project implementing organization itself reflects and models the standards of good governance (2 questions for a possible 2 points or loss of 2 points). (the overall potential scores for this section of 11 questions is 11 points.); 2) professional safeguards are in place against conflicts of interest and against unintended consequences that could distort other public or private systems: with a total of 9 questions broken into two sub-categories.  the first sub-category seeks to protect against conflicts of interest and against organizations searching for problems to fit the tool of “capacity building” or stakeholders to agree to projects (4 questions and a total loss of 4 points).  the second sub-category seeks to protect against negative or adverse impacts on the overall political/government system and on related business and civil society systems (5 questions and a potential loss of 9 points). (the overall potential score for this section of 9 questions is to maintain the points awarded in the first section.) european journal of government and economics 4(2) 163 the first category (and particularly the first sub-category of 7 questions) is itself a screening to test whether a project actually achieves anything in the area of capacity building at all, and whether it meets minimal basic competence in the field. questions marked with an asterisk (*) are those that apply to any development intervention. they represent the basic steps of project design (root cause analysis; benchmarks and cost benefit of interventions that are tied to specific measurable results and indicators; sustainability of the input), of screening to assure appropriate development impact, and tests of the donor organization to assure best practices are followed. note that some of the questions below may seem lengthy, and this “simple” list may not look that easy to use on first glance. the questions are lengthy so that the goal of each question is clear and readers can train themselves to effectively score the question. with some short practice (working through the example in the appendix, in section v, and also considering the scoring for several projects, presented in the appendix, in section iv), readers will be able to see how the indicator works to distinguish interventions and to offer a valid way to score and improve them. 1. proper application of the basic principles and standards of capacity building (5 of the 5 recognized principles of capacity building, including diagnostic of the three levels of analysis of capacity) and of development interventions with the donor organization also serving as a model of accountability: this is the category that can be used for screening whether the project and spending really have any substance and fit the basic professional competence of capacity building and of development interventions. (11 questions and a potential score of 11 points) a project that does not score more than 4 points in this category is already partly suspect as being driven by an outside agenda to favour a specific group rather than to promote real democratization and good governance. at least 5 of these questions (marked with an asterisk) are directly applicable to any development intervention. the other 6 are specifically applicable to “capacity building” though they could also be applicable to other development interventions with slight modification. 1.a) project meets 4 of 5 of the recognized principles of capacity building, including diagnostic of the three levels of analysis of capacity, and principles of professionalism for development interventions: this is the heart of the indicator. a project that does not score at least 4 points here is probably not competent in the field. (7 questions and a potential score of 7 points) question 1*: country and cultural fit of the development intervention for sustainable development: fixing what is broken. the project is not formulaic or sector specific (“to train judges”; “to build a modern x system”) but is fit specifically into the local cultures and their needs for maintaining or returning to sustainability and fixing a system that has broken, with the analysis beginning with an assessment of local needs. the question is whether the capacity building is relevant to restore the sustainability of the local culture(s) and repairs an existing and identifiable underlying problem that has moved the culture away from sustainability and needs to be “fixed” in line with international principles of cultural survival and sustainable development. the overall approach carefully incorporates the requirements of sustainable development established by the rio declaration (balance of population and consumption with productivity and resources) in its approach to governance as the basic principle of the role of governance in sustainable development (united nations conference on environment and development42; united nations 42 u.n. declaration on the rights of indigenous peoples (2008) url (consulted 17 june 2015): http://daccessdds.un.org/doc/undoc/gen/n06/512/07/pdf/n0651207.pdf?openelement. lempert ● a quick indicator of “capacity building” initiatives 164 declaration on the rights of indigenous peoples43). if the approach is one that simply targets a “sector” in a “sector wide approach,” and/or that is an outside determination based on a formula or a comparison with what “developed” countries or cultures have, then the score here is zero because it has not started with the cultural fit as a whole and may simply be developing one system to the detriment of the overall balance for sustainability and appropriate governance. (for expanded applications, see lempert and nguyen, 2008.) scoring: yes 1 debatable 0.5 no 0 question 2*: governance or civil society functions promoted are appropriate parts of the key governmental or non-governmental organizational missions through which there is direct public accountability. the project is clearly designed to promote either a specific government function of protecting and/or promoting (separately and as a competing function in a structured system) an asset or resource of specific ethnic groups and of the country in ways that maintain or increase the per capita assets of the culture and the country, or to promote a non-governmental organization function that is specifically separate from government and which is directly accountable to all of its owner, consumer, neighbour stakeholders. there is a clear and measurable mission of accountable governance with measurable costs and benefits to specific units of public spending or resources for which specific government or non-government officials can be held directly accountable by citizens (removal, punishment, or rewards). interventions that “promote regional integration” or “monitor foreign development funds” or “strengthen civil society” do not earn points. government missions should be to “promote, measure, and maintain” assets. those that “promote justice” or “improve health” or “protect resources” but do not assure that there are measurable benefits to specific cultures and/or per capita long-term improvement that are appropriate to their sustainability and adaptability, or that “develop” a resource without also protecting it, may actually distort effective governance systems and earn no points. scoring: yes 1 debatable 0.5 no 0 question 3: full framework analysis includes three of the multiple dimensions or levels of capacity to determine all of the failures and their relationships to each other. regardless of any initial area that the project might pre-select as an intervention, the project withholds judgment on the nature of solutions or problem and conducts a full diagnosis to get at the root causes of the failures. the project design includes a thorough assessment of the multiple dimensions of “capacity” that includes all of the following as a key to understanding the human dimensions and behaviours that are the source of failures of systems to meet needs for sustainable development – 1) the societal and cultural level, including ideologies, mal-adaptations to environmental changes, distributions of power and organization of political and 43 united nations conference on environment and development (1992) rio declaration on environment and development. url (consulted 17 june 2015): http://www.un.org/documents/ga/conf151/aconf15126-1annex1.htm. european journal of government and economics 4(2) 165 legal institutions (including legacies of colonialism or current outside pressures) that interfere with sustainability; 2) failures at the organizational and institutional level for fulfilling specific missions; and 3) human resources failures. the analysis distinguishes symptoms from causes and also models how the entire system would work if it were returned or moved towards sustainability. there is commentary on the status of each of the four areas indicating whether or not there are failures at each level and determining with evidence whether or not specific areas can be approached independently to reach solutions or whether only comprehensive and linked solutions can achieve solutions to the problem and move towards a sustainable model. scoring: yes – 1 (all four of the levels are fully considered) debatable 0.5 (all four levels are considered but measures may be sloppy.) no 0 question 4*: root cause analysis and problem trees. the project focuses on those areas where there are failures and where specific and comprehensive changes can lead to measurable performance improvements. the project design includes a thorough assessment of the root causes for the failures that can be corrected through “building capacity.” problems are mapped in a problem tree and the root causes of the problem, with identifiable target actors whose behaviours need to be changed, are presented in a systematic and linked way. the potential intervention is tied directly to the problems at every one of the identified steps in the sequence where there are failures, with specific measurable outputs per unit of input at each stage, with particular attention to human behaviours at various levels to be changed that underlie the problem (rather than symptoms of weaknesses) that lead to measurable performance outcomes/results. symptoms (low skills; lack of transparency or incentives, weak management, low resources) are not root causes, only individual and social and organizational behaviours are, and these are fully recognized in a problem statement and root cause analysis. the logframe shows specific measurable behaviour changes in cost savings, cost effectiveness, and other service delivery and social indicators, as direct output evidence of performance improvement outcomes, rather than simply throw inputs at a symptom and assume that there will be “better governance” or “improvement” or “greater capacity” or “more efficiency” because resources have been transferred. scoring: yes 1 debatable 0.5 no – 0 (no problem statement or root cause or problem tree) question 5*: logframe specifically targets the root causes in an appropriate sequence, showing how specific measured inputs lead to specific measured performance changes over a long period of time, with cost effectiveness ratios of inputs to outputs included based on benchmarks of outcomes. the project logframe demonstrates specific measurable behaviour changes as outputs that lead to performance improvements, rather than simply throws inputs at symptoms and assumes that there will be “better governance” or “improvement” or “greater capacity” because resources have been transferred, documents have been drafted (laws, action plans or policies), agreements have been signed, people have been “trained,” or new offices or institutions have been established or tasked. inputs are not turned into outputs because they are delivered or received or because something is to be produced. to earn points there must be clear benchmarks of lempert ● a quick indicator of “capacity building” initiatives 166 changed service delivery that are not simply accounting measures of delivery of inputs or intermediate inputs and there should be benchmarks linking units of spending to specific behavioural changes of specific numbers of a target population, with those specific behaviours leading to performance results that move a culture back to a path of sustainable development (living within its resource base). scoring: yes 1 debatable 0.5 no – 0 (no cost benefit analysis of outputs; no logframe or logframe where inputs and outputs are confused and do not meet professional standards) question 6: public accountability mechanisms are part of the assessment (fourth principle of capacity building, in analysis) and the project begins with concern of placing capacity in public hands, starting with youth as the first and priority alternative of any capacity building intervention. the project design includes a thorough assessment of the accountability of the systems to the public as a key to the social context. there is a focus on the laws, the incentives of officials, the feedback for funding, the monitoring of funds, and the public tasking and monitoring of results of the institutions for which capacity is to be “built.” there is a focus on increasing the capacity of the public, first, since all government and non-governmental functions and cultural decisions and understandings ultimately depend on the skills of the public. building capacity of the public is considered first before any decision is made to work directly with the public’s agents or representatives, in order to ensure that specialists will be fully accountable to and controlled by the public. projects to build civil society or governance start with assessments of the general skills, powers, and capacities in the population for such oversight and works to built them at the level of basic socializing institutions with young people, first, rather than with adults. the project clearly explains any decisions not to put all such skills and capacities into the public starting with youth and in models with youth (model civic actions and democratic governance of youth, model youth courts and public investigations, youth newspapers and ngos, youth businesses, etc.) scoring: yes 1 debatable 0.5 no 0 question 7: agreement is transparent to citizens and they are included in the project oversight and approval (fourth principle of capacity building, in implementation). the project transfer of inputs to government or ngo stakeholders is conducted in ways that include screening and monitoring directly by citizens of the host population, including not only the direct approval of their legislative bodies but openness to media and outreach and inclusion of the public, so as to avoid any agreements that are government to government or bureaucrat to bureaucrat transfers. scoring: yes – 1, if the project shows awareness of this and protects against harm debatable or not relevant 0 no – (-1) (loss of a point) european journal of government and economics 4(2) 167 1.b) sustainability of the impact through systematization and institutionalization: the fifth principle of capacity building and one common to other development interventions: the project doesn’t just seek short-term impact but institutionalizes a process of continued responsiveness and adjustments of capacity, as well as identification of needs and building of new capacity, in the governmental system and/or culture. (2 questions and a potential score of 2 points, to be awarded only if the project has achieved a score of at least 4 points on the first 7 questions. otherwise the measures are rewarding sustainability of a project that may actually undermine appropriate capacity building.) question 8: intervention institutionalizes change at the root of the problem, fixing a broken system, and is not funding an institution or activity for a problem that starts elsewhere in the society or culture (e.g. offering remedial training of adults, professionals or employees). the project repairs a system failure (e.g., basic parental or public education for adult responsibilities; university education; professional education; a personnel selection system; salary structure; information system) and assures that is adaptive to change, for multi-generational continuing impact rather than offering a short term transfer or quick fix to one target stakeholder or recipient group. an education problem in skills, an awareness or consciousness issue, an issue of funding or overall behaviours, is traced back to family socialization and the formal and informal (media) education system with a focus on changing those for the long term rather than simply treating the symptom by working with adults. scoring: yes 1 debatable 0.5 no 0 question 9*: sustainability of the intervention and impact. the intervention builds a continuing system of measurement, monitoring, determining of actual need and value, and appropriate advocacy and receipt of future funding. the intervention has built institutional mechanisms that are self-sustainable within the country’s resources and have continued local financing and management, with freedom from continued foreign or institutional funding that would create dependency on outsiders for achievement of the project goals. the project builds and institutionalizes an incountry monitoring system of capacity and of costs and benefits in the services that are being supported, such that the system adapts to changing conditions and measures (with adequate oversight to assure no conflicts of interest in the data) that there is appropriate advocacy for funds and receipt of support in an effective balance with other competing systems through a fair process. there is a specific determination of “how much” capacity is needed to solve a particular problem and at a measurable cost, with costs and benefits considered to the country and how the country will be able to pay for it out of the stream of future benefits. it is not just considered to be good for its own sake, without justification. scoring: yes 1 debatable 0.5 no 0 1.c) internal procedures of the project, itself, reflect the values of accountability and self-reliance: the project organization is itself a model of good governance, accountability, efficiency, transparency and appropriate capacity (2 questions and a potential score of 2 points but also a potential loss of 2 points if the project itself sends a message that contradicts and undermines what it claims to be achieving) lempert ● a quick indicator of “capacity building” initiatives 168 question 10: accountability of the donors: the project itself (the donor organization) is a model of transparency and direct accountability to beneficiaries and to citizens, and does not hide behind barriers that require citizens or beneficiaries to demand that government representatives or other elites bestow accountability. projects that meet this requirement will have open books, clear professional ethics codes, full published reports on their projects, and full use of measurement tools of benefits in the profession (cost-benefit, baselines, comparative indicators, and industry benchmarks). moreover, the donor organization, itself, has conducted the very same capacity analysis subject to the same questions as in this test, to assure that it has a clear and strategic mission, that its own results are measurable with benchmarks and cost effectiveness data, that its employees meet specific tests of certified competence in their fields, that evaluations are done independently and objectively not with management oversight but with the oversight of funders and beneficiary stakeholders, with results available and with feedback systems that assure accountability and implementation of the results. scoring: yes – 1, if the project shows awareness of this and protects against harm debatable or not relevant 0 no – (-1) (loss of a point) question 11: rewards and incentives: the project rewards behaviours that promote independence and sustainability of recipients and punishes behaviours that promote dependency or donor relationships, with no opportunity for collusion to prolong funding in the absence of meeting stringent conditions for results, established in advance. delayed projects are not rewarded because of good “relationships” have been built and officials or beneficiaries “appreciate” the assistance, but because there are clear standards showing progress towards measurable results and sustainability and that exceed standard benchmarks for projects resolving similar root causes of the problems. grants are given with real conditions that have enforceability and consequences without paternalistic justifications that recipients cannot or should not be held to real standards. success of a project does not lead to replication in additional areas using more outside funds but towards promotion of copying by others with their own resources. failure of a project to be sustainable or to show strong benefits does not invite additional funding because of “continued poverty” or “need” but triggers an immediate change and possible liability. the original setting of conditions on the project meets the most stringent of international treaty standards, benchmarks and objectives and was not simply a politically negotiated transfer of support to facilitate project inception. scoring: yes – 1, if the project shows awareness of this and protects against harm debatable or not relevant 0 no – (-1) (loss of a point) 2. professional safeguards are in place against conflicts of interest and against unintended consequences that could distort other public or private systems: these questions are challenge tests to assure that the design process is actually being used in ways that it is supposed to work. for this reason, scoring here is negative, subtracting points for failures. questions are in two categories: 4 questions on conflicts of interest and 5 questions on negative impacts on public and private systems. (9 questions and a potential loss of 9 points) european journal of government and economics 4(2) 169 2.a) the “capacity building” tool is one tool to fit an objectively measured sustainable development need, rather than a tool searching for a need where it can be used or as a political facilitator of some other transfer: no conflicts of interest: the project is not subject to the influence of any stakeholder groups in its design or implementation other than determinations of independent professionals held directly accountable to ethics standards (4 questions and a potential loss of 4 points) question 12: assessment of the need is protected against subjective biases of the partners/ stakeholder recipients. the project clearly distinguishes the difference between sustainable development “needs” and partner/stakeholder “wants.” needs for capacity building are assessed as relevant to fixing a root cause of a development problem with failures assessed before the project begins and before any funds are committed, using independent, objective measures such as skills testing, production and service analyses, monitoring of performance and outputs, strategic management tools (for potential consulting inputs), etc. the project is not a santa claus, measuring its success with “smile sheets” of how happy partners or stakeholder recipients are to receive support and asking them to measure their “needs” with questionnaires or group meetings (what they would “like to learn” or where they could “use consultants”) in ways that offer funds first and then seek to justify their transfer with the subjective or political determinations. scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 13: delivery of inputs does not personally benefit the stakeholders/partners but benefits everyone in the society/culture equally. the project includes no incentive payments in the form of perquisites to convince or ease participation of officials (per diems or other payments that are above salary; travel and tours; scholarships and education resulting in degrees or certificates that can be used in other jobs; vehicles, computers, or other office equipment). anything delivered is measured in a specific test of benefits to the country versus benefits to the implementing stakeholders. such payments and transfers are also those that members of the public are informed about and approve. scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 14: there are no linked inputs or grant funds for officials to manage that do not directly improve capacity for already existing budgets and target root causes of capacity failures rather than symptoms. the project includes no pump priming grants to government agencies that increases their existing budgets without starting directly on the root causes of administrative system failures. anything delivered is measured in a specific test of solving the problem of current lack of lempert ● a quick indicator of “capacity building” initiatives 170 capacity for managing existing budgets or for making the case for increased taxation rather than as a sweetener that could hide another objective. questions are asked immediately about why the agency currently does not tax effectively for funds if they do lead to public benefit, or why they have not closed loopholes of corruption or established appropriate development priorities, before anything is given to a partner to manage. scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 15: screening and delineation of all conflicts of interest of international “partner” organizations for “twinning” projects, for consultants and international firms, and of all implementing and donor agents/ stakeholders is combined with legal and ethical oversight. the project documents all possible biases and benefits to the participating partners, donor agents, and donor staff to assure that there are no future financial or personal benefit conflicts of interest beyond pure humanitarian goals driving projects or methods for specific benefits to stakeholders that are in any conflict with the funders and/or the public beneficiaries that drive project decisions in any way. strict mechanisms are in place for outside challenges, for review, and for stopping projects or penalizing offenders for any such conflicts. scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) 2.b) no negative or adverse impacts on the political/government system or on related private systems: the project does not distort the overall governmental system or the roles or balance of governmental or non-governmental (business and community based) institutions but supports their appropriate roles. (5 questions and a potential loss of 5 points) question 16: citizen powers and enforceable protections are strengthened as the initial and linked goal and result of all interventions, with real safeguards against transferring new powers to officials or to large private organizations (business or civil society, including foreign government, business and ngo actors) in the name of citizen protection, or any other changes making decisions more opaque or hierarchical. the project does not increase the power of officials or designated representatives over citizens (e.g., supporting strengthening of an “institution” that empowers its officials such as parliament or an “ombudsman” or “prosecutor” rather than increasing the role of the public using the government function, such as empowering the public with legislative power and skills, increasing public ability to sue and remove government officials and to be the judicial deciders, or allowing for class action suits and other “private prosecutor” mechanisms.) the intervention does not create new public bureaucracies without a real change in citizen power. scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm european journal of government and economics 4(2) 171 no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 17: interventions do not improve efficiency of systems that may be undemocratic, inequitable or unjust but begin with a focus on accountability, participation, and equity. the project recognizes the danger of “efficiency” of systems that are inherently unrepresentative and does not simply promote efficiency given the potential for continued rights abuses by an authoritarian or unaccountable regime. projects working with businesses, community based organizations or any aspects of government where decisions are made (particularly the justice system but also parliament and any ministries with discretionary and/or policy authority) work to link efficiency interventions directly with equity and direct citizen control (removal, sanctions, legal challenges, equal participation as jurors). scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 18: donor-to-stakeholder beneficiary financial transfers (grant, loan, gift) or commitments that are given as part of “capacity building” do not create future obligations in ways that could subsidize a favoured social group, distort taxes, or negatively influence priorities for sustainability. the project meets strict public finance criteria for grant, loan/investment, or subsidy and funds are allocated appropriately, with public oversight to assure that any loans are approved by the public and that the public is able and willing to pay for each item that is linked to “capacity building.” the impact of the assistance does not end up subsidizing some other improper spending (allowing a transfer from one category of spending into something else as a result of the gift), reducing pressure on elites who should tax themselves to fund the project, or end up distorting capital markets in the country by offering a gift or subsidy for a kind of productive investment that should be in the form of a loan or in the form of a loan at competitive market rates. if the project merely seeks to find “poor” people to help by building the “capacity” to help them, but does not analyze the responsibility of the elites in the country to fulfill obligations of social solidarity with people in their own country, the project is part of a collusion in detaching elites globally from their local responsibilities and has a negative impact. if the project does not do an analysis of existing government spending and tax policies, it is likely that the project is offering money for the “poor” at the same time that an excessive amount of the budget is being used for military and police spending to control the poor, and the project is actually subsidizing this pattern. if the project includes a gift rather than a loan and does not include standard financial controls and conditions, it is likely being given with the knowledge that corruption will occur and as a subsidy for corruption and waste, including purchase of luxuries and foreign goods. scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) lempert ● a quick indicator of “capacity building” initiatives 172 no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 19: government functions are respected and the balance with civil society is promoted appropriately and internally, with effective public regulation of the nongovernmental sectors. the intervention does not replace a government function with foreign paternalism building parallel systems that are better than government but that do not improve what is wrong, or transfer a government function to another place like civil society because of current underperformance. nor does the project seek to build a civil society that must rely on funds from donors or from elites, in place of direct support and control by the beneficiaries, themselves. nor does it seek to build a strong private sector without effective public regulation of that sector and restrictions on the powers it could exert through financial pressures. (ngos are not public service providers but provide for private needs and have a role in trying to improve government action; businesses are not “corporate citizens” but are producers to be taxed and regulated to fund public functions, etc. the appropriate role of an ngo is to model a new behaviour to try to convince government to change, and to advocate for the special interests of a group with sustainable funding and accountability to that beneficiary group, but not reliant on foreign or other outside funding. the appropriate role of funding for ngos must also be to ensure that they are sustainable with funding from the beneficiaries and with accountability directly to the beneficiaries, and not dependent on funds from foreigners or from elites who are disconnected to benefits to the beneficiaries and who have different interests.) scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) question 20: the intervention promotes competition among providers in the area where capacity is built and does not distort market systems. the project seeks to build the infrastructure in society for government consulting, training systems/ educational institutions, information systems and other organizations that public institutions can hire and fund through competitive bids, and does not seek to monopolize a particular training or provide it directly where such services can be provided. donors do not pick “winner” ngos or businesses or agencies to receive support in ways that could disfavour competitors from developing. an assessment is done at the beginning of the project of the means of developing this kind of service market for the capacity needs of governmental and non-governmental organizations (following a business development services (bds) model). scoring: yes – 0 (no loss of points), if the project shows awareness of this and protects against harm no explicit policy at the organization to protect against this, but no clear sign that abuse has occurred – (-0.5) (loss of half of a point) no, and there are signs that this abuse has occurred or is very likely to occur – (-1) (loss of a point) european journal of government and economics 4(2) 173 2.5. how some organizations do after understanding how the indicator works, it is easy to apply to every new case in just a few minutes and with close agreement among anyone using it. in the appendix, in section iv, are several examples including many of the standard approaches that are now widespread in the field, showing how different organizations and projects score, from best to worst. rather than score specific projects in particular countries, some of the projects are generalized in project categories that are common in the field, showing the range of scores that they earn depending on which particular features are included in certain types of projects by specific donors and proponents. 3. post-script: solutions the irony of exposing the flaws in development projects today is that the “experts” who are in the position to make changes have little incentive to change, while those who are best protected by change are the least informed and organized about where or how to begin to push for reforms. an indicator can facilitate change, but like other improved tools, it must be in the hands of those willing and able to use it. organizations that score the worst on the new indicator in this article will likely not even recognize their failures. (they are likely to say that professional compliance standards do not apply or that projects are “too diverse” or that measurement is “too difficult” or “subjective”. they may say that this business-like approach that introduces a variety of professional expertise takes the artistry and “humanistic” or “human” judgment out of their work, though in fact it does the opposite by applying their own standards to their work. they are likely to respond defensively to suggestions for more public oversight of their work and to claim that accountability is a form of “policing,” even though they accept the idea of “accountability” as one of the key principles of effective capacity building. they are likely to say that oversight implies “mistrust” and that their good faith is being questioned, in the premise that they are above the law and the public is (by their design) ignorant and uninformed about what they do. they may say that holding a government official accountable for results is unfair because there are “too many factors”) overall, such responses from many “professionals” in “good governance” will demonstrate exactly why many of the people in place in current systems are part of the problem and not the solution. indeed, the only real solution is mobilization of the public. this author has suggested the formation of donor monitor ngos that act as public advocates (lempert, 2008) and has designed a full set of other governance reforms in media, organizational oversight, private attorneys general and other citizen powers that would promote professionalism and accountability at the level of constitutional changes (lempert, 1997) as well as educational and cultural reforms. but who will fund and promote them? in short, foxes have entered the henhouse in design and implementation of capacity building projects in international development as well as many other governmental systems, and there is a need to devise better oversight systems. the only way that change can really occur is if those public voices who have an interest in the oversight act collectively to protect their interests. in the case of capacity building for “governance” those interests are a stable and democratic world as well as control over billions of dollars in public money that is diverted. this article offers one tool, a weapon of empowerment, to at least facilitate that effort, as part of a codification of laws and standards that could ultimately be enforceable both by those paying for the interventions and those at the receiving end. this indicator takes away excuses that oversight is too difficult for ordinary citizens and that we must simply wait, pray, and rely on experts to change in ways lempert ● a quick indicator of “capacity building” initiatives 174 they have little incentive to change, rather than to take on the burdens of citizenship to protect the public interest in promoting effective, efficient and law abiding development interventions. references acharya, meena and art wright (2000) ‘an evaluation of the impact of the united nations system on capacity building for poverty eradication’, a report for the department of economic and social affairs of the united nations, december 15. boesen, nils and ole therkildsen (2004) between naivety and cynicism: a pragmatic approach to donor support for publicsector capacity development. copenhagen: ministry of foreign affairs. boesen, nils, peter christensen and ole therkildsen, ole. (2002) ‘evaluation of capacity development in a sector programme support context: contributions to an analytical framework’, copenhagen: danish international development agency (danida). url (consulted 17 june 2015): http://www.um.dk/publikationer/danida/english/evaluations/capacitydevelopment2 00501/11636_cdoe_asp/annex1.asp busan partnership for effective development cooperation. (2011) ‘fourth high level forum on aid effectiveness’, url (consulted 17 june2015): http://effectivecooperation.org/resources/ ember, carol r. and melvin r. embert (2006) cultural anthropology. 12th edition. united states: prentice hall. christensen, robert k., helen k. liu, stephanie moulton, becky nesbit and eva witesman (2006) ‘assessing the effectiveness of capacity measures and frameworks in public and nonprofit performance evaluation’, url (consulted 17 june 2015): http://arnova.omnibooksonline.com/2006/data/papers/pn062016.2.pdf consulting, elton (2002) ‘the bronte catchment project: a case study in capacity building’, elton consulting march 2002. url (consulted 17 june 2015): http://www.elton.com.au/pdfs/capacity.pdf dresang, dennis l. (2009) personnel management in government agencies and non-profit organizations. london: longman. enemark, stig (2003) ‘best practice in capacity building: understanding the concept of capacity building and the nature of land administration systems’, fig working week, paris, france. emmanuel, clive, kennet merchant and david otley (1990) accounting for management control. london: chapman & hall. european commission (undated, accessed in 2003) handbook on promoting good governance in ec development and cooperation. european aid cooperation office, publication j54-07/20. european commission (2002) measuring democracy and good governance. conference, munich. url (consulted 17 june 2015): http://www.inwent.org/imperia/md/content/bereich4-intranet/abteilung4-06/d1.pdf forrester, jay (1994) ‘systems dynamics, systems thinking and soft or’, system dynamics review 10(2):245-256. freedom house. (2006) methodology. url (consulted 17 june 2015): http://www.freedomhouse.org/template.cfm?page=352&ana_page=330&year=200 6 garrison, ray h., eric noreen and peter c. brewer (2005) managerial accounting. new york: mcgraw-hill irwin. european journal of government and economics 4(2) 175 gubbels, peter and catheryn koss (2000) from the roots up: strengthening organizational capacity through guided self-assessment. oklahoma: world neighbors. honadle, beth walter (1981) ‘a capacity building framework: a search for concept an purpose’, public administration review 41(5): 575-580. kaufmann, daniel, aart kraay and massimo mastruzzi (2007) ‘governance matters vi: aggregate and individual governance indicators 1996-2006’, the world bank, policy research working paper, no. 4280. kekic, laza. (2007) ‘the economist intelligence unit’s index of democracy’, url (consulted 17 june 2015): http://www.economist.com/media/pdf/democracy_index_2007_v3.pdf lempert, david (1997) ‘holding the powers that be accountable to our ethics code to protect our integrity and the peoples we serve’, human rights 24(2):2224. lempert, david (2008) ‘why we need an international development donor monitor’, policy innovations. url (consulted 17 june 2015): http://www.policyinnovations.org/ideas/policy_library/data/01464 lempert, david (2009a) ‘a dependency in development indicator for ngos and international organizations’, global jurist 9(2). url (consulted 17 june 2015): http://www.degruyter.com/view/j/gj.2009.9.2/gj.2009.9.2.1296/gj.2009.9.2.1296.xml ?format=int lempert, david (2009b) ‘why government and ngo projects fail despite ‘evaluations’: an indicator to measure whether evaluation systems incorporate the rules of good governance’, journal of multi-disciplinary evaluation 6(13):58-108. lempert, david. (2010a) ‘a human rights education project indicator for ngos and international organizations’, interamerican journal of education for democracy 3(1): 45-72. lempert, david (2011) ‘democracy building democratic project indicators for ngos and international organizations’, global jurist 11(2) doi: 10.2202/19342640.1385 lempert, david (2012) ‘a quick indicator of effectiveness of “income generation” and “sustainable business initiatives” in international development’, economology journal 2(2):28-90. lempert, david (2014) ‘universal development goals for this millennium’, consilience: the journal of sustainable development 12(1): 1-24. lempert, david (1996) escape from the ivory tower: student adventures in democratic experiential education. michigan: jossey-bass publishers. lempert, david, nguyen, hue nhu (2008) ‘how to tell if development projects are doing sustainable development: an indicator for ngos and international organizations’, international journal of sustainable societies 1(1):44-54. linnell, deborah (2003) evaluation of capacity building: lessons from the field. washington: alliance for nonprofit management. lusthaus, charles, marie-hélène adrien, gary anderson, fred garden and george p. montalván (eds) (2002) organizational assessment: a framework for improving performance. new york: international development research centre. marshall, monty, keith jaggers and ted r. gurr (2006) polity iv project: political regime characteristics and transitions 1800-2006. url (consulted 17 june 2015): http://www.systemicpeace.org/polity/polity4.htm lempert ● a quick indicator of “capacity building” initiatives 176 mayo, dambisa (2009) dead aid: why aid is not working and how there is a better way for africa. new york: farrar, straus and giroux, mckinsey and company and venture philanthropy partners (2001) effective capacity building in non-profit organisations. url (consulted 17 june 2015): http://74.125.153.132/search?q=cache:px5folpu5yj:venturephilanthropypartners.org/learning/reports/capacity/full_rpt.pdf+effective +capacity+building+in+nonprofit+organisations.+mckinsey+and+company.+2001&cd=1&hl=en&ct=clnk measure evaluation (1999) measuring capacity building in health and population programs,” conference proceedings for u.s. agency for international development. arlington, virginia. url (consulted 17 june 2015): http://pdf.usaid.gov/pdf_docs/pnacj170.pdf muller, duane (2007) usaid’s approach to monitoring capacity building activities: experiences, lessons learned, and best practices. url (consulted 17 june 2015): http://74.125.155.132/search?q=cache:vrmlior3eqwj:www.encora.eu/coastalwiki/i nstitutional_capacity_building+institutional+capacity+building&cd=5&hl=en&ct=cln k. musgrave, richard and peggy b. musgrave (1973) public finance in theory and practice. new york: mcgraw hill. nelson, debra l., james c. quick (2005) understanding organizational behavior. mason: thomson south-western, 2nd edition. noe, raymont a., john r. hollenbeck, barry gerhart and patrick m. wright (2007) human resources management. new york: mcgraw hill. nye, joseph s. (2004) soft power: the means to success in world politics. new york: public affairs. ogiogio, gene (2005) ‘measuring performance of interventions in capacity building: some fundamentals’, africa capacity building foundation, occasional paper no 4. organization for economic cooperation and development. (2008) aid targets slipping out of reach? france: paris. url (consulted 17 june 2015): http://www.oecd.org/dataoecd/47/25/41724314.pdf otoo, samuel, natalia agapitova and joy behrens (2009) the capacity development results framework. a strategic and results-oriented approach to learning for capacity. washington d.c.: the world bank. paris declaration on aid effectiveness. (2005) url (consulted 17 june 2015): http://www.oecd.org/dataoecd/11/41/34428351.pdf raffer, kunibert and hans w. singer (1996) the foreign aid business. london: edward elgar. robbins, stephen p. (2002) organizational behavior. london: prentice hall. samuel, otto, natalia agopitova and joy behrens, joy. (2009). capacity development results framework. washington: world bank institute. schachter, mark (2000) capacity building: a new way of doing business for development assistance organizations. ottawa: ottawa institute on governance. seidman, ann and nalin abeyesekere (2000) legislative drafting for democratic social change: a manual for drafters. london: kluwer law international. stiftung bertelsmann (2006) ’bertelsmann transformation index: political management in international comparison’, url (consulted 17 june 2015): european journal of government and economics 4(2) 177 http://www.bertelsmann-transformationindex.de/fileadmin/pdf/bti_2006_brosch_re_gb.pdf todd, landman. (2006) indicators for human rights based approaches to development in undp programming: a users’ guide. new york and oslo: united nations development programme. url (consulted 17 june 2015): http://www.undp.org/oslocentre/ usaid (1998) handbook of democracy and governance program indicators. washington, d.c.: center for democracy and governance. vanhanen, tatu (2000) ‘a new dataset for measuring democracy’, 1810-1998’, journal of peace research 37(2):251-265. world bank (2000) reforming public institutions and strengthening governance. washington, d.c.: the world bank publishin world bank (2001) committee of donor agencies for small enterprise development, business development services for small enterprises: guiding principles for donor intervention. world bank (2005) capacity building in africa: an independent evaluation group evaluation of world bank support. washington d.c: world bank. lempert ● a quick indicator of “capacity building” initiatives 178 appendix section i: principles of "capacity building" two long-standing definitions offered by two major international donor organizations, undp and the european commission, dating back 10 to 20 years, touch on what are essentially five key elements for capacity building in working with foreign governments and civil society in development. these key elements are highlighted in the text with numbers, with the first three relating to the three critical levels at which capacity building interventions must operate and the fourth referring to a principle of good governance. indeed, they come out of the development literature going back at least thirty years (honadle, 1981). undp (1991) defined 'capacity building' as: 1) the creation of an enabling environment with appropriate policy and legal frameworks, 2) institutional development, including community participation (of women in particular), 3) human resources development and strengthening of managerial systems. 4) undp recognizes that capacity building is a long-term, continuing process, in which all stakeholders participate (ministries, local authorities, nongovernmental organizations and water user groups, professional associations, academics and others). the european commission definition highlights these four areas with a particular emphasis on the aspects of accountability to the public (the fourth category above): ‘to develop and strengthen structures, institutions and procedures that help to ensure: transparent and accountable governance in all public institutions; improve capacity to analyze, plan, formulate and implement policies in economic, social, environmental, research, science and technology fields; and in critical areas such as international negotiation.’ (world bank, 2005:26). several documents of major international donor organizations have elaborated on these four areas, above, in documents that are readily available and repeated in multiple undp, world bank, and other sources, including even wikipedia’s page for ‘capacity development.’ the wording is sometimes a bit different, but the concepts are the same. to help make the jargon for the three different levels of organizational performance intelligible, the terms can be clustered as follows.  the ‘enabling environment’ – the legal and political authority, resources and incentives, for an organization to fulfill its legal, public, established purpose is sometimes referred to as the ‘institutional capacity’ (slightly confusing it with ‘institutional development’) to denote the “rules of the game”; the legal or cultural environment in which organizations are constrained and directed (world bank, 2005). the undp refers to this as the ‘broader system/societal level’ or ‘systems level’ (undp, 1998). added to legal and regulatory framework and policy concerns are issues of resources, management, and accountability. major donors do not often use the word “culture” (sometimes they use the euphemism, “social capital” to assume that all societies are on a single path to development and simply lack certain aspects), but of course the cultural fit of a particular institutional function is a necessary part of an analysis of a sustainable system in the development context and is part of a complete analysis of the operating environment.  “institutional development” for effective and efficient management is sometimes defined as ‘organizational development’ or ‘organizational capacity’ (world bank, 2005) with ‘institution’ referring to the overall political or government system as an institution and parts within it or alongside it (private european journal of government and economics 4(2) 179 businesses or community based organizations) as the ‘organizations’ or ‘entities’ (undp, 1998). the capacity building work here is that applied regularly in the business and community based organization (cbo) sectors to improve efficiency and performance through strategic planning and improvement of mission as well as building institutional resources and applying effective financial and organizational management, staffing, accounting and control, including the quality of feedback and evaluation systems.  “human resources” (skills) is sometimes referred to as ‘human capacity’ (world bank, 2005:27). the undp also refers to this as the ‘group of people level/individual development’ (undp, 1998). sometimes ‘education and training’ is also further elaborated as skills, information and perspectives (three sub-areas) that can all be objectively tested and tied to performance.  in addition to the three levels above and accountability to the public, most definitions implicitly offer a fifth category; that of the long term sustainability of the capacity to perform the necessary function and the ability to attract resources to assure that sustainability. a undp definition made it clear that capacity was part of a: 5) ‘continuing process’ whereby individuals or organizations units perform functions ‘sustainably’ as well as effectively and efficiently (undp, 1998:10). others reiterate this as ‘the attraction, management and absorption of resources’ (honadle, 1981) or as an ‘ongoing and sustainable fashion’ (elton consulting, 2002) or with the flexibility to deal continually with a changing environment. there is also an understanding that these five elements must all be taken together. the undp makes clear that, ‘capacity building should be seen as a comprehensive methodology aiming to provide a sustainable outcome through assessing and addressing a whole range of relevant issues and their interrelationships’ with none of these areas to be viewed alone (undp 1998). these elements and their relationship are presented in the chart below. lempert ● a quick indicator of “capacity building” initiatives 180 figure: the five key principles of capacity building in conducting capacity building with non-profit organizations, consulting firms like mckinsey have introduced other frameworks that highlight concerns using different words but essentially reiterate the three ‘systems levels’ along with other elements like ‘culture,’ ‘strategy,’ ‘aspirations’ (mission and vision) and ‘organizational management’ that are already included components of the framework above (mckinsey, 2001). that, in itself, is the basic standard of the tool. like any tool, it also must be part of an established set of routine procedures for development interventions. there are a variety of studies of capacity building by donors that simply put them within the context of ordinary procedures for results based management, appropriate measures of the problems to be solved, the root causes of the problems, use of logframes to match interventions to steps in the root causes of problems, and use of standard cost-benefit analysis procedures and baselines to assure efficient use of resources to achieve measurable results promoting appropriate development objectives (boesen and therkildsen (danida), 2004; boesen, christensen and therkildsen, 2002). this is also essentially what the world bank evaluations and various undp evaluations refer to when they fault their organizations for not applying the standard procedures of the field to these particular tools (undp, 1998, 2002, 2004; world bank, 2000, 2005; acharya and wright, 2000; linnell, 2003). as evaluators have noted, there is nothing magical about capacity building. it simply requires following standard technical practices that are well established in the fields of public administration, law, general business management (strategic planning, personnel management, accounting for managerial control). these are basic textbook skills.   1. policy and legal framework  3. human capacity  2. institutional development/ management efficiency  organizational function: 3 levels of capacity 5. sustainability 4. public accountability european journal of government and economics 4(2) 181 there is certainly room for difference on achieving higher quality or applying new techniques for better results, as in all specialized and technical fields. but the key criticisms that have been launched in this field are not over the use of innovative new practices. they are simply about whether the most basic standards are being applied. and these can very easily be stated, referenced in standard texts, and put into an indicator to see whether they are being followed. just as in medical practice, a standard diagnosis requires certain tests, adherence to certain standards of cleanliness, patient ethics and care, and drug protocols, the basic test of competence is whether these are all followed. once the basics are adhered to for licensing purposes, then there can be disagreements about new tests and procedures and their value. at this stage with “capacity building,” however, we are still at the basic stages simply of establishing whether practitioners meet the basic “licensing” standards for establishing minimal competence. given that organizations have already defined the goals and purposes of capacity building and given that the procedures for design and evaluation of development interventions is also routine, it is relatively simple to run through a checklist to determine whether all of these elements are actually being followed or not in any specific project intervention and for holding development actors accountable. so why has no one sought to, or been able to do it? section ii: indicators in the field and the lack of an indicator for “capacity development” interventions practitioners have sought to create measurements for what is most difficult and what is fraught with political and disciplinary issues, but they have yet to offer a measurement in the area where there is already political and professional agreement; over whether their own activities comply with legal and professional standards; something that can be determined with a simple checklist. political scientists and development organizations have created a series of indicators to measure whether countries as a whole have the attributes of “good governance” that industrialized donor nations (politically) determine reflect the appropriate standard. these generally seek to measure the quality of governance in their entirety and in the aggregate, but without looking at whether specific projects actually improve governance and meet international standards (defined in united nations treaties) for sustainably protecting cultural or community assets. for example, the world bank’s devised indicator uses a definition of governance that reflects its own goals of effectively managing the loans it gives to government officials to ensure that the bank’s objectives are adhered to. the indicator aggregates subjective views on ‘voice and accountability, political stability, government effectiveness, regulatory quality, rule of law and control of corruption’ (kaufmann, kraay, mastruzzi, 1997). similarly, freedom house, a u.s. government funded organization, measures whether certain processes meet their subjective standards in seven areas: (electoral process; civil society; independent media; national democratic governance; local democratic governance; judicial framework and independence; and corruption) and scores whether a country is ‘democratic’ according to its own cultural preferences, rather than whether governance is effective in meeting local needs or even if it is in keeping with international law. a country could score well on this index but still destroy all of its minorities or sell all of its resources in violation of international treaties (freedom house, 2006).) a number of other indicators that are used in political science data sets do not seem to be used at all by development organizations and do not have direct applicability to measuring the quality of capacity building. these indicators rank countries (polity iv; marshall et al, 2006), pick specific attributes (polyarchy 1.2 that measures ‘competitiveness and political participation’ (vanhanen, 2000); lempert ● a quick indicator of “capacity building” initiatives 182 generates an ‘index of democracy’ based on five categories (used by the economist magazine, kekic, 2007)) or substitutes goals of economic transformation that are in direct contradiction to the fundamental principles of good governance that require protecting assets and promoting sustainability (bertelsmann transformation index in stiftung bertelsmann, 2006). though one might expect practitioners to have clear and simple indicators to support and justify program spending, the reality is that the development bureaucracies seem to be even more confused (or politicized) in setting the goals of governance and capacity building interventions than political scientists and in agreeing on what to measure. among development organizations, the one that has come closest to developing an indicator for ‘capacity building’ is the u.s. agency for international development (usaid). usaid’s failure in this area, however, is that it does not start with a basic indicator for ‘capacity building’ or governance initiatives but tries instead to define every kind of outcome or outcome measure they would like to see for specific interventions. in other words, they substitute individual trees for the goal of protecting a forest. for example, usaid’s measure of effective capacity building in the training of officials is whether or not it leads to ‘elected officials who have been trained’ (a restatement of the input), ‘and who say that they are using their new skills on the job’ with ‘examples of how they are using it’ (usaid, 1988, page 166). indeed, as an effective measure, this one violates the basic principles of governance since it turns an input (training) into the measured output and offers only subjective reports from stakeholders with no objective measures of improved performance to citizen beneficiaries. this is more an example of what is wrong than an attempt to provide measures. though usaid also mentions that it would be useful to measure the ‘skill level’ of ‘salaried staff,’ this also has no relation to performances and outcomes. to put it more bluntly, the increased efficiency of an authoritarian or fascist regime in its executions, due to higher skill level, would receive high marks on a usaid indicator. so would simple bribes to government officials or indoctrination that was disguised as ‘training.’ a more recent document on ‘usaid’s approach to monitoring capacity building activities’ simply offers some formulaic suggestions such as: ‘avoid broad statements,’ ‘be inclusive’ and ‘be selective’ (muller, 2007). a usaid review of recent indicators showed little in the way of other approaches (measure, 1999). rather than offer indicators, most development organizations simply offer checklists of ideas to consider when doing “capacity building.” these can more easily be described as a set of cookbook recipes than as a legal means of applying standards of accountability. organizations offering this “recipe” approach are the world bank and asian development bank (ogiogio, 2005; otoo et al, 2009), the u. n. system (that also offers national surveys of public perceptions rather than any objective professional standards, landman, 2006; undp 2004), and the european union (ec, 2002; 2001; ec undated, pre-2003). in many of these documents the definitions are circular and the goals become ‘partnerships’ and ‘policy’ improvement without any measurable content. among the most recent of these, one published by the world bank institute, there is still confusion between the tool of “capacity building” and any kind of development intervention, as well as between the content delivered and the form of delivery. the ‘capacity development results framework’ is described as doing everything that development does – ‘design, implementation, monitoring, management, and evaluation of development projects’ – in ways that are so broad as to make it useless (otoo et al, 2009). such ‘frameworks’ do not start with analysis of the problem, with problem trees, as a diagnostic for lack of capacity but seek to cover everything, including country-wide programme cycles, ‘national development strategies, 5-year plans, and visions for the future’ (otoo et al, 2009:11). essentially, in place of examination of capacity building, they have european journal of government and economics 4(2) 183 substituted frameworks for how to create logical frameworks for any kind of development intervention through any kind of modality. moreover, most of the “framework” (rather than “measurement”) literature comes from publications of the donors, themselves, with little outside to create accountability. indeed, there still does not appear to be even one international journal devoted to “capacity building” anywhere. the result is that the real problems in the field, of hidden agendas for corruption and dependency, are never examined other than in vague euphemisms to protect those responsible, using bureaucratic jargon like ‘unintended negative consequences’ and lack of ‘clarity of mission’ or lack of ‘supportiveness of stakeholders’ among a long list of ‘capacity factors’ (otoo et al, 2009:12). at best, the focus is only on ‘educational impact analysis’. the problems remain hidden, by design. similarly, a number of diagnostics have been developed in the actual practice of “capacity building,” mostly with community based organizations (mckinsey, 2001; christensen et. al., undated; gubbels and koss, 2000; lusthaus et al, 2002). mostly these parallel and actualize approaches that in the business literature would be called ‘strategic management’ or ‘accounting for management control’ (emmanuel, merchant and otley, 1990). however, they have yet to be applied as accountability tools to the billions of dollars in international interventions. it is ironic that while there are two recent international treaties that guide international interventions and reference the idea of standards – the paris declaration on aid effectiveness (2005) and the accra agenda for action (2008)44, now followed and reinforced by the international aid transparency initiative signed in busan (2011) – they offer few specifics and little enforcement, thus reflecting the exact problems they claim they are trying to solve. with so much effort to develop “indicators,” the fact that none of them seem to do what is so simple seems to suggest how deeply rooted are the cultural and institutional barriers to holding even governments in western “democracies” accountable for billions of dollars of public spending. section iii: the problem with many “capacity building” projects and the real value of an indicator the world bank and other organizations’ evaluations readily admit the problems of current capacity building approaches, though they do so euphemistically. it is easy to take their own words and to expose what is going on in simpler language. what the donors’ own internal critics have exposed is that project fail because donors want them to fail to protect other agendas and because major international organizations are using “capacity building” as a cover to bribe or co-opt officials for foreign agendas. one world bank study admitted both the goals and the mechanisms of how this works, noting that ‘governments generally are inclined to improve services demanded by powerful interests’ and then noting specifically that it is ‘donors’ payments’ that ‘subordinate the coherence of the machinery of government to the narrower goal [and] short term gains … of project implementation’ for the donors (world bank, 2000:41). the mechanisms are also clear. often the corruption that foreign donors claim they are seeking to stop are actually initiated or reinforced by donors themselves, with “capacity building” one of the ways of buying off of government officials directly through what the world bank itself admits are ‘donors’ payments of salary supplements’ (i.e., what could be seen as the equivalent of illegal bribes under the 44 accra agenda for action (2008) url (consulted 17 june 2015): http://siteresources.worldbank.org/accraext/resources/4700790-1217425866038/accra_4_ september_final_16h00.pdf. lempert ● a quick indicator of “capacity building” initiatives 184 u.n. declaration against corruption and bribery45), in addition to ‘grants and concessional loans’ to ‘encourage line ministries … to “market” to donors’ (world bank, 2000:41). the world bank notes that the approach is usually to exclude the public, ‘interacting exclusively with government interlocutors’ while disguising bribes in the form of ‘computers and other inputs’ where it is obvious that no changes in services will occur because of ‘the absence of deep and sustainable demand for institutional reform.’ moreover, they note that, ‘this applies to much of the donor community’ (world bank, 2000:14). indeed, this reinforces what the previous indicators designed by this author to test foreign projects for ‘dependency/colonialism,’ ‘sustainability,’ and ‘democracy’ helped reveal; that many of the large donors are in fact continuing to pursue a colonial agenda with little oversight or accountability and the creation of tools that hide their underlying intent (lempert, 2009a, 2011; lempert and nguyen, 2008). most often the ministries or government organizations to be ‘built’ (what many observers would view as a euphemism for co-opted with foreign gifts, funds and advisors) are the ministry of planning and investment, to serve as a (dependent or ‘colonial’) intermediary of international investment banks (mayo, 2009), the judiciary in order to protect international businesses; the parliament in order to write the laws that open their economies and systems to foreign businesses and other influences and then to promote these changes to the public; the head ministers in order to write the development plans and policies and claim they are domestic aspirations rather than foreign driven; ministry of health to stop contagious diseases from leaving the country, etc. the goal in building capacity in the non-governmental sector is often to generate business joint ventures to the benefit of foreign business, to influence current and emerging leaders of political parties while improving their ability to manipulate their citizenry to support foreign agendas, and to build a foreign funded rather than locally accountable or sustainable civil society to lobby for foreign interests (generally for foreign investment; often for ‘women’s rights’ as a way to destroy traditional practices and free women for work in export processing zones or as administrators for the foreign sector) (lempert, 2009a, 2012). in the author’s experience over 30 years, in almost all of these projects, the public is excluded by design and “accountability” is really accountability to the agenda of the donors, not of any parts of government to citizen oversight, and the amount of secrecy has actually been increasing with consultants forced to sign statements of confidentiality that the author, as a lawyer, believes to be in direct violation of most public transparency laws (lempert, 2009b). as the world bank itself noted, the public ‘voice’ is systematically excluded (world bank, 2000:43). the fact that this happens is an open secret with little or no influence either from the taxpayer/donor public or from the publics in countries where the projects are being run. in fact, most international development organizations have now established secrecy clauses (in violation of public laws in both donor and recipient countries) to hide information about these projects under the pretext that officials in recipient countries would not candidly discuss their capacity “needs” if they believed the public might learn of their incompetence and fallibility. some vignettes taken from some actual projects in which the author is familiar (in the section below) highlight the absurdities and flagrant abuses that now occur. in presenting these cases, the author draws on professional experience and participant observer methodologies used in the field of social anthropology as well as determinations applying methodologies of law and public administration. the mechanisms that cloak these bribes as “capacity building” are relatively transparent and are easy to reveal with a good indicator that exposes lack of 45 u.n. declaration against corruption and bribery in international commercial transactions (1996): a/res/51/191. url (consulted 17 june 2015): http://www.un.org/documents/ga/res/51/a51r191.htm european journal of government and economics 4(2) 185 safeguards against conflicts of interest of donors themselves, and of recipient government “stakeholders” and the lack of transparency has nothing to do with encouraging reform and everything to do with covering up the abuses. the extent of these abuses now appears so widespread that most professionals in development can easily list the forms of what could best be described as “shadow bribes” take without a moment’s hesitation. this author has personally witnessed them delivered, in violation of local and international laws, by almost every donor organization in the form of attendance fees to government officials for simply coming to donor workshops (listed as “per diems” or envelope money) so that donors can then seek and spend more money on the pretext that officials are eager for such training and that they have an impact, “study tours” that are disguised junkets, as well as hotel buffets and entertainment (usually in lavish western settings) in recipient countries. officials are also easily bought in the name of capacity building; with donor organizations working in fact as tour agents and caterers, running parties, giving away branded items with logos (including umbrellas, travel bags, and briefcases). institutions like the un appear to have been turned into lobbying agencies or missionaries, for hire by the larger donors who pay them to use government contacts to promote specific agendas of the international banks or of economic blocs seeking to expand their influence in trade agreements or “harmonization” of laws, convincing national officials to legislate and adopt such policies and to claim that it is their local people who have demanded them. many other projects appear to be simply naïve, throwing money at symptoms of incompetence and inefficiency with no idea how organizations work, what the incentives are of individuals within them, or what they are even supposed to do in a functioning and sustainable system and this is a result both of the hiring of staff with few skills (other than language) and little adherence to codes of professional responsibility, with almost no outside direct public oversight. project approval procedures have largely been corroded in order to facilitate (what can be described as “corrupt”) agendas for purchasing government systems that keep money flowing, many projects are implemented with no standard procedures to follow. as a more recent world bank review of ‘capacity building’ projects admitted, most such projects ‘do not specify the capacity building objectives,’ lack ‘adequate needs assessments’ and do not even consider ‘processes of organizational and behavioural change’ (ogiogio, 2005). thus, large numbers of projects come to be funded where the goal of “training officials” is about as effective and well thought out as schemes teaching elephants to fly. officials who lack the salary, the ideology, the incentives or the capacity will not produce services no matter how much education they have. in other cases, government systems that are riddled with nepotism or politics, receive capacity building funds that essentially seek to turn secretaries into brain surgeons without recognizing that the processes of recruitment that are politicized and corrupted and are the source of the problem. rather than improve organizations, foreign funds allocated for “capacity building” are often propping up former generals and their children in regimes that are largely recognized as corrupt (by transparency international and other organizations measuring accountability) in order to keep them employed and in power; building the “capacity” of dictatorships to maintain their rule. “awareness raising” projects for “accountability” and “combating of corruption” often throw training at the very people who are the perpetrators of the crimes and as effective and well thought out as schemes trying to turn wolves into herbivores. the donor justification is that transferring funds to government officials to help “demonstrate” how to run fair judicial systems or equitable public services, can help convince them to “buy in” to the idea. yet, without any pressures or conditions to pressure them to do that or convince them why it might be in their interest to do so, the approach is akin to believing that giving money to criminals for charity will turn them into humanitarians rather than simply end up in their foreign bank accounts. lempert ● a quick indicator of “capacity building” initiatives 186 other projects offer remedial education to specialists or adults when the real failures have occurred in the basic education system, in professional schools, or in selection procedures. a lack of proper analysis of the sources of the problem again results in treating symptoms rather than the disease. the irony of “capacity building” is that the organizations claiming to be the experts are often the last to have their own internal experts who have read the basic textbooks in the field at the university or masters level. few seem to follow the basics of development project design and “results-based management” (with standard practices like problem trees, root cause analysis, appropriate identification of inputs-outputs-outcomes, and use of baselines, and cost benefit analysis) or who even apply their own international treaty agreements and mission statements in areas of sustainable development and protection of cultural rights and diversity (rather than promotion of industrialization, consumption and economic exploitation) (lempert, 2014). there is nothing here that cannot be improved simply by holding donor organizations to safeguards against conflicts of interest and to textbook principles. a list of some of the standard textbooks that apply to capacity building in this field would include those of overall systems and system change (forrester, 1994), principles of legal drafting for democracy and efficacy (seidman and abeyesekere, 2000), understanding of human cultural differentiation and sustainability (ember, 2006), basic business analysis and organizational strategy for effective management control (garrison, noreen and brewer, 2005), principles of public finance (musgrave, 1973), organizational behaviour (robbins, 2002; nelson and quick, 2005), personnel management (noe, gerhart, wright and hollenback, 2007; dresang, 2009), training and education theories to differentiate skills, perspectives, and information (lempert et. al., 1995), and current models of promoting competitive business development services (world bank, 2001). others might be added in the areas of appropriate roles of ngos (innovation and advocacy) versus government (services), and mechanisms for accountability and avoiding conflicts of interest. the point is simply that the disciplines and approaches are well established but many donors seem not to even know their own profession; perhaps deliberately. vignettes: what really happens in "capacity building" projects some standard project types and how they are rigged or doomed to “fail” from the start are described below. projects to turn wolves into doves international projects work with all three branches of foreign governments in approaches that seemed designed to keep the wrongdoers in power by assuring that only they have the “expertise” to run government (in ways acceptable to the donor). while projects with ministries seek to improve them through proselytizing rights treaties to the very people who are the abusers and who have no incentive to change, the classic examples of capacity building failures are those that work with the legislative and judicial branches of governments in developing countries. the responsibilities of these political institutions are to represent the public will in ways that balance interests of different cultures (to assure their sustainability) and of individual rights and preferences, through legislation (legislatures) and through fair and representative means of resolving disputes (adjudication). the “capacity” they need is the capacity to protect and reflect these interests in the process of legislation/policy and in conflict resolution among different interests. that is not what foreign projects seek to promote. what they do, instead, is the following in three basic types that this author has worked on directly.  undp parliamentary capacity building projects with donor support – “parliamentary capacity building” projects are run almost everywhere, with european journal of government and economics 4(2) 187 funding from almost every major donor. those that the author has seen in several regions show no intent to address the real failings and capacity needs of national or local legislatures. if parliaments are failing, it usually means that democracy itself is missing since this is the real capacity underlying a legislature: the public is badly represented, unable to control its representatives, unable to understand what legislation is and how to offer it, and/or that the legislature itself is a symbolic puppet (a legacy of colonial rule) that has no real power to challenge military, police, or economic (domestic or foreign power). it may mean that the legislature is simply a ritual body with no real direct control of most government spending (that may come from overseas governments rather than from taxes or from elite control over some resource that generates funds). in a functioning parliament, the public sets the agenda for sustainable development, the courts assure that the legislature upholds international treaty goals for sustainable development, and competence of parliament is assured through the power of parliament to increase its funding through taxes and to hire the best people to carry out its tasks. failed officials are voted out of power and their failed staffs go with them. what happens in these projects, however, is that the foreign donors seek to reinforce the idea of permanent, “efficient” and “expert” parliamentary staff alongside career parliamentarians who remain no matter what election results say, with training in foreign languages and in passing of foreign laws favoured by the donors. training includes instruction to elected deputies on how to appear democratic by meeting (often for the first time) with their constituents to show how “democratic” they are. the projects promote connections with foreign parliaments in donor countries as a way to create a brotherhood among these elites and in ways that reinforce dependency for future favours.  world bank ministry of justice projects to promote “rule of law” – while these projects in “efficiency” of “administration of justice” are most often supported by the world bank, they are copied by other major donors. as with parliaments, if judiciaries fail, the capacity problem is almost always in their democratic legitimacy rather than in their inability to efficiently or expertly administer (in)justice. where judiciaries fail, it usually means that the public is not represented as jurors or in democratic oversight and balance of judges, that citizens lack the power to enforce norms on military, police, or domestic and international economic powers, and that the public and judges lack the funding to be educated and involved in complex decisions. working judiciaries use the law to bring in the funding and expertise they need, with public support, to challenge abuses of power and imbalances in representation. what happens in these projects is that rather than focus on “justice” and equity, donors seek to reinforce the idea of permanent, “efficient” and “expert” elite selected judges who are “independent” of the public but still vulnerable to existing economic and military power. support for “capacity” includes study tours, computers and training of judges. projects for legal “access” mostly seek to assure that the legal order is efficiently enforced on individuals in minor matters and that edicts are known to the population rather than subject to change. on one japanese funded project that the author worked on for the world bank in a country described as having no rule of law and where project funding was slowed due to violation of bank regulations against insider dealing of bank funds (detailed in world bank documents), a ministry of justice not only gave an ultimatum that the author steer loan funds to a committee of their friends to allocate funds for study tours and computers, but they also demanded a kickback from my salary to the ministry. the author’s approach was to offer loans to students and to promote lawyers representing citizens, farmers and workers to benefit small business and sustainable development, rather than to just promote business law and extraction of that country’s resources (oil). after more than 15 years, the world bank lempert ● a quick indicator of “capacity building” initiatives 188 continues to support the wrongdoing (and the same ruler and legal system remain, with the country increasing their oil contracts).  major donor projects on “human rights” and “anti-corruption” with ministry officials – in many fields, capacity building has developed as part of an “industry” by certain donors to propagate certain treaties or moral values on the pretext that such proselytizing solves problems. donors rushing to offer support given on the size of the perceived problem (violations of rights or corruption). in most cases, the real problem is that ministry officials have no real power or incentives to confront wrongdoing and the public (and communities) are too weak to protect and enforce their own established rights traditions (in many cases, traditions that were at times more progressive than those of developed countries but that were and continue to be destroyed by colonialism and globalization). what happens on these projects is that rather than describe the power imbalances or the reasons why cultural systems that protected rights were destroyed, the projects focus on these government officials who are either powerless, indifferent, or the source of the problem; identifying them as “duty bearers” rather than as oppressors or spectators. the approach is purely symbolic and fits the definition of a public relations ‘whitewash’ or ‘rights wash’ (lempert, 2011). typical projects are those offering “human rights training” for police and prison guards who themselves may have been victims of violence or who are trained in violence and lack the capacity to use other approaches. similar are courses in “anti-corruption” for military or officials who are simultaneously being purchased by foreign donors or pressured and bought by foreign governments and corporations. most of what is really learned, as the author has observed by analyzing the curriculum and measuring the impacts, is how to mouth slogans and falsify international reporting on compliance. the effect is to undermine real public pressures for change by demonstrating the networks that the abusers have now established with international authorities. meanwhile, those who attend project workshops not only benefit from the disguised bribes of free trips and buffets and per diems, but they also walk away with everything from logo blazoned umbrellas (a undp “anti-discrimination” project that the author evaluated in a formerly eastern bloc country, that actually taught officials how to destroy cultural differences and force everyone to the lowest common denominator as a familiar stalinist example of non-discrimination) to “certificates” of course completion that are used as chits for study scholarships and job opportunities. projects teaching elephants to fly like birds international projects not only work with governments but also with “civil society” including businesses and civil society organizations in claimed attempts to build capacity for “market economies” or for “democratic oversight.” a typical failure in both the government and public sector is that they train their “partners” with the very skills that the organizations lack but in areas where the specific individuals will never have the ability to perform those skills. in developed countries, organizations hire people who have the ability to work as managers or fundraisers or advertisers or creative product designers to expand organizations, recognizing that people have different and specific abilities. but in the development community, where it is determined that organizations cannot fire poor performers or hire competence, the idea is that capacity can be built by turning health ministry secretaries into brain surgeons, state lumber company managers into environmentalists, state coal company officials into computer software entrepreneurs, and military officials at national and local levels into legislative drafters for democratic participatory oversight laws. examples of two common types of failures are the following.  ngo capacity building to promote civil society – almost every donor now seeks to support “civil society” projects. what that usually means, in the european journal of government and economics 4(2) 189 author’s experience evaluating projects in this sector, is throwing money at those organizations that have been created as dependent branches of foreign ngos to oppose or perform the functions of governments in recipient countries, rather than to actually rise up as citizen funded representative organizations that act as monitors of government and business to assure their accountability. since these organizations are established as administrative arms of foreigners that have no understanding of how to respond to citizens in their own country by offering them real benefits and asking for their funds, attempts to build their capacity and make them sustainable fall on deaf ears. often what they really teach is the fundraising skills of how to keep appealing to foreigners for money. with little or no management ability, little or no ability to plan, and no understanding of the actual role of civil society, “capacity building” here in skills like management, strategic planning, or even higher skills of statistics and lobbying, is almost always the equivalent of the blind leading the blind.  european commission network of schools of political studies to teach tolerance and rights – a typical capacity building project supported by various donors is on reaching out to young elites in developing countries in a purported attempt to train them in new concepts such as “democracy” as those countries are brought into the orbit of the donor. the european commission, for example, in a project evaluated by this author, spent millions of euro identifying young political elites (usually ruling party members, on career tracks of political patronage) in former soviet and eastern european countries being brought into europe, on the pretext of teaching them “human rights”, “tolerance” and other european “democratic” values. in fact, what these projects appear to actually teach is how to conduct political discussions behind closed doors with political elites, including those of europe, and how to maintain “networks” with each other through meetings in international hotels, rather than through meetings and accountability to their own publics. the moscow branch of this school brought henry kissinger to russia to explain to these identified future elites how he ran an unaccountable u.s. foreign policy under the nixon administration. section iv: how some organizations do in "capacity building", scored using the proposed indicator below are the results of use of the indicator offered in this article on several different categories of international capacity building interventions that are described in international development literature. some of them are from the author’s first hand experiences in more than 30 years in international and community development work. from the categories, it is clear how the scoring works to separate different types of projects in terms of quality as well as to expose hidden agendas and failures and to note where there is a need for accountability and improvement. note that even though not every question applies to every kind of project, the scoring is still designed to yield a scoring spread that leads to categorization and comparison and that also shows how some projects in a category can do better or worse depending on their attention to specific project features that are highlighted in the scoring system. before reading these results, consider the following. most “self-rating” systems using indicators grossly over-inflate results because of the natural tendency to look uncritically at one’s own projects (why there is a need for clear and objective grading standards) and because there is a tendency to avoid considering several organizations at once when rating those organizations one favours. any rating instrument needs to be “calibrated”; i.e., tested for consistency using the same test question multiple times on multiple organizations in order to reveal differences. each observer doing the test ultimately reaches some internal consistency after a lempert ● a quick indicator of “capacity building” initiatives 190 number of tests, but different observers are likely to come up with different results because they are “harder” or “softer.” the scores below are those consistent with the judgment of the author and they are an example of strict application of the ideas, such that weaknesses are revealed as areas where improvement is needed. if such a tool is ultimately adapted by professionals and subject to multiple tests, there would ultimately be a consensus on the scaling and the rating system. the scoring below is not an “absolute” and there is not enough space in this article to present the full detail and evidence to fully explain the scoring for every organization. a sample of the process with full detail for one organization is provided in appendix section v. here, the reader should view the scores as coming from a larger set of data and the professional judgments of the author in following standard social science disciplinary protocols. models of comprehensive capacity building: 8-11 points. the examples that fall into this category are rare. the marshall plan de-nazification of (west) germany after world war ii – though it is a bit of historical guess-work, probably the allied reconstruction of germany after world war ii would earn 9 to 11 points, at the top of the scale. there was little danger of collusion with the nazi officials since they were sentenced for war crimes, and the approach was to rebuild democracy and rule of law at all levels from the schools to the courts to the constitutional and political system. given the understanding of sustainability in 1945, the approach fulfilled it (and tried to begin to rebuild cultures of minorities like jews and roma/gypsies). of course there was self-interest on the part of the u.s. and u.s. corporate and global interests (possibly losing a point on question 10), though the u.s. also understood the long term future interest of a sustainable germany and built a country that competed with the u.s. economically. there are also questions as to whether the u.s. military and government that were implementing this aid were really subject to clear accountability to the u.s. public (though possibly more so then than today!). this is a model intervention, but it also occurred under rare and specific circumstances. for comparison, you can compare the “de-capacitization” that occurred in east germany under the soviets. “street law” civic education projects, “where there is no doctor” training of basic primary health care, grameen bank community self-financed lending, “manage your own business” training for family and household businesses (gtz and other donors), and small scale participatory user-group projects (such as afap’s local contributory irrigation projects – the common dimension of these projects is that they start with a basic need (legal skills, health skills, business and lending skills) and work directly with community groups to build those skills and to change cultures at grassroots levels. in doing so, they can earn about 8 points. these ngo projects are all examples of competence in capacity building in specific fields, building sector and governance skills and institutionalizing them, where major donors fail. the weaknesses that these organizations share is that they do not consider the overall sustainable development impact of their interventions that could be negative (question 1) and they don’t look at the larger systems that have failed to seek ways to get to the real root causes of the problem (question 3). however, they score well in other areas. there are no conflicts of interest in their work but they potentially distort existing service systems (question 18). minimally competent approach to capacity building: 4-7 points. examples here are of interventions that are technically competent but that may be largely flawed from a development perspective as destructive of local cultures. british establishment of public administration schools in colonial africa in 1950 as part of state-building; french support for public health systems (hospitals, “pasteur” and other institutes) in colonial africa and indochina – the “positive” aspects of the european colonial legacy are the institutions that have carried over european journal of government and economics 4(2) 191 in those countries in the post-colonial period and both of these interventions score about 4-5 points. the areas in which they are deficient are clear: the goal was not to promote sustainable development of the natives but to “civilize” them in ways that destroyed their cultures and “modernized” them to follow european approaches. systems were not “fixed” but destroyed. so there are no points on questions 1 or 8, and colonial regimes are not models of accountability (question 10), but these projects score 7 of the 11 positive points. these colonial interventions obviously created and empowered local elites (questions 13 and 16) and there were conflicts of interest on the parts of the colonial implementers (question 15), for a loss of about 3 points. narrow or weak intervention: 0 – 3 points. projects that are technically competent and that offer public training through market solutions can also score points on competency but fail if they are promoting a particular agenda that is not sustainable development and that could distort overall government functions. agricultural extension projects of major donors and small business promotion projects (of donors like the world bank’s international finance corporation/mekong project development facility in southeast asia) -these projects can earn between 2 to 6 points, depending on how carefully they actually promote a “business development services” approach that builds private sector capacity, rather than seeks to replace it. among specific skills promotion projects that work widely with the public on training and outreach and that offer fee based courses or services, these approaches are technically competent in the very basics of capacity building. the reason they only earn about 4 points of the first 7 is that they do not consider overall development (they start with an ideology of productivity) and do not effectively measure the overall function of learning and applying information (about business or agriculture). the organizations running these projects are not accountable and the approaches they promote are sometimes sustainable and long-term but sometimes only short-term inputs. they may lose points for subsidizing elites or for failing to promote truly comprehensive and market based solutions. administration of justice projects, judicial training projects, human rights training for police and judges, projects of the world bank, european commission (ec), usaid, undp, and other donors – these projects, described in vignettes, sometimes show slight benefits in improving efficiency or in ameliorating symptoms, but they lose points by failing to focus on democracy and equity and by entrenching existing elites and their inequities, and at best score 3 points if they save public funds and address inefficiencies in spending and government action, though more often go slightly negative. incompetent projects with hidden agendas that have been corrupted either by the donor agency, stakeholders in a developing country bureaucracy, or both: < 0 points. international organizations working in areas like “justice” and “antidiscrimination” and building “parliamentary” and “government” “capacity,” claim to be doing much more than they really are and the test exposes them, quickly as promoting hidden agendas that undermine democracy rather than promote it, with scores of 0 to as low as minus 7 points (-7). united nations capital development fund (uncdf) and other “local development” projects supported in almost the same model by a large number of donors (e.g., the seila project in cambodia with dfid, world bank, and undp support, local governance projects of gtz, snv, lux-dev, and belgian development assistance in vietnam) – all of these projects work on a common model of transferring funds to local government officials to promote necessary “decentralization,” with claims that they are simultaneously building local government “capacity” and supporting “participation/democracy” by asking some local citizens what gifts they would prefer from the foreign donors in a list of giveaways. a typical score for these projects is strongly negative (-4.5 points), lempert ● a quick indicator of “capacity building” initiatives 192 suggesting that not only do they lack competence but that they are being used to manipulate local governments and destroy local systems for a colonial objective of the donor. the detailed analysis in the appendix (section v) of uncdf demonstrates the conflicts of interest and the lack of professionalism built into the design of these approaches. these projects fail to examine what systems are broken from the perspective of local cultures and are used to promote agendas of globalization/export and industrialization that are favourable to donors and in violation of international treaty agreements. school of the americas (u.s. military) and other military and police capacity building programs in iraq and elsewhere – as expected these projects to build “professionalism” may have good log frames and partly support the role of defence and security (+1.5 points) but they do not look at root causes of instability and work to militarize conflicts, building systems that weaken citizen control over their militaries and elites, militarize conflicts, and are in the interests of the u.s. more than local citizens (-5 points), leading to a score of minus 3.5 points (-3.5). they actually score slightly higher, ironically, than “local development” projects because they are strengthening a system, despite the negative effects of that system. ‘civil society promotion’ projects of the european union (eu) and usaid in the balkans and elsewhere, ashoka foundation civil society grants – the standard “civil society” promotion project of major donors transfers funds and builds skills of a select number of western-style ngos (rather than already existing community organizations of political, tribal, or religious identity) in ways that undermine their sustainability and seek to transform rather than repair local cultural mechanisms; earning minus 5 points (-5). the roles of ngos are defined as providing services that government should provide or lobbying for foreign concerns rather than aggregating local interests with local control and sustainable funding. the projects score no positive points because they do not examine local needs and do not build overall skills or functions, but pick organizations and partners to receive funds. partners in developing countries who receive funds and counterparts who administer the projects introduce biases (loss of 3 points) and the projects actually weaken citizens relative to these foreign-funded organizations, while also distorting civil society and the appropriate role of government and civil society in services (loss of 2 points). instead, these projects should start at the school level, teaching skills, and work with existing organizations that were weakened by colonial rule. standard undp, eu, and world bank capacity building at the national government level with parliaments, planning and investment ministry, and line ministries – while these expensive projects should reflect the highest competence and the state of the art in procedures, the reality is that they are at the bottom of the scale, failing in almost every way, and scoring minus 7 points (-7). the indicator easily exposes them as indicative of an agenda that is designed to buy off foreign governments and to make them unaccountable to their publics in support of an external agenda, rather than to promote capacity in line with international law and basic principles of development. they score no positive points because they do not seek to balance sustainable systems, address existing problems, find what is broken and needs to be fixed, or promote necessary functions (such as “legislation” or “asset protection”). instead of defining and measuring a problem, projects are targeted at specific government entities and justified on the basis of symptoms (“weak governance,” “corruption”) and the need for “governance” projects to “promote development” or “promote the millennium development goals.” there is no measure of output or outcome other than receipt of transfers of money and services that are then defined as “strengthened.” almost all potential conflicts of interest are present (loss of 4 points) and the ability of officials to influence the projects also promotes weakening of citizen powers and oversight and subsidization of country elites (further loss of 3 points). european journal of government and economics 4(2) 193 section v: an example of applying the indicator: the united nations capital development fund scoring of united nations capital development fund (uncdf) on the 20 component questions of the indicator preliminary information for assessment uncdf’s organizational mission (claim) and role of capacity building according to uncdf website ‘the united nations capital development fund (uncdf) offers a unique combination of investment capital, capacity building and technical advisory services to promote microfinance and local development in the least developed countries (ldcs)’ through its programmes, uncdf strives to contribute to the attainment of the millennium development goals (mdgs) and to the implementation of the brussels programme of action for ldcs in a direct, concrete and measurable way. there is growing consensus that democratic governance creates the conditions for sustainable development and poverty reduction. similarly, it is increasingly accepted that achieving the mdgs and eradicating poverty needs to be done at the local level and thus requires the involvement of local authorities. uncdf's local development programmes support national decentralization strategies in the ldcs and seek to improve social services, governance and pro-poor economic infrastructure at the local level by providing technical assistance and investment capital directly to local authorities. uncdf's investment capital is flexible, high-risk and innovative. it is channeled primarily to poor rural areas in the ldcs where poverty reduction, capacity and governance challenges are typically the greatest. uncdf’s “capacity building” activities in practice, according to uncdf website overview of claimed activities: uncdf's local development programmes (ldps) introduce participatory planning and budgeting systems at the local level. these programmes seek to ensure a voice for women and other disadvantaged groups in local public decision-making. the programmes also work within, and support, the national system of central-local government institutional and fiscal relations. description of actual interventions: emphasis on local level institutional development: improving procedures and practices for local level resource mobilization and public expenditure management (including development planning, investment programming, performance budgeting, procurement, implementation, asset management and internal controls) to enhance the efficiency, effectiveness and accountability of local authorities in poverty reduction activities. performance-linked block-grant funding facility: providing local authorities with general purpose development budget support for sustainable, small-scale, local investments in social and economic infrastructure, such as schools, health clinics, rural roads, water and sanitation and natural resources management. this support is linked to agreed measures of local performance and serves as an incentive for local capacity building. local economic development: this approach emphasizes the importance and leadership of local authorities in encouraging and supporting local entrepreneurship and local enterprise creation. overall analysis of uncdf in using the tool of “capacity building” uncdf has no clear mission statement and starts off its mission by defining its tools rather than its ends. it confuses several potential end goals that it offers as political slogans without definitions – sustainable development, the mdgs (that do not promote sustainable development if improperly applied), eradicating poverty, and local development – with means to those ends such as microfinance, governance, and decentralization. rather than offer measurements of results, it justifies actions on the basis of politics and slogan of a “growing consensus” that has no clear link to beneficiaries or the public that funds the organization. the role of “capacity building” along with other tools like investment capital is equally confused, with the role of these tools in helping to achieve the means to an ends or the ends unclear. there seems to be a formulaic approach to building local government infrastructure for delivering grants to increase productivity, rather than supporting democratic accountability or sustainable development, or protecting communities or cultural groups. it looks like the organization is opportunistic rather than goal driven and this waves red flags for the use of “capacity building” as an appropriate development tool. uncdf offers several slogans about how it is “participatory,” builds local tax systems and other forms of “accountability” but none of this is demonstrated in operating procedures or design given that it simply funnels money and training to local officials without assessing the problems or building citizen skills. an organization that defines itself by shifting of money and resources (here simply “investment capital”) and promoting “growth” without clear development ends is almost certain to score miserably on a development index that tests how it seeks to use a tool properly and professionally for development objectives. lempert ● a quick indicator of “capacity building” initiatives 194 analysis question indicator scoring i. proper application of the basic principles and standards of capacity building and donor interventions scoring in the first category demonstrates that uncdf’s approach to capacity building is an afterthought or opportunistic slogan at best, to promote an agenda that is not one of sustainable development and that does not follow the basic competence and requirements of sustainable development. 1.5 points i. a.) project meets 4 of 5 recognized principles of capacity building and of basic professionalism in the initial review of competence, with a checklist of 7 basic attributes of capacity building, uncdf does not achieve more than 1. this is a “capacity building” organization that lacks the basic competence in its own field. this initial scoring raising red flags that uncdf is either opportunistic, completely mismanaged, or hiding another agenda. 1 point. 1. intervention is fit to country and cultural needs for sustainable development uncdf does no independent analysis of minority cultural sustainability even though it purports to work on “decentralization” as a goal, and it does not conduct overall analysis of sustainability. its focus is only on economic production not on achieving the development balance to protect resources or to balance consumption with production. this is business investment, not development that meets international treaty standards. 0 points. 2. governance or civil society functions are appropriate and key and there is direct public accountability several of the functions that uncdf promotes are relevant government functions: capital budgeting and investment policy to protect and promote assets. it is debatable whether it is really doing so in ways that really protect funds and capital investments in communities or not. it is also debatable whether it is promoting these assets on a per capita or community basis. they appear to be promoting the private sector though it is not clear whether they are protecting the public against its power. assume they know their business and award 1 point, though a more stringent scoring would award 0.5 points. 3. diagnosis includes full framework analysis of all levels affecting capacity there is no evidence that uncdf analyzes the problems in each country or locality before moving to apply a formulaic tool and approach. they are starting with the tools before analyzing all of the dimensions of the problem. 0 points. 4. intervention targets root causes and problems, not symptoms, are defined and mapped the country analyses that they have done to date show no evidence of problem trees or root cause analysis and simply list symptoms of weak government management of various sectors. 0 points. 5. logframe targets root causes with benchmarks and cost effectiveness measures uncdf’s logframes bear the hallmark of lack of competence of most projects in the u.n. system in that the “outputs” are simply restatements of the inputs with no benchmarks or cost-benefits and no true measurable outputs in terms of service performance and capacity. 0 points. 6. public accountability mechanisms are part of the assessment and a first priority uncdf claims that it builds democracy and accountability but the reality is that its activities are with government officials. there is no attempt to start teaching basic concepts of finance, investment, and government oversight to citizens as part of basic civic skills, which would truly meet the goals of improving local government performance and accountability. 0 points. 7. the project agreement is transparent and accountable to citizens uncdf actually worsens the problem it claims to be solving with regard to democratization and accountability to citizens. if the goal is accountability, the project agreements need to start with the public and model public oversight, rather than start with government officials who are the symptoms of the problem (unrepresentative, non-transparent, possibly corrupt). 0 points. i. b.) sustainability of impact (5th principle of capacity building) 0 points are to be awarded here because uncdf does not meet the threshold of at least 4 points in the first seven questions. however, even if it did, it still would not receive points for these two questions. 8. intervention fixes a broken system at the heart of the problem uncdf never identifies the system that is broken and the root causes to be fixed, nor does it institutionalize a new kind of system. it simply trains existing officials who may or may not be in place in the future rather than creating institutions to add needed skills into the society. 0 points. 9. intervention institutionalizes a flexible, funded, sustainable system uncdf’s small credit projects may be sustainable if they are properly managed and safeguarded but there is no evidence to suggest that local government financial systems will maintain themselves after the intervention or that the categories of block grants will be maintained after the intervention ends. 0 points. i. c.) internal project procedures model good governance and capacity the organization itself is not open and accountable but it does have at least a partial commitment to the idea of conditionality. 0.5 points. european journal of government and economics 4(2) 195 10. donor agency is itself a model of accountability and capacity uncdf is not directly accountable to the public that funds it or the beneficiary public where it works and its procedures bear the hallmark of an unaccountable bureaucracy largely shrouded in secrecy, typical of the un system and large donors. 0 points. 11. project conditions aid on clear capacity improvements with measurable beneficiary results uncdf claims that it sets performance conditions though performance seems to be based not on capacity and services but rather on economic productivity gains or transfers that occur as a result of assistance. 0.5 points given that the kind of conditionality uncdf offers has debatable impact. ii. professional safeguards in place against conflicts of interest and unintended consequences though claiming to promote both government and the private sector, uncdf seems to promote the worst features of both with no safeguards. -6 points ii. a.) no conflicts of interest uncdf is highly vulnerable to conflicts of interest given the way that it operates to toss funds into localities and to promote business. -2.5 points 12. needs assessment protects against biases of stakeholder recipients it seems that uncdf already has a formulaic approach and is not much influenced by local priorities. however, its commitment to “participation” and to offering funds along with capacity building is a sign that it is vulnerable to using funds to pander to local interests for political reasons. -0.5 points. 13. inputs do not personally benefit stakeholder/partners there is no indication from project documents that uncdf is offering salary supplements to local officials or that recipients of training are going on to better paying jobs. however, there are no clear policies to screen against it. -0.5 points. 14. grant funds are directly linked to capacity improvements and not budget subsidies uncdf offers grant funds in order to fund economic growth and certain policies, even while claiming that it is part of capacity building. this is an example of failure. -1 point 15. donor implementing agents are screened for conflicts of interest there is no direct evidence that uncdf partners or consultants are profiting directly from export businesses and deals with localities where they are sent, but there is an inference within the organization that the goal is to hire business experts to promote business and trade rather than to assure sustainable local business development, and there are no ethics codes or other forms of screening in this area. -0.5 points ii. b.) no negative or adverse impacts on public or private systems uncdf’s projects appear to exist to promote uncdf’s glory in promoting business and productivity rather than in protecting communities or local systems. uncdf lacks the discipline either of a bank or a government agency and seems to introduce the worst characteristics of both. -3.5 points 16. citizen powers are strengthened relative to officials uncdf’s modality of working directly with government officials suggests that it is rooting authority and connections in particular individuals and at particular places rather than in the society in ways that will disempower citizens. -1 point 17. focus is on equity, participation, accountability and not efficiency the focus of uncdf’s interventions seems to be on governmental efficiency rather than on equity or participation, though the small credit and community approach partly counters this. -0.5 points 18. aid does not distort financial systems or subsidize elites uncdf’s choice to award grants rather than to work as a bank and award loans, and its lack of assessment of community savings and tax policies as a prelude to grants suggests that it is subsidizing elites and distorting the very fiscal policies it claims to be building. -1 point 19. public and private functions are kept in an appropriate balance with proper roles it is questionable whether uncdf’s intervention in small credit is interfering with the private sector. there are also questions as to whether uncdf’s probusiness ideology includes any protections for consumers or communities through regulation. -0.5 points 20. market systems are promoted where possible uncdf’s capacity building through local governments is done through outside consultants even though building financial expertise is best left to business schools and government training centers. this suggests a potential distortion of these private and public services. -0.5 points total: uncdf’s score in capacity building is strongly negative, suggesting that it is using “capacity building” simply as a slogan to promote a corporatist and business agenda that is in fact undermining development and governance objectives rather than promoting them. -4.5 points lempert ● a quick indicator of “capacity building” initiatives 196 analysis of uncdf is based not only on its own statements on its website but also from review of some 8 project documents in as many countries (a 20 percent sample of 40 project countries in 2009) including its logframes, problem statements, results measures, inputs and overall goals and project logic at the request of one of uncdf’s regional offices. a copy of this review of tensions in uncdf’s overall mission and overall deficiencies in standards, is on file with the author. neither this review nor any similar critique has been posted on uncdf’s website and provides evidence of the organization’s inability to follow such claimed procedures of transparency and stakeholder participation in its own activities. european journal of government and economics volume 5, number 2 (december 2016) issn: 2254-7088 120 hidden costs of cuts: austerity, civil service management and the motivation of public officials in central and eastern europe after the crisis jan-hinrik meyer-sahling, university of nottingham, united kingdom barbara janta, rand europe, united kingdom iveta reinholde, university of latvia, latvia christian van stolk, rand europe, united kingdom abstract the implementation of austerity measures presents a dilemma for governments. while austerity measures such as cutbacks aim to reduce costs and enhance public sector efficiency, the same measures might undermine the motivation of employees and, consequently, the prospects of effectively implementing austerity programmes. based on a survey of ministerial officials in poland and latvia, this article finds that the scale of cutbacks explains a larger decline of staff motivation in latvia than in poland. the article further shows that motivation was more likely to decrease after the crisis if austerity measures involved cutbacks such as staff reductions, recruitment freezes, and a reduction of training opportunities. keywords economic crisis; austerity; cutbacks; civil service reform; motivation; central and eastern europe. jel classification h50; j30; d73. european journal of government and economics 5(2) 121 introduction the global financial crisis has provoked a new debate over the status and contribution of public administration research in the age of austerity (kelly and dodds 2012; lodge and wegrich 2012; potter 2012). at the macro-level, research has concentrated on the vulnerability of the state in member states of the organisation for economic co-operation and development (oecd) (lodge and hood 2012), the type of responses that governments have selected to tackle the crisis (armingeon 2012), and the political and economic determinants of fiscal consolidation (lodge and rodríguez-vives 2013; kickert et al 2015). with regard to the organisation of public administration, research has explored the impact of the crisis on administrative reform (randma-liiv and kickert 2016), decision-making and coordination in public administration (peters et al 2011; savi/randma-liiv 2015), human resources management practices (parrado 2010; demmke 2016), compensation packages (oecd 2012) and leadership competencies (leslie and canwell 2010; lodge and hood 2012). much less is still known about the position of public officials vis-à-vis the economic crisis and the austerity measures that have since been passed to varying extent in european democracies. this is especially true for the motivation of public officials after the crisis. the motivation of employees is widely recognised as a key determinant of organisational performance in the public and private sector (huselid 1995; hondeghem and perry 2009; brewer and brewer 2011). the motivation of officials and hence their willingness to exert effort is therefore an important condition for the successful implementation of austerity programmes. however, governments that seek to implement austerity programmes face a dilemma, in that public officials are simultaneously agents and objects of reform. on the one hand, austerity measures such as cutbacks, which are the focus of this article, tend to be associated with staff reductions, wage and pension cuts, hiring and promotion freezes, and fewer resources for training and development (see, for instance, parrado 2010; undp 2010; oecd 2012; raudla et al 2015; demmke 2016). austerity measures might therefore involve hidden costs, in that they have an inherently de-motivating effect on public officials. at the same time, governments have to rely on public officials to implement austerity measures that target public administration and public policy programmes. in the worst case, the efficiency gains from austerity measures are offset by motivational losses trapping government in a vicious circle cutting costs and motivation at the same time. this article addresses this dilemma. it builds on the literature on human resource management, motivation and performance in the public and private sector to examine the impact of post-crisis cutbacks in the area of civil service management on the motivation of public officials. the article focuses on the impact of human resource management practices that follow the enactment of cutback measures. it examines five areas of civil service management including post-crisis changes in recruitment, promotion, dismissal, salary management and training policy in order to identify which of these areas have had an effect on the motivation of officials. in addition, the article assesses the extent to which the wider context such as the scale of a country’s austerity programme and the type of ministry as the immediate work context affect the motivation of officials. empirically, the article is based on a survey of ministerial officials in two countries from central and eastern europe: latvia and poland. latvia belongs the group of european countries that was hardest hit by the global financial crisis. latvia had to apply for support from the international monetary fund (imf) and the european commission in order to avoid default. the international loans were followed by a large-scale restructuring of the public sector including major cutbacks in the civil service. poland, by contrast, is among the european union (eu) member states that was least affected by the crisis insofar as growth of gross domestic product (gdp) remained positive during the post-crisis period. consequently, public sector meyer-sahling ● hidden costs of cuts 122 restructuring efforts have been far more moderate. the comparison of poland and latvia therefore provides an opportunity to disentangle the impact of country-level austerity measures and individual-level experience of changing management practices on the motivation of public officials after the global financial crisis hit these two countries. the analysis shows that latvian officials perceive a much larger decline of motivation than polish officials. this difference is likely to reflect the scale of cutback measures in latvia. at the same time, no effect is found within countries for officials who work in finance and economics ministries. with regard to differences between areas of civil service management, the analysis reveals that motivation among ministerial officials was likely to decrease if they experienced hiring freezes, staff cuts and a reduction of training opportunities. by contrast, promotion freezes and salary cuts are not found to have a consistent effect on the motivation of officials. austerity and motivation: what to expect for public officials after the crisis the motivation of employees is an important determinant of organisational performance in the public and private sector (huselid 1995; guest 1997; hondeghem and perry 2009; brewer and brewer 2011). most generally, motivation refers to a personal disposition, a psychological trait to act towards a desired goal. it is typically associated with the willingness to exert effort. in human resource management, motivation is widely regarded as a critical outcome of human resource management practices and one of the key mechanisms to generate firm performance (rainey 2001). more specifically, it ‘consists of an individual’s direction, intensity, and duration of effort. motivation manifests itself in the individual’s choices to exert effort, choices of how much effort to exert, and choices of how long to exert the effort’ (lepak et al 2006: 232).16 research on human resource management has paid particular attention to the impact of human resource management strategies and practices on organizational performance. approaches differ with regard to the scope of management practices that they examine and the type of performance outcome they seek to observe (wright and boswell 2002). broad, inclusive perspectives assess the impact of human resource management systems, that is, a combination of human resource management functions and their relation to performance (huselid 1995). by contrast, narrow approaches focus on individual management practices, for instance, the impact of incentive pay schemes on performance in public and private sector organisations (marsden 2010; dahlstroem and lapuente 2010). second, performance outcomes are either observed at the level of organisations or at the level of individuals. the former focuses on indicators such as the financial performance of firms. by contrast, at the individual level studies examine the job satisfaction, commitment, turnover and personal health and wellbeing. the fourfold distinction of approaches to the study of human resource management and organizational performance provides a helpful foundation for assessing the impact of austerity measures on the motivation of public officials, 16 the difficulty to define motivation or more specifically ‘work motivation’, which focuses on work-related behaviour, is widely recognised (rainey 2001, wright 2001). yet definitions tend to find common ground in their focus on effort exerted by the employee and the reference to intensity, direction and persistence/duration of effort. note also that our understanding of motivation differs from the notion of ‘public service motivation’. the latter is a specific type of motivation defined as ‘an individual predisposition to respond to motives grounded primarily or uniquely in public institutions’ (perry/wise 1990: 368, see hondeghem/perry 2009 for a recent review). however, the insights of this body of literature are relevant for the present article and will be referred to below. european journal of government and economics 5(2) 123 which is the focus of this article. first, among austerity measures we focus on cutback measures in relation to civil service management. cutbacks in the area of civil service management primarily fall into the category of reducing operational expenditure (raudla et al 2015). in this article, we focus on five areas of human resources management, including austerity-driven changes in recruitment, promotion, dismissal, salary management and training (see demmke 2016 for a similar approach). we develop hypotheses for each of these human resource management functions and assess the relation to individual perceptions of motivation. second, the research on work motivation in the public and private sector directs attention at other factors that might affect change in the context of austerity. for this study, we will later control for basic contextual parameters such as the severity of cutback measures in a given country and the position of finance and economics ministries within each country. hiring freezes recruitment policy refers here to the process of hiring new staff. this may involve the expansion of an organisation following the creation of new positions or simply the replacement of staff who have departed and hence the filling of vacancies. in the context of austerity, hiring freezes have been the main change in the area of recruitment policy (parrado 2010; undp 2010; demmke 2016). governments have stopped to expand the civil service and they have stopped to replace staff and hence to fill vacancies. hiring freezes may be associated with a cut of positions, while vacancies are simply maintained for a definite or indefinite time in others. for existing staff, hiring freezes should not directly affect their motivation. after all, these officials do already have a job in the civil service and they do no longer have to compete for a position. however, hiring freezes have an indirect effect in that they affect the workload of officials. if vacancies are not filled, the remaining officials will have to cover and hence do a fair bit of extra work. as a result, hiring freezes have been argued to have negative effects on the job satisfaction of officials (demmke 2016). we should therefore expect that hiring freezes are associated with a decline of motivation among public officials. h1: the motivation of public officials is more likely to decrease if post-crisis recruitment policy has changed including hiring freezes (recruitment hypothesis). staff cuts dismissal policy refers here to the process of firing officials from the civil service. generally, it might include transfers and relocations in the civil service, but dismissals are clearly the most serious of these options in that they involve a termination of the public service relationship. in the context of austerity, dismissal policy plays a very important role for government. mass work force cuts, quotas to reduce the number of staff, early retirement plans, the abolition of positions and terminations in the wake of administrative re-organisations are key instruments for governments to reduce fiscal costs and enhance the efficiency of public administration (lodge and hood 2012; raudla et al 2015). for public officials, a change in dismissal policy might not have a direct effect on their motivation at work. however, it has considerable threat potential and might undermine staff morale, as officials might fear to lose their job in the future. moreover, it may be associated with higher workload comparable to the impact of hiring freezes because the same amount of work will have to be done by fewer people. we should therefore expect a negative effect of staff cuts on the motivation of public officials. meyer-sahling ● hidden costs of cuts 124 h2: the motivation of public officials is more likely to decrease if post-crisis dismissal policy has changed including an increase in workforce cuts (dismissal hypothesis). promotion freezes promotion policy refers to the upward mobility and career advancement of officials within an organisation. it might involve automatic promotions, competitions and simply nominations to higher ranks and positions. promotions are usually associated with higher salary levels and hence tend to have direct effect on the reward levels of officials. in the context of austerity, promotion policy might simply lead to a freeze for upward mobility comparable to hiring freezes. vacant positions at the higher level might also be kept open or they might even be cut, as a result of which promotion prospects narrow for officials in lower ranks. for public officials a change in promotion policy should have a direct effect on their motivation, as it reduces opportunities for career advancement and salary increases. in rationalist terms, there are fewer incentives to work hard because there will be no reward. in addition, it might affect the motivation of officials because opportunities for the recognition of good performance and contributions to the organisation are no longer available. we should therefore expect that postcrisis promotion freezes should have a negative effect on the motivation of officials. h3: the motivation of public officials is more likely to decrease if post-crisis promotion policy has changed including a narrowing of promotion opportunities (promotion hypothesis). salary cuts salary policy is primarily concerned with the monetary compensation of officials. in remuneration terms this includes typically a fixed salary plus discretionary and nondiscretionary salary elements. the former include performance bonuses of various kinds, while the latter refers to rank-specific supplements, age-related supplements etc. in addition, one might consider non-monetary rewards and rewards that are paid after the end of the public service relations, in particular, pensions in this category (brans and peters 2012). in the context of cutbacks, most elements of salary management tend to be affected (demmke 2016). fixed salaries might be frozen or even reduced. similarly, bonuses might be frozen or cut. non-monetary rewards also tend to be on the chopping board during periods of crisis. public sector pensions are among the key areas in which governments aim to make savings, not less because the cuts in reward levels are not immediately felt by public officials. for public officials, a change in salary policy should have a direct effect on their motivation (guthrie 2008). salary cuts, in particular, undermine the material incentives for public officials to work hard. bonus cuts, pension cuts etc should have the same kind of effect. to be sure, public officials might be happy to have a job at all when considering that unemployment might go up in the private sector. yet compared to the pre-crisis period, public officials are still experiencing a lower level of compensation for the same or even a higher level of workload. we therefore expect that austerity-driven changes in salary policy are associated with a decrease in motivation among public officials. h4: the motivation of public officials is more likely to decrease if post-crisis salary policy has changed including salary cuts (salary hypothesis). cutting training opportunities training policy refers here to the development of skills and competencies of public officials in the context of their employment relation. it differs from pre-service european journal of government and economics 5(2) 125 training such as university education. in-service training can take a large number of forms such as short and long courses funded and provided by ministries and agencies, national schools and institute for the training of officials, universities, international organisations and the private sector. it might also include support for university study at home and abroad, exchanges and increasingly online education. cutback measures in the area of training policy most naturally involve the reduction in the provision of training opportunities (metsma 2014). this may range from a reduction of budgetary resources to the closure of public training institutes. moreover, training budgets tend to be soft budgets in the sense that governments find it less difficult to cut funding in this area during a period of crisis (demmke 2016). for public officials, we should expect that a cut in the provision of training opportunities lowers the motivation of officials. training provides an opportunity to develop skills and hence to perform well on the job. participation in training courses also tends to provide an incentive for staff, in particular, when it involves travel abroad or the acquisition of higher-level qualifications such as via study abroad programmes. we therefore expect cutbacks in the area of training policy after the crisis to have a negative effect on the motivation of officials. h5: the motivation of public officials is more likely to decrease if post-crisis training policy has changed including a reduction in training opportunities (training hypothesis). post-crisis civil service reforms in latvia and poland the analysis is based on a survey of ministerial officials that was conducted in 2010/2011 in latvia and poland. latvia and poland are two new member states from post-communist central and eastern europe that joined the eu in 2004. with regard to the impact of the global economic crisis, the two countries differ significantly (bideleux 2011). latvia belongs to the eu members that was hardest hit by the crisis, while poland belongs to the member states that was least affected. in latvia, in 2009 economic output declined by 17.7 per cent, the highest slump in economic growth in the eu. the government deficit rose to 9.8 of gdp in and unemployment went from 6.5 per cent in 2007 to 18.2 per cent in 2009 and 19.8 per cent in 2010.17 in the winter of 2008/2009, latvia asked the imf and the eu for financial support in order to avoid default following the nationalisation of the latvian bank parex. the international loans were linked to a clear commitment from the latvian government to reform its public finance by means of fiscal measures, social welfare reform, economic restructuring and specific measures for the reform of the financial sector. the programme for economic stabilisation and growth included specific measures to reform public administration by means of reorganisation and expenditure cuts (government of latvia 2008). in the context of the so-called optimisation plan for the reform of public administration, the government passed a range of civil service reform measures most of which aimed to cut operational expenditures. they included:  staff cuts by over 10 per cent (25.000 out of 205.000 officials were to be cut) primarily in the context of the re-organisation of public administration (leta 2009). 17 for growth and unemployment figures, please see the official eurostat figures available at http://epp.eurostat.ec.europa.eu/portal/page/portal/eurostat/home. meyer-sahling ● hidden costs of cuts 126  the government did not take an explicit decision on recruitment. however, due to need to downsize staff, it became impossible for ministries and agencies to hire new staff.  promotion freezes and formal downgrades in the context of the structural reorganisation of ministries and agencies in many ministries, while other ministries chose to dismiss officials. for heads of units, for instance, this meant that over a short period of time they became so-called senior officers and, as a result, lost their managerial attributes. promotions were not formally frozen but in practice they took no longer place.  salary cuts by 15 per cent in january 2009 and again by around 20 per cent for top and middle staff in june 2009. managers were given discretion to decide whether they would prefer to cut salaries at the top or mid-level. this led to considerable differences in the way salary cuts were experienced by civil servants (memorandum 2009; darzina 2009).  the central government cut the funding for training institutions and training activities. the latvian school of public administration lost 99 per cent of its funding (lspa 2011). training had to be paid for by ministries, agencies or personally by civil servants, leading to a major decline in training activities. in contrast to latvia, poland experienced a softer landing after the global financial crisis hit europe (bideleux 2011). economic growth remained positive for the entire post-crisis period reaching a low point of 1.6 per cent of gdp in 2009 but accumulating 17 per cent during the period from 2008 to 2012; the highest growth rate among eu member states. in 2008 and 2009, the unemployment rate varied between 7 and 9 per cent and until the end of 2012 it never rose above 10.5 per cent. despite these positive figures, poland experienced a high government deficit of 7.4 per cent of gdp in 2009 and public debt has continuously risen since the beginning of the global economic crisis towards its constitutional limit (see rae 2013). given the far more moderate impact of the global economic crisis, the government of poland did not embark on any major public administration reform. the civil service law was amended in november 2008 but the change was not related to the looming crisis in the region. yet several measures were still taken to address the changing context of public administration.  the area of recruitment was affected in two ways. on the one hand, bida (2009) argues that the number of vacancies that was publicly advertised increased. moreover, he argues that both the number and quality of applicants per post increased compared to the pre-crisis period. on the other hand, the government reduced the quota of civil servants who were given the status of ‘nominated civil servant’, poland differs from other countries in central and eastern europe in that it distinguishes ‘nominated civil servants’ and ‘civil service employees’, the former have to undergo a lengthy recruitment and selection procedure, a preparatory service, they enjoy full tenure and significantly higher salary levels. currently, just under 5 per cent of all civil servants (6.000 out of 122.000) are ‘nominated civil servants’, initially the government aimed to raise this proportion to 10 per cent of all civil servants. however, in the context of austerity the government admitted a declining proportion of candidates who had successfully passed the nomination exam and even reduced the quota for new nominations to merely 200 in 2013 and 2014 (gazeta prawna 2011). european journal of government and economics 5(2) 127  the government did not establish specific targets to reduce the staff numbers in central government ministries and agencies. measures were taken to facilitate dismissals by means of disciplinary proceedings that were brought in line with the labour code but no efforts were made to downsize the civil service.  the government did not pass specific measures to reduce promotion prospects for civil servants. however, following lengthy discussion the government decided that nominated civil servants would not be granted an automatic promotion if they pass two performance reviews successfully. this measure can be considered as a way of narrowing the promotion prospects for higher and senior civil servants (gazeta prawna 2013a).  in the area of salary management, the government reviewed the salary bands shortly after the economic crisis emerged (zieliński 2009). even if this measure required additional government funds for some ministries, it was also decided that civil service salaries would be frozen from 2008. since then they have increased but below the level of inflation, implying real wage decline for civil servants (gazeta prawna 2013b).  with regard to training policy the government reviewed the process of selecting training providers. at the same time, training provision for civil servants was maintained at similar levels compared to the pre-crisis period. the comparison of austerity-driven civil service reform strategies in the two countries suggests that human resource management practices were likely to change to a much larger extent in latvia than in poland. yet in both cases, reforms are associated with a ‘negative’ development in that staff levels were cut, fewer officials were hired, promotion prospects narrowed, salaries were frozen or even cut and the provision of training was reduced. we would expect these developments to be reflected in the survey results and we would certainly expect that the motivation of officials in latvia decreased to a larger extent than in poland. data, operationalization and descriptive results the empirical analysis is based on a survey of public officials that was conducted in the winter of 2010/2011 in latvia and in the spring of 2011 in poland (see also meyer-sahling and mikkelsen 2016). it targeted civil servants employed at central government ministries. the survey was conducted in local languages as an online survey. it generated 1520 responses, including 972 from poland and 547 from latvia. the reliance on a web-based survey implies that we had no perfect control over the population of respondents, that is, we do not know with certainty who precisely was invited to participate in the online survey and whether officials who were invited also had a chance to complete the survey. in order to maximise the consistency of the data across the two countries, we applied the same sampling procedure. in a first step, personal meetings were arranged with senior officials from the civil service department in poland and the state chancellery in latvia in order to explain the nature and purpose of the survey. next, we sent the survey link together with a detailed explanation to our partners at the central institutions who then distributed the link to the heads of personnel in the ministries. in latvia, this role is performed by the state secretary. in poland, the director general is responsible for personnel affairs. heads of personnel then sent the survey link to all civil servants employed in the core structure of their ministry. as to the representativeness of the sample, we were able to check it in poland against the population of officials in civil service ranks. it suggests that our sample meyer-sahling ● hidden costs of cuts 128 is reasonably representative. men are slightly over-represented in the sample (42 per cent as opposed to 35 per cent in the polish civil service) and – as one might expect when conducting a web-based survey – young civil servants are overrepresented (29 per cent as opposed to 23 per cent in the polish civil service). we assume a similar level of representativeness for the latvian sample but have not been able to secure population-based data that would have allowed for a valid comparison against our sample data. for the analysis below we address potential biases by controlling for demographic variables such as sex, age and experience in public administration. the survey primarily asked questions with regard to the implementation of civil service policy and the attitudes of officials towards principles of civil service management. a subset of questions directly addressed the experience of respondents with post-crisis civil service management. answers were measured on the likert scale, ranging from 1 (strongly disagree) to 5 (strongly agree). respondents were given the option ‘don’t know’ if they could not or did not want to answer a question. for the analysis ‘don’t know’ responses were re-coded as ‘neutral’ responses.18 ideally, we would be able to measure the motivation of officials by means of panel data that compares the preand post-crisis situation in the civil service in the two countries. unfortunately, we do not have directly comparable data for such two points in time. in order to address this limitation, the survey included several questions that directly addressed the experience with civil service management since the onset of the global economic crisis. for instance, we asked ‘the practice of promoting staff in my ministry has not changed since the beginning of the global financial crisis and its repercussions for my country’, it should be noted that we asked for ‘non-change’ therefore assuming that civil service management practices might have persisted since the pre-crisis period. for the analysis we reversed the coding of the variable for ease of interpretation. throughout the analysis in this article, a higher value therefore indicates more change in the respective area of human resource management. second, asking for change does of course not tell us ‘what kind of change’ nor ‘what direction of change’, we addressed this issue by asking a follow-up question in which respondents could specify what, according to them, had changed, for instance, in the practice of promoting staff. a large proportion of respondents completed the open questions and has provided rich insight into the perception of officials vis-à-vis post-crisis civil service reforms. most answers were short but they would usually be unambiguous specifying ‘salary cuts’, ‘bonus cuts’, ‘no recruitment of new staff’, ‘no more promotions’, ‘no more trainings’, or in many cases simply ‘no change’, the open answers have allowed us to interpret the likert scale questions and to assume relatively confidently that ‘change’ is by and large synonymous with ‘negative change’, ‘decline’ or ‘cuts’, for each of the five areas of civil service management, we have been able to identify patterns that allow us to associate change with ‘hiring freezes’ in the case of recruitment policy, ‘a growing number of dismissals’ for dismissal policy, ‘promotion freezes’ for promotion policy, ‘salary cuts’ for salary management and a ‘reduction of training opportunities’ for training policy. these patterns apply to both countries. they simply differ in the extent to which respondents report change of this kind (see below, table 1). the survey also included one question that asked for the change in motivation of officials. the motivation question explicitly asked ‘officials in my ministry are as 18 the analysis below was also conducted without including ‘don’t know’ responses. substantively, the results are the same. it should be noted that the survey contains selective non-response, leaving up to 15 per cent of some of the variables missing. european journal of government and economics 5(2) 129 motivated to do their job as they were before the global financial crisis and its repercussions for my country.’ asking with regard to the departmental level rather than the individual level has the advantage that answers might be subject to less social desirability bias, as respondents might not have wanted to indicate openly that their own motivation level has declined since the onset of the crisis.19 for the question on the motivation of officials we did not ask an open follow-up question in the survey. we can therefore not rely on a second, open-ended survey response to ascertain that a change in motivation among ministerial officials is equivalent to a decrease in motivation. however, before and after conducting the survey we were able to conduct personal conversations with managing and nonmanaging civil servants in the two countries.20 they indicated that a negative connotation is plausible in that officials were – to varying degrees – disappointed, frustrated and over-worked after the austerity measures hit the civil service.21 we therefore assume for the empirical analysis that change in motivation is equal to a decline in motivation after the crisis. table 1 lists the means and standard deviations for the dependent variable and the main independent variables. as mentioned earlier, a higher score indicates a higher degree of change, while a score of 1 indicates no change at all. for motivational change, the two-country mean is close to a neutral response of 3. the higher mean for latvia suggests that the austerity measures had a much greater impact on the motivation of public officials in latvia than in poland. table 1. descriptive statistics poland latvia total mean std.dev. mean std.dev. mean std.dev. motivation change 2.79 0.91 3.28 0.98 2.96 0.97 recruitment change 2.79 0.86 2.69 0.85 2.75 0.86 promotions change 2.71 0.77 2.74 0.83 2.72 0.80 dismissals change 2.71 0.75 2.92 0.78 2.79 0.76 salaries change 2.65 1.05 3.31 0.99 2.88 1.08 training change 2.76 0.82 3.54 0.97 3.04 0.96 table 1 also shows important differences with regard to the independent variables, in particular, the experience with post-crisis civil service management. the values for recruitment and promotion policy, for instance, do not differ much, and in both countries, they indicate similar developments. by contrast, in the areas of dismissals and, in particular, salary and training policy the differences between latvia and poland are larger. in these cases, the open answers echo the austerity measures outlined above. a large proportion of latvian respondents simply answered, ‘salary cut by 30 per cent’, ‘salary cut by 40 per cent’, ‘no more 19 while the survey question provides a general indicator of the degree of motivation among ministerial staff and a first attempt to address the problem of post-crisis change in motivation, we are acutely aware that motivation is ideally measured by more than one item (beulens/van den broeck 2007, wright 2001). 20 we did not conduct systematic interviews. however, the regular conversations with staff in the context of the preparation and implementation of the survey have given us valuable insights for the interpretation of the data. 21 it should be recognised of course that for many there was also a sense of relief in that they could hold on to their job while unemployment rates rose. meyer-sahling ● hidden costs of cuts 130 bonuses’, ‘no more training’ etc. by contrast, for polish respondent, a much larger proportion suggested ‘no change’ or merely ‘no salary increase’, ‘salaries have been frozen’ etc. in other words, the cutback measures are clearly reflected respondents’ statements. one of the major merits of our survey is that it captures how public officials appear to experience cutbacks in different ways. indeed, it is likely that cutbacks in the area of civil service management are not uniformly implemented across public administration, as it is more than common that certain institutions and jobs are sheltered while others are earmarked for change (see parrado 2010). moreover, even if a much larger proportion of latvian officials evidently ‘complained’ about salary cuts, the open survey responses indicated that there were considerable differences in the depth of the cuts, ranging from 5 – 50 per cent. this clearly reflects the government’s post-crisis policy, whereby senior managers were given discretion to allocate the salary cuts of the second stage of reforms in the summer of 2009. for the assessment of our hypotheses, we include several control variables. first, we include basic demographic variables such as sex, age, experience in public administration and education in order to control for potential biases in our sample and to take into account the literature on work motivation in the public and private sector (for example, buelens and van den broeck 2007). for sex we include a dummy variable for ‘female’, both age and experience are measured in bands and dummy variables are included accordingly. with regard to the level of education it is worth mentioning that the survey distinguishes degrees that have been completed and degrees that are still in progress. for the analysis, we code finished and unfinished degrees in the same category (e.g. ‘finished and unfinished phd degree’, which acts as the reference category below), as we consider participation in an ma programme to be closer in educational experience to the completion of an ma than to a completed ba programme. further, the rank of respondents is included, as managers with more responsibilities and control within an organisation are likely to be more motivated. they might also be more optimistic in their assessment of the motivational levels of their staff (rainey 2001; buelens and van den broeck 2007). we therefore include a dummy variable for ‘manager’ in our models. finally, we control for the country of origin and the ministry in which the official works. as outlined above, we expect that motivational change is greater in latvia than in poland due to the severity of the crisis and the scale of the subsequent austerity measures. finally, we include a dummy variable for officials who work in finance and economics ministries. we expect that they remain more motivated after the crisis because they might be less affected by the cutbacks. many of the austerity measures – outlined above – target social policy, while officials in finance and economic affairs are more likely to be sheltered from negative effects. moreover, their status may even be elevated in that they might see themselves as the ones who solve the economic crisis. results and discussion table 2 presents the results of the analysis for the explanation of motivation among public officials. the table consists of four models. model 1 shows the impact of five human resource management practices on motivation without controlling for country and ministry context and without taking into account demographic confounders. model 2 adds the country and the position of finance and economics ministries in order to take into account broad features of the work context. model 3 presents the main results of the linear regression analysis. it shows the five main variables together with the contextual and the demographic variables. model 4 is primarily meant as a robustness check. it presents the results of a european journal of government and economics 5(2) 131 logistic regression analysis showing the odds that a change in motivation occurs. the presentation of the logistic regression analysis is limited to the model that includes the main variables and the control variables. looking at the results, model 3 shows that changes in recruitment, dismissal and training policy decrease motivation among public officials. the three variables are significant at the 1 per cent level and confirm the hypotheses that hiring freezes, staff cuts and a reduction of training opportunities reduce the motivation of ministerial staff. in relative terms, changes in the provision of training have the largest effect (b = 0.16, t = 4.48, p < 0.001). the magnitude of the effect of hiring freezes (b = 0.11, t = 2.90, p < 0.004) and staff cuts (b = 0.11, t = 2.41, p < 0.016) is slightly smaller. table 2. change of motivation since the beginning of the global economic crisis model 1 model 2 model 3 model 4 b b b exp(b) (constant) 1.604*** 1.626*** 0.506 0.434 recruitment policy change 0.094*** 0.107*** 0.112*** 1.254** promotion policy change 0.001 -0.012 -0.014 1.075 dismissal policy change 0.069** 0.11** 0.105*** 1.396*** salary policy change 0.094*** 0.047 0.045 1.117 training policy change 0.206*** 0.158*** 0.155*** 1.251*** latvia 0.352*** 0.301*** 1.49** finance ministries 0.061 0.042 0.958 manager 0.022 0.993 female -0.035 1.129 age (ref. cat. ‘under 30 years’) 30 to 40 years -0.142* 0.841 40 to 50 years -0.094 0.922 50 years and older -0.345*** -0.643* experience in public admin (ref.cat. ‘< 1 year’) 1 to 4 years -0.069 0.207*** 5 to 10 years 0.071 0.262*** 11 and more years 0.048 0.225*** education (ref.cat. ‘phd degree’) ma degree 0.125 1.349 ba degree 0.123 1.405 no higher education 0.184 1.781 r square 0.099 0.125 0.133 nagelkerke r square 0.139 note: *p<.1; **p<.05; ***p<.01 by far the largest significant coefficient belongs to the country dummy variable and hence the differences between latvia and poland (b = 0.30, t = 4.06, p < 0.001). it suggests that even if the five human resource management variables are likely to absorb a great deal of the differences in officials’ experience with cutback measures, a range of wider country-level factors must be taken into account when seeking to understand differential change in motivation among officials. by contrast, employment in finance and economics ministries does not have an effect on motivation. country differences and differences in the way cutbacks are meyer-sahling ● hidden costs of cuts 132 experienced appears to be more relevant for motivational change than the ministerial work context within countries. model 4, which presents the results of the logistic regression analysis, confirms the significant effect of changes in recruitment, dismissals and training policy on motivation. the odds of observing a change in motivation are 39.6 per cent higher for officials who have experienced major staff cuts in their department. for officials who have experienced hiring freezes and a cut in training provision the odds of perceiving a change in motivation are slightly lower at 25 per cent. model 4 also confirms the major effect of the country context and hence the scale of the cutback measures. for latvian officials the odds of having experienced motivational change in their ministry are 49 per cent higher than for polish officials. model 4 indicates that officials from finance and economics ministry do not differ in their perception of motivation after the implementation of the cutbacks. table 2 further suggests that salary cuts and promotion freezes are not associated with a change in motivation. the effect of changes in salary policy shrinks and loses significance when the country and ministerial contexts are included in the analysis (compare models 1 and 2). promotion freezes, by contrast, do not have any significant effect on motivational change among public officials. the structure of the civil service systems in poland and latvia is likely to be responsible for the irrelevance of promotion freezes. both countries operate a position-based system that does not provide much room for promotions (meyer-sahling 2009; demmke and moilanen 2010). career advancement tends to require participation in several rounds of competition procedures rather than any prospect of automatic promotion. the relative importance of changes in recruitment policy, which refers to all levels in the hierarchy rather than just the entry level, captures this aspect of the civil service systems in the two countries very well. the secondary role of salary policy might be more surprising, as conventional wisdom would initially point to the detrimental effect of salary cuts for the motivation of employees. however, research that compares the motivational basis of private and public sector employees tends to argue that public sector employees are less motivated by monetary rewards (boyne 2002; beulens and van den broeck 2007). they tend to value job security, a supportive work environment and a better work-life balance more than their private sector peers (chen and hsieh 2015). to be sure, the lower importance of mechanisms of extrinsic motivation such as salaries and promotions also resonates closely with the research on ‘public service motivation’ which stress that public officials are motivated by working for a ‘good cause’ and not just for monetary reward (for a recent review, see perry et al 2010). the importance of training, dismissal and recruitment policy relative to salary and promotion policy resonates closely with this line of argumentation. training budgets tend to be one of the first items that are cut when savings have to be made (demmke 2016). yet the analysis supports the view that public officials value the opportunity to develop their skills and competences to better perform their job. both recruitment freezes and staff cuts also relate closely to the quality of the work environment and hence the working conditions. they inevitably increase the workload of officials and, in addition, one should not underestimate the negative impact on staff morale in the departments following staff cuts and re-organisations. model 3 and 4 also show the role of demographic factors on motivation among officials in latvia and poland. model 3 and 4 suggest that older officials are less likely to perceive motivational change within their institution. moreover, according to model 4 officials with more years of experience are less likely to perceive a change of motivation. the results suggest that the older generation has responded less negatively to cutbacks in the civil service. older as well as more experienced officials have fewer job opportunities outside the administration and they are more likely to seek (to continue) a long-term career in the civil service given their european journal of government and economics 5(2) 133 experience so far. they might therefore be more willing to bear the consequences of the cutbacks and they appear to be more likely to accept the need to get on with the job even under adverse conditions. conclusion this article has examined the impact of cutbacks in the area of civil service management on the motivation of public officials in latvia and poland. based on a survey of ministerial officials, it has shown that latvian officials are more likely to perceive changes in motivation than their peers from poland. the differences clearly reflect the scale of the austerity-driven cutbacks in latvia in comparison to poland. with regard to different areas of civil service management, the article has shown that changes in recruitment, dismissals and training policy undermine the motivation of staff in central government ministries. in the context of cutback management, these changes tend to refer to hiring freezes, staff cuts and a reduction in the provision of training. by contrast, salary cuts and promotion freezes did not negatively affect the motivation of public officials. the results suggest that public officials in latvia and poland are relatively less motivated by monetary rewards and prospects of career advancement. this finding resonates with research on employee motivation in the private and public sector. extrinsic motivation, which focuses on material incentives, tends to be less relevant for public sector employees (boyne 2002; buelens and van den broeck 2007, chen and hsieh 2015). by contrast, recruitment freezes, staff cuts and cuts in training provision are more closely related to workload considerations, the quality of the working conditions and interest in self-development. these factors tend to be valued highly by public sector employees. the findings of this article suggest that these differences also apply in the context of austerity programmes. the article relied on the analysis of post-crisis survey data rather than a comparison of data from before and after the crisis hit the two central and eastern european countries. inevitably there is some uncertainty regarding the findings. moreover, for the purpose the analysis we had to rely on comparably simple measures of motivation and motivational change. however, the differences between the two countries and the perception of austerity-driven civil service reforms as reported in quantitative and qualitative answers are reasonably plausible and provide a first glimpse at the trade-offs that governments face when seeking to implement austerity-oriented reforms. cutbacks in the area of civil service management have been and are likely to remain key instruments for governments to reduce costs and increase efficiency in public administration during times of crisis. this article has shown that the implementation of these kinds of measures entails additional, hidden costs for government, in that they appear to undermine the motivation of public officials. however, the costs of reform depend on the scale of the cutbacks and they are not uniform across areas of civil service management, giving governments some discretion over the choice of instruments. it remains to be seen to what extent the results for latvia and poland hold up to the inclusion of more countries and a longer-term comparison over time. references armingeon, klaus (2012) ‘the politics of fiscal responses to the crisis of 2008– 2009’, governance, 25(4): 543–565. bida, a, (2009) ‘problemy motywacji zatrudnienia w służbie cywilnej (problems with civil servants job motivation)’ in w mikułowski and a jezierska (eds) wyzwania rozwoju zasobów ludzkich administracji publicznej w dobie kryzysu ekonomicznego meyer-sahling ● hidden costs of cuts 134 (challenges of public administration human resources development at the time of economic crisis, conference proceedings, warszawa (warsaw), 10-11 december 2009, pp. 109 – 113. bideleux, robert (2011) ‘contrasting responses to the international economic crisis of 2008–10 in the 11 cis countries and in the 10 post-communist eu member countries’, journal of communist studies and transition politics 27(3-4): 338-363. boyne, george a. (2002) ‘public and private management: what’s the difference?’, journal of management studie, 39(1): 97–122. brans, marleen and b. guy peters (eds) (2012) rewards for high public office in europe and north america. london: routledge. brewer, gene a. and gene a. jr brewer (2011) ‘parsing public/private differences in work motivation and performance: an experimental study’, journal of public administration research and theor, 21(2): 347–362. buelens, mark and herman van den broeck (2007) ‘an analysis of differences in work motivation between public and private sector organizations’, public administration review, 67(1): 65–74. chen, chung-an and chih-wei hsieh (2015) ‘does pursuing external incentives compromise public service motivation? comparing the effects of job security and high pay’, public management review, 17(8):1190-1213. dahlström, carl and victor lapuente (2010) ‘explaining cross-country differences in performance-related pay in the public sector’, journal of public administration research and theory, 20, pp. 577600. darziņa, lidija (2009) ‘nosaka robežsummas atalgojuma samazināšanai valsts pārvaldē (ceiling for salary cuts in public administration has been defined)’, latvijas vēstnesis, 01.07.2009. available at http://www.lvportals.lv/print.php?id=194174. demmke, christoph (2016) ‘effects of budgetary constraints on hr reforms in central public administrations and the importance of institutional context’, international journal of public administration, early online 7 nov 2016. demmke, christoph and timo moilanen (2010) civil services in the eu of 27: reform outcomes and the future of the civil service, frankfurt: peter lang. gazeta prawna (2011) służba cywilna: niewielu kandydatów do awansu w urzędach (civil service: not many candidates to be nominated), gazeta prawna online version. available at http://praca.gazetaprawna.pl/artykuly/507642,sluzba_cywilna_niewielu_kandydato w_do_awansu_w_urzedach.html. gazeta prawna (2013a) ‘podwyżek dla urzędników nie ma. i nie będzie (no salary increases for public officials)’, gazeta prawna online version. available at http://serwisy.gazetaprawna.pl/praca-ikariera/artykuly/712547,podwyzek_dla_urzednikow_nie_ma_i_nie_bedzie.html. gazeta prawna (2013b) ‘rząd utrudnia awans urzędnikom. kto może liczyć na wyższą pensję i stabilną pracę? (government makes promotions of public officials more difficult. who can expect higher salary and job stability?)’, gazeta prawna online version. available at http://serwisy.gazetaprawna.pl/praca-ikariera/artykuly/708324,rzad_utrudnia_awans_urzednikom_kto_moze_liczyc_na_w yzsza_pensje_i_stabilna_prace.html. government of latvia (2008) par ministru kabineta 2008.gada 9.decembra sēdes protokollēmumu (prot. nr.89 1.§) "likumprojekts "grozījumi likumā "par valsts budžetu 2009.gadam"", un latvijas ekonomiskās stabilizācijas un izaugsmes atjaunošanas programma" (on the decision of the cabinet of ministers made during the meeting on 9 december 2008 – “amendment to the state budget law for european journal of government and economics 5(2) 135 2009” and “programme of economic stabilisation and growth”). available at http://www.mk.gov.lv/lv/mk/tap/?pid=40102497&mode=mk&date=2008-12-09 guest, david e. (1997) ‘human resource management and performance: a review and 
research agenda’, international journal of human resource management 8(3): 263276. guthrie, james p (2008) ‘remuneration: pay effects at work, in peter boxall’, john purcell and patrick wright (eds), the oxford handbook of human resource management, oxford: oxford university press, pp.344-363 hondeghem, annie and james l. perry (2009) ‘egpa symposium on public service motivation and performance: introduction’, international review of administrative sciences 75(1), pp. 5–9. huselid, mark a. (1995) ‘the impact of human resource management practices on turnover, productivity, and corporate financial performance’, academy of management journal 38(3): 635−672. kelly, josie and anneliese dodds (2012) ‘public administration in an age of austerity: the future of the discipline’, public policy and administration 27(3):199211. kickert, walter, and tiina randma-liiv (2016) ‘the politics of cutback management in thirteen european countries: statistical evidence on causes and effects’, public management review, 19(2):175-193. kickert, walter, tiina randma-liiv and riin savi (2015) ‘politics of fiscal consolidation in europe: a comparative analysis’, international review of administrative sciences 81(3): 562–584. lepak, david p. hui liao, yunhyun chung and erika e. harden (2006) ‘a conceptual review of human resource management systems in strategic human resource management research’, research in personnel and human resources management 25, 217–271. leslie, keith and adam canwell (2010) ‘leadership at all levels: leading public sector organisations in an age of austerity’, european management journal, 28(4): 297– 305. leta, (2009) ‘valdība apņemas ierēdņu skaitu samazināt līdz 8% no iedzīvotāju skaita (the government decided to cut employees in public administration by 8% of total number)’, september 15, 2009. available at http://www.tvnet.lv/zinas/latvija/195494valdiba_apnemas_ierednu_skaitu_samazinat_lidz_8_no_iedzivotaju_skaita. lodge, david and marta rodríguez-vives (2013) ‘how long can austerity persist? the factors that sustain fiscal consolidations’, european journal of government and economics 2(1): 5-24. lodge, martin and christopher hood (2012) ‘into an age of multiple austerities? public management and public service bargains across oecd countries’, governance 25(1):79-101. lodge, martin and kai wegrich (eds) (2012) executive politics in times of crisis, basingstoke: palgrave. marsden, david (2010) ‘the paradox of performance-related pay systems: why do we keep adopting them in the face of evidence that they fail to motivate?’, in christopher hood, helen margetts and perri (eds) paradoxes of modernization: unintended consequences of public policy reforms, oxford: oup, pp. 185-202. memorandum (2009) memorandum of understanding between the european community and the republic of latvia, 29.01.2009. available at www.ec.europa.eu/latvija/documents/pievienotie_faili/29.01.09.mou.doc meyer-sahling ● hidden costs of cuts 136 metsma, merilin (2014) ‘the impact of cutback management on civil service training: the case of estonia’, administrative culture 15(1):58-79. meyer-sahling, jan-hinrik (2009) the sustainability of civil service reform in central and eastern europe five years after accession. sigma paper nr 44. paris: oecd publications. meyer-sahling, jan-hinrik and kim sass mikkelsen (2016) ‘civil service laws, merit, politicization, and corruption: the perspective of public officials from five east european countries’, public administration, 94(4):1105-1123. ministry of finance (2012) finanšu ministrija. dati par ekonomikas izaugsmi 2011.gadā turpina iepriecināt (the ministry of finance of the republic of latvia. data on economic growth in 2011 continues to please). available at 2012 http://www.fm.gov.lv/?lat/aktualitates/jaunumi/50908. oecd (2012) public sector compensation in times of austerity, paris:oecd publications. parrado, salvador (2010) sustainability of human resources in central public administration in the eu member states, brussels: european public administration network. perry, james l. and lois r. wise (1990) ‘the motivational bases of public service, public administration review 50(3):367–373. perry, james l., annie hondeghem and lois r. wise (2010) ‘revisiting the motivational basis of public service: twenty years of research and an agenda for the future’, public administration review 70(5): 681-690. peters, b. guy, jon pierre and tiina randma-liiv (2011) global financial crisis, public administration and governance: do new problems require new solutions?, public organization review 11(1):13-27. potter, michael (2012) ‘still learning to speak and still not being heard: public administration and the latest financial crisis’, administration & society 44(3):367373. rae, g, (2013) ‘avoiding the economic crisis: pragmatic liberalism and divisions over economic policy in poland’, europe-asia studies 64(3):411-425. rainey, hal g. (2001) ‘work motivation’, in r.t. golembiewski (ed.) handbook of organisational behaviour, 2nd edition, , new york: marcel dekker, pp. 19 – 42. raudla, ringa, riin savi and tiina randma-liiv (2015) ‘cutback management literature in the 1970s and 1980s: taking stock, international review of administrative sciences 81(3): 433 – 456. savi, riin and tiina randma-liiv (2015) ‘decision-making in time of crisis: cutback management in estonia’ international review of administrative sciences 81(3): 479-497. undp (2010) challenges for public sector human resources management in the recent economic downturn, bratislava: regional centre for public administration reform of undp. wright, bradley e. (2001) ‘public-sector work motivation: a review of the current literature and a revised conceptual model’, journal of public administration research and theory 11(4):559-586. wright, patrick m, and wendy r. boswell (2002) ‘desegregating hrm: a review and synthesis of micro and macro human resource management research’, journal of management 28(3):247–276. zieliński, wojciech (2009) ‘modernizacja zarządzania zasobami ludzkimi w administracji publicznej (modernisation of human resources management in public european journal of government and economics 5(2) 137 administration)’, in w. mikułowski and a. jezierska (eds) wyzwania rozwoju zasobów ludzkich administracji publicznej w dobie kryzysu ekonomicznego (challenges of public administration human resources development at the time of economic crisis), conference proceedings, warszawa (warsaw), 10-11 december 2009, pp. 99-106. european journal of government and economics volume 4, number 1 (june 2015) issn: 2254-7088 a sharpe-ratio-based measure for currencies javier prado-dominguez, universidade da coruña, spain carlos fernández-herráiz, caia, spain abstract the sharpe ratio offers an excellent summary of the excess return required per unit of risk invested. this work presents an adaptation of the ex-ante sharpe ratio for currencies where we consider a random walk approach for the currency behavior and implied volatility as a proxy for market expectations of future realized volatility. the outcome of the proposed measure seems to gauge some information on the expected required return attached to the “peso problem”. keywords sharpe ratio; peso problem; carry trade; currency strategies. jel codes g11; g13; g15; g17. 67 pere ● governance and economic development of western balkan countries introduction the reward to variability ratio has been around for nearly 40 years since the seminal work by sharpe (1966). since his paper, the so called sharpe ratio spawned various performance measures, including several adaptations of the ratio, which offer different flavors of the risk reward relationship. the success of the ratio is rooted in the combination of two opposing but complementary concepts: risk and return. the ratio is always stated as the excess return of a portfolio or strategy per unit of risk handled. it is a simple but effective measure of how well rewarded are the risks taken if some assumptions about markets and investors are made. the ratio could help selecting among strategies or portfolios. when an investor makes use of the sharpe ratio, she must pay attention to the various assumptions involved in the measure. the ratio implies that mean and standard deviation for one period are sufficient statistics for evaluating the attractiveness of a portfolio. but comparisons based on the first two moments of the distribution do not take into account possible differences among portfolios in other moments. in fact, goetzmann et al (2002), offer a “sharp” analysis of the issue, proving that the distribution of returns that maximizes the ratio is the one with significant third and fourth moments. those moments imply fat left tails (big unexpected losses more frequent than expected if the returns followed a normal distribution), and also a higher concentration of returns around the expected return of the distribution (offering a false sensation of security to investors). the field of currency trading is well aware of this kind of phenomenon. it is one of the main explanations adduced to justify the conundrum of the forward puzzle. other common explanation for the forward puzzle is the “peso problem”, see fama (1984). the “peso problem” is another conundrum on its own. it states that if an investor finds an asset with a potential excess return, she may suspect there is a risk behind a classic assumption of the modern portfolio theory. and if she is not able to find where the potential risk is hidden, then it is probable that the market is pricing a “peso problem”: a low frequency but massive negative result. the original example was apparently named in this way by milton friedman in the seventies, see sill (2000). the mexican peso was offering a spread return versus the dollar risk free rate even though a policy of fixed exchange rates was in place since a long time ago between the us dollar and mexican peso. after the late seventies peso devaluation, the saying was that the spread between the currencies interest rate returns was the market price for the potential disruptive but unexpected devaluation event. again, the “peso problem” advances a distribution of returns with fat tails and a high concentration of returns around the average. currency crashes, the peso problem and potential investor biases are the more accepted explanations of the failure of the uncovered interest rate parity. even though it is not still possible to find a consensus among researchers regarding the origin of the currency risk premium, it is difficult to simply deny the existence of such a premium. the analysis of naïve carry trade strategy returns over the last thirty years shows the existence of a persistent differential return. more refined strategies offer even better and more difficult to explain spread returns. see jordá and taylor (2009), clarida et al (2009), menkhoff et al (2014), ilmanen (2014), to name a just a few. as a final comment, john maynard keynes (1924) noted a long time ago that the interest rate parities may only hold, and it is even uncertain, if we expect no capital controls, perfect convertibility, neither sovereign risk nor country risk, and other different assumptions. but if the investor finds riskier countries or riskier currencies, whatever the definition of riskier we employ, we should expect excess expected returns for a currency strategy and positive risk premiums for the risky currency investments. so the assumption of zero excess return should not hold, and there is little sense in looking for the accomplishing of the uncovered interest rate parity in every possible currency pair. that is why a reflection regarding the use and 68 prado-dominguez and fernández-herráiz ● a sharpe-ratio-based measure adaptation of the sharpe´s reward to variability ratio to currency strategies could lead to a further understanding of the matter. in section 2 we define our sharpe ratio adaptation for evaluating currency expected returns and risks. section 3 shows two examples of potential latin american currencies with “peso problems” embedded and the information offered by our adaptation of the sharpe ratio. the paper ends with some conclusions and suggested paths for related research. a sharpe-ratio-based measure for currencies the raison d´être of the ratio, as stated with the very same words by sharpe (1994), is the analysis of zero investment strategies. in our view, strategies involving currencies, and particularly that carry trade, are appropriate examples of that way of thinking about differential returns. a currency is always quoted in relation to other currency. and the strategy of running short currencies with the lower rate and long currencies with the higher rate is a perfect zero investment one. the traditional building blocks of this strategy would be a short leg, borrowing money in the currency with the lower rate and a long leg, converting the proceeds into the investment currency and investing them at the higher rate. we will try to express a natural way of transposing the risk to variability ratio into the field of a simple currency strategy. first we will define the numerator of the ratio, the differential return, which will be called from now on indistinctly as excess return or time varying risk premium; or just the risk premium. the excess returns for the period 𝑡𝑡 to 𝑡𝑡 + 1, 𝑥𝑥𝑟𝑟𝑡𝑡+1, are defined as follows: 𝑥𝑥𝑟𝑟𝑡𝑡+1 = (𝑖𝑖𝑡𝑡∗ − 𝑖𝑖𝑡𝑡) + 𝑟𝑟𝑡𝑡+1 − 𝑟𝑟𝑡𝑡 where 𝑖𝑖𝑡𝑡∗ is the one period interest rate return for the investment currency and 𝑖𝑖𝑡𝑡 is the investor currency or risk free currency – we may consider these two definitions as equivalent for this exercise. and where 𝑟𝑟𝑡𝑡+1and 𝑟𝑟𝑡𝑡 are both the log of the exchange rate at those periods, expressed as units of investor currency (the risk free currency) per investment currency. when we represent the exchange rate in this manner, it is quite clear for the reader that an increase in the exchange rate implies an appreciation of the investment currency, also implying good news for a potential investor. it is equally possible to think about exchange rate the other way around, but we favored that representation for the sake of clarity. if there is an appreciation of the investment currency, the investor receives the interest rate spread for the period, but also the appreciation of the currency. both things sum up to the left hand side of the equation, 𝑥𝑥𝑟𝑟𝑡𝑡+1, the excess returns. as stated, when the investment period comes to an end, the realized return for the investor is the sum of both the interest rate spread and the variation in the exchange rate. nowadays, it is convenient to realize that the effect of the exchange rate could be positive, negative, or even zero. if we were to follow sharpe in the definition of the ex post or historical ratio, we would think about averaging the 𝑥𝑥𝑟𝑟𝑡𝑡+1observations for the t period so: 𝑥𝑥𝑟𝑟��� = 𝑇𝑇−1 � 𝑥𝑥𝑟𝑟𝑡𝑡+1 𝑇𝑇 𝑡𝑡=𝑡𝑡+1 we would then compute the standard deviation of the excess returns, 𝜎𝜎𝑥𝑥𝑥𝑥, and define the ratio as: 𝑆𝑆ℎ∗ = 𝑥𝑥𝑟𝑟��� 𝜎𝜎𝑥𝑥𝑥𝑥 and finally, for purposes of standardization, take the annualized version of the ratio multiplying by the square root of the number of periods in a year. 69 pere ● governance and economic development of western balkan countries 𝑆𝑆ℎ = 𝑥𝑥𝑟𝑟��� 𝜎𝜎𝑥𝑥𝑥𝑥 �𝑡𝑡𝑦𝑦 where 𝑡𝑡𝑦𝑦 is the number of periods in the year, being 12 for monthly excess returns, 52 for weekly excess returns, and so on and so forth. that would be a correct transposition of the historic sharpe ratio for currency investing ex post analysis. unfortunately, this may be a worthless exercise if we do not offer an answer to the definition of the ex-ante risk to variability ratio. the ex-ante sharpe ratio would be helpful for making investment decisions if the proper assumptions are respected, but is also the key to move the discussion towards a theoretical approach. so, what should be the definition of the ex-ante risk to variability ratio in a context of currency investing? let us begin with the more general formulation: 𝑆𝑆𝑡𝑡 = 𝐸𝐸𝑡𝑡(𝑥𝑥𝑟𝑟𝑡𝑡+1) 𝐸𝐸𝑡𝑡(𝜎𝜎𝑥𝑥𝑥𝑥) �𝑡𝑡𝑦𝑦 where the ratio is built including the expected excess return in t+1 conditioned on the information known at t in the numerator. the denominator will be the expected volatility of the excess return again conditioned on the information known at t. we invoke the former definition of excess returns for the period t+1. but now we will express the expected return based on the information known in t. so the expected excess return will be: 𝐸𝐸𝑡𝑡(𝑥𝑥𝑟𝑟𝑡𝑡+1) = (𝑖𝑖𝑡𝑡∗ − 𝑖𝑖𝑡𝑡) + 𝐸𝐸𝑡𝑡(𝑟𝑟𝑡𝑡+1) − 𝑟𝑟𝑡𝑡 as a matter of fact, interest rates for both currencies are known at t, and the currency spot rate at t is also known. but an expression of the expected nominal exchange rate in t+1 is needed if we want to draw an informed decision. there are different ways of coping with this issue, and the literature has developed appealing alternatives. for a promising proposition on currency valuation and predictability see menkhoff et al (2014). as noted by meese and rogoff (1983), nominal exchange rates are difficult to predict. they are so difficult to predict that even a random walk does a decent job compared to some structural models of exchange rates. since their seminal but discouraging paper, a lot of work and some progress have been made on the predictability of currencies. sarno and taylor (2003) offer an extensive review and cover these and other topics on currencies. that said, even today, the random walk seems the hardest benchmark for every new potential currency model. see rossi (2013) for a review of factors, models and results. we could assume that the nominal exchange rate in our model is well represented by a random walk. then, we could write that: 𝑟𝑟𝑡𝑡+1 = 𝑟𝑟𝑡𝑡 + 𝑟𝑟𝑡𝑡 where 𝑟𝑟𝑡𝑡 is white noise. if we take expectations on t, note that: 𝐸𝐸(𝑟𝑟𝑡𝑡+1) = 𝑟𝑟𝑡𝑡 assuming that the spot nominal rate follows a random walk, the expected currency excess returns simplify to: 𝐸𝐸(𝑥𝑥𝑟𝑟𝑡𝑡+1) = (𝑖𝑖𝑡𝑡∗ − 𝑖𝑖𝑡𝑡) the random walk simplification is used sometimes in the currency trading models of the industry, as ilmanen (2014) briefly notes. interestingly, the expected excess return ends up as the spread between the fixed return offered by the investment currency and the fixed return offered by the risk free currency. this naïve model of expected returns would suggest selecting the investing currency with the highest 70 prado-dominguez and fernández-herráiz ● a sharpe-ratio-based measure interest rate spread in order to gain access to the highest expected return. unfortunately, it is well known that spreads entail risks, and high spreads may imply high expected risks. in a freely convertible currency, and imposing a no arbitrage framework, expected currency volatility should also give the investor some ex-ante information about the potential risks of the carry trade strategy. it is sometimes possible to model volatility, and even to obtain an outlook of its next developments. there is a huge literature on this issue, but the reading of poon and granger (2003) may offer an extended critical survey on the matter. drawing on their conclusions, it seems that implied volatility extracted from quoted options on exchange rates does a decent job in predicting expected volatility. it usually beats even complex arch family models and other suggested procedures. a potential proxy for the expected excess returns volatility could be the implied at the money volatility. the implied volatility is available as the volatility expected for an asset which is “implied” in the quotations of the “at the money” options on the asset. taking the implied volatility, 𝑣𝑣𝑡𝑡 as the representation of the expected volatility, we will say that: 𝐸𝐸(𝜎𝜎𝑥𝑥𝑥𝑥) = 𝑣𝑣𝑡𝑡 finally we are able to construct the proposed ratio for our model of currency investing, so the ex-ante return to variability ratio may be: 𝑆𝑆𝑡𝑡 = (𝑖𝑖𝑡𝑡∗ − 𝑖𝑖𝑡𝑡) 𝑣𝑣𝑡𝑡 �𝑡𝑡𝑦𝑦 all variables were previously defined, and we prefer to keep the annualized formulation following sharpe (1994) once more. if we decide to use nominal year interest rates and annualized implied volatility, the ratio simplifies to: 𝑆𝑆𝑡𝑡 = (𝑖𝑖𝑡𝑡∗ − 𝑖𝑖𝑡𝑡) 𝑣𝑣𝑡𝑡 the proposed sharpe ratio shows the interest rate differential return per unit of implied volatility, and do not try to make a judgement regarding the future path of the investment currency. in fact, the ratio assumes that the currency follows a random walk with no drift. actually, ex-ante judgements about future currency outcomes are embedded into the implied volatility data. due to the potential predictability of future currency volatility patterns, the implied volatility data introduces the market view on potential future risks. but sometimes, for reasons that could be rooted in the ideas presented on the next paragraphs, a disconnection between risk expectations based on volatility and interest rate differentials between currencies may arise. this is a key question to consider because, at the end of the day, the interest rate spread is another way of looking at currency risks. in fact, interest rate spread is the risk premium of the asset: the investment currency for our purposes. we will make a first empirical approach coming back to the currencies where the “peso problem” was first noticed: the latin american currencies. ex-ante currency sharpe ratio and selected latin american currencies: seeking potential peso problems as a first approach towards the use of the proposed currency sharpe ratio for detecting peso problems, we will look at two of the latin american (latam) currencies with the highest interest rate spreads versus the dollar during the period of study. the currencies are the brazilian real (brlusd) and the argentinean peso (arsusd). we collected daily data for both currencies since january 2007 until march 2015 from the bloomberg professional lp database. the data starts well before the 71 pere ● governance and economic development of western balkan countries beginning of the last bear market in risky assets. in particular, naïve carry trade strategies suffered severe losses during the period, concentrated in the second half of 2008. the currencies evaluated are the brazilian real (brlusd) and the argentinean peso (arsusd) versus the us dollar. the series include the one-month-deposit interest rate for each currency including the us dollar; the implied at-the-money onemonth volatility for each currency, and the currency spot return for each month. we built the monthly ex-ante reward to variability ratios for each currency (arsusd_sr and brlusd_sr) using the proposed measure in section 2. the main descriptive statistics for the series of sharpe ratios are shown in table 1, and the evolution of the ratios during the period are shown in graphs 1 and 2. table 1. descriptive statistics for the ex-ante sharpe ratio series (january 2007 to march 2015) arsusd_sr brlusd_sr mean 1.97 0.7 median 1.74 0.69 maximum 7.12 1.31 minimum 0.23 0.26 std. dev. 1.14 0.24 skewness 1.26 0.49 kurtosis 5.89 2.49 jarque-bera 60.84 5.17 probability 0.000000 0.075446 sum 195.3151 69.5206 sum sq.dev. 128.3374 5.727 observations 99 99 a quick look at the data shows an unusually high ex-ante risk to variability ratio for the argentinean peso (arsusd_sr). average levels around 2 imply an expected 2 percent return per unit of implied volatility. clearly investors demand a high interest rate return differential for investing in pesos. the spread is so high that it obscures the level of volatility of the currency by 2 to 1. these results were expected, because the argentinean peso has been seen as one of the toughest currencies to invest in the last fifteen years. the market is clearly pricing the ugly reputation of the arsusd as an investment. ex-ante sharpe levels for the brazilian real (brlusd_sr) are clearly lower than the levels registered for ars, but still high compared to sharpe ratios offered by passive strategies. 72 prado-dominguez and fernández-herráiz ● a sharpe-ratio-based measure graph 1. ex-ante sharpe ratio for argentinean peso (arsusd_sr) (january 2007 to march 2015) graph 1 uncovers a demanding sharpe ratio for the argentinean peso during the period. for illustrative purposes, we drew a line at the 0.5 sharpe ratio level. sharpe ratios of 0.5 or between 0.3 and 0.5 are typical levels for bechmarking passive strategies. we would say that an asset with an sr of 0.5 will offer five points of return per each ten points of volatility. the average level of the ex-ante sharpe ratio for the sample period is around 1.97. levels of around 0.5 of ex-ante sr for ars are only attained during stress periods and reverse quickly towards higher levels. see, for example, the valleys in the graph during the summer of 2007, the last quarter of 2008, or the recent “tapering sell off” during may of 2013. paradoxically, local events like the big depreciation of january 2014 appear more like volatile periods for the ratio, but also potentially well rewarded by interest rate differentials, offering the investor willing to ride the potential “peso problem” a reward for the high risk involved in the trade. 0 1 2 3 4 5 6 7 8 2007 2008 2009 2010 2011 2012 2013 2014 2015 arsusd_sr 73 pere ● governance and economic development of western balkan countries graph 2. ex-ante sharpe ratios for brazilian real (brlusd_sr) (jan-2007 to mar-2015) although brl investors require a less demanding sharpe ratio for investing in the currency during the period of study, graph 2 shows that it is still well beyond the levels suggested for pure passive buy and hold strategies. the average value of the ex-ante sharpe ratio for the sample period is around 0,7. once more, the ratio may respond to potential peso problems embedded in the brl investing strategy. stressed markets constitute inflection points, as we saw in the ars case. the explosive volatility developments send the ratio down the elevator offering potential opportunities when the realized return is diluted due to exchange rate depreciation. conclusion the reward to variability ratio is a commonly used measure of investment efficiency. the ex-ante version helps in debating theoretical questions. our proposed adaptation to the ex-ante ratio for currencies leaves the potentially unpredictable currency movement apart, making use of the random walk hypothesis. we also consider implied volatility as a proxy for market expected volatility. following that path, we are able to construct an ex-ante sharpe ratio including only known variables at time t. it that sense we do not introduce any expectation on the potential return of the spot currency, so we are explicitly not considering the uncovered interest rate parity projected return. this way of presenting the ratio introduces a substantial bias towards selecting currencies with high interest spreads and low expected volatility, if we were to follow the classic approach of selecting strategies with the highest sharpe ratio. perhaps this proposal of a sharpe ratio for currencies may also serve investors in a different way. the ratio may uncover potential risks hidden in the structural relationship between expected volatility and expected interest rate spreads (the risk premium). once the ratio is built following those premises, a new and potentially interesting interpretation of the ratio arises. as we see it, a high risk premium per unit of implied volatility may signal a potential peso problem perception from market participants. 0.2 0.4 0.6 0.8 1.0 1.2 1.4 2007 2008 2009 2010 2011 2012 2013 2014 2015 brlusd_sr 74 prado-dominguez and fernández-herráiz ● a sharpe-ratio-based measure this could open a new line of financial research that could assess the investment currency required return due to “peso problems”, and even isolate it from other potential risk factors involved in currency investment strategies. but, is it a behavioral problem or a rational response to future potential very negative outcomes? both explanations may offer compelling arguments. we could conclude that high sharpe ratios signal a behavioral bias response for very negative past outcomes. it is easy to build the story for the currencies studied in the article. it could also be argued that these currencies require a high risk premium going forward due to a combination of political, macroeconomic and other risks. references clarida, richard, josh davis and niels pedersen (2009) ‘currency carry trade regimes: beyond the fama regression’, nber working paper no. 15523. fama, eugene (1984) ‘forward and spot exchange rates’, journal of monetary economics 14(3): 319-338. goetzmann, william, jonathan ingersoll, mattew i. spiege and ivo welch (2002) ‘sharpening sharpe ratios’, nber working paper no. 9116. ilmanen, antti (2014) expected returns. west sussex: john wiley & sons. jordà, oscar and alan m. taylor (2009) ‘the carry trade and fundamentals: nothing to fear but feer itself’, nber working paper no. 15518. keynes, john m. (1924) a tract on monetary reform. london: macmillan and co. meese, richard and kenneth rogoff (1983) ‘empirical exchange rate models of the seventies. do they fit out of sample?’, the journal of international economics 14: 3-24. menkhoff, lukas, lucio sarno, maik schmeling and andreas schrimpf (2014) ‘currency value’, kiel institute for the world economy, university of kiel. poon, ser-huang and clive granger (2003) ‘forecasting volatility in financial markets: a review’, journal of economic literature 41(2): 478-539. rossi, barbara (2013) ‘exchange rate predictability’, cepr discussion paper no. dp9575. sarno, lucio and mark p. taylor (2003) the economics of exchange rates. cambridge: cambridge university press. sharpe, william f. (1966) ‘mutual fund performance’, journal of business 39(1): 119-138. sharpe, william f. (1994) ‘the sharpe ratio’, the journal of portfolio management 21(1): 49-58. sill, keith. (2000) ‘understanding asset values: stock prices, exchange rates, and the “peso problem”’, federal reserve bank of philadelphia business review september/october: 3-13. 75 government balance-consistent economic growth rates and their implications european journal of government and economics 9(3), december 2020, 220-231 220 european journal of government and economics issn: 2254-7088 government balance-consistent economic growth rates and their implications: a study of the euro area countries charalambos pattichisa* a york business school, york st john university, york, united kingdom * corresponding author at: c.pattichis@yorksj.ac.uk abstract. using the model derived by bajo-rubio (2014), this paper estimates government budget balance-consistent growth rates for the euro area countries. a comparison of these estimates to their actual growth rates indicates that most of these countries are growing at rates that are very similar to their government balance-consistent growth rates. this finding implies that many euro area countries would not be experiencing excessive imbalances in their government budget over the long-run that could harm future economic growth. the analysis has also shown that for many euro area countries, the predictions of the model seem to be broadly consistent with their actual fiscal experience. keywords. balanced budget; budget deficits; economic growth jel codes. e60; h60; h68 doi. https://doi.org/10.17979/ejge.2020.9.3.5776 1. introduction in a recent paper, bajo-rubio (2014) has shown that thirlwall’s law (thirlwall, 1979) can be extended to derive the economic growth rate that is consistent with equilibrium between any two variables, e.g. between government revenue and government expenditure. thirlwall’s law has long been used to estimate an economy’s balance of payments-constrained growth rate (see thirlwall (2011) for a discussion and a review of some studies using the law). however, there do not seem to have been any studies done to calculate government budget balance-consistent growth rates. the aim of the present paper is to address this gap in the literature by investigating whether the euro area countries are growing at rates that are consistent with a balanced government budget and to discuss the resulting policy implications from the analysis. using the model derived by bajo-rubio (2014), this paper estimates government budget balance-consistent growth rates for the euro area countries and then compares these estimates to the actual growth rates of these economies. this comparison enables conclusions to be drawn as to whether government deficits may pose a threat to the future growth prospects of these countries. as the next section argues, this threat arises in various ways but also from the fact that a country may need to take corrective measures to tackle any persistent and excessive budget deficits, as the resulting increases in government debt cannot continue indefinitely without leading to a future fiscal crisis. these corrective (contractionary) measures may, therefore, have a negative impact on future economic growth. the approach taken by this paper is that if countries are growing at rates that are similar mailto:c.pattichis@yorksj.ac.uk https://doi.org/10.17979/ejge.2020.9.3.5776 charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 221 to their government budget balance-consistent growth rates, such negative effects on growth would be avoided. another aim of this paper is to investigate the predictions of the model and how they relate to the actual experience of these countries. in summary, the main results and implications of the analysis are: (a) the majority of euro area countries are growing at rates which are very similar to their government balance-consistent growth rates. this suggests that many of these countries would not be experiencing excessive imbalances in their government budget over the long-run and, therefore, there would be no need to take significant corrective measures that could harm future economic growth; (b) for many countries in the sample, the predictions of the model seem to be broadly consistent with their actual fiscal experience; and, (c) for those countries whose budget deficit may pose a serious threat to future economic growth, this paper discusses ways to reduce their government balanceconsistent growth rate. the remainder of this paper is organised as follows. the next section focuses on the effects of budget deficits, while section 3 explains the theoretical model and research approach. section 4 presents and discusses the data, results and their implications while section 5 checks the robustness of the results. the last section offers some concluding remarks. 2. the effects of budget deficits the empirical literature on the effects of budget deficits on economic growth is extensive. it is not the aim of this paper to provide an extensive review of this literature as the present paper is employing a different theoretical approach (which is discussed in the next section). as far as it can be ascertained, there are no other empirical studies that have employed the same approach as the present paper. thus, the discussion in this section is intended to highlight the importance of studying this topic by noting the likely effects of budget deficits on economic growth with some results from the literature. budget deficits may affect economic growth in various ways. there is, for example, a huge literature analysing the effects of budget deficits on interest rates, see amongst others, the studies of laubach (2009), kameda (2014), and cebula (2018). the theoretical argument here is that higher budget deficits may increase interest rates as the government tries to finance the deficit with increased borrowing. in addition, higher budget deficits may increase consumption and reduce national saving, thus, leading to a rise in interest rates. the resulting higher interest rates may crowd-out private investment, reduce the capital stock and, hence, reduce long-term economic growth. thus, any short-run positive effects on consumption and growth via keynesian multiplier effects may be eliminated in the long-run. furthermore, economic theory suggests that budget deficits may be inflationary if, for example, they are financed by an increase in the money supply. for example, studies by jalil et al (2014) and ahmad and aworinde (2019) found that budget deficits are indeed inflationary. on the other hand, catão and terrones (2005) have investigated this issue for various groups of countries and have reported that budget deficits increase inflation in developing countries and in countries with charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 222 already high inflation but do not in the case of advanced countries with already low inflation. similarly, lin and chu (2013), using panel data for 91 countries over 47 years, looked at episodes of low, middle and high inflation and have found that budget deficits increase the inflation rate in the latter two episodes; however, in the case of low-inflation episodes, they have reported a weak impact of budget deficits on inflation. another possible negative effect of higher budget deficits is that they can lead to higher current account deficits (the twin deficits hypothesis), see for example, the evidence presented by forte and magazzino (2013) and trachanas and katrakilidis (2013). this negative effect on the current account deficit may come about via the various consequences of budget deficits discussed above. for example, if budget deficits increase interest rates, the resulting increase in financial inflows will cause an exchange rate appreciation with negative effects on the current account deficit. furthermore, if budget deficits lead to inflation, the resulting loss in competitiveness will also increase the current account deficit. in addition, if budget deficits lead to an increase in private consumption, the current account deficit will increase as expenditure on imported goods is likely to increase. the literature investigating the effects of budget deficits on economic growth is voluminous and shows mixed results. cebula (1995), for example, has investigated the effect of us budget deficits on us economic growth and has found a negative effect. in contrast, the results of taylor et al. (2012), using us data from 1961-2011, suggest that higher budget deficits increase economic growth. adam and bevan (2005) found that reducing the budget deficit to around 1.5% of gdp enhances growth (but any further reduction does not). castro (2011) has investigated the effect of the fiscal rules of the european union on growth and did not find a negative effect. finally, lau and yip (2019) have analysed the effect of budget deficits on the economic growth of asean countries both before and after the global financial crisis and have reported that budget deficits had a negative effect before the crisis but a positive effect after it. 3. theoretical model and research approach in a recent paper, bajo-rubio (2014) has shown that thirlwall’s law (thirlwall, 1979) can be generalised to calculate the growth rate that is consistent with an equilibrium between any two variables. focusing on the government budget balance, where the relevant economic variables are government revenue (t) and government expenditure (g), bajo-rubio (2014) has shown that the rate of economic growth that is consistent with a balanced budget is given by equation (1) below: ŷgb = ġ / ety (1) in equation (1), ŷgb is the government balance-consistent growth rate (i.e. the rate of growth of an economy that is consistent with a balanced government budget), ġ is the rate of growth of government expenditure, and ety is the income elasticity of tax revenues (a measure of the responsiveness of tax revenues to a change in national income). charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 223 in deriving equation (1), bajo-rubio (2014) assumes that initially g=t and, therefore, equation (1) shows that the budget balance will be maintained if the growth rate of government expenditure is equal to the growth rate of government revenue, with the latter growth rate estimated in the model as (ŷgb times ety). equation (1), therefore, shows the rate of economic growth consistent with the equality of government revenue and government expenditure. this government balanceconsistent growth rate effectively defines the lowest rate at which an economy can grow without experiencing budget deficits.1 as bajo-rubio (2014) stated, if an economy grows at a rate greater than ŷgb it will experience a budget surplus and if it grows at a rate lower than ŷgb it will experience a budget deficit. in the latter case, the resulting budget deficits may affect the growth prospects of an economy in various ways, as discussed in the previous section. for example, a country may need to take corrective measures to tackle the persistent budget deficits as the resulting increases in government debt are a burden on future generations that cannot continue indefinitely without leading to a future fiscal crisis. these measures may come through a reduction in government expenditure or an increase in taxation, which may have a negative effect on future economic growth. thus, the eventual tackling of persistent budget deficits and the resulting government debt may negatively affect future economic growth. using equation (1), this paper estimates the government budget balance-consistent growth rates for all nineteen euro area countries and then compares these estimates to the actual growth rates of these economies. this comparison enables conclusions to be drawn as to whether fiscal imbalances are likely to be a serious problem. this paper also investigates the predictions of the model and how they relate to the actual experience of these countries.2 the next section presents and discusses the data, results and implications of the analysis. 4. data, results and implications 4.1. data to apply the above framework, the analysis requires estimates of the growth rate of government expenditure (ġ) and of the income elasticity of tax revenue (ety) for all 19 member states of the euro area. the analysis also requires estimates of the actual growth rate of real gdp (�̇�𝑦) for each country. as proxies for ġ and �̇�𝑦, the paper uses the estimated coefficient of the time trend in a regression of the natural logarithm of each variable in level form on a constant and a time trend. the data used in both cases is over the period 1995-2018 and were taken from the annual macroeconomic database of the european commission’s directorate general for economic and financial affairs (ameco). the data for government expenditure refers to the real total expenditure of general government (code: outg), while data for the income variable refer to real 1 bajo-rubio (2014) refers to this as the government balance-constrained growth rate. this paper prefers the term government balance-consistent growth rate since an economy can grow at above this rate and experience a budget surplus, which is not necessarily a constraint. 2 as mentioned in the previous section, there do not seem to be any other empirical studies that have employed the same approach as the present paper. there is, however, a huge literature investigating the sustainability of fiscal policy by focusing on the intertemporal budget constraint, see for example, the recent papers by brady and magazzino (2018a, 2018b) and the numerous studies cited therein. charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 224 gross domestic product at 2010 reference levels (code ovgd). data for the real total expenditure of general government were only available for all countries from 1995, and this dictated the sample size. estimates for the income elasticity of tax revenue (ety) were taken from koester and priesmeier (2017), but the robustness of the results was checked by using other tax elasticity estimates as explained later in this paper. 4.2. results the results of the estimated government balance-consistent economic growth rates are summarised in table 1. the following points provide a discussion of the results presented in table 1. (a) the first observation that can be drawn from these results is that most euro area countries are growing at rates (�̇�𝑦) which are very similar to their government balance-consistent growth rates (ŷgb). this suggests that many of these countries should not be experiencing excessive imbalances in their government budget over the long-run. thus, for many of these countries, the budget deficit may not pose a serious threat to future economic growth as corrective (contractionary) measures to tackle it would not be necessary. as can be seen in table 1, six countries are growing at rates above their government balance-consistent growth rates and are, therefore, expected to experience budget surpluses. the remaining countries are growing at rates that are below their government balance-consistent growth rates and are, therefore, expected to experience budget deficits. however, for many of the latter countries, the actual growth rate is very close to the government balance-consistent growth rate and, as such, the resulting budget deficits may not be a serious problem. (section 4.3 discusses further implications of the analysis, especially for those countries which are growing at rates that are below their government balanceconsistent growth rate). (b) the last two columns of table 1 summarise the actual experience of these countries: they show, respectively, the average budget deficit and average budget surplus (as a percentage of gdp) over the period 1995-2018, i.e. the average deficit of those years where a country had a deficit and the average surplus of those years where a country had a surplus. this was calculated using data on net lending/net borrowing of the general government as a percentage of gdp (variable code ublg in the ameco database). as can be seen, most of these countries had small budget imbalances over the period which tends to support the predictions of the model (as many of them were growing at rates which are very similar to their government balance-consistent growth rates).3 it is important to mention here that this may be the result of the stability and growth pact (sgp) which constrains, in general, euro area countries to budget deficits of no greater than 3% of their gdp. this sgp requirement imposes discipline on the growth of government expenditure in these countries which, via equation (1), lowers their government balance-consistent growth rate. 3 it goes without saying that if this paper had averaged deficits and surpluses together over the period, the resulting budget imbalances will be lower for most countries than those shown in table 1. charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 225 table 1. estimates of the government balance-consistent economic growth rates. ety ġ ŷgb �̇�𝑦 average budget deficit average budget surplus austria 0.99 1.42 1.43 1.70 -2.61 0.11 belgium 1.06 2.00 1.89 1.67 -2.41 0.13 cyprus 1.40 3.56 2.54 2.35 -3.95 1.54 estonia 0.96 3.86 4.02 3.67 -1.00 1.30 finland 1.07 2.01 1.88 1.81 -2.35 3.61 france 1.05 1.83 1.74 1.47 -3.60 nap germany 1.04 0.76 0.73 1.33 -2.88 0.87 greece 1.21 1.15 0.95 0.41 -7.86 0.76 ireland 0.88 4.14 4.70 4.52 -6.88 1.72 italy 1.15 0.46 0.40 0.36 -3.37 nap latvia 0.98 4.17 4.26 3.77 -2.37 0.50 lithuania 1.07 3.48 3.25 3.99 -3.32 0.46 luxembourg 1.00 3.46 3.46 3.15 -0.87 2.28 malta 1.17 2.90 2.48 3.25 -4.42 2.11 netherlands 0.96 1.51 1.57 1.67 -2.78 0.63 portugal 1.20 1.39 1.16 0.81 -4.84 nap slovakia 0.80 3.07 3.84 3.89 -4.57 nap slovenia 0.98 2.46 2.51 2.34 -3.75 0.37 spain 1.06 2.26 2.13 1.88 -4.58 1.78 notes and sources (table 1): ety are the long-run income elasticities of tax revenue taken from koester and priesmeier (2017) except for lithuania which was not part of their study. the estimate for lithuania is taken from european commission (2014, pg. 45); ġ is the (percentage) growth rate of real government expenditure; ŷgb is the government balance-consistent economic growth rate calculated using equation 1; and ẏ is the actual (percentage) growth rate of real gdp. the last two columns show the average budget deficit and average budget surplus (as a percentage of gdp) over the period 1995-2018 (i.e. the average deficit of those years where a country had a deficit and the average surplus of those years where a country had a surplus, respectively); nap stands for not applicable. (c) the paper takes now a closer look at each country individually. as explained earlier in the paper, if an economy grows at a rate greater than ŷgb, the model predicts that it will have a budget surplus and if it grows at a rate lower than ŷgb it will have a budget deficit. therefore, as already mentioned, the model predicts a deficit for 13 euro area countries and a surplus for the remaining six. in several cases, these predictions seem to be consistent with the experience of these countries, but in some cases, they are not. let’s consider each country in turn by comparing the prediction of the model with the actual fiscal experience of each country over the 1995-2018 period: • austria: the model predicts a budget surplus. however, for most of the 1995-2018 period, the country has experienced a deficit, albeit a small one on average. • belgium: the model predicts a budget deficit. this is consistent with the experience of the country over most of the 1995-2018 period. • cyprus: the model predicts a budget deficit. this is consistent with the experience of the country over most of the 1995-2018 period. • estonia: the model predicts a budget deficit. however, for about half of the years over the 1995-2018 period, the country has experienced a small surplus and for the remaining years a small deficit (i.e. overall, the country had a fairly balanced budget). charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 226 • finland: the model predicts a budget deficit. this is consistent with the post-2009 experience of the country but not with the pre-2009 experience. • france: the model predicts a budget deficit. this is consistent with the experience of the country over the period as france has experienced deficits in all years of the 1995-2018 period. • germany: the model predicts a budget surplus. even though germany had a surplus in the post-2014 period, for almost all previous years it has experienced deficits. • greece: the model predicts a budget deficit. this is consistent with the experience of the country over most of the period. • ireland: the model predicts a budget deficit. however, for about half of the years over the 1995-2018 period, the country has experienced a surplus and for the remaining years a deficit. • italy: the model predicts a budget deficit. italy has experienced deficits in all years of the 1995-2018 period. • latvia: the model predicts a budget deficit. this is consistent with the experience of the country over most of the period. • lithuania: the model predicts a budget surplus. this is not consistent with the experience of the country over most of the period. • luxembourg: the model predicts a budget deficit. this is not consistent with the experience of the country over most of the period. • malta: the model predicts a budget surplus. this is not consistent with the experience of the country over most of the period. • netherlands: the model predicts a budget surplus. the country had experienced surpluses in seven years over the period and deficits in the rest. these imbalances have been generally small. • portugal: the model predicts a budget deficit. portugal has experienced deficits in all years of the 1995-2018 period. • slovakia: the model predicts a budget surplus. the model prediction is inconsistent with actual experience as slovakia has experienced deficits in all years of the 1995-2018 period. • slovenia: the model predicts a budget deficit. this is consistent with the experience of the country over most of the period. • spain: the model predicts a budget deficit. this is consistent with the experience of the country over most of the period. it is important to stress that the above (individual country) predictions should be interpreted with care as most of the euro area countries were growing at rates (�̇�𝑦) which were very similar to their government balance-consistent growth rates (ŷgb). thus, even though some of the individual predictions are not consistent with actual experience, the imbalances for most countries are predicted to be small (which was the case for many of these countries). this is also supported by the fact that most of these countries have experienced deficits in some years and surpluses in charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 227 others over the 1995-2018 period. 4.3. implications this section focuses on the implications of the analysis, especially for those countries whose actual growth rate is below their government balance-consistent growth rate and, as such, the resulting budget deficits may be a serious problem. apart from the obvious implication that such countries should try to increase their actual rates of economic growth (in ways that do not increase their budget deficit), they may also try to take measures to reduce their government balanceconsistent growth rate. it should be clear from equation (1) that this can be done in two ways: firstly, by reducing the growth rate of government expenditure (ġ); and, secondly, by adopting measures that increase the income elasticity of tax revenue (ety). if reducing the growth rate of government expenditure may have a negative impact on economic growth, what conclusion can be drawn about the appropriate rate of growth of government expenditure? to answer this question, we need to look at the evidence on tax revenue elasticities. the empirical literature seems to suggest that total tax revenue elasticities are not different from one for many countries.4 for example, deli et al (2018), using panel data of 25 oecd countries from 1965-2015, have shown that the long-run estimates of total revenue elasticities for the panel are not different from one. they have also shown that this result holds for most countries in their sample (for 16 out of 25 countries). similar results have also been obtained by dudine and jalles (2018). belinga et al (2014) have also found that, for about half of the countries in their sample, the long-run total tax buoyancies were not statistically different from one.5 using equation (1), these results from the literature suggest that, for many countries, a balanced government budget requires that the growth rate of government expenditure (ġ) must be equal to the rate of economic growth (�̇�𝑦); or, to avoid deficits, the growth rate of government expenditure must not exceed the rate of economic growth. put differently, any percentage change in government expenditure must lead to an at least equal percentage change in aggregate income.6 a similar point has been made by dudine and jalles (2018). another possible course of action is for countries to change the composition (but not the level) of their government expenditure in ways that enhance economic growth. for example, bussière et al (2017) have shown that an increase in government investment of 1% of gdp that is compensated by a similar fall in government consumption will stimulate economic growth. 4 even though koester and priesmeier (2017) have found that, for most of the countries in their sample, the long-run tax revenue elasticities are statistically different from one, the estimated value for many of them is close to one. this is also supported by the data given in table c1 of their paper that summarises tax revenue elasticities estimated in other studies. as koester and priesmeier (2017) note, the difference in these elasticity estimates is small despite the fact that some of the estimates are based on gdp growth and others on the output gap. 5 see deli et al (2018), dudine and jalles (2018) and koester and priesmeier (2017) for a discussion of the differences between tax buoyancies and tax elasticities. 6 this does not mean that the fiscal multiplier effect is not present. this is because, in absolute monetary (euro) terms, a given percentage change in gdp is a bigger value than the same percentage change in government spending. for example, a 1% increase in government spending may lead to a 1% increase in gdp but the latter change is bigger in absolute monetary terms (due to the multiplier effect). however, as ramey (2019) shows, recent evidence suggests that government expenditure multipliers range from 0.6-1, i.e. they may be less than 1. charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 228 the second course of action for a country is to adopt measures that increase the income elasticity of tax revenue. there are many factors that could influence this. as koester and priesmeier (2017) argue, the composition of tax revenue (as between various direct and indirect taxes) is relevant here. for example, empirical evidence tends to suggest that corporate taxes are more responsive to gdp growth and have a long-run tax buoyancy greater than one in advanced economies, see for example the evidence presented by belinga et al (2014), dudine and jalles (2018) and deli et al (2018).7 thus, if total tax revenue relies more heavily on more buoyant taxes, then the income elasticity of total tax revenue will be higher (dudine and jalles, 2018). furthermore, as dudine and jalles (2018) and koester and priesmeier (2017) argue, taking measures to improve tax collection in ways that enhance tax compliance would also increase the income elasticity of total tax revenue. in addition, dudine and jalles (2018) have shown that more openness to trade, higher human capital, and lower volatility in output and inflation would also increase tax buoyancies. koester and priesmeier (2017) have also argued that a more progressive tax structure could increase tax revenue elasticities. the above factors, therefore, could be considered by a country trying to increase its income elasticity of total tax revenue. for a more detailed discussion of these factors see dudine and jalles (2018), koester and priesmeier (2017) and belinga et al (2014). 5. checking the robustness of the results to check the robustness of the conclusions, this paper has also calculated government balanceconsistent growth rates using elasticity estimates from two other sources: (a) the long-run total tax buoyancies estimated by dudine and jalles (2018) (for those countries whose estimates were statistically significant); and, (b) the long-run total tax buoyancies estimated by belinga et al (2014). the results of these robustness checks are presented in table 2. (please note that not all euro area countries were covered by the above two studies). comparing the actual growth rates of these countries with the government balance-consistent growth rates calculated using the new elasticity estimates (given in table 2) reveals that the conclusions remain the same for most countries. thus, the results and conclusions based on table 1 are robust with the following four exceptions: • ireland: the conclusions for ireland remain the same when using the dudine and jalles (2018) elasticity estimate but not when using the belinga et al. (2014) estimate. in the latter case, the model predicts a budget surplus. as has been noted earlier, for about half of the years over the 1995-2018 period, the country had experienced a surplus and for the remaining years a deficit. 7 the latter two studies have also shown that corporate income tax buoyancies are greater during downturns than during expansions of economic activity. for another analysis of the effects of the business cycle on tax elasticities, see boschi and d’addona (2017). charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 229 table 2. robustness checks. ety (b) ety (dj) ŷgb (b) ŷgb (dj) �̇�𝑦 austria 1.07 1.0 1.33 1.42 1.70 belgium 1.03 1.013 1.94 1.97 1.67 cyprus na 1.446 na 2.46 2.35 estonia 0.98 1.019 3.94 3.79 3.67 finland 1.08 1.007 1.86 2.00 1.81 france 1.12 1.054 1.63 1.74 1.47 germany 1.05 1.422 0.72 0.53 1.33 greece 1.09 1.953 1.06 0.59 0.41 ireland 0.98 0.911 4.22 4.54 4.52 italy 1.13 0.807 0.41 0.57 0.36 latvia na na na na 3.77 lithuania na na na na 3.99 luxembourg 1.05 ns 3.30 na 3.15 malta na na na na 3.25 netherlands 0.86 ns 1.76 na 1.67 portugal 1.12 1.295 1.24 1.07 0.81 slovakia 0.75 0.798 4.09 3.85 3.89 slovenia 1.01 1.017 2.44 2.42 2.34 spain 1.21 ns 1.87 na 1.88 notes and sources (table 2): ety (b) are the long-run total tax buoyancies taken from table 1 of belinga et al (2014); ety (dj) are the long-run total tax buoyancies taken from table a1a of dudine and jalles (2018); ŷgb (b) and ŷgb (dj) are the respective government balance-consistent growth rates calculated using equation (1); ẏ is the actual (percentage) growth rate of real gdp; na implies that data were not available and ns implies that the estimates were not statistically significant and, therefore, not utilised. • netherlands: using the belinga et al. (2014) elasticity estimate, the model now predicts a budget deficit. as has been noted earlier, the country had experienced surpluses in seven years over the period and deficits in the rest (but these imbalances have generally been small). • slovakia: the conclusions for slovakia remain the same when using the dudine and jalles (2018) elasticity estimate but not when using the belinga et al. (2014) estimate. in the latter case, the model now predicts a budget deficit which is consistent with actual experience, as slovakia had experienced deficits in all the years of the 1995-2018 period. • spain: using the belinga et al. (2014) elasticity estimate, the model now predicts a balanced budget as the actual growth rate is almost the same as the government balanceconsistent growth rate. as has been noted earlier, though, spain has experienced budget deficits over most of the 1995-2018 period. 6. conclusions an important issue in macroeconomics is the link between economic growth and the government fiscal balance. using the model derived by bajo-rubio (2014), this paper has estimated government budget balance-consistent growth rates for the euro area countries and compared these estimates to the actual growth rates of these economies. this comparison enabled an charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 230 assessment as to whether any fiscal imbalances may pose a threat to future economic growth. this threat could arise, for example, from the fact that a country may need to take corrective measures to tackle any persistent budget deficits, as the resulting increases in government debt cannot continue indefinitely without leading to a future financial crisis. these corrective (contractionary) fiscal measures may, therefore, have a negative impact on future economic growth. such corrective measures may also be necessary if budget deficits lead to higher inflation and higher current account deficits. furthermore, budget deficits may lower economic growth if they lead to higher interest rates that may crowd-out private investment. the analysis has revealed that most euro area countries are growing at rates that are very similar to their government balance-consistent growth rates. this finding suggests that many of these countries would not be experiencing excessive fiscal imbalances. therefore, there would be no need for major contractionary measures that could harm future economic growth. this finding may be the result of the fiscal discipline imposed on these countries by the stability and growth pact. the analysis has also shown that, for many countries in the sample, the predictions of the model seem to be broadly consistent with their actual fiscal experience. in terms of policy implications, this paper has summarised ways to reduce the government balance-consistent growth rate by increasing the income elasticity of tax revenue and/or by reducing the growth rate of government expenditure. acknowledgements i am grateful to two anonymous referees of this journal for useful comments, but i am solely responsible for any remaining errors. references adam, c.s. and bevan, d.l. (2005). fiscal deficits and growth in developing countries. journal of public economics, 89, 571-597. https://doi.org/10.1016/j.jpubeco.2004.02.006 ahmad, a.h. and aworinde, o.b. (2019). are fiscal deficits inflationary in african countries? a new evidence from an asymmetric cointegration analysis. north american journal of economics and finance, 50. https://doi.org/10.1016/j.najef.2019.100999 bajo-rubio, o. (2014). balance-constrained growth rates: generalizing thirlwall’s law. applied economics letters, 21(9), 593–596. https://doi.org/10.1080/13504851.2013.877563 belinga, v., benedek, d., de mooij, r. and norregaard, j. (2014). tax buoyancy in oecd countries. international monetary fund, imf working paper wp/14/110. https://doi.org/10.5089/9781498305075.001 boschi, m. and d’addona, s. (2017). the stability of tax elasticities over the business cycle in european countries. australian national university, cama working paper 44/2017. https://doi.org/10.2139/ssrn.2967524 brady, g. l. and magazzino, c. (2018a). fiscal sustainability in the eu, atlantic economic journal, 46(3), 297-311, doi:10.1007/s11293-018-9588-4. https://doi.org/10.1007/s11293018-9588-4 brady, g. l. and magazzino, c. (2018b). government debt in emu countries, the journal of economic asymmetries, 18. https://doi.org/10.1016/j.jeca.2018.e00096 https://doi.org/10.1016/j.jpubeco.2004.02.006 https://doi.org/10.1016/j.najef.2019.100999 https://doi.org/10.1080/13504851.2013.877563 https://doi.org/10.5089/9781498305075.001 https://doi.org/10.2139/ssrn.2967524 https://doi.org/10.1007/s11293-018-9588-4 https://doi.org/10.1007/s11293-018-9588-4 https://doi.org/10.1016/j.jeca.2018.e00096 charalambos pattichis / european journal of government and economics 9(3), december 2020, 220-231 231 bussière, m., ferrara, l., juillard, m. and siena, d. (2017). can fiscal budget-neutral reforms stimulate growth? model-based results. banque de france working paper no. 625. https://doi.org/10.2139/ssrn.2955120 castro, v. (2011). the impact of the european union fiscal rules on economic growth. journal of macroeconomics, 33(2), 313-326. https://doi.org/10.1016/j.jmacro.2010.09.003 catão, l.a.v. and terrones, m.e. (2005). fiscal deficits and inflation. journal of monetary economics, 52(3), 529-554. https://doi.org/10.1016/j.jmoneco.2004.06.003 cebula, r.j. (1995). the impact of federal budget deficits on economic growth in the united states: an empirical investigation, 1955-1992. international review of economics and finance, 4(3), 245-252. https://doi.org/10.1016/1059-0560(95)90042-x cebula, r.j. (2018). impact of us federal government budget deficits on the ex ante real interest rate yield on 10-year treasury notes during the post-bretton woods (1972-2016) era. journal of economics and finance, 42, 828-835. https://doi.org/10.1007/s12197-017-9424-8 deli, y., rodriguez, a. g., kostarakos, i. and varthalitis, p. (2018). dynamic tax revenue buoyancy estimates for a panel of oecd countries. the economic and social research institute (esri) dublin, esri working paper no. 592. dudine, p. and jalles, j. t. (2018). how buoyant is the tax system? new evidence from a large heterogeneous panel. journal of international development, 30(6), 961-991. https://doi.org/ 10.1002/jid.3332 european commission (2014). report on public finances in emu 2014. european economy 92014. https://doi.org/10.2765/7626 forte, f. and magazzino, c. (2013). twin deficits in european countries. international advances in economics research, 19(3), 289-310, https://doi.org/10.1007/s11294-013-9406-3 jalil, a., tariq, r. and bibi, n. (2014). fiscal deficit and inflation: new evidences from pakistan using a bounds testing approach. economic modelling, 37, 120-126. https://doi.org/10.1016/j.econmod.2013.10.029 kameda, k. (2014). budget deficits, government debt, and long-term interest rates in japan. journal of japanese and international economies, 32, 105-124. https://doi.org/10.1016/ j.jjie.2014.02.001 koester, g. and priesmeier, c. (2017). revenue elasticities in euro area countries. european central bank, ecb working paper 1989. lau, w-y. and yip, t-m. (2019). the nexus between fiscal deficits and economic growth in asean. journal of southeast asian economies, 36(1), 25-36. https://doi.org/10.1355/ae361d laubach, t. (2009). new evidence on the interest rate effects of budget deficits and debt. journal of the european economic association, 7(4), 858-885. https://doi.org/10.1162/jeea. 2009.7.4.858 lin, h-y. and chu, h-p. (2013). are fiscal deficits inflationary? journal of international money and finance, 32, 214-233. https://doi.org/10.1016/j.jimonfin.2012.04.006 ramey, v.a. (2019). ten years after the financial crisis: what have we learned from the renaissance in fiscal research? journal of economic perspectives, 33(2), 89-114. https://doi.org/10.1257/jep.33.2.89 taylor, l., proaño, c.r., de carvalho, l. and barbosa, n. (2012). fiscal deficits, economic growth and government debt in the usa. cambridge journal of economics, 36 (1), 189–204. https://doi.org/10.1093/cje/ber041 thirlwall, a. p. (1979). the balance of payments constraint as an explanation of international growth rate differences. banca nazionale del lavoro quarterly review, 32(128), 45–53, (republished in 2011 in psl quarterly review, 64(259). thirlwall, a. p. (2011). balance of payments constrained growth models: history and overview. psl quarterly review, 64(259), 307–351. trachanas, e. and katrakilidis, c. (2013). the dynamic linkages of fiscal and current account deficits: new evidence from five highly indebted european countries accounting for regime shifts and asymmetries. economic modelling, 31, 502-510. https://doi.org/10.1016/ j.econmod.2012.12.026 https://doi.org/10.2139/ssrn.2955120 https://doi.org/10.1016/j.jmacro.2010.09.003 https://doi.org/10.1016/j.jmoneco.2004.06.003 https://doi.org/10.1016/1059-0560(95)90042-x https://doi.org/10.1007/s12197-017-9424-8 https://doi.org/10.1002/jid.3332 https://doi.org/10.1002/jid.3332 https://doi.org/10.2765/7626 https://doi.org/10.1007/s11294-013-9406-3 https://doi.org/10.1016/j.econmod.2013.10.029 https://doi.org/10.1016/j.jjie.2014.02.001 https://doi.org/10.1016/j.jjie.2014.02.001 https://doi.org/10.1355/ae36-1d https://doi.org/10.1355/ae36-1d https://doi.org/10.1162/jeea.2009.7.4.858 https://doi.org/10.1162/jeea.2009.7.4.858 https://doi.org/10.1016/j.jimonfin.2012.04.006 https://doi.org/10.1257/jep.33.2.89 https://doi.org/10.1093/cje/ber041 https://doi.org/10.1016/j.econmod.2012.12.026 https://doi.org/10.1016/j.econmod.2012.12.026 european journal of government and economics volume 5, number 2 (december 2016) issn: 2254-7088 104 switching costs in the european postal service. are there any solutions? carlos pateiro-rodríguez, universidade da coruña, spain javier prado-domínguez, universidade da coruña, spain jesús m. garcía-iglesias, universidad de extremadura, spain josé m. barreiro-viñán, universidade da coruña, spain abstract this article examines the costs of switching that may exist in the european postal sector, where it is carried out an ambitious process of opening to competition since 1997. inadequate regulation of the access to some elements of postal infrastructure or services within the scope of the universal postal service exists. this article proposes adaptations to ensure transparent and non-discriminatory access conditions to elements of postal infrastructure in line with the sectorial directives aimed at strengthening competition in the long term in the postal market. the proposed adaptations focus on services such as postcode systems, address databases, post office boxes, delivery boxes, re-direction and return to sender services. all of them can help reduce the switching cost and thus strengthen competition. keywords liberalization; universal postal service; switching costs; postal network. jel classification l87; m31; l32; l51. european journal of government and economics 5(2) 105 introduction prior to its liberalization, the european postal sector was characterized by typical monopolistic characteristics: it was a market where a dominant provider (usually publically owned) controlled close to 100 per cent of the market share. in response to this situation, three directives were introduced between 1997 and 2008. the three directives were aimed at promoting the opening of the european postal market. the three postal directives maintain the universal postal service (ups), from which the universal service obligations (usos) are derived, beyond the full market opening (fmo). keeping the usos may well put the universal service provider (usp) economic equilibrium in danger, and hence, the directives design mechanisms in order to finance the additional costs of such usos: (i) a mechanism to compensate the undertaking concerned from public funds or (ii) a mechanism for the sharing of the net costs of the universal service obligations between providers of services and/or users. the ups extends to a set of high-quality postal services with which all users are permanently provided at an affordable price throughout the territory. the usp must assume the usos. these can work as a retention mechanism on the part of providers, and can strengthen switching costs because many of the customers wishing to switch providers need to return to their incumbent provider for some aspect of their mail services. this is due to the potentially incomplete territorial network coverage of an entrant who has not offered ups or due to inefficiencies in the access regulations of the postal network. eighteen years after the first directive, the opening of the postal market in the main eu states has not reached the levels expected in the ambitious liberalization process. the usps that existed before the liberalization process had still retained a high market share, approximately 90 per cent, in terms of main postal products. furthermore, as pointed out by jonsson and selander (2006) and pateiro et al (2013a; 2013b), some entry attempts have only reached very low market shares or even failed. though the markets will be open to competition, the lack of legal restrictions to competition does not mean that competition will be present (okholm et al 2010: 80). for example, although the markets of estonia, finland, germany, spain, sweden, and the united kingdom (uk) are fully liberalized legally, the actual competition level in these countries is low or has been undisclosed. the dominant provider continues to hold dominance in correspondence and direct mail segment (more than 90 per cent) in most of the european markets: germany, hungary, iceland, luxembourg, netherlands, spain, portugal, slovenia, and slovakia. the most competitive markets are unaddressed delivery and parcel (about 50 per cent). historically, these are the markets where competition has been allowed for the longest period of time. are there any exceptions? a parliamentary decision of 1992 ended with the letter mail monopoly of the incumbent operator swedish post (posten ab). according to jonsson and selander (2006: 364), the entry of citymail, in 1991, was an important trigger for the liberalization. city mail’s bumpy road towards being established as the world’s first entrant with a significant portion of a domestic letter mail market has included two bankruptcies and several changes in ownership. according to the findings of cohen et al (2004), citymail would not have been able to attract investors if its original business plan had been analyzed correctly. the reasons for this outcome may stem from the impact of new communication technologies on traditional postal communication (fève et al 2010) and the current economic crisis (martin et al 2012; trinkner and grossmann 2006). in this context, a declining traditional postal market will not attract a large number of new entrants. in view of the introduction of substitute electronic communication and the opening of the market, the council of the european union recommended diversifying the activities of ups providers by providing electronic business services or other information services. many european ups providers, undisputed leaders in the pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 106 declining traditional post, have directed their strategies toward the parcel sector and e-commerce, where its market share was relatively low. such providers are moving towards a wider range of not strictly postal products to make ensure the profitability of postal network and their human and technical resources. in addition to the above factors, particularly the impact of new communication technologies on traditional postal communication and the current economic crisis, which have made entering the postal sector less attractive, the authors can also include provider switching costs as an entry barrier. it is difficult for customers to switch providers due to increasing switching costs and a lack of information on existing alternatives (patterson and smith 2003; sharma and patterson 2000; white and yanamandram 2007). furthermore, an established relationship, which is inherent in a historically monopolized postal model, may generate shared values, a stronger identification between the parties, and added profits. these factors act as an entry barrier to competitors (reinartz and kumar 2003; deligonul et al 2006). finally, the benefit of ups demands conditions of permanence, ubiquity and frequency, which accentuate the consumer attraction, as opposed to the potential weakness of an entrant who offers incomplete services in the market. this particular characteristic of the regulated postal market, with universal service obligations, encourages customers to stay with their present provider. there is a set of elements of the postal network and some services to which access can present limitations. among others, the post office boxes, delivery boxes, postcode system, address database, information on change of address, redirection service and return to sender service. the member states shall ensure that transparent, non-discriminatory access conditions are available. access to these elements of the postal infrastructure or services represents real switching costs. the fourth section addresses each and solutions to reduce costs are proposed. after this introduction, the paper is structured as follows: the second section present a summary of the general regulatory framework for the postal sector in the european union since 1997. the third section addresses the issue of the switching costs due to its potential impact on competition in the postal sector. in the fourth section, the authors analyze access to specific postal network elements, such as access to po (post office) boxes, access to postcodes, redirection of mail service, change of address, return to sender service, and access to letterboxes. for each of them the authors propose answers that can help mitigate the switching costs. in the fifth and sixth sections, in this order, the costs of information and search for available choices and termination of a long-term relationship with the incumbent are discussed. finally conclusions and references are presented. liberalization regulatory framework in the european postal sector three postal directives regulate the process of liberalization of the european postal market, as follows: the first postal directive is 97/67 ec stated that measures to ensure the gradual and controlled liberalization of the market and to secure a proper balance in doing so were necessary to guarantee the obligations and rights of universal service providers (usp), and the free provision of services in the postal sector. the universal service obligations consist of the permanent provision of a service of specified quality at all points in the national territory at affordable prices for all users. quality, ubiquity, permanence and affordability are the characteristics of ups7. to the extent necessary to ensure the maintenance of ups, the directive 7 for other features of the us see d 97/67, articles 5-6. european journal of government and economics 5(2) 107 (see article 7) creates an area reserved to universal service provider on a part of the postal activity that includes a high proportion of letters. the parameters in terms of weight and price were successively reduced until they disappeared in 2011. the second is the 2002/39 ec directive. according to this directive, member states may continue to reserve services to a universal service provider. at the same time it proposes, if appropriate, the date of 2009 for the full accomplishment of the postal internal market. the third is the 2008/6 ec directive. this directive then stated that an fmo would occur by 2009 (this actually occurred in 2011). it also removed the market share (the reserved area) that directive 97/67 ec had exclusively ascribed to usp8 and stated that the progressive and gradual opening of postal markets to competition had provided usp with sufficient time to implement the necessary modernization and restructuring measures required to ensure their long-term viability under new market conditions. the directive also required member states to adapt their regulatory systems to a more open environment, in order to improve social welfare (crew and kleindorfer 2006). each member state has adapted its internal regulations based on laws of their parliaments and provisions of the respective governments. in turn, the universal service providers adapted their organizational structures and operational frameworks. provider switching costs customer dissatisfaction diminishes a postal service provider’s customer base, forces the operator to rely on a more volatile customer mix, and erodes the firm’s reputation (levesque and mcdougall 1996). this is particularly true in network industries, as the postal service. while some customers take no action at all when dissatisfied, others may take action such complaining directly to the provider about the service or switch suppliers. customers can feel compelled to continue their relationship with a service provider due to the user’s perceptions of high switching costs (porter 1980) even if the relationship is not a satisfactory one. the may incur three types of costs if they switch: i) financial costs derived from the switching process; ii) information and search costs on available alternatives, and iii) the termination cost of ending a long relationship with a regular provider. switching costs are defined as the customer’s perceptions of the additional costs of terminating the present relationship and finding another provider (patterson and smith 2003: 108). switching costs constitute any factor that makes it difficult to or increases the price of a customer switching providers (valenzuela et al 2005: 243). when the customer leaves their usual provider, they incur two types of loss: i) the loss of the advantages the company had created with its marketing strategy of relationships producing social and economic benefits and empathy, and customization costs and ii) the assumption of direct financial, time and effort costs. these are the two types of positive and negative barriers to switching as stated in the literature on switching costs (jones et al 2007; valenzuela et al 2005). table 1 summarizes the switching cost classifications of burnham et al (2003) and jones et al (2002). the authors have adapted this table assuming that the customer may change the postal service provider. 8 an operator such as this will be a designated usp or an incumbent; this term has become widespread in regulatory literature. pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 108 table 1. classification of switching costs burnham, frel, mahajan (2003) examples financial costs loss of volume or length of relationship discounts (+)a, payments for breaking the contract with the present operator (-)b, cost of new contracts and guarantees (-), mail redirection costs (-), return to sender costs (-). process costs time cost to find a new postal operator (-), and learning effort to adapt to a new system of identification, containerization, deposit, collection and invoicing of the postal deliveries (-). costs to learn new routines reconfigure hardware and software to be compatible (-). relationship costs breakdown of the affective bonds created between the customer management and employees and the postal operator employees or loss of the brand relationship (+). loss of the positive effects on our brand generated by a job well done by the incumbent (+). costs to reestablish communication networks with other users (-) jones, mothersbaugh, beatty (2002) examples continuity costs loss of the profits guaranteed by the length of the relationship with the present postal provider: discounts in volume and/or the composition of deliveries destination (+), permanence (+), perception of a probable worsening of the postal service provided by the entrant operator (-), perception of the damage which a deficient postal service can cause to our postal communications with our own customers (-). risk of loss of customers, both current and potential (-). learning costs loss of searching time of other attractive operators (-), adaptation costs to new preparing, packing and deposit systems of postal deliveries (-), adaptation costs to new computing supports for the identification of the postal deliveries and management costs of the new operator virtual office (-). adaptation costs to the opening hours (-). adaptation costs in our resources to transport logistics of the new postal operator (-). sunk costs previous costs generated to establish the relationship with our present provider: acquisition of containerization, load and unload and delivery storage elements, computer programs for invoicing, documents adapted to the operator methodology (+). adaptation costs in transport elements (-). adaptation costs to loading and unloading systems of the new operator (-). other costs (-). (+) positive barrier to switching; (-) negative barrier to switching. source: burnham et al (2003) and jones et al (2002); modify and adapted by authors part of the economic literature assumes that switching costs are anti-competitive. klemplerer (1995) argues that switching costs generally raise prices and create deadweight losses of the usual kind in a closed oligopoly and may also discourage new entry and so further reduce the market’s competitiveness. because switching costs tend to reduce competition, firms may dissipate more social surplus in costly activities to create them. in the same vein, armstrong and sappington (2006: 350) state that reduce customer switching and search costs is the purpose of a policy of liberalization that “can stimulate vibrant, enduring competition that may ultimately substitute for regulatory oversight”9 insofar as consumers are able to identify and secure postal service provision by another provider, the usp or incumbent will be compelled to offer the lowest price and/or the highest quality, and competition will be reinforced. however, the presumption that switching costs are anti-competitive is misplaced as a general statement. fabra and garcía (2015), in an analysis based upon a continuous-time dynamic equilibrium model in which switching costs are independent and identically distributed across consumers and over time, find that when firms cannot price-discriminate between old and new consumers, the effects of switching costs on prices critically depends on the degree of market share asymmetries. in the case that the market shares become sufficiently symmetric, price competition turns fiercer, and in the long-run, switching cost have a procompetitive effect. however, this result can hardly be in the postal sector, where 9 see prado and pateiro (2010) for the application of policies to the postal sector as recommended by armstrong and sappington (2006). european journal of government and economics 5(2) 109 the usp that existed before the liberalization process have still retained a high market share, as indicated above. in response to the entry of competitors in this sector, the usp can adopt different strategies: (i) the incumbent tries to defend its absolute market power by creative renegotiation of terms of agreements with its customers or by offering price cuts in the competitive segment of the market,10 (ii) the incumbent intends to finance price cuts in the competitive area by increasing prices in the non-competitive area;11 (iii) advantages bound to exclusive provider contracts; (iv) a pricing policy in profitable areas to better combat cream skimming practices.12 the costs derived from such maintenance strategies can be considered as customer retention costs. thus, the usp will support retention costs in the same way that the customer or entrant support the switching costs. in the third section the authors analyze typical provider switching costs. a clear example of a provider switching cost in telephone service is the cost of not being able to transfer one’s current telephone number where portability13 is not guaranteed. ‘absent such number portability, a consumer might be reluctant to switch suppliers because the switch would require the consumer to inform all friends and associates of her new number or reprint business cards, for example’ (armstrong and sappington 2006: 351). shi et al, (2006) studied the effects of number portability in the cellular phone industry. viard (2007) studied the effect of portability on competition in the phone market. similarly, in the postal service, switching costs could be charged if each new entrant adopted a different postcode system, or if the incumbent and the competitor did not have some form of cooperation (a reciprocal access) for the delivery of mail to recipients that have signed a po box with any of the operators or for delivery in the same home delivery boxes. in the postal service there is a series of matters which would probably cause important switching costs, both in terms of financial resources and time spent. in addition to access to postcode databases or access to po boxes, there are other issues that can generate switching such as information on change of address, redirection of mail service, return to sender service or access to letterboxes. network and databases access this section analyzes access to some elements of the postal network, as follows: access to po boxes, access to postcodes, redirection of mail service, change of address, return to sender service, and access to letterboxes. access to po boxes delivery to po boxes is an alternative to home delivery. clients choose this delivery either because they want to have their mail delivered in the morning earlier than at the regular postman’s visit, or because they do not want it delivered at home for whatever reason. the rental of the box is charged to the client-recipient. 10 the sweden usp posten ab and other european usps followed this strategy at the beginning of the 1990s. many of these contracts were annulled or modified by competition courts. 11 this strategy can lead a dynamic process of graveyard spiral. see crew & kleindorfer (2005), prado & pateiro (2012). 12 cream skimming practices in the postal sector are common in profitable urban areas. 13 portability is the guarantee that the customer will keep the same telephone number when they switch service provider. pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 110 the address on the envelope specifies the name of the addressee and the number of the box, as well as the postcode to identify the delivery unit of destination. if the consignor changes the provider, the new operator needs to enter the incumbent´s premises in order to leave the deliveries into the box appointed for the addressee. the competitor cannot inject its mail anywhere in the entrance of the delivery unit: this would imply problems with regard to contracting and transferring responsibility. access for all competitors and the incumbent to the physical space of the po boxes is a complex issue. in france, political advisors expressed the idea about building a private corridor behind the po boxes locked with a key, which would be made available to all licensees as well as to the incumbent. this proposal was abandoned because of the great technical difficulty and prohibitive costs. therefore, the most practical solution is handing over the mail to the incumbent for loading the po boxes. figure 1 represents the case in which an entrant injects the postal items in the incumbent distribution center for delivery to po boxes. in this case the entrant does not install po boxes in their locale. figure 1. entrant access to p.o box of the incumbent source: the authors competitors can place po boxes in their offices in the same way as the incumbent, but there is no sense in addressees signing the po box service with each competitor. access to postcodes access to postcodes is not identified by most national regulatory authorities (nras) as a bottleneck issue calling for regulatory intervention. thanks to new technological means, the use of gps coordinates coding each address is probably a smarter way to proceed (fratini et al 2010). however, the basic discount appears when the clients agree to have to present their mail sorted by postcode (work-sharing discounts in the upstream access), allowing the incumbent to avoid sorting. if the client changes the service provider, he could obtain a similar work-sharing discount. but if the incumbent changes the postcode system, the client incurs switching cots because the adaptation of his sorting system by postcode is necessary. to defend the interests of customers and to strengthen long-term competition, the mandatory access to postcodes could be imposed in order to allow competitors to rightly deliver. it is a question of avoiding the postcode operator in each member state can freely modify the postcode system. in sweden, postcode changes are handled by a postcode council with representatives of posten ab, city mail, the confederation of private postal colection sortingadmission sorting transport transport distribution p.o box  delivery colection admission incumbent network entrant network access downstream e1 european journal of government and economics 5(2) 111 operators and government agencies. the regulator can put system changes on hold. in the uk, royal mail is committed to a policy of no change wherever possible. postcode changes are only made if it results in a major benefit to the service they provide and new postcodes work in tandem with the old ones for an overlapping 12-month period. in 2015 the partnership on transport and logistics companies working in spain submitted a report to the ministry for public works in order to modify the spanish postcode (from 5 digits to 8 digits). the companies argue the economic and environmental profits stemming from the new postcode. the switching costs in the postcode to users and operators must be weighed up against the expected profits. in any case, it is necessary that the postcode be unique. in that respect, the change of the code would be neutral in terms of switching costs. redirection of mail the information on redirection service is undoubtedly an important issue that can cause high switching costs if there is not a perfect coordination among operators. traditionally letters, postcards, direct mail were redirected at the request of the addressee, without costs. this situation has changed in many states where postal operators have established redirection services that allow redirecting postal items to any new national or international address. so, in spain, france and the uk, the forwarding is done at the request of the addressee for the time period stipulated after having paid the agreed price.14 if a competitor has no access to information about customers of the incumbent redirections, the switching costs are evident: the addressees would have to enter into a redirection contract with each operator. and the senders of the forwardings would lose some communications, which would have a negative effect on their business. just like in the case of the return to sender service, an additional problem arises when an entrant covering part of the territory may need to deliver mail outside its area (see figure 3). figure 2 represents an ideal redirection service. consider a postal item from a to b, addressee mr. x. the sender entrusts the postal item to any of the operators: incumbent, entrant 1 or entrant 2. if mr. x moves to the city c, with which of these operators would mr. x hire the redirection service? it no longer really makes sense that a client temporarily changing his address should be obliged to give the information to all licensees. the addressee (in our case mr. x) hires the redirection service with the redirection service center (rsc) shared by all operators, and he pays a fee for administrative costs. the rsc transmits this information to all providers in real time. the financial costs and time costs are lower for the user. in summary, if there is an rsc, the switching costs are reduced when a customer changes the supplier and thus competition is promoted. 14 spain: contracts: 1, 2 or 6 months: national: €33.28, €49.90, €66.09. international: €50.19, €75.65, €99.86. france: contracts 6 or 12 months: national €24.50, €44. international: €69, €124. uk: 3, 6, or 12 months from £24.99 for each last name (in all cases prices in 2015). pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 112 figure 2. centralized redirection service source: the authors. green=information redirection; red=postal delivery route; dashed line=redirection route. change of address access to the change of address is a more important issue than access to the postcode database for delivering mail (fratini et al 2010). when someone changes his address, the information is given to the incumbent, which has the most reliable database, when compared with banks, telecom operators or energy providers. in france, the information is centralized in a database that traces the old and the new addresses. if the new provider has not immediate access to the change of address database, the client will incur switching costs since he needs to communicate his change of address to each operator. as in the case of redirection service, it does not really make sense that a client should be obliged to hire the change of address service to give this information to all competitors. that is why sharing information on address changes could be required by law. in france, new entrants have a cnil15 authorization, on the same terms as la poste, to maintain a file containing both the old and new addresses. the following step is about stating the mandatory access to the change of address database. return to sender service consider a postal mailing deposited in a and destined to b, both within the territory of the entrant (figure 3). the sender printed the address c on the envelope for return. the incumbent obligation to provide access to a return to sender service reflects the idea that an entrant covering part of the territory may need to deliver mail outside its area depending on where the sender is located (c in our case). thus the entrant would need to use the incumbent’s network to return undeliverable mail. the directive 2008/6 ce says that where several universal service providers with regional postal networks exist, member states should also assess and, where necessary, ensure their interoperability in order to prevent impediments to the prompt transport of postal items. as the legal and market situation of these elements or services is different among the member states it is appropriate to only require member states to adopt an informed decision on the need, extent and choice of the regulatory instrument, including where appropriate on cost sharing. 15 the commission nationale de l’information et des libertés protects citizens’ personal information to prevent any general cross-filing of the population. european journal of government and economics 5(2) 113 this provision is without prejudice to the right of member states to adopt measures to ensure access to the postal network under conditions of transparency and nondiscrimination. figure 3. return service outside area of the entrant source: the authors cases like those in figure 3 are not very frequent, but a decision must be adopted. according to ecorys (2008) the main challenges for the nras include arranging interoperability in a multi-operator market. in france, uk, spain, and italy different procedures are in place. the more elegant solution has been adopted in sweden: the total cost of the return to sender service has been estimated and shared by all operators. thus, the interoperability of networks is secured for this service. we propose a solution in this area, in line with that adopted in sweden. the nra estimates the cost of return to sender service of postal items that are to be returned to locations outside the territory of the first operator, when he is working in a limited geographical area. this cost will be shared by each of the postal operators in proportion to their volume of mail. this solution reduces the three common types of switching costs: procedural, financial and relational. access to letterboxes the matter about the access to letterboxes depends on issues such as its owners, location and access to them from the outside of the buildings. in the most of countries such as spain, france, portugal, uk, etc., the residents are owners of the letterboxes. as regards its access, there is no problem when it is outside (this happens in portugal and single-family homes in many countries). a problem arises when the access is from the inside of the buildings or when the incumbent is their owner because it is necessary to authorize the access to the property. in some countries, the access is restricted to the incumbent by law (united states of america, or austria). in france, poland, hungary, slovakia, germany, inter alia, the access to the private mailbox has been subject to discussion, regulation and litigation. in most cases, the keys or codes of the entrance slots held by the incumbent are not shared with competitive operators for reasons of safety, privacy, exclusive property and liability. there have been many discussions regarding the ownership of the letterboxes. it can be argued that these infrastructure elements are not essential facilities or actual bottlenecks in the sense of stable entry barriers, but there may be common ground to assert that the competitive development of the market may be hindered if access to the mentioned elements is refused, according to plaut economics (2007: 16). to avoid switching costs such as those described above and to promote competition in the postal sector, directive 2008/6/ec states that whenever necessary to protect the interest of users and/or to promote effective competition, territory of the incumbent territory of the entrant a b c pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 114 and in the light of national conditions and national legislation, member states shall ensure that transparent, non-discriminatory access conditions are available to elements of postal infrastructure or services provided within the scope of the ups, such as delivery boxes, among others ecorys (2008) states that access to the postal infrastructure consists of access to, among others, the letterboxes of individual consumers and businesses, po boxes, address databases and postal codes. here we have another reason why sharing information for letter boxes access could be required by law, provided that security rules are complied with. this is our proposal for access to letterboxes. information costs and the search for available choices varela et al (2009) found that a lack of attractive choices in the market or perceiving the present provider as more attractive than others can reduce the likelihood of customer abandonment. when a customer is exposed to negative experiences with his regular provider, he leaves his provider and seeks a more suitable alternative (sharma and patterson 2000). however, if alternatives do not exist or are not well known, the customer maintains the relationship with the provider even though it is unsatisfactory. the search costs and the quality information provided by the firms is analyzed by ghosh and galbreth (2013). competition can compel providers to deliver high-quality products to consumers at low prices if consumers can easily identify and secure service from the firms that offer the best products at the lowest prices (armstrong and sappington 2006: 350). in this sense, liberalization policies and national regulatory authorities (nras) will have to ensure truthful, transparent and objective low-cost information so that consumers are aware of the available alternatives and, where possible, can switch providers. measures recommended on this point include i) fast access to information on the existence of competitors, including accessibility, schedules, prices and other supply conditions; ii) cost reductions for the consumer when he switches provider (i.e., to retain the same po box number and postal code); and iii) a reduction in asymmetric costs that the consumer incurs when he chooses different providers. directive 2008/6/ec highlights the importance of information about universal postal services as well as about the characteristics of specific services and their access. article 6 states that member states will adopt the necessary measures to ensure that users and postal service providers regularly receive up-to-date information and with sufficient accuracy from ups providers regarding their universal services. furthermore, special reference must be made to the access conditions regarding these services as well as to prices and quality levels. the information must be published in an appropriate manner. in the process of provider switching, the postal user makes a comparative analysis about the quality provided by the new supplier against the previous one. taking into account this matter, directive 97/67 ce states that the quality of service expected by users constitutes an essential aspect of the services provided; the evaluation standards for this quality of service and the levels of quality achieved must be published in the interest of users. although the legislation enforces the duty to provide information about ups by the usp, this measure would have to extend to every postal service provider, both within the scope of the universal service and including all other services. thus, it would ensure that the consumer can analyze and compare the conditions of all alternative services, including deliveries, prices, packaging, quality, claims, indemnification, schedules, access conditions, bonuses, and territorial scope of application. european journal of government and economics 5(2) 115 the termination of a long-term relationship with the incumbent some service providers have enjoyed their market power for many decades, operating as monopolies or with a high market share. the eu postal sector possessed monopolistic characteristics when the liberalization process began 15 years ago and even today most providers still operate above the market power threshold. the conventional workings of the postal sector have ensured relationships of extraordinary longevity. long links with a service provider result in a relationship that is more complex and produces larger commercial deals (reinartz and kumar 2003). long-term relationships contribute to shared values, identification between the parts, mutual information and certainty regarding behavior (palmatier et al 2006). in so far as the provider offers a quality service and satisfactorily remedies any failures, the length of service in the relationship reinforces brand loyalty. empirical studies, such as woistschläger et al (2011), reveal that satisfaction, economic switching barriers, social ties, and habits are the drivers of customer loyalty. empirical support for inter-temporal dependencies in brand choices, also known as state dependence effects, has been obtained in various contexts (seetharaman 2003; abramson et al 2000; seetharaman et al 1999). the role of the informational advantage is studied by villas-boas (2004; 2006). this advantage may work as a barrier to entry because consumers tend to be loyal to the pioneering brands. white and yanamandram (2006) explore the mediating effects of dependence and commitment on the relationship between switching costs and behavioral loyalty. loyalty to brand is considered one of the factors behind consumer reluctance to abandon the usp (jonsson and selander 2006). brand loyalty reduces the vulnerability of the dominant operator against competition and acts as an entry barrier. if entrants into the market offer greater quality services, loyalty delays customer decisions to separate from their present provider and, at the same time, the incumbent has time to introduce quality improvements oriented to maintain customer portfolios. therefore, brand loyalty constitutes an essential element of entry barriers in the postal sector. along the same lines, stigler (1968), dijl et al (2006), and mcafee et al (2004) maintained that economies of scale would constitute an entry barrier only if consumers were loyal to the incumbent. that is to say, only where economies of scale are accompanied by a strong inertia on the demand side one can speak of authentic entry barriers in the postal sector. thus, scale economies alone do not constitute an entry barrier. discomfort experienced by customers of a new supplier when adapting to the change are an unquantifiable cost that requires the estimator’s best judgment. while this is an important factor, it must not be overemphasized. conclusions a part of provider switching costs are related to the access to some incumbent infrastructures such as the postcode system, address database, post office boxes, delivery boxes, information on change of address, re-direction service and return to sender service. this provision shall be without prejudice to the right of member states to adopt measures to ensure access to the postal network under transparent, proportional and non-discriminatory conditions. directive 2008/6 states that member states shall ensure that transparent, non-discriminatory access conditions are available to these elements of postal infrastructure or services. parties can best define among themselves the most appropriate way to ensure interoperability. the examples in the postal sector show that ex ante soft regulation, where the nra intervenes only in the case of disputes, seems to be the emerging model. pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 116 we propose solutions that can help to reduce the problems of access to the postal network elements studied in this paper. such solutions contribute undoubtedly to strengthen competition in the european postal market. first, the access to postcodes is not identified by the nras as a bottleneck issue that calls for regulatory intervention. however, we assume that to defend the interest of the customers and to strengthen long-term competition, a mandatory access to postcodes could be imposed in order to allow competitors to deliver properly. second, with respect to the access to po boxes, we propose a downstream access to the distribution center of the incumbent operator, as shown in figure 1. in this case, the entrant does not install po boxes in their locale. the senders of postal items do not need to subscribe to po boxes in each operator, thereby reducing the switching cost. third, regarding the redirection of mail service and the change of address service, we propose an ideal redirection service, which consists in sharing the rsc by all operators, as it is shown in figure 2. the addressee hires the redirection service with the rsc and pays a fee for the administrative costs. given that the financial and time costs are lower for the user, the switching costs are reduced when a customer changes the supplier and thus competition is promoted. fourth, the solution proposed here for the return to sender service is in line with that adopted in sweden. the cost of return to sender service of postal items that have to be returned to locations outside the territory of the first operator (see figure 3), when this operator is working in a limited geographical area, will be shared by each of the postal operators in proportion to the volume of mail. this solution reduces the three common types of switching costs: procedural, financial and relational costs. fifth, regarding the access to letterboxes, it is difficult to reach an optimal solution because the access to this element depends on issues such as its owners, location and access to them from inside or outside the buildings. in this case, the optimal solution is to pass a law that forces the postal operators to share the information on access to letterboxes, provided that security rules are complied with. the proposed solutions reduce the switching costs and enhance the long term competition in the postal sector. regulation on upstream and downstream access to the postal network is not homogeneous across states and, moreover its development has not reached a sufficient level so far. competition in the postal sector has experienced a limited progress compared to the initial proposals. the incumbents still retain a high market share. the causes of limited competition in the postal market lie in the insufficient and overly slow access regulations, the evolution of communication technologies to the detriment of classical postal communications, the economic crisis, as well as switching costs to a lesser extent. references abramson, charles, rick l. andrews, imram s. currim and morgan jones (2000) ‘parameter bias from unobserved effects in the multinomial logit model of consumer choice’, journal of marketing research 37(4): 410–26. armstrong, mark s. and david e. sappington (2006) ‘regulation, competition and liberalization’, journal of economic literature 44(2): 326-66. burnham, thomas a., judy k. frels and vijay mahajan (2003) ‘consumer switching costs: a typology, antecedents, and consequences’, journal of the academy of marketing science 31(2): 109-26. european journal of government and economics 5(2) 117 crew, michael a. and paul r. kleindorfer (2005) ‘competition, universal service and the graveyard spiral’, in michael a. crew and paul r. kleindorfer (eds) regulatory and economic challenges in the postal and delivery sector. new york, ny: kluwer, pp. 1-30. crew, michael a. and paul r. kleindorfer (2006) ‘the welfare effects of entry and strategies for maintaining the uso in the postal sector’, in michael a. crew and paul r. kleindorfer (eds) progress toward liberalization of the postal and delivery sector. new york, ny: springer, pp. 3-22. dijl, paul, eric van damme and pierre larouche (2006). ‘regulating access to stimulate competition in postal markets? ’, in michael a. crew and paul r. kleindorfer (eds) progress toward liberalization of the postal and delivery sector. new york, ny: springer, pp. 153-62. deligonul, seyda, daekwan kim, anthony s. roath and erin cavusgil (2006) ‘the achilles’ heel of an enduring relationship: appropriation of rents between a manufacturer and its foreign distributor’, journal of business research 59(7): 802810. directive 97/67/ec on common rules for the development of the internal market of community postal services and the improvement of quality of service. official journal of the european communities 21.1.98. l15/14-25. directive 2002/39/ec amending directive 97/67/ec with regard to the further opening to competition of community postal services. official journal of the european communities 5.7.2002. l176/21-25. directive 2008/6/ec amending d/ec with regard to the full accomplishment of the internal market of community postal services. official journal of the european union 27.2.2008. l52/3-20. ecorys nederland bv (2008) ‘main developments in the postal sector (20062008) ’, report for the ec, september. fabra, natalia and alfredo garcía (2015) ‘dynamic price competition with switching costs’, dynamic games and applications 5(4): 540-567. fève, fréderique, jean-pierre florens, frank rodriguez and soterios soteri (2010) ‘forecasting mail volumes in an envolving markete invironment’, in michael a. crew and paul r. kleindorfer (eds) heightening competition in the postal and delivery sector. northampton, ma: edward elgar publishing limited, pp. 116-134. fratini, alessandra, bernard roy and joost vantomme (2010) ‘access to infrastructure and service elements in the postal sector’, in michael a. crew and paul r. kleindorfer (eds) heightening competition in the postal and delivery sector. northampton, ma: edward elgar publishing limited, pp. 22-37. ghosh, bikram and michael r. galbreth (2013) ‘the impact of consumer attentiveness and search costs on firm quality disclosure. a competitive analysis’, management science 59(11): 2604-2621. jones, michael a., david l. mothersbaugh and sharon e. beatty (2002) ‘why customers stay: measuring the underlying dimensions of services switching costs and managing their differential strategic outcomes’, journal of business research 55(6): 441-450. jones, michael a., kristy e. reynolds, david l. mothersbaugh and sharon e. beatty (2007) ‘the positive and negative effects of switching costs on relational outcomes’, journal of service research 9(4): 335-355. jonsson, per, and sten selander (2006) ‘the real graveyard spiral. experiences from the liberalized swedish postal market’, in michael a. crew and paul r. kleindorfer (eds) progress toward liberalization of the postal and delivery sector. new york, ny: springer, pp. 359-366. pateiro-rodríguez, prado-domínguez, garcía-iglesias, and barreiro-viñán ● switching costs in the european postal service 118 klemperer, paul (1995) ‘competition when consumers have switching cost: an overview with applications to industrial organization, macroeconomics, and international trade’, review of economic studies 62(4): 515-539. levesque terrence j., and gordon h. mcdougall (1996) ‘customer dissatisfaction: the relationship between types of problems and customer response’, canadian journal of administrative sciences 13(3): 264-76. martin vance l., chris paterson and jessie x. wang (2012) ‘forecasting letter volumes: augmenting econometric baseline projections’, in michael a. crew and paul r. kleindorfer (eds), multi-modal competition and the future of mail, edward elgar publishing limited, northampton, massachusetts, pp. 60-76. mcafee preston, hugo mialon and williams, michael a. (2004) ‘what is a barrier to entry?. american economic review 94(4):461-65. okholm h., winiarczyk m., möller a. and nielsen c. (2010) main developments in the postal sector (2008-2010), copenhagen economics. www.copenhageneconomics.com palmatier robert w., rajiv p. dant, dhruv grewal and kenneth r. evans (2006) ‘factors influencing the effectiveness of relationship marketing: a meta-analysis’, journal of marketing 70(4):136-53. pateiro rodríguez carlos and javier prado domínguez (2010) ‘un análisis de la transformación del servicio universal y el ámbito reservado en las políticas para la competencia del sector postal en la unión europea’, gestión y política pública 19, 187-237. pateiro rodríguez carlos, garcía iglesias jesús m. and josé m. barreiro viñán (2013a) ‘las dimensiones territoriales urbana e interurbana de la liberalización del sector postal y cuestiones de ineficiencia del acceso a la red’, urban public economics review 17, 78-105. pateiro rodríguez carlos, josé m. barreiro viñán and garcía iglesias jesús m. (2013b) ‘efectos de los costes de cambio y de retención sobre el bienestar social en el proceso de apertura del servicio postal europeo’, anales de economía aplicada 27, 209-230. prado domínguez, javier and carlos pateiro rodríguez (2012) ‘la prestación del servicio postal universal ¿conlleva un riesgo inexorable de quiebra?’, revista de ciencias sociales 83 (1):1-9. patterson paul g. and tasman smith (2003) ‘a cross-cultural study of witching barriers and propensity to stay with service providers’, journal of retailing 79(2): 107-20. plaut economics (2007) ‘opening postal markets in switzerland. monopolistic bottlenecks, technical co-ordination and social regulations’, olten, january. porter, michael e. (1980) competitive strategy: techniques for analysing industries and competitors, free press, new york. reinartz, werner j. and vinod kumar (2003) ‘the impact of customer characteristics on profitable lifetime duration’, journal of marketing 67(1):77-99. seetharaman, p.b. (2003) ‘probabilistic versus random-utility models of state dependence: an empirical comparison’, journal of research in marketing 20(1): 87-96. seetharaman, p.b., andrew ainslie and pradeep k. chintagunta (1999) ‘investigating household state dependence effects across categories’, journal of marketing research 36(4):488-500. european journal of government and economics 5(2) 119 sharma, neeru and paul g. patterson (2000) ‘switching costs, alternative attractiveness and experience as moderators of relationship commitment in professional, consumer services’, international journal of service industry management 11(5):470-90. shi, mengze, jeonweng chiang and byong-duk rhee (2006) ‘price competition with reduced consumer switching costs: the case of wireless number portability in the cellular phone industry’, management science 52(1), 27-38. stigler, george j. (1968) ‘barriers to entry, economies of scale, and firm size’, in george j. stigler (ed.) the organization of industry , homewood, il: irwin, pp. 6770. trinkner, urs and martin grossman (2006) ‘forecasting swiss mail demand’, michael a. crew and paul r. kleindorfer (eds), progress toward liberalization of the postal and delivery sector. springer, new york, pp. 267-80. valenzuela, fredy, david pearson and roger epworth (2005) ‘influence of switching barriers on service recovery evaluation’, journal of service research, special issue, december: 239-57. varela, concepción, rodolfo vázquez and victor iglesias (2009) ‘comportamiento de abandono de la relación de un cliente con la empresa en un contexto de fallo y recuperación del servicio’, cuadernos de economía y dirección de la empresa 12(40):143-70. viard, v. brian (2007) ‘do switching costs make markets more or less competitive? the case of 800-number portability’, rand journal of economics 38(1): 146-63. villas-boas, j. miguel (2004) ‘consumer learning, brand loyalty, and competition’, marketing science 23(1): 134-45. villas-boas, j. miguel (2006) ‘dynamic competition with experience goods’, journal of economics and management strategy 15(1):34-66. white, lesley and venkata k. yanamandram (2006) ‘exploring the switching costsbehavioural loyalty relationship amongst dissatisfied customers in the b2b services sector’, research online, university of wollongong. library. available at: http://ro.uow.edu.au/commpapers/235/ white, lesley and venkata k. yanamandram (2007) ‘a model of customer retention of dissatisfied business services customers’, managing service quality 17(3):298-316. woisetschläger, david m., patrick lentz and heiner evaschitzky (2011) ‘how habits, social ties, and economic switching barriers affect customer loyalty in contractual service setting’, journal of business research 64(8): 800-8. european journal of government and economics volume 5, number 2 (december 2016) issn: 2254-7088 78 editorial statement: the first five years of the european journal of government and economics diego varela, universitatea alexandru ioan cuza din iasi, romania giacomo benedetto, royal holloway, university of london, united kingdom jose manuel sanchez-santos, universidade da coruña, spain abstract this editorial statement reflects on the experience from the first five years of the european journal of government and economics and proposes some broad ideas about what we believe should be the future of the journal in the following years. the developments and ideas presented here are divided in three parts: achievements and difficulties of the past five-year period, the renewal of the editorial team, and new challenges for the future. keywords academic publishing; journal management. jel classification a12; a30; p27. european journal of government and economics 5(2) 79 this editorial statement reflects on our experience as editors of the european journal of government and economics during its first five years of existence. it also presents some broad ideas about what we believe should be the future of the journal in the following years. the exposition is divided in three parts: achievements and difficulties of the past five-year period, the renewal of the editorial team, and new challenges for the future of the journal. achievements and difficulties the first five years of the european journal of government and economics have been very intense ones in which we have made important achievements but also faced considerable difficulties. as far as the achievements are concerned, besides managing to regularly publish two issues per year, which is an achievement in itself, we have also managed to include the journal in some globally recognised academic indices, such as a more stringent directory of open access journals, econlit, latindex and scopus. currently our journal is fully indexed by these services, which has increased its standing vis-à-vis fellow academics and authorities in a number of countries. achievements have also come on a personal side for the editors, and include the recent achievement of two jean monnet chairs selected by the european commission to be co-funded by the erasmus+ programme of the european union, in european political economy and eu budget policy, respectively. we have also faced difficulties, some of which still persist to a greater or lesser extent. one of these unresolved issues has been the understaffing of the journal, which has meant that the editors have had to undertake ourselves many of the tasks usually reserved for specialists, such as website management, copyediting or galley formatting of papers. this not only means an overload for the editors that detracts from their editorial duties, but also a gap in quality with respect to other journals. a second difficulty that still persists is the lack of integration between the fields of government and economics. although we are increasingly receiving a number of articles in areas that interlink both fields of politics and economics, the truth is that in many cases both fields have remained as watertight compartments. the increased level of technical sophistication of papers means that many experts in government do not understand or simply are not interested in the papers in the field of economics, and vice versa. another issue is related to the vocation of the journal to expand to other countries, and in particular to contribute to the integration of former communist countries of eastern europe. the difficulty in this case comes from the lack of economic and social cohesion in europe, which translates into great differences as far as research quality is concerned. in practical terms, this means that the distance between authors and reviewers across regions is so great that it is rather difficult to find an able reviewer or to get a paper accepted, unless regions are kept segregated, which is the opposite to our initial intentions as editors. but all in all the balance is positive, the main benefit being the experience we have acquired through the management of the journal, the contacts we have made, and the cohesion we have developed among the members of the editorial team. renewal of the editorial team these initial five years would not be possible without the help of our editorial board, which has brought together more than 30 colleagues from a number of countries in europe and beyond. after five years of service, time has come to thank them expressly for their efforts. besides us, this editorial board has included the following members: varela, benedetto and sanchez-santos ● the first five years of ejge 80  manel antelo, universidad de santiago de compostela, spain  eduardo bandres, universidad de zaragoza, spain  michael w. bauer, humboldt university berlin, germany  peter benczur, european commission, joint research centre, ispra, italy  oscar calvo-gonzalez, world bank, washington dc, united states  jose ramon cancelo, universidade da coruña, spain  raul caruso, università cattolica del sacro cuore, milano, italy  luiz de mello, organisation for economic co-operation and development, france  michael ehrmann, european central bank, frankfurt, germany  fabio franchino, università degli studi di milano, italy  miriam hartlapp, wissenschaftszentrum berlin für sozialforschung, germany  bjorn hoyland, university of oslo, norway  raya kardasheva, kings college london, united kingdom  santiago lago-peñas, universidad de vigo, spain  thierry madies, université de fribourg, switzerland  diego martinez-lopez, universidad pablo de olavide, sevilla, spain  jorge martinez-vazquez, georgia state university, united states  jan-hinrik meyer-sahling, university of nottingham, united kingdom  carmen nastase, universitatea stefan cel mare din suceava, romania.  jorge onrubia, universidad complutense de madrid, spain  fabio padovano, università degli studi roma tre, italy  antti pajala, university of turku, finland  george poede, universitatea alexandru ioan cuza din iasi, romania  lucia quaglia, university of york, united kingdom  yvon rocaboy, université de rennes 1, france  andres rodriguez-pose, london school of economics and political science, united kingdom  fritz sager, universität bern, switzerland  roger scully, cardiff university, united kingdom  stephan stetter, universität der bundeswehr münchen, germany  virgil stoica, universitatea alexandru ioan cuza din iasi, romania  robert thomson, university of strathclyde, united kingdom  frantisek turnovec, univerzita karlova v praze, czech republic  richard c. whitaker, university of leicester, united kingdom after five years, a new cycle starts that will surely bring a renewed editorial team with new faces both among editors and editorial board members, a team that will be able to base its project on the work initiated by these colleagues to whom we hereby thank, but that will also bring new ideas and enthusiasm with them. new challenges for the future the challenges for the new editorial team are related to the difficulties that we have mentioned above, the first of which will be, in our opinion, the needed professionalization of the journal. assuring the specialisation of functions such as copyediting, proofreading or formatting will allow editors to focus on their own functions and increase the quality of the journal. it is an open question whether this can be better undertaken at home or by means of outsourcing, and whether the european journal of government and economics 5(2) 81 funding for such improvements may come from subscription fees, author publishing fees, or some kind of institutional sponsorship. other unresolved issues that have been mentioned above are the integration of the fields of government and economics, and the integration of academics from different regions of the world with different academic backgrounds. finally, and most importantly, the new editorial team will need to continue our efforts to increase the quality of published papers. this, in turn, will allow the journal to rise in citation rankings and, ultimately, to be admitted to additional indices such as the journal of citation reports, and further increase in recognition and influence (varela, 2013). all these challenges are difficult ones in themselves, but especially if we consider that they interact among them. for instance, rising in established citation rankings may conflict with the objective to open up the journal to new countries and regions. institutional sponsorship arrangements needed for the professionalisation of the journal may also conflict with the journal’s autonomy. all in all, these are just a few of the open questions that will make the next cycle a challenging one for the editorial team of the european journal of government and economics. references varela, diego (2013) 'the contribution of isi indexing to a paper's citations: results of a natural experiment', european political science 12(2): 245-53. sources of economic fluctuations in france: a structural var model european journal of government and economics volume 1, number 1 (june 2012) issn: 2254-7088 66 sources of economic fluctuations in france: a structural var model nabil ben-arfa, university of nice sophia antipolis, france* abstract this paper studies the economic fluctuations of an open economy such as the french economy. a system of variables containing output, price level, trade balance, real exchange rate and oil prices is analyzed by applying the structural vector autoregressive (svar) methodology initiated by sims (1980). this set of variables allows to evaluate the main sources of impulses of the french economy fluctuations. the results show that five structural shocks are identified using the long-run constraints implemented by blanchard and quah (1989). from the svar dynamic properties, impulse response functions and variance decomposition, the french economy is shown to be particularly vulnerable to supply and oil price shocks, where these two shocks respectively contribute to 40% and 35% of the economic disturbance. france is also hit by important external shocks which damage its trade balance position. finally, it is found that shocks related to economic policy (demand shocks) have a quite limited impact on the economic activity. jel classification e32; f41; c22 keywords economic fluctuations; external shocks; internal shocks; oil price shock; svar model * address for correspondence: nabil ben arfa, cemafi, university of nice sophia-antipolis, résidence les imperators, immeuble le constantin a, chemin de la lauve, 83700 saint raphaël, france. e-mail : nabil_ar@yahoo.fr. doi: https://doi.org/10.17979/ejge.2012.1.1.4277 european journal of government and economics 1(1) 67 introduction the 1980s have operated a methodological and a theoretical revival on the economic fluctuations analysis. the aim of this paper is to deal with empirical treatment of economic disturbances. sims (1980) was the pioneer of the fluctuations analysis within the vectorial autoregressive model, where impulses are apprehended as innovations in a statistical term. these var models were introduced as an alternative to the traditional econometric models. sims proposed a new form of modeling based on no a priori and where no distinction is made between exogenous and endogenous variables. since pioneer work of sims (1980), the main empirical work dealing with the sources of economic fluctuations lay on autoregressive vectorial model. these canonical var models however posed some problems related to the shocks identification, they faced a lot of criticisms, qualifying them as “atheoretical” models. these criticisms lead to the birth of structural var models, models in which shocks identification is conducted by the imposition of constraints drawn from economic theory. it is this methodology of structural var which will be applied to the french economy. we will apply a structural var model to the french economy in order to identify the main shocks which are the origin of the economic activity fluctuations. in the second section we will reconsider the theoretical and the methodological revival of fluctuations analysis. in the third and the fourth part of the paper we explore the data used and estimate the structural var model. finally, we interpret the results. the methodological and theoretical revival of the fluctuation analysis: the 1980s has operated a methodological and a theoretical revival on the analysis of economic fluctuations. the methodological revival was initiated by sims (1980); it was inscribed on the line of the impulse-propagation approach suggested by frisch (1933) and slutsky (1927). on a theoretical level, real business cycle theory constitutes a true theoretical revival on the fluctuations analysis: it proposes to explain the main part of the economic fluctuations within the neo-classic growth model disturbed only by shocks affecting the total factor productivity. this marks the abandonment of the debate on the relative importance of monetary versus fiscal shocks. the debate on the relative importance of supply and demand shocks emerges. real business cycle theory at the beginning of the eighties, the relevance of the equilibrium monetary theory was rejected in a theoretical as in an empirical level. it is in this context that appears the real business cycles theory, or rbc1, with the pioneers’ models of kydland and prescott (1982) and long and plosser (1983) in closed economy. the real business cycle theory considers economic fluctuations as the optimal response of economic agents to shocks on the total factor productivity. the models of real business cycle thus conceive the evolution of economic aggregates as the decision result of a great number of agents seeking to maximize their utility and only constrained by technological resource. the real business cycle theory attributes an insignificant role, even no role, to the monetary policy 1 for real business cycles. european journal of government and economics 1(1) 68 these basic models were followed by many extensions: extensions to open economies, with the international real business cycle of backus, kehoe and kydland (1992, 1994, and 1995). extensions to others shock in addition to the technological shock, by borrowing theoretical assumptions from the keynesian theory. criticisms addressed to the basic real business cycle models lead to the development of an abundant literature, with increasingly sophisticated models. results of these developments were not always satisfactory especially concerning the reproduction of the stylized facts. the methodological contribution of the real business cycles theory is however admitted by a large part of economists. parallel to this movement within the real business cycle theory, a new school of thought was emerging; it is the new keynesian macroeconomics. the new keynesian (nk) shares with the partisans of the real business cycle theory the fact that macroeconomic requires more microeconomic bases. however, nk economists believe that market imperfections are the key to understanding the real-world. the introduction of nk ideas into rbc models seems to make results definitely more satisfactory, in the sense that these models are accepted by economics profession and that their empirical results are more realistic. the introduction of the prices rigidity was sufficient to join again with the monetary policy, neutral and without effect in basic rbc models. some economists saw in this “marriage” between rbc and keynesian, the birth of “the new neo-classical synthesis” (goodfriend and king 1997). nowadays, macroeconomic models incorporate the principal theoretical elements of rbc models. they adopt their general structure; seek to identify the impulses response function of agent in a general equilibrium structure. on the other hand, the way in which the models define and identify the cycles is substantially different from the original contributions, various types of imperfections and rigidities are introduced. these imperfections proposed by new keynesian are related to the imperfect nature of competition on goods market, the specificity of financial market exchange, etc. during last decades, rbc initials disappeared gradually and those of dsge appear (dynamic stochastic general equilibrium). the methodological revival: var model sims (1980) proposed a tool for fluctuations analysis based on impulses, defined as statistical innovations. since sims contribution of 1980, the mains empirical work on the economic fluctuations sources lay on autoregressive vectorial methodology. the purpose of sims consists in evaluating the contribution of various innovations of a system to the dynamics of each variable. to distinguish the impulses response from the propagation mechanisms, he proposes the choleski method of orthogonalization. following criticisms and in particular those concerning the impossible interpretation of shocks economically through the choleski decomposition, many authors suggest to base the orthogonalization of shocks on structural model of innovations, the structural var model. shapiro and watson (1988), blanchard and quah (1989) and gali (1992), proposed to identify structural impulses, which are interpretable economically: supply shocks, demand, economic policy…their methods of identification are based on restrictions drawn from the economic theory. from an econometric point of view the structural impulses are estimated as a function of the canonical innovations, obeying to constraints resulting from the economic theory. the imposed restrictions can be of different kind and their economic implications european journal of government and economics 1(1) 69 diametrically opposite. one distinguishes two types of restrictions used in the recent literature: short term restrictions and long run restrictions. the short run constraints relate to the instantaneous answers of variable to shocks. long run restrictions are related to the long term shocks responses. those developments make the birth of the structural var model, i.e var models where it is possible to give economic definition to various shocks. data characteristics and var model estimation the purpose of this section is to analyze the economic disturbances in an open economy, the french economy. empirical study presented in this section is based on the var methodology incited by sims (1980). these recent developments on time series econometrics are applied to a system of variable including output, prices, trade balance, real exchange rate and oil price. this system of variables makes possible the evaluation of the main source of disturbance in the french economy. so, in an autoregressive vectorial model including these variables, five structural shocks are identified with the help of the blanchard and quah (1989) method of decomposition. long term characteristics of the data before the model estimation, we must preliminary check the order of integration and test the possible presence of cointegration relationship between variables. we use quarterly data extending from 1978q1 to 2007q4:2 y: gdp logarithm p: logarithm of consumer price index se : logarithm of trade balance tc: logarithm of the real effective exchange rate pp: logarithm of the oil price tests of stationarity to analyze the long-term properties of the data, we use three different methods: augmented dickey-fuller test (1979), phillips-perron test (1988) and kwiatkowskiphillips-schmidt-shin test (1992). 2 our data comes from the insee (institut national de la statistique et des etudes economiques) database and the ifs (international financial statistics) from the imf (international monetary fund). q for quarter. all series are seasonally adjusted. the oil price is in us dollars. the real effective exchange rate is computed with ulcs. european journal of government and economics 1(1) 70 table 1. unit root tests augmented dickey fuller (adf) phillips-perron (pp) kwiatkowski-phillips-schmidt-shin (kpss) variables statistics of the test critical value (5%) statistics of the test critical value (5%) statistics of the test critical value (5%) gdp 5.63 -1.95 13.17 -1.95 1.30 0.46 ∆ gdp -4.92 -2.89* -7.55 -2.89* 0.05 0.46 trade balance -0.17 1.95 -0.03 1.95 0.60 0.46 ∆ (trade balance) -5.35 -1.95 -10.60 -1.95 0.19 0.46 real foreign exchange rate -1.10 -1.95 -1.10 -1.95 1.20 0.46 ∆ (real foreign exchange rate) -9.66 -1.95 -9.66 -1.95 0.07 0.46 cpi -0.05 -1.95 -0.05 -1.95 0.54 0.46 ∆ (cpi) -8.72 -1.95 -8.72 -1.95 0.05 0.46 the oil price 1.05 -1.95 1.01 -1.95 0.87 0.46 ∆ (of the oil price) -8.57 -1.95 -8.60 -1.95 0.15 0.46 notes: * this critical value is relating to the model with constant and without trend. the character ∆, indicates the first difference of the variable. all the variables are in logarithm. according to unit root tests, it appears that all the variables of the model are nonstationary; they are integrated of order one. cointegration relationship to test the possible existence of cointegration between variables, we use the test implemented by johansen (1991) and johansen and juselius (1990).3 so we suppose the vector x of dimension (5×1): x =  pptcsepy ,,,, the general representation of the model in vecm4 form is given by the following expression:  tx = c + ttptpt xxx   11111 ... where matrices i (i = 1,…, p) are of size (n×n). the method suggested by johansen and juselius is based on two assumptions: on one hand the vector x must be i (1) and in addition the vector of the residual  must be a white noise. the strategy of the test consists in analyzing the rank of 3 the advantage of this method is that it allows for the identification of multiple cointegrating vectors. the engle-granger cointegration methodology (engle and granger, 1987) is limited to testing only for one cointegrating vector. 4 vecm for vector error correction model. european journal of government and economics 1(1) 71 the matrix . if the rank of  is zero then there is no cointegration between the variables. if the rank of the matrix  is r, there exist two matrices of dimension (n×r),  and  as:  =  ' ' is a matrix which contains the r vectors of cointegration.  is a matrix which contains the weights associated to each vector of cointegration. to determine the number of vectors of cointegration r, johansen proposes two statistics: the trace test and the maximum eigenvalue test. the trace statistic is the following: tr = t )1log( 1     n qi i the ho hypothesis is: r ≤ q, i.e. there are at least r vectors of cointegration. this test is equivalent to test the rank of the matrix  since testing the existence of r vectors is equivalent to test the following null assumption: rg ( ) = r three cases can be presented: rg ( ) = 0, this means that r = 0: there is no cointegration, in other words tx is integrated of order 1 but not cointegrated. it is then possible to estimate a var model on tx . rg ( ) = r, with r<n: tx is cointegrated with r rank, thus r relations of cointegration exist. a vecm can be estimated. rg ( ) = n, in this case the matrix  is full rank, tx is stationary, and there is no cointegration. a var model can be estimated directly on tx . the maximum eigenvalue test statistic is: max 1 log(1 ) q vp t       the ho hypothesis is r= q, the alternative one is r= q+1. the results of these tests are conditional to the var estimation, and consequently to the lag choice. we have used aic criterion, the selected lag number is equal to 3. we carry out tests by supposing the absence of trend and constant in the relationship of cointegration and in the vecm model. european journal of government and economics 1(1) 72 table 2. cointegration test: trace test assumption eigenvalue trace statistic critical value: 5% probability r = 0 0.193010 48.02 69.81 0.7207 r ≤ 1 0.082414 23.15 47.85 0.9585 r ≤ 2 0.075934 13.17 29.79 0.8835 r ≤ 3 0.024574 4.01 15.49 0.9024 r≤4 0.009677 1.12 3.84 0.2882 according to the results appearing in the table 2, we accept the null hypothesis of absence of cointegration (48.02< 69.81). max-eigenvalue test (table 3) also indicates no cointegration at the 0.05 level. table 3. cointegration test: maximum eigenvalue test assumption eigenvalue max-eigen statistic critical value: 5% probability r = 0 0.193010 24.87 33.87 0.3935 r ≤ 1 0.082414 9.97 27.58 0.9852 r ≤ 2 0.075934 9.16 21.13 0.8193 r ≤ 3 0.024574 2.88 14.26 0.9540 r≤4 0.009677 1.12 3.841 0.2882 finally, the various tests carried out, we conclude by the rejection of the stationarity and cointegration of the five series in level, so we specify a model in first difference. the var model will be consequently built on the growth rate of gdp ( y ), on the first difference of real effective exchange rate ( )tc , on the first difference of trade balance rate ( se ), on the inflation rate ( p ) and on the first difference of oil price (∆pp). var model identification basing on results found below, the var model in matrix form is the following: tx = tit p i i xa    1 tx = (∆pp, se , y , tc , p ,) ' , the column vector of the explained variables which depends on its p lags. ia , are the square matrices of the coefficients to be estimated. t , is the vector of the residuals. t = ( pp , se , y , tc , p ,) ' represents at each moment t, the value of tx which is not explained by its past. these residuals are also regarded as innovations or impulses. from the vector of variables tx , we will define five shocks: a supply shock, a real demand shock, a nominal demand shock, an external shock (shock on the trade balance) and an oil price shock. we consider the residual resulting from the first equation as an oil price shock, the one resulting from the second equation as an external shock, that resulting from european journal of government and economics 1(1) 73 the gdp equation as a supply shock and finally residuals resulting from the fourth and the fifth equation are supposed to be demand shocks (real and nominal). however, this definition of the shocks is likely to be false since the residuals resulting from a canonical var model are often correlated. consequently, we will adopt the structural var method in which shocks identification is based on constraints resulting from the economic theory. identifying constraints we adopt the structural var methodology developed by blanchard and watson (1986), bernanke (1986), shapiro and watson (1988), blanchard and quah (1989). two types of restrictions are mentioned in the literature; short-run restrictions and long-run restrictions. to identify the structural shocks, we choose the identification of blanchard and quah (1989) imposing long run constraints.5 in other words we will constrain certain variables not to have long-term effects. we attempt to estimate the following structural var model:                      p tc y se pp = )(ld                        p tc y se pp      the left part of this equality describes the vector of variables entering into the var system. the right-hand side, is a mixture of the structural shocks, (exogenous forces of the system), and of the matrix d (l) which describes the coefficients associated to these shocks. the identification of the structural shocks is done by imposing long run constraints using the matrix d (1). the long-term answers to shocks are defined by this matrix: )1(d =                 )1()1()1()1()1( )1()1()1()1()1( )1()1()1()1()1( )1()1()1()1()1( )1()1()1()1()1( 5554535251 4544434241 3534333231 2524232221 1514131211 ddddd ddddd ddddd ddddd ddddd the identification of shocks in a system of 5 variables requires 10 constraints: first constraints are drawn from the open economy assumption. it results from this hypothesis that domestic shocks do not affect the variables generating the external shocks: 0)1()1()1()1()1()1( 252423151413  dddddd . we can then identify 6 constraints. 5 our choice is justified; this kind of restriction is applied only in the case of integrated variable which is the case of our model. european journal of government and economics 1(1) 74 following constraints are drawn from the theoretical assumption commonly accepted since the work of blanchard and quah (1989). it is about the distinction between supply and demand shocks. indeed, economic theory supposes that supply shocks can affect economic activity on the long term. whereas demand shocks affect economic activity only on the short term. it results two additional constraints: 0)1()1( 3534  dd concerning the constraints that allow the distinction between the two demand shocks, real demand shock (generally assimilated to a fiscal shock or a foreign exchange rate shock) and the nominal demand shock (monetary shock for example). a real demand shock is supposed to have an effect on real foreign exchange rate whereas a nominal shock does not affect the foreign exchange rate. it results an additional constraint: .0)1(45 d finally the last constraint is related to the oil price: an oil price shock is supposed to be the only shock which can have long-term effects on the oil price. this assumption gives us a last constraint: 6 . thus the matrix representing the effects of structural shocks on the model variables is the following one: )1(d =                 )1()1()1()1()1( 0)1()1()1()1( 00)1()1()1( 000)1()1( 0000)1( 5554535251 44434241 333231 2221 11 ddddd dddd ddd dd d after the constraints identification, the estimation of the var model requires the determination of the lag. we use the criterion test of information. lag selection criteria the results of aic, sc and hq are exposed in the following table. table 3. lag selection criteria of the var model lag logl lr fpe aic sc hq 0 1497.285 na 1.45e-18 -26.88801 -26.76596* -26.83850* 1 1531.364 64.47411 1.23e-18 -27.05160 -26.31930 -26.75453 2 1559.340 50.40804* 1.17e-18* -27.10523* -25.76267 -26.56059 3 1570.737 19.50838 1.50e-18 -26.86013 -24.90732 -26.06793 4 1589.531 30.47622 1.70e-18 -26.74831 -24.18524 -25.70855 * indicates lag order selected by the criterion 6 this assumption allows the identification of these four constraints: 0)1()1()1()1( 15141312  dddd . we take the first one, 0)1(12 d , since the three remaining ones were already taken into account using the assumption of independence between internal shocks and external shocks. 0 ) 1( 12  d european journal of government and economics 1(1) 75 the results of aic tests, sc and hq, give different conclusions: the order of the var model is equal to 1 according to criteria sc and hq, it is equal to 3 according to the criterion of aic, fpe and lr the majority of lag length criteria suggest the use of a lag of 3 in the analysis. we then take p = 3. estimation of the structural var model: impulse response functions and variance decomposition impulse response functions and variance decomposition svar model estimation is conducted by estimating the ordinary var model7, then we apply the long term identifying constraints previously identified, and finally we determinate the structural shocks, their impacts and their contributions to the french economy fluctuations. sources of growth rate fluctuations chart 1 represents the response impulse functions of growth rate to the various shocks. it shows the effects of the five shocks on the growth rate. we notice a positive and significantly persistent effect of supply shocks in short-term us in the long run. however, the real and nominal demand shocks affect the economic activity only transitorily. 7 main validation tests of the var model are reported in the appendix. european journal of government and economics 1(1) 76 chart 1. impulse response functions of the growth rate -.006 -.004 -.002 .000 .002 .004 .006 .008 5 10 15 20 25 30 response of the growth rate to external shock -.006 -.004 -.002 .000 .002 .004 .006 .008 5 10 15 20 25 30 response of the growth rate to supply shock -.006 -.004 -.002 .000 .002 .004 .006 .008 5 10 15 20 25 30 response of the growth rate to real demand shock -.006 -.004 -.002 .000 .002 .004 .006 .008 5 10 15 20 25 30 response of the growth rate to nominal demand shock -.0002 .0000 .0002 .0004 .0006 .0008 .0010 .0012 5 10 15 20 25 30 response ot the growth rate to oil price shock indeed, as it was predicted by economic theory, a positive supply shock involves an improvement of the activity level, this improvement remains rather bearable in long-term. however, demand shocks have insignificant impact, their impacts tend to be equal to zero in long term. concerning the external shocks, the impulse response functions show a negative effect of these shocks on the trade balance. this negative effect can be explained by the degradation of price competitiveness of france resulting from the progressive appreciation of the euro compared to the dollar, which leads to a fall in exports and to an economic activity decrease. this degradation can also be explained by the rise of the oil price or other reasons. moreover and basing on the model results we notice the importance of the oil price effect on economic activity. the gdp impulse response function to shocks shows an important effect of oil price shock on the activity level. the growth rate responds negatively to oil price shock. the initial response is the largest one; in long run the european journal of government and economics 1(1) 77 effect tends to be insignificant. jalles (2009) adopts several oil price specifications and also found a significant impact of oil price shock on french economic activity. table 4. variance decomposition of the growth rate quarters oil price shock external shock supply shock real demand shock nominal demand shock 1 42.28251 22.08729 21.31262 2.140815 12.17676 4 39.02417 15.70284 34.68763 2.843572 7.741792 8 35.08653 14.36379 41.12939 2.342595 7.077686 12 34.60134 13.57045 43.22037 2.148490 6.459353 16 34.42554 13.13336 44.23477 2.048563 6.157772 20 34.33957 12.85764 44.77958 1.996534 6.026670 24 34.29383 12.67877 45.06698 1.968955 5.991462 28 34.25937 12.56120 45.22237 1.954872 6.002194 30 34.24477 12.51787 45.27105 1.950650 6.015659 the table above reveals the contribution of each shock to the growth rate fluctuations. we notice the prevalence of supply and oil price shocks in the explanation of the growth rate dynamics. indeed, whatever the chosen horizon, short or long run, oil price shock explains between 40% and 35% of the activity variability, the supply shock explains between 21 % and 45% of the variability. france, as an importer of oil energy, is particularly vulnerable to oil price shock. concerning demand shocks, it arises from the variance decomposition, that these shocks have a limited contribution on the long-term economic activity. this consolidates our theoretical assumptions, stipulating that demand shocks do not have a permanent effect on the gdp. demand shocks (real or nominal) contribute to less than 6 % in the economic activity variability. in addition, the introduction of the assumption of an open economy into our model enables us to evaluate the contribution of external shocks apprehended in this case by shocks on the trade balance. this shock has a considerable effect on the economic activity; its contribution turns around 12% in the long term and exceeds 22% in the short term. to summarize, the variance decomposition of gdp shows a prevalence of supply and oil price shocks in the economic fluctuation explanation. sources of price fluctuations chart 2 indicates a positive impact of nominal demand shocks. these nominal demand shocks reflect the evolution of the money supply and highlight the narrow correlation between the price level and the monetary aggregates. chart 2 highlights the fact that a nominal demand shock leads to an increase in the general prices level. the raising of prices remains constant on the long term. in addition, the impulse response functions reveal the importance of supply shocks contribution in the variation of the price level. a positive supply shock (technological shock for example) will make the production more efficient, leads to an increase in output and thus lower the price level as it shown in the chart below. the effects of supply shocks are also more important than those of the real demand shocks. this result can be explained by the limited effects of the french fiscal policy since the european pact of stability constraints the budget deficit to only 3% of the gdp. european journal of government and economics 1(1) 78 chart 2. impulse response functions of the inflation rate -.016 -.012 -.008 -.004 .000 .004 .008 .012 .016 5 10 15 20 25 30 response of inflation rate to oil price -.016 -.012 -.008 -.004 .000 .004 .008 .012 .016 5 10 15 20 25 30 response of inflation rate to external shock -.016 -.012 -.008 -.004 .000 .004 .008 .012 .016 5 10 15 20 25 30 response of inflation rate to supply shock -.016 -.012 -.008 -.004 .000 .004 .008 .012 .016 5 10 15 20 25 30 response of inflation rate to real demand shock -.016 -.012 -.008 -.004 .000 .004 .008 .012 .016 5 10 15 20 25 30 response of inflation rate to nominal demand shock european journal of government and economics 1(1) 79 table 5. variance decomposition of the inflation rate quarters oil price shock external shock supply shock real demand shock nominal demand shock 1 0.013192 1.025571 58.29633 14.97440 25.69051 4 4.960914 3.455207 48.21400 11.05525 32.31463 8 7.848685 5.708743 38.26513 8.535835 39.64161 12 10.09025 5.150766 36.00894 7.568884 41.18117 16 11.47424 4.805916 35.80208 7.048559 40.86920 20 12.38600 4.537079 36.00211 6.705005 40.36980 24 13.03210 4.340852 36.28109 6.471194 39.87476 28 13.48659 4.200097 36.54344 6.309716 39.46016 30 13.66247 4.145376 36.65760 6.247748 39.28680 the variance decomposition of inflation rate shows a very substantial contribution of impulses conducted by the economic policy. the contribution is more important for nominal demand shocks, about 40%. this prevalence of nominal demand shocks remains stable whatever the time horizon. in addition, as we have notice when dealing with impulse functions, supply shocks contribute to a significant part on inflation rate variation, a contribution which turns around 35%. finally, considering the importance of the french economy openness degree, it would be interesting to highlight the external shocks impact on the price level. through the variance decomposition, we notice a very weak contribution of the trade balance shock on prices disturbances. this contribution is approximately about 4% in the long term, less than 1% in the short term. we also notice that an oil price shock is accompanied by an increase in the price level essentially in long term. in the short term the relative contribution of the oil price shock is almost equal to zero, however in the long term it turns around 13%. this increase (even if it remains controlled enough) cuts down the purchasing power of households, decreases consumption and consequently the growth rate. the oil prices rise of these last years, combined with an unstable geopolitical environment, seems to be durable not temporally. one can expect a continuous rise of oil prices; the economic policies should take into account this new international evidence by defining more rigorous energetic policies. european journal of government and economics 1(1) 80 sources of foreign exchange rate fluctuations chart 3. impulse response functions of the foreign exchange rate -.015 -.010 -.005 .000 .005 .010 .015 5 10 15 20 25 30 response of the foreign exchange rate to oil price shock -.015 -.010 -.005 .000 .005 .010 .015 5 10 15 20 25 30 response of the foreign exchange rate to external shock -.015 -.010 -.005 .000 .005 .010 .015 5 10 15 20 25 30 response of the foreign exchange rate to supply shock -.015 -.010 -.005 .000 .005 .010 .015 5 10 15 20 25 30 response of the foreign exchange rate to real demand shock -.015 -.010 -.005 .000 .005 .010 .015 5 10 15 20 25 30 response of the foreign exchange rate to nominal demand shock european journal of government and economics 1(1) 81 table 6. the variance decomposition of the foreign exchange rate quarters oil price shock external shock supply shock real demand shock nominal demand shock 1 4.913600 8.474808 36.02216 50.33014 0.259295 4 6.114626 10.06832 34.76513 47.35129 1.700633 8 6.894016 11.43277 33.69233 45.72826 2.252634 12 7.018748 11.34274 33.99508 45.26362 2.379807 16 7.080580 11.33318 34.02144 45.18350 2.381303 20 7.098915 11.32942 34.04914 45.14051 2.382014 24 7.111069 11.32563 34.06474 45.11766 2.380897 28 7.117800 11.32325 34.07302 45.10537 2.380558 30 7.120032 11.32233 34.07584 45.10117 2.380622 from the table above, we notice a prevalence of domestic shocks in the explanation of exchange foreign rate fluctuations; particularly the real demand shock, which is assimilated in our model to a fiscal shock or to an adjustment of foreign exchange rate. indeed, they contribute to approximately 50 % of the foreign exchange rate disturbance. concerning the nature of the impact, theoretically a real demand shock due to the aggravation of budget deficit involves an appreciation of the real foreign exchange rate and a deterioration of the external position. our empirical results are in conformity with this theoretical assumption. impulses response functions show an appreciation of the real foreign exchange rate when the economy is hit by a real demand shock which is probably caused by the overvaluation of the euro/dollar exchange rate since the fiscal policy in france is relatively controlled (the european pact of stability). concerning the remainder of domestic shocks, we notice a quite high contribution of supply shocks and a very limited contribution of nominal demand shocks; they respectively contribute to 36% and 2% of disturbances. impulses response functions reveal that the effects of nominal demand shocks are unstable and close to zero. in other hand a positive supply shocks involves a depreciation of the real foreign exchange rate. finally concerning the contribution of external and oil price shocks to foreign exchange rate fluctuations; it appears from the variance decomposition that these two shocks have a significant contribution but still lower in comparison to internal shocks. indeed, the external shocks contribute to approximately 11% of the foreign exchange rate variability, oil price shocks contribute to 7%. taking into account these results, we can conclude that foreign exchange rate disturbance is due primarily to real demand shocks; they contribute to approximately 50%. european journal of government and economics 1(1) 82 sources of trade balance fluctuations chart 4. impulse response functions of the trade balance -.004 .000 .004 .008 .012 .016 .020 .024 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 response of the trade balance to oil price shock -.004 .000 .004 .008 .012 .016 .020 .024 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 response of the trade balance to external shock -.004 .000 .004 .008 .012 .016 .020 .024 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 response of the trade balance to supply shock -.004 .000 .004 .008 .012 .016 .020 .024 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 response of the trade balance to real demand shock -.004 .000 .004 .008 .012 .016 .020 .024 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 response of the trade balance to nominal demand shock european journal of government and economics 1(1) 83 table 7. variance decomposition of the trade balance quarters oil price shock external shock supply shock real demand shock nominal demand shock 1 0.482441 50.32980 6.364164 7.407492 35.41610 4 4.023561 46.26756 7.478640 10.91904 31.31119 8 6.074570 42.74174 8.056011 11.02374 32.10393 12 6.318011 42.39966 8.081979 11.24891 31.95144 16 6.328384 42.36155 8.091934 11.25460 31.96354 20 6.329172 42.35446 8.094050 11.25585 31.96647 24 6.329163 42.35283 8.093642 11.25604 31.96833 28 6.329101 42.35188 8.093429 11.25587 31.96971 30 6.329075 42.35161 8.093417 11.25587 31.97003 based on the response impulse functions and the variance decomposition table, one notices the importance of nominal demand shocks; they contribute to approximately 30% of the trade balance variation in short as in long term. the remainder shocks have an insignificant effect on the trade balance. however, it is important to underline that an increase in oil prices contributes to the appreciation of real exchange rate in the short term and to its depreciation in the long term. since the emu, appreciation of the euro/dollar exchange rate worsens the french external position. the improvement of the french external balance position should be made by the amelioration of terms of trade, by the diversification of its business partners especially by the acquisition of new market shares (emergent market), by reducing labour cost, or by investing in research and innovation, etc. conclusion the results founded in this paper underline the french vulnerability to internal as to external shocks. they especially highlight the importance of supply and oil price shocks in economic fluctuations. indeed, through the two var models instruments (impulse response function and variance decomposition), we clearly notice the prevalence of supply shocks in the explanation of gdp fluctuations (between 21% and 45% depending in time horizon). oil prices shocks explain between 40% and 35% of economic disturbance. france, as a net importer of oil energy, is particularly vulnerable to oil price shocks; oil shock has a negative and durable effect on the economic activity. so it is important to find alternative solutions to reduce the economy dependence on oil energy. concerning demand shocks, particularly those relating to monetary and fiscal policies. results show that their effects are quite limited; this is probably due to the restrictive european economic policies adopted since the 1990s, and which is strongly controlled since the 1st january 1999 with the establishment of the european monetary union. the improvement of the french economic activity requires a better coordination, a better governance of the european economic policies. it also requires the improvement of the energy policies in order to attenuate the french economy dependence. regarding the french external position, a position to be taken seriously into account, france must encourage firms to invest and innovate to improve the competitiveness. france should benefit from the strong growth of emergent countries like china or india by orientating its exports towards sectors in expansion. it should also follows the german example by encouraging investments in small and medium-size companies. european journal of government and economics 1(1) 84 references backus, david and patrick kehoe (1992) ‘international evidence of the historical properties of business cycles’, american economic review, 82(4): 864-888. backus, david, patrick kehoe and finn e. kydland (1992) ’international real business cycles’, journal of political economy, 100(4): 745-775, backus, david, patrick kehoe and finn e. kydland (1994) ‘dynamics of the trade balance and the terms of trade: the j-curve? ‘, american economic review, 84(1): 84-103. backus, david, patrick kehoe and finn e. kydland (1995) ’international business cycles: theory and evidence’, in thomas f. cooley (ed) frontiers of business cycle research, pp. 331-356. princeton, nj: princeton university press. bernanke, ben s. (1986) ‘alternative explanations of the money-income correlation’, carnegie-rochester conference series on public policy, 25: 49–99. blanchard, oliver j. and mark w. watson (1986) ’are business cycle all alike’, in robert gordon (ed.), the american business cycle: continuity and change, pp. 123-156. chicago: university of chicago press. blanchard, oliver j. and danny quah (1989) ‘the dynamic effects of aggregate demand and supply disturbances’, american economic review, 79(4): 655-673. blinder, alan and stephen m. goldfeld s (1976) ’new measures of fiscal and monetary policy, 1958-73’, american economic review, 66(5): 780-796. dickey, david and wayne fuller (1979) ‘distribution of the estimators for autoregressive time series with a unit root,’ journal of the american statistical association, 74(366): 427-431. engle, robert f. and clive w. j. granger c. (1987) ‘cointegration and error correction representation, estimation and testing’, econometrica, 55(2) :251-276. frisch, ragnar (1933) ’propagation and impulses problems in dynamic economics’, in economics essays in honour of gustav cassel, pp. 171-205. london: allenunwin georges. gali, jordi (1992) ‘how well does the is-lm model fit post war u.s data?’, quarterly journal of economics, 107(2): 709-738. goodfreind, marvin and robert g. king (1997) ’the new classical synthesis and the role of monetary policy’, nber macroeconomics annual, cambridge, mit press. jalles, joao t. (2009) ‘do oil prices matter? the case of a small open economy’, annals of economics and finance 10(1): 65-87 johansen, soren (1991) ‘estimation and hypothesis testing of cointegration vector in gaussian vector autoregressive models’, econometrica, 59(6): 15511580. johansen, soren and katarina juselius. (1990) ‘maximum likelihood estimation and inference on cointegration-with applications for the demand for money’, oxford bulletin of economics and statistics, 52(2): 169-210. keating, john w. (1992) ‘structural approaches to vector autoregressions’, federal reserve bank of st louis journal review, 74(4): 37-57. kwiatkowski, denis, peter c.b. phillips, peter schmidt and yongcheol shin(1992) ‘testing the null hypothesis of stationarity against the alternative of a unit root,’ journal of econometrics, 54(1-3): 159-178. kydland, finn e. and edward c. prescott (1982) ‘time to build and aggregate fluctuations’, econometrica, 50(6): 1345-1370. european journal of government and economics 1(1) 85 kydland, finn e. and edward c. prescott (1991)’ the econometrics of the general equilibrium approach to business cycles’, scandinavian journal of economics, 93(2): 161-178. long, john b. and charles plosser (1983) ‘real business cycle’, journal of political economy, 91(1): 39-69. mccallum, bennett t. (1989) ’real business cycles models’, in robert j. barro (ed) modern business cycles theory, pp. 16-50. oxford : basil blackwell. phillips, peter c.b. and pierre perron (1988). ‘testing for unit roots in time series regression,’ biometrika, 75(2), 335-346. rebelo, sergio (2005) ‘real business cycle models: past, present, and future’, nber working paper, n°11401. shapiro, mattew and mark w. watson (1988) ‘sources of business cycle fluctuations ‘, cowles foundation discussion paper, n°870. sims, christopher a. (1980) ‘macroeconomics and reality’, econometrica, 48(1): 1-48. slutsky, eugen (1927) ’the summation of random waves as the sources of cyclic processes’, in problems of economic conditions, ed, conjuncture institute, moskva, 3(1). revised english version, 1937, in econometrica, 5(2): 105-146. sin título the short selling regulation in the european union european journal of government and economics 6(1), june 2017, 5-23. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 the short selling regulation in the european union: assessing the authorization granted for the european securities and markets authority to prohibit short selling matias huhtilainen a, * a university of eastern finland, finland. * corresponding author at: university of eastern finland, yliopistokatu 2. 80100 joensuu, finland. matias.huhtilainen@uef.fi article history. received 7 may 2016; first revision required 12 december 2016; accepted 27 january 2017. abstract. this paper discusses the renewed short selling regulation (regulation (eu) no 236/2012) in the european union. the focus is on the provisions that deal with prohibiting short selling in exceptional market circumstances. the regulation further enforces certain obligations to report and disclose short positions. it is concluded that banning short selling is not an effective tool to contain extreme price volatility. the difference-in-differences regression and repeated measures glm were used to test whether short selling bans were successful in containing volatility of those spanish and italian stocks that were subject to two back-to-back prohibitions during the years 2011-2013. the results are consistent with the majority of previous research, suggesting that the effectiveness of short sale constraints in reducing volatility is limited at best. furthermore, there are evidence of counterproductive effects: constraints on short selling may actually increase volatility as well as deteriorate liquidity. however, based on theory and previous studies, reporting and disclosure requirements shall be favored provided they improve market efficiency as well as supervisory work of regulatory bodies. keywords. short selling; regulation; intervention; treatment effect jel classification. g01; g18; k20 1. introduction the financial crisis in 2008 initiated a global recession which eventually, but only partly contributed to the sovereign debt crisis in the eurozone. it was argued in the liikanen report (2012) that these events (should) provide impetus for the legislative reforms concerning the financial and capital markets in the european union. the group further emphasized that even though the reforms are, by nature, pre-emptive, the aim shall be in providing authorities such tools that would effectively control the already existing market disorders. the distinctive objective is to create a market infrastructure in which the efficient capital allocation would not be suppressed, but at the same time systemic risks could be monitored and managed properly in order to protect the economic stability of the member states. in consequence, the european commission has http://www.ejge.org/ 6 m. huhtilainen / european journal of government and economics 6(1), 5-23 proposed about 30 sets of rules since 2010 regarding the common financial and capital markets in the european union (eu). the most essential reforms include the implementation of basel 3 framework into the eu banking regulation (crd iv) as well as in further accordance with global collaboration the european market infrastructure regulation (eu) no 648/2012 and alternative investment fund managers directive (2011/61/eu), which provide comprehensive rules for overthe-counter derivatives trading and for the business activity conducted by property, private equity, commodity and hedge funds. it is noteworthy though, that this type of reactive regulatory work should not be viewed as anything unique. as dobravolskas and seiranov (2011) address, the “regulatory reforms are reaction to market failures” and the evolution of regulation can be comprehended as a long regulatory cycle where “periods of tightened regulation are changed with lax regulation or deregulation”. quite interestingly, the previous financial markets regulatory cycle in the eu from 1999 to 2004 also addressed issues such as supervision, supranational regulation and harmonized frameworks (quaglia 2007), but was unable to provide sufficient systemic resiliency against the upcoming meltdown in 2008. relevant to this paper, however, is the report “model for a pan-european short selling disclosure regime regulation” by the committee of european securities regulators (2010). in the immediate aftermath of lehman brothers collapse both the united states (us) securities and exchange commission and the regulatory body of the united kingdom (uk), the financial services authority, had temporarily banned short selling in order to protect the markets and reduce downward pressure on prices. the ban covered 29 financial stocks in london stock exchange and eventually over 900 stocks in the us. subsequently, 24 other countries enforced varying constraints on short selling between september and october in 2008. (hansson and fors 2009; beber and pagano 2013). following the diversified measures by regulatory bodies, cesr remarked in the report that market participants had been negatively affected by the numerous and varying rulebooks across market places while allowing other, usually larger institutional market participants to benefit from regulatory arbitrage. the report formed the basis for the legislative process that would eventually lead to the introduction of regulation (eu) no 236/2012 on short selling and certain aspects of credit default swaps, which became fully applicable on 1 november 2012. the regulatory ambition has been to (i) increase the transparency of short positions; (ii) strengthen the stability of sovereign debt markets; (iii) provide sufficient tools for authorities to intervene markets in exceptional times; and to (iv) ensure a proper coordination between national authorities and european securities and markets authority (esma). the focus of this study is on the short sale constraints from both legal and empirical perspectives. first, the key provisions of the new regulation regarding short sale constraints, reporting requirements and supervisory work are addressed. secondly, theory on price formation and the impact of trading constraints is discussed. lastly, the difference-in-differences regression and repeated measures glm are applied to test whether the short selling bans conducted in spain and italy were successful in containing price volatility during the years 2011-2013. in contrast to a trading venue where only one-time intervention is observed, both spain and italy enforced two subsequent short selling bans on financial stocks following each other in a relatively 7 m. huhtilainen / european journal of government and economics 6(1), 5-23 short period of time. this provides a research design where the same treatment (=short selling ban) is repeated twice on same patients (=financial stocks), allowing any repetitive intervention effect to be observed with higher degree of certainty. the success of these constraints is concluded by reflecting the empirical results against the arguments that were used to advocate the said measures. as the bans in spain and italy were explicitly justified by the excessive price volatility, a logical research problem is therefore to assess whether the bans issued by the regulatory bodies were successful in containing the volatility of those financial stocks that were concerned by the prohibitions. subsequently, a conclusion of the necessity of such powers granted for market authorities is also drawn. 2. the practice of short selling under new regulatory regime the first significant revision is the prohibition of naked short selling. according to preamble 18 of the regulation the “uncovered short selling of shares and sovereign debt is sometimes viewed as increasing the potential risk of settlement failure and volatility”, and in order to reduce such risks “it is appropriate to place proportionate restrictions on uncovered short selling of such instruments”. in consequence, the regulation obliges to either borrow or to arrange the borrowing or have an arrangement with a third party that has confirmed that the securities have been located before entering into a short position. thus, prohibition of naked positions applies equally to equity and sovereign debt instruments. concerns regarding the stability of sovereign debt markets on one hand and the excessive sovereign debt of certain member states on the other have been highlighted since the escalation of european debt crisis. due to the possible adverse impact on debt market stability, regulation now prohibits purchasing credit default swaps without having a long position in underlying sovereign debt instrument. thus, sovereign credit default swaps shall be based on the insurable interest principle: only legitimate reason to enter into a cds contract is to hedge against the default risk of the issuer, which naturally requires an ownership in the underlying security. conversely, should a natural or a legal person buy a credit default swap without purchasing the underlying debt instrument first, it would be in his best interest that the issuer defaults. further practical change is the obligation to report and disclose significant net short positions to national competent authorities (nca) and to the public, respectively. apart from exemptions granted for market makers and authorized primary dealers, the reporting requirement applies to every market participant, but in respect to a predefined threshold. a significant net short position in shares equals or exceeds 0.2 per cent of issued share capital. notification obligation further applies to every 0.1 per cent change above the threshold. respective initial thresholds in sovereign debt are set at 0.1 and 0.5 per cent. the former applies if the total outstanding issued debt does not exceed €500 billion, and vice versa. however, the threshold of 0.5 per cent may apply irrespective to the value of outstanding debt provided there is a liquid futures market for that particular sovereign debt. moreover, additional incremental levels are set at 50 per cent of the initial threshold and shall be each 0.05 per cent change above the initial notification threshold of 8 m. huhtilainen / european journal of government and economics 6(1), 5-23 0.1 per cent starting at 0.15 per cent, as well as each 0.25 per cent change above the initial threshold of 0.5 per cent starting at 0.75 per cent. the purpose of the reporting obligation is twofold. first, it enables esma to monitor market moves in general and large positions in particular, and in consequence manage risks that dwell within. secondly, ncas are required to disclose reported net short positions in shares in a comprehensive and easily accessible manner, which shall be expected to contribute to more transparent market fluctuations. the disclosure threshold is set at 0.5 per cent of issued share capital, and each 0.1 per cent move above the initial threshold in addition is further disclosed. christophe et al. (2004) suggest that authorities should disclose more detailed and timely information on short positions as it would provide more accurate picture of the market sentiment, thus increasing the pricing efficiency. based on trading data from 913 nasdaq-listed stocks, they concluded that short sellers are information-oriented traders: results showed that abnormal short selling activity before the earnings announcement was significantly linked with the stock price reaction after the earnings went public, indicating a positive relationship between overvalued stocks and short sale volume. lastly, the regulation provides ncas an authorization to enforce temporary bans or introduce other constraints of similar effect, provided it notifies esma beforehand, which then coordinates and implements the proposed measures. however, according to article 28(1), esma is permitted to disregard any nca and directly order emergency measures and conduct direct operational decisions anywhere in the european economic area. intervention may only take place, however, if market conditions favor taking measures. such conditions pose an apparent threat to the functionality of financial markets or to the stability of the whole or part of the financial system in the union (“cross-border implications”). interestingly, the european court of justice (ecj) rejected the uk’s lawsuit in january 2014. in the case “c-270/12, united kingdom of great britain and northern ireland v. council of the european union, european parliament”, the uk argued that emergency powers handed to esma were illegal and that constraints on short selling have a negative impact on market efficiency. the ecj ruled that the new powers were compatible with eu law, dismissing the legal case in its entirety. 2.1 which circumstances count as “exceptional”? the legal mandate for the competent authority to prohibit or impose conditions to natural or legal persons entering into a short sale is provided in article 20(2). measures are further applicable to transactions concerning all financial instruments and transactions other than short sales, provided the pursued effect of the transaction is to confer a financial advantage in the event of a decrease in price or value of another financial instrument. in accordance with article 20(1)(a), the definition of exceptional circumstances is proportional to the market preconditions that include adverse events or developments which constitute a serious threat to financial stability or to market confidence in one or more member states. article 24(1) of commission delegated regulation (eu) no 918/2012 (referred to hereinafter 9 m. huhtilainen / european journal of government and economics 6(1), 5-23 if not mentioned otherwise) supplements the regulation with specific characteristics of adverse events and developments. they include any act, result, fact, or event that is or could reasonably be expected to cause serious financial, monetary or budgetary problems which may lead to financial instability concerning a member state or a bank and other financial institutions deemed important to the global financial system; a rating action or a default by any member state or banks and other financial institutions deemed important to the global financial system that causes or could reasonably be expected to cause severe uncertainty about their solvency; substantial selling pressures or unusual volatility causing significant downward spirals in any financial instrument related to any banks and other financial institutions and sovereign issuers deemed important to the global financial system; any relevant damage to the physical structures of important financial issuers, market infrastructures, clearing and settlement systems, and supervisors which may adversely affect markets in particular where such damage results from a natural disaster or terrorist attack; and any relevant disruption in any payment system or settlement process, in particular when it is related to interbank operations, that causes or may cause significant payments or settlement failures or delays within the union payment systems, especially when these may lead to the propagation of financial or economic stress in a bank and other financial institutions deemed important to the global financial system or in a member state. lastly, additional consideration should take into account the violent intra-day price changes. thus, should a price of a financial instrument fall significantly during a single trading day, the national competent authority is handed powers to prohibit or restrict short selling in order to prevent a disorderly decline in the price of the financial instrument (article 23(1)). a liquid share is determined to experience a significant fall in price when close-to-close return yields -10 per cent or more. this applies to illiquid shares as well, provided the company is included in the main national equity index and is the underlying for a derivative contract traded in a trading venue. otherwise a significant decrease in price is defined as 20 per cent or more for a share which price is €0.50 or higher, or the equivalent in the local currency. in all other cases, the daily decrease in value shall be 40 per cent or more to be considered significant. for sovereign and corporate bonds, increases of 7 and 10 per cent in the yield, respectively, are deemed to be significant falls in value (article 23(2-3)). quite interestingly, certain market conditions that would be later identified by the regulation were already broadly taken into account in a public statement esma released on 11 august 2011. in it esma noted that the european financial markets had been very volatile in the preceding weeks and that the “developments have raised concerns for securities markets regulators across the european union”. the following day spain, france, belgium and italy prohibited short selling with several publicly traded financial and insurance companies. the bans were originally set to last either 15 days or until further notice. belgium and france lifted the ban on 11 february 2012. france did not validate the decision, but belgium pointed to the reduced market volatility. greece had already banned short selling on 9 august 2011. the greek regulator hcmc made the decision while taking into account the conditions prevailing in the greek markets. after several extensions, the ban was eventually lifted on 15 july 2013. 10 m. huhtilainen / european journal of government and economics 6(1), 5-23 however, the regulatory bodies in mainland europe were not the first ones to implicate a relationship between high volatility and short selling. on 19 september 2008, sec described the market conditions as a period of unusual and extraordinary market volatility, while it appeared that “unbridled short selling is contributing to the recent, sudden price declines in the securities of financial institutions unrelated to true price valuation.” a day earlier fsa justified its own ban by stating that the measurements would protect and stabilize the markets. the chief executive hector sants was quoted that “while we still regard short selling as a legitimate investment technique in normal market conditions, the current extreme circumstances have given rise to disorderly markets”. australian market authority asic (2012) remarked how during the financial crisis countries around the world took steps to strengthen their financial systems due to the widespread concern that short selling was contributing to market volatility and putting enough pressure on market confidence to be systematically relevant to the global financial system and economy. 2.2 remarks on price discovery process among academics and market participants, the debate about trading constraints is partly preoccupied with interest on their impact on price discovery process and allocation of capital. miller (1977) proposed that overvaluation of stocks follows circumstances where short selling is prohibited, yet investors hold heterogeneous beliefs on the stock’s fair value. constraints on short selling only allow bullish or optimistic views to be reflected into security prices, provided that current shareholders are indifferent of owning that stock, while bearish investors who do not own shares cannot participate in trading. boulton and braga-alves (2010) provided results consistent with miller’s conclusion about the linkage between constraints and overpricing of securities. hence, considering the previously discussed statements regulatory bodies released during the global wave of interventions, the following problematization of causality shall be assessed: either short selling is the reason and securities trading below their fair values the result; or securities are being aggressively shorted because their current market value does not correspond with the underlying fundamentals and future expectations. under efficient market conditions, the first scenario seems unlikely. the effect of downward price manipulation would disappear due to the increased demand for the respective security trading at below its fair value. baker and wurgler (2006) interpret efficient markets as a competition among rational investors eventually leading to equilibrium in which prices equal the rationally discounted value of expected cash flows. the latter scenario is further supported by diamond and verrecchia (1987) who argued that due to the relatively high costs associated with short sales, only informed traders, and specifically those traders with bearish view, engage in short selling. reed (2003) tested the diamond-verrecchia-model and provided results consistent to those by diamond and verrecchia. werner (2010) points out to evidence on the negative relationship between changes in short interest and subsequent stock returns, suggesting that short sellers are at least somewhat rational participants whose decision making is based on 11 m. huhtilainen / european journal of government and economics 6(1), 5-23 fundamental valuation. however, short selling serves other practical purposes as well. namely, as hedging positions or portfolios, and as a part of market makers’ toolkit (gregoriou 2012). on the contrary, several studies dismiss the assumption of perfect rationality. biased investors’ decision-making is correlated with the rest of the markets, thus leading to irrational behavior (huang et al. 2015). lee et al. (2002) find excessive returns to be correlated with shifts in sentiment. they further conclude that bearish turnover in sentiment leads to increased volatility along with decreased future returns. brown and cliff (2005) also find a positive relationship between pricing errors and market sentiment; periods of high sentiment are followed by low longterm future returns. based on the presented evidence, a case could be made for the short selling prohibitions that took place in the aftermath of financial crisis. if during market turmoil herd behavior builds up and decision making among market participants becomes highly correlated, aggressive short selling may strengthen the excessive downward pressure, thus pushing the security prices below their intrinsic values for a considerable amount of time. adverse effects are numerous, such as margin calls and enforced collateral sales, which would further deepen the pro-cyclicality effect. 2.3 market effects of short selling constraints the evidence regarding the relationship between short selling and volatility, liquidity and price efficiency is somewhat consistent. charoenrook and daouk (2005) provide results which point to less volatile aggregate returns as well as greater liquidity in the markets when short selling is not subject to restrictions. stocks with higher short-sale constraints are affected by deteriorated price efficiency and vice versa; neither do relieved constraints lead to price instability or extreme negative returns (saffi and sigurdsson 2011; bai et al. 2006; boehmer and wu 2013; mckenzie 2012). boehmer et al. (2013) and beber and pagano (2013) concluded that constraints reduce liquidity; this is particularly undesirable in a crisis situation where bid-ask-spreads are already large. evidence indicating failure of the bans during the financial crisis and later in europe during the years 2011 and 2012 have been provided by saastamoinen and suhonen (2013); helmes et al. (2011); lobanova et al. (2010); boulton and braga-alves (2012). in addition, bernal et al. (2014) concluded that all types of short sale constraints inflict negative impact on market efficiency and that no constraint is effective against price decline. however, they proposed that naked short selling prohibition is the least damaging one from the perspective of market efficiency. on the other hand, chang et al. (2007) find less positively skewed returns as well as increased volatility for individual stocks coincidental to short sale volume. furthermore, lecce et al. (2008) and setzu and marchesi (2006) present evidence that suggest increased volatility in relation to short selling. appel and fohlin (2010) argue that short selling bans in 2008-2009 improved market liquidity. bris et al. (2007) presented results that suggested less negatively skewed returns in countries where short selling was prohibited. they further propose that short selling bans may alleviate panic in the markets. 12 m. huhtilainen / european journal of government and economics 6(1), 5-23 it is worth noting that the bans were targeted distinctly at financial stocks, which is suggestive of concerns on contagion of losses subsequent to liquidity run in a highly concentrated and interconnected network of financial institutions. the sec press release on 19 september 2008 further implicated those concerns as sec suggested there was an essential link between the stock price and confidence in the financial institution and that institutions were particularly depended on the confidence of their trading counterparties in the conduct of their core business. vitali et al. (2011) provide evidence on the global structure of the control network of transnational corporations. they further conclude that while the financial network is seemingly robust in a nonvolatile environment, in bad times firms go into distress simultaneously due to the contagion effect in a densely connected system. 3. did the two subsequent bans in italy and spain contain volatility? according to theory of government failure, poor public policies generate counterproductive effects, which lead to an environment that is more inefficient or otherwise non-optimal, vis-à-vis, to the preceding situation where the intervention did not take place (datta-chaudhuri 1990). thus, good goals per se cannot be satisfactory if the costs exceed the benefits. rent control, for instance, is suggested to distort the residential construction and lower the housing quality. the target (an affordable residential district) would be achieved but only with costs that are not being sufficiently compensated (moon and stotsky 1993). interestingly, the regulation clearly takes into account the necessity to compare benefits and costs before conducting any interventions: article 20(1)(b) permits competent authorities to act only when the “measure is necessary to address the threat and will not have a detrimental effect on the efficiency of financial markets which is disproportionate to its benefits”. the sec chairman christopher cox stated in an interview on 31 december 2008 that, quote, “knowing what we know now, i believe on balance, the commission would not do it (short-sale ban on financials) again. the costs appear to outweigh the benefits.” whether market authorities are affected by action bias during market turmoil is worth considering. notable market panic could add further pressure simply to do anything in order to satisfy the action-seeking public. should taking measures become the priority in itself rather than the actual outcome, it could heighten the risk of intervention’s “detrimental effects disproportionate to its benefits”. although the methodical approach is similar to saastamoinen and suhonen (2013), only spanish and italian stocks are considered for they were “treated” twice in both countries, as explained earlier. the paper by saastamoinen and suhonen on the other hand studies the first (and onetime) ban conducted in spain, france, belgium and italy from 12 august 2011. the treatment indicator is defined in table 1. the treatment indicator values 0 and 1 indicate non-treated and treated time periods, respectively. the timetable is split up in five portions using the exact dates each short selling ban was first introduced and later lifted in spain and italy. the non-treatment periods were composed to be as identical in duration as possible. 13 m. huhtilainen / european journal of government and economics 6(1), 5-23 table 1. treatment indicator. 0 1 0 1 0 spain 01/02/2011 11/08/2011 12/08/2011 15/02/2012 16/02/2012 22/07/2012 23/07/2012 31/01/2013 01/02/2013 31/08/2013 italy 01/02/2011 11/08/2011 12/08/2011 24/02/2012 25/02/2012 22/07/2012 23/07/2012 14/09/2012 15/09/2012 31/03/2013 similarly to the above-mentioned paper, though, the possible treatment impact on price volatility is estimated on close-to-close prices; see figure 1. the (potential) treatment effect. is observed from the post intervention difference-in-differences between the treated group a and control group b. figure 1. (potential) treatment effect observed from the post intervention difference-in-differences between the treated group a and control group b. 3.1 data and methodology the data contain those stocks that were subject to the first ban in spain and italy. stocks under the scope of the second ban were the same in italy, but in spain every listed company in the madrid stock exchange was concerned. however, only those 16 stocks that were affected by the first ban are included. in addition, the italian financial market authority consob imposed a oneday prohibition on four stocks on 27 february 2013, but this was not taken into account. a control group was formed with 30 non-treated swedish, finnish, dutch, irish, slovene, austrian, maltese, cypriot, german and portuguese financial services companies in respect to a considerable market value as well as sufficient liquidity and trading volume. due to the mergers and acquisitions the number of companies in the data did not remain constant. the following companies in spain were delisted at some point between december 2011 and august 2013: banca cívica, banco de valencia, banco español de crédito, banco pastor, caja de ahorros del mediterráneo. in italy, credito artigiano was acquired by credito valtellinese in september 2012. additionally, bankia launched an ipo in june and the shares began trading on 20 july 2011. banca cívica went public on 21 july 2011. the number of observations (trading 0 5 10 15 20 25 t 1 t 2 b a a' 14 m. huhtilainen / european journal of government and economics 6(1), 5-23 days) for bankia and banca cívica preceding the first ban totaled 17 and 16, respectively. the number of stocks in the control group remained the same. due to the different treatment cutoff points at 𝑡𝑡𝑛𝑛 between italian and spanish markets, the average group volatility for control stocks is computed twice using both spain’s and italy’s exact dates, see table 1. stock’s daily log return r is calculated with close-to-close prices (adjusted with dividends and splits): 𝑟𝑟𝑡𝑡 = ln � 𝑝𝑝𝑡𝑡 𝑝𝑝𝑡𝑡−1 � [1] where 𝑝𝑝𝑡𝑡 denotes asset price at time t. the realized variance (rv) method is followed to compute the stock’s realized volatility as the sum of daily squared log returns. the stock’s average rv for each time period is further obtained by dividing the sum with the number of close-to-close returns: 𝑟𝑟𝑟𝑟= 𝑛𝑛 ∑ 𝑡𝑡=1 𝑟𝑟𝑡𝑡 2 𝑛𝑛 [2] the average group rv for each time period is then computed by dividing the sum of mean rvs with number of companies belonging to that group in question: 𝑅𝑅𝑅𝑅𝚤𝚤,𝑡𝑡������ = 𝑛𝑛 ∑ 𝑖𝑖=1 𝑟𝑟𝑟𝑟𝑖𝑖 𝑁𝑁 . [3] table 2 presents the average group rvs. time period indicator values are as described in table 1. as noted earlier, control group volatility is estimated twice using both spain’s and italy’s exact treatment moments table 2. average group rvs. 0 1 0 1 0 spain .000420409 .001664344 .001262440 .002076911 .002330436 italy .000803200 .001699800 .004650600 .001834200 .000576400 control spain .001321801 .001609780 .000928883 .000878727 .000911369 control italy .001321801 .001614430 .000877246 .001214762 .000722796 15 m. huhtilainen / european journal of government and economics 6(1), 5-23 3.1.1 difference-in-differences framework the estimate of difference-in-differences between treatment and control groups after the treatment is obtained from (µ11 − µ10) − (µ01 − µ00), [4] where µ𝑖𝑖𝑡𝑡 is the (mean) outcome of group i at time t. the dummy value 0 in subscript either denotes control group or pre-treatment period, and vice versa. table 3. volatility during bans. spain italy first ban second ban first ban second ban (µ11 − µ10) .020292487 .010042306 .012888638 −.025367707 (µ01 − µ00) .003765482 −.000834253 .003823385 .005235096 d-d estimate .016527005 .010876559 .009065253 −.030602803 as shown in table 3, volatility in spain exceeded the preceding period’s volatility during both bans. the time period under the second ban was less volatile in italy compared to the preceding non-ban period. the treatment effect as the coefficient of the interaction 𝛽𝛽3 is obtained from the difference-indifferences linear regression model: 𝑦𝑦𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1(𝑡𝑡𝑟𝑟𝑡𝑡𝑡𝑡𝑡𝑡𝑖𝑖) + 𝛽𝛽2(𝑡𝑡𝑎𝑎𝑡𝑡𝑡𝑡𝑟𝑟𝑖𝑖) + 𝛽𝛽3(𝑡𝑡𝑟𝑟𝑡𝑡𝑡𝑡𝑡𝑡𝑖𝑖 × 𝑡𝑡𝑎𝑎𝑡𝑡𝑡𝑡𝑟𝑟𝑖𝑖) + ԑ𝑖𝑖, [5] where: 𝑦𝑦𝑖𝑖 = the treatment effect; 𝑡𝑡𝑟𝑟𝑡𝑡𝑡𝑡𝑡𝑡𝑖𝑖 = 1 if in treatment group; 𝑡𝑡𝑟𝑟𝑡𝑡𝑡𝑡𝑡𝑡𝑖𝑖 = 0 if in control group; 𝑡𝑡𝑎𝑎𝑡𝑡𝑡𝑡𝑟𝑟𝑖𝑖 = 1 if after the introduction of treatment; 𝑡𝑡𝑎𝑎𝑡𝑡𝑡𝑡𝑟𝑟𝑖𝑖 = 0 if before the introduction of treatment. baseline is the non-treatment time period which is followed by a period of short sale prohibition (follow up). the upper section of the table merely reports the number of companies in each group during time period i. treatments effects for both spain and italy are shown in table 4 and table 5, respectively. neither treatment significantly reduced targeted groups’ volatility relative to the change in control group stocks; see table 5. 16 m. huhtilainen / european journal of government and economics 6(1), 5-23 table 4. the treatment effects for both bans in spain. first ban second ban baseline follow up baseline follow up control 30 30 30 30 spain 16 16 15 14 total 46 46 45 44 r2 0.0256 0.0456 baseline follow up baseline follow up control 0.001 0.002 0.001 0.001 spain 0.000 0.002 0.001 0.002 diff -0.001 0.000 0.000 0.001 diff-in-diff 0.001 0.001 std. error 0.001 0.001 t 0.80 0.97 p>|t| 0.428** 0.337** notes. * means and standard errors are estimated by linear regression. ** inference: *** p<0.01; ** p<0.05; *p<0.1 table 5. the treatment effects for both bans in italy. first ban second ban baseline follow up baseline follow up control 30 30 30 30 italy 29 29 29 29 total 59 59 59 59 r-square 0.0210 0.0292 baseline follow up baseline follow up control 0.001 0.002 0.001 0.001 italy 0.001 0.002 0.005 0.002 diff -0.001 0.000 0.004 0.001 diff-in-diff 0.001 -0.003 std. error 0.001 0.003 t 0.68 -0.99 p>|t| 0.500** 0.327** notes. * means and standard errors are estimated by linear regression. ** inference: *** p<0.01; ** p<0.05; *p<0.1 3.1.2 repeated measures glm in contrast to computing the average daily rv of control group stocks for each time period twice using both spain’s and italy’s exact intervention moments separately, the dates that divide control group data into five subsequent portions are redefined as to concurrently respond both italian and spanish time periods as inclusively as possible. it is thus important to address that “manipulating” data in this manner will produce at least somewhat inaccurate results. table 6 shows the redifined 17 m. huhtilainen / european journal of government and economics 6(1), 5-23 time periods, and table 7 some descriptive statistics. the model omits those stocks that have missing values in either time period. therefore n is constant for each group throughout the dataset. table 6. redefined time periods. control 1 2 3 4 5 time period 01/02/2011 11/08/2011 12/08/2011 24/02/2012 25/02/2012 22/07/2012 23/07/2012 23/11/2012 24/11/2012 14/06/2013 table 7. descriptive statistics for the redefined time periods. group mean n time 1 spain .0003886263 13 italy .0008225523 28 control .0013218015 30 time 2 spain .0008748838 13 italy .0017307628 28 control .0016144301 30 time 3 spain .0011986207 13 italy .0047757355 28 control .0008772460 30 time 4 spain .0021373749 13 italy .0018406059 28 control .0010071807 30 time 5 spain .0023304357 13 italy .0005764257 28 control .0007745044 30 the following analysis includes one between-subjects factor “group” with three levels (spain, italy and control group) and one within-subjects factor “time” with five levels (time periods 1 to 5). the interest is on whether the difference between levels of group is different when comparing within-subjects factor levels 2 and 1, 4 and 3 and 5 and 1. contrast coefficients matrix of each possible pairwise comparison of group levels is thus: 𝐿𝐿 = ⎝ ⎜⎜ ⎛ 1 −1 0 1 0 −1 0 1 −1 0 −1 1 −1 1 0 −1 0 1 ⎠ ⎟⎟ ⎞ , 18 m. huhtilainen / european journal of government and economics 6(1), 5-23 where l1: (1) spain (-1) italy, l2: (1) spain (0) control, l3: (0) italy (1) control, l4: (0) control (1) italy, l5: (-1) italy (1) spain and l6: (-1) control (0) spain. for within-subjects factor time, levels 2 and 4 notate the first and second bans, respectively. interaction contrasts for group levels and difference between levels 2 and 1, 4 and 3 and 5 and 1 of time are provided. thus, for l2 “time 2 v time 1”: (2spain 2control) (1spain 1control) =.000. the last contrast for the difference between last and first time period is merely anecdotal, as if whether there was a significant difference in volatility after two nearly consecutive prohibition periods vis-à-vis to the base point when no actions had been taken by the authorities. table 8. contrast results. contrast time 2 v time 1 time 4 v time 3 time 5 v time 1 l1 contrast estimate .000 .004 .002 hypothesized value 0 0 0 difference (est. – hypothesized) .000 .004 .002 std. error .001 .003 .001 sig. .477 .161 .037 95 % conf. for diff. lower bound -.002 -.002 .000 95 % conf. for diff. upper bound .001 .009 .004 l2 contrast estimate .000 .001 .002 hypothesized value 0 0 0 difference (est. – hypothesized) .000 .001 .002 std. error .001 .003 .001 sig. .741 .766 .017 95 % conf. for diff. lower bound -.001 -.005 .000 95 % conf. for diff. upper bound .001 .006 .005 l3 contrast estimate .001 -.003 .000 hypothesized value 0 0 0 difference (est. – hypothesized) .001 -.003 .000 std. error .000 .002 .001 sig. .187 .157 .710 95 % conf. for diff. lower bound .000 -.007 -.001 95 % conf. for diff. upper bound .002 .001 .002 l4 contrast estimate -.001 .003 .000 hypothesized value 0 0 0 difference (est. – hypothesized) -.001 .003 .000 std. error .000 .002 .001 sig. .187 .157 .710 95 % conf. for diff. lower bound -.002 -.001 -.002 95 % conf. for diff. upper bound .000 .007 .001 l5 contrast estimate .000 -.004 -.002 hypothesized value 0 0 0 difference (est. – hypothesized) .000 -.004 -.002 std. error .001 .003 .001 sig. .477 .161 .037 95 % conf. for diff. lower bound -.001 -.009 -.004 95 % conf. for diff. upper bound .002 .002 .000 l6 contrast estimate .000 -.001 -.002 hypothesized value 0 0 0 difference (est. – hypothesized) .000 -.001 -.002 std. error .001 .003 .001 sig. .741 .766 .017 95 % conf. for diff. lower bound -.001 -.006 -.005 95 % conf. for diff. upper bound .001 .005 .000 19 m. huhtilainen / european journal of government and economics 6(1), 5-23 table 8 presents the contrast results. no statistical significance for the differences between levels of group is found when comparing levels of “time” of 2 and 1, and 4 and 3. however, for l2 “5 v 1” a p-value of .017 is obtained: treated stocks in spain traded in a more volatile situation as opposed to stocks in control group. in addition, a significant p-value is found for l5 (and therefore for l1), although the comparison is between two treated groups. nevertheless, it is noteworthy that volatility developed to opposite directions in spain and italy when comparing time periods 5 and 1. 4. conclusions the short selling bans of 2011-13 did not contain the price volatility of spanish and italian financial stocks that were subjected to trading constraints. therefore, the target that market authorities had set for the interventions was not accomplished. the results are consistent with the bulk of previous studies on the usefulness of short sale bans to contain stock market volatility. based on the findings of this and previous studies, esma and ncas should maintain particularly strict criterion according for which to assess the necessity to conduct interventions in order to “correct” the markets. as the sec chairman christopher cox stated earlier in this study, “the costs appear to outweigh the benefits”. however, evidence to support the actions taken by regulators are provided, for instance, by brunnermeier and oehmke (2014) who show that financial institutions are vulnerable to predatory short selling through enforced liquidation of long-term asset holdings following the withdrawal of depositors and short-term creditors. as the authors conclude, implications on the resilience of balance sheets may justify temporary restrictions on short selling of fragile institutions. it is also true that the approach of this study solely relies on daily closing prices and linear models, which lead to certain limitations. the alternative ways to assess the impact of short sale constraints have been provided by bernal et al. (2014), fung and draper (1999) and kolasinski et al. (2013) who focused on trading volume and bid-ask spreads, mispricing of index futures and the information content of prices, respectively. further, as explained earlier, the generalized linear model was applied to time periods that did not reflect the actual treatment periods. nonetheless, it is questionable whether short sale bans or other constraints of similar effect should be justified by excessive volatility for two reasons in particular: volatility is a subjective metric in a sense that market participants tolerate risk differently. thus, it is impossible for a regulatory body to define an exact moment after which the market becomes too risky or too volatile for everybody or even for the majority of participants, yet the measures are collective in nature. secondly, financial markets are not a closed system; if the risk level does not satisfy the riskaverse person or entity, they are free to exit specific positions or markets in general, or not to utilize the possibility of free entry in the first place. 20 m. huhtilainen / european journal of government and economics 6(1), 5-23 references appel, i., & fohlin, c. (2010). 'shooting the messenger?' the impact of short sale bans in times of crisis. http://dx.doi.org/10.2139/ssrn.1595003 australian securities and investments commission (2012). short selling: post-implementation review. retrived from http://asic.gov.au/regulatory-resources/find-a-document/reports/rep302-short-selling-post-implementation-review/ bai, y., chang, e. c., & wang, j. (2006). asset prices under short-sale constraints. retrived from https://stuff.mit.edu/people/wangj/pap/bcw_061112.pdf baker, m., & wurgler, j. (2006). investor sentiment and the cross-section of stock returns. the journal of finance 61(4), 1645-1680. https://doi.org/10.1111/j.1540-6261.2006.00885.x beber, a., & pagano, m. (2013). short-selling bans around the world: evidence from the 200709 crisis. the journal of finance 68(1), 343-381. https://doi.org/10.1111/j.15406261.2012.01802.x bernal, o., herinckx, a., & szafarz, a. (2014). which short-selling regulation is the least damaging to market efficiency? evidence from europe. international review of law and economics 37(c), 244-256. https://doi.org/10.1016/j.irle.2013.12.002 boehmer, e., jones, c. m., & zhang, x. (2013). shackling short sellers: the 2008 shorting ban. the review of financial studies 26(6), 1363-1400. https://doi.org/10.1093/rfs/hht017 boehmer, e., & wu, j. (2013). short selling and the informational efficiency of prices. the review of financial studies 26(2), 287-322. https://doi.org/10.1093/rfs/hhs097 boulton, t. j., & braga-alves. m. v. (2012). naked short selling and market returns. the journal of portfolio management 38(3), 133-142. https://doi.org/10.3905/jpm.2012.38.3.133 boulton, t. j., & braga-alves, m. v. (2010). the skinny on the 2008 naked short sale restrictions. journal of financial markets 13(4), 397421. https://doi.org/10.1016/j.finmar.2010.05.002 bris, a., goetzmann, w. n., & zhu, n. (2007). efficiency and the bear: short sales and markets around the world. the journal of finance 62(3), 1029-1079. https://doi.org/10.1111/j.15406261.2007.01230.x brown, g. w., & cliff, m. t. (2005). investor sentiment and asset valuation. the journal of business 78(2), 405-440. https://doi.org/10.1086/427633 brunnermeier, m. k., & oehmke, m. (2014). predatory short selling. review of finance, european finance association 18(6), 2153-2195. https://doi.org/10.1093/rof/rft043 chang, e. c., cheng, j. w., & yu, y. (2007). short-sales constraints and price discovery: evidence from the hong kong market. the journal of finance 62(5), 20972121. https://doi.org/10.1111/j.1540-6261.2007.01270.x charoenrook, a., & daouk, h. (2005). a study of market-wide short-selling restrictions. cornell university working paper. retrived from https://papers.ssrn.com/sol3/papers.cfm?abstract_id=687562 christophe, s. e., ferri, m. g., & angel, j. j. (2004). short-selling prior to earnings http://dx.doi.org/10.2139/ssrn.1595003 http://asic.gov.au/regulatory-resources/find-a-document/reports/rep-302-short-selling-post-implementation-review/ http://asic.gov.au/regulatory-resources/find-a-document/reports/rep-302-short-selling-post-implementation-review/ https://stuff.mit.edu/people/wangj/pap/bcw_061112.pdf https://doi.org/10.1111/j.1540-6261.2006.00885.x https://doi.org/10.1111/j.1540-6261.2012.01802.x https://doi.org/10.1111/j.1540-6261.2012.01802.x https://doi.org/10.1016/j.irle.2013.12.002 https://doi.org/10.1093/rfs/hht017 https://doi.org/10.1093/rfs/hhs097 https://doi.org/10.3905/jpm.2012.38.3.133 https://doi.org/10.1016/j.finmar.2010.05.002 https://doi.org/10.1111/j.1540-6261.2007.01230.x https://doi.org/10.1111/j.1540-6261.2007.01230.x https://doi.org/10.1086/427633 https://doi.org/10.1093/rof/rft043 https://doi.org/10.1111/j.1540-6261.2007.01270.x https://papers.ssrn.com/sol3/papers.cfm?abstract_id=687562 21 m. huhtilainen / european journal of government and economics 6(1), 5-23 announcements. journal of finance 59(4), 1845-1875. https://doi.org/10.1111/j.15406261.2004.00681.x committee of european securities regulators (2010). model for a pan-european short selling disclosure regime. retrived from https://www.esma.europa.eu/sites/default/files/library/2015/11/10_088.pdf datta-chaudhuri, m. (1990). market failure and government failure. the journal of economic perspectives 4(3), 25-39. https://doi.org/10.1257/jep.4.3.25 diamond, d. w., & verrecchia, r. e. (1987). constraints on short-selling and asset price adjustment to private information. journal of financial economics 18(2), 277311. https://doi.org/10.1016/0304-405x(87)90042-0 dobravolskas, a., & seiranov, j. (2011). financial stability as the goal of post-crisis regulatory reforms. verslo sistemos ir ekonomika 1(1), 101-114. european commission (2014). a comprehensive eu response to the financial crisis: substantial progress towards a strong financial framework for europe and a banking union for the eurozone. retrived from url: http://europa.eu/rapid/press-release_memo-1457_en.htm european securities and markets authority (2011). esma promotes harmonised regulatory action on short-selling in the eu. retrived from https://www.esma.europa.eu/sites/default/files/library/2015/11/esma_2011_266_uu_pu blic_statement_on_short_selling.pdf financial services and markets authority (2012). the fsma modifies short selling rules. retrived from http://www.fsma.be/en/article/press/div/2012/2012-02-13_shortselling.aspx fung, j. k. w., & draper, p. (1999). mispricing of index futures contracts and short sales constraints. journal of futures markets 19(6), 695-715. https://doi.org/10.1002/(sici)10969934(199909)19:6<695::aid-fut4>3.0.co;2-h gregoriou, g. n. (2011). handbook of short selling. waltham, ma, usa: academic press. hansson, f., & fors, e. (2009). get shorty? market impact of the 2008-09 u.k. short selling ban. university of gothenburg working papers in economics no. 365. retrived from https://gupea.ub.gu.se/handle/2077/20407 hellenic capital market commission (2010). guidance on hcmc’s decision regarding short selling ban. helmes, u., henker, j. & henker, t. (2011). the effect of the ban on short selling on market efficiency and volatility. paper presented at the 2011 financial management association annual meeting, denver. http://dx.doi.org/10.2139/ssrn.1688135 huang, d., jiang, f., tu, j., & zhou, g. (2015). investor sentiment aligned: a powerful predictor of stock returns. review of financial studies 28(3), 791837. https://doi.org/10.1093/rfs/hhu080 kolasinski, a. c., reed, a., & thornock, j. r. (2013). can short restrictions actually increase informed short selling? financial management 42(1), 155-181. https://doi.org/10.1111/j.1755053x.2012.01223.x https://doi.org/10.1111/j.1540-6261.2004.00681.x https://doi.org/10.1111/j.1540-6261.2004.00681.x https://www.esma.europa.eu/sites/default/files/library/2015/11/10_088.pdf https://doi.org/10.1257/jep.4.3.25 https://doi.org/10.1016/0304-405x(87)90042-0 http://europa.eu/rapid/press-release_memo-14-57_en.htm http://europa.eu/rapid/press-release_memo-14-57_en.htm https://www.esma.europa.eu/sites/default/files/library/2015/11/esma_2011_266_uu_public_statement_on_short_selling.pdf https://www.esma.europa.eu/sites/default/files/library/2015/11/esma_2011_266_uu_public_statement_on_short_selling.pdf http://www.fsma.be/en/article/press/div/2012/2012-02-13_shortselling.aspx https://doi.org/10.1002/(sici)1096-9934(199909)19:6%3c695::aid-fut4%3e3.0.co;2-h https://doi.org/10.1002/(sici)1096-9934(199909)19:6%3c695::aid-fut4%3e3.0.co;2-h https://gupea.ub.gu.se/handle/2077/20407 http://dx.doi.org/10.2139/ssrn.1688135 https://doi.org/10.1093/rfs/hhu080 https://doi.org/10.1111/j.1755-053x.2012.01223.x https://doi.org/10.1111/j.1755-053x.2012.01223.x 22 m. huhtilainen / european journal of government and economics 6(1), 5-23 lecce, s., lepone, a., & segara, r. (2008). the impact of naked short-sales on returns, volatility and liquidity: evidence from the australian securities exchange. 21st australasian finance and banking conference 2008 paper. https://dx.doi.org/10.2139/ssrn.1253176 lee, w., jiang, c., & indro, d. (2002). stock market volatility, excess returns, and the role of investor sentiment. journal of banking and finance 26(12), 22772299. https://doi.org/10.1016/s0378-4266(01)00202-3 liikanen, e. (2012). high-level expert group on reforming the structure of the eu banking sector. final report. retrived from http://ec.europa.eu/internal_market/bank/docs/highlevel_expert_group/report_en.pdf lobanova, o., hamid, s. s., & prakash, a. j. (2010). the impact of short-sale restrictions on volatility, liquidity and market efficiency: the evidence from the short-sale ban in the u.s. http://dx.doi.org/10.2139/ssrn.1571601 mckenzie, m. d. (2012). should short selling be banned during periods of market turmoil? jassa 2, 8-13. miller, e. m. (1977). risk, uncertainty, and divergence of opinion. the journal of finance 32(4), 1151-1168. https://doi.org/10.1111/j.1540-6261.1977.tb03317.x moloney, n. (2014). eu securities and financial markets regulation. oxford university press. moon, c.-g., & stotsky, j. g. (1993). the effect of rent control on housing quality change: a longitudinal analysis. journal of political economy 101(6), 11141148. https://doi.org/10.1086/261917 quaglia, l. (2007). the politics of financial services regulation and supervision reform in the european union. european journal of political research 46(2), 269290. https://doi.org/10.1111/j.1475-6765.2006.00695.x reed, a. v. (2002). costly short-selling and stock price adjustment to earnings announcements. dissertations available from proquest. aai3073047. retrived from http://repository.upenn.edu/dissertations/aai3073047 saastamoinen, j., & suhonen, n. (2013). were the european short selling bans of 2011 effective? economics bulletin 33(3), 1847-1851. retrived from http://www.accessecon.com/pubs/eb/2013/volume33/eb-13-v33-i3-p173.pdf saffi, p. a. c., & sigurdsson, k. (2011). price efficiency and short selling. the review of financial studies 24(3), 821-852. https://doi.org/10.1093/rfs/hhq124 securities and exchange commission (2008). sec halts short selling of financial stocks to protect investors and markets. retrived from http://www.sec.gov/news/press/2008/2008211.htm setzu, a., & marchesi, m. (2006). the effects of short-selling and margin trading: a simulation analysis. university of cagliari working paper. retrived from https://www.semanticscholar.org/paper/the-effects-of-short-selling-and-margintrading-a-setzu-marchesi/e4a412a782035b8141ed641d4c0e19b596fed2b1 vitali, s., glattfelder, j. b., & battiston, s. (2011). the network of global corporate control. plos one 6(10). https://doi.org/10.1371/journal.pone.0025995 https://dx.doi.org/10.2139/ssrn.1253176 https://doi.org/10.1016/s0378-4266(01)00202-3 http://ec.europa.eu/internal_market/bank/docs/high-level_expert_group/report_en.pdf http://ec.europa.eu/internal_market/bank/docs/high-level_expert_group/report_en.pdf http://dx.doi.org/10.2139/ssrn.1571601 https://doi.org/10.1111/j.1540-6261.1977.tb03317.x https://doi.org/10.1086/261917 https://doi.org/10.1111/j.1475-6765.2006.00695.x http://repository.upenn.edu/dissertations/aai3073047 http://www.accessecon.com/pubs/eb/2013/volume33/eb-13-v33-i3-p173.pdf https://doi.org/10.1093/rfs/hhq124 http://www.sec.gov/news/press/2008/2008-211.htm http://www.sec.gov/news/press/2008/2008-211.htm https://www.semanticscholar.org/paper/the-effects-of-short-selling-and-margin-trading-a-setzu-marchesi/e4a412a782035b8141ed641d4c0e19b596fed2b1 https://www.semanticscholar.org/paper/the-effects-of-short-selling-and-margin-trading-a-setzu-marchesi/e4a412a782035b8141ed641d4c0e19b596fed2b1 https://doi.org/10.1371/journal.pone.0025995 23 m. huhtilainen / european journal of government and economics 6(1), 5-23 werner, s. (2010). short selling activities and convertible bond arbitrage empirical evidence from the new york stock exchange. vol. 75. springer science & business media. https://doi.org/10.1007/978-3-8349-6003-0 https://doi.org/10.1007/978-3-8349-6003-0 abstract. this paper discusses the renewed short selling regulation (regulation (eu) no 236/2012) in the european union. the focus is on the provisions that deal with prohibiting short selling in exceptional market circumstances. the regulation furthe... keywords. short selling; regulation; intervention; treatment effect jel classification. g01; g18; k20 2. the practice of short selling under new regulatory regime nonetheless, it is questionable whether short sale bans or other constraints of similar effect should be justified by excessive volatility for two reasons in particular: volatility is a subjective metric in a sense that market participants tolerate ri... references the politics of renewable power in spain vol.9 • no.1 2020 issn: 2254-7088 european journal of government and economics 9(1), june 2020. european journal of government and economics issn: 2254-7088 number 9, issue 1, june 2020 doi: https://doi.org/10.17979/ejge.2020.9.1 the politics of renewable power in spain 5-25 doi: https://doi.org/10.17979/ejge.2020.9.1.5231 john s. duffield voting turnout in greece: expressive or instrumental? 26-45 doi: https://doi.org/10.17979/ejge.2020.9.1.5426 irene daskalopoulou the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach 46-73 doi: https://doi.org/10.17979/ejge.2020.9.1.5714 luzius stricker and moreno baruffini do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey 74-94 doi: https://doi.org/10.17979/ejge.2020.9.1.5948 esra n. kilci a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 95-114 doi: https://doi.org/10.17979/ejge.2020.9.1.5847 barbara pawełek https://doi.org/10.17979/ejge.2020.9.1 https://doi.org/10.17979/ejge.2020.9.1.5231 https://doi.org/10.17979/ejge.2020.9.1.5426 https://doi.org/10.17979/ejge.2020.9.1.5714 https://doi.org/10.17979/ejge.2020.9.1.5948 https://doi.org/10.17979/ejge.2020.9.1.5847 european journal of government and economics 9(1), june 2020, 5-25 5 european journal of government and economics issn: 2254-7088 the politics of renewable power in spain john s duffielda* a department of political science, georgia state university, atlanta, united states * corresponding author at: duffield@gsu.edu article history. received 3 april 2019; first revision required 8 october 2019; accepted 13 january 2020. abstract. government support for renewable power in spain has varied over time. after fostering a boom in the first decade of the 2000s, public support dried up in the early 2010s before making a reappearance in the last several years. this paper details and explains the fluctuating political fortunes of renewable power in spain. it argues that the recent revival of support for renewable power should no come as no surprise. rather, it reflects a reassertion of the underlying imperatives that prompted support by governments of different parties in previous years. nevertheless, the role of government in promoting renewable power is now undergoing a fundamental change, as it shifts from directly subsidizing new generating capacity to removing barriers that might discourage private investment and facilitating the operation of market forces. keywords. spain; spanish politics; renewable power; wind power; solar power jel codes. q42; q48 doi. https://doi.org/10.17979/ejge.2020.9.1.5231 1. introduction in the first decade of the 2000s, spain emerged as a global leader in the deployment of renewable power. a series of support schemes triggered rapid growth in first wind, then solar photovoltaic (pv), and finally concentrated solar power (csp, also called solar thermoelectric) installations. as a result, the country attained some of the highest levels of generating capacity and electricity output by these technologies in the world. then, in the early 2010s, the deployment of renewable power in spain ground to a halt. to close a growing gap between the revenues and expenditures of the electric power system, successive governments reduced the remuneration that renewable power plants could receive and then imposed a moratorium on subsidies for new installations. as a result, almost no renewable generating capacity was added after 2012. in the past several years, however, renewable power, especially wind and solar pv, has experienced a revival in spain. the people’s party government of mariano rajoy (2011-2018) held a series of auctions for renewable generating capacity to be built by 2020. then, the subsequent socialist government of pedro sánchez (2018-2019) took steps to remove obstacles that had blocked some forms of renewable power and threatened to hinder the deployment of the capacity awarded in the actions. and independently, a growing number of mailto:duffield@gsu.edu https://doi.org/10.17979/ejge.2020.9.1.5231 john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 6 private actors decided to build new wind and pv power plants without any government subsidies. as a result, thousands of new megawatts of renewable capacity went online in 2019, with more expected to follow. this paper describes and explains the fluctuating political fortunes of renewable power in spain. the focus is on broad trends in central government support for renewable power rather than the details of various promotional schemes1. along the way, it makes three main arguments. first, the recent revival of support for renewable power should come as no surprise. rather, it reflects a reassertion of the underlying imperatives that prompted government support for renewable power in previous years. the pause in subsidies and deployment was just a temporary departure from the norm, reflecting an unusual concatenation of circumstances. second, precisely because of these strong long-standing motivations, support for renewable power has enjoyed substantial cross-party political support at the national level. the ups and downs in renewable capacity deployment are not primarily the result of political differences, despite changes of government every seven to eight years. with one principal exception, which is described below, the major parties have generally pursued similar goals with regard to renewable power. third, the role of government in promoting renewable power is now undergoing a fundamental change. in the past, the central government played a critical role by providing financial support for investments that would otherwise be unprofitable. increasingly, as renewable power has become cost competitive with conventional sources of power, the government’s principal task has become removing barriers that might discourage private investment and facilitating the operation of market forces, not directly subsidizing new power plants. the paper begins by identifying the motivations that have undergirded spain’s support for renewable power over the years. it then reviews the country’s initial efforts to promote renewable power and the boom and bust that followed. the following section examines the recent revival of government support for renewable power, and a final section explores the longer-term prospects for renewable power in spain and the changing role of government in promoting it. the paper focuses on two particular sources of renewable electricity: wind and solar pv. wind was the primary beneficiary of initial efforts to promote renewable power and has become the largest single source of renewable electricity in spain. solar pv, for its part, has always had the greatest resource potential, although that potential took longer to be realized because of higher costs, and it is now poised to become the leading area of investment. in contrast, hydropower, the traditional source of renewable electricity in spain, has been a much more mature technology, with relatively few remaining opportunities for exploitation. finally, although spain has been the world leader in csp, which uses mirrors to concentrate the sun’s rays, it constitutes a smaller share of spain’s renewable capacity than either wind or pv and is likely to continue to do so because of its relatively higher cost. 1 for the latter, particularly useful sources are del río 2008, mir-artigues 2012, and del río 2017. john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 7 a further limitation in scope is the paper’s focus on politics at the national level. spain has a relatively decentralized political system, and the regions (autonomous communities) have at times played important roles in both promoting and hindering the growth of renewable power. the authority to approve power plants of up to 50 megawatts (mw) of capacity has resided with regional governments, and many have engaged in extensive planning and offered their own promotional schemes. nevertheless, the principal impetus for the growth of renewable power, whether in the form of legislation, regulation, or financial support, has come from the central government, with regional policies tending to shape the outcomes mainly at the margins. 2. why promote renewable power in spain? successive spanish national governments have been motivated to promote renewable power for multiple reasons. these reasons have not been unique to spain, but they have been powerful motivations nonetheless. three general imperatives have stood out: the desire for greater security of energy supplies, environmental protection, and expected socioeconomic benefits. the security benefits of renewable energy were noted as early as the 1980 energy conservation law, the first major energy legislation adopted under the 1978 constitution, which aimed to reduce spain’s dependence on external sources of hydrocarbons (ley 82/1980). since then, they have been a constant theme in spain’s successive renewable energy plans. indeed, the importance of this motivation only grew in the 1980s, 1990s, and early 2000s, as spain’s dependence on fossil fuels from foreign sources steadily increased, both in absolute and relative terms, with rapidly rising energy consumption. until the economic crisis of 2008, fossil fuels accounted for roughly 50 percent of electricity generation, and the share of coal, the one fossil fuel that had been produced domestically, had been steadily declining, especially after imported natural gas was introduced in the early 2000s (british petroleum 2018). renewable energy would contribute to energy security not only by reducing imports but also by diversifying spain’s primary energy sources (pfer 1999, p. 1; per 2005, pp. 13, 332-34; per 2011, p. xxxiv). the environmental benefits of renewable energy gained prominence in the 1990s, especially after the adoption of the kyoto protocol, which committed spain to limiting the growth of its greenhouse gas emissions. along with energy conservation and efficiency, the substitution of renewable energy for fossil fuels, especially in the generation of electricity, was regarded as a key tool for reducing emissions of greenhouse and other noxious gases and for limiting the environmental impact of the energy system more generally. by one estimate, the renewable energy sources in place by 2010 were already reducing co2 emissions by more than 50 million tonnes per year, or approximately one-sixth of spain’s actual emissions, and about two-thirds of the reductions were attributable to renewable power (pfer 1999, pp. 9-15; per 2005, pp. 8-9, 23-24; per 2011, pp. 654-660). the potential socioeconomic benefits of renewable energy were also recognized by the john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 8 1990s, but the importance attributed to them grew in tandem with actual--and estimates of potential future--renewable energy deployments. the specific benefits were manifold and included technological development leading to greater industrial competitiveness, the creation of new businesses and jobs, especially in rural and remote areas, and thus regional development, and overall economic growth. indeed, the number of companies and jobs in the sector as well as its contribution to spain’s gdp grew steadily in the 1990s and 2000s. by 2011, the renewables sector was expected to support some 300,000 jobs and generate about 18 billion euros in wealth per year by 2020, with renewable power again contributing about twothirds of the total (pfer 1999, pp. 16-25; per 2005, pp. 14, 25-27, 337; per 2011, pp. 627-33, 646-52). underpinning these views about the benefits of renewable power was spain’s prodigious resource potential. from the beginning, spain’s solar resources were regarded as immense, as much as several terawatts, or much more than the country could ever use, thanks to spain’s extensive land area and some of the highest levels of solar radiation in europe. for many years, the challenge consisted primarily of bringing down the price of pv installations to the point where the resource could be exploited cost-effectively. meanwhile, estimates for spain’s wind power potential marched steadily upward, from about 15 gigawatts (gw) in 1999 to more than 40 gw in 2005 and then to as many as 330 gw in 2011, as technology improved over time (pfer 1999, pp. 58-59; per 2005, pp. 42, 161; per 2011, pp. xxxvi, 237, 379-81). further shaping spanish policy on renewable power have been a series of national targets generated by the european union. a 1997 white paper established a goal of providing 12 percent of primary energy consumption from renewable sources by 2010. a 2001 directive set a target of 29.4 percent of electricity from renewables by 2010. and a 2009 directive required spain to generate 20 percent of its final energy consumption from renewable sources. these targets in turn were incorporated into the series of energy plans developed by successive governments between 1999 and 2015. because of rising energy consumption, however, the specific targets for renewable power sources grew steadily during the first decade of the 2000s. for example, the 2010 target for wind more than doubled, from 8974 megawatts (mw) in 1999 to 20,155 mw in 2005, while that for solar pv jumped from 144 mw in 1999 to 400 mw in 2005. and in 2010, even higher targets were set–38,000 mw for wind and 8367 mw for solar pv--for 2020, although these were subsequently reduced by about one-quarter (pfer 1999; per 2005; paner 2010; minetur 2015b). these consistent motivations have been the basis for a national consensus on the importance of promoting renewable power. with one principal exception, discussed in the last section, there has been little difference in the degree of support offered by the major parties2. as the following sections demonstrate, governments of different political orientations have pursued similar goals in similar ways. 2 see also del río 2008 (pp. 2917-18), which notes continuity and stability in the promotion of renewable power under different governments. john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 9 3. initial efforts to promote renewable power efforts to promote renewable energy date back at least as far as the energy conservation law of 1980, which was adopted at the height of the second oil shock (ley 82/1980). it was not until the following decade, however, that the socialist government of felipe gonzález (1982-1996) undertook the first comprehensive effort to promote renewable power generation. a 1994 regulation established a ‘special regime’ for new renewable power installations of up to 100 mw. qualifying installations would receive access to the grid and the right to sell any surplus production to power distribution companies at a price determined by a complex formula and under contracts lasting at least five years (rd 2366/1994). the special regime was further developed by the subsequent people’s party government of josé maría aznar (1996-2004), which initiated the liberalization of the spanish electricity market in consonance with eu directive 96/92/ec via the electricity sector act of 1997. although the regime was now limited to installations of up to 50 mw (and only 10 mw for hydroelectric plants), qualifying producers would receive a premium (prima) set at somewhere between 80 and 90 percent of the average electricity price, and possibly higher for solar power (ley 54/1997). over the following 10 years, successive governments of both parties made repeated efforts to fine tune the details of the special regime so that spain would meet its growing renewable power targets3. in 1998, the aznar government adopted a regulation that sought to incentivize investment in several ways. it gave producers a choice between two alternative forms of support--either a fixed premium on top of the market price or a fixed total price (feed-in tariff or fit)–that could be adjusted annually. it waived any time limit on how long producers could receive support. and it differentiated support levels by technology (solar, wind, geothermal, hydro, biomass, etc.), offering particularly large subsidies to solar installations (rd 2818/1998). in 2004, shortly before the spanish general election that year, the aznar government made a second attempt to get the incentives right. the new regulation created an additional incentive for renewable producers to sell their electricity directly to the wholesale market. it also sought to stabilize subsidy levels and create more certainty for investors by guaranteeing support for the lifetime of the plant and reviewing fits and premiums only every four years (instead of annually, as under the previous regulation). and it raised the threshold for the highest level of support for solar pv installations, from 5 kilowatts (kw) to 100 kw (rd 436/2004). a third effort was made in 2007 by the socialist government of josé luis rodríguez zapatero (2004-2011). under the existing regulation, the tariffs, premiums, and other incentives fluctuated with the average electricity tariff and there were no lower or upper bounds, creating uncertainty and risks for both investors, the government, and consumers, who would ultimately pay for the support. to address the problem, a new regulation delinked support levels from electricity prices while setting caps and floors for the amount of support that generators opting for the premium would receive. although the premium option was eliminated for solar pv, it continued to receive by far the highest fits, and the fit for pv installations of between 100 kw 3 for additional details, see especially del río 2008 and del río and mir-artigues 2012. john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 10 and 10 mw was increased by 82 percent, or to nearly the level of the smallest installations (rd 661/2007). figure 1. renewable power capacity (megawatts). source: cnmc. thanks in no small part to these consistent efforts to promote renewable power by successive governments, the amount of generating capacity grew substantially between 1997 and 2010 (figure 1). wind was the first technology to take off. by 2002, spain had more wind power capacity than any country but germany, and it remained in the top three through the end of the decade, when it nearly achieved the much higher target of 20,155 mw set in 2005 (british petroleum 2018). solar pv took longer to gain momentum, but investment in it eventually exploded. in 2007, when capacity almost quintupled, and in 2008, it increased by 400 percent. suddenly, spain ranked second in the world in pv capacity, and it exceeded its 2010 target by nearly ten-fold. 4. the tariff deficit and spain’s response primarily because of the growth in first wind and then solar pv power, spain experienced a rapid rise in the amount of electricity generated from renewable sources, which roughly doubled during the first decade of the 2000s. because of the decline in electricity consumption occasioned by the economic crisis at the end of the decade, moreover, the relative share of 420 886 1686 2296 3508 5066 6324 8532 10095 11897 14536 16323 18855 19700 21063 22630 22994 23014 23029 23063 23091 23148 1 1 2 2 4 7 11 23 47 146 690 3398 3391 3829 4233 4523 4646 4654 4666 4673 4678 4700 0 5000 10000 15000 20000 25000 wind solar pv john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 11 renewable power grew even faster. by 2010, approximately one-third of all electricity generated in spain came from renewable sources (cnmc 2013). the growth in renewable power came at a substantial cost, however, in the form of an even more rapid increase in the support renewable generators received. these payments jumped from less than 1 billion euros per year in the mid-2000s to more than 5 billion euros per year after 2010, of which approximately half went to solar pv installations because of their much higher support levels. these payments, in turn, contributed to a large and expanding ‘tariff deficit’ (déficit de tarifa), which reflected the difference between the regulated costs of the electric power system and the access tariffs paid by consumers. between 2008 and 2011, the tariff deficit grew at an average rate of around 5 billion euros per year, reaching some 30 billion euros. this amount represented approximately three percent of spain’s gdp, at a time when the government was under intense pressure to reduce spending (cnmc 2013; cnmc 2017). to be sure, support for renewable power was not the only cause of the tariff deficit. a more fundamental reason was the government’s policy, in place since the liberalization of the power market in 1997, of controlling electricity prices to protect consumers against sudden increases. nor were payments to renewables the only regulated costs of the system. others included transmission and distribution, subsidies for non-peninsular systems and energy poverty, the annuities required to cover the deficits of previous years, and support for the non-renewable power sources– co-generation and waste – in the special regime. as a result, the annual deficit had reached high levels even in years before renewables took off. but by the late 2000s, support for renewable power had become the single largest component of the regulated costs, and, for that reason, the subject of particular scrutiny, with solar pv heading the list of perceived culprits. (espinosa 2013; gómez, dopazo, and fueyo 2016, 435; iea 2015, 129; unef 2013). it should also be noted that the growth of renewables also exerted downward pressure on wholesale electricity prices by displacing other forms of electricity generation with higher marginal costs, especially fossil fuels. a substantial technical literature exists on the overall impact of renewables on electricity prices, and whether the cost of support was greater or less than this positive market effect. in fact, fine-grained studies have found that it varied substantially by technology, with wind yielding a net gain and solar pv a net loss (sáenz, del río, and vizcaino 2008; ciarreta, espinosa, and pizarro 2014; costa-campi and trujillo-baute 2015; carvalho figueriredo and pereira da silva 2018). in any case, this benefit tended to be overlooked in the rush to address the burgeoning tariff deficit. a further impetus to take action was provided by the negative impact of renewables on other power sources. the boom, along with a decline in demand associated with the economic crisis, resulted in a substantial overcapacity in spain’s generation system, which saw the overall average load factor decrease from 44 percent in 2000 to 31 percent in 2012. that, in combination with the downward pressure on wholesale electricity prices, threatened the finances of the major utilities. particularly hard hit were the 26 gw of natural gas-fired combined cycle (gfcc) plants built after 2000 in anticipation of substantial growth in spain’s electricity demand, making gas the single largest source of installed capacity. this brand new fleet saw its john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 12 overall load factor drop from around 50 percent in the mid-2000s to below 20 percent in 2011 and then continue to fall, almost to single digits in 2014 (moreno and martínez-val 2011; mcgovern 2011; gómez, dopazo, and fueyo 2016, 434, 436, 444). as a result of these developments, the promotion of renewable power went, in just a few years, from enjoying broad public support, or at least tolerance, to facing substantial opposition. particularly outspoken in their criticism were large electricity consumers and the major utilities, whose returns on investment in non-renewable generating capacity was being put at risk (linares and labandeira 2013). these various pressures in turn prompted similar policy responses by successive governments led by different parties. beginning as early as 2008, the zapatero government took steps to control the cost of renewable power. when these actions proved inadequate, the subsequent rajoy government adopted even stronger measures, culminating in a complete overhaul of the renewable support system4. one set of common efforts focused on slowing the growth in renewable power support (del río 2016). targeting solar pv initially, the zapatero government limited the size of the plants that could receive subsidies, established an overall quota of new capacity per quarter, cut the size of the fit, and added a mechanism for automatically reducing the fit if more than 75 percent of the quarterly quota was met (rd 1578/2008). then, it further reduced the fit for new pv installations by applying ‘correction’ factors of up to 55 percent and extended similar measures to renewable power installations of other types (rd 1565/2010; rdl 6/2009; rd 1614/2010). unsatisfied with the results of these efforts, the new rajoy government took the drastic step of imposing an open-ended moratorium on support for new renewable power plants as one of its first actions, in january 2012 (rdl 1/2012)5. governments of both parties also adopted measures aimed at generating additional revenues to help reduce the tariff deficit directly. the zapatero government imposed a grid access charge for all power producers (rdl 14/2010). the rajoy government then established a generation charge and created new taxes on nuclear spent fuel and the use of coal and oil to generate electricity (ley 15/2012). finally, neither government shied away from controversial measures to modify, retroactively, the terms of compensation for pre-existing renewable power plants. the zapatero government limited the years solar pv installations could receive the fit and capped the number of equivalent operating hours that were eligible for compensation. for its part, the rajoy government revised the formula for updating the tariffs for all renewable sources in a way that was expected to result in slower growth--and possibly even negative growth, in real terms—and eliminated the option of receiving a premium on top of the wholesale price of electricity (rd 1565/2010; rdl 14/2010; rdl 2/2013). ultimately, in 2013 and 2014, the rajoy government replaced the fit/premium model for promoting renewable power with an entirely new legal and regulatory framework involving primarily investment-based support. at that point, the tariff deficit had continued to grow, despite the previous efforts by both governments to reduce it, and spain was under pressure from both 4 for a much more detailed discussion, see del río and mir-artigues 2012. 5 table a5 in del río and mir-artigues (2014, p. 51) shows the evolution of the fit for pv from 1998 to 2011. john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 13 the eu and the imf to take more decisive action to reform the electricity sector (iea 2015, pp. 100-102; espinosa 2013, pp. 1-2). in a series of laws and regulations (rdl 9/2013; ley 24/2013; rd 413/2014), the rajoy government abolished the special regime and required all renewable producers to sell their power directly to the wholesale electricity market. instead of a fit or premium based on the amount of power generated, they would receive, if necessary, a subsidy designed to ensure a reasonable profitability (rentabilidad razonable) on their investments, which would closely track the amount of installed capacity, not output (iea 2015, pp. 129-130; espinosa 2013, p. 35; del río 2016, pp. 9-12). all in all, these measures contributed to a significant slowdown and then a virtual halt in the deployment of new renewable generating capacity. from 2009 to 2013, wind power grew at roughly half the previous rate and then stalled at about 23 gw. over the same period, solar pv capacity rose by just 1250 mw before plateauing at about 4.6 gw. the boom in renewable power was over6. 5. the revival of renewable power this pause would be only temporary, however. indeed, hardly had the new legal and regulatory framework been put in place in 2013 and 2014 than pressure began to mount for a further increase in spain’s renewable power output. this pressure would eventually result in a series of auctions through which the government would authorize nearly 9 gw of new renewable capacity, to be completed by 2020, most of which would be eligible for remuneration under the new system. although spain had been meeting its annual renewable power targets, by 2015 it appeared that more capacity would be needed to reach the overall goal of 20 percent renewable energy in the country’s final energy consumption by 2020 (rd 947/2015; ec 2015). the 2011 renewable energy plan had set targets–35,750 mw of wind and 7250 mw of solar pv--that substantially exceeded the actual capacity, and even spain’s 2015 energy planning document anticipated the need for nearly 8 gw in additional wind and solar pv capacity by 2020 (per 2011, p. 469, minetur 2015b, p. 28). at the same time, the rajoy government desired to demonstrate its support for renewable energy prior to the general elections scheduled for december 2015 (del río 2016, pp. 12, 26). thus, widespread agreement existed on the need to increase spain’s renewable capacity. the principal question was how to do so at the lowest possible cost (del río 2017, p. 17). one promising approach was suggested by the guidelines for state aid issued by the eu in 2014, which proposed that members use auctions in the future to provide support for new renewable energy installations (ec 2014). in addition, the new electricity sector law and the subsequent regulation concerning renewable power had provided for the use of ‘competitive contests’ (concurrencia competitiva) to promote more renewable production when needed to meet eu targets or to reduce foreign energy dependence (ley 24/2013; rd 413/2014). 6 del río and mir-artigues (2012, p. 5562) also point to reduced access to credit following the economic crisis as a cause of the stagnation, but that alone cannot explain the abrupt halt in deployment that occurred after 2011. john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 14 as the 2015 elections approached, the rajoy government issued a further set of regulations and statements that laid the necessary groundwork for spain to proceed with its first auction (subasta). under the terms set by the government, the auction would be limited to small amounts of wind (500 mw) and biomass (200 mw) capacity; solar power in particular was excluded. and in contrast to most renewable power auctions held elsewhere, but in keeping with the new emphasis on investment-based support, bidders would offer a discount on the rate of return for the initial investment in a standard reference plant. government support would be provided only if the market price fell below the level needed to achieve the discounted rate of return, and winning bidders would have to complete their projects by the beginning of 2020 (rd 947/2015; minetur 2015a; minetur 2015c)7. this initial, very modest auction took place in january 2016, and was regarded as a success. the volume of wind power bids totaled five times the amount that could be awarded, and all the winning bids offered a 100 percent discount on the rate of return, meaning the government would never have to provide remuneration to the successful proposals. nevertheless, the renewable energy industry criticized the auction for its small size and the neglect of other technologies, especially solar pv (appa 2016; del río 2016, pp. 31-34; del río 2017, pp. 1820; minetur 2016). partly in response to these criticisms and in view of the continuing need to raise renewable power output by 2020, the government announced in mid-2016 that it would a hold a second, much larger auction. because spain needed to hold a second round of national elections, however, the details were not worked out until early the following year, after a new rajoy government had finally been seated, and the auction did not take place until may 2017. in addition to the much greater volume--up to 3000 mw--the auction was open to all technologies (rd 359/2017; minetur 2017a; minetur 2017b). once again, the auction was greatly oversubscribed, this time by approximately 300 percent, with bids roughly evenly divided between wind and solar pv projects. and as before, all the winning bids offered the maximum possible discount, although it was set at less than 100 percent of the investment this time. because of a provision that ties would be broken in favor of projects with greater hours of operation, however, virtually all of the capacity was awarded to wind projects, triggering an outcry from the solar pv sector (unef 2017; minetur 2017c). then, just days later, the government announced that yet another large auction would be held that summer. based on the experience of the may auction, officials realized that much more capacity could be available at a deep discount in the guaranteed rate of return, and there was no time to spare if new capacity were to be approved soon enough to come on line before 2020. because by far the greatest potential for low cost capacity lay with wind and pv, however, this third auction would be limited to those two technologies (rd 650/2017; minetur 2017d). the july auction was also massively oversubscribed. this time, however, changes in the parameters made pv relatively more attractive. some 3900 mw of pv and 1100 mw of wind were awarded at the maximum discount, making it the largest renewable power auction to date 7 a detailed description of the auctions can be found in del río 2016, pp. 11-20. john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 15 in terms of capacity (minetur 2017e). overall, the three auctions appeared to be a great success. if all the awarded capacity were built, spain would nearly double its solar pv capacity and would increase that of wind by about 20 percent. to be sure, some observers expressed concerns. one was that much of the awarded capacity might never materialize, because the winning offers appeared too low to be profitable, the result of strategic bidding based on the assumption that the last accepted bid – which would determine the rate of return would be higher. second, the short time frame for completing projects increased the potential for bottlenecks in the supply chain, especially for wind (del río 2016; inspiriata 2017). and the deployment of much new capacity was threatened by the scheduled expiration of many existing permits for grid access and connection at the end of 2018 as well as insufficient connection capacity in some regions (ojea 2018a; ojea 2018b). over the next two years, however, the obstacles were largely overcome, especially as the socialist government installed in june 2018 made the success of the auctions a priority. among other things, the new government extended the expiration dates of grid connection permits and, along with regional governments, took steps to expedite the approval processes (ojea 2018c; epe 2018a; epe 2018b; rdl 15/2018). partly as a result, the completion of new power plants surged in 2019. by the end of october, more than 3 gw of new capacity had been installed, and additional projects were coming on line at an accelerating rate as the end-of-year deadline approached. as much as 5 gw of new capacity was expected to be in place by then – what would constitute a single year record in spain and most of the rest of the capacity awarded in the auctions was expected to follow in 2020. one reason for the lag was that some developers were intentionally delaying the completion of their projects while attempting to secure financing on more favorable terms, even if it meant losing the guarantees that they had been required to deposit (roca 2019; ojea 2019b; epe 2019a). 6. future prospects for renewable power in spain what are the longer-term prospects for renewable power in spain, once the new capacity authorized by the auctions has been largely installed? the motives – security of supply, environmental protection, and socioeconomic benefits – that, along with spain’s abundant resource potential, shaped spanish policy in the past are likely to remain strong. indeed, they are being reinforced by important external developments. thus we should expect the recent revival of renewable power, especially solar pv and wind, to continue over the next decade. what will change is the role of the spanish government, which is shifting from providing financial support to renewable installations to facilitating private investment. 6.1. future prospects for renewable power in spain one development likely to drive additional renewable power deployment is the adoption of new, higher targets. in 2018, the european union established a framework for the promotion of john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 16 renewable energy over the next decade (ec 2018a). the overall binding target for 2030 of at least 32 percent represented a substantial increase over the 2020 goal. as a first step, specific national objectives would be set by the eu member states themselves. in february 2019, after months of speculation, the sánchez government provided spain’s initial response in the form of a long-awaited draft for a law on climate change and the energy transition (miteco 2019a) and the eu-mandated integrated national plan for energy and climate for 2021-2030 (plan nacional integrado de energía y clima 2021-2030, pniec) (miteco 2019b). the latter set an ambitious overall renewable energy target of 42 percent of final energy consumption by 2030, more than double the target for 2020, with no less than 74 percent of electricity coming from renewable sources. how much renewable generating capacity was spain likely to require? a first glimpse of what might be needed was provided by the report of the expert commission that was established in 2017 to analyze scenarios that would lead to the achievement of likely european energy and climate objectives. the report, issued in april 2018, employed a somewhat lower baseline scenario for 2030 in which 62 percent of electricity came from renewable sources. to achieve that level, spain would need a total of 31,000 mw of wind capacity and 47,150 mw of pv (comisión 2018, p. 8). given its more ambitious renewable electricity goal, the draft pniec set even higher targets. it called for 50,258 mw in wind, 36,882 mw in solar pv, and 7303 mw in csp power generating capacity (miteco 2019b). to achieve these objectives, the government would hold auctions of at least 3 gw in renewable capacity each year (miteco 2019a), although the details of the auctions remained to worked out through late 2019, at least in part because of the delay in forming a new government following the april national elections (epe 2019b). 6.2. falling costs of renewable power plants a second important development is the steady decline in the cost of renewable power, especially solar pv, to the point at which it can compete with conventional power sources without any financial support. by 2019, the cost of pv modules and entire pv systems had plunged by 90 percent and 80 percent, respectively, in a decade. as a result, the levelized cost of electricity in parts of spain had dropped as low as 24 e/mwh, well below the wholesale market price, and it was expected to continue to fall (vartiainen et al. 2019). the sharp decline in costs is one reason for the success of the recent auctions, in which winners all submitted the lowest possible bids. but it is also behind a development of perhaps even greater potential to transform the renewable power landscape in spain: the decision of a many actors to invest in renewable power without reliance on any government financial support. in a growing number of cases, large electricity consumers have entered into power purchase agreements (ppas) with renewable producers, whereby the former will buy electricity directly from the latter rather than from traditional utilities. by offering long-term contracts at fixed prices, ppas enable consumers to hedge their electricity costs in the face of unpredictable and john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 17 generally rising fossil fuel prices while ensuring producers and their financial backers of a steady income stream. in addition, such agreements can help companies burnish their environmental image by using non-co2 producing electricity to power their operations (bellini 2017; gómez 2019). by late 2019, ppas involving some 1500 mw of pv alone had been signed (pérez galdón 2019). other cases have involved purely speculative merchant projects that would sell their electricity directly to the wholesale market. and in at least one case, the project developer entered into a financial hedge arrangement that would lock in prices for a number of years (ellomay capital 2018). as a result, spain has seen a steady drumbeat of announcements for new solar pv plants in particular, beginning in 2017, that were unrelated to the auctions. the first ppa was signed in july 2017, and by early 2018, according to one estimate, a new ppa was being reached every week (edpr news 2017). by mid-2018, some 23 gw in pv projects beyond those awarded in the auctions had submitted applications for approval (monforte 2018; parnell 2018). and by october 2019, the spanish national grid operator, red eléctrica de españa, had granted access permits to new renewable projects totaling nearly 100 gw, with another 36 gw of capacity waiting for approval and some 58.5 gw in applications denied because of a lack of connection points or capacity on the grid (ree 2019; see also ojea 2019a). 6.3. a new role for government as the costs of new renewable power installations continue to decline, the need for government financial support should decline as well, unless wholesale prices drop so much as to make renewables unprofitable. instead, the primary role of the central government is likely to shift to removing obstacles that would otherwise inhibit private investment and the operation of market forces. indeed, the draft pniec anticipated that virtually all of the new investment in renewable energy will come from private sources (miteco 2019b, pp. 145-146). a good example of this shift is spain’s evolving policy toward demand-side generation (dsg), and specifically the version involving the production of electricity primarily for one’s own use, or self-consumption (autoconsumo). this has been the aspect of renewable power over which the major political parties have been most deeply divided. in the late 1990s and 2000s, renewable support policy in spain gave priority to utility scale power plants designed tp inject their output into the grid. it was not until 2010 that selfconsumption began to receive serious consideration in national planning. over the following year and a half, the zapatero government developed a set of regulations, but these were put on hold with the election of the rajoy government in late 2011 (mir-artigues 2013, pp. 668-669). following the new government’s moratorium on subsidies for new renewable installations in early 2012, many regarded self-consumption as a way to boost the fortunes of the struggling pv sector in particular. when the rajoy government finally revealed its own draft regulation in mid2013, however, it contained provisions that would make the deployment of dsg with selfconsumption unprofitable. of particular concern was a special back-up charge (peaje de john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 18 respaldo) on self-consumed electricity. although intended to ensure that self-consumers would help pay for the fixed costs of the power system, the charge had the effect of making selfconsumed power more expensive than electricity bought from the grid (unef 2013; mirartigues, del río, and cerdá 2018). the proposed regulation was roundly criticized by environmentalists and the renewable power industry. even the national energy commission (cne) issued a report calling the proposal ‘discriminatory’ in comparison with other measures consumers could take to reduce their purchases from the grid (cne 2013). in the face of this criticism, the rajoy government waited until october 2015, shortly before the next national election, to issue a final regulation. this version contained most of the barriers found in the original draft, including the back-up charge (rd 900/2015). the conservative government continued to prioritize reducing the tariff deficit over promoting renewable power. the rajoy government’s approach to self-consumption prompted widespread political opposition. prior to the december 2015 general election, all the opposition parties signed a letter promising to eliminate the back-up charge if they were to win a majority in the congress, and early the following year they committed to removing the obstacles in the 2015 regulation. in 2017 and again in 2018, the opposition parties even introduced bills to promote selfconsumption. nevertheless, the rajoy government was able to use its control over the legislative process, even after losing its majority in the congress in 2015, to prevent any such proposals from advancing (kenning 2015; kenning 2016; fernández 2018). the prospects for self-consumption suddenly brightened in mid-2018, when the socialists took the reins of power following a successful vote of no confidence in the rajoy government. a substantial majority of representatives quickly voted to lift the obstacles to adopting a new law to promote self-consumption (ojea 2018d). in the fall, the sánchez government eliminated the back-up charge and simplified administrative procedures for approving self-consumption projects as part of a broader package of urgent measures to promote the energy transition (rdl 15/2018). and in april 2019, the government issued a comprehensive new regulation for selfconsumption that received a positive reception from the renewable power industry (rd 244/2019). as a result, the pv sector anticipated that the amount of such capacity would grow by 400 mw per year (unef 2019). nevertheless, the job of the government will not be completely confined to removing administrative, technical, economic, and fiscal barriers to investment. there will continue to be areas where a more active government role will be needed. an important one will be balancing supply and demand. unlike conventional sources of power, most renewable power is intermittent and cannot be simply dispatched as needed. thus, as the share of renewable power grows, there could increasingly be occasions when the amount of electricity available exceeds or falls short of demand. the government could help address this challenge in several ways. on the demand side, spain has recently established an interruptibility service whereby large consumers agree, for a fee, to reduce their usage by pre-determined amounts when ordered by the grid operator, but john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 19 there is much more the government can do to promote demand-side management more broadly, as recognized in the draft pniec (roldán, burgos, riquelme, and trigo 2016; miteco 2019b, pp. 67-69). similar opportunities exist on the supply side in the form of capacity mechanisms, whereby the system operator pays producers to create or maintain backup power plants that can be brought online as needed. alternatively, and of particular interest in the case of renewable power, the government could help to promote various forms of electricity storage. thus far, spain’s efforts have focused on pumped hydroelectric storage, but the draft pniec anticipated a growing reliance on battery storage on a large scale, possibly facilitated by the use of auctions, as well as thermal storage associated with csp installations (miteco 2019b, pp. 67, 170). finally, the government will be central to developing additional interconnections with the broader eu electricity market through france. because spain, and the iberian peninsula more generally, is an isolated ‘energy island,’ interconnections that enable spain to export and import electricity as needed are vital for balancing supply and demand. recently, spain doubled the amount of power it can export across its northern border, to nearly 3 gw or about six percent of peak power consumption, and spain and france plan to build an underwater interconnection in the bay of biscay with a capacity of 2 gw, to be completed by 2025, while two additional crossings in the pyrenees that could increase the total to 8 gw are under consideration. the ultimate goal is to bring spain’s total interconnection capacity, including that with portugal, up to the eu’s target for 2030 of 15 percent of spain’s electricity consumption (ec 2018b; miteco 2019b: 14, 28). 7. conclusions over more than thirty years, the fortunes of renewable power have waxed and waned in spain. in the 1990s and much of the 2000s, renewable power received substantial government support, primarily in the form of generous feed-in tariffs and premiums. then, for the better part of a decade, that support was steadily cut back and, in some respects, eliminated. in the last several years, however, renewable power has experienced a resurgence of support, although not to the extent previously enjoyed. the deployment of new renewable generating capacity has paralleled these fluctuations in support levels, first rising, then stagnating, and once again increasing in recent years. one striking aspect of this story is that it cannot simply be attributed to shifting political winds. although the center-left and center-right parties have differed on many issues, they have generally shared similar views when it comes to support for renewable power. both conservative and socialist governments backed the incentives that made possible the initial take off of renewable power. likewise, governments of both parties participated in the cuts that followed. and although less government support has been needed in recent years, thanks to the declining cost of renewable projects, both conservative and socialist governments have taken steps to ensure that spain deploys enough new generating capacity to meet its overall john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 20 renewable energy targets for 2020. policy on renewable power, whether supportive or restrictive, has generally enjoyed a broad consensus. of course, this finding raises the question of what will happen in the future, given the recent fragmentation and polarization of the party system. since 2015, spain has been ruled by minority governments that have been less able to pass new laws, such as the long promised law on climate change and energy transition. nevertheless, when it comes to renewable power, these political developments may matter less than they might have in the past, because of the changing role of government. in the future, thanks to the increasing cost competitiveness of renewables, government will be needed less to provide financial support, a potentially controversial role, than to facilitate the operation of market forces. and whatever the precise role that spanish governments end up playing, the overall trend is clear. future governments, regardless of their political orientations, are likely to preside over a prolonged and substantial increase in renewable power that reflects the country’s long-standing security, environmental, and socioeconomic motivations as well as its extensive renewable resources. that rising eu targets and declining costs point in the same direction will only reinforce these underlying forces. less clear is what lessons can be generalized from the spanish case. is it unique, or at least rare, in the extent to which renewable power has enjoyed – and may continue to enjoy -broad political support? a potentially fruitful line of research would be to examine where, when, and why support for renewables has tended to command political consensus. a quick look at the united states, for example, indicates that consensus can be elusive and that support for renewables, like the closely related issue of climate change, can be a divisive political issue. the u.s. case also suggests some potential hypotheses to explore with regard to the causes of such differences, such as the strength of interest groups that might benefit from or be harmed by the growth of renewable power and the scope and geographical distribution of renewable resources. suffice it to say, the examination of the politics of renewable power in comparative perspective is likely to yield fruitful insights. aknowledgements this work was supported by funding from the college of arts and sciences and the department of political science at georgia state university. i also wish to thank the many spanish experts in government, universities, research centers, industry associations, and advocacy groups who have helped me to understand this subject. nevertheless, i am responsible for any errors of fact or interpretation. references asociación de empresas de energías renovables (appa) (2016). el resultado de la subasta eléctrica añade más incertidumbre al sector removable (15 jan.). retrieved from https://www.appa.es/el-resultado-de-la-subasta-electrica-anade-mas-incertidumbre-aljohn s. duffield / european journal of government and economics 9(1), june 2020, 5-25 21 sector-renovable bellini, e. (2017). spanish pv could go merchant but market must change. pv magazine (3 nov.). retrieved from https://www.pv-magazine.com/2017/11/03/spanish-pv-could-gomerchant-but-market-must-change british petroleum. (2018). statistical review of world energy. retrieved from https://www.bp. com/en/global/corporate/energy-economics/statistical-review-of-world-energy.html carvalho figueriredo, n., and pereira da silva, p. (2018). the price of wind power generation in iberia and the merit-order effect. international journal of sustainable energy planning and management 15: 21–30. ciarreta, a., paz espinosa, m., and pisarro-irizar, c. (2014). is green energy expensive? empirical evidence from the spanish electricity market. energy policy 69: 205-215. https://doi.org/10.1016/j.enpol.2014.02.025 comisión de expertos de transición energética. (2018). análisis y propuestas para la descarbonización: resumen ejecutivo. retrieved from https://elperiodicodelaenergia.com /wp-content/uploads/2018/04/resumen-20180402-veditado.pdf comisión nacional de la energía (cne). (2013). informe 19/2013 (4 sept.). retrieved from https://www.cnmc.es/sites/default/files/1552840_7.pdf comisión nacional de los mercados y la competencia (cnmc). información mensual de estadísticas sobre las ventas de régimen especial (various months and years). retrieved from https://www.cnmc.es/estadisticas?hidtipo=12749 costa-campi, m.t., and trujillo-baute, e. (2015). retail price effects of feed-in tariff regulation. energy economics 51: 157-165. https://doi.org/10.1016/j.eneco.2015.06.002 del río gonzález, p. (2008). ten years of renewable electricity policies in spain: an analysis of successive feed-in tariff reforms. energy policy 36, 2917-2929. https://doi.org/10.1016 /j.enpol.2008.03.025 del río, p. (2016). implementation of auctions for renewable energy support in spain, report d7.1-es (march). retrieved from http://auresproject.eu/sites/aures.eu/files/media/documents /wp7_-_case_study_report_spain_1.pdf del río, p. (2017). assessing the design elements in the spanish renewable electricity auction: an international comparison, working paper 6/2017, real instituto elcano (17 april). retrieved from http://www.realinstitutoelcano.org/wps/wcm/connect/435f8ff1-8e2d-498b9c60-9ac1183d2fa0/dt6-2017-delrio-design-spanish-renewable-electricityauction.pdf?mod=ajperes&cacheid=435f8ff1-8e2d-498b-9c60-9ac1183d2fa0 del río, p., mir-artigues, p. (2012). support for solar pv deployment in spain: some policy lessons. renewable and sustainable energy reviews 16, 5557-5566. https://doi.org /10.1016/j.rser.2012.05.011 del río, p., mir-artigues, p. (2014). a cautionary tale: spain’s solar pv investment bubble, international institute for sustainable development (feb.). retrieved from https://www.iisd.org/gsi/sites/default/files/rens_ct_spain.pdf edpr news (2017). the edp group and calidad pascual sign the first power purchase agreement (ppa) (25 july). retrieved from https://www.edpr.com/en/news/edp-group-andcalidad-pascual-sign-first-power-purchase-agreement-ppa ellomay capital (2018). ellomay capital enters into a binding term sheet for a power financial hedge for the talasol project (24 jan.). retrieved from https://ellomay.com/pressreleases/ellomay-capital-enters-into-a-binding-term-sheet-for-a-power-financial-hedge-forthe-talasol-project el periódico de la energía (epe). (2018a). apoyo unánime del parlamento andaluz para simplificar la normativa de trámites administrativos para proyectos renovables (18 july). retrieved from https://elperiodicodelaenergia.com/apoyo-unanime-del-parlamento-andaluzpara-simplificar-la-normativa-de-tramites-administrativos-para-proyectos-renovables el periódico de la energía (pe). (2018b). el gobierno de aragón autoriza la construcción de 53 proyectos eólicos para cumplir con los plazos del ministerio (6 august). retrieved from https://elperiodicodelaenergia.com/el-gobierno-de-aragon-autoriza-la-construccion-de-53proyectos-eolicos-para-cumplir-con-los-plazos-del-ministerio https://www.pv-magazine.com/2017/11/03/spanish-pv-could-go-merchant-but-market-must-change https://www.pv-magazine.com/2017/11/03/spanish-pv-could-go-merchant-but-market-must-change https://www.bp.com/en/global/corporate/energy-economics/statistical-review-of-world-energy.html https://www.bp.com/en/global/corporate/energy-economics/statistical-review-of-world-energy.html https://doi.org/10.1016/j.enpol.2014.02.025 https://elperiodicodelaenergia.com/wp-content/uploads/2018/04/resumen-20180402-veditado.pdf https://elperiodicodelaenergia.com/wp-content/uploads/2018/04/resumen-20180402-veditado.pdf https://www.cnmc.es/sites/default/files/1552840_7.pdf https://www.cnmc.es/estadisticas?hidtipo=12749 https://doi.org/10.1016/j.eneco.2015.06.002 https://doi.org/10.1016/j.enpol.2008.03.025 https://doi.org/10.1016/j.enpol.2008.03.025 http://auresproject.eu/sites/aures.eu/files/media/documents/wp7_-_case_study_report_spain_1.pdf http://auresproject.eu/sites/aures.eu/files/media/documents/wp7_-_case_study_report_spain_1.pdf http://www.realinstitutoelcano.org/wps/wcm/connect/435f8ff1-8e2d-498b-9c60-9ac1183d2fa0/dt6-2017-delrio-design-spanish-renewable-electricity-auction.pdf?mod=ajperes&cacheid=435f8ff1-8e2d-498b-9c60-9ac1183d2fa0 http://www.realinstitutoelcano.org/wps/wcm/connect/435f8ff1-8e2d-498b-9c60-9ac1183d2fa0/dt6-2017-delrio-design-spanish-renewable-electricity-auction.pdf?mod=ajperes&cacheid=435f8ff1-8e2d-498b-9c60-9ac1183d2fa0 http://www.realinstitutoelcano.org/wps/wcm/connect/435f8ff1-8e2d-498b-9c60-9ac1183d2fa0/dt6-2017-delrio-design-spanish-renewable-electricity-auction.pdf?mod=ajperes&cacheid=435f8ff1-8e2d-498b-9c60-9ac1183d2fa0 https://doi.org/10.1016/j.rser.2012.05.011 https://doi.org/10.1016/j.rser.2012.05.011 https://www.iisd.org/gsi/sites/default/files/rens_ct_spain.pdf https://www.edpr.com/en/news/edp-group-and-calidad-pascual-sign-first-power-purchase-agreement-ppa https://www.edpr.com/en/news/edp-group-and-calidad-pascual-sign-first-power-purchase-agreement-ppa https://ellomay.com/press-releases/ellomay-capital-enters-into-a-binding-term-sheet-for-a-power-financial-hedge-for-the-talasol-project https://ellomay.com/press-releases/ellomay-capital-enters-into-a-binding-term-sheet-for-a-power-financial-hedge-for-the-talasol-project https://ellomay.com/press-releases/ellomay-capital-enters-into-a-binding-term-sheet-for-a-power-financial-hedge-for-the-talasol-project https://elperiodicodelaenergia.com/apoyo-unanime-del-parlamento-andaluz-para-simplificar-la-normativa-de-tramites-administrativos-para-proyectos-renovables https://elperiodicodelaenergia.com/apoyo-unanime-del-parlamento-andaluz-para-simplificar-la-normativa-de-tramites-administrativos-para-proyectos-renovables https://elperiodicodelaenergia.com/el-gobierno-de-aragon-autoriza-la-construccion-de-53-proyectos-eolicos-para-cumplir-con-los-plazos-del-ministerio https://elperiodicodelaenergia.com/el-gobierno-de-aragon-autoriza-la-construccion-de-53-proyectos-eolicos-para-cumplir-con-los-plazos-del-ministerio john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 22 el periódico de la energía (pe). (2019a). la fotovoltaica duplicará su potencia en 2020, hasta los 9.000 mw, y la eólica crecerá un 20%, según informa. (24 sept). retrieved from https://elperiodicodelaenergia.com/la-fotovoltaica-duplicara-su-potencia-en-2020-hasta-los9-000-mw-y-la-eolica-crecera-un-20-segun-informa el periódico de la energía (pe). (2019b). ribera: las nuevas subastas de renovables se aprobarán «en cuanto puedan». (21 nov.). retrieved from https://elperiodicodelaenergia.com/ribera-las-nuevas-subastas-de-renovables-se-aprobaranen-cuanto-puedan espinosa, m.p. (2013). an austerity-driven energy reform (oct. 2013), researchgate. retrieved from https://www.researchgate.net/publication/258809913_an_austerity-driven_energy_ reform european commission (ec). (2014). guidelines on state aid for environmental protection and energy 2014-2020 (2014/c 200/01). retrieved from https://eur-lex.europa.eu/legalcontent/en/txt/pdf/?uri=celex:52014xc0628(01)&from=en european commission (ec). (2015). renewable energy progress report. com(2015) 293 final (brussels: 15 june). retrieved from https://eur-lex.europa.eu/resource.html?uri=cellar: 4f8722ce-1347-11e5-8817-01aa75ed71a1.0001.02/doc_1&format=pdf european commission (ec). (2018a). europe leads the global clean energy transition (14 june). retrieved from http://europa.eu/rapid/press-release_statement-18-4155_en.htm european commission (ec). (2018b). european solidarity on energy (27 july). retrieved from http://europa.eu/rapid/press-release_ip-18-4621_en.htm fernández, m. (2018). la oposición (sin ciudadanos) lleva de nuevo al congreso una iniciativa para tumbar el 'impuesto al sol'. el boletin (10 may). retrieved from https://www.elboletin.com/noticia/162920/economia/la-oposicion-sin-ciudadanos-lleva-denuevo-al-congreso-una-iniciativa-para-tumbar-el-impuesto-al-sol.html gómez, a., dopazo, c., and fueyo, n. (2016). the “cost of not doing” energy planning: the spanish energy bubble. energy 101: 434-446. https://doi.org/10.1016/j.energy.2016.02.004 gómez, r. (2019). el auge de los ppas (14 jan.). retrieved from https://elperiodicodelaenergia.com/el-auge-de-los-ppas/ inspiriata. (2017). financing spanish renewables projects. retrieved from https://www.capitaliq.com/ciqdotnet/creditresearch/renderarticle.aspx?articleid=1990704 &sctartid=448388&from=cm&nsl_code=lime&sourceobjectid=10410209&sourcerevid=3& fee_ind=n&exp_date=20280213-22:39:51 international energy agency (iea). (2015). the energy policies of iea countries: spain: 2015 review. paris: iea. kenning, t. (2015). spanish government under fire after approving “sun tax”. pv tech (12 oct.). retrieved from https://www.pv-tech.org/news/spains_sun_tax_approved_to_avoid _development_of_pv kenning, t. (2016). spanish parliament rallies against punitive “sun tax”. pv tech (9 march). retrieved from https://www.pv-tech.org/news/spanish-sun-tax-on-brink-of-removal ley 82/1980, de 30 de diciembre, sobre conservación de energía (30 dec. 1980). retrieved from https://www.boe.es/diario_boe/txt.php?id=boe-a-1981-1898 ley 54/1997, de 27 de noviembre, del sector eléctrico (27 nov. 1997). retrieved from https://boe.es/buscar/doc.php?id=boe-a-1997-25340 ley 15/2012, de 27 de diciembre, de medidas fiscales para la sostenibilidad energética (27 dec. 2012). retrieved from https://www.boe.es/buscar/doc.php?id=boe-a-2012-15649 ley 24/2013, de 26 de diciembre, del sector eléctrico (26 dec. 2013). retrieved from https://www.boe.es/buscar/doc.php?id=boe-a-2013-13645 linares, p., and labandeira, x. (2013). renewable electricity support in spain: a natural policy experiment. economics for energy wp 04/2013. mcgovern, mike. (2011). ¿los eólicos prefieren el gas? energías renovables (6 june). retrieved from https://www.energias-renovables.com/eolica/iquest-los-eolicos-prefieren-elgas ministerio de industria, energía y turismo (minetur). (2015a). orden iet/2212/2015, de 23 de octubre (23 oct.). retrieved from https://www.boe.es/boe/dias/2015/10/24/pdfs/boe-a2015-11432.pdf https://elperiodicodelaenergia.com/la-fotovoltaica-duplicara-su-potencia-en-2020-hasta-los-9-000-mw-y-la-eolica-crecera-un-20-segun-informa https://elperiodicodelaenergia.com/la-fotovoltaica-duplicara-su-potencia-en-2020-hasta-los-9-000-mw-y-la-eolica-crecera-un-20-segun-informa https://elperiodicodelaenergia.com/ribera-las-nuevas-subastas-de-renovables-se-aprobaran-en-cuanto-puedan https://elperiodicodelaenergia.com/ribera-las-nuevas-subastas-de-renovables-se-aprobaran-en-cuanto-puedan https://www.researchgate.net/publication/258809913_an_austerity-driven_energy_reform https://www.researchgate.net/publication/258809913_an_austerity-driven_energy_reform https://eur-lex.europa.eu/legal-content/en/txt/pdf/?uri=celex:52014xc0628(01)&from=en https://eur-lex.europa.eu/legal-content/en/txt/pdf/?uri=celex:52014xc0628(01)&from=en https://eur-lex.europa.eu/resource.html?uri=cellar:4f8722ce-1347-11e5-8817-01aa75ed71a1.0001.02/doc_1&format=pdf https://eur-lex.europa.eu/resource.html?uri=cellar:4f8722ce-1347-11e5-8817-01aa75ed71a1.0001.02/doc_1&format=pdf http://europa.eu/rapid/press-release_statement-18-4155_en.htm http://europa.eu/rapid/press-release_ip-18-4621_en.htm https://www.elboletin.com/noticia/162920/economia/la-oposicion-sin-ciudadanos-lleva-de-nuevo-al-congreso-una-iniciativa-para-tumbar-el-impuesto-al-sol.html https://www.elboletin.com/noticia/162920/economia/la-oposicion-sin-ciudadanos-lleva-de-nuevo-al-congreso-una-iniciativa-para-tumbar-el-impuesto-al-sol.html https://doi.org/10.1016/j.energy.2016.02.004 https://elperiodicodelaenergia.com/el-auge-de-los-ppas/ http://www.osborneclarke.com/wp-content/uploads/2017/09/madrid-roundtable-background-reading.pdf https://www.capitaliq.com/ciqdotnet/creditresearch/renderarticle.aspx?articleid=1990704&sctartid=448388&from=cm&nsl_code=lime&sourceobjectid=10410209&sourcerevid=3&fee_ind=n&exp_date=20280213-22:39:51 https://www.capitaliq.com/ciqdotnet/creditresearch/renderarticle.aspx?articleid=1990704&sctartid=448388&from=cm&nsl_code=lime&sourceobjectid=10410209&sourcerevid=3&fee_ind=n&exp_date=20280213-22:39:51 https://www.capitaliq.com/ciqdotnet/creditresearch/renderarticle.aspx?articleid=1990704&sctartid=448388&from=cm&nsl_code=lime&sourceobjectid=10410209&sourcerevid=3&fee_ind=n&exp_date=20280213-22:39:51 https://www.pv-tech.org/news/spains_sun_tax_approved_to_avoid_development_of_pv https://www.pv-tech.org/news/spains_sun_tax_approved_to_avoid_development_of_pv https://www.pv-tech.org/news/spanish-sun-tax-on-brink-of-removal https://www.boe.es/diario_boe/txt.php?id=boe-a-1981-1898 https://boe.es/buscar/doc.php?id=boe-a-1997-25340 https://www.boe.es/buscar/doc.php?id=boe-a-2012-15649 https://www.boe.es/buscar/doc.php?id=boe-a-2013-13645 https://www.energias-renovables.com/eolica/iquest-los-eolicos-prefieren-el-gas https://www.energias-renovables.com/eolica/iquest-los-eolicos-prefieren-el-gas https://www.boe.es/boe/dias/2015/10/24/pdfs/boe-a-2015-11432.pdf https://www.boe.es/boe/dias/2015/10/24/pdfs/boe-a-2015-11432.pdf john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 23 ministerio de industria, energía y turismo (minetur). (2015b). planificatión energética: plan de desarrollo de la red de transporte de energía 2015-2020. retrieved from https://www.mincotur.gob.es/energia/planificacion/planificacionelectricidadygas/desarrollo20 15-2020/documents/planificaci%c3%b3n%202015_2020%20%202016_11_28%20v publicaci%c3%b3n.pdf ministerio de industria, energía y turismo (minetur). (2015c). resolución de 30 de noviembre de 2015 (30 nov.). retrieved from https://www.aeeolica.org/uploads/documents/10228-resolucirn-de-30-de-noviembre-de2015-por-la-que-se-convoca-la-subasta-para-la-asignacirn-del-rrgimen-retributivoespecrfico-a-nuevas-instalaciones-de-produccirn-de-energra-elrctrica-a-partir-de-bioma.pdf ministerio de industria, energía y turismo (minetur). (2016). resolución de 18 de enero de 2016. retrieved from http://www.subastasrenovables.omie.es/files/boe-a-2016-552_0.pdf ministerio de industria, energía y turismo (minetur). (2017a). orden etu/315/2017, de 6 de abril (8 april). retrieved from https://www.boe.es/diario_boe/txt.php?id=boe-a-2017-3880 ministerio de industria, energía y turismo (minetur). (2017b). resolución de 10 de abril de 2017 (12 april). retrieved from https://www.boe.es/boe/dias/2017/04/12/pdfs/boe-a-20174094.pdf ministerio de industria, energía y turismo (minetur). (2017c). resolución de 19 de mayo de 2017 (26 may). retrieved from http://www.subastasrenovables.omie.es/files/resultados_i _subasta_ifer_0.pdf ministerio de industria, energía y turismo (minetur). (2017d). orden etu/615/2017, de 27 de junio (28 june). retrieved from https://www.boe.es/boe/dias/2017/06/28/pdfs/boe-a-20177389.pdf ministerio de industria, energía y turismo (minetur). (2017e). resolución de 27 de julio de 2017 (28 july). retrieved from http://www.subastasrenovables.omie.es/files/boe-a-20178997.pdf ministerio para la transición ecológica (miteco). (2019a). anteproyecto de ley de cambio climático y transición energética (20 feb.). retrieved from https://www.miteco.gob.es/images/es/1anteproyectoleyccyte_tcm30-487336.pdf ministerio para la transición ecológica (miteco). (2019b). borrador del plan nacional integrado de energía y clima 2021-2030 (feb.). retrieved from https://www.miteco.gob.es/images/es/documentoparticipacionpublicaborradordelplannaciona lintegradodeenergiayclima2021-2030_tcm30-487344.pdf mir-artigues, p. (2012). economía de la generación eléctrica solar: la regulación fotovoltaica y solar termoelétrica en espańa. civitas/thomson reuters. https://doi.org/10.32796/cice. 2012.83.6039 mir-artigues, p. (2013). the spanish regulation of the photovoltaic demand-side generation. energy policy 63, 664-273. https://doi.org/10.1016/j.enpol.2013.09.019 mir-artigues, p., del río, p., and cerdá, e. (2018). the impact of regulation on demand-side generation. energy policy 121, 286-291. https://doi.org/10.1016/j.enpol.2018.05.008 moreno, f., and martínez-val, j. (2011). collateral effects of renewable energies deployment in spain: impact on thermal power plants performance and management. energy policy 39:6561-6574. doi: https://doi.org/10.1016/j.enpol.2011.07.061 monforte, c. (2018). la fotovoltaica tramita proyectos por 20.000 millones de inversion. cinco días (19 march). retrieved from https://cincodias.elpais.com/cincodias/2018/03/16 /companias/1521226891_225916.html ojea, l. (2018a). las renovables temen que el rd de acceso y conexión no llegue a tiempo y pierdan sus permisos y avales. el periódico de la energía (4 july). retrieved from https://elperiodicodelaenergia.com/las-renovables-temen-que-el-rd-de-acceso-y-conexionno-llegue-a-tiempo-y-pierdan-sus-permisos-y-avales ojea, l. (2018b). todos los proyectos de renovables ganadores de las subastas estarán en funcionamiento el 1 de enero de 2020. el periódico de la energía (24 sept.). retrieved from https://elperiodicodelaenergia.com/todos-los-proyectos-de-renovables-ganadores-de-lassubastas-estaran-en-funcionamiento-el-1-de-enero-de-2020 https://www.mincotur.gob.es/energia/planificacion/planificacionelectricidadygas/desarrollo2015-2020/documents/planificaci%c3%b3n%202015_2020%20%202016_11_28%20vpublicaci%c3%b3n.pdf https://www.mincotur.gob.es/energia/planificacion/planificacionelectricidadygas/desarrollo2015-2020/documents/planificaci%c3%b3n%202015_2020%20%202016_11_28%20vpublicaci%c3%b3n.pdf https://www.mincotur.gob.es/energia/planificacion/planificacionelectricidadygas/desarrollo2015-2020/documents/planificaci%c3%b3n%202015_2020%20%202016_11_28%20vpublicaci%c3%b3n.pdf https://www.aeeolica.org/uploads/documents/10228-resolucirn-de-30-de-noviembre-de-2015-por-la-que-se-convoca-la-subasta-para-la-asignacirn-del-rrgimen-retributivo-especrfico-a-nuevas-instalaciones-de-produccirn-de-energra-elrctrica-a-partir-de-bioma.pdf https://www.aeeolica.org/uploads/documents/10228-resolucirn-de-30-de-noviembre-de-2015-por-la-que-se-convoca-la-subasta-para-la-asignacirn-del-rrgimen-retributivo-especrfico-a-nuevas-instalaciones-de-produccirn-de-energra-elrctrica-a-partir-de-bioma.pdf https://www.aeeolica.org/uploads/documents/10228-resolucirn-de-30-de-noviembre-de-2015-por-la-que-se-convoca-la-subasta-para-la-asignacirn-del-rrgimen-retributivo-especrfico-a-nuevas-instalaciones-de-produccirn-de-energra-elrctrica-a-partir-de-bioma.pdf http://www.subastasrenovables.omie.es/files/boe-a-2016-552_0.pdf https://www.boe.es/diario_boe/txt.php?id=boe-a-2017-3880 https://www.boe.es/boe/dias/2017/04/12/pdfs/boe-a-2017-4094.pdf https://www.boe.es/boe/dias/2017/04/12/pdfs/boe-a-2017-4094.pdf http://www.subastasrenovables.omie.es/files/resultados_i_subasta_ifer_0.pdf http://www.subastasrenovables.omie.es/files/resultados_i_subasta_ifer_0.pdf https://www.boe.es/boe/dias/2017/06/28/pdfs/boe-a-2017-7389.pdf https://www.boe.es/boe/dias/2017/06/28/pdfs/boe-a-2017-7389.pdf http://www.subastasrenovables.omie.es/files/boe-a-2017-8997.pdf http://www.subastasrenovables.omie.es/files/boe-a-2017-8997.pdf https://www.miteco.gob.es/images/es/1anteproyectoleyccyte_tcm30-487336.pdf https://www.miteco.gob.es/images/es/documentoparticipacionpublicaborradordelplannacionalintegradodeenergiayclima2021-2030_tcm30-487344.pdf https://www.miteco.gob.es/images/es/documentoparticipacionpublicaborradordelplannacionalintegradodeenergiayclima2021-2030_tcm30-487344.pdf https://doi.org/10.32796/cice.2012.83.6039 https://doi.org/10.32796/cice.2012.83.6039 https://doi.org/10.1016/j.enpol.2013.09.019 https://doi.org/10.1016/j.enpol.2018.05.008 https://doi.org/10.1016/j.enpol.2011.07.061 https://cincodias.elpais.com/cincodias/2018/03/16/companias/1521226891_225916.html https://cincodias.elpais.com/cincodias/2018/03/16/companias/1521226891_225916.html https://elperiodicodelaenergia.com/las-renovables-temen-que-el-rd-de-acceso-y-conexion-no-llegue-a-tiempo-y-pierdan-sus-permisos-y-avales https://elperiodicodelaenergia.com/las-renovables-temen-que-el-rd-de-acceso-y-conexion-no-llegue-a-tiempo-y-pierdan-sus-permisos-y-avales https://elperiodicodelaenergia.com/todos-los-proyectos-de-renovables-ganadores-de-las-subastas-estaran-en-funcionamiento-el-1-de-enero-de-2020 https://elperiodicodelaenergia.com/todos-los-proyectos-de-renovables-ganadores-de-las-subastas-estaran-en-funcionamiento-el-1-de-enero-de-2020 john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 24 ojea, l. (2018c). el gobierno agilizará los trámites de impacto ambiental de las nuevas plantas de renovables para llegar al 20% en 2020. el periódico de la energía (22 june). retrieved from https://elperiodicodelaenergia.com/el-gobierno-agilizara-los-tramites-de-impactoambiental-de-las-nuevas-plantas-de-renovables-para-llegar-al-20-en-2020 ojea, l. (2018d). la ley que acabará con el ‘impuesto al sol’ tendrá que esperar a septiembre. el periódico de la energía (25 june). retrieved from https://elperiodicodelaenergia.com/laley-que-acabara-con-el-impuesto-al-sol-tendra-que-esperar-a-septiembre ojea, l. (2019a). locura renovable en españa. el periódico de la energía (31 jan.). retrieved from https://elperiodicodelaenergia.com/locura-renovable-en-espana-quieren-instalarsemas-de-87-gw-la-mitad-ya-con-el-permiso-de-acceso-concedido-por-ree ojea, l. (2019b). los proyectos renovables ganadores de las subastas prefieren perder los avales a cambio de firmar un ppa. el periódico de la energía (26 sept.). retrieved from https://elperiodicodelaenergia.com/los-proyectos-renovables-ganadores-de-las-subastasprefieren-perder-los-avales-a-cambio-de-firmar-un-ppa parnell, j. (2018). economics not tenders driving spain’s solar resurgence. pv tech (23 april). retrieved from https://www.pv-tech.org/news/economics-not-tenders-driving-spains-solarresurgence pérez galdón, b. (2019). don rodrigo inicia la nueva era de plantas solares sin subvención. cinco días (nov. 9). retrieved from https://cincodias.elpais.com/cincodias/2018/11/09/companias/1541758756_915913.html plan de acción nacional de energías renovables de españa (paner) 2011 – 2020 (2010). retrieved from https://www.mincotur.gob.es/energia/desarrollo/energiarenovable /documents/20100630_paner_espanaversion_final.pdf plan de energías renovables en españa (per) 2005-2010 (2005). retrieved from http://www .idae.es/uploads/documentos/documentos_per_2005-2010_8_de_gosto2005_completo.(modificacionpag_63)_copia_2_301254a0.pdf plan de energías renovables (per) 2011-20 (2011). retrieved from http://www.idae.es/file /9712/download?token=6moebdcb plan de fomento de las energías renovables en españa (pfer) (1999). retrieved from http:// www.idae.es/uploads/documentos/documentos_4044_pfer2000-10_1999_1cd4b316.pdf real decreto (rd) 2366/1994, de 9 de diciembre (9 dec. 1994). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-1994-28980 real decreto (rd) 2818/1998, de 23 de diciembre (23 dec. 1998). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-1998-30041 real decreto (rd) 436/2004, de 12 de marzo (12 march 2004). retrieved from https:// www.boe.es/buscar/doc.php?id=boe-a-2004-5562 real decreto (rd) 661/2007, de 25 de mayo (25 may 2007). retrieved from https:// www.boe.es/buscar/act.php?id=boe-a-2007-10556 real decreto (rd) 1578/2008, de 26 de septiembre (26 sept. 2008). retrieved from https:// boe.es/buscar/doc.php?id=boe-a-2008-15595 real decreto (rd) 1565/2010, de 19 de noviembre (19 nov. 2010). retrieved from https:// www.boe.es/buscar/doc.php?id=boe-a-2010-17976 real decreto (rd) 1614/2010, de 7 de diciembre (7 dec. 2010). retrieved from https:// boe.es/diario_boe/txt.php?id=boe-a-2010-18915 real decreto (rd) 413/2014, de 6 de junio (6 june 2014). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-2014-6123 real decreto (rd) 900/2015, de 9 de octubre (10 oct. 2015). retrieved from https:// www.boe.es/buscar/act.php?id=boe-a-2015-10927 real decreto (rd) 947/2015, de 16 de octubre (16 oct. 2015). retrieved from https:// www.boe.es/boe/dias/2015/10/17/pdfs/boe-a-2015-11200.pdf real decreto (rd) 359/2017, de 31 de marzo (1 april 2017). retrieved from https:// boe.es/buscar/doc.php?id=boe-a-2017-3639 real decreto (rd) 650/2017, de 16 de junio (17 june 2017). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-2017-6940 real decreto (rd) 244/2019 de 5 de abril (6 april 2019). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-2019-5089. https://elperiodicodelaenergia.com/el-gobierno-agilizara-los-tramites-de-impacto-ambiental-de-las-nuevas-plantas-de-renovables-para-llegar-al-20-en-2020 https://elperiodicodelaenergia.com/el-gobierno-agilizara-los-tramites-de-impacto-ambiental-de-las-nuevas-plantas-de-renovables-para-llegar-al-20-en-2020 https://elperiodicodelaenergia.com/la-ley-que-acabara-con-el-impuesto-al-sol-tendra-que-esperar-a-septiembre https://elperiodicodelaenergia.com/la-ley-que-acabara-con-el-impuesto-al-sol-tendra-que-esperar-a-septiembre https://elperiodicodelaenergia.com/locura-renovable-en-espana-quieren-instalarse-mas-de-87-gw-la-mitad-ya-con-el-permiso-de-acceso-concedido-por-ree https://elperiodicodelaenergia.com/locura-renovable-en-espana-quieren-instalarse-mas-de-87-gw-la-mitad-ya-con-el-permiso-de-acceso-concedido-por-ree https://elperiodicodelaenergia.com/los-proyectos-renovables-ganadores-de-las-subastas-prefieren-perder-los-avales-a-cambio-de-firmar-un-ppa https://elperiodicodelaenergia.com/los-proyectos-renovables-ganadores-de-las-subastas-prefieren-perder-los-avales-a-cambio-de-firmar-un-ppa https://www.pv-tech.org/news/economics-not-tenders-driving-spains-solar-resurgence https://www.pv-tech.org/news/economics-not-tenders-driving-spains-solar-resurgence https://cincodias.elpais.com/cincodias/2018/11/09/companias/1541758756_915913.html https://www.mincotur.gob.es/energia/desarrollo/energiarenovable/documents/20100630_paner_espanaversion_final.pdf https://www.mincotur.gob.es/energia/desarrollo/energiarenovable/documents/20100630_paner_espanaversion_final.pdf http://www.idae.es/uploads/documentos/documentos_per_2005-2010_8_de_gosto-2005_completo.(modificacionpag_63)_copia_2_301254a0.pdf http://www.idae.es/uploads/documentos/documentos_per_2005-2010_8_de_gosto-2005_completo.(modificacionpag_63)_copia_2_301254a0.pdf http://www.idae.es/uploads/documentos/documentos_per_2005-2010_8_de_gosto-2005_completo.(modificacionpag_63)_copia_2_301254a0.pdf http://www.idae.es/file/9712/download?token=6moebdcb http://www.idae.es/file/9712/download?token=6moebdcb http://www.idae.es/uploads/documentos/documentos_4044_pfer2000-10_1999_1cd4b316.pdf http://www.idae.es/uploads/documentos/documentos_4044_pfer2000-10_1999_1cd4b316.pdf https://www.boe.es/diario_boe/txt.php?id=boe-a-1994-28980 https://www.boe.es/diario_boe/txt.php?id=boe-a-1994-28980 https://www.boe.es/diario_boe/txt.php?id=boe-a-1998-30041 https://www.boe.es/diario_boe/txt.php?id=boe-a-1998-30041 https://www.boe.es/buscar/doc.php?id=boe-a-2004-5562 https://www.boe.es/buscar/doc.php?id=boe-a-2004-5562 https://www.boe.es/buscar/act.php?id=boe-a-2007-10556 https://www.boe.es/buscar/act.php?id=boe-a-2007-10556 https://boe.es/buscar/doc.php?id=boe-a-2008-15595 https://boe.es/buscar/doc.php?id=boe-a-2008-15595 https://www.boe.es/buscar/doc.php?id=boe-a-2010-17976 https://www.boe.es/buscar/doc.php?id=boe-a-2010-17976 https://boe.es/diario_boe/txt.php?id=boe-a-2010-18915 https://boe.es/diario_boe/txt.php?id=boe-a-2010-18915 https://www.boe.es/diario_boe/txt.php?id=boe-a-2014-6123 https://www.boe.es/diario_boe/txt.php?id=boe-a-2014-6123 https://www.boe.es/buscar/act.php?id=boe-a-2015-10927 https://www.boe.es/buscar/act.php?id=boe-a-2015-10927 https://www.boe.es/boe/dias/2015/10/17/pdfs/boe-a-2015-11200.pdf https://www.boe.es/boe/dias/2015/10/17/pdfs/boe-a-2015-11200.pdf https://boe.es/buscar/doc.php?id=boe-a-2017-3639 https://boe.es/buscar/doc.php?id=boe-a-2017-3639 https://www.boe.es/diario_boe/txt.php?id=boe-a-2017-6940 https://www.boe.es/diario_boe/txt.php?id=boe-a-2017-6940 https://www.boe.es/diario_boe/txt.php?id=boe-a-2019-5089 https://www.boe.es/diario_boe/txt.php?id=boe-a-2019-5089 john s. duffield / european journal of government and economics 9(1), june 2020, 5-25 25 real decreto-ley (rdl) 6/2009, de 30 de abril (30 april 2009). retrieved from https:// www.boe.es/buscar/doc.php?id=boe-a-2009-7581 real decreto-ley (rdl) 14/2010, de 23 de diciembre (23 dec. 2010). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-2010-19757 real decreto-ley (rdl) 1/2012, de 27 de enero (28 jan. 2012). retrieved from https:// www.boe.es/buscar/doc.php?id=boe-a-2012-1310 real decreto-ley (rdl) 2/2013, de 1 de febrero (1 feb. 2013). retrieved from https:// www.boe.es/buscar/pdf/2013/boe-a-2013-1117-consolidado.pdf real decreto-ley (rdl) 9/2013, de 12 de julio (12 july 2013). retrieved from https:// www.boe.es/buscar/doc.php?id=boe-a-2013-7705 real decreto-ley (rdl) 15/2018, de 5 de octubre (6 oct. 2018). retrieved from https:// www.boe.es/diario_boe/txt.php?id=boe-a-2018-13593 red eléctrica de espańa (ree). (2019) evolución de la tramitación de los procedimientos de acceso a la red de la generación eólica y solar fotovoltaica gestionados por red eléctrica (oct.). retrieved from https://www.ree.es/sites/default/files/01_actividades/documentos /accesored/informacion_gestion_acceso_oct19.pdf roca, r. (2019). españa bate el récord de potencia instalada con más de 106 gw gracias a la nueva capacidad de renovables. el periódico de la energía (20 nov.). retrieved from https://elperiodicodelaenergia.com/espana-bate-el-record-de-potencia-instalada-con-masde-106-gw-gracias-a-la-nueva-capacidad-de-renovables/ roldán fernández, j.m., burgos payán, m., riquelme santos, j.m., and luis, a. (2016). renewables versus efficiency: a comparison for spain. energy procedia 106, 14-23. https://doi.org/10.1016/j.egypro.2016.12.101 sáenz de miera, g., del río, p., and vizcaino, v. (2008). analysing the impact of renewable electricity support schemes on power prices: the case of wind electricity in spain. energy policy 36: 3345-3359. https://doi.org/10.1016/j.enpol.2008.04.022 unión española fotovoltaica (unef). (2013). el ministerio de industria impide el autoconsumo de electricidad (19 july). retrieved from http://www.energiza.org/ index.php?option=com_k2&view=item&id=608:unef-%e2%80%98el-ministerio-de-industriaimpide-el-autoconsumo-de-electricidad unión española fotovoltaica (unef). (2017). el resultado de la subasta demuestra que la fotovoltaica ha sido discriminada (17 may). retrieved from https://unef.es/2017/05/elresultado-de-la-subasta-demuestra-que-la-fotovoltaica-ha-sido-discriminada unión española fotovoltaica (unef). (2019). el rd de autoconsumo sitúa al ciudadano en el centro del modelo energético al garantizar el libre acceso a la energía (5 april). retreived from https://unef.es/2019/04/el-rd-situa-al-ciudadano-en-el-centro-del-modelo-energetico-algarantizar-el-libre-acceso-a-la-energia vartiainen, e., masson, g., breyer, c., moser, d., and román medina, e. (2019). impact of weighted average cost of capital, capital expenditure, and other parameters on future utility‐ scale pv levelised cost of electricity. retreived from https://onlinelibrary.wiley.com /doi/epdf/10.1002/pip.3189. https://doi.org/10.1002/pip.3189 https://www.boe.es/buscar/doc.php?id=boe-a-2009-7581 https://www.boe.es/buscar/doc.php?id=boe-a-2009-7581 https://www.boe.es/diario_boe/txt.php?id=boe-a-2010-19757 https://www.boe.es/diario_boe/txt.php?id=boe-a-2010-19757 https://www.boe.es/buscar/doc.php?id=boe-a-2012-1310 https://www.boe.es/buscar/doc.php?id=boe-a-2012-1310 https://www.boe.es/buscar/pdf/2013/boe-a-2013-1117-consolidado.pdf https://www.boe.es/buscar/pdf/2013/boe-a-2013-1117-consolidado.pdf https://www.boe.es/buscar/doc.php?id=boe-a-2013-7705 https://www.boe.es/buscar/doc.php?id=boe-a-2013-7705 https://www.boe.es/diario_boe/txt.php?id=boe-a-2018-13593 https://www.boe.es/diario_boe/txt.php?id=boe-a-2018-13593 https://www.ree.es/sites/default/files/01_actividades/documentos/accesored/informacion_gestion_acceso_oct19.pdf https://www.ree.es/sites/default/files/01_actividades/documentos/accesored/informacion_gestion_acceso_oct19.pdf https://elperiodicodelaenergia.com/espana-bate-el-record-de-potencia-instalada-con-mas-de-106-gw-gracias-a-la-nueva-capacidad-de-renovables/ https://elperiodicodelaenergia.com/espana-bate-el-record-de-potencia-instalada-con-mas-de-106-gw-gracias-a-la-nueva-capacidad-de-renovables/ https://doi.org/10.1016/j.egypro.2016.12.101 https://doi.org/10.1016/j.enpol.2008.04.022 http://www.energiza.org/index.php?option=com_k2&view=item&id=608:unef-%e2%80%98el-ministerio-de-industria-impide-el-autoconsumo-de-electricidad http://www.energiza.org/index.php?option=com_k2&view=item&id=608:unef-%e2%80%98el-ministerio-de-industria-impide-el-autoconsumo-de-electricidad http://www.energiza.org/index.php?option=com_k2&view=item&id=608:unef-%e2%80%98el-ministerio-de-industria-impide-el-autoconsumo-de-electricidad https://unef.es/2017/05/el-resultado-de-la-subasta-demuestra-que-la-fotovoltaica-ha-sido-discriminada https://unef.es/2017/05/el-resultado-de-la-subasta-demuestra-que-la-fotovoltaica-ha-sido-discriminada https://unef.es/2019/04/el-rd-situa-al-ciudadano-en-el-centro-del-modelo-energetico-al-garantizar-el-libre-acceso-a-la-energia https://unef.es/2019/04/el-rd-situa-al-ciudadano-en-el-centro-del-modelo-energetico-al-garantizar-el-libre-acceso-a-la-energia https://onlinelibrary.wiley.com/doi/epdf/10.1002/pip.3189 https://onlinelibrary.wiley.com/doi/epdf/10.1002/pip.3189 https://doi.org/10.1002/pip.3189 number 9, issue 1, june 2020 the politics of renewable power in spain 1. introduction 2. why promote renewable power in spain? 3. initial efforts to promote renewable power 4. the tariff deficit and spain’s response 5. the revival of renewable power 6. future prospects for renewable power in spain 7. conclusions aknowledgements references microsoft word ejge_04_2015-010.docx european journal of government and economics volume 4, number 2 (december 2015) issn: 2254-7088 79 does debt predict growth? an empirical analysis of the relationship between total debt and economic output willem vanlaer, hasselt university, belgium wim marneffe, hasselt university, belgium lode vereeck, hasselt university, belgium johan van overtveldt, hasselt university, belgium abstract although the recent global financial crisis has stimulated a vast amount of research on the impact of public debt on economic growth and also increasingly on the role of private credit, the total levels of indebtedness of an economy have largely been ignored. this paper studies the impact of the total level of and increases in debt-togdp on economic growth for 26 developed countries in the short, medium and longer term. we analyse whether we can predict the future level of growth, simply by looking at the total level of debt, or increases in that debt level. we find that there is a negative correlation between high levels of debt and short term economic growth, but that this effect tapers in the medium and long term. similarly, we find that rapid debt accumulation is negatively related to economic growth over the short term, the impact is less pronounced over the medium term and is nonexistent over the long term. jel classification h63; o40. keywords public debt; government debt; private debt; growth; oecd countries. acknowledgements we thank the participants of 5th global conference of the forum for economists international in amsterdam & pet15 in luxembourg for their very useful comments and insights on this paper. vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 80 1. introduction scholars and policy makers agree in general that debt, both public and private, has played an important role in the build up to the recent global financial crisis (e.g. reinhart and rogoff, 2010, mian and sufi, 2014). an increasing number of researchers assert that the high levels of debt commonly found in developed economies, play an important role in the subsequent slow recovery (e.g., reinhart, reinhart and rogoff, 2012 and chatterjee, 2013). one important strand of research looks into the role of public debt in destabilizing an economy. when there are increasing doubts about the fiscal sustainability of a country, households and corporations anticipate future elevated taxes, which results in reduced consumption and investment1. subsequently, as the risk of a default increases, government debt holders will demand higher interest rates, which makes it even harder for a government to service its debt. this point of view has dominated public policy – and research – in the immediate aftermath of the crisis. an excellent illustration is the preeminence of austerity measures to deal with the outfall of the global financial crisis, both in europe and in the united states and advocated by international institutions such as the imf (e.g. imf, 2010). only in the last few years, research has looked into the role of private debt, or credit2, on economic growth (e.g. taylor and corrado, 2012). previously, the narrative of hyperrational private households and companies who make decisions in their own interest that also benefit society as a whole, had prevailed. moreover, as net world debt is zero, the consensus view was that losses to creditors are automatically cancelled out by gains to debtors. the fallout of the collapse in house prices in the united states in 2007 clearly proved this assumption to be erroneous. little research, however, has been pursued on the impact of the total level of debt3 on an economy. nevertheless, public debt and private credit are interlinked. one example is provided by the recent bailouts of the financial sector in spain. as shown in figure 1 below, simply focusing on the level of public debt-to-gdp would erroneously give the impression that spain did not have a problem of excessive debt in the run-up to the global financial crisis, as it strictly adhered to the maastricht criterion of public debt levels of below 60%. simultaneously however, a large construction bubble was building up, as private debt levels soared from around 125% of gdp at the start of the decade to more than 220% when the crisis struck in 2007-2008. hence, if in case of failure of the private sector the public sector steps in, it can be argued that this amounts to an implicit government guarantee, making private debt indistinguishable from public debt (bhandari, haque and turnovsky, 1990). the recent financial crisis has indeed shown, both in the united states and in europe, the willingness of the sovereign to stand behind its banking sector. investors considered too-big-to-fail banks as creditworthy as the government of its host country. these financial institutions could borrow at near identical rates as the sovereign, which made their debt very similar to sovereign debt (see e.g. schich and lindh, 2012). similarly, a narrow focus on the level of public debt would suggest that the us economy is in dire straits as its public debt-to-gdp levels skyrocketed from 53% in 1 this line of thinking is heavily influenced by the ricardian equivalence theorem which states that consumers take the budgetary constraints of the government into account when making consumption decisions (barro, 1996). consequently, higher budget deficits now result in higher taxation in the future. forward looking consumers anticipate this and hence increase their savings to pay for these higher taxes. 2 in the remainder of this paper, private debt and credit will be used interchangeably. 3 in this article, we define the total level of debt in an economy as the sum of gross general government debt and private credit to households and non-financial companies. we exclude interbank lending in our analysis. european journal of government and economics 4(2) 81 2000 to close to 103% in 2012. on the other hand, american households and companies managed to deleverage after the crisis – albeit by a relatively small amount – which is likely to be beneficial for the long run sustainability of its economy. figure 1 evolution of public and private debt: examples at a more fundamental level, schmidt (1943) asserted that public and private debt are essentially equivalent. an argument which was further elaborated by buchanan (1958). the latter challenged the dominant theory on the distinction between public and private debt which hinges on the claim that private debt repayment represents a reduction in private net worth of the individual, but that public debt repayment does not constitute a reduction in the cumulative wealth of the community. however, when the taxpayer is taxed for servicing the public debt, this reduces his net worth, which is quite similar to private debt – and interest – repayments. this reduction in net worth must be offset, not against the increase in net worth enjoyed by the bondholder, but against the productivity of the public investments which are financed by the debt. indeed, the increase in net worth of the bondholder will take place, regardless of the productivity of the public project. it follows that the taxpayer (i.e. the public borrower) is at no time in a different position from the private borrower. if the latter invests foolishly, his real income is reduced when interest payments are due. similarly, if the state uses borrowed funds in an ill-advised way, this reduces the aggregate wealth of the community (bhatt, 1959). backhaus and wagner (2006) make a similar claim, stating that “public debt is just a particular form of private debt, where the borrower is especially powerful”. whereas a private citizen, responding to an unexpected drop in income, might borrow to smooth out this decline in spending power, so might the sovereign increase its borrowing when confronted with falling revenues, for example due to a recession. both for the sovereign and the private citizen, debt does not alter its net worth; it simply affects the timing of expenditure. public and private debt are not only interlinked and should therefore be examined simultaneously, they are also often interchangeable and should in many cases thus be aggregated to fully appreciate the extent to which debt has an impact on the economy. tuition fees for higher education exemplify this reasoning. a college 10 0 15 0 20 0 25 0 p ri va te d eb t a s % o f g d p 40 50 60 70 80 90 p u bl ic d eb t a s % o f g d p 2000 2002 2004 2006 2008 2010 2012 year public debt spain private debt spain spain 13 0 14 0 15 0 16 0 17 0 p ri va te d eb t a s % o f g d p 50 60 70 80 90 10 0 p u bl ic d eb t a s % o f g d p 2000 2002 2004 2006 2008 2010 2012 year public debt us private debt us us vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 82 education entails certain costs (e.g. paying salaries to teachers, building the necessary infrastructure). whether these costs are borne by the state, by providing free higher education, or by the students by paying full cost fees, does not change this fact4. so, it may follow that either the government has to take on debt in order to finance higher education, or that the individual has to take on a student loan to pay for school. a similar argument can be made for the provision of health care, which can be arranged by the state, e.g. in the form of single-payer healthcare, or via private insurers. likewise, governments can choose to outsource the exploitation of public infrastructure, such as toll roads, to private partners in exchange for co-financing of the initial investment. in essence, these are basically transfers on the same side of a consolidated balance sheet of the society. arguments on who – the public or the private sector should pay may depend on cost efficiency5 or ideology6. however, they do not alter the overall need for financing these projects and services. public debt and private credit are therefore communicating vessels and should not be analysed separately. focusing on only one is likely to yield a distorted view of the level indebtedness in and its impact on an economy. therefore, this paper studies the link between the total level of debt-to-gdp, and changes in that level, and economic growth. we analyse whether we can predict the future level of growth, simply by looking at the total level of debt, or increases in that debt level. 2. review of literature the debate on the relationship between public debt and economic growth was revived by reinhart and rogoff (2010), which was part of a much larger empirical analysis they performed in their book this time is different: eight centuries of financial folly (2009). their methods are temptingly straightforward. reinhart and rogoff group country-years in four categories by public debt-to-gdp ratios: 0-30 percent, 30-60 percent, 60-90 percent, and more than 90 percent. next, they compare real gdp growth rates across these different groups. they find that this relationship is rather weak for public debt-to-gdp ratios below a threshold of 90%. for debt levels above 90%, however, median growth rates go down by around one percent and average growth falls even more. this has subsequently been referred to as the nonlinear relationship between public debt and economic growth (minea and parent, 2012). afterwards, in a critical attempt to replicate the results of reinhart and rogoff’s 2009 paper, herndon, ash and pollin (2014) uncovered “data omissions, questionable methods of weighting and elementary coding errors”. they assert that, when these errors are corrected, average growth at public debt levels above the 90% threshold does not vary dramatically from average growth at lower debt levels. in a study by woo and kumar (2015), the impact of the initial debt-to-gdp level on subsequent gdp growth was explored. they authors find that a 10 percentage point increase in the initial public debt level reduces gdp per capita by around 0.2 percentage points a year. 4 of course, one can argue that a private institution is better managed than a college that is run by the state and that thus the total cost for society is lower in the former case, but that discussion is beyond the scope of this paper. 5 in some instances, it may be more efficient if the state pays for a particular service (e.g. health care), while in – many other case the private market produces more efficient outcomes (e.g. telecoms services). 6 some people simple prefer a smaller or a larger state than others. european journal of government and economics 4(2) 83 baum, checherita-westphal, and rother (2013) analysed the impact of public debt on gdp in the emu by using a dynamic threshold panel methodology. they identify an inverse u-shaped relationship between public debt and growth as their findings suggest that the short-run impact of debt is positive, decreases to zero at a public debt/gdp ratio of around 67% and is significantly negative at debt ratios above 90%. critics of the original reinhart and rogoff (2009) paper reply that, while there may very well be a negative relationship between public debt and economic growth, the effect works in the opposite direction to what the supporters of austerity claim. it is low growth that causes the state revenues to fall and public expenditures to rise, thus resulting in a higher level of public debt. lof and malinen (2014) tried to tackle the issue of reverse causality by using panel vector autoregressions that represent the dynamic relationship between gdp and public debt, decomposing cause and effect. they indeed conclude that the negative correlation between both variables is mainly driven by the impact of economic growth on sovereign debt, not the other way around. a recent paper by three economists at the imf research department (pescatori, sandri and simon, 2014) refuted the existence of debt thresholds after which economic growth significantly deteriorates. they used a novel empirical approach to determine the relationship between debt thresholds and growth prospects. more specifically, they take a sample of all episodes where public debt increases above a particular threshold and calculate real gdp growth per capita over the following h years, varying h from 1, 5, 10 to 15. although their method is quite similar to the one applied by reinhart and rogoff (2012), it differs in two crucial aspects. firstly, the range of debt thresholds being analysed is much broader than the 90 percent threshold on which the reinhart and rogoff paper focuses. secondly, economic growth over a particular time span is analysed, regardless of the debt outcome. in contrast, reinhart and rogoff only consider the period when debt persists above a certain threshold. taylor (2012) illustrated the importance of private credit in developed economies. he argues that past growth in private debt contains predictive information about the likelihood of a crisis occurring in the future. moreover, he finds that the recession after a credit boom is more severe than a “normal” recession. jorda, schularick and taylor (2013) built on this work by analyzing the co-evolution of sovereign debt and credit in developed countries. they find that the risks to economic and financial stability mainly come from booms in private debt rather than increases in public debt. nevertheless, when a country enters a crisis period with an already elevated level of public debt, this intensifies the crisis, presumably by the limited ability to introduce fiscal stimulus to uphold aggregate demand. a paper by randveer, uusküla and kulu (2011) researched the link between economic recovery after a crisis and growth of credit before a crisis. they find, quite counterintuitively, that rapid credit expansion before a crisis is related with higher economic growth after the crisis. this contrasts with the widespread belief that, after a crisis and especially after a balance sheet recession like the one we have recently experienced, household debt overhang and the subsequent process of deleveraging acts as a serious drag on economic recovery. that is also precisely what gärtner (2013) found in her study of the great depression. debt overhang of households was indeed an important factor in holding back the economic recovery after the very severe drop in output that followed the stock market crash of october, 1929. in their widely praised book house of debt (2014), atif mian and amir sufi provided compelling evidence for the case that the buildup of household debt was the main culprit for the recent recession. additionally, when they analyse previous economic vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 84 downturns, the increase in consumer debt often plays a very important role. they conclude that the bigger the increase in debt, the harder the fall in spending. cecchetti, mohanty and zampolli (2011) analysed government debt, corporate debt and household debt and find that each of these three types of debt becomes a drag on growth when it reaches a level above 85-90% of gdp. they conclude that debt improves welfare but can be damaging to the economy when levels are very high. as mentioned, the initial response to the global financial crisis was to focus on excessive government debt. afterwards, the focus shifted to private credit. our study aims to further the literature by looking at the overall debt level in an economy. this paper provides empirical evidence on the relationship between total debt and economic growth. we explore whether there is a specific debt threshold after which growth plummets. section 3 describes the data and methodology used in our analysis. in section 4, we provide empirical evidence on the link between total debt levels and growth for our panel of 26 oecd countries from 1961-2012 in. section 5, analyses changes in total debt-to-gdp and real gdp growth. in the subsequent section, we discuss our main findings. finally, section 7 presents the conclusions. 3. data and methodology we have gathered data on public debt levels from a comprehensive database on gross government debt, which is compiled by the imf fiscal affairs department7. data on private debt comes from a recent bis database8 on credit to the private non-financial sector. we acquire gdp data from the ameco database of the european commission. our dataset comprises annual data on 26 developed economies, over the period 1961-2012. table a in the appendix provides the summary statistics for public, private and total debt-to-gdp levels, yearly increases in total debt-to-gdp levels and real gdp per capita growth rates. in our sample, the average total debt-to-gdp level is 168%, the average yearly increase in debt is 3.65% and average real growth equals 2.48%. however, these averages disguise large discrepancies between countries. whereas countries like turkey and mexico have average total debt-to-gdp levels of around 70%, countries such as japan and luxembourg9 have an average debt level which is several times larger (363% and 257% respectively). we need to discuss several methodological issues when examining the relationship between debt and growth. firstly, it seems highly implausible that there is one common debt threshold after which growth plummets. if there exists a “dangerous” debt threshold, this will most likely vary across countries and across time, and will be dependent on numerous other factors (e.g. potential growth rate10, willingness to save of the private sector). another issue pertains to the causality between debt and growth. rather than being the cause, high debt may be the result of anaemic growth. if it is sluggish growth that causes high debt, it is less probable that a specific threshold is 7 see horton et al. (2010) for a complete description of the database, which can be found online on www.imf.org/external/datamapper/index.php?db=debt. 8 see dembiermont, drehmann and muksakunratana (2013) for a detailed description of the database: http://www.bis.org/statistics/credtopriv.htm. 9 luxembourg is a particular case due to its heavy reliance on financial services, resulting in a very high level of private debt. 10 as summers (2014) has argued repeatedly in his discussion of secular stagnation, in a world with economic output chronically below potential, debt-financed public projects might be needed to generate growth levels which are consistent with full employment and stable inflation. european journal of government and economics 4(2) 85 discerned. hence, if such a threshold is found, the likelihood that it is driven by a causal effect of debt on growth is relatively high. moreover, elevated debt levels might conceal an omitted variable, a common factor both increasing debt and reducing growth (e.g. a financial crisis or a war). this applies particularly when analyzing the short-term relationship between debt and growth, since a recession almost mechanically results in a higher debt ratio due to the denominator shrinking. one way to address the issue of causality is by adopting instrumental techniques. for example, panizza and presbitero (2014) use an instrument variable that takes into account valuation effects resulting from the interplay between exchange rate volatility and foreign currency debt. they find no evidence that debt has a causal effect on growth. next to the short-term relationship between debt and growth, which is the focus in most of the literature, we consider the long-term link. more concretely, we analyse the long-term relationship between today’s level of total debt to gdp, bt, and subsequent gdp growth in the next h-years, git (h) = yt+h/yt. this longer-term horizon allows for tempering the effects of reverse causality that a brief recession or boom might have on short-term economic growth. we apply the framework provided by pescatori, sandri and simon (2014). our analysis starts by taking a sample of all country episodes11 where total debt rose above a threshold τ. next, we look at real gdp per capita growth over the next h years, where h ϵ [1,5,10,15]. countries can have multiple episodes, but not overlapping ones. we consider the start of a rising debt episode to be the first year in which the total debt level exceeds the threshold τ, conditional on the level being below the threshold in the previous year. to put it more formally, for a country i and a threshold τ, the start of an episode has to meet the following conditions: bit ≥ τ, bit-1 < τ and ∄ j ϵ [1,…,h] s.t. bit-j ≥ τ, bit-j-1 < τ. there are several features of this model which are important to note. firstly, by analyzing economic performance over a given period regardless of the debt outcome, we avoid the truncation problem which arises when we only look at the period when debt remains above a certain level, in effect solely selecting ‘failures’. in our analysis, we admit countries that succeed in reducing total debt levels after surpassing a particular debt threshold along with the ‘failures’, i.e. countries whose debt level remains elevated. secondly, we allow each country to only have a limited number of episodes, due to the fact that we rule out overlapping episodes and by requiring that an episode starts when total debt exceeds the threshold from below. when calculating averages, the episodes are pooled together and weighted equally12. finally, in contrast to the growth regressions adopted in most other papers, our methodology does not impose a linear relationship between debt and economic growth. 4. total debt levels and economic growth 4.1. short-term (1 year) our analysis starts with the short-term relationship between the total debt level and economic growth. in figure 2, we present the average real gdp per capita growth rate in the year after the total debt level exceeds a certain threshold, i.e. h=1. more 11 a country episode is a combination of h consecutive country years (cf. supra). 12 the methodology adopted by reinhart and rogoff (2010) resulted in several countries having a lot more observations than other countries. alternative weighting methods can subsequently bring about significantly different conclusions. vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 86 precisely, we show a scatterplot of all the observations, plotting economic growth against total debt-to-gdp. we also include a locally fitted regression function13. we can observe that there are countries with higher debt which experience sound growth rates. reversely, there are countries with lower debt levels and that display limited or even negative economic growth14. in general, however, a higher level of debt leads to lower economic growth, but this relationship becomes less pronounced at very high total debt levels. in addition, a spearman's correlation was calculated to determine the relationship between debt and growth. it showed there was moderate, negative monotonic correlation between the two variables (ρ = 0.42). figure 2 total debt and short-term economic growth 4.2. mediumand longer-term (5, 10 and 15 years) however, it would be ill-advised to draw conclusions about the link between total debt and economic growth simply by analysing the year after a particular debt threshold is exceeded. the reason for this is straightforward; instead of increased debt leading to lower growth, the relation could run the other way, i.e. lower growth resulting in higher debt levels. this argument has been extensively discussed by critics of the original reinhart and rogoff paper (2009) (see e.g. herndon, ash and pollin, 2013 and pescatori, sandri and simon, 2014). when growth significantly slows down, and the economy enters into a recession, government revenues decline, e.g. due to a fall in personal taxes and income taxes. similarly, governement expenditures increase, as more people rely for instance on unemployment insurance. absent severe cuts in spending on other policy areas – 13 the locally smoothed regression function is estimated with the general additive model with integrated smoothness estimation using the gamfit package in stata. 14 our methodology can produce multiple observations for one country-year. for example, if a country’s total debt level jumps from 100% to 115% in one year, multiple thresholds (e.g. 102%, 104%, 106%, etc.) will be exceeded. -1 0 -5 0 5 1 0 r ea l g d p p e r ca pi ta g ro w th r at e (% ) 0 50 100 150 200 250 300 350 400 450 500 total debt to gdp thresholds (%) 1 year european journal of government and economics 4(2) 87 which would deepen the recession – this results in a larger budget deficit and consequently a higher level of government debt. correspondingly, job losses caused by the recession render households less able to service their debt, which in the short run makes it very hard to deleverage. some households will even have to take on additional loans in order to make ends meet. in summary, when economic growth slows down, or even becomes negative, the debt-to-gdp ratio almost mechanically increases; debt (i.e. the numerator) soars due to lower revenues and higher expenditures as gdp (i.e. the denominator) shrinks. we try to eliminate the bias which potentially arises from reverse causality as well as mitigate the effects of outliers by extending the horizon of our analysis. if high levels of total debt in an economy significantly hold back growth over the medium to long term, we would foresee that growth is not only supressed in the first year that debt surpasses the threshold, but also in the ensuing years. figure 3 shows economic growth over 5, 10 and 15 years respectively after a certain debt threshold has been reached15. to complete the picture, we have added the results from our previous analysis, i.e. short-term growth. when looking at economic growth over a 5-year horizon, the result of our previous analysis is confirmed; more debt results in lower growth. although growth performance improves noticably in the longer run, higher debt levels are still related to a lower growth rate. this is confirmed by spearman’s rank correlation coefficient, which declines from -0.45 for 5-year growth to -0.20 for growth over a 15-year horizon, which is a rather weak correlation. additionally, we compared the regression coefficients among these four groups to test the null hypothesis: h0: b1 = b2 = b3 = b4 where b1 is the regression for 1-year growth, b2 for 5-year growth, b3 for 10-year growth, and b4 for 15-year growth. our analysis reveals that the null hypothesis can be rejected (f=105.07, p = 0.0000). this means that the regression coefficients between debt and growth do significantly differ across the 4 groups (1 year, 5 year, 10 year, 15 year). we also tested the regression coefficients between two groups pairwise (e.g. comparing 10-year growth with 15-year growth). all regression coefficients significantly differ from each other (p = 0.0000). figure 3 total debt and mediumto long-term economic growth as a robustness test, we have rerun our analysis on a homogenous set of data. whereas we varied our exclusion window in figure 3 in correspondence with the period over which we calculate the average growth rate (e.g. the exclusion window for h=1is 2012 and for h=15 is 1998), in figure 4 the exclusion window is the same for each curve, i.e. 1998. in other words, each of the four graphs looks at the exact same debt episodes; only the period over which economic growth is averaged, 15 the exclusion window for the different horizons depends on the level of h, i.e. for h=1/5/10/15 we look at episodes until 2012/2007/2002/1997. -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 400 500 total debt to gdp thresholds (%) 1 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 400 total debt to gdp thresholds (%) 5 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 400 total debt to gdp thresholds (%) 10 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 total debt to gdp thresholds (%) 15 year vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 88 varies. the results are strikingly similar, with spearman’s rank correlation varying from -0.40 to -0.20. figure 4 total debt and mediumto long-term economic growth (homogenous dataset) of course, our study could be influenced by outliers. for example, several episodes of very low growth might skew our results. therefore, in figure 5 we repeat our analysis, but eliminate the observations with 2.5% lowest and 2.5% highest growth levels16. our previous results remain unaltered. in the short term, there is a relatively strong negative correlation between debt and growth, but this correlation attenuates when growth over a longer period is taken into consideration. figure 5 total debt and mediumto long-term economic growth (outliers removed) comparing the impact of public debt and private debt levels is beyond the scope of this paper. nevertheless, it is interesting to note that the general relationship between total debt and growth – relatively strong negative correlation in the short term, much less so on the long term – also holds for public and private debt, as can be seen in figure 6. however, the strength of the negative relationship with growth is more pronounced for private debt than for public debt. 16 we get similar results if we eliminate only the 1.25% highest and lowest values or if we eliminate the 5% highest and lowest values. -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 total debt to gdp thresholds (%) 1 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 total debt to gdp thresholds (%) 5 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 total debt to gdp thresholds (%) 10 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 total debt to gdp thresholds (%) 15 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) 0 100 200 300 400 500 total debt to gdp thresholds (%) 1 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) 0 100 200 300 400 total debt to gdp thresholds (%) 5 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) 0 100 200 300 400 total debt to gdp thresholds (%) 10 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) 0 100 200 300 total debt to gdp thresholds (%) 15 year european journal of government and economics 4(2) 89 figure 6 public debt & private debt levels and mediumto long-term economic growth 4.3. debt trajectory hitherto, our analysis has focused solely on episodes where the total debt level has exceeded a threshold. now, we look at the impact of decreasing total debt levels on economic growth. therefore, we study all episodes where the total debt level declines below a given threshold: bit ≤ τ, bit-1 ˃ τ and ∄ j ϵ [1,…,h] s.t. bit-j ≤ τ, bit-j-1 ˃ τ. in figure 7, we compare short-term economic growth (h=1) in these episodes with the ones identified in figure 2. we can observe that the trajectory of growth performance is quite similar between countries that exceed or fall below a debt threshold. yet countries on a declining debt path have slightly higher growth rates at high debt levels compared to those on an increasing path. in other words, high levels of debt are negatively associated with growth, but if these high debt levels are falling, growth performance improves slightly. in addition, when debt is falling, there appears to be no impact on growth until debt levels reach 200% of gdp, after which real gdp growth steadily deteriorates. -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 50 100 150 200 250 public debt to gdp thresholds (%) 1 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 50 100 150 200 public debt to gdp thresholds (%) 5 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 50 100 150 public debt to gdp thresholds (%) 10 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 50 100 150 public debt to gdp thresholds (%) 15 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 400 private debt to gdp thresholds (%) 1 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 300 400 private debt to gdp thresholds (%) 5 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 private debt to gdp thresholds (%) 10 year -1 0 -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) 0 100 200 private debt to gdp thresholds (%) 15 year vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 90 figure 7 debt trajectory and short-term economic growth figure 8 presents the results for h=15.17 we now find that the evolution of economic growth is very similar for countries with increasing and decreasing debt levels. higher levels of debt are still correlated with somewhat lower growth, but there is no impact on economic performance stemming from the increase or decrease of debt levels. 17 we get similar results for h=5 and h=10. -1 0 -5 0 5 10 r ea l g d p p er c ap ita g ro w th r at e (% ) 0 100 200 300 400 500 total debt to gdp thresholds (%) 1 year debt rising -1 0 -5 0 5 10 r ea l g d p p er c ap ita g ro w th r at e (% ) 0 100 200 300 400 500 total debt to gdp thresholds (%) 1 year debt falling european journal of government and economics 4(2) 91 figure 8 debt trajectory and long-term economic growth 5. changes in total debt levels and economic growth 5.1 general in our previous analysis, we have found that in the short term, high total debt levels in an economy are correlated to lower economic growth, but the effect becomes less pronounced in the the medium and long run. moreover, we have concluded that the debt trajectory has little to no impact on growth. in this section, we evaluate whether changes in debt levels have a significant impact on gdp per capita growth. heretofore, we have looked at the stock of debt; we will now scrutinize debt flows. 5.2 short-term (1 year) in figure 9, we plot the average growth rate of countries the year after they reached a particular change in total debt level θ, that is18: bit – bit-1 = θ. the graph does show that a rapid accumulation of debt is correlated with lower economic growth, but the relationship is rather weak. this is also confirmed by spearman’s rank correlation coefficient, which is only -0.08. 18 the reason why we use “=” in this part of our analysis instead of “≥” is twofold. firstly, we have positive as well as negative values for θ, which renders the notion of “exceeding a certain threshold” less straightforward to interpret. secondly, if we would use “≥”, then exceeding a threshold of e.g. 1% would imply that we use all observations not just where there is a debt increase of 1%, but also where debt increases by 2%, 3%, 4%, etc. this way, certain thresholds would have hundreds of observations, which would make our analysis rather meaningless. -1 0 -5 0 5 10 r ea l g d p p er c ap ita g ro w th r at e (% ) 0 100 200 300 total debt to gdp thresholds (%) 15 year debt rising -1 0 -5 0 5 10 r ea l g d p p er c ap ita g ro w th r at e (% ) 0 100 200 300 total debt to gdp thresholds (%) 15 year debt falling vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 92 figure 9 change in total debt and short-term economic growth we now extend our analysis in two ways. firstly, we look at changes in total debtto-gdp levels over multiple years d, with d ϵ [1,5,10,15]. secondly, we study the growth performance of countries over the medium to long term h, with h ϵ [1,5,10,15]: bit – bit-d = θ and ∄ j ϵ [1,…,h] s.t. bit-j – bit-d-j = θ. figure 10 shows short-term economic growth (h=1) and debt accumulated over 1, 5, 10 and 15 years (i.e. d=1,5,10,15). figure 10 growth performance over 1 year with debt accumulation over 1-15 year (h=1 and d=1,5,10,15) firstly, the graphs show that a rapid accumulation of debt is associated with lower growth. all four curves have a negative slope. it can be observed that this correlation is not particularly strong, as shown by spearman’s rank correlation coefficient, which never exceeds -0.17. secondly, the period of time over which debt has been built up matters. logically, an increase in the total debt-to-gdp level of 20% over 1 year is more negatively correlated to growth than a 15 year building up period. thirdly, whereas deleveraging (i.e. negative values of changes in total debt-to-gdp levels) does not appear to impact growth, economic performance starts to deteriorate as debt accumulates (i.e. positive values of changes in total debt-to-gdp levels). this holds for all four curves. the threshold for which growth -1 0 -5 0 5 10 r e al g d p p e r ca pi ta g ro w th r at e (% ) -20 -10 0 10 20 30 change in total debt to gdp (%) accumulated over 1 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -20 -10 0 10 20 30 change in total debt to gdp (%) accumulated over 1 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -25 0 25 50 75 change in total debt to gdp (%) accumulated over 5 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -50 -25 0 25 50 75 100 change in total debt to gdp (%) accumulated over 10 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -50 0 50 100 150 change in total debt to gdp (%) accumulated over 15 years european journal of government and economics 4(2) 93 starts to slow down, increases as the period over which debt is accumulated is extended. for example, economic performance remains quite constant for d=5 until a 25% increase in debt. for d=15, an increase in debt of 50% must be reached before we see a negative impact on growth. finally, all regression coefficients were compared and found to be significantly different across the four groups (p = 0.0000). again, our analysis does not change if we exclude the 2.5 lowest and 2.5% highest values of growth, as shown in appendix c. when comparing the difference between public and private debt, it is noteworthy that private debt again has the strongest negative correlation with economic growth, as can be seen in figure 11. figure 11 growth performance over 1 year with debt accumulation over 1-15 year (h=1 and d=1,5,10,15) – public & private debt 5.3. medium-term (5 years) from the analysis so far, we cannot conclude that the causal direction runs from the accumulation of debt to sluggish growth. as argued, a period of low growth may result into a rapid build-up of debt. in addition, an increase in debt in the order of magnitude of over 10% in one year is very likely to be accompanied by a severe crisis, which in itself slows down growth. therefore, we will analyse the link between debt accumulation and mediumto long-term growth. in figure 12, we provide the results for economic growth over 5 years (h=5) after reaching a threshold change in total debt levels, that has been accumulated over 1, 5, 10 and 15 years (i.e. d=1,5,10,15) respectively. -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -10 -5 0 5 10 15 change in public debt to gdp (%) accumulated over 1 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -30 -15 0 15 30 45 change in public debt to gdp (%) accumulated over 5 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -50 -25 0 25 50 75 change in public debt to gdp (%) accumulated over 10 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -50 -15 20 55 90 change in public debt to gdp (%) accumulated over 15 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -10 -5 0 5 10 15 change in private debt to gdp (%) accumulated over 1 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -25 0 25 50 75 change in private debt to gdp (%) accumulated over 5 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -25 0 25 50 75 100 change in private debt to gdp (%) accumulated over 10 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -40 0 40 80 120 change in private debt to gdp (%) accumulated over 15 years vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 94 figure 12 growth performance over 5 year with debt accumulation over 1-15 year (h=5 and d=1,5,10,15) again, we find that the period over which debt is accumulated is relevant. the shorter this period, the more negative the economic growth. however, this relationship is much less strong in the medium term than in the short term. when debt has been built up over 10 or 15 years, the relationship becomes practically non-existent19.this is supported by spearman’s rank correlation, which is a mere 0.06 for d=10 and 0.03 for d=15 and fails to be statistically significant (p=0.09 and 0.49 respectively). hence, there appears to be no negative impact of a large increase in debt compared to a smaller increase in debt, when debt is built up over an extended period. these outcomes do not change when we exclude outliers, as shown in appendix d. there is a clear difference between public and private debt. accumulation of public debt is not negatively correlated with medium-term growth; spearman’s rank correlation is positive and ranging between 0.10 and 0.20. in contrast, a negative relationship between private debt accumulation and medium-term growth is observed and verified by spearman’s rank correlation, which ranges between -0.16 and -0.22. figure 13 growth performance over 5 years with debt accumulation over 1-15 year (h=1 and d=1,5,10,15) – public & private debt 19 the null hypothesis that both coefficients are the same also could not be rejected (p=0.09). -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -15 -5 5 15 25 change in total debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in total debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 75 change in total debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 120 change in total debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -10 -5 0 5 10 15 change in public debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in public debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 75 change in public debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 75 change in public debt to gdp (%) accumulated over 15 years european journal of government and economics 4(2) 95 5.4. long-term (15 years) finally, figure 14 presents the results of our analysis on growth performance over a 15 year horizon (h=15) 20 and debt accumulating over 1, 5, 10 and 15 years (d=1,5,10,15). we find that countries experiencing a significant increase in their total debt level do not record worse economic performances over the longer term than countries with a constant or declining total debt level. if anything, there is a slightly positive relationship between debt accumulation and longer term growth; spearman’s rank correlation is positive for all four curves and ranges between 0.01 and 0.13. appendix h demonstrates that we get similar results once extreme values are removed. moreover, we compared the regression coefficients among the four groups. they do not siffer significantly (f=1.36, p=0.2540), which means that the period over which debt is accumulated is not relevant. the regression coefficients between two groups were also compared pairwise (e.g. comparing debt accumulation over 1 year with debt accumulation over 10 years). the null hypothesis could not be rejected for any comparison. figure 14 growth performance over 15 year with debt accumulation over 1-15 year (h=15 and d=1,15) figure 15 shows that there is hardly any difference in the correlation between debt and long-term growth between public debt and private debt, with the only exception being public debt accumulated over 1 year. 20 the analysis for economic growth over a 10 year period yields similar results, as can be seen from appendices e-g. -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -10 -5 0 5 10 15 change in private debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in private debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 75 change in private debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in private debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -15 -5 5 15 25 change in total debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 change in total debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in total debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in total debt to gdp (%) accumulated over 15 years vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 96 figure 15 growth performance over 15 years with debt accumulation over 1-15 year (h=15 and d=1,5,10,15) – public & private debt 6. discussion of results although the recent global financial crisis has produced a vast amount of research on the impact of public debt on economic growth and increasingly also on the role of private credit, the total levels of indebtedness of an economy have largely been ignored. our paper for the first time attempts to fill that void by analysing the link between total debt-to-gdp levels and economic growth for a panel of 26 oecd countries between 1961 and 2012. we investigate whether we can predict the future level of growth, simply by looking at the total level of debt, or increases in that debt level. we found little evidence to support the hypothesis that there is a critical threshold of the total debt-to-gdp ratio above which economic growth plummets. rather, growth performance worsens gradually as debt levels rise. in addition, this relationship is more pronounced for short-term growth than for mediumto longterm growth. it follows that the causal direction most likely runs from low growth to higher debt levels, not the other way around. another explanation is that, on average, the return on the investments financed by this debt compensates their cost. for example, companies take on more debt to improve their machinery, which raises their debt level in the short run. however, in the longer run, this results in higher productivity and consequently higher profits, with the increase in profits offsetting the interest payments on the debt. a similar argument can be made for public investments that increase public debt-to-gdp in the short run, but improves the long-term economic potential of a country. one can also argue that it is not so much the level of debt, or increases in that level, which matters. rather, the cost of servicing debt, i.e. bond yields on both corporate and government debt21, plays a far more important role. at lower interest 21 we assume that low levels of corporate and government bond yields are also translated into lower levels of interest on consumer loans. -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -10 -5 0 5 10 15 change in public debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in public debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in public debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in public debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -10 -5 0 5 10 15 change in private debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in private debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in private debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in private debt to gdp (%) accumulated over 15 years european journal of government and economics 4(2) 97 rates, it will be easier to service debt. simultaneously, certain investment projects become profitable due to the lower cost of financing it, which results in higher growth. of course, there is a feedback loop between the level of debt and the level of interest paid on it, but, ceteris paribus, a lower yields on bonds improves investments prospects and hence economic growth. we have also looked at the impact of the pace of debt accumulation on subsequent economic growth. we find that rapid debt accumulation is negatively related to economic growth over the short term, but that this effect is less pronounced over the medium term and is non-existent over the long term. this could be explained by the fact that countries experiencing a fast rising debt level are typically hit by an exogenous shock, which causes the public and private sector to take on more debt in the short term, without significantly altering the long-term fundamentals of the country. in our analysis, we have applied a non-parametric method to demonstrate the correlation between debt and growth. the appeal of this method is that the results can be interpreted rather intuitively. however, further research is warranted to identify the complex structural relationship between total debt levels and gdp growth. another area for future research is to decompose the aggregate level of debt in an economy in its composing parts and debt holders to find out what type of debt has significant impact on economic growth. for example, public debt can be disaggregated into internal and external debt, private debt into mortgage debt, credit card debt and others. this way, the impact of borrowing in a foreign currency, both for the public and private sector, can be studied. 7. conclusions in summary, this paper has two main contributions. the first is that the current emphasis on public debt is too narrow and that private debt should be included in analysing the level of indebtedness of an economy. as public and private debt in many cases are interchangeable, it is the effect of aggregated debt on growth that should be explored. second, it shows that there is no critical total debt-to-gdp ratio after which medium-term economic performance significantly worsens. hence, the excessive focus on a target value such of the debt-to-gdp ratio is misguided. there are other, more important factors which determine the impact of debt and growth, such as bond yields. an abstract level of debt does not have predictive power for the outlook of growth. although we find some evidence that, in the short term, higher debt levels are associated with lower growth, the relationship weakens significantly when we expand the time horizon. this supports the hypothesis that it is lower growth which leads to more debt and not vice versa. similarly, we find that the pace of debt accumulation negatively correlates to short-term economic growth and that this relationship weakens for medium-term growth. for long-term growth, the speed at which debt is building up does not substantially alter longer-term growth prospects. of course, our results do not imply that a country or the private sector can pile on large amounts of debt without careful analysis of the projects financed by this debt accumulation. our research simply suggests that historically, on average, government debt and private credit are mainly used in such a way that is conducive to economic growth. vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 98 references backhaus, jürgen and richard e. wagner (2006) handbook of public finance. new york: springer science & business media. barro, robert j. (1996) 'reflections on ricardian equivalence', national bureau of economic research working papers no. w5502. baum, anja, christina checherita-westphal and philipp rother (2013) 'debt and growth: new evidence for the euro area', journal of international money and finance 32: 809-821. bhandari, jagdeep s., nadeem ul haque and stephen j. turnovsky (1990) 'growth, external debt, and sovereign risk in a small open economy', staff papersinternational monetary fund staff papers: 388-417. bhatt, makesh (1959) 'orthodoxy: old vs new', the economic weekly 11(22): 717718. buchanan, james m. (1958) public principles of public debt a defense and restatement. cecchetti, stephen g., madhusudan s. mohanty and fabrizio zampolli (2011) 'the real effects of debt', bank for international settlements, monetary and economic department working papers no. 352. chatterjee, satyaijit (2013) 'debt overhang: why recovery from a financial crisis can be slow', business review q2: 1-9. dembiermont, christian, mathias drehmann and siriporn muksakunratana (2013) 'how much does the private sector really borrow? a new database for total credit to the private non-financial sector', bis quarterly review march. gärtner, kaharina (2013) 'household debt and economic recovery evidence from the us great depression', european historical economics society working papers no. 36. herndon, thomas, michael ash and robert pollin (2014) 'does high public debt consistently stifle economic growth? a critique of reinhart and rogoff', cambridge journal of economics 38(2): 257-279. horton, mark. a., ali s.m. abbas, nazim belhocine and asmaa a. el-ganainy (2010) 'a historical public debt database', imf working paper no. 10/245. international monetary fund (2010) world economic outlook: rebalancing growth. washington (april). jordà, oscar, moritz hp schularick and alan m. & taylor (2013) 'sovereigns versus banks: credit, crises and consequences', national bureau of economic research working papers no. w19506. lof, matthijs and tuomas malinen (2014) 'does sovereign debt weaken economic growth? a panel var analysis', economics letters 122(3): 403-407. mian, atif and amir sufi (2014) house of debt how they (and you) caused the great recession, and how we can prevent it from happening again. chicago: university of chicago press. minea, alexandru and antoine parent (2012) 'is high public debt always harmful to economic growth? reinhart and rogoff and some complex nonlinearities', cerdi working papers, no. 2012-18. panizza, ugo and andrea f. presbitero (2014) 'public debt and economic growth: is there a causal effect?', journal of macroeconomics 41: 21-41. pescatori, andrea, damiano sandri and john simon (2014) 'debt and growth: is there a magic threshold?', imf working papers no. 14/34. european journal of government and economics 4(2) 99 randveer, martti, liina kulu and lenno uusküla (2011) 'the impact of private debt on economic growth', eesti pank working papers no. 10/2011. reinhart, carmen. m and kenneth rogoff (2009) this time is different: eight centuries of financial folly. princeton: princeton university press. reinhart, carmen. m and kenneth rogoff (2010) 'growth in a time of debt', american economic review 100(2): 573-578. reinhart, carmen. m., vincent r. reinhart and kenneth rogoff (2012) 'public debt overhangs: advanced-economy episodes since 1800', the journal of economic perspectives 26(3): 69-86. schich, sebastian and sofia lindh (2012) 'implicit guarantees for bank debt.' oecd journal: financial market trends 2012(1): 45-63. schmidt, emerson p. (1943) 'private versus public debt', american economic review 33(1): 119-121. summers, larry h. (2014) 'us economic prospects: secular stagnation, hysteresis, and the zero lower bound', business economics 49(2): 65-73. taylor, alan m. and caro corrado (2012) 'the great leveraging', national bureau for economic research working papers no. w18290. woo, jaejoon and manmohan s. kumar (2010) 'public debt and growth', economica 82(328): 705–739. vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 100 appendix descriptive statistics country public debt to gdp private debt to gdp total debt to gdp yearly increase in total debt real gdp per capita growth australia 23,81% 103,66% 128,36% 2,11% 2,05% austria 44,86% 96,39% 142,33% 3,76% 2,57% belgium 88,24% 132,83% 229,69% 4,71% 2,34% canada 69,22% 129,70% 198,92% 2,49% 2,01% czech republic 25,58% 69,27% 94,85% 1,40% 1,82% denmark 36,36% 156,07% 192,49% 4,23% 2,01% finland 25,79% 119,74% 149,83% 3,34% 2,68% france 39,86% 106,58% 151,33% 3,88% 2,23% germany 42,71% 109,07% 151,96% 2,12% 2,17% greece 65,48% 54,84% 120,78% 5,26% 2,70% hungary 80,91% 86,14% 165,86% 5,42% 1,95% ireland 58,50% 139,23% 201,05% 7,75% 3,42% italy 79,09% 80,27% 159,36% 3,23% 2,27% japan 83,33% 168,83% 256,93% 5,99% 3,46% luxembourg 8,25% 351,04% 363,05% 15,47% 2,74% mexico 38,43% 23,52% 70,57% 0,18% 1,92% netherlands 58,17% 122,26% 180,44% 3,78% 2,11% norway 36,62% 143,06% 181,38% 2,01% 2,61% poland 51,53% 46,08% 94,65% 1,85% 3,77% portugal 45,00% 137,92% 182,92% 5,87% 3,08% spain 36,18% 124,02% 164,03% 4,82% 2,74% sweden 48,24% 148,32% 197,14% 3,53% 2,13% switzerland 45,51% 156,64% 207,46% 2,13% 1,41% turkey 36,78% 26,85% 70,08% 1,53% 4,27% united kingdom 60,39% 111,60% 169,91% 2,39% 2,01% united states 55,25% 116,43% 172,02% 2,41% 2,06% average 48,61% 116,51% 168,17% 3,65% 2,48% european journal of government and economics 4(2) 101 database coverage country public debt to gdp private debt to gdp total debt to gdp real gdp per capita growth australia 1961-20121 1961-2012 1961-20121 1961-2012 austria 1961-20121 1961-2012 1961-20121 1961-2012 belgium 1961-20122 1970-2012 1970-20122 1961-2012 canada 1961-2012 1961-2012 1961-2012 1961-2012 czech republic 1993-2012 1993-2012 1993-2012 1991-2012 denmark 1961-20123 1961-2012 1961-20123 1961-2012 finland 1961-20124 1970-2012 1970-20124 1961-2012 france 1961-20125 1970-2012 1970-20125 1961-2012 germany 1961-20126 1961-2012 1961-20126 1961-2012 greece 1961-20127 1961-2012 1961-20127 1961-2012 hungary 1982-20128 1989-2012 1989-20128 1992-2012 ireland 1961-2012 1971-2012 1971-2012 1961-2012 italy 1961-2012 1961-2012 1961-2012 1961-2012 japan 1961-2012 1964-2012 1964-2012 1961-2012 luxembourg 1974-20129 2003-2012 2003-2012 1961-2012 mexico 1961-201210 1980-2012 1980-201211 1961-2012 netherlands 1961-2012 1961-2012 1961-2012 1961-2012 norway 1961-201212 1961-2012 1961-201212 1961-2012 poland 1986-2012 1992-2012 1992-2012 1991-2012 portugal 1961-2012 1961-2012 1961-2012 1961-2012 spain 1961-201213 1970-2012 1970-2012 1961-2012 sweden 1961-201214 1961-2012 1961-201214 1961-2012 switzerland 1961-201215 1961-2012 1961-201215 1961-2012 turkey 1961-2012 1986-2012 1986-2012 1961-2012 united kingdom 1963-2012 1961-2012 1963-2012 1961-2012 united states 1961-2012 1961-2012 1961-2012 1961-2012 1missing data for 1965. 2missing data for 1965, 1980, 1981 and 1989. 3missing data for 1997. 4missing data for 1964-1966, 1979 and 1980. 5missing data for 1978 and 1979. 6missing data for 1976. 7missing data for 1976-1978. 8missing data for 1993 and 1994. 9missing data for 1990. 10missing data for 19621964, 1969 and 1981. 11missing data for 1981. 12missing data for 1966, 1981 and 1982. 13missing data for 1963 and 1964. 14missing data for 1965, 1966 and 2003 15missing data for 1964-1969 vanlaer, marneffe, vereeck, van overtveldt ● does debt predict growth? 102 growth performance over 1 year with debt accumulation over 1-15 year (h=1 and d=1,5,10,15) – outliers removed growth performance over 5 year with debt accumulation over 1-15 year (h=5 and d=1,5,10,15) – outliers removed growth performance over 10 year with debt accumulation over 1-15 year (h=10 and d=1,5,10,15) growth performance over 10 year with debt accumulation over 1-15 year (h=10 and d=1,5,10,15) – outliers removed -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -20 -10 0 10 20 30 change in total debt to gdp (%) accumulated over 1 year -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -25 0 25 50 75 change in total debt to gdp (%) accumulated over 5 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -50 -25 0 25 50 75 100 change in total debt to gdp (%) accumulated over 10 years -1 0 -5 0 5 1 0 r e a l g d p p e r ca p ita g ro w th r a te ( % ) -50 0 50 100 150 change in total debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -15 -5 5 15 25 change in total debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in total debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 75 change in total debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 120 change in total debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -20 -10 0 10 20 change in total debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in total debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in total debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in total debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -20 -10 0 10 20 change in total debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in total debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in total debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in total debt to gdp (%) accumulated over 15 years european journal of government and economics 4(2) 103 growth performance over 10 years with debt accumulation over 1-15 year (h=10 and d=1,5,10,15) – public & private debt growth performance over 15 year with debt accumulation over 1-15 year (h=15 and d=1,5,10,15) – outliers removed -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -10 -5 0 5 10 15 change in public debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in public debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 75 change in public debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -60 -30 0 30 60 90 change in public debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -10 -5 0 5 10 15 change in private debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 -15 0 15 30 45 change in private debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in private debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in private debt to gdp (%) accumulated over 15 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -15 -5 5 15 25 change in total debt to gdp (%) accumulated over 1 year -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -50 -25 0 25 50 change in total debt to gdp (%) accumulated over 5 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -25 0 25 50 75 change in total debt to gdp (%) accumulated over 10 years -5 0 5 1 0 r e al g d p p e r ca pi ta g ro w th r a te ( % ) -30 0 30 60 90 change in total debt to gdp (%) accumulated over 15 years european journal of government and economics volume 4, number 1 (june 2015) issn: 2254-7088 the impact of good governance in the economic development of western balkan countries engjell pere, european university of tirana, albania abstract after the socio-economic transformation and the establishment of the free market institutions, the development and improvement of living standards in post-transition economies are deemed to depend more and more on the so called secondary ‘generating’ reforms, at the core of which is the good governance. drawing from this approach, this article seeks to address the role and the effect of the good governance in the economic development of the western balkan countries. more specifically, the article investigates the impact of good governance in the rates of economic growth of gdp. the article adopts a quantitative methodology approach, i.e. an econometric model based on the examination of a panel – data of good governance indicators for western balkan countries for the period 1996 – 2012. the analysis concentrates on albania, bosnia and herzegovina, croatia, kosovo, macedonia, montenegro and serbia. keywords western balkans; good governance; economic growth; governance index. jel classification h110; o40; r58. 25 pere ● governance and economic development of western balkan countries introduction this article aims at identifying the correlation between good governance and economic growth in the western balkan countries. after the political change in the 1990s, the economic development of these countries, for several years, was characterized by high economic growth rates, benefiting from the new political system based on democracy and transformation of economy towards the free market one. growth rates of gdp, in most western balkan countries, particularly prior to 2008, have been much higher than in west european countries. however, this positive development is accompanied with many lacks and critics regarding good governance; particularly in the area of corruption, law enforcement, property rights, etc. many political and economic scholars argue that good governance is one of the main factors, not only for the democratic development of the country, but it is also one of the primary factors in the economic growth. but some others are critical of this approach, arguing that this correlation can be only theoretical and there is not enough evidence to support it. in this light, the main research question refers to the attempt to confirm the correlation between improvement of governance and economic growth in western balkan countries. based on the positive data of economic growth on one side, and problematic aspects of good governance in these countries on the other, the main hypotheses is that in the western balkans, there is no clear evidence to support a positive correlation between good governance and economic growth. the methodological approach of this article is rather simple and as follows: upon completion of a literature review, the article analyses through descriptive interpretation the data on economic growth in western balkan countries and at the same time, for the same period, the indicators of good governance are also examined as estimated by international institutions. further, the article presents an econometric model, which shows the relation between good governance and economic growth for different countries in the western balkans. from the statistical point of view, the article uses the databases of the world bank (world governance indicators wgi), which bring together considerable data and statistical indicators from different economic, social and political international institutions. it is necessary to emphasize that this article does not seek to evaluate or discuss neither the content nor the reliability of the data published by the wgi. the analysis, as stated above, relates only to the region of the western balkans, and thus includes seven countries: albania, bosnia & herzegovina, croatia, kosovo, macedonia (fyrom), montenegro and serbia. drawing conclusions on the correlation between good governance and economic growth based on data for only seven cases might certainly have some weaknesses in argumentation. however, the presented work does not aim to draw general and universal theoretical conclusions, but only to illustrate the above correlation in this region. it is also necessary to remark that the analysis focuses on the period from 1996 to 2012. this relatively short period is related to the radical political and economic change in the western balkans, which started in the beginning of the 1990s.for the purposes of this analysis, it is not relevant to take into analysis the period before these changes occurred in the. this implies that the conclusions of the article should be considered only for this specific period and can be different in the long run. following the introduction, the paper will continue with three other sections: section 2 will present the variation of good governance indicators in western balkan countries; section 3 shows the positive economic growth in the region; and section 4 analyzes the correlation between economic growth and good governance. the results of econometric model are shown in annexes at the end of the paper. 26 european journal of government and economics 4(1) literature review many theoretical as well as empirical studies are dedicated to evidencing the relation between good governance and economic growth. there is a common acceptance that good governance is one of the main factors, not only for the democratic development of the country, but it is also a primary factor in the economic development. in this regard, hall and jones (1999) stress that the difference between economic developments in different countries, productivity, accumulation of capital etc., can be explained essentially by the difference in social structure. ‘the central hypothesis … is that the primary, fundamental determinant of a country's long-run economic performance is its social infrastructure. by social infrastructure, we mean the institutions and government policies that provide the incentives for individuals and firms in an economy’ (hall and jones, 1999). in this regard the evaluation of kaufmann (2003) is rather interesting, in which he analyzes the economic development of a certain number of countries for a long period (from 1970 still the beginning of 2000), and concludes that the slowing of growth rate of this last period (2002-2003) is not related only to the macroeconomic situation of certain countries, but also to the stagnation or the lowering of some of wellbeing standards, as are quality of the institutional structure, independence of the judiciary, level of corruption and the ease of doing business. such results are also drawn by roll and talbott (2003), who conclude that about 80 percent of differences in gni per capita between different countries ,can be explained by such factors as property rights, political rights, governance expenses, freedom of speech, etc., while negative effects come from excessive administrative regulation, informal economy, trade barriers, etc. from a historical point of view, a particular approach is given by acemoglu, robinson and jonson (2002). based on empirical data, they argue why among colonized european countries, those who were wealthier in the 16th century, now after 500 years, are relatively poorer. according to them, this difference cannot be explained simply by the impact of geographic factors, but exactly from the role of political institutions in the economic development of these countries as many authors highlight, good governance is seen as a key element in the economic development particularly in the developing countries. this thesis is related mostly to the fact that these countries inherit a limited market infrastructure which requires reforming governance initiatives to speed up free market initiatives and raise production (khan, 2007). based on empirical analysis, khan (2007) argues for a substantial correlation between good governance and an increase of income per capita, seeing good governance as an important factor in economic growth. in this framework, the so called ‘secondary generating reforms’, which have in their core good governance, are considered nowadays more and more important for the eastern european countries, including the western balkans. why ‘secondary’? the primary reforms in these countries are considered those that are related to the transformation of the country from a centralized economy to a free market system, the essence of which is establishing of free market institutions, i.e. privatizing, liberalization of economy, fiscal reform, building up of a new financial system, etc. after these reforms take place, the development of these countries is primarily related to good governance, in all aspects of this concept. drawing from this interrelation between good governance and economic growth, the question posed is: how can we evaluate good governance? is it possible to have unique indicators/indices to determine all dimensions of ‘good governance’? as stated above, this is related to the market and particularly with its efficiency. in this way knack (1995) determines five main indicators characterizing good governance: (i) corruption; (ii) legal framework; (iii) public administration efficiency; (iv) lack of contract execution by the government; and (v) expropriation. knack aggregates above indices in one aggregated index, so named ‘property right index’, which is evaluated from 0 (lowest) to 50 (highest). 27 pere ● governance and economic development of western balkan countries the world bank has considered as a good governance index the one from kaufmann, kraay and mastruzzi (2005), which groups it in six main pillars: (i) accountability and responsibility of governance – assessment of political and human rights; (ii) political stability and lack of violence – assessment of violent and terroristic acts; (iii) governance efficiency – the quality of public services; (iv) legal framework – assessment of politics which stumble the free market; (v) law enforcement – implementation of contracts, court verdicts, etc. (vi) corruption control – abuse of office for personal profits. from a general review of the literature on economic growth and good governance, there should be mentioned that some authors are critical of this approach, showing that this correlation can be only theoretical and there is not enough evidence to support it (kurtz and and shrank, 2007). according to them the relation between economic growth and good governance can be obvious only in developed countries or in a very long period of time. in this context, this article seeks to test this hypothesis in the case of the western balkan countries. it aims at identifying the correlation between good governance and economic growth in this region. governance indicators in the western balkan countries when attempting to analyze the impact of good governance in economic growth, the first problem that rises is that of definition, i.e. what do we mean by ‘good governance’. it is a broad concept and in many aspects very difficult to measure, particularly in quantitative figures. for that reason in the political-economic literature there is no clear definition of this concept. however, the object of this article is not to analyze the problematic aspects related with indicators that are used to express the level of good governance of a country. from this point of view, this article adopts the definition of the world bank regarding ‘good governance’ as provided in 2002 and later in 2007. in 2002 the world bank defined the governance as ‘the ability of the state to provide those institutions that support growth and poverty reduction-often referred to as good governance-is essential to development’ (world bank, 2002).the later definition was within the same perception (world bank), in which governance is defined as ‘...the manner in which public officials and institutions acquire and exercise the authority to shape public policy and provide public goods and services.’ (world bank, 2007). to have a clearer definition, today we mostly refer to the kaufmann and kraay (2008) classification, which classifies governance indicators in two groups based on two main criteria: (a) what do they measure (b) on what sources and opinions they are based. further, the first group of indicators includes two categories: (a.1) indicators who value the law enforcement or specific rules; and (a.2) indicators who value the concrete results of governance, specific politics or the outputs of this politics. regarding the second group of indicators, which means where they are based, (sources, opinions, etc.), governance indicators are classified in two other categories: (b.1) indicators which result from evaluation of different experts, and (b.2) indicators which reflect the results of a sample’s observation from groups of individuals and firms. although both methods have their weaknesses, this type of classification is important for the analysis and the results presented below. in this framework, the analysis of good governance for western balkan countries is based on the six main indicators defined by the world bank, (kaufmann, kraay and mastruzzi, 2005), elaborated for the period 1996 – 2012. 1. accountability of governance. 2. political stability and lack of violence. 3. governance efficiency. 4. legal framework. 5. law enforcement. 28 european journal of government and economics 4(1) 6. corruption control. the evaluation of these indicators form the world bank is made by ranking 230 countries on the bases of percentiles, meaning the percentage of other countries that have a lower indicator that the given country. [information about the elaboration’s sources of indices can be found in ‘worldwide governance indicators’ for respective countries 1992–2012.] the better the ranking, the more positive is considered the index of that country, because the percentage of the other countries which have a lower performance will be higher. meantime, the world bank makes an evaluation of each indicator of governance from –2.5 (bad performance) to +2.5 (good performance). since this paper focuses on the analysis of the effect of good governance in western balkan countries, the analysis is not based on the ranking of the countries (percentile), but on the evaluation from -2.5 to +2.5. it is clear that the higher the evaluation, the higher the ranking of the country will be. the value for these indicators for western balkan countries in 2012 are shown in table 1. table 1. good governance indicators in western balkan countries in 2012 (index valuation and ranking in 230 countries) alb bih hrv ksv mkd mne srb voice and accountability index valuation 10 -0.14 0.48 -0.22 0.00 0.23 0.17 rank 50 45 64 42 50 56 56 political stability & no violence index valuation -0.16 -0.54 0.58 -1.15 -0.44 0.56 -0.22 rank 40 29 64 15 33 64 39 government effectiveness index valuation -0.28 -0.47 0.70 -0.39 -0.07 0.13 -0.11 rank 45 39 72 42 52 60 51 regulatory quality index valuation 0.17 -0.06 0.44 -0.04 0.35 0.01 -0.08 rank 56 51 67 53 61 53 51 rule of law index valuation -0.57 -0.23 0.21 -0.56 -0.24 -0.01 -0.39 rank 35 48 60 36 48 55 44 control of corruption index valuation -0.72 -0.30 -0.04 -0.62 0.02 -0.10 -0.31 rank 27 49 57 30 59 55 48 source: world bank, worldwide development indicators – data (wdi), 2013 the first indicator is the ‘accountability of governance’. the variation of the evaluation of this indicator for western balkan countries is shown in graph 1. 29 pere ● governance and economic development of western balkan countries graph 1.accountability of governance (1996 – 2012) source: world bank, worldwide governance indicators (wgi), 2013 as shown in the graph, this indicator shows an improvement for all the countries in the period 1996 2000 and a relative stabilization after 2001. in general, for this group of countries the index is between -0.5/+0.5 (ranking 45 – 55 from about 230 countries). the highest evaluation is for croatia while the lowest is for kosovo (-0.22 in 2012). graph 2. political stability and no violence (1996 – 2012) source: world bank, worldwide governance indicators (wgi), 2013. the data for montenegro are from 2006 and for kosovo from 2008. 30 european journal of government and economics 4(1) regarding the ‘political stability and lack of violence’ (the second indicator, graph 2), the indices vary between -1,0 / +1.0, and ranks vary from15 for kosovo, to 64 for croatia and montenegro in 2012.the overall and significant characteristic is that, with exception of croatia and montenegro, all other countries have currently a negative sign rating, lower than the average level (zero). for kosovo, although the index has been very positive in 2008 (the first year of calculation of this indicator), there has been a significant drop in the next years. a characteristic for three other countries, albania, bosnia herzegovina and macedonia, is that the index suffered a decrease in the same year, 2011, reflecting the local insurgences that year; later in 2012 the evaluation improved. graph 3. governance efficiency (1996 – 2012) source: world bank, worldwide governance indicators (wgi), 2013. the data for montenegro are from 2006 and for kosovo from 2008. graph 3 shows the dynamics of the third index, ‘governance efficiency’. even here it is shown that except for croatia and montenegro, the other countries have a negative index, with slight positive trends. the ranking of the countries (2012) is from 39 (bosnia and herzegovina), to 72 (croatia, distinguished from other countries). 31 pere ● governance and economic development of western balkan countries graph 4. legal framework (1996 – 2012) source: world bank, worldwide governance indicators (wgi), 2013. the data for montenegro are from 2005 and for kosovo from 2007. the fourth index ‘legal framework’ is seen in graph 4. from the graph a considerable improvement is seen, for almost all countries, especially in the years 2003 to 2004. a characteristic of croatia is that, although it is still better than the others, the index has slightly decreased after 2009. graph 5. law enforcement (1996 – 2012) source: world bank, worldwide governance indicators (wgi), 2013. the data for montenegro and for kosovo are from 2003. graph 5 presents the evaluation of the ‘law enforcement’ index (rule of law). it is undoubtedly one of the most negative indicators for the western balkans. with the exception of croatia, the assessment is negative, despite the improvements starting from 2003. in 2012, (excluding croatia), the evaluation is between -0.01 (montenegro) and -0.57 (albania). the ranking of the countries for 2012 is in the 32 european journal of government and economics 4(1) range between 35-55 respectively for albania and montenegro (excluding croatia which occupies the rank 60). this means that these countries have the better index regarding law enforcement, compared with only 35 55 percent of 230 countries surveyed in this analysis. graph 6. corruption control (1996 – 2012) source: world bank, worldwide governanceindicators (wgi), 2013. the data for montenegro and for kosovo are from 2003. graph 6 shows another negative indicator for the western balkans, the ‘control of corruption’. for 2012, with the exception of macedonia (0,02) all other countries are below zero and there is no improvement for the entire period analyzed. the index is quite problematic and varied from macedonia 0,02; croatia -0.04; montenegro -0.10; bosnia and herzegovina -0.30; serbia -0.21; to kosovo and albania respectively 0.62 -0.72. albania and kosovo have ranked respectively 27 and 30 (that are better compared with only 27 and 30 percent of countries) while other countries are in the range from 48 to 59. regarding good governance in western balkan countries, we may refer also to penev (2012), who conducts a profound analysis of some aspects of the reforms undertaken in the region and underlines in detail the weaknesses of their introduction and implementation. in this point of view, we should underline for example: gaps in low enforcement; corruption levels; weaknesses of legal system in the area of rule of law, property rights, and judicial independence; insufficient cooperation between the governments and the respective parliaments; lack of capacities for regulatory impact analysis, etc. economic growth in western balkan countries in macroeconomic terms, gdp has been one of the most positive and stable indicators in the western balkans. in 2007, the gdp growth in the region, compared with average growth of gdp in the 27 countries of the european union (eu 27), shows that the balkan countries have had a rapid economic growth. during the years 2005 2007, gdp growth in the region has been higher than the eu on average 6-7 percent. even in 2008, when the economic crisis began in european industrialized economies, western balkan countries had a positive growth on an average of 5.5 percent. the slowdown in economic growth in the region began in 33 pere ● governance and economic development of western balkan countries 2009 in almost all countries. the decline was particularly significant in croatia (6.8%), macedonia (-5.8%) and serbia (-3.1%). graph 7. growth of gdp (1996 – 2012) source: world bank, worldwide development indicators – data (wdi), 2013 graph 8. growth gdp per capita (1996 – 2012) source: world bank, worldwide development indicators – data (wdi), 2013 graphs 7 and 8 report the growth rates of gdp and gdp per capita for the western balkan in the period 1996-2012. as seen from the two graphs, rates of growth were satisfactory, but have been getting lower after 2008. the decline is particularly more evident in the years 2009-2012, where several countries have had also negative rates. alb bih hrv ksv mkd mne srb 1996 9,10 88,96 5,92 1,18 7,80 1998 12,70 15,60 1,98 3,38 4,90 0,70 2000 7,30 5,50 3,75 4,55 3,10 5,34 2002 2,90 5,30 4,88 -0,70 0,85 1,90 4,12 2004 5,90 6,10 4,13 2,61 4,63 4,40 9,30 2006 5,00 6,20 4,94 6,00 5,03 8,60 3,60 2008 7,70 5,42 2,08 6,90 4,95 6,90 3,80 2010 3,50 0,70 -1,41 3,90 2,89 2,50 1,01 2012 0,80 -0,70 -2,00 3,80 -0,27 0,50 -1,70 -3,00 2,00 7,00 12,00 17,00 alb bih hrv ksv mkd mne srb 1996 9,65 90,88 10,04 0,77 8,22 1998 12,90 12,27 3,59 2,28 5,05 1,79 2000 7,72 3,23 6,75 3,65 3,13 5,68 2002 3,61 4,81 4,88 -0,77 0,40 1,77 4,17 2004 6,66 6,35 4,15 2,54 4,38 4,19 9,56 2006 5,55 6,33 4,98 5,15 4,84 8,42 4,01 2008 8,03 5,62 2,13 6,04 4,84 6,75 4,24 2010 3,53 0,90 -1,16 3,07 2,81 2,39 1,42 2012 0,54 -0,56 -1,69 2,91 -0,35 0,43 -1,23 -3,00 2,00 7,00 12,00 17,00 34 european journal of government and economics 4(1) good governance and economic growth in western balkan countries from the previous presentation of governance indicators set by the world bank (graphs 1 to 6) and growth rates (graphs 7 and 8), is noted that while governance indicators have not been optimistic, the rates of economic growth (gdp) were satisfactory (at least until 2008). this indicates the fact that economic growth in these countries should not significantly have been affected by the improvement of good governance, but normally by other factors. among these, we may include: the primary reforms related to privatization, opening of these economies to international trade, changes in the structure of the economy, increasing demand in the unsaturated markets, etc. to analyze in more detail such indications, and in order to measure the impact of good governance in the economic growth in western balkan countries, the following econometric model is used: gdpoecdλgcflogλ extgdlogλ log 321 1996/ ++ ++++++++= crrlawrreeffstbaccgdpagr capgdp γϕεδβα (1) in this model the indicators of good governance are those set by the world bank (kaufmann, kraay, and mastruzzi, 2010). in (1) grgdp/cap represents economic growth (growth of gdp per capita); loggdp1996 is the logarithm of the gdp per capita for the base year of the study (1996); inclusion of this noted variable in the model becomes important because countries with lower income per capita, should have the largest percentage increase; acc represents the governance indicator that reflects the ‘governance accountability’ (voice and accountability); stb is an indicator representing ‘political stability and lack of violence’ (political stability and no violence); eff is the indicator of ‘governance efficiency’ (governance effectiveness); rre is the indicator which expresses the ‘administrative and regulatory framework’ (regulatory quality); law is the index that characterizes the level of ‘law enforcement’ (rule of law); crr represents index of corruption’s level ‘(control of corruption). to ensure the same assessment of impact on the dependent variable (grgdp/cap), the above indicators are normalized with x = 0 and sd = 1. in this model three other variables are included, which are considered to affect in a considerable way economic growth, although not directly related to governance): extgdlog, it is the logarithm of trade opening index of the economy, which shows the percentage of exports and imports to total gdp; gcflog, which is the logarithm of the ration of capital formation to gdp and gdpoecd, which represents economic growth of gdp per capita in oecd countries. inclusion of extgdlog based on the assumption that the more an economy is open to international markets, the more substantial its growth will be, while the variable which is linked to the capital formation, gcflog, presupposes that the growth of capital formation to gdp will affect faster economic growth. further, gdpoecd variable is included based on assuming that economic development of western balkan countries, as relatively small economies, is affected by the economic growth on developed countries (oecd). evaluation of coefficients in model (1) is based on the world bank database for the western balkan countries (albania, bosnia and herzegovina, croatia, kosovo, macedonia, montenegro and serbia). the period taken in analysis is from 1996 to 2012. for some years during this period, particularly for montenegro and kosovo, there are no records, so they are completed using noted statistical methods (moving average). drawing conclusions on the correlation between good governance and economic growth based on data for only seven cases might certainly have some weaknesses 35 pere ● governance and economic development of western balkan countries in argumentation. however, this work does not aim to draw general, overreaching and universal theoretical conclusions, but only to illustrate the above correlation in this region. it is also necessary to remark, that the analysis focuses on the period from 1996 to 2012. this relatively short period is related to the radical political and economic change in the western balkans, which started in the beginning of the 1990s. for the purposes of this analysis, it is not relevant to take into analysis the period before these changes occurred. this implies that the conclusions of the article should be considered only for this specific period and can be different in the long run. on this basis, panel data are built which are processed in stata. the results are presented in annex 1. in this annex it is noted that: growth of gdp per capita (gdpcap), is statistically significant correlated from: (1) good governance efficiency (eff, p = 000), (2) the administrative regulatory framework (rre, p = 0,076), (3) law enforcement (law, p = 0,089), (4) income per capita in the base year (log96, p = 0.039), (5) share of capital formation to gdp (gcflog, p = 0,003) and (6) growth of gdp per capita in developed oecd countries (gdpoecd, p = 0,029). the link between economic growth and political stability is not statistically significant (stb, p = 0.136). while the results for the other indicators (acc, crr and extgdlog) show that the correlation is not statistically significant. taking statistically significant results, the data indicate that there is a positive correlation between the economic growth and (1) rule of law (law), (2) economic growth in oecd countries (gdpoecd), and in particular (3), the rate of capital formation to gdp (gcflog). in the meantime, the data show anomalies in correlation between economic growth and governance efficiency (eff), administrative regulation (rre), and the level of corruption (crr). for these indicators the coefficients are negative (respectively -8.22, -2.59 and -0.58), which means that the improvement of these indicators has not contributed to economic growth as might be thought. anomalies can be noted also in positive dependence of economic growth, from the initial level of income (year 1996), as is normally assumed that the greater is the basis of departure (beginner level), the smaller will be the rates of growth. after eliminating from the model any variables that are not statistically significant, the regression is repeated and the results are presented in annex 2. results show that the correlation of economic growth from analyzed variables is statistically significant (although for stb, p = 0.166). from indicators that characterize good governance, positive impact on economic growth have been those related to political stability and absence of violence (stb) and the improvements in law enforcement (law). it is also a positive impact of capital formation to gdp (gcflog), and economic growth in oecd countries (gdpoecd). the anomalies are observed in the impact of governance efficiency (eff) and regulatory reforms in the administrative context (rre). as mentioned above, in the economic literature there is a common agreement that good governance is an important factor in economic growth. the data analyzed above shows that for the western balkan countries, only specific aspects that characterize good governance have a positive and reasonable impact on economic growth, namely law enforcement (law) and political stability (stb), although for the latter p = 0.166. for some other elements the data shows a negative correlation (governance efficiency and regulatory framework), while for some others, statistical data do not provide a meaningful response (governance accountability and anticorruption measures). in this case we may return to the literature. kurtz and shrank (2006) argue that the positive correlation of economic growth and good governance, cannot always be evident. their arguments are related to two main problems. the first is related to governance indicators assessments, which are based on perceptions and may have a biased content. the second problem is based in the assumption that not only good governance affects economic growth, but overall economic growth and development can bring improvements in governance. 36 european journal of government and economics 4(1) according to them, there may be other casual and unobserved factors, which make ‘…select countries into high-growth/good governance or low-growth/malgovernance equilibrium’ (kurtz and schrank, 2007). as underlined above, the aim of this paper is not to discuss the content of the indicators set by the world bank and they are taken for granted. however, if we go to the data of annexes 1 and 2, it can be said that the economic growth for the western balkan countries was not based particularly in good governance, such as improvements in administrative business regulation, anti-corruption measures or efficiency of public services. in fact, this conclusion is clearly shown in the graphs above regarding good governance indicators (graphs 1 to 6) and economic growth (graphs 7 to 8). although the growth has been stable (at least until 2008), the majority of good governance indicators were negative ones with little improvement. in this regard, the econometric model presented in function (1) is modified, including in it also the performance of economic growth in previous years. in this way, economic growth is seen not only dependent on factors set in function (1), but also from the development in previous years. analysis is done in two options. in the first way (a), the economic growth in the current period is seen depending on the indicators and other variables in function (1) taken from the same year, but also on the growth of gdp per capita, respectively in the first, second and third preceding years. while the second option (b), economic growth is seen depending on the indicators of governance and other variables in function (1) taken from the previous year, and the growth of gdp per capita respectively in the first, second and third preceding years. in option (a) processing data in stata has brought the results shown in annex 3. here gdcap1, gdcap2 and gdcap3 represent the increase in gdp per capita respectively in the first, second and third years before the actual period. the data show that the dependence of economic growth from some aspects of good governance cannot be considered statistically significant (acc, stb, rre, crr and extgdlog). not taking these variables into account, the regression is repeated and results are presented in annex 4. here, all variables are statistically significant (p <0.05). the data show an anomaly regarding the correlation of the economic growth from governance efficiency (eff) and the initial level of income (log96). on the other hand, law enforcement (variable law) has a significant impact on economic growth in the same period, also this growth depends on the growth in the previous years (gdpcap1 and gdpcap2) and economic development in oecd countries (gdpoecd). compared with the analysis previously done, in this case, the variant (a) shows that the ratio of capital formation to gdp is replaced with the variables that reflect economic growth in previous years. this is acceptable because both these variables in fact present the same thing: the basis of economic development in prior years. compared to the previous regression, the determination coefficient (r-sq) is 0.54, which is larger compared to the 0.32 in the regression shown, in annex 2. in the variant (b), the economic growth (gdp/capita) in a period, depends not only on the growth in previous periods, but also on the governance in these periods. alternatively, the economic impact of the governance is seen in growth of gdp in the next years. the first results of the regression are presented in annex 5, while the final results (after eliminating variable that are not statistically significant), appear in annex 6. in this case growth is influenced positively by the governance accountability index (acc), economic growth in the previous year and economic growth in developing countries (gdpoecd). although statistically acceptable, the data show an abnormal dependence of gdp growth per capita, from the regulatory reforms (rre). the determination coefficient here is above 0.54. in the analysis, it is also taken into consideration the assumption of correlation of economic growth on specific governance indicators, seeing their dependence on the 37 pere ● governance and economic development of western balkan countries growth of gdp per capita in the first, second and third preceding years. it means analyzing the dependence in inverse: the impact of economic growth on improving the governance indicators. also in this case, data do not show logical dependence or the conclusions are not statistically significant. conclusions despite a relatively long duration of social and economic transformation, the improvement of governance indicators in the western balkan countries during the period 1996-2012 has been slow and below the average level of assessment. croatia makes an exception, which is above the average level indicators and positive. from six main indicators defined by the world bank as indicators of good governance, about four of these are still negative for six countries (from seven in total). the negative value has held in the particular indicators of ‘corruption control’; ‘law enforcement’, ‘governance efficiency’ and ‘political stability and lack of violence’. regarding the correlation of economic growth from good governance, the analysis of regression for the period 1996-2012 presents a messy dependence of economic growth on the good governance level. in some cases the dependence of these indicators is negative (governance efficiency, regulatory framework and corruption), while in some other cases it is not statistically significant. croatia with the most positive indices in governance has the most modest growth compared to the other countries. not all aspects of good governance have the same impact on economic growth and for some of them this impact is faster than others. the statistical analysis shows that political stability, absence of violence (stb) and the strengthening of law enforcement (law) affect the growth of the same period, but it is not evident for other indicators. statistical analysis shows that some aspects of good governance can be better identified for their impact on economic growth, displaced in time. governance accountability (acc) affects economic growth in future periods, which means it has a slower future impact. taking for granted the assessment of governance indicators, analysis conducted in paper shows that the impact of good governance in economic development of the western balkan countries can be interpreted only in the long term. it is not clear from the overall evidence of this relation for the short period of 10 to12 years. for the western balkans, statistical analysis shows that there is no significant dependency of the improvement of governance indicators on the economic growth in the previous periods. thus, the assumption that governance can be improved as a result of economic development in general, at least for the analyzed period, 19962012, is not confirmed. references acemoglu, daron, simon johnson and james a. robinson (2002) ‘reversal of fortune: geography and institutions in the making of the modern world income distribution‘, the quarterly journal of economics 117(4): 1231-1294. gradstein, mark (2004) ‘governance and growth’, journal of development economics 73: 505-518. hall, robert e. and charles i.jones (1999) ‘why do some countries produce so much more output per worker than others?’, quarterly journal of economics 114(1): 83-116. hossein, jalilian, colin kirkpatrick and david parker (2007) ‘the impact of regulation on economic growth in developing countries: a cross-country analysis’, world development 35(1): 87-103. 38 european journal of government and economics 4(1) kaufmann, daniel (2003) ‘governance redux: the empirical challenge’, working paper, mpra paper 8210. kaufmann, daniel and aart kraay (2003) ‘growth without governance’, economia: the journal of the latin american and caribbean economic association 3(1): 169 215. kaufmann, daniel and aart kraay (2007) ‘governance indicators: where are we, where should we be going?’ world bank policy research working paper series 4370. kaufmann, daniel, aart kraay and massimo mastruzzi (2010) ‘response to: “the worldwide governance indicators: six, one, or none”’, brookings institution, world bank development research group, and world bank institute. kaufmann, daniel, aart kraay and massimo mastruzzi (2010) ‘the worldwide governance indicators: methodology and analytical issues’, the world bank, institute global governance, policy research working paper 5430. kaufmann, daniel, aart kraay and massimo mastruzzi (2005) ‘governance matters iv for 1996-2004’, the world bank world bank, institute global governance. policy research working paper. kaufmann, daniel, aart kraay and massimo mastruzzi (2006) ‘growth and governance: a reply’, the world bank (forthcoming, journal of politics 69(2)). kaufmann, daniel, aart kraay and pablo zoibo-lobatón (1999) ‘aggregating governance indicators’, the world bank world bank, institute global governance, policy research working paper 2195. kaufmann, daniel, aart kraay and pablo zoibo-lobatón (1996) ‘governance matters’, the world bank world bank, institute global governance, policy research working paper 2196. khan, mushtaq (2007) ‘governance and development: the perspective of growthenhancing governance’, desa working paper 54. knack, stephen and philip keefer (1995) ‘institutions and economic performance: cross-country tests using alternative measures’, economics and politics 7(3): 207227. kurtz, marcus j. and andrew schrank (2007) ‘growth and governance: models, measures, and mechanisms’, journal of politics 69(2): 538-554. la porta, rafael, florencio lopez-de-silanes, andrei shleifer and robert vishny (1999) ‘the quality of government’, journal of law, economics and organization 15(1): 222-279. thomas, melissa a. (2009) ‘what do the worldwide governance indicators measure?’ european journal of development research 22(1):31-54. mauro, paolo (2002) ‘the persistence of corruption and slow economic growth’, imf working paper wp/02/213. penev, slavica (2012) ‘economic and european perspectives of western balkan countries’, westminster foundation for democracy. roll, richard and johntalbott (2003) ‘political and economic freedoms and prosperity’, journal of democracy 14(3): 75-89. manasan, rosario g., eduardo t. gonzalez and romualdo b. gaffud (1999) ‘indicators of good governance: developing an index of governance quality at the lgu level’, no. jpd 1999 vol. xxvi no. 2-d. philippine institute for development studies. sharma, shalendra d. (2007) ‘democracy, good governance, and economic development’, taiwan journal of democracy 3(1): 29-62. 39 pere ● governance and economic development of western balkan countries siddiqui, danish a. and qazi m. ahmed (2009) ‘institutions and economic growth: a cross country evidence’, mpra paper 19747. 40 european journal of government and economics 4(1) appendix annex 1. growth of gdp per capita and governance indicators rho 0 (fraction of variance due to u_i) sigma_e 8.0817479 sigma_u 0 _cons -85.86295 46.36586 -1.85 0.064 -176.7384 5.012459 gdpoecd 1.085017 .497624 2.18 0.029 .1096917 2.060342 gcflog 25.54283 8.57272 2.98 0.003 8.740604 42.34505 extgdlog -4.841363 13.70471 -0.35 0.724 -31.70211 22.01938 log96 18.2358 8.855933 2.06 0.039 .8784875 35.59311 crr -.5827184 1.619033 -0.36 0.719 -3.755966 2.590529 law 3.181707 1.871013 1.70 0.089 -.4854125 6.848826 rre -2.587529 1.456549 -1.78 0.076 -5.442312 .2672543 eff -8.219662 2.33311 -3.52 0.000 -12.79247 -3.646851 stb 1.736147 1.164004 1.49 0.136 -.5452597 4.017553 acc .6123372 1.34217 0.46 0.648 -2.018269 3.242943 gdpcap coef. std. err. z p>|z| [95% conf. interval] corr(u_i, x) = 0 (assumed) prob > chi2 = 0.0000 wald chi2(10) = 47.24 overall = 0.3519 max = 14 between = 0.8939 avg = 14.0 r-sq: within = 0.3099 obs per group: min = 14 group variable: code number of groups = 7 random-effects gls regression number of obs = 98 . xtreg gdpcap acc stb eff rre law crr log96 extgdlog gcflog gdpoecd, re 41 pere ● governance and economic development of western balkan countries annex 2 growth of gdp per capita and good governance indicators (statistically significant) rho 0 (fraction of variance due to u_i) sigma_e 7.9839582 sigma_u 0 _cons -96.01482 30.81755 -3.12 0.002 -156.4161 -35.61354 gdpoecd 1.022126 .47998 2.13 0.033 .0813824 1.962869 gcflog 24.46416 7.894859 3.10 0.002 8.990521 39.9378 log96 19.2147 7.743185 2.48 0.013 4.038337 34.39106 law 2.791953 1.242372 2.25 0.025 .3569483 5.226958 rre -2.579118 1.348562 -1.91 0.056 -5.222252 .0640152 eff -7.944702 2.236542 -3.55 0.000 -12.32824 -3.561161 stb 1.539992 1.11113 1.39 0.166 -.6377835 3.717767 gdpcap coef. std. err. z p>|z| [95% conf. interval] corr(u_i, x) = 0 (assumed) prob > chi2 = 0.0000 wald chi2(7) = 48.03 overall = 0.3480 max = 14 between = 0.7939 avg = 14.0 r-sq: within = 0.3160 obs per group: min = 14 group variable: code number of groups = 7 random-effects gls regression number of obs = 98 . xtreg gdpcap stb eff rre law log96 gcflog gdpoecd, re 42 european journal of government and economics 4(1) annex 3. variation of gdp growth from governance indicators in actual period and economic growth in previous years rho 0 (fraction of variance due to u_i) sigma_e 6.6936012 sigma_u 0 _cons -32.55872 39.33558 -0.83 0.408 -109.655 44.5376 gdpoecd 1.127075 .4185967 2.69 0.007 .306641 1.94751 extgdlog -1.686349 11.1711 -0.15 0.880 -23.58131 20.20861 log96 10.05325 7.616994 1.32 0.187 -4.875785 24.98228 gdpcap3 .4760815 .0799733 5.95 0.000 .3193367 .6328262 gdpcap2 .283829 .081155 3.50 0.000 .1247681 .4428899 gdpcap1 .0781734 .1262794 0.62 0.536 -.1693297 .3256765 crr -.8051266 1.343033 -0.60 0.549 -3.437423 1.82717 law 2.347946 1.57322 1.49 0.136 -.7355079 5.4314 rre -.8417507 1.151782 -0.73 0.465 -3.099202 1.415701 eff -4.400297 2.050324 -2.15 0.032 -8.418858 -.381737 stb .5715056 .9906176 0.58 0.564 -1.370069 2.51308 acc 1.067719 1.107259 0.96 0.335 -1.102469 3.237908 gdpcap coef. std. err. z p>|z| [95% conf. interval] corr(u_i, x) = 0 (assumed) prob > chi2 = 0.0000 wald chi2(12) = 111.50 overall = 0.5674 max = 14 between = 0.9560 avg = 14.0 r-sq: within = 0.5441 obs per group: min = 14 group variable: code number of groups = 7 random-effects gls regression number of obs = 98 > poecd, re . xtreg gdpcap acc stb eff rre law crr gdpcap1 gdpcap2 gdpcap3 log96 extgdlog gd 43 pere ● governance and economic development of western balkan countries annex 4. variation of gdp growth from governance indicators in actual period and economic growth in previous years (statistically significant) annex 5. variation of gdp growth, from governance indicators and economic growth in previous years rho 0 (fraction of variance due to u_i) sigma_e 6.510881 sigma_u 0 _cons -40.57649 19.58747 -2.07 0.038 -78.96723 -2.185744 gdpoecd 1.071694 .3898761 2.75 0.006 .3075512 1.835837 log96 11.62513 5.831543 1.99 0.046 .1955118 23.05474 gdpcap3 .4842415 .0765319 6.33 0.000 .3342418 .6342413 gdpcap2 .3166192 .0671942 4.71 0.000 .184921 .4483174 law 2.068566 .9828004 2.10 0.035 .1423121 3.994819 eff -4.63443 1.490092 -3.11 0.002 -7.554956 -1.713903 gdpcap coef. std. err. z p>|z| [95% conf. interval] corr(u_i, x) = 0 (assumed) prob > chi2 = 0.0000 wald chi2(6) = 113.85 overall = 0.5558 max = 14 between = 0.8747 avg = 14.0 r-sq: within = 0.5392 obs per group: min = 14 group variable: code number of groups = 7 random-effects gls regression number of obs = 98 . xtreg gdpcap eff law gdpcap2 gdpcap3 log96 gdpoecd, re rho 0 (fraction of variance due to u_i) sigma_e 2.6655707 sigma_u 0 _cons -4.077269 9.321261 -0.44 0.662 -22.3466 14.19207 gdpoecd .5632867 .1694072 3.33 0.001 .2312547 .8953187 log96 1.703298 2.76614 0.62 0.538 -3.718237 7.124834 gdpcap3 .1017265 .0331055 3.07 0.002 .0368409 .1666121 gdpcap2 .0098275 .0335076 0.29 0.769 -.0558462 .0755012 gdpcap1 .0496679 .0514519 0.97 0.334 -.051176 .1505118 crr -.8081095 .5375017 -1.50 0.133 -1.861593 .2453745 law -.9459614 .6012155 -1.57 0.116 -2.124322 .2323994 rre -.7157147 .4764095 -1.50 0.133 -1.64946 .2180308 eff -.3183857 .8567224 -0.37 0.710 -1.997531 1.360759 stb .3111996 .4266384 0.73 0.466 -.5249963 1.147396 acc 1.713261 .4679364 3.66 0.000 .7961226 2.6304 gdpcapi coef. std. err. z p>|z| [95% conf. interval] corr(u_i, x) = 0 (assumed) prob > chi2 = 0.0000 wald chi2(11) = 52.47 overall = 0.3991 max = 13 between = 0.7289 avg = 13.0 r-sq: within = 0.3696 obs per group: min = 13 group variable: code number of groups = 7 random-effects gls regression number of obs = 91 > e . xtreg gdpcapi acc stb eff rre law crr gdpcap1 gdpcap2 gdpcap3 log96 gdpoecd, r 44 european journal of government and economics 4(1) annex 6. variation of gdp growth from governance indicators and economic growth in previous years (statistically significant) rho .10568209 (fraction of variance due to u_i) sigma_e 2.6058534 sigma_u .8957869 _cons 2.163959 .5591069 3.87 0.000 1.06813 3.259789 gdpoecd .5023464 .1614302 3.11 0.002 .1859491 .8187438 gdpcap3 .0961148 .0314896 3.05 0.002 .0343964 .1578332 rre -1.683388 .3976871 -4.23 0.000 -2.46284 -.9039352 acc .8629871 .3567695 2.42 0.016 .1637318 1.562242 gdpcapi coef. std. err. z p>|z| [95% conf. interval] corr(u_i, x) = 0 (assumed) prob > chi2 = 0.0000 wald chi2(4) = 45.28 overall = 0.3102 max = 13 between = 0.2080 avg = 13.0 r-sq: within = 0.3797 obs per group: min = 13 group variable: code number of groups = 7 random-effects gls regression number of obs = 91 . xtreg gdpcapi acc rre gdpcap3 gdpoecd, re 45 european journal of government and economics volume 2, number 1 (june 2013) issn: 2254-7088 41 government vs opposition voting in the finnish parliament eduskunta since world war ii antti pajala, university of turku, finland abstract in a parliamentary system it is by definition justified to assume the government parties voting almost always in a unitary manner in plenary votes. in a multiparty system it is, however, hard to predict how the opposition groups vote. few studies analysing government-opposition voting in the finnish parliament eduskunta were published during the 1960s and 1970s. this study provides similar analyses regarding the parliamentary years of 1991-2012. combined the studies provide an insight into the government-opposition relations since world war ii. the results show that before the 1990s the government-opposition division in plenary votes appeared rather clear and the political party groups’ positions followed the traditional left-right dimension. since the 1990s, the government-opposition division has become greater. the governing coalition acts almost as a bloc while the opposition groups are divided into moderate and hard opposition. the opposition groups, however, appear in a more or less random order. consequently, since the 1990s the left-right dimension has disappeared with respect to plenary voting. jel classification d70; d72 keywords voting; parliament; opposition; finland acknowledgements the author wishes to thank the three anonymous referees for helpful ideas and suggestions. this work is supported by the academy of finland. european journal of government and economics 2(1) 42 introduction in parliamentary systems it is rather obvious to assume that the government groups almost always vote together in plenary votes. in such multi-party systems it is much harder to predict how the opposition groups vote. for example, some groups might act as support parties for government coalitions while some groups might nearly always oppose the coalition for tactical or political reasons. this study analyses opposition and government groups’ plenary voting behaviour in the finnish parliament ‘eduskunta’. in the analyses below the data cover the parliamentary years of 1991-2012. together with previous finnish plenary voting studies published earlier in the late 1960s and early 1970s we are able to provide an insight into cooperation between government and opposition in eduskunta since world war ii. the context of parliamentary politics in finland has changed considerably over time. nousiainen (2000, 2006) divides finnish parliamentary history into three distinctive eras. the first of them lasted from the beginning of the country’s independence in 1917 to the late 1930s. representative of the time were short lived bourgeois minority coalitions. after the exceptional years of the war, distinctive to the second parliamentary era were majority coalitions which were formed around the social democratic party (sdp) and the finnish agrarian union (ml, later renamed as the centre party kesk). later in the 1960s and onwards the communist finnish people’s democratic league (skdl) also participated from time to time in government. these majority coalitions were still short lived. the party groups were somewhat inflexible with their governmental policies and negotiating a solution during governmental crisis was impossible more often than not. the second parliamentary era continued to the early 1980s. the prolonging of this era was caused by “issues of foreign policy”, i.e. relations with the soviet union (nousiainen 2006, 294). consequently, the national coalition (kok), a bourgeois party, was not considered as a plausible coalition partner even though it had grown in popularity over the decades. the third parliamentary era can be seen to have started after sorsa’s fourth cabinet (1983-1986). governments lasting for the whole electoral period have been the standard almost without exceptions since then. what is distinctive to the third parliamentary era was also the dismissal of the old triangle of sdp, skdl and kesk as the backbone of the coalitions. the following the holkeri government (1987-1990) had sdp and kok as the main coalition parties. after the holkeri government basically any majority coalition could have and has been possible as ideological issues have not been an obstacle. a new triangular phenomenon took place during 1991-2010: out of the three largest parties (kesk, kok, sdp) two were the main coalition partners and one was left in opposition. the 2011 elections resulted in a sudden success for the populist true finns (ps) and the party became the third largest party in eduskunta. previous studies together with the applied data and analyses below will provide a rough picture of two parliamentary eras since world war ii. in the previous context plenary votes are at the core of political decision-making, however, the mass media are not interested in the votes unless there are notable dissidents within party groups or there are dissident government groups, for example. what the mass media is publishing very frequently is discussion which highlights the division between government and opposition. a weakly justified belief might well be that the opposition always votes against the government for political and tactical reasons. the previous finnish studies, however, provide evidence that this is not the case in eduskunta. also the analyses below lend support to this observation. moreover, i show that the government-opposition relations have changed considerably between the last two parliamentary eras. pajala ● government vs opposition voting in the finnish parliament eduskunta 43 while eduskunta has its own peculiarities and nuances, it can still be regarded as a rather typical mainstream parliamentary legislature. the main research setting, which is elaborated later on, is the following: first, most probably the government groups act in a unitary manner in plenary votes while the opposition groups do not. there is no systematically coordinated opposition. moreover, it is often the case that the opposition is comprised of parties in opposite ends of the traditional leftright dimension. second, the opposition groups vote with the government in varying degrees while an extreme opposition always opposing the government is not likely to exist. this setting leads to the main research questions: how united is the government? how united is the opposition? into what extent do the opposition groups vote with the government? what can we say about cooperation among the party groups during the last two parliamentary eras? the remaining step in the analysis is the theoretical question of what explains the party groups’ observed behaviour. the theoretical framework presented in arter (2006, 180-2) suggests parties exercising various strategies in parliamentary activities. the office-seeking, policy-seeking and profile-seeking (vote-seeking) strategies refer to opposition, although they can be used with respect to the governing parties as well. while the strategies are not mutually exclusive, they can be used as an aid in explaining the voting behaviour of the party groups. this work draws heavily on pajala (2011) published in finnish. the data regarding the modern years in pajala (2011) were limited to 1991-2006. the applied data below include six additional parliamentary years and now cover the 2007-2010 electoral term as well as the first two years of the 2011-2014 term. here i also apply a variation of the cooperativeness illustration method by laakso (1972a) instead of the original version used in pajala (2011). following the introduction is a review of the previous finnish studies and their main results concerning the second parliamentary era. subsequent to this i introduce the theoretical framework together with an account on finnish parliamentary opposition and its changing role in eduskunta. the next chapter starts by presenting the data, which are followed by the yearly analyses of government-opposition voting. next, a variation of the cooperation measure developed by laakso (1972a) is introduced and the votes are pooled and analysed per electoral terms. in order to get a comprehensive picture between the two parliamentary eras, i compare the results with the earlier findings of laakso and others. a short discussion concludes. previous and related research the political dimensions constructed below are explicitly interpreted as the government–opposition dimensions. as such they cannot be treated as “ideological” dimensions, but rather as “practical” dimensions apparent in parliamentary systems. recently pajala (2012) analysed plenary votes in eduskunta using the optimal scaling (oc) method developed by poole (2000; 2005). oc and other scaling methods are able to provide the ideal points of the mps in a number of political dimensions. as in latent variable analysis, the political dimensions have to be interpreted by the researcher. one of pajala’s (2012) main results is that during the last few decades the dimension having the most explanatory power is the government–opposition dimension. higher, possibly ideological, dimensions had only very marginal explanatory power and could not be sensibly interpreted. put another way plenary votes seem to reflect ideological dimensions rather poorly. ideological dimensions colour the background while plenary voting is everyday politics. setting aside plenary votes, political dimensions and parties’ ideal points have been studied by other means as well. use of expert interviews by benoit and laver (2006) is one such possibility while the textual analysis of party programmes by klingemann et al. (2006) is another. these data have also been gathered regarding finland, however there is slight variation in the positioning of some of the european journal of government and economics 2(1) 44 parties between the expert interviews and textual analyses. recently paloheimo (2008, 54), using multidimensional scaling in conjunction with opinion poll data, concluded that the traditional left-right dimension is still the most important ideological dimension in finland. while plenary votes no longer highlight the left-right dimension, this was not the case in the past. regarding previous decades some pioneering plenary vote analyses were carried out by risto sänkiaho (1969), markku laakso (1972a; 1972b) and pekka nyholm (1969; 1972). this promising array of analyses seems, however, not to have continued afterwards. the data regarding the above studies concern the 1950s and 1960s and provide us with a rather good view on the second parliamentary era of finland. as below, the historical analyses were carried out in the ppg level. however, instead of having the government majority fixed to the extreme-right the studies had the national coalition fixed to the right (in a technical sense). the methods applied in the laakso, sänkiaho and nyholm studies varied, although the basic idea in all of them was to measure and evaluate ppg cooperation and distances by comparing group majorities vis-à-vis each other in plenary votes. sänkiaho also compared individual members of parliament (mps) somewhat along the lines of modern scaling methods. the results in the studies are rather similar regardless of the method used. the cooperation or distance measure introduced by laakso (1972a) together with a variation of his graphical presentation method is applied below. pekka nyholm (1972) relied on older technique and used a slightly simpler cooperation measure proposed by stuart rice (1928). although the research of sänkiaho (1969) is limited in data, it is methodologically the richest. he starts his account by finding out how the individual voting decisions of the mps correlate over the votes. in order to find out factors affecting the voting decisions of the mps the correlation matrix is then used for factor analysis. the most important factor thus obtained was the party affiliation of the mps. laakso, sänkiaho and nyholm studied various pre-defined ppg combinations over the votes as well. deviating from other papers sänkiaho reports an interesting detail according to which the mps voted but few times against their respective ppg majority. this result complements the earlier voting cohesion analysis of nyholm (1961) by showing that deviations from complete ppg voting unity are not a result of the same rebellious mps. two common denominators in the historical studies are: (1) plenary votes highlight the difference between government and opposition ppgs. as said above, this is a feature parliamentary systems, however compared with the results below, the distance between government and opposition now appear to be greater (2) plenary votes highlighted the ppgs appearing according to the traditional left-right dimension as well. here, however, the second and third parliamentary eras differ in one important aspect. previously, the main coalition partners were always adjacent groups in the left-right dimension. in more recent times (since the holkeri government 1987-1990), this has only once been the case. if the cabinet is not connected, this affects the plenary voting of the opposition as well. below we shall treat the ppgs as having just one voice. this assumption is realistic and justified as already the historical studies showed that the most important factor behind the voting decisions of the mps was found to be the party affiliation. indeed, intra-party voting cohesion in finland has been very high throughout the two latest parliamentary eras. over the decades, the voting cohesion has increased even further (nyholm and hagfors 1968; pajala and jakulin 2007). among the nordic countries voting cohesion in finland has been the lowest (jensen 2000). during the 2007-2010 electoral term, some of the ppgs had perfect voting unity, so the difference nowadays is likely to be hairbreadth. internationally, ppgs in parliamentary systems typically have very high internal voting cohesion. the rice cohesion index ranges from zero to one, and typically pajala ● government vs opposition voting in the finnish parliament eduskunta 45 the respective values are well over 0.9 (rice 1928, sieberer 2006). systems showing low ppg voting cohesion are usually non-parliamentary. the usual examples are the european parliament and the u.s. congress, for which the cohesion values are around 0.6-0.7. opposition in eduskunta finally, as the assumption of ppgs in the governing coalition voting similarly is at the very heart of parliamentary systems the main interest lies in the behaviour of the opposition. it seems there is rather little systematic research on parliamentary oppositions (for references see pajala 2011, 216). mainly, the literature focuses in defining and studying the nature of opposition and distinguishing different types of opposition in various countries. regarding finland the chapters in arter (2006) and the monograph by rantala (1982) are probably the most recent and comprehensive accounts. the standard finnish textbooks review the opposition only very briefly (nousiainen 1998; paloheimo and wiberg 1996; wiberg 2008). the book series published in the honour of the 100-year-old eduskunta include some articles (jyränki 2006; nousiainen 2006; ollila 2007). parliamentary opposition in finland is defined here to consist of those ppgs and mps who are not in the government coalition. the most important tasks of the opposition usually found in the literature are listed for example by helms (2008): (1) criticising the government, (2) scrutinising and checking governmental actions and policies, and (3) representing a credible ‘alternative government’. two kinds of opposition are identified by sartori (1971), which are relevant here: first there is the responsible constitutional opposition, which includes ppgs who are aware of a realistic possibility of being in the government in the future. the behaviour of these parties is usually rather modest and realistic in parliaments and in parliamentary elections. the second type is a constitutional but non-responsible opposition. parties belonging to this category are aware that their probability to govern is very low. these irresponsible opposition parties tend to be “promising wildly and outbidding” as sartori (1971, 35) puts it. in the newest i.e. the third finnish parliamentary era the opposition parties can all be categorised being responsible opposition parties. basically any of the ppgs could have been in government responsibility. in the preceding second era, at least two right-wing parties (including the national coalition kok) were in permanent opposition for some decades, however these parties still acted more or less as responsible oppositions. during this era, the strong left-right dimension limited the number of possible government coalitions, however the gradual disintegration of the dimension opened up new avenues. according to nousiainen (2000; 2006, 297) after the 1999 parliamentary elections any majority coalition could have been possible. ideological restrictions were not obstacles any more. in fact, probably the same would hold for at least two previous elections as well. as finland is not a two-party system the opposition cannot represent a credible alternative government, but its main task remains to criticise the government and to some extent scrutinise and check governmental actions and policies. deviating from finland the scandinavian neighbours have a long tradition of minority governments, which often need support from opposition parties in order to pass legislation. hence, the supporting parties might be in “opposition” in varying degrees (christensen and damgaard 2008). summing up the previous, opposition ppgs in eduskunta will present policy alternatives and hence will distinguish itself from the governing coalition. nousiainen (2006) characterises the current role and choice space of the opposition: as the finnish parliamentary system is currently a true majority one the opposition groups can be regarded as powerless spectators while the governmental parties negotiate and decide upon major political issues. the most important political decision making and negotiation arena is the cabinet negotiations and especially the writing of the government programme. this process lasts few weeks after the elections, however, the most important policy formulation european journal of government and economics 2(1) 46 takes place during this short period of time. the most visible arena the opposition is left with is the plenary hall (ollila 2007). during plenary sessions, the opposition has the opportunity to criticise the actions and policies of the government and provide their own alternatives to government bills by votes. also parliamentary interpellations as well as government reports and announcements (which can include the votes of confidence regarding a single minister or the whole cabinet) are handled in plenary sessions. the role of the opposition has not always been as limited. during the second parliamentary era, the postponement rule and especially two-thirds qualified majority requirement guaranteed the opposition, if united, the possibility to delay an ordinary law proposal to the next annual parliamentary session when it had to be adopted unchanged in order to become a law. should this happen it would have been a defeat for the government. sometimes only the threat to use the postponement possibility guaranteed the united opposition (minimum of 67 mps) negotiation leverage. the current “true” majority system has been seen being established along with the 1992 partial constitutional reform when the postponement rule and the qualified majority requirement regarding ordinary legislation were removed from the old 1906 constitution and the 1928 parliamentary act. previous partial reforms in the 1980s to the 1928 parliamentary act had changed the postponement of a law proposal only to the next annual parliamentary session instead of postponing it to the next electoral term (helander 1990; jyränki 2006, 102-105). the old regulations can be seen as being rather efficient as laws were very rarely postponed. according to helander (1990, 57) on average only 1.5 law proposals were postponed yearly during 1917-1986. modern finnish society in general does not resemble much of what it was during the 1950s, 1960s or even the 1970s. in the past, the class cleavages were substantially wider. since world war ii finland gradually became a very wealthy country and at the same time the class cleavages became much thinner. the vast majority of the finnish population can be seen belonging to the upper or lower middle class. at the same time also political parties have changed and their political programmes have significantly converged (paloheimo and raunio 2008; paloheimo 2008). parties have adopted catch-all strategies which aim at maximising votes in general elections. paloheimo and raunio (2008) call modern parties “election parties” (vaalipuolue). this progress continued more or less undisturbed for decades. only after the 2011 elections the party system experienced a major shock. the true finns (ps), a radical populist protest party, won the elections and suddenly became the third largest party in eduskunta. compared with the second parliamentary era the governing coalition can nowadays be regarded as a true bloc. instead of the short lived cabinets of the second era, which nousiainen (2000; 2006, 294) characterises as politically fast pulsing, overall rather unorganised and only weakly predictable, the governments have remained in power for the whole electoral periods since 1983. now the government programmes are long and detailed thus requiring a strong commitment of the coalition partners. in the 1970s, the government programmes were few pages in length while the current programme of the katainen government contains over 100 pages and is the result of intense negotiations after the 2011 elections. the coalition partners also sign a set of written rules guiding their behaviour in the parliament. hence, inside the coalition the ppgs or mps are not allowed much room to move and the bloc acts almost in a unitary manner. the opposition, in turn, is not systematically coordinated. its position in the parliament is hard and the main public working arena is the plenary hall. the role of the opposition is narrowed to criticise the actions of the government in speeches and providing alternatives to government bills (creating votes). chances having an effect on major political decisions or strategies are virtually non-existent. in order to predict how the opposition parties vote we shall turn into the basic framework of ppg strategies in parliamentary systems. accordingly, ppgs can adopt three main pajala ● government vs opposition voting in the finnish parliament eduskunta 47 strategies: first, an office-seeking strategy, second, a policy-seeking strategy, and finally, a profile-seeking (vote-seeking) strategy. the strategies are not mutually exclusive, but rather complimentary. (arter 2006, 181-182.) these strategies lead us to the following rank order prediction of opposition ppgs. the largest ppgs (kesk, kok, sdp and ps since 2011), when in opposition, will adopt mainly an office-seeking strategy. these parties will remain in moderate opposition as they do not risk the chance of being considered too far from the mainstream policy consensus. in opposition they will not irritate the governing coalition too much as this might have negative consequences in future government formation. other, smaller ppgs, while also interested in stepping into office, have lower chances of succeeding. these parties have a stronger need to appeal to their voters and thus need to exhibit more profile-seeking strategies when in opposition. therefore the distances of the smaller ppgs from the government majority are likely to be greater than the above large parties. the hardest (but not necessarily extreme) opposition is likely to consist of (small) radical parties (arter, 2006, 182). in finland these would include the radical populist ps and possibly vas, which is the leftmost ppg regarding the left-right dimension. these parties probably exhibit the most active policy-seeking strategy. finally, in the following section the voting patterns are studied per parliamentary year as well. when elections approach in the last year of an electoral term, parties need to stand out. with respect to plenary voting opposition parties can act in two ways: vote more frequently with the government or against it. the former would indicate a party getting closer to the mainstream political consensus presenting itself as a viable coalition partner. the latter would indicate a party presenting itself as a true alternative to the governing coalition. both strategies would be carried out in order to attract voters. however, as plenary votes are mostly a product of opposition activity and opposition parties support their own initiatives, intuitively it should be the case that opposition parties move apart from the government during the last year of an electoral period. data and analyses the applied data is collected by pajala and jakulin (2012) and include detailed information about all 12269 plenary votes taken during 1991-2010 parliamentary years. pajala (2013) has gathered 1077 votes for the 2011 and 2012 parliamentary years. roughly half of the votes are budget amendment votes. these votes take place in december when eduskunta decides upon the state budget for the next fiscal year. for more details on the votes see pajala (2006). as i shall consider aggregate level votes instead of individual mp vote decisions, two definitions are needed for the analyses below: first, the voting choice of a ppg is the one supported by the majority of the ppgs members. consequently, if the majority of the members of a ppg voted “yes” then the voting choice of the whole ppg is also “yes”. otherwise the ppgs voting choice is “no”. i shall not consider abstentions or absent mps. second and likewise, the voting choice of the government is the one supported by the majority of the government coalition mps. one person groups have been disregarded. for example, the true finns exist only since 2003 even though the party had one mp (raimo vistbacka) during 1995-2002. government-opposition voting per parliamentary year comparing ppg majorities with government majorities over plenary votes in a parliamentary year let us see in relative terms how often a ppg cooperates with the government. table 1 shows the results in columns 3-12. the second column indicates the number of plenary votes in the parliamentary year and the third column shows the share of votes where the government groups were not in european journal of government and economics 2(1) 48 complete unison. as can be seen the government groups have nearly always (in well over 90 percent of the votes) been in complete unison. on average only some 5 percent of the votes are such that one or more government groups voted against the government majority. since 2007 the disunity of the government has been less than 1 percent and during the past two parliamentary years the cabinet has been in complete unison. in this sense the last six years in table 1 deviate from preceding times. during 1994 and 2002 cabinet disunity is exceptionally high: regarding the former year the christian democrats (kd) resigned from the cabinet, as did the green league (vihr) regarding the latter year 2002. also in 1996 the government disunity is rather high as in 12.4 percent of the votes the government coalition was not in complete unison. while no obvious reason for this phenomenon is apparent the disunity can be seen throughout the lipponen ii government (1999-2002) coalition, however the sdp seems to toe the government line better than others. the same applies to 2002a, however this time it is an election year. in the other columns of table 1 we can see the cooperation scores. the bold and underline style government groups deviate clearly from the opposition groups as their cooperation scores are always very close to the maximum value 100. once again the change in 2007 and onwards is clear as the maximum value appears more often than not in the cabinet ppg cells. resignation from the cabinet substantially lowers the cooperation scores as can be seen in the cases of kd and vihr in rows 1994b and 2002b, respectively. at least as interesting are the opposition ppgs cooperation scores. an extreme opposition group, which would have cooperation score zero does not exist in table 1, however, after the 2011 parliamentary elections the true finns (ps) were nearly there in 2011 when their score was only 2.0 out of 100. indeed, the hardest opposition politics has been exercised by the ps, especially after the 2007 elections. also left-wing alliance vas (when in opposition) has shown similar behaviour, most visibly during the 2007-2010 term. the agreement score of the kd has been on the rise throughout their opposition period and was already over 50 in 2010. however, the most moderate group has been kok during 2003-2006. well over 50 percent of the votes were such that the majority of kok mps voted in line with the government. for other times the same holds for the young finns (nuors) and to some extent also vihr. whenever one of the largest groups (sdp, kesk or kok) was in opposition, the ppg exercised rather modest voting behaviour against the government, especially in the later years. a notable exception is ps during the last parliamentary term: ps is now the third largest group; however it is almost an extreme opposition party. only the swedish people’s party (sfp) has been continuously in the cabinet. an interesting detail is a cycle where opposition parties tend to get closer to government during the last year of an electoral term. this observation is against the prediction stated above. nyholm (1961, 124-125) defines two aspects with respect to matters processed in eduskunta. first, there are matters for which the technical contents are very important. second, there are matters for which the political significance is high. nyholm (1961, 124) continues to propose a dimension where one extreme would denote matters of pure technical interest. in the middle are matters with rather balanced political and technical importance. in the other extreme are matters of utmost political importance. these are votes of confidence such as parliamentary interpellations, government reports and announcements. nyholm (1961) observed that ppgs voting unity is higher in the votes of confidence as did pajala (2010, 1415) regarding the current data. keeping in mind the results in table 1 the ppgs seem to approach the vast majority of the votes considering the technical contents of the proposals. it would seem that the political importance of the votes is not very high as otherwise the scores in table 1 would be lower than they appear. following the idea of nyholm (1961) pajala (2011) separated and analysed the votes of confidence. the hypothesis was that compared with the results in table 1 the government groups should now show higher agreement scores and the opposition groups lower. while the prediction for the former was correct the latter pajala ● government vs opposition voting in the finnish parliament eduskunta 49 groups did not show lower scores, but in some cases even higher. the ppgs internal voting unity is higher regarding votes of confidence, however the opposition groups do not vote more often against the government in these votes as one might have assumed. table 1. government groups’ voting unity and shares (%) of plenary votes during 1991-2012, where other ppgs voted with the government majority year vot. gov . sdp kes k ko k vih r sfp kd vas ps vr var nu or s sm p 1991 621 2,6 51,2 99,8 100, 0 40,6 98,9 98,4 13,2 34,1 1992 834 4,9 26,5 99,6 99,4 26,1 98,3 97,2 17,4 24,1 1993 897 6,5 28,8 98,6 99,0 25,4 97,1 97,8 16,6 17,2 1994a 179 12,8 22,9 96,1 98,3 25,1 95,5 92,2 19,0 32,4 1994b 795 5,0 46,2 97,5 98,6 32,1 98,9 55,0 21,6 28,3 1995 273 1,8 100, 0 19,4 100, 0 98,5 100, 0 16,5 99,6 27,1 52,0 1996 315 12,4 98,1 24,8 95,6 95,2 93,0 26,3 97,5 48,9 53,3 1997 279 6,1 97,1 28,3 99,3 96,8 99,3 25,1 97,1 54,8 50,9 1998 372 7,5 98,7 34,7 98,9 95,2 97,8 26,3 96,8 58,1 62,4 1999 249 4,0 98,4 45,8 99,2 98,4 98,8 31,3 98,0 2000 253 5,9 98,0 34,0 98,8 95,7 98,4 20,9 96,0 2001 271 5,9 99,3 36,9 99,3 97,8 99,3 30,6 97,4 2002a 25 16,0 100, 0 24,0 96,0 84,0 92,0 24,0 96,0 2002b 456 3,7 98,7 48,7 99,3 61,4 98,9 37,7 97,6 2003 491 2,9 98,8 100, 0 68,2 51,9 98,4 45,6 25,7 15,7 2004 723 4,3 97,9 99,7 66,0 52,6 98,1 47,7 24,6 22,3 2005 815 1,3 99,8 100, 0 56,8 39,8 98,9 37,8 14,7 14,8 2006 1002 2,0 99,2 99,8 64,5 43,1 99,0 37,7 11,7 10,9 2007 668 0,2 65,6 100 99,9 99,9 100 51,5 4,2 9,6 2008 722 0,1 49,7 100 100 99,8 100 48,9 4,2 9,0 2009 924 0,1 50,3 100 100 100 99,9 49,6 4,5 8,5 2010 1099 0,9 57,6 99,7 99,8 99,5 99,9 54,8 8,2 7,6 2011 448 0 100 62,1 100 100 100 99,8 99,8 2,0 9,8 2012 629 0 100 60,3 100 100 100 100 100 9,2 8,3 notes: vot: number of plenary votes; gov: share (%) of votes, in which the government coalition did not vote complete unison; 1994b and 2002b: changes in government composition; party abbreviations not in text: vr and var are small defected factions from vas, the finnish rural party smp is the predecessor of ps; ppgs in governments are bolded and underlined. source: pajala and jakulin (2012) and own computations. turning further into the opposition results in table 2 provide us with an understanding how united the opposition ppgs are in plenary votes. the most important finding is the large variance in how many groups (group majorities) voted against the government. the range is from all opposition parties down to no opposition parties. in the latter case, the votes were the result of proposals by only one mp or a minority of mps from one or more opposition groups. votes like these appeared especially during the lipponen government (1999-2002) after which they seem to have disappeared almost completely. why this is the case remains an open question. during the aho (1991-1994) and lipponen ii governments the most frequent case was the whole opposition voting against the government. the number of opposition groups is at a minimum during the lipponen ii (kesk, kd) and katainen (kesk, ps, vr) governments yet only in roughly half of the votes the opposition has been united. preceding the katainen administration (2011-2012) kesk lost the elections while ps was the winner. both parties ended up in the opposition. apparently kesk lost votes especially to ps which would explain the reluctance of kesk to cooperate with ps. perhaps the same applies to the era of european journal of government and economics 2(1) 50 the lipponen ii government as to some extent the kd and kesk appeal to the same voters. for other times the opposition seems to be rather evenly scattered. since 2005 cases where one opposition group would not have been backed up by at least one other opposition group are rather seldom. why the opposition is so scattered is probably due to the fact that votes are created as a result of proposals from opposition groups. one opposition group might be interested in some topic and sees it important or potentially advantageous to register their opinion to the plenary minutes while other opposition groups might show no interest to the matter at hand. after all, opposition in eduskunta is not systematically coordinated. table 2. shares (%) of how many opposition groups voted against the government during 1991-2012 year 0 1 2 3 4 5  (%) 1991 0,6 21,7 21,7 27,9 28,0 100 1992 2,5 10,2 14,3 24,9 48,1 100 1993 2,2 9,7 14,8 20,3 53,0 100 1994a 1,7 14,0 11,7 27,4 45,3 100 1994b 2,5 8,2 21,0 25,8 23,3 19,2 100 1995 1,5 12,8 18,7 33,3 33,7 100 1996 4,4 18,7 21,9 35,6 19,4 100 1997 7,2 18,6 19,4 35,8 19,0 100 1998 9,9 19,4 23,9 35,8 11,0 100 1999 26,1 24,9 49,0 100 2000 18,2 18,6 63,2 100 2001 23,6 20,3 56,1 100 2002 22,2 23,7 28,5 25,6 100 2003 0,4 18,1 21,0 24,2 21,2 15,1 100 2004 1,1 16,3 24,3 24,2 20,9 13,1 100 2005 0,0 6,6 22,7 22,3 24,7 23,7 100 2006 0,6 5,9 23,9 22,5 24,9 22,4 100 2007 0 6,7 40,4 29,8 23,1 100 2008 0 4,4 34,1 30,3 31,2 100 2009 0 5,5 34,8 26,7 32,9 100 2010 0,2 6,9 40,3 26,1 26,5 100 2011 0 7,6 58,7 33,7 100 2012 0,1 8,7 59,8 31,3 100 notes: for example in 1991 four opposition groups voted against the government in 28 percent of the votes; 1994a: kd in government; 1994b: kd not in government. source: pajala and jakulin (2012) and own computations. government-opposition voting per electoral term so far we know the opposition groups are voting against the government with varying intensities. there are no extreme opposition groups which would always vote with or against the government. this is hardly surprising as there are no antiestablishment parties in eduskunta; moreover, minority governments (with their possible support parties) have not existed for decades. in what follows the analysis focuses on electoral terms instead of single parliamentary years in order to provide a more general picture on the cooperation among the ppgs. government majority, as previously, is assumed to be an ‘actor’ among the party groups. for the analyses, i shall use a variation of a measure of group cooperation (or group distance) defined by markku laakso (1972a; 1972b). this measure has a very intuitive vector presentation. basically, laakso’s measure is a more sophisticated version of the previous cooperation measure of rice (1928). the measures are, however, not comparable and rice’s measure does not have a vector interpretation (laakso 1972a, 9). from previous finnish literature only laakso’s results can and will be compared with the ones presented below. the rice (1928) and laakso (1972a) measures are basically just simple measures of distance between two voters over some set of votes. first we assume the voting choice of a ppg being the choice favoured by the majority of the ppg. the majority assumption is in our view rather justified as the voting minorities of the pajala ● government vs opposition voting in the finnish parliament eduskunta 51 ppgs, if any, are typically marginal (pajala 2010). next we assume the voting choice of the government being the one supported by the majority of government groups’ mps. there are four possibilities how two groups ai and aj can vote: both can vote “yes” (yy) or “no” (nn) or the groups can disagree (yn or ny). so, the more there are yy and nn cases thus lowering the number of yn and ny cases the closer the groups are. the difference between the voters’ cooperation and disagreement is then divided by the number of votes v. formally, laakso’s distance c in a vote can be obtained as     v nyynnnyy c ji aa   , where index ji aac can have values in the range of [-1, 1]. value 1 denotes complete agreement and -1 complete disagreement between two voters over the votes. at zero the voters (dis)agree on half of the votes. the computation over the votes will result in a distance matrix. the closeness or cooperation between two groups can now be illustrated geometrically by representing the groups as unit vectors and computing the angle between the vectors by using the values of ji aac . if we consider the ppgs as vectors in a space, their relation to the government majority can be visualised in the upper half of a unit circle (the length of the vectors is irrelevant). the government vector is fixed to the rightmost position i.e. the positive x-axis. starting from origin the end point of the government vector is thus at (1, 0). a ppg vector always disagreeing with the government and having an index value of -1 would be mirroring the government in the negative x-axis ending at (1-, 0). a voter having an index value of 0 would be (dis)agreeing with the government in exactly half of the votes having a vector equal to the positive y-axis ending at (0, 1). the index value is simply the angle between the unit vectors. in laakso’s original visualisation, the index value is the projection of a voter’s vector to the x-axis. now, it is of course a matter of taste which version is applied. we like to apply the simpler visualisation and highlight the closeness of the government parties by using the angle as the index value. the projection version diagrams would instead boost the differences with respect to the government parties due to the underlying non-linear trigonometry. for details on laakso’s original method see pajala (2011). as an example consider groups kok and the agrarian union (ml) in figure 1 having ji aac index value 0.83. converted to degrees we have 90*0.83 = 75 degrees (to the right of y-axis). likewise kok and sdp have ji aac = -0.36 in figure 1. converted to degrees we get 90*-0.36 = -32 degrees (to the left of the zero point i.e. the y-axis). instead of having the government majority as the rightmost voter laakso (1972a; 1972b) put kok as the rightmost vector. the choice was quite understandable as at the 1960s kok was considered the right-wing extreme in the technical sense that there was nothing to the right of kok. thus, all other groups could be measured against kok. this resulted in a sort of pre-defined left-right dimension. in order to shed light to the second parliamentary era of finland we turn to the three governments analysed in laakso (1972a). the oldest of these was lehto’s caretaker cabinet followed by virolainen’s (kok, ml, sfp, kp) right-wing coalition and finally the subsequent paasio government (sdp, skdl, kesk) which was a centre-left-wing coalition. figure 1 shows the political landscape during the virolainen government 1964-1966. the basic party setting and left-right dimension is almost identical regarding all three previous governments as it is in figure 1. although lehto’s preceding caretaker cabinet was not responsible politically, practically the only difference with regard to figure 1 was ml and vm being close together roughly where ml is in figure 1. during the paasio government, the leftwing coalition partners (sdp, skdl) together with tpsl were close together roughly where tpsl is in figure 1. also kesk was close to the left-wing coalition european journal of government and economics 2(1) 52 partners being only a hair to the right. the right-wing parties considerably resemble the respective setting in figure 1. the movement of ml from one side of the y-axis to the other was the largest ppg movement during these three governments of the second parliamentary era. all in all the division into socialist and non-socialist parties was very clear as figure 1 shows. figure 1. government and opposition voting during the virolainen government (1964–1966) notes: party abbreviations not in text: tpsl = social democratic league, vm = liberal league, kp = liberal people’s party. source: nyholm (1972a) and own computations. moving to the third parliamentary era figures 2a-2f shows the political landscapes of the aho (1991-1994), lipponen i (1995-1998), lipponen ii (1999-2002), vanhanen i (2003-2006) and vanhanen ii / kiviniemi (2007-2010) governments and the first two years of the katainen (2011-2012) governments. the government majority (gov) is always located as the rightmost vector. compared with laakso’s (1972a, 1972b) result the most obvious difference is the closeness among the government groups: with respect to every government in the third era the coalition partners virtually show a unitary voting behaviour regardless of the composition of the government. the only old cabinet which resembles the modern ones is at some extent paasio’s centre-left coalition. nowadays the vectors of the coalition partners are hardly distinguishable and opposition groups are very clearly apart from the cabinets. compared with the previous decades another important difference is that the left-right dimension cannot be seen any more. the government is a separate tight cluster and the opposition parties differ from the cabinet in a more or less random order. in figures 2a-2f, the y-axis (not drawn in the figures) for which the cooperation measure is zero can be interpreted as a divider: groups to the right of the positive y-axis are more likely to vote with the government than against it and on the left side of the y-axis the groups are more likely to oppose the government than to vote with it. the ppgs seem to form three clusters: one is the cabinet on the right. second is the moderate opposition. a good example is kd, vihr and kok in figure 2d, or kd and sdp in figure 2e. it appears every government has at least one opposition party which is more likely to vote with it than against it. examples are kok in figure 2d and kesk in figure 2f. the third cluster are the less moderate or hard opposition. examples are ps and vas in figures 2d-2e. the hardest opposition is also the most recent as is shown in figure 2f (vr and ps). in fact, the opposition has become harder over time in figures 2d-2f. only if we set the resigned kd (resignation took place almost at the end of the 1991-1994 electoral period) aside in figure 2a the political landscape during the aho government is perhaps the most polarised when compared with the other times. the christian democrats resigned from the cabinet in 1994 and the greens of pajala ● government vs opposition voting in the finnish parliament eduskunta 53 vihr in 2002. the effects can be seen in figures 2a and 2c. after the resignations both parties deviate significantly from the cabinet, but only became moderate opposition parties. during every government, the opposition is rather scattered lending support to the observation that the opposition is not systematically organised. figure 2a. government and opposition voting during the aho government (19911994) figure 2b. government and opposition voting during the lipponen i government (1995-1998) european journal of government and economics 2(1) 54 figure 2c. government and opposition voting during the lipponen ii government (1999–2002) figure 2d. government and opposition voting during the vanhanen i government (2003-2006) pajala ● government vs opposition voting in the finnish parliament eduskunta 55 figure 2e. government and opposition voting during the vanhanen ii and kiviniemi governments (2007-2010) figure 2f. government and opposition voting during the first two years of the katainen government (2011-2012) discussion to recap, the purpose of the paper was to measure cooperation among ppgs in plenary votes. more precisely, the idea was to measure cooperation between government and opposition groups using information obtained from the comparison of group majorities to government majorities over parliamentary votes during 19912012. by comparing these results with previous literature the aim was to map out the patterns of ppg cooperation in finland since world war ii. two main results could be found in the previous literature published in the late 1960s and early 1970s. first, plenary votes divided the government groups and the opposition groups into two separate and distinctive camps. second, the votes also showed the ppgs following the traditional left-right dimension. during the 1950s nearly half of the plenary votes were of type that pitted communists (skdl) against others (nyholm 1961, 134). the previous studies provide us with a rough picture on parliamentary dynamics until the mid-1980s. using the terminology of nousiainen (2000; 2006) the previous literature is about the second parliamentary era of finland. in comparison the above results provide us with a picture on the subsequent third era of majority parliamentarism. the most important result has continued to be the clear division into government and opposition groups as in european journal of government and economics 2(1) 56 previous times. moreover it is now the case that government groups are a bloc voting virtually always in a unitary manner. the difference with the second era governments is clearly observable. also, during the second parliamentary era the polarisation of the groups was rather high. the opposition groups were very far from the governing coalition. it seems there was less polarisation at the beginning of the third era, though since 2007 the polarisation seems to have been on the rise. a second important result was the absence of the left-right dimension with respect to the ppgs. this result is not entirely unexpected as for example the lipponen i and ii ‘rainbow’ coalitions practically included the whole dimension while kesk was in opposition. the same of course applies to katainen’s ‘six pack’ coalition in office at the time of writing. looking at the plenary votes it seems that ideological differences or variety among the political parties and hence also the ppgs have almost entirely disappeared. the former ‘ideological’ opposition has been replaced with its modern version which consists of ppgs that were left out of the coalition in the government negotiations. exacerbating the situation we could perhaps talk about a more or less ‘technical’ opposition. this, however, applies also to the governments as ideology was a strong factor in government formulation after world war ii and almost up to the 1990s. in those days, the lifespan of a government was considerably shorter in comparison with the modern ones lasting for the whole four-year electoral period. during the second parliamentary era, the parties were rather inflexible ideologically and politically causing frequent government crisis. since the 1990s basically any party combination including two out of the three (or four) largest parties has proven to be able to work. ideological issues have not been the main obstacles. the best examples of this are the oversized coalitions of lipponen and katainen. the rank order prediction of opposition ppgs was rather successful. the largest ppgs (kesk, kok, sdp), while in opposition, were found in the modest opposition and sometimes voted more likely with the government than against it. an exception to this was ps in the current electoral period. after the 2011 elections the party became the third largest ppg, however, adopted a completely different voting pattern being almost in extreme opposition. this is a new feature in finnish politics. as a small party, ps was predicted to be in the hardest opposition together with vas, which appeared to be the case. finally, other small parties were found to be farther away from the government as the above three large ones, as expected. against the somewhat intuitive prediction, the opposition parties moved closer to governments during the last year of an electoral period. details of this rather interesting observation are left for further research. what was not visible any more in the latest parliamentary era was the left-right dimension which was still so obvious in the preceding times. it is of course the case that above i have specifically analysed the government-opposition dimension. however, according to the scaling results in pajala (2012) this dimension has the most explanatory power with respect to plenary voting. higher dimensions were hardly interpretable and at the best showed only weak traces of a possible left-right dimension. it is of course possible to order the parties from left to right and several data sources provide estimations for such positions. still, the difficulty is that only vas is unanimously seen as the leftmost party, but for the rest there is no consensus. a further thing is that government coalitions are not comprised of ideologically adjacent parties in the left-right sense. this is probably the main reason why the left-right dimension is nowadays virtually non-existent regarding plenary voting. as a concluding statement, this study has shown that (with the possible exception of the most recent electoral term) eduskunta is not at all the divided and extremely polarised public arena that one might suspect by following the arguments in plenary popularised by the mass media. pajala ● government vs opposition voting in the finnish parliament eduskunta 57 references arter, david (2006) democracy in scandinavia: consensual, majoritarian or mixed? manchester: manchester university press. benoit, kenneth and laver, michael (2006) party policy in modern democracies. london: routledge. christiansen, flemming and damgaard, erik (2008) ‘parliamentary opposition under minority parliamentarism: scandinavia’, journal of legislative studies 14(12): 46-76. helander, voitto (1990) ‘lepäämäänjättämismekanismi: tutkimus vj 66.7§:n käytöstä itsenäisyyden aikana’, studies in political science 45: university of turku. helms, ludger (2008) ‘studying parliamentary opposition in old and new democracies: issues and perspectives’, journal of legislative studies 14(1-2): 619. jensen, torben (2000) ‘party cohesion’, in esaiasson, peter and knut heidar (eds.), beyond westminster and congress: the nordic experience. columbus: ohio state university press. jyränki, antero (2006) ’kansanedustuslaitos ja valtiosääntö 1906-2005’, in suomen eduskunta 100 vuotta series, part 2 eduskunnan muuttuva asema. helsinki: edita, pp. 1-179. klingemann, hans-dieter, andrea volkens, judith bara and ian budge (eds.). (2006) mapping policy preferences ii: estimates for parties, electors and governments in central and eastern europe, european union and oecd 19902003. oxford: oxford university press. laakso, markku (1972a) ‘eräitä eduskunnan päätöksenteon rakenteellisia piirteitä’. research report c: 21, department of political science, university of turku. laakso, markku (1972b) ‘william h. rikerin poliittisesta koalitioteoriasta: peliteoreettinen tarkastelu sekä empiirinen sovellutus eduskunnan äänestyskäyttäytymiseen vuosina 1964, 1965 ja 1966’. research report n 24/1972, department of political science, university of helsinki. nousiainen, jaakko (2006) ’suomalainen parlamentarismi’ in series suomen eduskunta 100 vuotta, part 2 eduskunnan muuttuva asema. helsinki: edita, pp. 180-335. nousiainen, jaakko (2000) ’suomalaisen parlamentarismin kolmas kehitysvaihe: konsensuaalinen enemmistöhallinta, vireytyvä eduskunta’, politiikka 42(2): 83-96. nousiainen, jaakko (1998) suomen poliittinen järjestelmä, 10th updated edition. porvoo: wsoy. nyholm, pekka (1972) parliament, government and multi-dimensional party relations in finland. commentationes scientarum socialium 2. fennica: helsinki. nyholm, pekka. (1969) ’eduskuntaryhmien kannanottojen keskinäisestä vertaamisesta’, politiikka 11(1): 6-15. nyholm, pekka (1961) suomen eduskuntaryhmien koheesio vuosien 1948-51 vaalikaudella ja vuoden 1954 valtiopäivillä. helsinki: societas scientiarum fennica. nyholm, pekka and hagfors, carl (1968) ryhmäyhtenäisyyden kehityksestä eduskunnassa 1930-1954. research report no. 14/1968, department of political science, university of helsinki. ollila, anne .(2007) ’ammattina kansan edustaminen’, in suomen eduskunta 100 vuotta series, part 5 kansanedustajan työ ja arki. helsinki: edita, pp. 10-173. european journal of government and economics 2(1) 58 pajala, antti (2013) ’plenary votes in the finnish parliament during 2011-2012: electronic vote data’, mimeo, university of turku. pajala, antti (2012) ’eduskunnan ristiriitaulottuvuudet täysistuntoäänestysten valossa vuosien 1991-2010 valtiopäivillä’, politiikka 54(2): 103-118. pajala, antti (2011) ’hallitusja oppositioryhmien suhde eduskunnassa ii maailmansodan jälkeisellä ajalla’, politiikka 53(3): 214-226. pajala, antti (2010) ‘plenary votes and voting cohesion in eduskunta: a descriptive analysis of the 1991-2006 parliamentary years’, acta politica aboensia, b2. pajala, antti (2006) ’eduskunnan täysistuntoäänestykset vuosien 1945 2005 valtiopäivillä’, politiikka 48(1): 64-75. pajala, antti and jakulin, aleks (2012) ‘plenary votes in the finnish parliament during years 1991-2010: electronic vote data’. finnish social science data archive fsd, data #2117. pajala, antti. and jakulin, aleks (2007) ’eduskuntaryhmien äänestyskoheesio vuosien 1991-2006 valtiopäivillä’, politiikka 49(3): 141-154. paloheimo, heikki (2008) ’ideologiat ja ristiriitaulottuvuudet’, in paloheimo, heikki and raunio, tapio (eds.) suomen puolueet ja puoluejärjestelmä. helsinki: wsoy, pp. 27-59. paloheimo, heikki and raunio, tapio (2008) ’puolueiden rooli ja tehtävät demokratiassa’, in paloheimo, heikki and raunio, tapio (eds.) suomen puolueet ja puoluejärjestelmä. helsinki: wsoy, pp. 11-25. paloheimo, heikki and wiberg, matti (1996) politiikan perusteet. porvoo: wsoy. poole, keith t. (2005) spatial models of parliamentary voting. cambridge: cambridge university press. poole, keith t. (2000) ‘non-parametric unfolding of binary choice data’, political analysis 8(3): 211-32. poole, keith and rosenthal, howard (1997) congress: a political-economic history of roll call voting. oxford: oxford university press. rantala, onni (1982) ‘enemmistövalta vai konsensus: parlamentaarinen oppositio isossa-britanniassa, ruotsissa ja suomessa’, studies in political science #41: university of turku. rice, stuart (1928) quantitative methods in politics. new york: knopf. sartori, giovanni (1971) ‘opposition and control: problems and prospects’, in barker, rodney (ed) studies in opposition. london and basingstoke: macmillan, pp. 31-38. sieberer, ulrich (2006) ‘party unity in parliamentary democracies: a comparative analysis’, journal of legislative studies 12: 150-78. sänkiaho, risto (1969) ‘eduskunnan äänestysdimensioista vaalikaudella 1962-66’, research report no. 17/1969, department of political science, university of helsinki. wiberg, matti (2006) politiikka suomessa. porvoo: wsoy. european journal of government and economics volume 2, number 1 (june 2013) issn: 2254-7088 59 ceausescu’s population policy: a moral or an economic choice between compulsory and voluntary incentivised motherhood? florin s. soare, university of bucharest and institute for the investigation of communist crimes and the memory of the romanian exile (iiccmer) abstract the purpose of this paper is to explain why, in 1966, the romanian leadership adopted a wholly restrictive pronatalist policy, based on the strict limitation of abortion, instead of one based on socioeconomic incentives to families, as suggested by technocrats. previous literature shows disagreement on whether the choice was motivated by moralistic or economic considerations. in order to find an answer to this question, hundreds of pages of archival material unpublished so far have been analysed, including the minutes of the central committee of the communist party, statistics, documents identified in the ministry of health archive, and the technical reports that were on the table at the time of the decision. the conclusion of this study, drawn on the basis of these documents, indicates that at the time of 1966, regardless of the suggestions of the technocrats, a decision had already been taken by ceausescu himself. this decision was influenced directly by economic considerations, namely the wish to obtain the maximum pronatalist effect at a minimum budgetary cost. jel classification j10; j13; j18 keywords pronatalist policy; abortion; romanian communist party; birth rate microsoft word ejge_05_2016-009 european journal of government and economics volume 5, number 1 (june 2016) issn: 2254-7088 29 the occupy and indignados movement and the importance of political context: differences between occasionals and regulars in spain and the uk cristina gomez-roman, universidade de santiago de compostela, spain jose manuel sabucedo, universidade de santiago de compostela, spain abstract in this work we look into the individuals’ reasons that led occupy protesters to participate in this movement. we interviewed occupy participants in the uk and spain. we wanted to find out if the reasons that lead them to participate are the same in both countries. the context of the country where the demonstration was held was taken into account, as well as the differences there may be in the reasons for participation if we consider whether they are occasional or regular participants in collective protest actions. using a two-by-two design defined by country of demonstration and the history of mobilisation (occasional vs. regular), our results show important differences in both factors. the study contributes to the literature, highlighting the importance of analysing the context where the demonstration takes place and which motives must be underlined in order to attract participants to protests. keywords political context; participation motives; occupy; spain; uk. jel classification d74; d83; e65. gomez-roman ● occupy and indignados movement 30 in may 2011, thousands of people throughout spain, outraged by socio-economic prospects and the lack of political responsibility, organised marches and protests in 58 different cities (feixa, 2013). beginning in madrid and spreading quickly throughout the country, protesters took over squares and set up encampments, some lasting for weeks. in these camps, participants debated and managed the occupation through general assemblies and various issue-specific working groups. on 15 october 2011, a similar scene played out next to st paul's cathedral in london, where a group of activists met and began an occupation, which lasted until 14 june 2012. inspired by the camps in spain, participants debated and managed the occupation through general assemblies and smaller commissions, streaming their protests live on social media. participants of both countries have been referred to as the indignados and occupy movement, respectively. these movements do not arise in a vacuum; their appearance was inspired by the global protest wave that started in tunisia, later spread across north africa and the middle east, in what was called the arab spring (costanza-chock, 2012; gledhill, 2012; tejerina, perugorría, benski, & langman, 2013). these protests were also related to the political-economic crisis taking place in europe (halvorsen, 2012), and as a result of dissatisfaction with the global crisis and the inefficient decisions by governments in responding to it. they rejected “politics as usual,” demanding a more equal and transparent financial/economic system (democracia real ya, 2014; occupy london, 2014). these movements were the beginning of a wave of protests that took place in 2011-2012 resulting in what is known as a new cycle of protest (tarrow, 1991). these movements were able to expand a discourse fitting with the general feeling of indignation. they were not claiming for specific issues (for instance: rejecting labour reform, demonstrating against abortion or climate change), but demanding a radical change in applying democracy, a radical change of the political system itself (van stekelenburg, 2012; langman, 2013). in addition, the spanish and british examples had a contagion effect on other countries: the usa, turkey, the netherlands, among others, all subsequently started their own similar movements. the squares in several international cities were also similarly occupied, according to van stekelenburg (2012: 224): ‘in february 2012, there were squares occupied in 1590 cities worldwide.’ this movement was a convergence of tactics and ideas, not novel on their own, but which, when combined, were able to energise activism that hadn’t been seen for several years, attracting not only media attention but the interest of many academics worldwide (pickerill & krinsky, 2012). a similar movement occurring in several countries provides a unique opportunity for academic study. previous research in the occupy/indignados movement has studied this protest related to its particular organisational structure (manilov, 2013), or has executed single case studies in different regions of occupy demonstrations (monographic in social movement studies, 2012; pickerill & krinsky, 2012; howard & pratt-boyden, 2013; halvorsen, 2015). however, these studies didn’t make comparisons in order to establish if participants in the protest have the same motives to participate. the analysis of political protests, and more specifically the occupy/indignados movement, cannot be carried out in isolation, studying one single action. as klandermans and smith (2002: 6) pointed out, ‘comparative research of movement participation is important. it tells us that what holds for a participant in one movement, or at one point in time, or in one place is not necessarily true for a participant in another movement, or at a different time or place’. in this case, comparing the same movement in two countries, we can examine how variations result from differences in the context in which the protest takes place, and how they interact with these contexts, increasing our knowledge in the general dynamics of protests. european journal of government and economics 5(1) 31 in this work we are interested in establishing the individual reasons leading protesters to participate in the actions of the indignados and occupy movement: what has led them to take part in the protests? in order to do so, we have interviewed occupy participants in the uk and spain. we wanted to find out if the reasons that lead them to participate are the same in both countries and if there are differences among them. we take into account the context of the country where the demonstration was held, and also the differences there may be in the reasons to participate, differentiating whether they are occasional or regular participants in collective protest actions. in line with recent work (verlhust & walgrave, 2009; saunders, grasso, olcese, rainsford, & rootes, 2012; rüdig & karyotis, 2013), we understand that even within the same demonstration different profiles of protester may exist: those who have been taking part in protest all their lives (regulars) and those who, due to crisis circumstances and the appearance of a new movement (far from standard political parties or trade unions), are occasionals (those who don’t have a long tradition of participation). context that matters: country where the demonstration takes place. although the assumption that the increase in the number of people who claim to have taken part in protests in recent decades has not been without controversy (biggs, 2014), the fact is in recent years there has been objective data showing a considerable increase in the number of demonstrations (ess, in jiménez, 2011; thomas & louis, 2013). this increase in the number of protests has probably been due to the effect of the economic crisis and citizens’ disenchantment with the policies undertaken to respond to it. dissatisfaction with democracy and poor trust in governments and politicians appears to be important in the participation of citizens (gómez-román & sabucedo, 2014). our approach assumes that in protest dynamics we need to consider the political context, that is, the perception and interpretation people hold on the political environment where they are embedded. we need to take into consideration these perceptions of the political context to understand the differences between participants. depending on these interpretations, they will activate different motivations encouraging individual participation. the context will have a bearing on the profile of the participant and on motivations for participation. social movements have a crucial role in creating discourses fitting with these perceptions. they must be able to frame the situation in an attractive way so as to bring as many people possible to demonstrations. one of the first attempts to study the complex relationship between protest and context was a longitudinal study of four centuries of action repertoires in france, by charles tilly (1986), who demonstrated that the repertoires of action changed according to an evolving context. another more recent attempt to study this relationship between protest and context was a comparative study by walgrave and rucht (2010). they formulated the term ‘issue-specific context’ to refer to the interactions between national context and the issue(s) of the demonstration. the authors’ evidence stems from the demonstrations against the war in iraq. in their study, the issue of the demonstration was the same in each of the countries but the national context differed considerably. the authors showed that the composition of the crowd varied depending on this contextual variation. in a more recent work, klandermans, van stekelenburg, damen, van leeuwen and van troost (2014) observed that at least the actors on stage may vary considerably as a result of the contextual setting in which they are staged. they found that the national context within which these demonstrations are taking place added information to the scene; furthermore they observed consistent differences due to this contextual variation. gomez-roman ● occupy and indignados movement 32 in this work, we compare two occupy demonstrations in two countries of europe: spain and the uk. these countries are similar because the crisis hit both hard and the two have traditionally low trust in government. the uk and spain have become more indebted as a result of the great bailout and according to mcnally (2010) public debt in these countries is now above 60 per cent. both countries have had to take austerity measures to confront this difficult situation. for instance, in spain, among other austerity measures, the socialist government included an increase in higher rate income tax and 8 per cent spending cuts, public sector workers have had their pay cut by 5 per cent and salaries frozen, and the retirement age was raised to 67. in the same line, in the uk, the conservative-liberal democrat coalition government took the biggest cuts in state spending since the second world war, as well as increasing the retirement age to 66, cutting 490,000 public sector jobs, and making budget cuts of 19 per cent on average up to 25 per cent in most government departments (farnsworth & irving, 2012).the situation in spain was clearly worse at that time: the level of unemployment was at that time 22.85 per cent, twice the average of the european union (ine, 2012; eurostat, 2012); while the uk’s unemployment was the worst in the country since 1994, it was only around 2.62 million (8.2 per cent). the spanish economy was also in the european union’s spotlight, as they decided on the necessity of a bailout plan. alongside the economic crisis in spain, there was also a political crisis taking place. politicians were identified as the third most important problem in spain, following unemployment and the economic crisis (anduiza, cristancho, & sabucedo, 2013). as for the influence of the adversity of the national setting, we presume that in spain, the perceptions of the political context underlying movement claims would influence motivations to participate in occupy demonstrations. different participants at the same demonstration: occasionals and regulars according to the european social survey, the number of people claiming to have participated in a demonstration in the past 12 months has increased between 2008 and 2012 (ess, 2012). this increase in the number of protests and the people who claim to have participated in it, are indicators that during the crisis period a large number of people who have not previously participated regularly may have been mobilised. the unfavourable political and economic environment and the appearance of a movement far from usual mobilisation agents (trade unions or political parties) encouraged the participation of citizens who were previously outside of politics. therefore, in this particular political and economic context, it is very important to study those who are participating not only in general terms, but to find out if their motives are the same for the occasionals (or sporadic participants) and the regulars (those who have been participating regularly). we strongly believe those who only participate occasionally in these times of crisis were not in the demonstration for the same reasons as the usual protesters. we understand that the economic and political context of hardship and the emergence of a movement like occupy have made the protest more attractive to those citizens who are not regular activists. while it is true that in recent years there has been a large number of empirical studies explaining the general causes of protest behaviour (van zomeren, postmes & spears, 2008; dalton, van sickle, & weldon, 2009; van stekelenburg & klandermans, 2010), the fact is there are still few works dealing with the existence of multiple profiles of protester within the same protest. in one of these studies, saunders et al. (2012), the authors classify four different profiles depending on the frequency and intensity of their participation: novices, returners, repeaters and stalwarts. verlhust and walgrave (2009) also studied the differences between firsttimers and stalwarts, testing a series of hypotheses to explain the differences between them. in a more recent work, rüdig and karyotis (2013) found new european journal of government and economics 5(1) 33 participants to be more similar to the general population than they are to established protesters. taking into account the findings of the three studies, we can conclude, in fact, there are significant differences among participants, especially considering their history of participation in protest. regulars participate more in various social and political organisations. they are ideologically situated on the left and they have a worse evaluation of how the system works. they have a more politically defined identity. in this paper, we use an excellent framework not only to establish if there are differences between occasionals and regulars, but moreover whether those profiles are closer when we consider the country of the protest. that is, are the occupy occasionals from both countries more similar to each other? or are protest participants, regardless of frequency of participation, more alike depending on the country in which the protest takes place? is it the context of the country where the event takes place that creates similarities between participants and their reasons for participating, or is it the fact that they are occasionals or regulars? explanations of collective action as mentioned above, how people interpret the political context in which they are immersed can be decisive in overcoming the barriers of participation and can also define the individual reasons why someone participates. classic studies of social psychology on the influence of context are very clear examples of how human behaviour is a function of the environment (zimbardo, 2007; lewin, 1936). how people perceive and interpret their environment influences their behaviour. if the interpretation of that context is so important, then we must consider these interpretations in order to better understand what has led people to participate in a protest as unique as occupy. we shall describe those socio-political and psychological variables that we believe can help us understand how participants interpreted this political context and how these perceptions are crucial to their participation. in what follows we provide explanations for collective action, running from perceptions of the political context (trust in institutions, satisfaction with democracy, political orientation and political values) to psychological motives (anger, efficacy and identity). we relate those variables to the country where the demonstration takes place as well as mobilisation history, thereafter formulating hypotheses to guide our analyses. the lack of trust in institutions and poor satisfaction with democracy can feed the intention to participate in protest actions (anderson & mendes, 2006; dalton et al., 2009). since the system does not respond to the demands of their people, they choose alternatives to change the state of things. therefore, in this study we asked participants in occupy demonstrations for their feelings of trust and satisfaction with the system. we understand these levels will be low, but they will be significantly lower among spanish participants (h1a), due to the worse economic circumstances of that country. furthermore, we understand these levels will be very poor among regular protesters, mainly due to the disappointment created by a system they have been fighting against repeatedly (gómez-román & sabucedo, 2014). occasionals, though disgusted with the system itself, are not as much so as regulars, since they are only sporadically repudiate the mentioned system (h1b). another variable to help us better understand how people interpret this political context is political orientation. how a person identifies on the scale of ideological left/right position is fundamental to interpretations of the environment around them. the political orientation of citizens provides a general framework for understanding context. it can even be understood as a heuristic from which people turn to explain many of the phenomena occurring in the environment. this political orientation provides a clear guide to understanding certain political issues and what position should be taken on these issues (hooghe & keern, 2013). having defined a political orientation, individuals create particular political attitudes (nie, verba, & gomez-roman ● occupy and indignados movement 34 petrocik, 1979) and react accordingly. participants understand that the occupy movement, as can be understood from its discourse, should be intermediate on the left / right scale, and should not be identified with any ideology. however, per previous literature (verlhust & walgrave, 2009; saunders et al., 2012), we expect there will be differences in this respect between occasionals and regulars. it has been found protest participation is more common among those who identify with the left (dalton et al., 2009), so we expect the regulars to be located along this part of the ideological spectrum (h1c). as commented before, individuals behave based on how they interpret reality. political values, close to ideological orientation, can also act as a guide for interpreting the environment. there are several classifications and scales for political values (inglehart, 1990; alexander, inglehart, & welzel, 2012; grasso y giugni, 2013). one such classification proposal, by heath, evans and martin (1994), differentiates between general orientations of an economic left-right scale (related to economic and political equality and government intervention versus free enterprise) on the one hand, and individual liberties or libertarian-authoritarian on the other (items related with freedom of thought, conscience or the right of association and the relationship person-political institutions). given the impaired economic situation, we believe the protesters in the spanish protest will score more highly on the scale of economic values, because they perceive a threat to their basic needs and will therefore maintain more material concerns (h1d). we also expect differences between occasionals and regulars in both countries. we understand occasionals take to the streets because their quality of life has been threatened and they will be motivated by economic values (especially in spain). regular activists, however, are demanding not only an economic change, but also a modification of the system’s functioning. this position is more relevant to individual liberties (h1e). once we know that the interpretation of political context by participants may impact their reasons for engaging in collective action, we would like to expound the psychological theories that might explain participation in demonstrations. in social psychology, there is some consensus on the existence of three psychosociological explanations for collective action, those being injustice, efficacy and identity (van zomeren et. al, 2008; van stekelenburg & klandermans, 2010). these three frames, proposed by gamson (1992), are fundamental in understanding participation, serve to interpret reality, are socially constructed and culturally contextualised, and encourages participation in collective action (gamson, 1992). the first frame corresponds to the perception of injustice. when one perceives oneself as not being treated as deserved, or that an experienced situation is unfair, then one would like to do something to change it (gurr, 1970). it is important to note that the situation does not always have to be objectively unjust. it must be perceived subjectively as unfair (kelly & breinlinger, 1996). moreover, if one considers oneself to be poor, but finds an acceptable justification for one’s poverty, even if the situation actually is objectively unfair, one will not perceive it as such and will do nothing to change it. mobilised citizens understand themselves to be unfairly treated and so have decided it is time to act. this injustice frame has a cognitive component, grievances, and an emotional one, anger. according to a metaanalysis by van zomeren, et al. (2008), the component to best explain collective action is anger. accordingly, we expect participants in both countries to show high levels of anger, but because the grievance situation in spain is worse, these levels should be higher there (h2a). for occasionals and regulars, we expect, given the regulars have been participating in the past in similar actions, allowing them to more quickly access a sense of grievance (tversky and kahneman, 1981), so that is why we expect them to be angrier (h2b). another frame proposed by gamson is the efficacy frame. people would not go to the streets to protest if they did not believe they could get something out of it. european journal of government and economics 5(1) 35 protesters do cost-benefit calculation for participation (klandermans, 1984). if individuals think fatalistically, e.g., "nothing i do can change things”, or “it is not in my power to do anything", they will not participate. but when they believe that through joint effort, the group will be able to respond to relevant events, overcome obstacles and achieve collective intermediate goals, they will have sufficient reason to join the mobilisation (hornsey, et al., 2006). accordingly, participants in both countries are expected to show feelings of efficacy in some way, but because the occupy movement had already been successful in spain and wall street at the time it took hold in the uk, we expect the anticipation of success to be greater among british respondents (h2c). regarding occasionals and regulars, as the regulars have previous experience of successful participation, we expect regulars will be more optimistic than occasionals in consideration of protest as effective (h2d). the third collective action model frame that gamson suggested is the identity frame. protest participation requires a strong sense of collective identification with the group for fighting collectively against a common disadvantage (simon et al., 1998). this shared feeling can provoke the necessity to act jointly against the same problem (sabucedo, durán, & alzate, 2010). according to stürmer and simon (2004), when people identify themselves with a group and assume that the adverse condition is motivated by an intergroup power conflict, it becomes what is called politicised identity. the more people identify with a group, the more they are inclined to protest on behalf of that group (reicher, 1996). according to this framework, the sense of identification should be present among the participants of spain and the uk, although in the former, given the political situation is worse and as they may feel more aggrieved by the out-group, they potentially would be showing more identification (h2e). in the case of mobilisation history, being a regular implies having been in previous situations where they shared experiences with other participants, so this identification should be stronger among regulars, although certainly not absent in occasionals (h2f). if we base our ideas on the precepts above, the principal objectives of the investigation are as follows. first, we shall examine whether there are differences between the different motivations for participation in occupy participants in spain and the uk. it is expected that the motives will differ in intensity between both countries. second, we shall examine whether there are differences in the motives for participation between occasionals and regulars. it is expected that regulars will score more highly in motives determining participation in collective action. third, we shall try to ascertain whether the interaction between the location and mobilisation history produces changes in the motivations for these protesters to participate. it is expected that there will be stronger motives for participation among regulars, especially in spain, and that their perceptions of the political context will have been affected, given that they are experiencing a worse economic scenario. method participants and procedure our data on participants in protests self-styled by the movement as occupy were collected in spain and the uk in 2011, following the methodology suggested by the collaborative european research project caught in the act of protest: contextualising contestation. for further information we recommend van stekelenburg, walgrave, klandermans, & verlhust (2012). respondents completed questionnaires―so-called protest-surveys―distributed during the demonstration (n=500-1000)―to be returned to the university using prestamped envelopes. the response rate for the demonstrations fluctuated between 25 and 35 per cent. all questionnaires and procedures are standardised. in order to control for response biases we also conducted short (2-3 minutes) interviews gomez-roman ● occupy and indignados movement 36 with a subsample of the respondents (n=100-200) comprising questions identical to those in the printed questionnaire. by comparing the face-to-face interviewees response to the identical questions in the returned questionnaires we can estimate the response bias. on average, somewhat older and more highly educated demonstrators were more likely to return the questionnaire. this non-response did not result in biased findings. as for the sampling of participants, we designed a sampling strategy such that each participant had the same likelihood of being selected. demonstrations were covered by a team consisting of 3-4 pointers, and 12-15 interviewers. each pointer had a team of 4-5 interviewers. the pointers selected the interviewees, while interviewers conducted the interviews and handed out questionnaires. separating these two roles appeared to be crucial in preventing sampling biases (walgrave & verhulst, 2011). as interviewers tend to select people they believe to be willing to cooperate, they often end up producing biased samples. interview teams started at different points of the procession and worked towards each other, approaching every n-th person in every n-th row. the result was samples we believe to be representative of the demonstrators present. our final sample consists of 175 participants: 122 in spain (55.8 per cent male with a mean age of 40, sd=14.23); and 53 in the uk (61.5 per cent male with a mean age of 40, sd=15.31). there were no differences among participants in terms of socio-economic class or educational level. design in this study we propose a 2x2 factorial design (country where the protest was held: spain vs. the uk; and participant profile: occasional vs. regulars). instruments the participants responded to a questionnaire, which included the following variables: independent variables place of demonstration: the uk or spain occasional vs. regular: we separated participants by taking into account their frequency of participation. they were to answer the following questions: “how many times have you taken part in a demonstration in the past? ever?” and “in the past twelve months?” there were 5 possible answers: “never”, “1 to 5”, “6 to 10”, “11 to 20” and “more than 20 times”. occasional are those ones who answered 1-5 in the item “ever” and never or 1-5 in “the past 12 months”. regulars are those who answered more than 21 “ever” and more than 6 “in the past 12 months”. dependent variables most questions were set out on a likert scale with 5 possible answers ranging from 1 (not at all) to 5 (very much). those cases where a different measure was used are indicated. anger. this was measured with the question, “to what extent do you feel anger when you consider the current economic crisis?” efficacy. this was measured from the following statements: “organized groups of citizens can have a lot of impact on public policies in this country” and “if citizens from different countries join forces, they can have a lot of impact on international politics.” (α = .71) collective identity. this was measured with the following questions: “to what extent do you identify with the other people present at the demonstration?”, “to european journal of government and economics 5(1) 37 what extent do you identify with any organisation staging the demonstration?” (α = .58) satisfaction with democracy. the participants had to answer the question, “in general, how satisfied or dissatisfied are you with the functioning of democracy in your country?” where 0 was not at all satisfied and 10 was very satisfied. political trust. participants were given a list of institutions (national government, national parliament, political parties, trade unions, judicial system and european union) and were asked to indicate how much trust they had in them. (α = .81) left/right placement: the participants had to answer the question, “in politics people sometimes talk of ‘left’ and ‘right’. where would you place yourself on this scale, where 0 means the left and 10 means the right?” political values: we asked participants to what extent they agree or disagree with the following statements; for the economic left-right values we used the following: “government should redistribute income from the better off to those who are less well off”, “even the most important public services and industries are best left to private enterprise” (reversed coded). (α =.54). authoritarian/libertarian values were measured with the items: “children should be taught to obey authority” (reverse coded), and “people from other countries should be allowed to come to my country and live here permanently if they want to.” (α =.53). results first, we present data showing the distribution of occasionals and regulars in both countries (table i). second, we present the manova with country of demonstration (spain vs. the uk) and mobilisation history (occasional vs. regulars) as fixed factors; and anger, efficacy, identity, and perception of political context as dependent variables (table ii). table i. distribution of respondents according to their mobilisation history and country as expected, occupy demonstrations attract more occasionals (54.9 per cent) and fewer regulars (45.1 per cent), in both countries. these data are relevant because they show how new organisations can serve to channel discontent in times of economic crisis and political disaffection, mobilising a great number of sporadic participants. the high number of regular protesters reported in the spanish case compared to the british data must be also noted (47.5 vs. 39.6 per cent). this is related to the high number of protests taking place in spain every year (jiménez, 2011) and the high number of people who acknowledge to participating frequently in demonstrations (ess, 2012), confirming once again that protest is one of the most representative forms of collective action in spain. although we do not have large numbers and the sample is unbalanced, we must highlight the relevance of these data, especially in spain’s occupy, the first demonstration of the movement, and as such ours are unique data. spain uk total occasional 64 (52.5%) 32 (60.4%) 96 (54.9%) regular 58(47.5%) 21 (39.6%) 79 (45.1%) total 122 53 175 gomez-roman ● occupy and indignados movement 38 table ii. manova analysis comparing country and mobilisation history factor dvs factors m sd f sig η2 country trust in institutions spain 1.92 0.53 11.26(1, 175) .001*** .06 uk 2.31 0.75 libertarian values spain 3.57 0.78 4.19(1, 175) .04* .02 uk 3.77 0.99 occasionals vs. regulars anger occasionals 4.29 0.77 5.06(1, 175) .03* .03 regulars 4.46 0.87 identity occasionals 3.76 0.72 14.59 (1, 175) .001*** .08 regulars 4.06 0.66 efficacy occasionals 3.99 0.75 9.59(1, 175) .002** .05 regulars 4.26 0.69 trust in institutions occasionals 2.12 0.66 7.30(1, 175) .008** .04 regulars 1.94 0.59 satisfaction with democracy occasionals 2.01 1.03 6.79(1, 175) .01** .0.4 regulars 1.70 0.88 left/right occasionals 2.96 1.67 53.55(1, 175) .001*** .24 regulars 1.02 1.25 lefteconomic values occasionals 4.08 0.76 45.14 (1, 175) .001*** .21 regulars 4.79 0.41 libertarian values occasionals 3.29 0.82 37.95(1, 175) .001*** .18 regulars 4.06 0.69 country x mobilisation history anger spain occasion als 4.30 0.66 4.09(1, 175) .04* .02 regulars 4.33 0.96 uk occasion als 4.28 0.96 regulars 4.86 0.36 identity spain occasion als 3.89 0.62 9.52(1, 175) .002** .05 regulars 3.97 0.63 uk occasion als 3.5 0.85 regulars 4.28 0.70 trust in institutions spain occasion als 1.91 .051 8.08(1, 175) .005** .04 regulars 1.92 0.56 uk occasion als 2.53 0.73 regulars 1.98 0.66 *p<.05 **p<.01 ***p<.001 per the results observed in the manova analysis (table ii), in the case of the variable “country”, there are significant differences between spain and the uk in trust in institutions (f [1, 175] = 11.26; p = .001) and libertarian values (f [1, 175] = 4.19; p = .04), where the results are higher in the uk, partially confirming hypothesis h1a. with regard to the variable “mobilisation history” we can see in table ii that there are significant differences among occasional and regulars in all the variables under study. thus, as expected, regulars of both countries are angrier (f [1, 175]= 10.69 p = .001) (h2b), identified more with the organisation and participants (f [1, 175]= 17.41; p = .001) (h2f), and perceived themselves as efficacious (f [1, 175]= 10.05; p european journal of government and economics 5(1) 39 = .002) (h2d), however trust less in institutions (f [1, 175] = 7.30; p = .008) and are less satisfied with democracy (f [1, 175] = 6.79; p = .01) (h1b). they are located further to the left on the ideological spectrum (f [1, 175] = 53.55; p = .001) (h1c) and show higher scores in libertarian (f [1, 175] = 37.94; p = .001) and left-economic values (f [1, 175] = 45.14; p = .001) (h1e). we can see that interaction between both independent variables exists and the results are significant for the dependent variables: anger (f [1, 175] = 4.09; p = .04), identity (f [1, 175] = 9.52; p = .002), and trust (f [1, 175] = 8.08; p = .005). figures 1 to 3 show those dependent variables in which there is significant interaction between country and mobilisation history. figure 1. anger differences between regulars and occasionals are very clear in terms of anger in the uk, where regulars are angrier. even more, regulars in the uk are the angriest of the four groups. contrarily, in spain, regulars and occasionals share similar levels of anger, very high in both cases, but lower than average british respondents. gomez-roman ● occupy and indignados movement 40 figure 2. identity as predicted, levels of identity are high in both countries and both profiles, but contrary to the expected, identity is higher among english regulars. similarities between participants are very interesting in the case of spain, where both occasionals and regulars are similarly identified with participants and the organisation. figure 3. trust in institutions european journal of government and economics 5(1) 41 in figure 3, we can see again the curious effect among spanish participants, where differences among regulars and occasionals are minimal. british occasionals are the most trustful but the four groups score quite low. in short, we could extract a general idea from figures 1 to 3, which is the similarity among spanish participants in the occupy protest. contrary to what is seen in the uk, where the disparities between occasionals and regulars are notable, in spain, for both occasional and regular, scores on the variables studied are very similar. apart from these significant differences, the groups show important similarities in key variables for political mobilisation. this shows that these variables are important in the context of political protest. having said that, it is also worth noting that the trend of the results in most cases was in line with our hypotheses. thus, for instance, regulars were angrier and, more than occasionals, perceived themselves as efficacious. they identified the strongest with the organisation and other participants and hold stronger political values. the results for trust in institutions are notable, where the worse economic situation in spain makes regulars and occasionals show similarly low results. discussion in this study, we analysed the motivations of demonstrators in spain and the uk to participate in occupy mobilisation. the objective was to compare and examine if the country where the demonstration takes place and the mobilisation history (being an occasional or a regular), changed the motives for participation. first, we shall discuss what happens to motives when the country is taken into account. as in the case of walgrave and rucht’s work (2010), those motives determining participation varied depending on this contextual variation. the results indicate that there are significant differences between spain and the uk, specifically in two of the variables under study: trust in institutions and libertarian values. as predicted in our hypothesis, spain’s worse political and economic context evokes differences in how this context is perceived. this is critical to determine participation. thus, the spanish protesters displayed very poor results in trust, while it is true that in the uk the results were also very low. as for libertarian values, and quite related to the relationship of person to institution, it is once again the british demonstrators who score more highly. the interpretation of the crisis situation seems to be what motivates participants in spain. the disenchantment with austerity policies and the many cases of corruption have taken their toll on how citizens perceive government institutions and its relationship with them. the indignados movement in spain has been critical, as it has brought together discontent and political disaffection, challenging the dominant discourse (moscovici, 1979), creating new frames about reality and introducing new issues in the public agenda. this has allowed them to attract occasional participants to their cause, which explains the high number of participants with this profile in occupy demonstrations. second, we would like to mention the results obtained taking into account the mobilisation history. as can be seen in the data, we found significant differences among occasionals and regulars. as expected, regulars are angrier, identify more with participants and the organisation and they consider themselves more efficacious. they are more disappointed with institutions and democracy, scoring quite low in trust and satisfaction. despite the occupy claim, “we are not on the left or right” (democracia real ya, 2014), regular participants, as documented with activists in other studies on protest (dalton et al., 2009) are more leftist. they score more highly in political values, both economic-left and libertarian. that is to say, those people defending justice in the distribution of wealth will be mobilised to obtain those objectives, demanding not only an economic change but that of operation of the system itself. for this reason they also show higher results in gomez-roman ● occupy and indignados movement 42 those values associated with a change in the relationship between person and political institution (alexander et al., 2012; grasso y giugni, 2013; inglehart, 1990). third, we wanted to examine whether an interaction existed between country and mobilisation history, and how this affected the motives to participate. this interaction is significant for three dependent variables: anger, identity and trust in institutions. before analysing every interaction individually, we wanted to emphasise the fact that although the indignados/occupy demonstration in spain was able to attract a wide variety of people (fundamentally due to the organisational characteristics and objectives), both occasionals and regulars have a similar emphasis in the motives to participate. this result can be explained by the novelty of the movement and its capacity to create an inclusive identity capable of attracting a varied profile of people. the indignados/occupy movement created an interpretive scheme of reality capable of collecting the political disenchantment of citizens, transforming into the outrage expressed in that mobilisation. for this reason, the levels of anger are high in both countries. as the occupy movement based its claims on public outrage, it is therefore not surprising that participants score highly in this variable. scores in the four groups were very high, always above 4, on a scale of 5 points. yet, it is also worth noting that the trend of the results regarding the history of mobilisation roughly matches our hypothesis. as such, the regulars express more anger in both countries; however, and contrary to expectations, british demonstrators scored more highly on this variable. this can be explained by the temporary difference between the two movements. the uk’s occupy started later, and british respondents were able to monitor its worldwide evolution, from initial euphoria to the subsequent apathy shown by institutions, which could explain the higher anger score among british regulars. this also could explain the differences found in identity results. when occupy emerged in spain, it was the very first time the term was mentioned. the novelty and freshness of the movement were initially very inclusive, which made both occasionals and regulars feel identified to the same extent, hence the similarity in scores of both spanish profiles. however, as the movement grew and gained international recognition, their image took a more defined drawing, which reduced the initially wide, blurred identity of the movement. when it erupted in the uk, both international media and governments had been responsible for disseminating and redefining the identity of the movement itself. and even though the scores of participants in the uk are high, both for occasional and regular, those with more experience in mobilisations seem to identify more with the occupy movement, with the british regulars being those who obtained higher results in this variable. lastly, regulars in both countries show lower values in trust in institutions. the most interesting finding, however, is with occasionals. as predicted, the worse political and economical context makes results of distrust more similar to regulars in the spanish case. so both hypotheses are true: the negative perception of context and the repeated struggle against the system itself are crucial to the interpretation of the context and the different reasons for participation. these results are consistent with classical theories of social psychology. as lewin pointed out (1936), people react to the context. socially constructed perceptions produce changes in the way people react to that context, in this case through collective action. to conclude, it is important to recognise certain limitations to our study. our methodology allowed us to study real participants at demonstrations, which forces us to adapt our method to these particular circumstances. while it is true that the occupy movement has been named so because of their camping in city squares, the data in spain were collected in the previous demonstration leading to the camp protest in puerta del sol. this can produce some bias about the commitment and participation of the surveyed protesters. in the british case we certainly know they european journal of government and economics 5(1) 43 camped, while in spain the indignados may or may not have ended up in camp when the demonstration finished. this has also produced differences in the amount of participants for the study: 122 in the spanish demonstration, and only 53 in the english protest. even following the same collecting procedure for our data, there were more people protesting at the demonstration in spain than in london, ergo more respondents to our questions, more people to approach, thus our unbalanced sample. despite this limitation, we believe the strength of these data lie in the fact that they allow us to compare real participants in collective action. however, we should be cautious with the importance we give to these results. the real value of these findings will become apparent when compared with other studies – those using both the same design models and different ones – in order to overcome aforementioned and unforeseen limitations. finally, we would like to stress here the importance of taking into account the history of mobilisation. as mentioned at the beginning of this work, cycles of protest encourage people to engage in collective action. some will remain occasionals because they will stop protesting as soon as their demand is met, while others will become activists (regular participants), because they will value their experience. moreover, we cannot forget that activists will be critical for the continuity of such movements. the results obtained in this study allow us, therefore, to conclude that the interpretation of the context where people are involved is crucial to determining their participation. those reasons also vary depending on mobilisation history. social movements’ organisations need to take this information into consideration when they create the frames for participation. they need to define which profile of participant they want to attract to emphasise certain aspects of their speech. this is crucial especially in attracting occasional participants with potential for regular activism in that movement. it is also important to highlight that the appearance of the occupy movement has substantially changed the dynamics of protest. while it is true this movement followed the dynamics of most protest groups--criticising the government and organising demonstrations against the ruling of the state--its main identity was opposition to the forces of the hegemonic right and left, and questioning the economic and political system as a whole. governments must consider how important it is to learn from past experiences and remain open to new ways of conceptualising politics. the public sphere cannot be underestimated because this form of action has changed the nature of contemporary political action. there are a variety of occupy-related collectives continuing to organise (howard & prattboyden, 2013). even in spain, new political parties have risen after these protests, so understanding their participants’ motivations can help in negotiating and reaching agreements in the new way of doing politics. as participants posit in their mottos: “from the streets to the parliament, looking for more direct democracy” (democracia real ya, 2014), governments should take into consideration the new political actor when making decisions. references alexander, amy c., ronald inglehart, and christian welzel (2012) ‘measuring effective democracy: a defense’, international political science review 33(1): 41– 62. anderson, chistopher j., and silvia m. mendes (2006) ‘learning to lose: election outcomes, democratic experience and political protest potential’, british journal of political science 36(1): 91-111. anduiza, eva, camilo cristancho, and josé m. sabucedo (2013) ‘mobilization through online social networks: the political protest of the indignados in spain’ information, communication & society, 17(6): 750-764. gomez-roman ● occupy and indignados movement 44 biggs, michael (2014) ‘has protest increased since the 1970s? how a survey question can construct a spurious trend’ the british journal of sociology 66(1): 142-161. costanza-chock, sasha (2012) ‘mic check! media cultures and the occupy movement’ social movement studies 11(3-4): 375-385. dalton, russell, alix van sickle, and steven weldon (2009) ‘the individualinstitutional nexus of protest behavior’ british journal of political science 40(1): 5173. democracia real ya! (2014). manifiesto. retrieved from http://www.democraciarealya.es/. european social survey (2012): data file edition 6.2. norwegian social science data services. retrieved from http://ess.nsd.uib.no/, in march, 2012. eurostat (2012) euro-indicators. retrieved online 2015-02-04 from http://ec.europa.eu/eurostat farnsworth, kevin, and zoë irving (2012) ‘varieties of crisis, varieties of austerity: social policy in challenging times’, journal of poverty and social justice 20(2): 133-147. feixa, c. (2013) ’the #spanishrevolution and beyond, fieldsights. hot spots, cultural anthropology online, retrieved from http://www.culanth.org/fieldsights/68-the-spanishrevolution-and-beyond on february 14th, 2013. gamson, william a. (1992) talking politics. cambridge: cambridge university press. gledhill, jim (2012) ‘collecting occupy london: public collecting institutions and social protest movements in the 21st century’ social movement studies 11(3-4): 342-348. gómez-román, cristina, and josé m. sabucedo (2014) ‘the importance of political context: motives to participate in a protest before and after the labor reform in spain’ international sociology 29(6): 546-564. grasso, maria, and marco giugni (2013) anti-austerity movements: old wine in new vessels? paper prepared for the xxvii meeting of the italian political science association (sisp), 12-14 september, florence. gurr, ted r. (1970) why men rebel. princeton: princeton university press. halvorsen, sam (2012) ‘beyond the network? occupy london and the global movement’, social movement studies, 11(3-4): 427-433. halvorsen, sam (2015) ‘taking space: movements of rupture and everyday life in occupy london’, antipode 47(2). 401-417. heath, anthony, geoffrey evans, and jean martin (1994) ‘the measurement of core beliefs and values: the development of balanced socialist/laissez faire and libertarian/authoritarian scales’, british journal of political science 24(1): 115-132. hooghe, marc, and anna keern (2013) ‘party membership and closeness and the development of trust in political institutions: an analysis of the european social survey, 2002–2010’ party politics, 21(6): 944-956. hornsey, matthew j., leda blackwood, winnifred louis, kelly fielding, ken mavor, thomas morton, anne o’brien, karl-erik paasonen, joanne smith, and katherine m. white (2006) ‘why do people engage in collective action? revisiting the role of perceived effectiveness’, journal of applied social psychology 36(7): 1701–1722. european journal of government and economics 5(1) 45 howard, neil, and keyra pratt-boyden (2013) ‘occupy london as pre-figurative political action’, development in practice 23(5-6): 729-741 inglehart, ronald (1990) culture shift in advanced industrial society. new jersey: princeton university press. instituto nacional de estadística (2011) encuesta de población activa. retrieved online 2015-02-04 from http://www.ine.es/ jiménez-sánchez, manuel (2011) la normalización de la protesta. el caso de las manifestaciones en españa (1980-2008). madrid: centro de investigaciones sociológicas. klandermans, bert (1984) ‘mobilization and participation: social-psychological expansions of resource mobilization theory’, american sociological review, 49(5): 583–600. klandermans, bert, and jackie smith (2002) survey research: a case for comparative designs, in bert klandermans, and suzanne staggenborg (eds.), methods of social movement research. minneapolis: university of minnesota press, pp. 3-31. klandermans, bert, jacquelien van stekelenburg, marie-louise van damen, anouk van leeuwen, and dunya van troost (2014). mobilization without organization: the case of unaffiliated demonstrators. european sociological review, 30(6): 702-716. kelly, caroline, and sara breinlinger (1996) the social psychology of collective action. basingstoke, england: taylor & francis. langman, lauren (2013) ‘occupy: a new social movement’, current sociology 1(4): 510-524. lewin, kurt (1936) principles of topological psychology. nueva york: mcgraw-hill manilov, marianne (2013) ‘occupy at one year: growing the roots of a movement’, the sociological quarterly 54(2): 206-213. mcnally, david (2010) global slump: the economics and politics of crisis and resistance. oakland, ca: pm press/spectre. moscovici, serge (1979) el psicoanálisis, su imagen y su público. buenos aires: huemul s.a. nie, norman h., sydney verba, and john r. petrocik (1979) the changing american voter. cambridge, ma: harvard university press. occupy london (2014) occupy london statements. retrieved from http://occupylondon.org.uk/ pickerill, jenny, and john krinsky (2012) ‘why does occupy matter?’, social movement studies 11(3-4): 279-287. reicher. steve d. (1996) ‘the battle of westminster: developing the social identity model of crowd behaviour in order to explain the initiation and development of collective conflict’, european journal of social psychology 26(1): 115-134. rüdig, wolfgang, and georgios karyotis (2013) ‘beyond the usual suspects? new participants in anti-austerity protests in greece’, mobilization: an international journal 18(3): 313-330. sabucedo, josé m., mar durán, and mónica alzate (2010) ‘identidad colectiva movilizada’, revista de psicología socia, 25(2): 189-201. saunders, clare, maria grasso cristiana olcese, emily rainsford, and christopher rootes (2012) ‘explaining differential protest participation: novices, returners, repeaters, and stalwarts’, mobilization: an international journal 17(3): 263-280. gomez-roman ● occupy and indignados movement 46 simon, bernd, stegan stürmer, michael loewy, ulrike weber, peter freytag, corinna habig, claudia kampmeier, and peter spahlinger (1998) ‘collective identification and social movement participation’, journal of personality and social psychology 74(3): 646-658. stürmer, stefan, and bernd simon (2004) ‘collective action: towards a dualpathway model, european review of social psychology 15(1): 59-99. tarrow, sidney g. (1991) struggle, politics, and reform: collective action, social movements and cycles of protest. ithaca, ny: center for international studies, cornell university. tejerina, benjamín, ignacia perugorría, tova benski, and lauren langman (2013) ‘from indignation to occupation: a new wave of global movilization’, current sociology 64(1): 377-392. thomas, emma f., and winnifred r. louis (2013) ‘doing democracy: the social psychological mobilization and consequences of collective action’, social issues and policy review 7(1); 173—200 tilly, charles (1986) the contentious french. four centuries of popular struggle. cambridge: harvard university press. tversky, amos, and daniel kahneman (1981) the framing of decisions and the psychology of choice, science 211(4481); 455-458. van stekelenburg, j., and bert klandermans (2010) ‘individuals in movements: a social psychology of contention’ in bert klandermans and conny m. roggeband (eds.), the handbook of social movements across disciplines. new york: springer, pp. 157–204. van stekelenburg, jacquelien (2012) ‘the occupy movement: product of this time’ development 55(2): 224-231. van stelekelenburg, jacquelien, stefaan walgrave, bert klandermans, and joris verlhust (2012) ‘contextualizing contestation: framework, design, and data’, mobilization: an international journal 17(3): 249-262. van zomeren, martijn, tom postmes, and russell spears (2008) ‘toward an integrative social identity model of collective action: a quantitative research synthesis of three socio-psyhological perspectives’ psychological bulletin 134(4); 353-372. verhulst, joris, and stefaan walgrave (2009) ‘the first time is the hardest? a cross-national and cross-issue comparison of first-time protest participants’, political behavior 31(3); 455-484. walgrave, stefaan, and dieter rucht (2010) the world says no to war. demonstrations against the war on angerq. minneapolis, mn: university of minnesota press. walgrave, stefaan, and joris verhulst (2011) ‘selection and response bias in protest surveys’, mobilization: an international quarterly 16(2); 203-222. zimbardo, philip g. (2007) the lucifer effect. understanding how good people turn evil. new york: random house. european journal of government and economics volume 4, number 1 (june 2015) issn: 2254-7088 inhibited privatization: a hurdle race over vested interests marcin senderski, kozminski university, poland abstract this paper recognizes vested interests as one of the primary premises that reduce the effectiveness of privatization policy, stall its momentum and produce structural problems in the long-run. both exogenous and endogenous drawbacks are cited, but the main focus is put on the dynamism of vested interests’ character, interconnectedness and evolution. policy makers have been long aware of the existence of activities rooted in vested interests including empire building behaviors, creation of sinecures or extravagant management style. hence, the fundamental effort here is put on the identification of emerging vested interests that were typically not considered by scholars. the channels through which conventional vested interests have snowballed over time are emphasized. this includes casting a closer glance at family employment, as well as at sports sponsorship arrangements, which emerge as the favorite domain of marketing activity for polish state-owned enterprises. the research of available literature is performed, along with its application to the polish case, and insightful observations concerning the anatomy of privatization-related reluctance. rough policy recommendations conclude the paper. keywords privatization; poland; vested interests; state-owned enterprises. jel classification l33; d72; d73. 46 senderski ● inhibited privatization introduction the principal goal of the paper is to perform an in-depth investigation into the character of privatization-related vested interests that hamper the pace and depth of ownership transformation. the case of poland is used as an example, both because of its familiarity to the author, and the presence of phenomena that are tackled in this paper, the pervasiveness of which may have been observed throughout recent years. vested interests were of course perceptible throughout consecutive stages of privatization in poland from its very beginning, but the evidence here is designed rather to illustrate the contemporary struggle to overcome the residual resistance. the term “vested interests” embodies various political interests expressed by individuals or groups, deriving rents from a given state-owned entity and capable of forming pressure aimed at the preservation of status quo, which is favorable for them, but which puts taxpayers at a disadvantage. another interesting frame for the notion of vested interests has been presented in krusell and rios-rull (1996) who studied political obstacles to technological innovations. they depict this phenomenon by the incumbents blocking the emergence of novelties by influencing the regulatory process. there are undoubtedly some merits behind the government’s presence in the corporate world and the intention of this article is neither to analyze its legitimacy, accuracy and effectiveness, nor to argue in favor of privatization’s inherent value. in fact, as hertog (2010) proved for the persian gulf monarchies, it would be too bold to make an identity relationship between public ownership and ineffectiveness. it is equally troublesome to infer high-level consequences of too much state in the economy, given the plethora of intermediate variables, differences in initial conditions and adopted development models. it is also ambiguous what consequences political connectedness has for companies involved in political campaigns (see newton and uysal, 2013). hence, instead of seeking strict evidence for the ineffectiveness of state-owned enterprises (later referred to as soes), this paper rather relies on a generic, primitive assumption that private property in the broad economy is more effective than public property. the paper is solely to stress that there is a number of supplementary motivations that derail soes from their shareholder value-oriented regime and make policy makers not cut the umbilical cord that feeds them with generous governmental perks. at the same time, being aware of the consequences of flawed state ownership is important as it attests to the importance of this article. first, vested interests and rent seeking behaviors can distort the economic efficiency of policy design and implementation in the domain of management of economic resources (jamasb, 2006:24). second, they might have prolonged consequences for the level of development of the private sector and financial markets (roland, 2001). third, protracted privatization may have insidious influence on public finance, leading to its sub-optimal structure and putting off necessary reforms. a certain sort of extra income dependency can be observed, meaning that the government takes yearly privatization revenue for granted, treating it as filling material for budget deficit (see e.g. iimi, 2003). henceforth, gradual and sluggish privatization postpones the need to introduce substantial, but openly disliked policy shifts. with swift privatization, government would be injected with a one-off cash boost, but in the coming years more courageous policies would have to be implemented. rosa and pérard (2010) stress that privatization-related literature is concentrated around two central subject matters: optimal privatization methods and efficiency gap between soes and private companies. particularly, the ideological component that drives decision makers is unexploited. although there is a broad literature also on the motivation of political actors to privatize or externalities associated with privatization, there is still a considerable deficit in comprehending the inhibitors to ownership transformation. naturally, the political dimension is recognized and the political theory of privatization is developed in such papers as boycko, shleifer and 47 pere ● governance and economic development of western balkan countries vishny (1996), opper (2004) or bortolotti and pinotti (2008), but the existing literature more often than not adopts a static stance towards interest groups, usually ignoring the fact their interests do evolve over time. on a statistical note, the history of privatization in poland dates back to 1981 (the introduction of market socialism policy), but it is not before the collapse of communist regime in 1989 and pushing through a series of privatization bills until the property transformation gathered its momentum (for more extended recapitulation refer e.g. to kowalik, 1991). between 1990 and 2008, nearly 7,500 (85.4 percent as of 1990) state enterprises have been put under some form of privatization. however, this figure is remarkably deceiving, as more than every fifth enterprise was just ‘commercialized’ into ‘sole-shareholder company of the state treasury’, which is quite immaterial alteration as compared with the status of soe. moreover, about 1,000 firms from the (statistical) privatization pool were dissolved, while even more than that were handed over to a government-dependent entity, the agricultural property agency (see central statistical office in poland, 2009). naturally, privatization in poland is by far not a textbook example of perfectly failed policy. tache (2008:12-14) enumerates such patterns, providing an example of her own country, romania, as well as russia, bulgaria and serbia, branding this model as a “patrimonial capitalism”. an extreme politicization of the underlying privatization schemes led virtually to a disappearance of hidden intrigues and charades, with vested interests being overtly displayed instead. for instance, members of the romanian parliament, sheltered by immunity, did not hesitate to perform managerial functions in the soes. in poland, misconduct of such gravity has not been seen. to provide a complete picture, however, it is important to add that vested interests related to privatization do not necessarily ruin the whole market-oriented effort. for instance, as rondinelli and yurkiewicz (1996:145) write, ‘the development of small and medium-sized enterprises and the spread of entrepreneurial activity were far more important than privatization of state enterprises in moving poland toward a market system’. nevertheless, in some jurisdictions, such as russia, the spoils system determined the quality of transition, and paralyzed market-oriented reforms (moors, 1997; fischer and sahay, 2000:21), as ‘the state was easily captured by well-organized industrial interests’ (kapstein and milanovic, 2000:29). the paper is structured as follows. this introduction, containing the very brief outlook of privatization-related struggles in poland, is followed by the discursive chapters on exogenous and endogenous sources of vested interests. the borderline between them happens to be blurred, but in general the exogenous originate outside the domain of policy makers, while the endogenous are rooted in politics. afterwards, the dichotomy between conventional and emerging vested interests is presented, the latter being the phenomena that arose relatively recently and have not been so far extensively covered by scientific research. what comes next is the illustration of the emerging vested interests on the basis of a concise case study carried out with regard to the largest polish soes. policy recommendations and further research proposals are derived in the final chapter. the paper’s contribution to the literature comes in two forms. first, it firmly introduces vested interests as a dynamic phenomenon, whereas the past literature was at best agnostic about their nature, or portrayed them as a static occurrence. second, the supportive value of this paper is that it offers a taxonomy of vested interests, not only dividing them into conventional vs. emerging, but also into exogenous vs. endogenous. the dynamics between these interconnected groups is also presented throughout the article.the conclusions may both enrich policy makers’ comprehension of vested interests domain, and assist in counteracting the overly improper use of political clout. as a rough introduction to the insights into exogenous and endogenous sources of privatization reluctance, one may review the following chart that should facilitate to keep up with the reasoning presented in this article. 48 senderski ● inhibited privatization figure 1. exogenous and endogenous sources of privatization reluctance source: own taxonomy. exogenous determinants of sluggish privatization although the intention of this paper is more to explore the endogenous dimension of privatization-related negligence, the brief recapitulation of exogenous factors is also desirable. as bel and fageda (2008:6) write, ‘it is commonly assumed that public employees and unions are in favor of internal production, while industrial interests have a greater preference for privatization’. the privatization effort is indeed quite often impeded by the trade unions empowered by directly ‘touched’ employees (warner and hebdon, 2001; careja and emmenegger, 2009). privatization issues are thus still central to positioning strategies espoused by political parties in transition states. in aggregate, one should agree with bortolotti and faccio (2004:2) remarking that there are few stories that document governments privatizing enthusiastically, and’most of the news stories depict governments that are faced with stiff opposition from many vested interests, but are forced to privatize because of budgetary shortfalls’. hence, exogenous determinants of sluggish privatization are, at least by some scholars, seen as dominant and powerful. the table on the next page shows the selected data for state-owned entities in oecd economies in 2009. based on the inputs presented in this table, the naïve correlations are as follows (figures in brackets indicate the correlation coefficient after excluding norway, which is outstandingly peculiar with its sovereign wealth funds): • employees as percent of population vs. gdp per capita: 0.043 (-0.342) • market value as percent of gdp vs. gdp per capita: 0.372 (0.023) • employees as percent of population vs. corruption perception index: -0.053 (0.193) • market value as percent of gdp vs. corruption perception index: 0.180 (0.072) • employees as percent of population vs. index of economic freedom: -0.383 (0.431) • market value as percent of gdp vs. index of economic freedom: -0.167 (-0.201) political conflict and/or indecisiveness exogenous (arisen in the broad social environment) endogenous (arisen among the parties with decision power) time weak domestic capital lack of domestic managerial capacity unwillingness to sell to foreigners, compounded with: deep belief in the „strategic importance” of a given soe social resistance as a result of one or several out of these phenomena: unfavorable academic climate or media coverage generic fear of change high perceived risk of layoffs or price increases apparent wrongdoings in privatizations conducted thus far, reinforcing the conviction of low ethical standards in the economy and politics „innovations” privileged positions within soes, both pecuniarily and in terms of prestige; no hard budget contraints that leads to empire building behaviors mutual benefits for politicians with nomination power and their nominees (kickbacks) convenience of catering for groups of interests extravagant wages and fringe benefits to earn or to give away nepotism csr lavish marketing budgets sports sponsorship all exploited and magnified by reluctant policy makers of politicians: of nominees: 49 pere ● governance and economic development of western balkan countries the caveats are particularly that correlation coefficients say nothing about causality and that it is a narrow and static approach, as the readily available data do not allow for full panel analysis. in fact, in order to advance to a more comprehensive picture, instead of having considered listed and non-listed enterprises, one should also include the statutory corporations, i.e. the units created by statutes. although christiansen (2011) has in fact done it, the oecd data set he uses has many nonnegligible blanks. referring to the academic discussion, not everyone had incentives to make key industries free from the shackles of communist continuum back in the early 1990s. some, “shock therapists”, worried that if the privatization did not proceed quickly, the overwhelming pool of vested interests would stimulate a reversion to socialism. others, ‘gradualists’ were uneasy about hurried reforms, claiming ‘they would be a disaster – economic failures compounded by political corruption – opening up the way to a backlash’ (stiglitz, 2002:27). the main arguments of the big bang enthusiasts centered around two focal points: detaining asset stripping and precluding vested interest groups from blocking privatization later on (fischer and sahay, 2000:12). indeed, radical actions seem to be superior to gradual actions when it comes to tackling vested interests, since the latter allow ‘defenders of the status quo far greater latitude to sabotage reform’ (megginson and netter, 1998:29). the burden of vested interests tied to the ancien regime was also recognized by earle and gehlbach (2002:5) who suggested a ‘strong state’ as a firewall against this threat. whereas in the 1990s “shock therapists” received vast support provided by an empirical literature, they were finally challenged by “gradualists”, who have revisited the models and pointed out their substantial and methodological flaws. overlooking initial conditions and institutional frameworks, and confusing the speed and the depth of reforms were the focus of critique (campos and coricelli, 2002:51; godoy and stirlitz, 2006:20).even sachs himself admitted afterwards that privatization policy prescriptions should contain less ideology, and be rather tailored to the country’s characteristics, since while ownership certainly matters, institutional infrastructure matters just as much (estrin et al., 2007:37; wallsten, 2002; zinnes, eilat and sachs, 2001:166). the break through periods are inseparably associated with extraordinary policies, initiated immediately as the country embarked on transition. first, ‘the magic of the moment’ facilitated winning, at low or no cost, public consent for reforms. second, it made it easier to dismantle disorganized vested interests (earle and gehlbach, 2002:6). however, much time has passed since eastern european societies last had an occasion to take advantage of this approach, because the period of extraordinary policies has left for good. while at the very beginning of the transition road, the reformative enthusiasm and people’s approval of undergoing economic changes indeed provided an auspicious environment, it deteriorated quickly after the wrong doings of privatization have been gradually made apparent. members of nomenklatura, supposedly most proficient when it came to understanding a complicated legal framework, have engaged in ‘spontaneous privatization’ right after new economic policies came into effect, and managed to strip ‘some assets of their firms through transfer pricing, sweet deals with foreigners, and other plundering devices’ (tirole, 1991:222). pittance savings of the ordinary population precluded it to take its slice of the new shareholding structures. as spotted by stiglitz (2002:143), even if some would be eager to make use of their nest egg to take over an enterprise, they would probably not be able to restructure it, given galloping inflation and risk averse or infant financing providers. as domestic investment capabilities were scarce at the beginning of the 1990s, the immense amounts of state-owned capital failed to attract broad masses. 50 senderski ● inhibited privatization table 1. statistics of selected oecd member countries in 2009. country gdp p.c. (ppp) 000 usd rankings enterprises majority-owned by the central level of government relative measures indicating the size of state-owned units listed non-listed total cpi ioef units employees market value units employees market value units employees market value employees as % of population market value as % of gdp aus 41.43 8.7 82.6 0 0 0.0 7 8 283 4.2 7 8 283 4.2 0.04 0.47 aut 39.26 7.9 71.2 2 28 741 8.2 6 50 459 7.8 8 79 200 16.0 0.95 4.87 bel 36.90 7.1 72.1 1 17 371 13.2 7 74 990 44.6 8 92 361 57.8 0.86 14.51 can 38.63 8.7 80.5 0 0 0.0 33 105 296 21.6 33 105 296 21.6 0.31 1.66 chl 15.92 6.7 78.3 1 156 0.2 9 5 559 2.7 10 5 715 2.9 0.03 1.08 cze 25.88 4.9 69.4 1 33 000 25.3 82 38 200 9.9 83 71 200 35.2 0.68 12.97 dnk 38.63 9.3 79.6 0 0 0.0 11 8 680 8.3 11 8 680 8.3 0.16 3.89 est 19.98 6.6 76.4 0 0 0.0 32 16 261 2.9 32 16 261 2.9 1.21 10.85 fin 35.87 8.9 74.5 3 24 844 29.4 28 61 187 16.3 31 86 031 45.7 1.61 23.86 fra 35.11 6.9 63.3 2 176 347 116.1 30 120 386 41.6 32 296 733 157.7 0.47 7.17 deu 35.96 8.0 70.5 0 0 0.0 57 66 419 22.9 57 66 419 22.9 0.08 0.78 hun 20.44 5.1 66.8 0 0 0.0 346 150 528 6.7 346 150 528 6.7 1.50 3.27 isr 27.71 6.1 67.6 0 0 0.0 29 50 264 43.2 29 50 264 43.2 0.67 20.83 ita 32.36 4.3 61.4 0 0 0.0 25 289 329 105.4 25 289 329 105.4 0.48 5.38 kor 26.22 5.5 68.1 8 39 599 38.3 48 81 056 139.4 56 120 655 177.7 0.24 13.72 mex 14.19 3.3 65.8 0 0 0.0 45 n/a 2.2 45 n/a 2.2 n/a 0.14 nld 41.16 8.9 77.0 0 0 0.0 28 60 355 74.1 28 60 355 74.1 0.36 10.84 nzl 29.87 9.4 82.0 1 10 726 0.5 17 17 107 9.1 18 27 833 9.6 0.64 7.36 nor 54.61 8.6 70.2 3 74 723 104.7 33 50 479 18.3 36 125 202 123.0 2.56 46.06 pol 18.79 5.0 60.3 13 184 079 59.5 573 542 082 34.0 586 726 161 93.5 1.89 12.92 prt 25.27 5.8 64.9 0 0 0.0 42 81 465 16.6 42 81 465 16.6 0.77 6.21 svn 26.93 6.6 62.9 3 3 048 0.9 33 22 276 3.1 36 25 324 4.0 1.24 7.25 esp 32.15 6.1 70.1 0 0 0.0 115 106 963 36.3 115 106 963 36.3 0.23 2.45 swe 37.29 9.2 70.5 0 0 0.0 43 143 253 66.1 43 143 253 66.1 1.53 18.90 che 46.81 9.0 79.4 1 19 813 19.8 1 7 534 0.7 2 27 347 20.5 0.35 5.59 gbr 35.36 7.7 79.0 1 160 900 50.7 12 202 668 5.5 13 363 568 56.2 0.59 2.59 source: christiansen, h. (2011) (data-related reservations outlined in this paper apply); transparency international, http://www.transparency.org/research/cpi/cpi_2009; the heritage foundation & wall street journal, www.heritage.org; oecd, www.stats.oecd.org. cpi stands for the corruption perception index, while ioef stands for the index of economic freedom. 51 pere ● governance and economic development of western balkan countries clearly, following the years of communist propaganda according to which the state’s property belonged to people, it was problematic to make people satisfied as many of them felt empty-handed. such an inference may be justified on the basis of pentor’s survey about privatization sentiments (see chart below). nevertheless, the research has been ceased after 2004, yet before most recent and highly profitable initial public offerings (ipos) of soes have been concluded (megginson and bortolotti, 2011), and further estimates are not available. now it seems that a certain reluctance prevails if the vested interests of citizens are endangered rather than those of politicians. therefore, ownership transformation is hampered in health care and utilities. nevertheless, it is beyond discussion that privatization is a two-edged sword for the state authorities, and some modes of ownership transformation (e.g. czechoslovakia’s voucher scheme, or – more up-to-date – poland’s ipos) may prove helpful in building up political capital and thus in increasing government’s popularity. privatization of soes through the mass enfranchisement of employees may also contribute to mitigating the anti-privatization movement within the company (shafik, 1996), but its track record in central and eastern europe is not very impressive, presumably due to mentality issues (castater, 2002:14-20; kozarzewski, 2008:257-260). moreover, the choice of a scheme is tricky as well. whereas czechoslovak privatization was in fact a ‘give-away to outsiders’ (glorious but inefficient), the russian case was rather a ‘give-away to insiders’ (definitely inglorious) according to roland’s (2000) typology. the search for a golden mean continues. figure 2. a survey regularly carried out by pentor until 2004, in which the question asked is: ‘do you generally support the efforts aimed at privatization of the polish economy?’ source: ministry of treasury (2005:279). the exogenous sources of policy makers’ reluctance towards privatization are also fueled by the prevalent term ‘thievish privatization’, implying that ethical standards lag behind the pace of structural transformation. this finding was confirmed by denisova, eller, frye and zhuravskaya (2010:4) who claim that ‘political elites have used public sentiment of illegitimacy of privatization to redistribute assets to themselves or their supporters’. a similar survey has been conducted for latin america by checchi, florio and carrera (2005), who found out that privatization discontents are predominantly poor, live in countries with high income inequalities, whereas privatizations in question were large, quick and involved such public utilities as water or electricity. yusuf (2009:73) enumerates a number of “sins” of privatization in transition economies, and to name just a few: sales to insiders, asset stripping, weak managerial capabilities, limited competition, feeble regulation, all amplified by resistance from vested interests. godoy and stirlitz (2006:22) hypothesize that the countries that embarked on rapid privatizations may suffer from the prolonged problem that property rights lack legitimacy in society. the societal approval of the pace of economic reforms was underscored by watts, walstad and skiba (2002). 57% 55% 60% 49% 37% 38% 40%28% 27% 23% 33% 39% 40% 43% 15% 18% 17% 18% 23% 21% 17% 1992 1996 1997 2000 2002 2003 2004 support objection indifference 52 senderski ● inhibited privatization state authorities, in fact, tend more to explore the public concerns than to undertake education campaigns to minimize them. they usually take public unrest at its face value and reinforce it, by emphasizing the argument of allegedly strategic importance of particular enterprises or branches, which seems to be a plausible excuse. the strategic importance is to some extent reflected in the polish corporate law. radwan and regucki (2012:17-18) present the articles of association of pgnig (a soe dealing with exploration and production of natural gas and crude oil) affirming that ‘the state may consent to the company taking actions or making investments that permanently or temporarily reduce the company’s economic efficiency, but are necessary for maintaining the energy security of the country’. this passage instructs that the state intervention is rationalized, as it is capable of endorsing social benefit at the cost of economic benefit of the corporation. the tensions between the commercial and non-commercial nature of soes are more thoroughly explored by christiansen (2013). in contrast, raszewski (2012:130-131) warns that labeling something as a security issue (i.e. a strategically important matter) allows for dealing with such a problem ‘behind closed doors’. in fact, such a securitization becomes an extreme form of politicization. however, strategic status may be ascribed only to a fraction of soes (e.g. energy industry, petroleum industry, rail transport, aviation, shipbuilding, arms industry, coal mining industry), whereas still hundreds of enterprises remain de facto isolated from stimulating free market mechanisms. this “excuse” is amplified by another. many people are hesitant towards potential buyers, mainly foreign companies, which do not take into account polish national interests. however, the argument of national interests is incomplete, as markets deemed strategic are usually strictly supervised by domestic regulatory bodies that restrain potentially abusive behaviors of foreign agents. moreover, the dual role of state being both the owner and the regulator may give birth to conflicts of interest (bauer, 2003). also, the question of whether public authority makes shareholders better off, as compared with private management, remains highly controversial. an overpaid (2,34 bn usd for the control stake) deal of pkn orlenpurchasing lithuanian mažeikiairefinery (mažeikiųnafta) in 2006 would have probably never taken place if pkn orlen had not been controlled by the state treasury. the transaction was carried through due to political reasons, which were later confirmed by both orlen’s executives and political leaders, but the refinery’s operations are significantly hampered by the anti-polish activity of russian oligarchs (vitkus, 2009:31-33). on the borderline between exogenous and endogenous vested interests lies the ideological fragmentation of the political scene, i.e. the internal conflict of political elites, though existing as a consequence of inconclusive results of democratic elections. as bortolotti and pinotti (2008:18) argue, greater political fragmentation (precisely: the number of agents with veto power) is the factor entailing longer delays to implement large scale divestiture. the same is confirmed for economies considered developed, like italy, where economic benefits from privatization have not been realized in full partly because of the ‘inherent weakness of the coalition governments that have run the country’ (goldstein and nicoletti, 2003:30). needless to add, an unstable political setting is highly likely to arise in transition states (lipton and sachs, 1990:298-299; stark, 1990; šikulová and frank, 2013:30). consequently, at the very beginning of the transition route, foreign investors are usually concerned about huge political risks. they not only risk low returns, but also, if unlucky, may have their assets seized by the next government, after being accused of, say, illegitimate privatization (stiglitz, 2002:144). nevertheless, as rattsø (2012) evidenced for norway, political conflict there also raises the influence of interest groups around soes. the example of an abundant state struggling with the same shortcoming as transition economies adds up to the universality of this finding. on a related note, imai (2009), who examined the privatization of japan’s postal saving system, found robust evidence for incredibly complex legislative behavior with plethora of political factors involved and with conflicts not only between, but also within political parties. 53 pere ● governance and economic development of western balkan countries after more than 20 years of successful market economy in poland, the existence of state-owned banks, power plants, refineries and coal mines, in addition to still considerable amount of minority stakes, is difficult to apprehend. since at the beginning of transition political agents may truly worry about layoffs and broadly understood social cost, these factors should diminish as time goes on (agarwal and nunnenkamp, 1992:18; godoy and stirlitz, 2006:18; gupta, ham and svejnar, 2008:188-189). overall, exogenous sources of vested interests typically lie behind the relatively idealistic interpretation of government inactivity in terms of ownership transformation of soes. as megginson and netter (1998:7) put it, benevolent governments, in this optimistic view, act ‘in the (perceived) best interest of their citizenry’ and are reluctant to resign from control fearing that the private sector would fail in providing necessary (strategically important) goods at an acceptable cost. using less lofty language, the government simply takes exogenous vested interests for granted. the next chapter shows that the driving force of government’s procrastination may equally well lie elsewhere. endogenous sources of privatization aversion despite the unambiguously important influence of civic dissatisfaction, resistance or anxiety towards privatization, a deal of effort in this paper is aimed at exploring the supply-side of privatization-related negligence. as we already know what societal convictions feed political agents and underpin their privatization aversion, let’s now investigate the endogenous factors that flourish solely within the political environment. this is along the lines of olofsgård (2003:11) who insists that ‘vested interest groups do not necessarily come from outside’, but the resistance to reform may be produced within the political elite. anti-privatization vested interests’ sources are twofold. the first cradle constitutes nomenklatura being somehow in charge of a given enterprise’s operations. they would oppose privatization as long as it takes in order to retain their own positions. van de walle (1989), who has not seen by then the post-communist privatization programs ‘in all their glory’, stated optimistically that political opposition to privatization is usually limited to the state bureaucracy. although he also warned that this opposition may extend from corporate level to the level of the ministry overseeing this corporation, he rather failed to predict that these were politicians and not the bureaucracy that took the lead in impeding privatization as time went on. hence, the second source is even more sustainable and also more sophisticated. implicit knowledge instructs that having an umbrella over public enterprises is a valuable fringe benefit for politicians capable of taking advantage of it. either directly or indirectly, supervisory boards of such entities are nominated by the ones currently in power. this creates boundless opportunities of appointing one’s close associates and allies, possibly incompetent or not caring about the business. decent wage, prestige and almost sure windfall severance package in case the political landscape changes, and the purge led by new government reshuffles the lineup of supervisory or management board members. assumingit is not very unlikely that the enterprise’s corporate governance may be rotten to the core, identical mechanism is due to be replicated downwards throughout the organizational chart, magnifying the impact of a political steering wheel. this renowned mechanism, coined as a “spoils system” (alternatively, “patronage system”), originated in the united states and reached its climax during andrew jackson’s presidency (1829-1837).however, it has been eradicated until nowadays, neither in poland (see e.g. majcherkiewicz and gadowska, 2005; nowakowski, 2005; gwiazda, 2008) nor in its cradle, the united states (see e.g. lewis, 2009). promises of positions in return for political support, e.g. during an electoral campaign, are not the only way to abuse one’s mandate. politicians have in fact a broad portfolio of opportunities to endorse their proponents and temptations are all 54 senderski ● inhibited privatization around, multiplied by prospective kickbacks in the form of pure cash or campaign contributions. moreover, dinç and gupta (2011) found country-specific evidencethat no firm located in the state from which the indian minister comes from (one who has jurisdiction over that firm), was ever privatized. as a matter of fact, india’s overly bureaucratic and discretionary system was enough to create vested interests that prevented reform in the past decades (singh, 2010:5). this result suggests that political patronage has a significant impact on the privatization decision. these mechanisms may be attenuated if politicians are restricted by the fear of private media coverage (schoenman, 2005:50-51), but it materializes only provided that the abuse is evident and truly scandalous. the recent press coverage for lubin, a headquarters of the state-owned giant kghm, indicates that many local politicians sat in supervisory boards of kghm or communal firms, although the remuneration taken from these firms has to be disclosed to the public, according to the domestic law (see open society institute, 2002:418-419, for earlier instances of such misconducts). politicians’ vested interests in poland have been catalyzed for a long time via the “chimney bill” (limiting compensation for management in soes to six times the national average wage) that heavily curtailed top managers’ willingness to work for soes. according to dla piper, a global law firm, the bill of this kind was unprecedented in the european union (money.pl, 2008). therefore, it is not surprising that comparative literature in this matter is non-existent. the bill was essentially a deterrent for top-notch specialists, and an open invitation to appoint less endowed executives. when the shortlist of contenders is unimpressive, it is relatively easy to pick pre-agreed candidates. as noted by bohdan wyżnikiewicz from the gdansk institute for market economics, the pathway to state-controlled management boards was open for young and inexperienced rookies, ‘almostpensioners’ or those attracted by political careers (rynekinfrastruktury.pl, 2012). the same finding, though indirect and deprived of detailed description, was revealed even by the ministry of treasury in poland (2005:124) in its annual report on privatization. the “chimney bill” was alleviated in 2010 by donald tusk’s government, privatizing quite eagerly as compared with preceding cabinets, though still exhibiting symptoms of excess procrastination. according to media reports, the “chimney bill” is now virtually “dead”, with top soes, such as polskagrupaenergetyczna (pge), enea, tauron, polish state railways (pkp), jastrzębskaspółkawęglowa (jsw) or kompaniawęglowa, bypassing the bill by making use of “managerial contracts”. a separate issue in the context of soes is the rotation of ceos, implying the lack of long-term strategic stability.for instance, the current ceo of lot polish airlines is the eighth since 2005 (not counting the interim ones). similarly, państwowyzakładubezpieczeń (pzu) had six ceos between 2001 and 2007, and kghm polskamiedź (kghm) had seven ceos between 2001 and 2009. whereas in a free market company, the management is preoccupied with profit maximization tasks, the management of a state-controlled entity, regardless of their will, is tied by motivations of political rather than economic nature. the managers of statecontrolled enterprise are expected to maintain advantageous relationships with their benefactors, as numerous studies have validated the importance of managerial political connections (faccio, 2010; hillman, 2005; khwaja and mian, 2005; wu and cheng, 2011). bortolotti and pinotti (2003:9) remind that it has already been largely documented that soes are a source of political rent for elected politicians, ‘who can interfere in the operating activity of the company in order to cater specific interest groups’. later on, they urge about most common modi operandi. these may include e.g. keeping redundant workers to push away the threat of unemployment and earn voters’ admiration. this is what rosa and pérard (2010:112) would nickname as ‘interestgroup real politik’, while boycko, shleifer and vishny (1996:310) argue that the spending politician may still be willing to subsidize a firm even after it is privatized, 55 pere ● governance and economic development of western balkan countries with the intent to ‘buy’ excess labor spending. all of this despite the fact that employment rose and the working class was better off after privatization (galal et al., 1994; megginson, nash and van randenborgh, 1994), contrary to anecdotal evidence. bjørnskov and potrafke (2011:206) remark that vested interests more typically pertain to left-wing governments. by the same token, li and lui (2004:212-213) argue that right-wing governments may have greater motivation in privatization or liquidation of soes, as this would weaken trade unions, their natural foe. in the light of the above mentioned arguments, schindele’s (2003:13) view, underlining a substantial role of politicians’ interest in the privatization process, seems to be an accurate synopsis: ‘being able to use the basic trade-off between the costs of restructuring the economy and efficient privatization, politicians might obtain private benefits from hindering privatization’. conventional versus emerging vested interests traditional sources of vested interests include the temptation to forego economic rationality in exchange for “empire building” behaviors. most of these issues can be viewed as classical principal-agent problems (vickers and yarrow, 1989). in the context of a non-privatized company, this boils down to three particular situations: • “pumping up” the size of the company, e.g. through unnecessary investments or foreign expansion, which lacks economic rationale; • the exploitation of sinecures, used to bolster one’s political influence; • unnecessary, non-austere perks for the management, such as luxurious limousines, leather armchairs, but also the new headquarters. for instance, one of the official chinese reports, quoted by cheng (2012:359) indicates that ‘in a lot of loss-making soes, where workers could not receive wages, managers paid themselves full wages, bonuses, and subsidies, and rode around in luxury cars’. the vested interests that may be deemed new to this universe are the following: • creating additional channels of political influence and expanding the catalogue of incentives, with which politicians may please their networks (e.g. tolerating extravagant activity in the area of corporate social responsibility, sports sponsorship arrangements, lavish marketing expenses); • proactive family employment (nepotism), of substantial value in the age of turbulent market environments. shleifer and vishny (1994:1024) argue that the privatization is most likely to occur when politicians cannot obtain large benefits from public firms any more. this is an apparently obvious axiom, but what the literature frequently omits is that political benefits, as shown above, may alter. for instance, once public opinion stops to accept politicians deriving a particular benefit from public firms, politicians might “invent” a new, less controversial genre of benefits. as it seems, the new vested interests display two traits. first, they are less unequivocal in terms of moral assessment, which means they are not unambiguously wrongdoings, at least at first glance. such an inference is justified on the basis of a recent public opinion research center report (centrum badaniaopiniispołecznej, 2012), where nepotism was not unequivocally despised. second, they generally account for a more proactive role of the managers put in charge of the soes. in this view, the manager’s (endowed’s) vision is not solely the derivative of the politician’s (endower’s) vision. it is more independent and individualistic. 56 senderski ● inhibited privatization it is generally difficult to decide whether these two are more exogenous or endogenous vested interests. on the one hand, they are derivatives of the independence of soes’ management and as such, individuals in charge of soes are the main beneficiaries, enjoying prestige and respect. on the other hand, however, given complicated mutual relations and reciprocation schemes between politicians and their nominees, politicians do also have incentives to maintain a pool of attractive job posts to fill. and these two additional benefits undoubtedly add up to the attractiveness of managerial positions in soes. illustrations of the problem one of the most prevalent bonuses associated with not privatized enterprises is the relatively convenient opportunity to spend money for non-business activity, including sports team ownership or sports sponsorship schemes. by participating in these, politicians associate themselves with positive emotions, playing the role of the “sugar daddy”. in the communist era, economic flows activated not only by regular and fullfledged soes but also by state-owned farms on the more local level, aimed at supporting sport constituted an obvious necessity given no alternative source of financing (andreff, 2008:23; micek, neo and górecki, 2011:48). contrary to the united states, where the so-called sports industry is resilient enough to secure vast resources, in much part of europe the idea of subsidized sport outlived the totalitarian regimes. nevertheless, public opinion is usually sensitive to the mismanagement of taxpayers’ money by a beneficiary sporting entity (andreff, 2006:9). table 2. a summary of soe-related sports sponsorship activities for key professional leagues in poland. note: names of companies were put in parentheses only if they are not included in club’s official name does soe sponsors the league? do soes sponsor particular clubs? football men no pge gks bełchatów, ruchchorzów (węglokoks), while kghm zagłębielubin is fully owned by soe basketball men yes (tauron basket liga) energa czarni słupsk, pge turów zgorzelec, polpharma starogard gdański (tauron), stelmet zielona góra (pgnig, pkp energetyka) women no energatoruń volleyball men no pge skra bełchatów, jastrzębski węgiel (jsw), lotos trefl gdańsk women yes (orlen liga) atom trefl sopot (pge), tauron mks dąbrowa górnicza handball men yes (pgnigsuperliga) orlenwisłapłock, azotypuławy, tauronstalmielec women yes (pgnigsuperliga) kghm metraco zagłębie lubin, spr lublin (pge) source: proprietary analysis based on relevant corporate websites. it is not feasible to gather wide-ranging and accurate data on corporate involvement in sponsorship schemes. even listed companies, although subject to rigorous discipline in terms of investor relations, do not have a duty to disclose such precise figures. this already should be a red flag for soes supervisors. screening the yearly reports of twenty largest enterprises traded on the warsaw stock exchange, reveals that whether firms share the details or not remains their discretionary choice. nevertheless, code of best practice for wse listed companies (warsaw stock exchange, 2011), updated 19 october 2011, says: ‘if a company supports different forms or artistic and cultural expression, sport activities, educational or scientific activities, and considers its activity in this area to be a part of its business mission and development strategy, impacting the innovativeness and competitiveness of the enterprise, it is good practice to publish, in a mode adopted by the company, the rules of its activity in this area.’ 57 pere ● governance and economic development of western balkan countries this rule should be applied not later than 1 january 2013. based on yearly reports covering 2011, soes most often conceal their activity in this field. such giants as orlen, pgnig or pge, known for their active participation in sponsorship schemes, do not mention a word on it in their reports. table 3. a summary of 2011 management reports issued by the 20 largest companies on the warsaw stock exchange (wig20 is a capitalization-weighted stock market index) no. company treasuryst ake (direct) free float (mil pln) wig20 share nature of involvement in sports, according to the 2011 official yearly report 1 kghm 31.79% 26,395 15.086% 100% ownership in zagłębielubin s.a., management of football section, organization of professional sporting events, enlisted as part of csr 2 pkobp 33.39% 24,036 13.738% mentions several running events that the bank supported 3 pzu 35.1875% 23,674 13.531% none 4 pekao 17,960 10.265% none 5 pknorlen 27.52% 15,903 9.089% none 6 pge 61.89% 13,654 7.804% none 7 pgnig 73.50% 8,596 4.913% none 8 tpsa 8,092 4.625% none 9 tauronpe 30.06% 5,145 2.940% none 10 bogdanka 4,587 2.622% admits that there is a separate advertizing budget dedicated to sports sponsorship (7,932,500 pln), along with the very broad clarification of why the sports sponsorship is important to the firm. the report enumerates clubs that have been supported. 11 bre 4,162 2.378% none 12 jsw 55.16% 3,612 2.064% none 13 handlowy 3,247 1.856% none 14 kernel 3,231 1.847% none 15 assecopol 3,167 1.810% mentions sponsorship agreements with assecoprokom gdynia (agreement set to expire on 31 july 2015) and assecoresovia (until 31 may 2014). 16 synthos 2,742 1.567% none 17 lotos 53.18% 2,468 1.411% none 18 gtc 2,153 1.230% none 19 tvn 1,568 0.896% none 20 boryszew 572 0.327% none source: www.stooq.pl [accessed: 5 january 2013]; applicable corporate reports. although sports sponsorship has been incorporated into the corporate marketing function over time (see farrelly, quester and burton, 1997), there are doubts on whether the scope of support is not too lavish in the case of soes. there are also numerous examples of municipal firms, or municipalities directly, that sponsor professional sports. however, although controversial, this model does not lead to a geographical discrimination. supporting the local team from local taxes may be considered as part of municipality’s strategy to attract new inhabitants to the town or to provide them with entertainment. on the contrary, country-wide soes, if they invest locally, discriminate against other locations, which are deprived of their own sports team. cherry-picking local teams to support at the cost of other regions seems to be a breach of neutrality. since corporations back professional teams more enthusiastically than amateur sports and physical activity in general, the underlying moral case to be tackled is even more pronounced. 58 senderski ● inhibited privatization another problem associated with soe-related vested interests is family employment. this not only concerns large nationwide corporations where such schemes are difficult to detect, but particularly the small-scale municipally-owned entities. such a category usually comprises bus operators, waterworks companies, provision of public utilities, road maintenance or city cleaning services. in august 2012, “rzeczpospolita” disclosed that 40 percent of warsaw city council members work for various governmental or self-governmental bodies. overall, polish mayors and heads of counties employ 250,000 people, whereas the central administration hires 180,000. in the age of unemployment and bleak economic prospects, the ability to create jobs and hand-pick employees adds up much to the prestige of an official, who may derive rents and expect for reciprocation from those employed. it is not uncommon for the local governor to be a leading employer in a commune or in a county. sixteen voivodeship capitals operate 280 municipally-owned entities (including 44 in warsaw only). municipally-owned firms are attractive to territorial governments, since in many cases they allow for hiding debt. since poland’s constitution caps debt to a pre-determined threshold, territorial governments have limited options for boosting their expenditures and thus municipally-owned entities become helpful in deceiving their true financial condition.in general, the privatization at the local level is a separate issue, with different set of incentives and disincentives. the level of complexity varies and while some scholars have found evidence that fiscal constraints, political and ideological considerations are decisive determinants of privatization policy (bel and fageda, 2009:116), others rather point out pragmatic concerns such as service quality (warner and hebdon, 2001). a scandal that broke out in 2012 around the agency for restructuring and modernisation of agriculture (arimr) is edifying. as it became clear for the public, the agency, whose task is particularly to transfer the eu agriculture funds to polish beneficiaries, has numerous staff. apart from the headquarters with 25 departments and 16 regional offices, there are 314 local offices. it has over 10,000 employees in total and a considerable potential for sinecure creation. simultaneosuly, the agricultural market agency (arr), with tasks partially overlapping with arimr, hires 1,200 people and – to make the picture full –there are additional 1,000 jobs in the already mentioned agricultural property agency. some theoretical explanation of the phenomenon of nepotism in soes is offered by santalainen, baliga and leimann (2003:78-79), who underline that the need to adjust to the free market standards inevitably leads to the disruption of prevailing corporate norms and values, part of which is providing employment to friends and relatives. they also highlight the excess employment that typically prevails in soes, and wrote about the problem of altered ‘psychological contract’ that invades safety of employees and gives rise to their disappointment. gabris and simo (1995) tried to discern whether the career motivations of public sector employees and private sector employees are contradictory, which would partially explain the perplexity of soes’ employees when faced with privatization plans, but they did not arrive at any convincing results. having said that, we may conclude that soes have become full-grown extensions of the regular public authority. state-owned business units serve no more solely to protect the country’s economic interest. the above mentioned case studies illustrate the multidimensional expansion of vested interests that indicates that they flourish rather than fall into decline. precisely, the corollary from screening these several cases is that the magnitude of the phenomenon is considerable and the innovation in inventing and exploiting the new sources of vested interests fell into a habit of the polish government, territorial governments and soes. whereas public opinion has already learnt how to blow the whistle to condemn the soes’ behavior when conventional vested interests are manifested (which is not to say that these conventional sources are in decline), it still 59 pere ● governance and economic development of western balkan countries does not have tools, or will, to face the new wave of spoils. the delicateness of sports sponsorship, csr or nepotism is evident. criticizing the sponsorship of “glorious” goals may sometimes fail to attract many admirers. criticizing nepotism must make anybody think twice, whether the claimant herself and her closest relatives are above suspicion. in the interconnected world of politics and bureaucracy, sinlessness is in deficit. conclusions and policy implications this paper recognizes vested interests as one of the primary premises that reduce the effectiveness of privatization policy, stall its momentum and produce structural problems in the long-run. since empire building behaviors, creation of sinecures or extravagant management style have all been identified by scholars long ago, this paper strove to unravel more up-to-date, emerging and yet unexploited sources of vested interests. the channels through which conventional vested interests have snowballed over time were briefly described. among this new set of privatization constraints, a closer look was cast at employing relatives and engaging in lavish sports sponsorship arrangements. the latter appears to be the favorite domain of marketing activity for polish soes. policy recommendations provided by the literature as regards the solution of the underlying deficiency read as follows. alesina and tabellini (2004:23) urge that if vested interests are in place, it is more efficient to let a non-elective bureaucrat handle a given problem than to have it retained by a politician, inherently vulnerable to voters’ rage. a similar argument emerges from the inferences of earle and gehlbach (2002:5), as the absence of an established reform-oriented constituency is likely to produce a backlash when the inevitable privatization-related short-term costs occur. to mitigate the risk of premature policy reversal, some kind of “insulation” of the polity from the public may be considered. moreover, bortolotti and pinotti (2003:9) seem to have found that the electoral system is important: the ones with majoritarian systems perform better and politicians are less likely to pursue rent seeking behaviors, as compared with the system based on proportional representation. schindele (2003:13) puts forward a solution, too: bargaining models of privatization suggest that ‘in order to achieve efficient privatization, corruption should be impossible and firms should face hard budget constraints’, which necessitates the full coverage of costs with revenues. in fact, neither of these conditions has ever been met in full in poland. the issue of implementation of any guidelines in the underlying matter remains difficult, due to the self-policing problem (nemoiudex in causa sua). as far as the conclusions stemming directly from this paper are concerned, two policy recommendations seem justified. first, given no rapid privatization, the remuneration schemes in public enterprises call for amendment. the pattern of the compensation should be benchmarked to the market, reflecting the significance of the function, as well as ensure the appropriate level of motivation for the manager. the current “chimney bill”, approved in 2000 under the populist pressure, is by no means the replication of the stimulating free market remuneration plans. naturally, this will exert a tangible impact on the minimization of interest group influence only if it suffices to attract top-notch independent managers to soe, and if it is accompanied by a fair and unbiased manager selection process. second, the multilateral improvement in transparency should be demanded. this may capture both the overt and detailed information of all marketing arrangements that soes take up (what has been partly stipulated by the already mentioned code of best practice for wse listed companies), as well as the disclosure of all relevant conflicts of interest that decision makers might have. as soon as such conflicts arise, a politician or a bureaucrat should be advised to dissociate from the decision making, since the quoted examples from india or japan proved that the detachment of reason and the subordination to pressure group interests are quite common when vested interests come into play. the antidote for the expanding set of vested interests lies obviously 60 senderski ● inhibited privatization also in the ethical conduct of the ones in power, high standard of which being always deficient in transition economies. this, however, is the starting point for a completely different study. there are several directions in which further investigation into the roots of vested interests should proceed. first, the paramount challenge is measurement. the quantitative tools are of limited value due to the poor data sets, data opacity and problematic sampling. methods of how to navigate this paucity need to be improved. second, the issue of universality seems interesting. since this paper tries to display the polish pattern of petrifying vested interests, it does not aspire to be considered ubiquitous. third, the actual strength and durability of vested interests in the various stages of privatization processes is interesting. with regard to the last proposal, two hypotheses may be crafted to stimulate future research. one, vested interests have limited impact on the very beginning of transformation, when reformist enthusiasm is larger. two, the nature of vested interests changes over time and new ones crowd out the older ones, what was partially argued in this paper. references agarwal, jamuna p. and peter nunnenkamp (1992) ‘methods and sequencing of privatization: what post-socialist countries can learn from chile’, kiel institute for the world economy, kiel working paper no. 527. alesina, alberto f. and guido e. tabellini (2004) ‘bureaucrats or politicians?’, center for economic studies & ifo institute for economic research, working paper series no. 1130. andreff, wladimir (2006) ‘new perspectives in sports economics: a european view’, aies/iase, working paper series no. 06-05. andreff, wladimir (2008) ‘globalization of the sports economy’, rivista di diritto ed economia dello sport 4(3): 13-32. bauer, johannes m. (2003) ‘the coexistence of regulation, state ownership and competition in infrastructure industries: quagmire or institutional innovation’, quello center for telecommunication management and law, working paper 003-2003. bel, germà and xavier fageda (2008) ‘local privatization, intermunicipal cooperation, transaction costs and political interests: evidence from spain’, research institute of applied economics, working paper no. 2008/04. bel, germà and xavier fageda (2009) ‘factors explaining local privatization: a meta-regression analysis’, public choice 139(1):05-119. bjørnskov, christian and niklas potrafke (2011) ‘politics and privatization in central and eastern europe: a panel data analysis’, economics of transition 19(2): 201230. bortolotti, bernardo and mara faccio (2004) ‘reluctant privatization’, fondazione eni enrico mattei, working paper no. 2004.130. bortolotti, bernardo and paolo pinotti (2003) ‘the political economy of privatization’, fondazione eni enrico mattei, working paper no. 2003.45. bortolotti, bernardo and paolo pinotti (2008) ‘delayed privatization’, bank of italy, working paper no. 663. boycko, maxim, andrei shleifer and robert b. vishny (1996) ‘a theory of privatization’, economic journal 106(435):09-319. campos, nauro f. and fabrizio coricelli (2002) ‘growth in transition: what we know, what we don’t, and what we should’, william davidson institute, working paper no. 470. 61 pere ● governance and economic development of western balkan countries careja, romana and patrick emmenegger (2009) ‘the politics of public spending in post-communist countries’, east european politics and societies 23(2):165-184. castater, nichole m. (2002) ‘privatization as a means to societal transformation: an empirical study of privatization in central and eastern europe and the former soviet union’, fondazione eni enrico mattei, working paper 2002.76. central statistical office in poland [głównyurządstatystyczny] (2009) ‘prywatyzacjaprzedsiębiorstwpaństwowych w 2008 r.’ [privatization of state-owned enterprises in 2008], warsaw. centrum badaniaopiniispołecznej [public opinion research center]. (2012) ‘polacy o nepotyzmie w życiupublicznym’ [opinions about nepotism in public life],research report no. bs/124/2012. checchi, daniele, massimo florio and jorge carrera (2006) ‘privatization discontent and its determinants: evidence from latin america’, iza discussion paper no. 1587. cheng, xiaonong (2012) ‘the self-empowered new master at the workplace: privatization in russia and china’, department of sociology at princeton university, november, mimeo. christiansen, hans (2011) ‘the size and composition of the soe sector in oecd countries’, oecd corporate governance, working paper no. 5. christiansen, hans (2013) ‘balancing commercial and non-commercial priorities of state-owned enterrises’, oecd corporate governance, working paper no. 6. denisova, irina, markus eller, timothy fryeand and ekaterina zhuravskaya (2010) ‘everyone hates privatization, but why? survey evidence from 28 post-communist countries’, centre for economic and financial research at new economic school, working paper no. 143. dinç, serdar and nandini gupta (2011) ‘the decision to privatize: finance and politics’, journal of finance 66(1):241-269. earle, john s. and scott gehlbach (2002) ‘a spoonful of sugar: privatization and popular support for reform in the czech republic’, upjohn institute staff, working paper no. 02–79. estrin, saul, jan hanousek, evzen kocenda and jan svejnar (2007) ‘effects of privatization and ownership in transition economies’, ipc working paper series no. 30. faccio, mara (2010) ‘differences between politically connected and nonconnected firms: a cross-country analysis’, financial management 39(3):905–928. farrelly, francis j., pascale g. quester and richard burton (1997) ‘integrating sports sponsorship into the corporate marketing function: an international comparative study’, international marketing review 14(3):170-182. fischer, stanley and ratna sahay (2000) ‘the transition economies after ten years’, national bureau of economic research, working paper 7664. gabris, gerald t. and gloria simo (1995) ‘public sector motivation as independent variable affecting career decisions’, public personnel management 24(1): 33-51. galal, ahmed, leory jones, pankaj tandon and ingo vogelsang (1994) welfare consequences of selling public enterprises. washington, dc: world bank. godoy, sergio and joseph e. stiglitz (2006) ‘growth, initial conditions, law and speed of privatization in transition countries: 11 years later’, national bureau of economic research working paper 11992. goldstein, andrea and giuseppe nicoletti (2003) ‘privatization in italy 1993-2002: goals, institutions, outcomes, and outstanding issues’, prepared for the cesifo 62 senderski ● inhibited privatization conference on privatization experiences in the eu, held in munich on 10/11 january 2003. gupta, nandini, john c. ham and jan svejnar (2008) ‘priorities and sequencing in privatization: evidence from czech firm panel data’, european economic review 52(2):183-208. gwiazda, anna (2008) ‘party patronage in poland: the democratic left alliance and law and justice compared’, east european politics and societies 22(4): 802-827. hertog, steffen (2010) ‘defying the resource curse: explaining the successful state-owned enterprises in rentier states’, world politics 62(2): 261-301. hillman, amy j. (2005) ‘politicians on the board of directors: do connections affect the bottom line?’, journal of management 31(3): 464-481. iimi, atsushi (2002) ‘an empirical note: privatization transaction and macroeconomy in developing countries’, jbici working paper no. 10. imai, masami (2009) ‘ideologies, vested interest groups, and postal saving privatization in japan’, public choice 138(1/2): 137-160. jamasb, tooraj (2006) ‘between the state and market: electricity sector reform in developing countries’, utilities policy 14(1): 14-30. kapstein, ethan b. and branko milanovic (2000) ‘dividing the spoils: pensions, privatization, and reform in russia’s transition’, world bank policy research working paper no. 2292. khwaja, asimijaz and atif mian (2005) ‘do lenders favor politically connected firms? rent provision in an emerging financial market’, the quarterly journal of economics 120(4):1371-1411. kowalik, tadeusz (1991) ‘marketization and privatization: the polish case’, in miliband, ralph, panitch, leo. (eds.) communist regimes: the aftermath, london: merlin, pp. 259-278. kozarzewski, piotr (2008) ‘corporate governance formation in poland, kyrgyzstan, russia and ukraine’, in mcgee, robert w. (eds.) corporate governance in transition economics, london: springer, pp. 251-288. krusell, per and josé victor rios-rull (1996) ‘vested interests in a positive theory of stagnation and growth’, the review of economic studies 63(2):301-329. lewis, david e. (2009) ‘patronage appointments in the modern presidency: evidence from a survey of federal executives’, vanderbilt law and economics research paper no. 09-24, vanderbilt public law research paper no. 09-20. li, david d. and francis t. lui (2004) ‘why do governments dump state enterprises: evidence from china’, in takatoshi ito and anne o. krueger (eds) governance, regulation, and privatization in the asia-pacific region, nber east asia seminar on economics 12:211-230. lipton, david and jeffrey sachs (1990) ‘privatization in eastern europe: the case of poland’, brookings papers on economic activity 2: 293-341. majcherkiewicz, tatiana, gadowska, kaja (2005) ‘political clientelism in public administration: a case study of institutional changes in the post-communist state of poland’, paper presented at the 13th nispacee annual conference: ‘democratic governance for the xxi century: challenges and responses in cee countries’, 1921 may 2005, moscow. megginson, william l., bortolotti, bernardo (2011) ‘privatization trends and major deals in 2010 and 1h2011’,url (consulted may 2014): http://ssrn.com/abstract=2136780. 63 pere ● governance and economic development of western balkan countries megginson, william l., robert c. nash and matthias van randenborgh (1994) ‘the financial and operating performance of newly privatized firms: an international empirical analysis’, the journal of finance 49(2):403-452. megginson, william l. and jeffry m. netter (1998) ‘from state to market: a survey of empirical studies on privatization’, nyse working paper 98-05. micek, grzegorz, harvey neo and janusz górecki (2011) ‘foreign direct investment, institutional context and the changing polish pig industry’, geografiskaannaler: series b, human geography 93(1):41-55. ministry of treasury in poland (2005) ‘raport o przekształceniachwłasnościowych w 2004 roku’ [report on ownership transformation in 2004], warsaw: author. money.pl (2008) ‘polskaustawakominowajedyna w krajach ue’ [polish “chimney bill” is the only one in the eu], url (consulted march 2014): http://prawo.money.pl/ aktualnosci/wiadomosci/artykul/polska;ustawa;kominowa;jedyna;w;krajach;ue,120, 0,330360.html. moors, kent f. (1997) ‘the failure of russian privatization 1992-1994: how the industrial nomenklatura prevented genuine reform’, journal of international legal studies 3(1):1-52. newton, ashley n., uysal, vahap b. (2013) the impact of political connectedness on firm value and corporate policies: evidence from citizens united. mimeo: university of oklahoma. nowakowski, krzysztof (2005) ‘problems of clientelism and corruption in poland’, review of sociology 11(2): 75-86. olofsgård, anders (2003) ‘the political economy of reform: institutional change as a tool for political credibility’, background paper to the world bank’s 2005 world development report. open society institute. (2002) ‘corruption and anti-corruption policy in poland’, in monitoring the eu accession process: corruption and anti-corruption policy, new york: author, pp. 393-450. opper, sonja (2004) ‘the political economy of privatization: empirical evidence from transition economies’, kyklos 57(4):559-586. radwan, arkadiusz and tomasz regucki (2012) ‘the possibilities for and barriers to sustainable companies in polish company law’, university of oslo faculty of law research paper no. 2012-32. raszewski, sławomir (2012) ‘security and the economics of energy in north east europe’, in kuzemko, caroline, belyi, andrei, goldthau, andreas, keating, michael f. (eds) dynamics of energy governance in europe and russia, london: palgrave macmillan, pp. 130-148. rattsø, jørn, sørensen, rune j. (2012) ‘political control of government enterprises: who controls whom?’, norwegian university of science and technology and norwegian business school. roland, gérard (2000) transition and economics. london and cambridge: mit press. roland, gérard (2001) ‘corporate governance systems and restructuring: the lessons from the transition experience’, in annual bank conference on development economics 2000. washington: world bank, pp. 331–352. rondinelli, dennis a. and jay b. yurkiewicz (1996) ‘privatization and economic restructuring in poland: an assessment of transition policies’, american journal of economics and sociology 55(2):145-160. 64 senderski ● inhibited privatization rosa, jean-jacques and edouard pérard (2010) ‘when to privatize? when to nationalize? a competition for ownership approach’, kyklos 63(1):110-132. rynekinfrastruktury.pl (2012) ‘rząd pracuje nad zmianami w ustawie kominowej.ekspercichcąjejlikwidacji’ [the government plans to update the “chimney bill”. experts demand their abolition], 19 july, url (consulted march 2014): http://www.rynekinfrastruktury.pl/artykul/112/1/rzad-pracuje-nad-zmianami-wustawie-kominowej-eksperci-chca-jej-likwidacji.html. santalainen, timo j., ram b. baliga and jaak leimann (2003) ‘straddling for market space: transforming estonian state-owned enterprises toward a free-market orientation’, international business & economics research journal 2(9):75-87. schindele, ibolya (2003) ‘theory of privatization in eastern europe: literature review’, fondazione eni enrico mattei, working paper no. 2003.2. schoenman, roger (2005) ‘captains or pirates? state-business relations in postsocialist poland’, east european politics and societies 19(1): 40-75. shafik, nemat (1996) ‘selling privatization politically’, international journal of the economics of business 3(3): 367-378. shleifer, andrei and robert w. vishny (1994) ‘politicians and firms’, quarterly journal of economics 109(4): 995-1025. šikulová, ivana and karol frank (2013) ‘the slovak experience with transition to market economy’, ekonomický ústav sav working paper no. 49. singh, nirvikar (2010) ‘the dynamics and status of india’s economic reforms’, mpra paper no. 24479. stark, david (1990) ‘privatization in hungary: from plan to market or from plan to clan’, east european politics and societies 4(3): 351-392. stiglitz, joseph (2002) globalization and its discontents. london: penguin books. tache, ileana (2008) the mass privatization process in romania: a case of failed anglo-saxon capitalism. mimeo: transilvania university of brasov. tirole, jean (1991) ‘privatization in eastern europe: incentives and the economics of transition’, in blanchard, olivier j., fisher, stanley. (eds) nber macroeconomics annual 1991, 6, cambridge: mit press. van de walle, nicolas (1989) ‘privatization in developing countries: a review of the issues’, world development 17(5):601-615. vickers, john and george yarrow (1989) privatisation: an economic analysis. london: mit press. vitkus, gediminas (2009) ‘russian pipeline diplomacy: a lithuanian response’, acta slavica iaponica 26:25-46. wallsten, scott j. (2002) ‘does sequencing matter? regulation and privatization in telecommunications reforms’, development research group, world bank. warner, mildredand and robert hebdon (2001) ‘local government restructuring: privatization and its alternatives’, journal of policy analysis and management 20(2): 315-336. warsaw stock exchange (2011) ‘code of best practice for wse listed companies:appendix to resolution no. 20/1287/2011 of the exchange supervisory board’, 19 october 2011, url (consulted may 2014): http://www.corpgov.gpw.pl/assets/library/english/ regulacje/bestpractices%2019_10_2011_en.pdf. watts, michael, willian b. walstad and alexander skiba (2002) ‘attitudes toward markets and market reforms in the former soviet union and eastern europe’, in 65 pere ● governance and economic development of western balkan countries watts, michael and willian b. walstad (eds.) reforming economics and economics teaching in the transition economies: from marx to markets in the classroom, cheltenham and northampton, ma: edward elgar publishing, pp. 8-34. wu, jianfeng and menita liu cheng (2011) ‘the impact of managerial political connections and quality on government subsidies: evidence from chinese listed firms’, chinese management studies 5(2): 207-226. yusuf, shahid (2009) development economics through the decades: a critical look at thirty years of the world development report. washington, dc: world bank. zinnes, clifford, yiar eilat and jeffrey sachs (2001) ‘the gains from privatization in transition economies: is “change of ownership” enough?’, imf staff papers 48(special issue):146-170. 66 european journal of government and economics 9(3), december 2020, 252-264 252 european journal of government and economics issn: 2254-7088 inequality and female labour force participation in west africa chimere okechukwu iheonu a*, ozoemena stanley nwodo b, uchechi anaduaka c, ugochinyere ekpo c a centre for the study of the economies of africa, abuja, nigeria b coal city university, department of economics, enugu, nigeria c university of nigeria, departments of economics, nsukka, nigeria * corresponding author at: iheonuchimere@yahoo.com abstract. this study examined the impact of income inequality on female labour force participation in west africa for the period 2004 to 2016. the study employed the gini coefficient, the atkinson index and the palma ratio as measures of income inequality. for robustness, the study also utilises female employment and female unemployment as measures of female labour force participation. the study employed the instrumental variable fixed effects model with driscoll and kraay standard errors to account for simultaneity/reverse causality, serial correlation, groupwise heteroskedasticity and crosssectional dependence. the empirical results reveal that the three measures of income inequality significantly reduce the participation of women in the labour force in west africa. the study also revealed that domestic credit, remittances and female education are positively associated with female labour force participation in the sub-region. further findings reveal that economic development reduces the participation of women in the labour force in west africa with the u-shaped feminisation theory not valid for the west african region. the study, however, revealed an inverted u-shaped relationship between inequality and female unemployment. policy recommendations based on these findings are discussed. keywords. female labour; fixed effects; inequality; instrumental variables; west africa jel codes. c23; d31; j21 doi. https://doi.org/10.17979/ejge.2020.9.3.6717 1. introduction achieving food security, poverty alleviation, gender equality, full employment and economic development for all in africa, and particularly west africa, will be a mirage if the challenge of poor female labour participation is not addressed. to achieve the sustainable development goals (sdgs) by 2030, closing the gender gap in the labour force is quite indispensable (united nations, 2018), and explains why researchers are now paying greater attention to female labour force participation. women in west africa, like their counterparts elsewhere, are very resourceful economic agents whose full participation in the economy’s labour force can contribute greatly to economic growth. in many developed economies, women make up almost half of the labour force (world development indicators (wdi), 2020), which could partly be the reason for the fast developmental pace of these economies, and implies that women make many significant contributions to economic development when harnessed. according to the international monetary fund (imf, 2018), increasing women’s engagement in the economy’s labour force leads to significant increases in productivity and impact positively on economic growth. akinyemi, solanke and odimegwu (2018) also observe that female mailto:iheonuchimere@yahoo.com https://doi.org/10.17979/ejge.2020.9.3.6717 iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 253 employment reduces the level of infant and child mortality. despite its positive impact, data from wdi (2020) has consistently revealed that across west african countries, female labour force participation has remained dismal. abney and laya (2018) has noted that there are fewer women in the formal economic sector, which is particularly unfavourable for economic and human prosperity, as economies are unable to reap the trickle down positive spill overs of female labour force participation on other socio-economic indicators. in addition to the low participation in the labour market, women in west africa face significant discriminations. in the world of labour in west africa, patriarchal ideology, educational attainment, household responsibilities and maternal commitment have led to the historic marginalisation of women. these factors automatically transmit to unequal opportunities for men and women. the feminization hypothesis also posits that in the early stages of economic development, female labour force participation declines. this is as a result of the adjustments in the structure of the economy from one which is based on agriculture to one based on industrialisation. however, in the later stages of economic development, with economies transforming into modern economies, fertility rates tend to decline accompanied with the rise in female education, leading to the rise in female labour force participation. this denotes a u-shaped relationship between economic development and female labour force participation. however, one economic fundamental that has not gained much prominence in economic discourse pertaining to female labour force participation in west africa is income inequality. several studies have established links between income inequality and women participation in the labour force. studies such as (gronau, 1982; nelissen, 1990; björklund, 1992; del boca & pasqua 2003; amin & davanzo 2004; western, bloome & percheski 2008; harkness, 2010) found a strong link between income inequality and female employment. these studies reflect that while female employment may reduce income inequality, income inequality can also act as a determining factor which could deteriorate female labour participation. a consistent theory on the effect of gender inequality on employment by klasen and lamanna (2009) is that income inequality changes economic realities negatively and motivates externalities that limit opportunities for women, leading to increase in female unemployment and participation in the labour market. this belief pushed further by asongu and odhiambo (2019) are broadly consistent with the literature on relationships between unemployment, income inequality and economic prosperity (witte & witt, 2001; brush, 2007; odedokun & round, 2001; perugini & martino, 2008; van der hoeven, 2010). furthermore, income inequality also drives poverty which remains a determining factor for not just the participation of women in the labour market but also the participation of men in the labour market. hence, to improve the female labour force participation rate, it becomes necessary to study how income inequality influences the labour force participation of women. there are relatively few studies on income inequality and the economic participation of women in africa. one such study is that of asongu and odhiambo (2019). this study, however, drifts away from their study by (1) accounting for cross-sectional dependence in the modelling exercise, and (2) employing three different measures of female labour force participation to account for iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 254 robustness. this study also concentrates on west africa, a region where poverty is substantially prevalent with a high rate of infant and child mortality. another study that investigated the influence of inequality on female labour force participation in africa is that of idowu and owoeye (2019). their study, however, also failed to account for cross-sectional dependence which, according to iheonu (2019), can lead to estimation bias. the present study whose scope covers 16 west african economies from 2004 to 2016 is novel in several ways. first, the study analysed the relationship between income inequality and female labour force participation in west africa, a region where there is a dearth of studies despite the vulnerable nature of the region, and the need to attain the sdgs. second, the study employed three measures of income inequalitythe gini coefficient, the palma ratio and the atkinson index, and female labour force participation rate to capture female labour participation. for robustness purpose, we utilise female employment and the female unemployment rate as they are measures of the economic participation of women (asongu and odhiambo (2019), (3) the study applied the instrumental variable fixed effects (fe) model to account for simultaneity/reverse causality as well as the driscoll and kraay (1998) standard errors to account for heteroskedasticity, serial correlation and cross sectional dependence. based on the objective of this study, the testable hypothesis is that income inequality has no significant influence on female labour force participation in west africa. the remainder of this study is structured thus: section two presents a review of the related theoretical and empirical literature, section three discusses the methodology and, while section four presents the estimation results and discusses the findings. section five concludes the study and provides policy recommendations. 2. literature review theories on female labour force participation can be traced back to the classical labour theory (clt). between 1960 and 1970, the fundamental speculations on female labour force participation rate rose significantly. the theories include the work-leisure hypothesis of mincer (1962), the human capital theory of becker (1964), and the time allocation hypothesis of becker (1965). the theoretical underpinning of the study, however, follows that of asongu and odhiambo (2019), and klasen and lamanna (2009). this is based on the premise that income inequality alters the workings of the economy and propels negative externalities that limit the opportunities of women. the limited opportunities of women further deteriorate other socio-economic factors which can affect children. empirically, female labour force participation has received some attention in the literature on income inequality. these studies have however been focused on few economies. while there is extensive evidence for the american economy (larrimore 2014, valletta 2006; pencavel 2006; cancian & reed 1999), there is fewer empirical research for the european economies (del boca & pasqua (2003) on italy and breen & salazar (2010) on the united kingdom). there are equally fewer studies in africa (asongu & odhiambo (2019); idowu & owoeye (2019)). asongu and odhiambo (2019) investigated the effect of household income inequality on the iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 255 economic participation of women in sub-saharan africa (ssa) for the period 2004 to 2014 using the generalised method of moment (gmm) methodology which accounts for endogeneity, serial correlation and heteroskedasticity. their result revealed that income inequality (measured as the gini coefficient and palma ratio) reduces female employment and increases female unemployment rate (when income inequality is proxied by the palma ratio). gradin and tarp (2019) examined the gender employment gap in the non-subsistence economic sector in mozambique. their findings show a wider gender employment gap, and the gap is associated with low levels of female human capital and educational attainment. idowu and owoeye (2019) equally investigated the determinants of the demand and supply of female labour force participation rate in 20 selected african economies applying the seemingly unrelated regression estimation technique. their study found that female labour force demand is positively associated with the growth rate of gross domestic product (gdp) and income inequality, and negatively associated with wages, the growth rate of gdp per capita and poverty. moreover, they reported an inverted u-curve relationship between economic growth and female labour supply in africa. the study of alcaino (2009) used the ordinary least square (ols) and the instrumental variable econometric technique to study the determinants of female labour force participation rate in chile for the period of 1854 to 2000. the result revealed that while wages significantly and positively affect female labour participation rate, consumption, life expectancy and fertility are negative determinants of female labour force participation rate. yakubu (2010) found a positive and significant influence of education on female labour force participation rate for south africa. also, forgha and mbella (2016) studied the determinants of female labour force participation in cameroon for 37 years using gmm estimation technique and revealed that fertility rate, dependency ratio, per capita income and male labour force were significantly the determinants of female labour force participation rate in cameroon. verme (2015) used an unbalanced panel data to test the validity of the u-shaped feminisation theory on the relationship between economic development and female labour force participation in the middle east and north africa (mena) for the period 1990 to 2012. the study revealed that there was no clear statistical evidence of a u-shaped linkage between economic development and the female labour participation rate. this finding is contrary to the findings of olivetti (2013), who studied 16 developed american economies. chapman (2015) employed the pooled ols in a panel of 20 mena economies for the period of 1990 to 2012 to test the u-shape relationship between economic development and female labour force participation. the result revealed the existence of a u-shaped relationship between economic development and female labour force participation. this means that in the early stages of economic development, the participation of women in the labour market declines but starts increasing at later stages of economic development. iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 256 3. method and data 3.1. method in understanding the influence of income inequality on female labour force participation in west africa, the study employs the instrumental variable fixed effects model due to the following reasons (1) the instrumental variable fixed effects model is robust to simultaneity/reverse causality – a notable feature in income inequality and female labour force participation relationship, and (2) it accounts for unobserved heterogeneity in the modelling exercise. following the study of efobi et al. (2019), the issue of simultaneity/reverse causality is tackled through the process of instrumenting the indicators of inequality as well as the control variables with their first lags. 𝑋𝑋𝑖𝑖,𝑡𝑡 = 𝛼𝛼0 + 𝛼𝛼𝑗𝑗�𝑋𝑋𝑖𝑖,𝑡𝑡−1� + 𝜀𝜀𝑖𝑖,𝑡𝑡 , (1) where 𝑋𝑋𝑖𝑖,𝑡𝑡 is an explanatory variable (i.e. income inequality indicators and other control variables) in country i at time t. 𝛼𝛼0 is a constant, 𝑋𝑋𝑖𝑖,𝑡𝑡−1 is the first lag of the explanatory variable and 𝜀𝜀𝑖𝑖,𝑡𝑡 is the error term. in obtaining the instruments for the explanatory variables for the fixed effects model, we save the fitted values from the ordinary least square (ols) regression in equation (1). these fitted values are then utilised as instruments for each of the regressors in the model. the instrumentation process is executed by means of employing standard errors which are heteroskedastic and autocorrelation (hac) consistent in nature. the study presents a fixed effects model, where 𝑓𝑓𝑖𝑖,𝑡𝑡 = 𝛽𝛽0 + 𝛽𝛽1𝐼𝐼𝑖𝑖,𝑡𝑡 + 𝑋𝑋𝑖𝑖,𝑡𝑡 + 𝜗𝜗𝑖𝑖 + 𝜇𝜇𝑖𝑖,𝑡𝑡. (2) here, 𝑓𝑓𝑖𝑖,𝑡𝑡 is female labour force participation rate. female labour participation rate captures the ratio of total female labour force to total female population. for robustness checks, we re-define the female labour force participation as female employment and female unemployment rates, respectively. while female employment encompasses the total number of employed females in an economy, female unemployment captures the total number of unemployed females in the economy. 𝐼𝐼𝑖𝑖,𝑡𝑡 represents three indicators of income inequality which includes gini coefficient, atkinson index and the palma ratio. 𝑋𝑋𝑖𝑖,𝑡𝑡 consist of control variables which includes domestic credit, remittances, female secondary school enrolment, a proxy for female education, and per capita gross domestic product (gdp) and its square to analyse the possibility of a u-shaped relationship between economic development and female labour force participation as observed in the literature (altuzarra, galvez-galvez & gonzalez-flores, 2019). domestic credit to the private sector is included in the model based on the notion and verified by verick (2014) that the availability of credit improves the participation of women in the labour market. the study as well includes remittances as a control variable based on extant literature which reveals a linkage between remittances and female labour force participation (khan and valatheeswaran, 2016; azizi, 2018; asiedu and chimbar, 2020). in this study, we posit a positive relationship between iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 257 remittances and the participation of women in the labour market, particularly in the informal sector. we also include female education into the model as education is a significant determinant of female labour force participation (ince, 2010) via the increase in human capital. 𝜗𝜗𝑖𝑖 is country specific effect and 𝜇𝜇𝑖𝑖,𝑡𝑡 is the error term. the estimation of the fixed effects model utilises the driscoll and kraay (1998) standard errors which is robust to serial correlation, groupwise heteroskedasticity, and cross-sectional dependence (see, iheonu, 2019; iheonu et al., 2019). 3.2. data the study employs a panel of 16 west african countries from the period 2004 to 2016. the period captured in the study is due to data availability constraints in the sub-region. the three indicators of female labour force participation are sourced from the world development indicators (2019), while the three indicators of income inequality are sourced from the global consumption and income project (gcip). the control variables would be captured by the following variables: (1) domestic credit to the private sector (% of gdp), (2) remittances (% of gdp), (3) female secondary school enrolment (% gross), (4) per capita gdp (constant us$), and (5) per capita gdp (constant us$) squared. table 1. list of variables and sources. variables definitions source female labour force participation rate labour force participation rate, female (% of female population 15-64), modelled ilo estimate wdi (2019) female employment employment to population ratio, 15+, female (%), modelled ilo estimate wdi (2019) female unemployment unemployment, female (% of female labour force), modelled ilo estimate wdi (2019) gini coefficient gini coefficient gcip atkinson atkinson index gcip palma palma ratio gcip domestic credit domestic credit to the private sector (% of gdp) wdi (2019) remittances remittances (% of gdp) wdi (2019) female education female secondary school enrolment (% gross) wdi (2019) gdp per capita gdp per capita (constant us$) wdi (2019) gdp per capitasquare the square of gdp per capita (constant us$) authors computation source: authors compilation. the gini coefficient is complemented in this study because it does not account for extreme values of the income distribution (asongu & odhiambo, 2019). according to zhang and naceur (2019), it is difficult for the gini coefficient to show welfare of low-and-high income groups. the atkinson and palma ratio controls for extreme values of the income distribution (cobham, schlogl & summer, 2015; asongu & odhiambo, 2019). while the atkinson index essentially validates which end of the income distribution contributes most to income inequality, the palma ratio addresses the gini coefficient’s non-sensitive nature to changes at the top and bottom of the income distribution. for ease of interpretation, the study converts gdp per capita and its square to their natural logarithm. the countries employed in this study are benin, burkina faso, cabo verde, cote d’ivoire, gambia, ghana, guinea, guinea-bissau, liberia, mali, mauritania, niger, nigeria, senegal, sierra leone, and togo. iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 258 4. presentation and analysis of results 4.1. descriptive statistics the descriptive statistics presents the means, the minimum and maximum values, the standard deviation of each of the variables as well as the total number of observations. first, table 2 shows that the dataset is unbalanced. it is also revealed that there is no large difference in the minimum and maximum values of the income inequality indicators, which suggests similarities across space and time. the study, however, reveals substantial differences in the minimum and maximum values of the indicators of labour force participation, which invariably reflects variations across west africa and the observed time period. domestic credit to the private sector and female education exhibits similar characteristics. remittances and the natural logarithm of gdp per capita, however, do not have a substantial difference between the minimum and maximum values which is clearly shown by their standard deviations of 0.89 and 1.48, respectively. table 2. descriptive statistics of the variables. variables observations mean standard deviation minimum maximum gini 207 0.5749 0.0227 0.4882 0.6108 atkinson 207 0.6809 0.0483 0.5098 0.7527 palma 207 5.8497 0.8998 3.0159 7.5243 female labour participation 208 57.9203 12.7308 30.868 82.138 female employment 208 53.4298 13.5246 26.014 79.309 female unemployment 208 6.1215 4.3123 0.219 13.912 domestic credit 207 17.0186 12.7121 1.2480 65.7418 remittances 208 5.8497 0.8998 3.0159 7.5243 female education 129 37.1401 22.1352 6.6145 95.4110 gdp 208 22.6124 1.4818 20.3531 26.8637 source: authors compilation. 4.2. results empirical results in table 3, table 4 and table 5 show the relationship between income inequality and female labour force participation in west africa. the results are divided into three tables, with each table having a different dependent variable to capture female labour force participation. results from table 3 reveal that all three measures of income inequality reduce female labour force participation significantly in west africa. this can be revealed by their probability values which are statistically significant at conventional levels. it is also revealed from table 3 that domestic credit, remittances and female education significantly improves the participation of women in the labour force. the positive relationship between domestic credit and female labour force participation supports the conclusion of field, martinez and pande (2016) where access to credit integrate women into the labour force. however, the remittance-female labour force participation relationship does not correspond with the findings of azizi (2018), who found out that remittances reduce female labour force participation. this disparity in findings could be as a result of the region-specific characteristics of west africa. on the other hand, gdp per capita and the square of gdp per capita significantly reduces female labour force participation in west africa. iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 259 the negative relationship between gdp per capita and female labour force participation is in line with the study of altuzarra, galvez-galvez and gonzalez-flores (2019). in west africa, it is plausible that high female labour force participation falls with economic development and rises with an economic decline. this is based on the intuition that women tend to support their families when the household is faced with financial constraint, particularly in times when the economy is on a decline. however, it is revealed that a further increase in gdp per capita tends to reduce the negative impact of gdp per capita on female labour force participation, as revealed by the coefficient value. this result contradicts the u-shaped hypothesis on the relationship between economic development (as measured by gdp per capita) and female labour force participation. in west africa, later stages of economic development also have a negative consequence on female labour force participation. table 3. instrumental variable fixed effects results (1). panel a: dependent variable: female labour force participation rate variable (1) (2) (3) gini -76.2262*** (0.007) atkinson -34.2202*** (0.003) palma -2.0532*** (0.004) domestic credit 0.0861** (0.012) 0.0786*** (0.012) 0.0852** (0.010) remittances 0.9422*** (0.000) 0.9033*** (0.000) 0.9120*** (0.000) female education 0.3224*** (0.000) 0.3263*** (0.000) 0.3279*** (0.000) gdp -8.3661*** (0.000) -7.8659*** (0.001) -9.0311*** (0.000) gdp2 -3.0105** (0.002) -3.5341*** (0.001) -2.4547 (0.002) constant 417.3367*** (0.000) 410.3331*** (0.000) 374.3462*** (0.000) r2 within 0.4281 0.4281 0.4219 f statistics 247.76*** 234.47*** 348.30*** observations 116 116 116 source: authors compilation. note: *** and ** denotes statistical significance at 1% and 5% respectively. in able 4, the results reveal that the three indicators of income inequality significantly reduce female employment in west africa. this transmits to the fact that income inequality leads to a fall in the participation of women in the labour force. the result supports the finding of asongu and odhiambo (2019). their study revealed that income inequality reduces the participation of women in economic activities. domestic credit, remittances and female education have a positive and significant impact on female employment in west africa. however, gdp per capita and the square of gdp per capita significantly reduces female employment in the sub-region. these results highlight the importance of credit access, remittances and education to female employment. employing the female employment rate as a proxy for female labour force participation supports the findings in table 3, where a u-shaped relationship between gdp and female labour force participation do not exist. instead, it is revealed that the rise in gdp per capita initially has a more pronounced negative consequence on female employment in west africa. however, the negative consequence becomes less pronounced over time. iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 260 table 4. instrumental variable fixed effects results (2). panel b: dependent variable: female employment variable (1) (2) (3) gini -72.1611*** (0.003) atkinson -31.4244*** (0.001) palma -1.9751*** (0.002) domestic credit 0.0789*** (0.009) 0.0713*** (0.009) 0.0784*** (0.006) remittances 0.9134*** (0.001) 0.8668*** (0.000) 0.8902*** (0.000) female education 0.2914*** (0.000) 0.2941*** (0.000) 0.2973*** (0.000) gdp -8.6224*** (0.000) -8.0797*** (0.000) -9.3071*** (0.000) gdp2 -1.5560** (0.019) -1.9499** (0.014) -1.0682** (0.044) constant 348.5136*** (0.000) 334.9005*** (0.000) 311.0345*** (0.000) r2 within 0.3721 0.3690 0.3683 f statistics 41.38*** 42.46*** 47.31*** observations 116 116 116 source: authors compilation. note: *** and ** denotes statistical significance at 1% and 5% respectively. in table 5, female unemployment is utilised as the proxy for female labour force participation. it is revealed that female unemployment is an increasing function of income inequality in west africa. however, we do not see any statistically significant relationship between income inequality and female unemployment rate in west africa. table 5. instrumental variable fixed effects results (3). panel c: dependent variable: female unemployment variables (1) (2) (3) gini 8.3113 (0.151) atkinson 2.4112 (0.493) palma 0.2828 (0.175) domestic credit -0.0041 (0.751) -0.0026 (0.836) -0.0047 (0.725) remittances -0.1505* (0.053) -0.1331** (0.024) -0.1573** (0.014) female education -0.0452 (0.266) -0.0441 (0.257) -0.0469 (0.219) gdp 2.4087* (0.073) 2.2600* (0.080) 2.5847* (0.027) gdp2 -1.1225* (0.056) -1.2037* (0.087) -1.1112* (0.052) constant 2.3031 (0.927) 12.5605 (0.566) 0.9543 (0.956) r2 within 0.0747 0.0726 0.0763 f statistics 93.21*** 87.62*** 40.11*** observations 116 116 116 source: authors compilation. note: *** and ** denotes statistical significance at 1% and 5% respectively. furthermore, it is also revealed that remittances significantly reduce female unemployment while domestic credit and female education reduces female unemployment insignificantly. furthermore, gdp per capita is revealed to significantly increase the female unemployment rate in west africa. however, the square of gdp is seen to significantly reduce the female iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 261 unemployment rate in the region. this result shows the existence of an inverted u-shaped relationship between economic development and female unemployment. earlier stages of economic development results in the increase in the female unemployment rate. however, later stages of economic development eventually result in a fall in the female unemployment rate. 5. conclusion and relevant policy recommendations the study has investigated the impact of income inequality on female labour force participation in west africa utilising three measures of income inequality. the study employed the instrumental variable fixed effects model in order to correct for the possibility of simultaneity/reverse causality and individual heterogeneity. the study finds that the gini coefficient, atkinson index and the palma ratio significantly influences female labour force participation rate and female employment negatively. however, the three indicators of income inequality increase female unemployment rate but not at any level of statistical significance. this result is consistent with the findings of asongu and odhiambo (2019). further findings from the control variable show that domestic credit to the private sector, remittances and female education significantly increases the female labour force participation rate and female employment in west africa. also, remittances and female education reduce female unemployment in west africa. while remittances are revealed to be significant, female education is seen to be insignificant. the result has revealed the importance of migrant remittance to female labour force participation and as such, the importance of removing bottlenecks to enable the smooth flow of remittances into west africa. findings also reveal that gdp per capita and the square of gdp per capita decreases female labour force participation rate and female employment which disproves the u-shaped hypothesis for west africa. we, however, found an inverted u-shaped relationship between gdp per capita and female unemployment, revealing that an initial response to economic development is the rise in female unemployment with a subsequent decline in female unemployment in west africa at later stages of economic development. the study recommends the following based on findings, (1) there is an essential need for the adoption of inequality reducing policies in west africa. the governments in the sub-region should also invest in sectors where women have a greater likelihood of participation. there is also need for quality infrastructure that drives economic activities as well as the adoption of a tax system that is progressive in nature to reduce income inequality, (2) it is necessary that the sub-region concentrates in developing the financial sector to improve access to credit. this can be achieved through a significant reduction in interest rates in financial institutions to enable borrowing, (3) the development of the financial sector would also reduce bottlenecks which would aid the increase in remittances flow into the west african subregion, (4) human capital development in the form of female education should also be prioritised in west africa. the positive linkage between female education and female labour force participation should be a focal point for west african countries on the need to pursue the girl child education. finally, the study also reflects on the importance of changing the socio-cultural norms in africa, where many women participate more in the labour market only when economic activities iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 262 are declining in order to support their spouses. as highlighted, in periods of rising per capita gdp, women draw out of the labour market and tend to cater more for their husbands and children leaving their husbands to cater for the family. conclusively, the feminisation theory fails to hold in west africa. references abney, d., and laya, a. (2018). this is why women must play a greater role in the global economy. retrieved from world economic forum: https://www.weforum.org/agenda/2018/ 01/this-is-why-women-must-play-a-greater-role-in-the-global-economy/ akinyemi, j., solanke, b., and odimegwu, c. (2018). maternal employment and child survival during the era of sustainable development goals: insights from proportional hazards modelling of nigeria birth history data. annals of global health, 84(1), 15-30. https://doi.org/10.29024/aogh.11 altuzarra, a., galvez-galvez, c., and gonzalez-flores, a. (2019). economic development and female labour force participation: the case of european union countries. sustainability, 11. https://doi.org/10.3390/su11071962 amin, s., and davanzo, j. (2004). the impact of wives' earnings on earnings inequality among married-couple households in malaysia. journal of asian economies, 15, 49-70. https://doi.org/10.1016/j.asieco.2003.12.002 asiedu, e., and chimbar, n. (2020). impact of remittances on male and female labour force participation patterns in africa: quasi-experimental evidence from ghana. review of development economics. https://doi.org/10.1111/rode.12668 asongu, s., and odhiambo, n. (2019). inequality and the economic participation of women in sub saharan africa: an empirical investigation. working papers of the african governance and development institute 19/027. https://doi.org/10.1108/ajems-01-2019-0016 azizi, s. (2018). the impact of workers' remittances on human capital and labour supply in developing countries. economic modelling, 75, 377-396. https://doi.org/10.1016/j.econmod .2018.07.011 becker, g. (1964). human capital: a theoretical and empirical analysis, with special reference to education. university of illinois at urbana-champaign's academy for entrepreneurial leadership historical research reference in entrepreneurship. becker, g. (1965). a theory of the allocation of time. the economic journal, 493-517. https://doi.org/10.2307/2228949 breen, r., and salazar, l. (2010). has increased women's educational attainment led to greater earnings inequality in the united kingdom? a multivariate decomposition analysis. european sociological review, 26(2), 143-157. https://doi.org/10.1093/esr/jcp01 brush, j. (2007). does income inequality lead to more crime? a comparison of cross-sectional and time-series analyses of united states countries. economics letters, 96(2), 264-268. https://doi.org/10.1016/j.econlet.2007.01.012 burtless, g. (1999). effects of growing wage disparities and changing family composition on the us income distribution. european economic review, 43, 853-865. https://doi.org/10.1016/s0014-2921(98)00099-3 chapman, k. (2015). economic development and female labour force participation in the middle east and north africa: a test of the 'u' shape hypothesis. gettysburg economic review, 8(3), 5-22. cobham, a., schlogl, l., and summer, a. (2015). inequality and the tails: the palma proposition and ratio revisited. united nations, department of economics and social affairs working papers 143. cox alcaino, m. l. (2009). women's labour force participation in chile, 1854-2000. pontificia universidad catolica de chile, instituto de sociologia. retrieved from https://polisci.mit.edu /files/ps/imce/gradstudents/papers/cox.labor.force.pdf https://polisci.mit.edu/files/ps/imce/gradstudents/papers/cox.labor.force.pdf https://polisci.mit.edu/files/ps/imce/gradstudents/papers/cox.labor.force.pdf iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 263 del boca, d., and pasqua, s. (2003). employment patterns of husbands and wives and family income distribution in italy (1977-98). review of income and wealth, 49(2), 221-245. https://doi.org/10.1111/1475-4991.00084 driscoll, j., and kraay, a. (1998). consistent covariance matrix estimation with spatial dependent data. review of economic and statistics, 80, 549-560. https://doi.org/10.1162/ 003465398557825 efobi, u., asongu, s., okafor, c., tchamyou, v., and tanankem, b. (2019). journal of multinational remittances, finance and industrialisation in africa. journal of multinational financial management, 49, 54-66. https://doi.org/10.1016/j.mulfin.2019.02.002 elveren, a., and elgin, c. (2019). informality, inequality and feminization of labour. political economy research institute working paper. esteve-volart, b. (n.d.). gender discrimination and growth: theory and evidence from india. sticerd discussion papers no. dedps42. field, e., martinez, j., and pande, r. (2016). does women's anking matter for women? evidence from urban india. international growth centre working paper s-35306-inc-1. forgha, n., and mbella, m. (2016). implication of labour force participation rate on economic growth in cameroon. international journal of development and economic sustainability, 4(1), 34-47. fund, i. m. (2018, may 31). pursuing women's economic empowerment. retrieved from https://www.imf.org/en/publications/policy-papers/issues/2018/05/31/pp053118 pursuing-womens-economic-empowerment gradin, c., and tarp, f. (2019). gender inequality in employment in mozambique. south african journal of economics, 87(2), 180-199. https://doi.org/10.1111/saje.12220 gronau, r. (1982). inequality of family income: do wives' earnings matter? population and development review, 8, 119-136. https://doi.org/10.2307/2808109 harkness, s. (2010). the contribution of women's employment and earnings to household income inequality: a cross-country analysis. lis working paper series. idowu, o., and owoeye, t. (2019). the supply of female labour force participation in selected african countries. business & social sciences journal, 14-30. iheonu, c. (2019). governance and domestic investment in africa. european journal of government and economics, 8(1), 63-80. https://doi.org/10.17979/ejge.2019.8.1.4565 iheonu, c., agbutun, s., omenihu, c., ihedimma, g., and osuagwu, v. (2019). the impact of governance quality on mortality rates in sub saharan africa. african population studies, 33(1), 4655-4668. https://doi.org/10.11564/33-1-1353 ince, m. (2010). how the education affects female labour force? empirical evidence from turkey. procedia social and behavioral sciences, 2(2), 634-639. https://doi.org/10.1016/ j.sbspro.2010.03.076 khan, m., and valatheeswaran, c. (2016). international migration, remittances and labour force participation of left-behind family members: a study of kerala. margin: the journal of applied economic research, 10(1), 86-118. https://doi.org/10.1177/0973801015612669 klasen, s., and lamanna, f. (2009). the impact of gender inequality in education and employment on economic growth: new evidence for a panel of countries. feminist economics, 15(3), 91-132. https://doi.org/10.1080/13545700902893106 larrimore, j. (2014). accounting for united states household income inequality trends: the changing importance of hosehold structure and male and female labour earnings inequality. review of income and wealth, 60(4). https://doi.org/10.1111/roiw.12043 mincer, j. (1962). labour force participation of married women: a study of labour supply. in aspects of labour economics (pp. 63-105). princeton university press. nelissen, j. (1990). the effect of increased labour force participation of married women on the distribution of family income in the netherlands. de economist, 138, 47-62. https://doi.org/10.1007/bf01718389 odedokun, m., and round, j. (n.d.). determinants of income inequality and its effects on economic growth: evidence from african countries. unu/wider working paper no. 103. pencavel, j. (2006). a life cycle perspective on changes in earnings inequality among married men and women. the review of economics and statistics, 88(2), 232-242. https://doi.org/10.1162/rest.88.2.232 https://doi.org/10.1016/j.mulfin.2019.02.002 https://www.imf.org/en/publications/policy-papers/issues/2018/05/31/pp053118pursuing-womens-economic-empowerment https://www.imf.org/en/publications/policy-papers/issues/2018/05/31/pp053118pursuing-womens-economic-empowerment https://doi.org/10.1111/saje.12220 https://doi.org/10.2307/2808109 https://doi.org/10.17979/ejge.2019.8.1.4565 https://doi.org/10.11564/33-1-1353 https://doi.org/10.1016/j.iref.2018.12.015 https://doi.org/10.1016/j.iref.2018.12.015 https://doi.org/10.1080/13545700902893106 https://doi.org/10.1111/roiw.12043 https://doi.org/10.1007/bf01718389 https://doi.org/10.1162/rest.88.2.232 iheonu et al. / european journal of government and economics 9(3), december 2020, 252-264 264 perugini, c., and martino, g. (2008). income inequality within european regions: determinants and effects on growth. the review of income and wealth, 54(3), 373-406. https://doi.org/10.1111/j.1475-4991.2008.00280.x united nations. (2018). leave no one behind: a call to action for gender equality and women’s economic empowerment (un secretary general’s high-level panel on women’s economic empowerment) . retrieved from https://www.empowerwomen.org/-/media/files/un%20women /empowerwomen/resources/hlp%20briefs/unhlp%20full%20report.pdf?la=en&vs=2916 valletta, r. (2006). the ins and outs of poverty in advanced economies: government policy and poverty dynamics in canada, germany, great britain, and the united states. review of income and wealth, 52(2), 261-284. https://doi.org/10.1111/j.1475-4991.2006.00187.x van der hoeven, r. (2010). income inequality and employment revisited: can one make sense of economic policy? journal of human development and capibilities, 11(1), 67-84. https://doi.org/10.1080/19452820903481459 verick, s. (2014). female labour force participation in developing countries. iza world of labour, 37. https://doi.org/10.15185/izawol.87 verme, p. (2015). economic development and female labour force participation in middle east and north africa: a test of the 'u' shape hypothesis. iza journal of labour and development, 4(3), 1-21. https://doi.org/10.1186/s40175-014-0025-z western, b., bloome, d., and percheski, c. (2008). inequality among american families with children, 1975 to 2005. american sociological review, 73(6). https://doi.org/ 10.1177/000312240807300602 witte, a., and witt, r. (2001). crime causation: economic theories. encyclopaedia of crime and justice. new york: macmillan. yakubu, y. (2010). factors influencing female labour force participation in south africa in 2008. the african statistical journal, 11, 85-104. zhang, r., and ben naceur, s. (2019). financial development, inequality, and poverty: some international evidence. international review of economics & finance, 61(c), 1-16. https://doi.org/10.1016/j.iref.2018.12.015 https://doi.org/10.1111/j.1475-4991.2008.00280.x https://doi.org/10.1080/19452820903481459 https://doi.org/10.15185/izawol.87 https://doi.org/10.1186/s40175-014-0025-z https://doi.org/10.1016/j.iref.2018.12.015 https://doi.org/10.1016/j.iref.2018.12.015 https://doi.org/10.1016/j.iref.2018.12.015 corruption and growth: evidence from the italian regions european journal of government and economics volume 1, number 2 (december 2012) issn: 2254-7088 126 corruption and growth: evidence from the italian regions nadia fiorino, university of l’aquila, italy emma galli, ‘la sapienza’ university of rome, italy ilaria petrarca, university of verona, italy * abstract this paper investigates the impact of corruption on economic growth in the italian regions. we estimate a dynamic growth model for the period 1980-2004 addressing both the potential bias of the measures of corruption and the endogeneity between corruption and economic development. we find strong evidence of a negative correlation between corruption and growth. moreover, since government intervention has been traditionally used to reduce income differentials between the northern and the southern regions, we also analyse the interaction between corruption and government expenditure. our results indicate that corruption undermines the positive impact that public expenditures have on economic growth. jel classification h11; o40; o38 keywords corruption; associative crimes; economic growth; dynamic estimation * address for correspondence: ilaria.petrarca@univr.it. european journal of government and economics 1(2) 127 introduction does corruption ‘sand’ or ‘grease the wheels’ of economic development? for a long time, scholars have investigated the economic consequences of corruption, drawing an ambivalent picture: on the one hand, corruption promotes investments that would have been otherwise stalled by regulations and bureaucratic procedures; on the other, it reduces the incentives to invest in productive activities. therefore, the effect of corruption on growth remains an empirical question. we contribute to this debate by estimating the effect of corruption on economic growth in a panel dataset for the 20 italian regions during the period 1980-2004 to verify whether corruption played a role in the differentiated growth path of southern italy. italy is an interesting case in this perspective because regional inequalities still persist although different kinds of public policies have tried to reduce the per capita income differentials between the northern and the southern regions since the end of world war ii (see padovano, 2007). the distribution of corruption and of social capital across the country is not homogeneous even though the institutions and the policies aimed at punishing and preventing corruption are centralized at the national level, and this may contribute to explain the differences in the economic growth rates of the italian regions. so far the literature has not paid much attention to the italian case, mainly due to the low quality of available data until the very recent years. del monte and papagni (2001) investigated the link between corruption and economic growth in the italian regions for the period 1963-1991. they show that the efficiency of public investments is lower in regions where corruption is higher which in turn negatively affects economic growth. we re-address this issue for a more recent period of the italian history (1980-2004) characterised by high variability in both growth rates and corruption crimes. we use a newly assembled dataset which has the advantage of collecting economic, socio-demographic and politico-institutional variables at regional level and provide a methodological solution to the shortcomings of the commonly used measures of corruption. we also move a step forward with respect to the existing literature by using an estimation technique which reduces the endogeneity bias of the coefficients. the rest of the paper is organized as follows. section 2 reviews the literature on the nexus between corruption and economic growth; section 3 describes the institutional context of the italian regions and shows the time dynamics of our key variables, namely economic growth and corruption. in section 4 we formulate our empirical strategy, present the empirical model and discuss the results. the last section provides some concluding remarks. literature review many economists consider corruption as a major obstacle to economic growth (see, among others, myrdal, 1989; andvig and moene, 1990; shleifer and vishny, 1993; blackburn et al., 2006). the main argument is that a government official that controls the supply of an individually-demanded service may abuse his arbitrary power to restrict the supply, for example, denying or delaying permissions. the elimination of these barriers requires an extra-price of the service, i.e. a bribe that increases the bureaucrats' rent. nonetheless, the bribe removes also any incentive to invest and defines a sub-optimal rent-seeking equilibrium of human capital that hampers growth. mauro (1995), keefer and knack (1997), hall and jones (1999), la porta et al. (1999), li et al. (2000) and gyimah brempong (2002) estimate the impact of corruption on growth for a wide cross section of countries. they verify that higher levels of corruption significantly reduce both investment and economic growth. interestingly, larger levels of corruption are associated to a misallocation of public resources. mauro (1998), tanzi and davoodi (1998), gupta et al. (2001), baldacci et al. (2004) find that corruption distorts the composition of government european journal of government and economics 1(2) 128 expenditure towards less productive activities and creates large public sectors where resources are wasted through rent seeking. some scholars, on the other hand, argue that corruption is not a totally inefficient activity because in the short run it solves some government failures (leff , 1964; huntinghton, 1968). this ‘greasing the wheels’ hypothesis, however, limits the growth-enhancing effect of corruption to those situations where governance is lacking and/or economic policy is inefficient. lui (1985) formalizes this argument into a model where firms value the time to waste in a queue. the more productive firms want to waste as little time as possible in the queue; therefore they are willing to buy the priority of their activities by paying a bribe. along this line of research shleifer and vishny (1994) develop a model of bargaining between politicians and enterprises and show that corruption can facilitate an efficient allocation of resources. this is because bribes are a way to distribute wealth between politicians and agents in the private sector. in other words, corruption increases efficiency by allowing private sector agents to buy their way out of the inefficiencies that would otherwise be introduced by the politicians. in this perspective, the existence of a negative linear relationship between corruption and growth is challenged in favour of a non-linear one, which predicts a positive growth effect at low levels of corruption incidence (see, among others, kurrild-klitgaard, 1988 and acemoglu and verdier, 1998). the ‘greasing the wheels hypothesis’ also implies that the relationship between corruption and economic growth can be affected by both the quality of the institutions and the size of the public sector. ehrlich and lui (1999) develop an endogenous growth model that analyzes the effect of corruption on economic growth in different politico-institutional settings. they predict that the balanced growth in a democracy (or competitive regime) and in an autocracy (or monopolistic regime) is the outcome of interaction between accumulation of human capital (socially productive) which engenders growth, and accumulation of political capital (socially unproductive) which mainly assures bureaucratic power and potential corruption. a non-linear relationship between corruption and growth is empirically found only in democratic regimes. méndez and sepulveda (2006) distinguish between ‘free’ and ‘not-free’ countries and include a measure of government expenditures to capture its interaction with corruption. their findings show that in ‘free’ countries corruption results beneficial for economic growth at low levels of incidence and detrimental at high levels of incidence. this relationship is not modified by the size of government. economic growth and corruption in the italian regions: some stylized facts the institutional framework italy is divided into 20 regions, that represent the upper tier of sub-state government. five of them, established in the years between 1948 and 1963, enjoy a special statute (regioni a statuto speciale, or rss) because of their multilingual status and peculiar geographical and economic position. the other 15 regions, featuring an ordinary statute (regioni a statuto ordinario, or rso), were established in the 1970s. until mid-1990s, however, the regions heavily depended on the central government. in particular, they have expenditure autonomy but lacked tax autonomy; regional resources were represented by transfers from national taxes and grants from the central government, whose amounts were not modifiable by the regions. since the 1990s, several legislative and constitutional reforms changed the institutional framework and increased both tax autonomy and expenditure competences. currently regional governments levy taxes of their own (about 24% of total national fiscal revenue), as well as shares of national taxes and transfers (about 53%) (giardina et al, 2009), and are responsible for health care european journal of government and economics 1(2) 129 expenditure plus a share of social services, education, environment, local transportation, housing, culture and tourism. differences in competences between the rso and rss have been reduced. economic growth, corruption and associative crimes in the italian regions figure 1: time dynamics of gdp growth and per capita corruption crimes yearly data are averaged over the full sample. gdp growth is measured as the percentage change from the past years' gdp growth; crimes are measured as per capita prosecutions for corruption crimes. source: italian institute of statistics (istat). figure 1 illustrates an opposite trend in gdp growth and per capita prosecutions for corruption crimes. overall economic growth decreased in italy, with a significant fall in the early 90s. the growth rates always were under 5% after 1997 while corruption crimes increase steadily between the mid-1970s and the first half of the 1990s and slightly decrease after 1993 as a consequence of the so-called mani pulite (clean hands) campaign undertaken by the judicial system. european journal of government and economics 1(2) 130 figure 2. gdp growth (average annual data) north: piedmont, val d'aosta, lombardy, trentino-alto adige, veneto, friuli-venezia giulia, liguria and emilia romagna; centre: tuscany, umbria, marche, latium; south: abruzzo, molise, apulia, basilicata, sardinia. figure 2 shows that the level of gdp growth is quite homogeneous across the macro-areas, and displays a decreasing pattern during the period 1980 1995.1 within the south an important difference between the regions characterized by a pervasive presence of criminal organizations and the others emerges, which seems to account for 10%. 1 recently daniele and malanima (2007) describe the mezzogiorno gap in gdp per capita between 1861 and 2004. while uniformity characterizes the pre-industrial period, since the 1880s a long divergence phase starts between the industrial areas and those which were not able to create a manufacturing industry. this phase ends in 1951 when the gdp per capita in the southern regions was only 47% of that of the rest of italy, then the convergence process continued until the first half of the 1970s when the relative gdp per capita reached about 66%. during the 1980s a new phase of a divergence process begins until 2002. european journal of government and economics 1(2) 131 figure 3. corruption crimes (average annual data) note: north: piedmont, valle d'aosta, lombardy, trentino-alto adige, veneto, friuli-venezia giulia, liguria and emilia romagna; center: tuscany, umbria, marche, latium; south: abruzzo, molise, apulia, basilicata, sardinia. figure 3 shows that the number of corruption crimes decreases in the 1980s; a significant increase is recorded from 1991 to 1998, where the number of crimes jumps from around 600 to around 2400; finally, after a decrease in 2000, it increased again. sicily and campania appear as the most corrupted regions of italy, followed respectively by the northern, the central and the southern ones. european journal of government and economics 1(2) 132 table 1. gdp and prosecutions per capita (1980-2004, average annual data) region gdp pc region associative crimes region corruption crimes valle d'aosta 21.147 sicily 42.1 latium 961 trentino-alto adige 20.159 calabria 31.1 molise 901.3 lombardy 19.715 campania 30.6 valle d'aosta 787.9 emiliaromagna 19.030 apulia 21.4 liguria 775.1 piedmontpied mont 17.53 basilicata 18 calabria 699.8 veneto 17.22 liguria 15.6 sicily 622 tuscany 16.54 latium 15.2 sardinia 617.2 friuli-venezia giulia 16.1875 abruzzo 13.2 friuli-venezia giulia 607.7 latium 16.07 friuli-venezia giulia 13.1 abruzzo 601.2 liguria 15.79 emiliaromagna 12.7 campania 558.2 umbria 14.79 umbria 12.6 basilicata 494 marche 14.47 trentino-alto adige 11.7 tuscany 494 abruzzo 12.54 lombardy 11.6 apulia 483.6 sardinia 11.44 valle d'aosta 11.4 trentino-alto adige 459.8 molise 11.001 molise 11.1 umbria 456.7 sicily 10.42 veneto 10.9 piedmont 426.3 basilicata 9.68 tuscany 10.8 marche 393.8 apulia 9.61 piedmont 10 lombardy 375.3 campania 9.56 marche 8 veneto 374.6 calabria 8.71 sardinia 6.8 emiliaromagna 349.5 note: data are reported by descending order; per capita gdp is measured in thousands of euro; per capita crimes are measured per million of inhabitants. source: italian institute of statistics (istat) overall, gdp growth as well as corruption crimes change both over time and across regions. table 1 provides a ranking of the regions according to their average gdp and corruption crimes per capita. using prosecution data as a measure for corruption bumps against the circumstance that in corrupt regions the judicial system is itself corrupt and fewer people will be charged with corrupt practices. the effectiveness of a legal system is rooted not only in the formulation of laws but also in the ‘legal culture’, that is the expectations and practices that inform the way they are enforced (treisman, 2000). although the legal system is the same in all the italian regions, its degree of legitimacy is not. moreover, such a measure reflects only the ‘revealed’ corruption, most likely by leaving part of the phenomenon hidden. table 1 highlights this problem. the northern regions are less corrupt than the central and southern regions; however, the ranking is not completely in line with people’s common sense about the real distribution of corruption in italy. indeed, prosecutions for corrupt practices in the appeal court district of reggio calabria, one of the major towns of calabria, in the last twenty years resulted in two convictions only. nevertheless, similar conditions characterize the districts of other ‘perceived’ corrupt regions, like sicily, campania and sardinia (davigo and mannozzi, 2007). to take into account the hidden corruption and avoid potential bias between official statistics and ‘true’ data, we consider the existing link between corruption and associative crimes (crimes ex art. 416 and 416 bis of the italian criminal law). this european journal of government and economics 1(2) 133 implies that, as the so-called mani pulite criminal trials confirmed, corruption emerges not only as corrupt practices but also as associative crimes in the most ‘perceived’ corrupt regions. empirics empirical strategy and model specification there are several issues related with the characteristics of the dataset that give reason for the choice of the estimator. first, the literature on corruption and growth generally estimates cross sectional regressions by averaging the effect of temporary shocks and smoothing the cycling pattern of gdp. although apparently straightforward, cross-country analyses make the implicit assumptions that countries are positioned on their steady state equilibria values for both the level of corruption and growth rate of output. hence, averaging out data into a single observation for each region involves a loss of information and may also distort the analysis of the relationship between the two. since the gdp growth rate as well as the corruption rate are not homogeneously distributed across regions and change over time, we employ a dynamic panel regression accounting also for the time variability of the data. secondly, our dependent variable, gdp growth, reasonably follows an autoregressive trend and requires a dynamic specification. the small size of the dataset, limited to at most 20 observations, reduces the efficiency of the gmm estimators (arellano and bond, 1991; blundell and bond, 1998) and requires a correction (bruno, 2005a and 200b). another concern is related to data on prosecutions against corruption. this measure is problematic because a low rate of prosecutions may signal either a low effective incidence of corruption or a widespread hidden corruption. moreover, it hits against the circumstance that in corrupt regions the judicial system may be itself corrupt and consequently fewer people would be charged with corruption crimes. further, potential endogeneity may affect the relationship between corruption and growth. an extensive literature originated by lipset (1960) considers that low levels of income generally determine corruption, i.e. less developed regions are endemically more corrupt. if this is the case, corruption would be correlated with the error term in the ols regression and the estimates would be biased. to control for the problem of a two-way causality between corruption and growth, we estimate an equation of determinants of corruption through ols and insert the estimated fitted values of the parameters of interest in our dynamic growth equation (kelejian, 1971; petterson-lidblom and dahlberg, 2003).2 this procedure also allows us to take into account possible problems related to the quality of the legal system and the under-reporting. indeed, the corruption equation controls for the degree of social capital which captures the existence of regional differences in people’s general attitude towards corruption. based on these considerations we develop a two-step procedure. the first step estimates an equation of the determinants of corruption (see on this issue fiorino and galli, 2010). the fitted values obtained from this equation reduce the potential bias due to the heterogeneous distribution of social capital and quality of the legal system at regional level; moreover, the ols estimation of this equation allows for 2 many authors have also worked with five-year averages for similar purposes. the use of five-year averages reduces short run fluctuations and allows to concentrate on the relationships between corruption and growth. see, for example, li et al. (2000), paldam (2002), glaser and saks (2006), méndez and sepulveda (2006). european journal of government and economics 1(2) 134 removing endogeneity in the growth model.3 the second step estimates a growth equation and substitutes the corruption crimes with the fitted values obtained in first step. to control for the dynamic bias induced we follow the literature and apply a least squares dummy variables corrected model (lsdvc, bruno 2005a and 2005b). this estimator corrects the lsdv estimator4 for the small size of the sample and provides a significant reduction of the bias, performing as well as the gmm estimator properly identified.5 a slight limitation of this methodology is the requirement of exogenous right hand side variables. sample data and description of variables our dataset collects economic, socio-demographic and politico-institutional variables for the 20 italian regions during the period 1980-2004 and consists of 500 observations. the source, if not differently specified, is the italian institute of statistics (istat). the variables are summarized in table 2. 3 petterson-lidblom and dahlberg (2003) show in fact that the use of the ols estimates in the first stage allows us to obtain consistent estimates of the parameters of interest without the need to resort to the full blown functional form of the first stage. 4 that is an ols regression including regional dummies. we estimated also a set of lsdv uncorrected for the dynamic bias and obtain results consistent with the ones presented. 5 an alternative dynamic estimator is the system gmm developed by blundell and bond (1998). this estimator adds to the equation in levels instrumented with differences (arellano and bond, 1991), a second equation in differences, instrumented with the variables in levels. when the gmm estimators are applied to small samples, the number of instruments outnumbers the observations, and overfit the lagged dependent variable. instrument proliferation generates false positive results, reducing the reliability of the estimates. roodman (2009) proposes to collapse the matrix of instruments to decrease their number. a rule of the thumb, however, suggests that to obtain robust test statistics one needs a number of instruments not larger than the number of groups. the gmm estimation of equation 2 never satisfies the rule of thumb, as in the most parsimonious specification we use 23 instruments for 19 groups. european journal of government and economics 1(2) 135 table 2. descriptive statistics obs mean std. dev. min max gdp growth 484 0.084 0.051 -0.003 0.257 variables of interest: corruption cor, individual crimes per pop(*1000) 500 0.572 0.308 0.10 1.939 ascr, associative crimes per pop(*1000) 520 0.016 0.015 0 0.226 expenditure/gdp 456 0.46 6.127 0.003 130.992 current expenditure/gdp 444 -2.507 0.359 -3.355 -1.306 capital expenditure/gdp 444 -3.719 1.03 -5.870 -1.199 investments/gdp lag 460 0.067 0.024 0.029 0.197 public consumption/gd p 480 0.217 0.057 0.118 0.358 gini index 440 0.33 0.035 0.236 0.479 school attainment 454 0.046 0.007 0.02 0.065 labor force units 500 1150.13 971.99 56.8 4508.7 population 576 2874227 2259821 112262 9742676 number of laws 460 50.25 26.40 3 165 fragmentation 460 0.668 0.13 0.128 0.880 voluntary organizations 460 449 555 11 5362 diffusion of newspapers 580 231811 221249 5687 1098279 referendum voters 460 1428645 1283230 31059 6177641 our first step consists in estimating an equation of the determinants of corruption. the equation is defined as follows: [1] corruptionit = f(xit, eit) for i=1,...,20 and t=1980,...,2004, where x is a vector of explanatory variables and e is the error term. we measure corruption in two different ways: 1) the number of regional government officials prosecuted for corrupt practices relative to the population (cor). the crimes that we consider are based on the libro ii, titolo ii (crimes against the public administration) of the italian criminal law as reported in the annali di statistiche giudiziarie of the istat (various issues). 2) a composite index annually computed per each region as the sum of per capita prosecutions and per capita associative crimes (cor+ascr). this because the most important criminal trials against corruption in italy (the so-called mani pulite and maxi trial of the sicilian mafia) have confirmed that corrupt activities may emerge also in the form of other typologies of crimes like the associative crimes (crimes ex art. 416 and 416 bis of the italian criminal law) that cannot be strictly considered crimes of corruption. the explanatory variables are: european journal of government and economics 1(2) 136 a) population in millions of inhabitants, a proxy for the size of the region. if highlypopulated regions exploit economies of scale in supplying of public goods (alesina and wacziarg, 1997) and have a low ratio of public service outlets per population, individuals might revert to bribes to 'get ahead of the queue'. b) gdp growth is the annual growth of gdp per capita calculated starting from the yearly gdp data released by crenos (2004). the growth is defined as the ratio between the first difference and the lagged gdp, representing the percentage of shift from the previous year's aggregate output. thus, gdpgrowthit = (gdpit gdpi,t-1) / gdpi,t-1. data are in constant terms. this variable, as well as education, is included to investigate the so-called lipset hypothesis: voters with higher income (and education) are expected to be both more willing and capable to monitor public employees and to take action when the latter violate the law. then, we expect a negative sign associated to both the coefficients of these variables. c) school attainment is a proxy for the level of education in the regions, measured as the share of high school enrolment over labour force. d) gini index is the regional inequality level, built using micro-data on the households’ disposable income. these data come from the survey of household income and wealth (shiw) conducted by the bank of italy (several years). the rationale is that as voters become more diverse along the income line, they will focus on redistribution rather than on the honesty of government officials (mauro, 1995; alesina et al., 1996). we then expect that an increase in income inequality will positively affect the degree of corruption. e) current expenditure/gdp and capital expenditure/gdp capture the role of government size on corruption (and indirectly on growth).6 a larger government size may generate a potential for corruption by producing more resources to be stolen and more rules to be exploited or subverted. this can be the case for the italian regions where the public sector plays a quite relevant role in the economy. corruption alters the composition of government expenditures towards less productive activities and thus the greater the government expenditures the greater the negative effects of corruption (mauro, 1998, tanzi and davoodi, 1998; gupta et al., 2001). f) number of laws enacted by the region is an alternative way to capture the impact of government size on corruption (weingast et al., 1981). g) fragmentation of regional government. when governments consist of large coalitions characterized by a certain number of parties with conflicting interests, the members of the coalition face a prisoner’s dilemma with respect to expenditures decisions. each of the partners within the coalition has different distributional objectives and consequently an incentive to protect the budget share which may favor their own clientele (roubini and sachs, 1989a; 1989b). political fragmentation may then increase the distribution of rents among politicians and engender a higher level of corruption. the use of this variable is also suggested by a change of the regional electoral system that occurred in 1995. the mechanism by which the members of the regional council are elected switched from a pure proportional representation to a mixed one. a top-up number of seats for the winning coalition is also introduced, so that the absolute majority of the legislators will be held by the coalition linked to the regional list that has obtained the relative majority of the votes. furthermore, the law reduced the tenure length of the council from five to two years if the relationship of confidence between the council and the regional government breaks down during the first two years. this reform was completed in 1999 when it was established that the president of the region is 6 since until the late 1990s regional expenditure was financed through transfers from the central government, we do not include transfers among the independent variables. nonetheless, we estimated a set of regressions with the inclusion of this variable and the results do not change. european journal of government and economics 1(2) 137 elected by universal and direct suffrage. we measure government fragmentation with the herfindahl index for concentration. the index is built by using the seats of the majority supporting the regional government with respect to the overall legislature and ranges from 0 (a legislature in which each legislator belongs to a different party) to 1 (when all members belong to the same party).7 data come from the ministero dell’interno. on this variable we expect a negative coefficient. h) the share of voluntary organizations over the population, the local diffusion of newspapers and the share of voters that participate in referendums on the total of voters, referendum voters. these variables proxy the degree of civicness of italian regions as propensity of citizens to be politically involved and as general attitude towards corrupt practices. these three variables control for the degree of corruption generally ‘accepted and tolerated’ in each regional environment. fivol (federazione italiana per il volontariato) is the source of data on voluntary organizations and the ministero dell’interno on the referendum voters. data on the diffusion of newspapers come from ads (agenzia diffusione stampa). the second step consists in estimating the following growth equation: [2] gdpgrowthit = h(gdpgrowthit-1,yit, corruption_fit ,uit) for i=1,...,20 and t=1980,...,2004, where y is a vector of standard economic growth models explanatory variables and u is the error term. equation [2] substitutes the variables on corruption with the fitted values obtained in step 1, cor_f and cor+ascr_f. the dependent variable of equation [2] is gdpgrowth, the annual growth of gdp per capita as already defined. as measure of corruption we use the fitted values estimated in equation [1] (corruption_f). the vector y includes a number of socio-economic and socio-demographic variables: i) the lagged growth level, gdpgrowthit-1. the coefficient of this variable indicates the average regional trend of growth, that is not predictable a priori. j) fixed gross public and private investments, investments/gdp. to avoid reverse causality, we introduce this variable with a one year lag and expect a positive correlation with the gdp. k) public consumption/gdp and private consumption/gdp are expected to be negatively correlated with economic growth. l) expenditure/gdp is the total expenditure over the gdp, is expected to be positively correlated with economic growth, if productive. m) the first difference of the gini index to capture the effect of a variation of inequality on growth. an unequal distribution of income is a barrier to growth because it generates a pressure to adopt redistributive policies that have an adverse effect on investment (persson and tabellini, 1994); present wealth, moreover, may depend on past wealth. therefore, the more unequal a region is, the lower its growth rate. 7 to calculate this index, we sum the seats of each party i of the majority, calculate the percentage s that these represent on the total number of seats of the council and compute the herfindahl index:   n i sionfragmentat 1 2 , where n is the total number of seats of the council. we then use the normalized herfindahl index that ranges from 0 to 1 and is computed as follows: h*= (h-1/n)/(1-1/n), where again, n is the total number of seats of the council and h is the usual herfindahl index, as above. european journal of government and economics 1(2) 138 n) school attainment is a proxy for the level of human capital in the regions, as previously defined. o) labour force is the size of the labor force, that is to say the share of units of labor over the regional population. while labor force is an indicator of the efficiency of the input labor, school attainment measures the quality of the input. we expect a positive sign associated to both these coefficients. results table 3 presents the results of the estimation of equation [1]. consistently with our predictions, high levels of corruption are associated with low levels of gdp growth and civicness, as the negative coefficients of voluntary organizations, diffusion of newspapers and referendum voters show. the positive sign of the diffusion of newspapers t-1 variable suggests that corruption is sensitive only to current information; in other words, the diffusion of newspapers is not an effective tool against corruption. this outcome seems to be in line with the positive sign on school attainment. contrary to the prediction of the literature, our estimates show that education does not play a role in reducing corruption. instead, as it clearly emerged in the so-called mani pulite investigation that signed the passing from a system of political patronage to a system of corruption that involved legislators, bureaucrats and businessmen, corruption in the 1980s and 1990s was typically a ‘white collar’ phenomenon. table 3. estimation of equation 1 cor cor+ascr population -0.05*** -0.017 gdpgrowth -0.28*** -0.31*** school attainment 0.3* 0.35** gini index -0.09 0.46 current expenditure/gdp -0.04 0.16 capital expenditure/gdp 0.08 0.053 number of laws 0.13** 0.12** fragmentation 0.58*** 0.63*** voluntary organizations -0.11** -0.13*** diffusion of newspapers -0.33*** -0.36*** diffusion of newspapers t-1 0.45*** 0.48*** referendum voters -0.06 -0.07 ascr, associative corruption crimes 0.17*** constant 2.88** 3.35*** observations 396 399 r2 0.485 0.413 note: ols regression, robust option specified. cor = individual corruption crimes; cor+ascr = sum of all the crimes. continuous variables in natural log. significance level: * p<0.05; ** p<0.01; *** p<0.001 the number of laws and fragmentation are, as predicted, positively correlated with corruption. finally, the presence of associative crimes (ascr) increases the level of individual crimes, bridging the two types of corruption. european journal of government and economics 1(2) 139 table 4. estimates of equation 2 model 1 model 2 model 3 model 4 model 5 model 6 gdp growth t-1 0.386*** 0.313*** 0.665*** 0.599*** 0.391*** 0.335*** expenditure/gdp 0.002 0.107 investments/gdp lag 0.05 0.138 0.577 0.131 0.036 0.136 public consumption/gdp -1.22** -1.009** 0.007 -0.830 -1.235* -0.991* ∆ gini index -0.28 -0.009 0.512 0.496 -0.199 0.160 school attainment 0.59** 0.75*** 0.065 0.749 0.679 0.886* labour force -2.33** -2.48** -1.979 -2.481 -2.32* -2.462** cor_f -8.87*** -8.36*** cor_f2 -0.50 -0.47*** (cor+ascr_f) -2.3 -2.20*** (cor+ascr_f) 2 -0.64*** -0.58*** expenditure/gdp *cor_f -0.254 expenditure/gdp *(cor+ascr_f) -4.51*** note: dependent variable, natural log of gdp growth. lsdvc estimation initialized with ab estimator, 50 bootstrap repetitions. 396 observations. continuous variables in natural log. significance level: *p<0.05; ** p<0.01; *** p<0.001 table 4 presents the results of the growth equation using the lsdvc estimator, including the fitted values of the estimation of the corruption equation.8 the six models differ with respect to the explanatory variables they consider: models 1 4 include the lagged value of investments, while models 5 6 also analyze the amount of public expenditure as a share of the gdp. finally, models 3 4 interact the proxies for corruption with public expenditure to control for their combined effect on growth.9 this effect turns out to be negative and significant, suggesting that the presence of corruption nullifies the positive impact that public expenditures, when productive, have on growth. this result emphasizes the 'sanding' impact of corruption. the variables of interest, i.e. the measures of corruption, are negative and significant as expected. in particular, cor_f and cor+ascr_f are the elasticity of growth to corruption. a marginal variation of cor_f is associated with an opposite variation of economic growth of about 8%; the effect reduces to 2.2% for the sum of the crimes. the smaller coefficient associated to cor+ascr_f indicates that the two types of corruption interact with the economic environment in a way that is complementary and not substitute to the economy itself. associative crimes, in this perspective, seem to mitigate the impact of corruption on growth. the hypothesis of a non-linear relationship between corruption and growth has been tested by including the squared variables of corruption. the coefficients, however, do not show a growth-maximizing effect and rather describe a stable negative impact of corruption on growth. in other words, our results suggest that italy is located on the negative slope of the non-linear relationship after the maximum point as the incidence of corruption is persistently high. the coefficients associated with the lagged dependent variable indicate a pattern of positive growth that is robust across the estimations. the regions have not yet 8 the lsdvc estimates are robust to the application of the system gmm estimator and the estimation of a first difference iv regressions, instrumenting the lagged dependent variables with the lagged independent variables, and assuming corruption to be exogenous. as already said, the gmm estimation of equation 2 never satisfies the rule of thumb, as in the most parsimonious specification we use 23 instruments for 19 groups. 9 the results are robust to the introduction of time effects. european journal of government and economics 1(2) 140 reached a long-term equilibrium path of growth; full convergence is far from being reached. the coefficient on investments/gdp lag is never significant, probably because the lack of disaggregated data on public and private investment hides large inefficiencies in public investment and generates this unexpected result.10 expenditure/gdp, similarly, is never statistically significant but shows the expected positive sign; public consumption negatively impacts on growth, but the coefficient is significant only in the models excluding the interaction between expenditures and corruption. the first difference of the gini index, ∆gini, is never significant and changes the sign across the models. interestingly, school attainment is positive and labor force is negative; both these covariates, however, are significant when lagged investments are included in the specification. the sign of these coefficients is motivated by a larger importance given to the quality of the labor force with respect to its size. since government intervention has been the major policy to reduce income differentials between the north and the south of the country, we further investigate the role of public expenditures on economic growth in the italian regions.11 in table 5 we present the estimation of a set of regression where total expenditure has been disaggregated into its main components12, i.e. current and capital expenditure.13 the effects of the control variables are consistent with the results of table 4. the variable expenditure/gdp is never significant and also its disaggregation is not a significant determinant for growth. in particular, capital expenditure/gdp fosters growth when controlling for corruption (model 3), but it is not significant when controlling also for associative crimes (model 4). this result is unexpected but suggests that the presence of criminal associations in some regions nullifies the productive content of capital investments.14 since local governments have a certain degree of discretion to direct public investments, the efficiency of capital expenditure reasonably changes according to the economic environment where it is implemented. as matter of fact, the composition of crimes varies across observations and some regional economies are more affected by criminal infiltrations than others. current expenditure/gdp, on the other hand, is usually more rigid as it includes mainly personnel wages, transfers to municipalities and local health units (called asl). these items are not expected to stimulate growth but to face everyday needs; therefore, the negative but not significant signs we find in models 3 and 4 do not contradict any theoretical prediction in the literature. in models 5 8, where we use the lagged values of expenditure to overcome an eventual simultaneity of expenditure and growth, this pattern is confirmed. 10 the disaggregated series of data on infrastructure expenditure, that better proxies public investment expenditure, is available only until 1991 (del monte and papagni, 2001), and it has not been included in the analyses to avoid such a huge loss of information. 11 on the nexus between the composition of public expenditure, corruption and growth see, among others, goel and nelson (1998), mauro (1998), tanzi and davoodi (1998), pieroni and d'agostino (2009). 12 the results are robust to the introduction of time effects. 13 to avoid collinearity issues we exclude the lagged investments. 14 caruso (2009) shows that in the italian regions in the period 1997-2003 investments in real estate sector, public investments and health expenditures are positively correlated with corruption, while social protection expenditure and private investments are negatively correlated with corruption. european journal of government and economics 1(2) 141 table 5. estimates of equation 2, disaggregated expenditure model 1 model 2 model 3 model 4 model 5 model 6 model 7 model 8 gdp growth t-1 0.397*** 0.352*** 0.352*** 0.307*** 0.374*** 0.327*** 0.377*** 0.316*** expenditure /gdp 0.002 0.110 current expenditure /gdp -0.412 -0.253 capital expenditure /gdp 0.130* 0.093 total expenditure /gdp t-1 -0.104 0.012 current expenditure /gdp t-1 -0.040 -0.013 capital expenditure /gdp t-1 0.115 0.118* public consumpt. -1.231** -0.960* -0.942 -0.818 -1.329** -1.026* -1.263** -1.048* ∆ gini index -0.186 0.185 -0.376 -0.059 -0.304 0.023 -0.196 0.103 school attainment 0.691 0.906** 0.567** 0.748*** 0.633 0.849** 0.593** 0.765*** labour force -2.258** -2.286** -2.231** -2.31*** -2.037* -2.22** -2.329** -2.40*** cor_f -8.56*** -8.55*** -9.4*** -8.39*** (cor_f)2 -0.48*** -0.48*** -0.53*** -0.47*** cor+ ascr_f -2.22*** -2.26*** -2.32*** -2.26*** (cor+ ascr_f)2 -0.58*** -0.62*** -0.63*** -0.61*** note: dependent variable: ln gdp growth. lsdvc estimation initialized with ab estimator, 50 bootstrap repetitions. 396 observations. continuous variables in natural log. significance level: * p<0.05; **p<0.01; *** p<0.001 the coefficients on cor_f and cor+ascr_f (table 5) are consistent with the prediction of the theory and with our previous estimations. they are significantly negative across all the models and show similar coefficients. in particular, as in table 4, the effect of corruption crimes (models 1, 3, 5 and 7) is always larger than the effect of the overall crimes (models 2, 4, 6 and 8). furthermore, the non-linear terms (cor_f)2 and (cor+ascr_f)2 remain negative and significant as well and smaller than the linear ones. the empirical analyses indicate a sound negative effect of corruption on growth which is robust to different specifications of the model. the results do not verify the growth-maximizing hypothesis. concluding remarks this paper investigates the effects of corruption on economic growth in the italian regions during the period 1980 2004. using a newly assembled data set that include economic, socio-demographic and politico-institutional variables, we address the potential bias of judicial measure of corruption and the endogeneity issue between corruption and growth by developing a two-step empirical strategy, our results show a negative correlation between corruption and economic growth that is robust to different specifications of the model and econometric techniques. we further investigate the role of public expenditure on economic growth since government intervention has been traditionally the major policy implemented to reduce income differentials between the north and the south of the country. total expenditure as well as its main components never turn out to be significant, suggesting that the presence of corruption undermines the positive impact that public expenditure generally has, if productive, on economic growth. european journal of government and economics 1(2) 142 references acemoglu, daron and thierry verdier (1998) ‘property rights, corruption and the allocation of talent: a general equilibrium approach’, economic journal 108(450): 1381-1403. alesina, alberto and romain wacziarg (1998) ‘openness, country size and government’, journal of public economics 69(3): 305-321. alesina, alberto, sule ozler, nouriel roubini and phillip swagel (1996) ‘political instability and economic growth’, journal of economic growth 1(2): 189-212. andvig, jens c. and karl o. moene (1990) ‘how corruption may corrupt’, journal of economic behaviour and organization 13(1): 63-76. arellano, manuel and stephen bond (1991) ‘some tests of specification for panel data: monte carlo evidence and an application to employment equations’, review of economic studies 58(2): 277-297. baldacci, emanuele, arye l. hillman and naoko c. kojo (2004) ‘growth, governance, and fiscal-policy transmission channels in low-income countries’, european journal of political economy 20(3): 517-549. blackburn, keith, niloy bose and m.e. haque (2006) ‘the incidence and persistence of corruption in economic development’, journal of economic dynamics and control 30(12): 2447-2467. blundell, richard and stephen bond (1998) ‘initial conditions and moment restrictions in dynamic panel data models’, journal of econometrics 87(1): 115143. bruno, giovanni s.f. (2005a) ‘approximating the bias of the lsdv estimator for dynamic unbalanced panel data models’, economics letters 87(3): 361-366. bruno, giovanni s.f. (2005b) ‘estimation and inference in dynamic unbalanced panel data models with a small number of individuals’. università bocconi-cespri wp no. 165. caruso raul (2009) ‘spesa pubblica e criminalità organizzata in italia, evidenza empirica su dati panel nel periodo 19972003’, economia e lavoro 43(1): 71-88. crenos (2004) regio-it 1970-2002. data-base on the italian regions, university of cagliari. daniele, vittorio and paolo malanima (2007) ‘il prodotto delle regioni e il divario nord-sud in italia (18612004)’, rivista di politica economica 97(2): 267-316. davigo, piercamillo and grazia mannozzi (2007) la corruzione in italia. percezione sociale e controllo penale. bari: editori laterza. del monte, alfredo and erasmo papagni (2001) ‘public expenditure, corruption and economic growth: the case of italy’, european journal of political economy 17(1): 1-16. del monte, alfredo and erasmo papagni (2007) ‘the determinants of corruption in italy: regional panel data analysis’, european journal of political economy 23(2): 379-396. elrich, isaac and francis t. lui (1999) ‘bureaucratic corruption and endogenous economic growth’, journal of political economy 107(s6): s270-s293. fiorino, nadia and emma galli (2010) ‘an analysis of the determinants of corruption: evidence from the italian regions’, polis working papers no. 171. giardina, emilio, marina cavalieri, calogero guccio and isidoro mazza (2009) ‘federalism, part competition and budget outcome: empirical findings on regional health expenditure in italy’, mpra paper no. 16437. european journal of government and economics 1(2) 143 goel, rajeev k. and michael a. nelson (1998) ‘corruption and government size: a disaggregated analysis’, public choice 97(1-2): 107-20. golden, miriam and lucio picci (2005) ‘proposal for a new measure of corruption, illustrated with italian data’, economics & politics 17(1): 37-75. gyimah-brempong, kwabena (2002) ‘corruption, economic growth, and income inequality in africa’, economics of governance 3(3): 183-209. gupta, sanjeev, luiz de mello and raju sharan (2001) ‘corruption and military spending’, european journal of political economy 17(4): 749-777. hall, robert e. and charles i. jones (1999) ‘why do some countries produce so much more output per worker than others?’, quarterly journal of economics 114(1): 83-116. huntington, samuel p. (1968) political order in changing societies. new haven, ct: yale university press. knack stephen and philip keefer (1997) ‘does social capital have an economic payoff? a cross-country investigation’, quarterly journal of economics 112(4): 1251-1288. klitgaard, robert (1988) controlling corruption. berkeley: university of california press. la porta, rafael, florencio lopez-de-silanes, andrei shleifer and robert vishny (1999) ‘the quality of government’, journal of economics, law and organization 15(1): 222-279. leff, nathaniel h. (1964) ‘economic development through bureaucratic corruption’, american behavioral scientist 8(3): 8-14. lui, francis t. (1985) ‘an equilibrium queuing model of bribery’, journal of political economy 93(4): 760-781. li, hongyi, lixin c. xu and heng-fu zou (2000) ‘corruption, income distribution and growth’, economics and politics 12(2): 155-182. mauro, paolo (1995) ‘corruption and growth’, quarterly journal of economics 110(3): 681-712. mauro, paolo (1998) ‘corruption and the composition of government expenditure’, journal of public economics 69(2): 263-279. méndez, fabio and facundo sepulveda (2006) ‘corruption, growth and political regimes: cross country evidence’, european journal of political economy 22(1): 82-98. myrdal, gunnar (1989) ‘corruption: its causes and effects’, in arnold j. heidenheimer, michael johnston and victor t. levine (eds) political corruption: a handbook, pp. 953-962. new brunswick, nj: transaction books. padovano, fabio (2007) the politics and economics of regional transfers. cheltenham: edward elgar. persson, torsten and guido tabellini (1994) ‘is inequality harmful for growth?’, american economic review 84(3): 600-621. pettersson-lidbom, per and matz dahlberg (2003) ‘an empirical approach for evaluating soft budget constraints’, uppsala university, department of economics working paper series no. 28. pieroni, luca and giorgio d'agostino (2009) ‘military spending, corruption and economic growth’, peace economics, peace science and public policy 14(3). european journal of government and economics 1(2) 144 pigliaru, francesco (2009) ‘persistent regional gaps and the role of social capital: hints from the italian mezzogiorno’s case’. crenos working paper no. 4. putnam, robert d. (1993) making democracy work: civic traditions in modern italy. princeton, nj: princeton university press. roodman, david (2009) ‘a note on the theme of too many instruments', oxford bulletin of economics and statistics 71(1): 135-158. roubini, nouriel and jeffrey d. sachs (1989a) ‘political and economic determinants of budget deficits in the industrial democracies’, european economic review 33(5): 903-933. roubini, nouriel and jeffrey sachs (1989b) ‘government spending and budget deficits in the industrialized countries’, economic policy 4(8): 99-132. shleifer, andrei and robert w. vishny (1993) ‘corruption’, quarterly journal of economics 108(3): 599-617. tanzi, vito and hamid davoodi (1998) ‘corruption, public investment and growth’, in hirofumi shibata and toshihiro ihori (eds) the welfare state, public investment and growth, pp. 41-60. tokyo: springer-verlag. weingast barry r., kenneth a. shepsle and christopher johnsen (1981) ‘the political economy of benefits and costs: a neoclassical approach to distributive politics’, journal of political economy 89(4): 642-664. can exit prizes induce lame ducks to shirk less? experimental evidence european journal of government and economics volume 1, number 2 (december 2012) issn: 2254-7088 106 can exit prizes induce lame ducks to shirk less? experimental evidence leif helland, bi norwegian business school, norway* jon hovi, university of oslo and cicero, norway lars monkerud, bi norwegian business school, norway abstract elected representatives serving their final period face only weak incentives to provide costly effort. however, overlapping generations (olg) models suggest that exit prizes sustained by trigger strategies can induce representatives in their final period to provide such effort. we evaluate this hypothesis using a simple olg public good experiment, the central treatment being whether exit prizes are permitted. we find that a significantly higher number of subjects in their final period contribute when exit prizes are permitted. however, this result does not originate from use of trigger strategies. more likely explanations include gift-exchange and focal-point effects. jel classification c7; c9; h4 keywords overlapping generations; last period effects; legislatures * address for correspondence: leif.helland@bi.no. european journal of government and economics 1(2) 107 introduction legislative parties consist of overlapping generations of elected representatives. having finite biological and political lives, such representatives eventually reach their final legislative period. when a representative realises that she has reached her final period, she becomes a lame duck. we use the term lame duck liberally, meaning representatives that know they are in their final period. in a strict sense the term lame duck characterizes a representative in the time between electoral defeat and exit from the legislature. in the us congress lame duck periods are fairly short; they last from the announcement of electoral loss until replacement by a winner a few weeks later. in parliamentary systems lame duck periods may last considerably longer: from when a representative is denied re-nomination prior to election until replacement takes place several months later. of course, a representative might be in her last period without recognizing it; for example, due to sudden and unexpected death or electoral defeat. in such cases she is not a lame duck, and does not face special incentives problems. since in this study we use an experimental setup in which reelection is exogenously given, a subject’s lame duck period unambiguously corresponds to the final period in which it makes a decision. re-election concerns can then no longer discipline her actions. this creates a tension; while a party's remaining (and future) representatives derive benefits from continued effort by lame ducks, a lame duck motivated by self interest will decline to provide costly effort unless she is incentivized to do so. whereas only few studies have focused on final-period behaviour in assemblies outside the united states,1 a voluminous body of research has considered finalperiod behavior in the us congress.2 these us studies have provided mixed results that to some extent contradict predictions derived from agency models. in particular, most studies have found some degree of final-period attendance shirking, but only scant evidence of deviant voting (ideological shirking). scholars often ascribe the latter finding to selection bias; only candidates with preferences akin to those of their constituencies will likely be elected. it is less clear why constituencies apparently do not select representatives that also continue to attend in their final period (rothenberg and sanders 2000).3 the choice of operational definitions impact on findings and may help explain the mixed results; for example, scholars differ over questions such as how ideological deviations should be measured and when the final period begins. this article serves a three-fold purpose. first, few previous studies have considered the important question of how political parties might use attractive retreat positions and other post-office perks to influence final-period behavior. we contribute towards filling this gap. second, we do so by using a novel experimental approach. we believe that the study of final-period behavior can profit from experimental control, because such control allows for explicit regulation of rewards, containment of reputational concerns to the legislative game, unambiguous separation of the final period from preceding periods, and elimination of selection bias. although many previous public goods experiments have used automatons, our approach is novel in that the automatons mimic an indefinite time horizon in the 1 notable exceptions include besley's and larcinese's (2011) study of attendance shirking in the british parliament, as well as lien's (2008) study of attendance shirking in the norwegian parliament. 2 bender and lott (1996) provide a critical review of the field. lott (1990) deals explicitly with postelective office employment. more recent contributions include rothenberg and sanders (2000), parker and powers (2002) and parker (2004). 3 our research is related to the literature on end-game effects. see brañas-garza and espinosa (2011) for many references to this literature. european journal of government and economics 1(2) 108 laboratory. finally, underlying many previous studies of final-period behavior is alchian's and demetz's (1972) framework (cf. crain, leavens and tollison 1986), which treats incentive schemes as exogenously given. taking the common view that control of procedures and institutions constitutes a defining characteristic of majority assemblies, this is clearly unsatisfactory (krehbiel 1990, shepsle 1986). our experimental design allows incentive schemes to be determined endogenously through majority choice. in the public goods olg model, sufficiently large exit prizes supported by trigger strategies can sustain a fully cooperative equilibrium in which even lame ducks contribute.4 the absence of exit prizes, in contrast, means no equilibrium exists in which lame ducks contribute. however, partially cooperative equilibria, in which all other representatives contribute, may still exist (shepsle and nalebuff 1990, cremer 1992, hammond 1975). with or without exit prizes, other equilibria also exist. in particular, playing the stage-game equilibrium (in which no player contributes) in every period is always an equilibrium.5 to fix ideas, one may think of exit prizes as attractive retreat positions. examples include a position as director or board member of a public company, an ambassador post, a leading position in a national or sub-national administration. related options include campaign support (financing and endorsement) for a top international position. political parties commonly award such positions to retired top politicians.6 we examine the impact of exit prizes in a simple olg experiment. our treatment is whether exit prizes are permitted. we treat electoral support as exogenously given, and representatives are re-elected a fixed number of times.7 in each period subjects choose simultaneously whether to contribute to the provision of a public good.8 we may think of such contributions as costly effort needed for the production of constituency services.9 as a simplifying assumption we let the quantity of the public good increase linearly in the number of contributions. monetary payoffs reflect the assumption that representatives value the public good, but dislike effort. by design, the exit prize 4 salant (1991:255) suggests, but does not analyze, the effect of exit prizes in the generic two-player public goods olg game. exit prizes in a purely distributive olg game are analyzed in alesina & spear (1988). in their model electoral competition is endogenous. there is a growing literature on the distribution of agenda setter rights and other valuable positions in legislative assemblies with overlapping generations of legislators (muthoo and shepsle 2007, 2004; shepsle, dickson and houweling 2002; diermeir 1995). the incentive schemes considered in this literature deviate slightly from ours, in that members gradually obtain more valuable positions in equilibrium (a "seniority system"). in these models (as in ours) electoral competition is exogenous. the impact of exit prizes is not considered. 5 kandori 1992 characterizes folk theorems for olg games with varying information about the history of the game. 6 political scientists debate whether attractive retreat positions serve primarily as rewards for services rendered, or as instruments for cementing partisan power in the longer run, referred to as "patronage" (see for instance pappas 2009; and kopecký and scherlis 2008, with references). we do not see a fundamental conflict here. as long as attractive retreat positions remain scarce, party leaders can use them both to discipline current legislators and to entrench power in the longer run. and even if such positions were to be used mainly for long-run entrenchment, party leaders selecting among otherwise comparable candidates should avoid candidates with a history of shirking. 7 this is clearly a simplifying assumption. it may well be that behavior would change in the experiment if re-election was endogenous (and linked to individual contributions or aggregate production). exogenous re-election allow us to focus more clearly on our main question; can exit prizes incentivize lame ducks. 8 this binary choice is invoked as a simplifying assumption; alternatively, we could have let subjects contribute part or all of a given endowment. 9 shepsle and nalebuff (1990) provide an alternative interpretation, in which aggregate efforts produce "power". european journal of government and economics 1(2) 109 offsets the lame duck’s cost of contributing. moreover, awarding the prize is cost free for remaining subjects.10 few previous experiments addressing the ability of trigger strategies to support high contribution levels in an olg public-goods game have been carried out. notable exceptions include van der heijden et al. (1998) and offerman, potters and verbon (2001). both of these experiments consider a game with two players per period, in which only the newborn player takes an action. in contrast, we consider a strategically richer olg game with three players per period, all of whom take actions. in keeping with the two mentioned studies, the subjects in our experiment face a binary choice (contribute vs. not contribute). van der heijden et al. (1998) study the impact of the game's information structure on contribution levels and on the variability of contributions. they find that knowledge of the game's history explains neither the level of contributions nor their variability. offerman, potters and verbon (2001) consider the impact of recommending the use of a trigger strategy. they find that such a recommendation increases the contribution level and causes many subjects to use trigger-like strategies. however, because of a remaining group of unconditional defectors, the fully cooperative outcome fails to materialize. scholars continue to discuss how one might best mimic an indefinite horizon in olg experiments. van der heijden et al. (1998) use a fixed and publicly known last period. because of this the cooperative outcome cannot be sustained as an equilibrium, which implies serious problems for studying the impact of trigger strategies. offerman, potters and verbon (2001) let subjects play a single olg game ("no reincarnation") with a constant and publicly known periodic stopprobability. as they note, with this procedure the number of subjects in the lab imposes a publicly known upper bound on the number of periods in the game. therefore, the game is not indefinitely repeated in a proper sense. olg experiments conducted in the context of monetary economics rely on a third approach (lim et al. 1994, marimon and sunder 1993), where subjects enter repeatedly in a longer game ("reincarnation"). the reincarnation approach entails that the experimental game becomes much more complicated than the olg models that motivated it (olg models do not admit "reincarnation"). we use a novel design in which automatons replace one subject in the penultimate period and two subjects in the ultimate period of a five-period game (see table 1).11 the automatons are programmed to follow a trigger strategy. this design solves several problems encountered in previous olg experiments. it also reduces the set of equilibria an attractive feature for our purposes. the reader might object that the automatons provide a powerful focal point (schelling 1960) that will likely impact on behavior. our response is two-fold. first, previous research shows that providing a focal point in the form of a strategy recommendation fails to bring out clear trigger-like behavior in olg public-goods experiments, and our own results suggest that the (even stronger) focal point provided by our automatons also fails to generate trigger-like behavior. second, while no-one has thus far come up with a procedure capable of perfectly mimicking an indefinite horizon in finite olg 10 we have resisted the temptation to motivate our experiment more broadly, say, by referring to a richer array of organizations with end-period challenges. in most organizations awarding a bonus for good behavior entails costs for the remaining players. legislatures are special in this regard. remaining legislators are not competing for the positions (= bonuses) concerned, so they do not incur a direct cost by awarding such a position to an outgoing representative. in addition, awarding a position to an outgoing representative (rather than to an outside competitor) entails only modest audience costs. retired legislators will likely be as competent as outside competitors for such positions. furthermore, the positions in question are usually long term, so current legislators will unlikely compete for them in the near future. interesting discussions of end-game effects in bureaucracies are provided by horn (1995). 11 increasing the number of periods further (say, to ten) would either reduce the number of independent observations or require a considerably larger number of subjects. european journal of government and economics 1(2) 110 experiments, our automatons provide a nice and novel way of circumventing this serious problem. we describe our automatons in detail in the section on experimental design. our research question is: do exit prizes induce lame ducks to contribute to the public good? we find that a significantly higher number of outgoing subjects contribute in the treatment with exit prizes than in the baseline treatment without exit prizes. however, our results suggest that this result does not derive from the use of trigger strategies. the next section depicts the olg public-good model underlying our experiment. the third section outlines our experimental design. the fourth section contains descriptive statistics. the fifth section presents the main results. finally, the sixth section concludes. model consider a t-person overlapping generations game under the standard assumptions of rational and self-regarding players. each period corresponds to a stage game with prisoners dilemma (pd) payoffs, in which n = 3 players move simultaneously. information is complete and (almost) perfect.12 it is common knowledge that the game ends in period t. in period t the stage game is played by two automatons and a human; in period t-1 the stage game is played by two humans and an automaton. in periods t = 1, 2,..., t-2 the stage game is played by three humans. a player is either in position a, b or c. the player in position a plays her first stage game, the player in position b plays her second (and next last) stage game, while the player in position c plays her last stage game. the automaton in t-1 plays in position a, while the two automatons in period t play in positions a and b, respectively. no prize the material payoff of player i in stage game t is:    n j jtitit zddzv 1 )1(  (1) in equation 1, z > 0 denotes player i's (exogenously given) endowment (players are assumed to have identical endowments). dit is a dummy that takes the value 1 if i contributes to the public good in stage t, and 0 otherwise. we assume that the material payoffs satisfy (1/n) < β < 1. this means that the payoffs conform to a prisoners dilemma game. since β < 1, setting dit = 0 is a dominant strategy in the stage game. the unique stage game equilibrium is thus dit = 0 ∀ i є t. since β > (1/n), dit = 1 ∀ i є t pareto-dominates the stage game equilibrium. let the state rt be: dit’ = 1 for i = (a, b) in all periods t′< t. a "grim trigger" strategy for the game can now be defined as follows: set dit = 1 if in position a or b and state is rt and set dit = 0 otherwise. the automatons used in the no-prize treatment were programmed to follow this grim trigger. on the equilibrium path of the grim trigger, a player in position a nets (6β + 1)z; a player in position b nets (4β + 1)z; and a player in position c nets (2β + 1)z. by a single deviation the player in a nets (β + 3)z; the player in b nets (β + 2)z; while the player in c nets 3βz. it follows that the player in position a cannot profit by a single 12 almost perfect information means that the history up to period t, but not including period t, is common knowledge. european journal of government and economics 1(2) 111 deviation from the equilibrium path of the grim trigger (by not contributing) if β ≥ (2/5). similarly, the player in position b cannot profit by such a deviation if β ≥ (1/3). finally, the player in position c cannot profit by deviating from the equilibrium path (by contributing) if β < 1. once the punishment path has been triggered, no player can profit by a single deviation from this path, since such a deviation would only reduce her payoff by (1 β)z. it follows that grim trigger is subgame perfect if β ≥ (1/3). given the way the automatons are programmed, only one other equilibrium exists. in this equilibrium all players set dit = 0. prize consider now the game with exit prizes. this game differs from the previous game in only one respect. after the players have made their contribution decisions and these decisions have been announced, players in positions a and b vote on whether to award a prize θz, with θ > 0, to the player in position c. the prize is awarded only if both of the players in positions a and b vote "yes". the material payoff to players in positions a and b is identical to the one described in equation (1). the material payoff to the player in position c is: zevw titit  (2) in equation (2) et is a dummy that indexes the outcome of the vote in stage game t. et takes the value 1 if both of the players in a and b vote "yes", and 0 otherwise. let state st be: dit’ = 1 for i = (a,b,c) and et’ = 1, in all periods t′< t. a grim trigger for the game with prize is defined as follows: if (a) in state st set dit = 1; (b) in state st, in position a or b, and dit+1 = 1 for i = (a,b,c), vote "yes". otherwise set dit =0 and do not vote "yes". as noted, the automatons in our prize treatment were programmed to follow this grim trigger. on the grim-trigger equilibrium path the player in position a nets (9β+θ)z; the player in b nets (6β+θ)z; while the player in c nets (3β+θ)z. consider first a single deviation dit = 0 from the equilibrium path of the grim trigger. such a deviation nets (2β+1)z for the player in position c; it nets (2β+2)z for the player in position b; and it nets (2β+3)z for the player in position a. so, the player in position c has no incentive to deviate if β ≥ (1-θ), while the player in position b has no incentive to deviate if β ≥ ((2-θ)/4) and the player in position a has no incentive to deviate if β ≥ ((3-θ)/7). if θ ≤ (2/3), then (1-θ) ≥ ((2-θ)/4) ≥ ((3-θ)/7). this means that if θ ≤ (2/3) and the player in position c has no incentive to deviate by playing not contribute, then nor will players in positions b or a have an incentive to deviate. next, consider the possibility of a single deviation from the equilibrium path of the grim trigger consisting of players a or b voting "no" to awarding the prize. by voting "no" a player in position b nets z, which is clearly less than what he nets if he deviates by setting dit = 0. by voting "no", a player in a nets 2z, which is clearly less than what he nets if he deviates by setting dit’ = 1. thus, players in positions a and b (who have a choice as to what kind of deviation to perform), will never deviate by voting "no" while continuing to contribute. finally, consider a single deviation from the punishment path of grim trigger, consisting of dit = 1. a player in position c nets βz by deviating, while he nets z if he abstains. a player in position b nets (β+1)z by deviating, while he nets 2z if he abstains. finally, a player in position a nets (β+2)z by deviating, while he nets 3z if he abstains. clearly, no player has an incentive to deviate from the punishment path of grim trigger by setting dit’ =1. the remaining question is whether players in positions a or b can profit by deviating from the punishment path of the grim trigger by voting "yes". the answer is clearly no. a single "yes" is insufficient to change the outcome from no-prize to european journal of government and economics 1(2) 112 prize, so the deviation cannot be profitable. because voting is cost free, players in a and b do not have a strict incentive to coordinate on a "no" vote. on the other hand, neither do they have a strict incentive to coordinate on a "yes" vote. summing up, in the prize treatment the grim trigger is subgame perfect if θ≤(2/3) and β≥(1-θ). given the way the automatons are programmed, there is only one other subgame-perfect equilibrium. in this equilibrium players in positions a, b and c always set dit =0 and never vote "yes". in the grim-trigger equilibrium, even a single defection will effectively end cooperation and will thus likely have a definitive effect on the full behavioral path of the group. moreover, players in positions a and b will have a strong incentive to vote “yes” as long as the players stay on the equilibrium path. design in our experimental design t=5, n=3, z=100 and β=0.6. this applied both to the treatment with prize and to the baseline treatment without prize. in the treatment with prize, θ=0.5. we specified payoffs in an "experimental currency" called schillings. every contribution (whether made by a subject or by an automaton) increased the sum of payoffs by 180 schillings, which were distributed equally among the three subjects who were active players in that period (60 schillings per subject), irrespective of the subjects' own decisions. in the prize treatment, lame ducks that were awarded a prize received an additional 50 schillings. we communicated the resulting payoff structure to the subjects via the instructions (available upon request). we conducted both the prize treatment and the baseline no-prize treatment with three subjects in periods 1, 2 and 3, two subjects and one automaton in period 4, and one subject and two automatons in period 5 (the final period). a game required 5 subjects altogether (cf. table 1). we played four games of the no-prize treatment, and four games of the prize treatment. no subject played more than one game, so this required 40 subjects; 20 subjects in the baseline no-prize treatment, and 20 subjects in the prize treatment. to expand the number of observations, we repeated this basic design with 40 additional subjects. the required 80 subjects were recruited (by e-mail) from undergraduate classes at the bi norwegian business school.13 since the lab had a capacity of 20 subjects we conducted four sessions, two with the no-prize treatment, and two with the prize treatment. the experiment was conducted on two separate days. each day we ran one session without prize and one with prize (in that order). table 1: structure of interaction position a position b position c period 1 subject iii subject ii subject i period 2 subject iv subject iii subject ii period 3 subject v subject iv subject iii period 4 automaton subject v subject iv period 5 automaton automaton subject v for each session we invited 24 subjects in total. whenever more than 20 subjects showed up, we withdrew excess subjects through random draws. a withdrawn subject received a show-up fee of 150 nok (approximately 23 usd). in contrast, 13 informed consent to particiption was obtained by positive responses to written e-mail invtations with a detailed description of the experimental situation. all procedures – including the execution of the experiment, the recording and management of data and presentation of results – were performed in compliance with applicable laws and regulations on subjects’ privacy rights. the experiment was endorsed by the school’s committee for coordinating surveys among students. european journal of government and economics 1(2) 113 participating subjects received no show-up fee. we randomly distributed the 20 subjects participating in a given session to the four groups. also, we randomly assigned numbers from i to v to the five subjects in each group. these numbers determined subjects’ starting positions. we communicated the programming of automatons to the subjects through the instructions. we informed subjects that they would receive their payoffs in cash once their session was over and that schillings would be converted to nok using the exchange rate 1 schilling =0.55 nok. in order to restrict social pressures, subjects entered a separate room one at a time to collect their earnings. because it was impossible to earn more than nok 1000 in this experiment, the payoffs were tax free by norwegian tax laws. as there was no show-up fee for participating subjects, their minimum theoretical earning was 99 nok (approximately 15 usd; in the unlikely event that a subject would be the sole contributor in all three periods). during the experiment all interaction took place via a computer network.14 before starting the experiment, the administrator distributed the instructions and read them aloud (to make them public knowledge). the administrator then asked the subjects several control questions to check that they understood the instructions and the payoff structure. we also conducted a test round (without monetary payoffs) to allow subjects to become familiar with the software. after the test round we randomly rematched groups and starting positions, in order to avoid reputation building based on test-round behavior. to compensate for the fact that participants acting as subject i could make a decision in only one period, we offered those participants additional earnings corresponding to twice the average per period payoff in the group. similarly, because participants acting as subject ii could make a decision in only two periods, we offered these participants additional earnings corresponding to the average per period payoff in the group. the compensation scheme was made public knowledge. the sessions with prize replicated the sessions without prize, except in the following respect: once the subjects had taken their contribution decisions, and these decisions were made known to other subjects in their group, the subjects/automatons in positions a and b voted on whether to award the subject in position c a prize of 50 schillings. awarding the prize required two votes in favor. descriptive statistics and analytical choices in this section we present descriptive statistics for the two types of decisions subjects made during the experiment (contribute vs. not contribute and award prize vs. not award prize). 14 the experiment was programmed in z-tree (fischbacher 1999). european journal of government and economics 1(2) 114 table 2: average percentage of contributions, contingent on position (a/b or c) and treatment (prize or no prize) decisions subjects groups position prize no prize prize no prize prize no prize a or b 69.6 46.4 70.3 48.4 69.6 46.4 (56) (56) (32) (32) (8) (8) c 75 15 75 15 75 15 (40) (40) (40) (40) (8) (8) percentage of decisions, averages over within-subject decision percentages, averages over withingroup decisions percentages (n). table 2 shows the average percentage of contributions, depending on position (a/b vs. c), and treatment (prize vs. no prize). table 3: average percentage of votes cast by subjects in position a or b to award the prize to the outgoing subject (position c) in the prize game (n) decisions subjects groups 80.4 82.8 80.4 (56) (32) (8) table 3 shows the average percentage of subjects in positions a or b voting in favor of awarding a prize to the outgoing subject (position c) in the prize treatment. tables 2 and 3 show little difference between, on one hand, averages based on within-subject means, and, on the other hand, total averages or averages based on group means (the latter two must necessarily yield identical results). as shown in table 2, our data leave us with a maximum of 56 decision-level observations, 40 subject-level observations or 8 group-level observations for contribution decisions within a treatment-position combination. similarly, table 3 shows that further analysis of subjects' propensity to award a prize can be based on a total of 56 decision-level observations, 32 subject-level observations or 8 group-level observations. we report results for group-level observations, which we believe bring us closest to the requirement of independent observations. first, we expect dependencies to exist across different decisions made by the same subject; in fact, it would be strange if no learning or adjustment took place. second, we expect dependencies to exist across subjects in the same group, because of strategic interaction and because subjects in a given group acquire information about the behavior of other players in that group. third, kruskal-wallis tests reveal significant across-group differences in contribution levels and voting patterns between subjects operating under the same experimental condition (i.e. the same combination of treatment and position). this indicates that within-group dependencies actually exist (tests are available upon request). we use only non-parametric hypothesis tests: mann-whitney u-tests (mwu) and kruskal-wallis (kw) tests for comparisons among group measures that vary only between groups (e.g. treatment effects and unobserved differences between individual groups); wilcoxon signed-rank tests (wsr) and friedman tests for comparisons of group measures that may vary within groups (e.g. differences in contributions between different portions or periods of the game); one-sample location tests (osl) for comparing certain outcomes with model equilibria. finally, we use the two-tailed p<0.10 as the criterion for a significant effect. european journal of government and economics 1(2) 115 results we organize the discussion in two parts. first, we demonstrate that lame ducks contribute more often in the prize treatment than in the no-prize treatment. second, we explore possible explanations for this finding. do prizes induce lame ducks to contribute? both treatments in our experiment include an equilibrium in which no player contributes. however, only the prize treatment also includes an equilibrium in which players in position c contribute. thus, our main hypothesis is that the fraction of position c players that contribute is on average at least as high in the prize treatment as in the no-prize treatment. table 2 shows that, in the no-prize treatment, only 15 per cent of the subjects in position c contribute on average. by contrast, in the prize treatment 75 per cent of the subjects in position c contribute. this 60 percent difference between the two treatments is highly significant in a mwu test (z=-3.41, p<0.000), and thus supports our main hypothesis.15 also, the average percentage of votes to award the prize to the outgoing subject in the prize treatment seems quite high in substantive terms (p=80.4; see table 3).16 to check the robustness of our finding that the prize treatment boosts contributions from the outgoing subject, we check whether a similar difference over treatments exists for positions a and b as well, and whether the difference for position c is stable across periods. a first and overall test can be conducted by comparing overall contribution levels for positions a and b in the prize treatment and the noprize treatment, respectively (i.e. the difference between p=69.6 and p=46.4 in table 2). it turns out that no significant difference exists here (mwu test, z=1:45,p=0.15). figure 1 provides additional evidence. 15 however, all four percentages in the rightmost column of table 2 are significantly off the mark in relation to any possible equilibrium outcome: osl tests for h0:p = 0 and h0: p = 100 for the prize-ab, prize-c and no-prize-ab conditions and h0: p = 0 test for the no-prize-c condition all generate p-values less than 0.10. 16 invoking the harsher criterion that data should conform not only to a model's directional predictions but also to its point predictions, we note that this percentage is significantly off the mark in relation to both of the model's equilibrium outcomes (specifically, osl tests of h0: p = 0 and h0: p = 100 both produce p = 0.01). european journal of government and economics 1(2) 116 figure 1: average number of contributions, contingent on position, treatment and period. n=8 for each data point figure 1 shows that in periods 1 through 3, across-treatment differences for subjects in position c are two to three times larger than the corresponding differences for subjects in positions a or b. in period 4 the across-treatment difference for subjects in position c is a sizeable 5 times larger than the corresponding difference for subjects in positions a or b. in the final period no subject in position c contributes in the no-prize treatment, whereas almost 40 per cent of the subjects in position c contribute in the prize treatment. the pattern of gradually diminishing contributions over time that revealed in figure 1 resembles the pattern usually observed in non-olg public goods games (without punishments). it is generally thought that such patterns are due to heterogeneity in social preferences.17 table 4: mann-whitney u tests for differences over treatments implied by figure 1 position treatment χ² p-value a or b prize* 9.15 0.03 no prize* 9.00 0.03 c prize** 9.23 0.06 no prize** 2.33 0.67 total n=16 for each test table 4 displays mwu tests of the across-treatment differences in contribution levels, conditioned on positions and periods. whereas a significant difference across treatments exists for the a/b condition in period 3, the other period-by 17 ledyard (1995) provides a rich survey of public goods experiments. see for instance fehr & fiscbacher (2002) for more recent work on social preferences in public goods experiments. european journal of government and economics 1(2) 117 period across-treatment differences are not significant for this condition. in contrast, all period-by-period across-treatment differences are significant for the c condition. these results make it evident that position c subjects behave differently in the prize and no-prize treatments. figure 1 also shows a marked tendency for the average number of contributors to decrease over the five periods. table 5 presents friedman tests for differences over the five periods. the results suggest that this tendency is significant for all experimental conditions except the c position in the no-prize game, in which contribution levels consistently remain low throughout. table 5: friedman tests for differences over periods implied by figure 1 period position 1 2 3 4 5 a or b z-value -1.47 -1.17 -1.83 -0.45 p-value 0.14 0.24 0.07 0.65 c z-value -2.38 -2.38 -2.84 -2.38 -1.78 p-value 0.02 0.02 0 0.02 0.07 *total n=32, **total n=40. that contribution levels decrease over the five periods is not necessarily incompatible with subjects using trigger strategies to sustain cooperation. we now check if subjects stick to the trigger's punishment phase whenever a deviation from the cooperative equilibrium path has occurred. to do this we condition contribution levels on whether the trigger's punishment phase has been activated. when this phase has been activated, the trigger instructs subjects to cease contributing and, if relevant, cease voting in favor of awarding a prize, for the remainder of the game. figure 2 conveys a first impression of the results. european journal of government and economics 1(2) 118 figure 2: total group contributions (hollow circles, left y-axis) and whether trigger is broken or not (right y-axis black discs) over periods, contingent on contract availability (1=prize; 0=no prize), group (1-4), and day (0=first day; 1=second day) total group contributions (hollow circles, left y-axis) and whether trigger is broken or not (right y-axis black discs) over periods, contingent on contract availability (1=prize; 0=no prize), group (1-4), and day (0=first day; 1=second day) the figure displays the evolution of contributions over the five periods, for the 16 groups in the experiment. the first two rows in the figure show play in the eight groups with no prize, whereas the last two rows show play in the eight groups with prize. hollow circles indicate the sum of contributions for the group (including contributions from the automatons). these are plotted on the left y-axis. black discs, plotted on the right y-axis, indicate whether the punishment path of the trigger is activated. the main picture is that contributions tend to fall once the punishment phase is activated, though not instantly (as it should according to theory). while only one group in the no-prize treatment manages to stay on the trigger's equilibrium path for all five periods, two groups manage to do so in the prize treatment (and one additional group stays on the equilibrium path until the final period). six groups in the no-prize treatment activate the punishment path by period two, compared to three groups in the prize treatment. in all, prizes clearly delay activation of the punishment path. table 6 displays average contribution levels in the two experimental treatments for the part of the game following a deviation from the cooperation path of the trigger strategy, conditioned on position. european journal of government and economics 1(2) 119 table 6: mean contribution levels over treatments in non-cooperative parts of the game, contingent on position (n) position prize no prize a or b 41.3 34.3 (5) (7) c 56.9 7.1 (6) (7) does the behavioral pattern correspond to what we would expect if subjects were to use trigger strategies? first, osl tests for all but one of the entries (p) in table 7 (the exception is position c contributions in the no-prize treatment) reveal that actual outcomes differ significantly from what one should expect if the trigger had not been activated. specifically, for these entries tests of h0: p=100 (i.e. the cooperative equilibrium outcome of an unbroken trigger) all show p-values less than 0.10, which suggests that activation of the trigger at the very least entails a reduction of contribution levels. these results are consistent with the conjectures provided by game theory. still, osl tests of h0: p=0 (i.e. the equilibrium outcome once the trigger has been broken) are also consistently significant (p<0.10) for all table entries, save for contributions from the c position in the no-prize treatment (p=7.1, p=0.50). in particular, contribution levels in the prize-treatment’s c position (p=56.9) are way off the equilibrium mark in substantive terms.18 gift exchange, efficiency, or focal point effect? why do we observe such a clear breach of equilibrium behavior in our data? in addressing this question we are constrained by our experimental design, and our answers can therefore be suggestive only. we focus on three possible explanations. first, outgoing subjects might be aiming for creating a gift-exchange relationship. by contributing, a position c subject makes a costly sacrifice for the benefit of the group. the remaining subjects can return this favor at no cost by awarding a prize to the outgoing subject. moreover, this holds regardless of whether the cooperative equilibrium path of the trigger has already been abandoned. previous research from other settings show that gift exchange may provide a stronger motivation for effort than hard incentives do.19 note that the presence of a gift-exchange relationship would undermine the credibility of the trigger's punishment phase. second, the high observed rates of "yes" votes and of awarded prizes might reflect efficiency concerns. specifically, awarding a prize increases the subjects' aggregate monetary payoffs, at no cost to those subjects that award the prize. there is an ongoing debate in the experimental economics literature regarding the extent to which efficiency concerns impact on behavior.20 18 although the p-value is a "mere" 0.06, this result is most likely influenced by a shortage of observations as some groups (2 out of 8) manage to stay on the cooperative path, as well as by the fact that group-level measures become more unstable as aggregates are based on fewer periods/decisions (only those following a switch to the punishment path of the trigger strategy, which, as noted above, tends to come relatively late in the prize treatment). 19 gift-exchange relationships have been studied experimentally by fehr, gächter and kirchsteiger (1997), and by fehr, kirchsteiger and reidl (1993, 1998). in fehr, klein and schmidt (2001) the contract design is endogenous (i.e. subjects choose a contract from a menu of available contracts). fehr and fischbacher (2005) provide a brief overview of gift-exchange experiments. 20 for a seminal paper dealing with this question in a social preferences framework, see charness and rabin (2000). engelman and strobel (2002) provide a review and some new experiments, which fehr, naef and schmidt (2005) criticize. güth, kliemt and ockenfels (2000) provide an application dealing european journal of government and economics 1(2) 120 finally, across-treatment differences in contribution and "yes-vote" levels may reflect a focal-point effect (schelling 1960): in addition to introducing an altered strategic environment, the prize treatment arguably creates a strong focus on cooperative behavior as such. we utilize the comparative-statics features of our experimental design, first with respect to contributions from the outgoing player and then with respect to "yesvotes". specifically, our comparisons exploit the fact that subjects' actions (contributions, voting decisions) cannot in any way be decisive for outcomes (awarding the prize) when (1) the trigger's cooperative equilibrium path has been abandoned and (2) automatons have entered the group as players (periods iv and v). table 7 lists mean contribution levels for different experimental conditions in which one or two but not all of the three mentioned explanations can be operative. if the trigger drives the subjects' behavior, one should observe low contribution levels and little across-treatment differences in cases where the trigger's cooperative equilibrium path has been abandoned. however, the two mwu-tests reported in table 7 show that contribution levels in such cases are in fact significantly higher in the prize treatment (42 to 70) than in the no-prize treatment (0 to 17). moreover, while the results for the upper row in table 8 (i.e. for the early periods of the game) are clearly at odds with trigger behavior, as well as with behavior driven by efficiency concerns, it is consistent with both the gift-exchange hypothesis and the focal-point hypothesis (i.e. the prize treatment creates a "cooperative focus"). in other words, this particular test cannot distinguish between the two latter hypotheses. table 7: mean contribution levels for outgoing subjects (position c) in periods where non-cooperative behavior has occurred, by treatment and period periods prize no prize z-value p-value 2 and 3 70 16.7 1.86 0.06 (5) (6) 4 and 5 41.7 0 2.38 0.02 (6) (7) z-statistics and p-values for the mann-whitney u test (n). therefore, it makes sense to compare the entries in the bottom row of table 7. whereas our two treatments differ in more than one respect in early post-defection periods (ii and iii), a nice feature of our design is that the treatments do not differ in more than one respect in later periods (iv and v). in our experiment it is public knowledge that automatons will necessarily vote against awarding a prize in periods iv and v, assuming that the cooperative equilibrium path has already been abandoned. hence, it would be pointless to aim at a gift exchange in these periods. however, the two experimental treatments differ in the focus they create on cooperation. the substantial and significant difference in contribution levels at this point (42 vs. 0) provides a strong indication of a focal-point effect. finally, if the gift-exchange hypothesis is correct, one should expect sharply declining contributions from outgoing subjects as opportunities for gift-exchange behavior wither: the pattern for the player in position c of the prize treatment shown in figure 1 is suggestive, with contribution levels decreasing from 87.5 in the first three periods to 50.0 in period iv and 37.5 in period v. however, it is not clear directly with public goods provision. hsu, anen and quartz (2008) explore the neurological correlates of equity and efficiency concerns. european journal of government and economics 1(2) 121 to what extent consistently cooperative groups manage to keep the aggregate at relatively high levels. the relevant pattern and test for group aggregates of contributions in situations where the cooperative equilibrium path has been abandoned concern the levels presented in the far left column of table 7. it turns out that a substantial 40 percentage point difference exists in contribution levels between (1) periods where outgoing subjects can be decisive (70) and (2) periods where they cannot be decisive (30, rather than the listed figure of 41.7, which includes a group that cooperates through periods 2 and 3). however, although sizable, this effect is not significant in a wsr test at conventional levels (s=3.0, p=0.25, n=5).21 we now turn to voting patterns in the prize treatment. table 8 lists the group mean of votes cast in favor of awarding a prize to the outgoing player, conditioned upon whether the latter chooses to contribute. table 8 includes only cases where voting occurs after the trigger's cooperative equilibrium path has been abandoned in periods 1, 2 or 3.22 table 8: mean share of votes in favor of awarding the prize to the outgoing subject (position c) in periods where non-cooperative behavior has occured before period 4, by contribution from the outgoing player (n) contribution from c no contribution from c total 0.8 0.63 0.75 (5) (2) (7) the overall share of votes cast in favor of awarding a prize after abandoning the cooperative equilibrium path is high (0.75,n=7) and significantly different from zero in a simple osl test (p=0.02). while at odds with the hypothesis that subjects play trigger strategies, this finding is consistent with the focal-point hypothesis. however, it is also consistent with the gift-exchange hypothesis and the efficiencyconcerns hypothesis. while having the right "reciprocal sign" according to the giftexchange hypothesis, the difference between the contribute condition and the not contribute condition is insignificant. table 9 shows that only two group-level units have observations in both conditions. the average percentage of "yes-votes" is 0.63 in the not contribute condition, and increases only moderately to 0.80 in the contribute condition for the same two groups. needless to say, applying the appropriate wsr test with only two group-level units proves futile (s=0.5, p=1.00).23 in essence, our experiment fails to provide statistically strong support for the giftexchange hypothesis. however, as we explained in our analysis of contribution patterns, it seems that outgoing subjects do not expect a prize to be awarded 21 five groups in the prize treatment have observations both in early periods (ii and iii) and in late periods (iv and v). in other words, three out of the eight groups are naturally excluded from the analysis: the two groups that manage to stay on the cooperative equilibrium path, and the one group that abandons this path in period iv, so that it plays under the broken trigger only in period v. 22 we include decisions in period 1, since the trigger strategy may shift to the punishment path not only as a consequence of non-cooperative play in any previous period, but also as a consequence of any departure from a pattern of across-the-board contributions in the current period of play. also, when looking at voting behavior in post-defection parts of the game we completely disregard patterns in period 4 since any level here is consistent with both the gift-exchange and the efficiency-concerns hypotheses: a voting decision in this situation is not only costless, it is also haphazard since subjects are completely disenfranchised (the outcome of no prize assured by the automatons inevitable exercise of its veto powers). 23 departing from our analytic choice of utilizing the group level as our unit of analysis and/or from the sounder practice of applying within-unit tests (wsr) where this is feasible, we also note that the listed difference of 0.80-0.63=0.17 in vote share averages between the two situations is not significant in the more lenient mwu test between groups (z=-0.63, =0.53, n=7) nor in an mwu test with individual decisions as units of analysis (diff.=0.80-0.67=0.13, z=-0.63, p=0.53, n=26). european journal of government and economics 1(2) 122 regardless of whether they contribute. in the prize treatment, outgoing subjects usually choose to contribute rather than to keep their endowment hoping that remaining subjects will nevertheless award them a prize (see table 6). in conclusion, outgoing subjects contribute far more often in the prize treatment than in the baseline no-prize treatment. however, our data indicate that this result does not originate from use of trigger strategies. several statistical tests reveal that behavior differs significantly from what one would expect if trigger strategies were to drive behavior. in the prize treatment, the focal-point effect likely explains the observed behavior; in fact, this is the only alternative explanation for which we find statistically significant evidence. although there is also some evidence that subjects engage in gift-exchange behavior, the small number of observations in our design makes statistically significant evidence for this proposition infeasible. conclusion many scholars consider that lame ducks will continue to service their party only if being incentivized. according to the model motivating our experiment, remaining party representatives can discipline lame ducks by credibly promising them exit prizes (such as attractive retreat positions or other post-office perks) if they do not shirk. we have examined this proposition experimentally. novel use of automatons allowed us to sidestep the fundamental problem of mimicking parties with indefinite lives in the lab. in one treatment remaining subjects could award cost-free exit prizes to the lame duck. the other treatment did not permit such prizes. in line with the model’s prediction, we found that a significantly higher proportion of outgoing subjects contribute in the prize treatment than in the baseline no-prize treatment. closer inspection, however, revealed that many outgoing subjects who chose to contribute received a prize irrespective of whether the trigger's punishment path had been activated. while inconsistent with the incentive scheme provided by trigger strategies, this behavioral pattern may be explained in several ways. we have indicated three explanations. first, the strategic situation generates a gift-exchange relationship. sheer decency suggests that an outgoing player who contributes should be awarded a prize regardless of play in previous periods. the fact that awarding a prize entails zero cost for remaining players makes this moral imperative particularly compelling. second, because failure to award a prize would leave money on the table, efficiency concerns imply that a prize should be awarded regardless of previous play. finally, the prize treatment makes cooperative behavior a powerful focal point, which might explain the observed behavior. our data cast considerable doubt on efficiency concerns as a likely explanation, while not allowing us to discriminate between gift-exchange and focal-point effects. the presence of either of these two motives, however, might undermine the credibility of the trigger strategy that supposedly sustains the proposed incentive scheme. the environment we study is very simple: the structure of the game is public knowledge; legislative (mis)behavior cannot impact on re-election prospects or on future job prospects (through reputational concerns); end-game behavior is automated; there is no ambiguity with respect to lame-duck periods; all interaction is anonymous; and rewards and punishments are fully controlled by the majority of the party. clearly, it bears little resemblance to the environment of real-world legislative politics. this, however, is hardly an objection. our environment was designed to control the impact of such other factors, so as to obtain a rigorous test of one particular hypothesis about legislative behavior: are lame ducks disciplined by endogenously determined exit prizes? our experiment indicates that they are, although in ways inconsistent with the standard theory of rational, self-regarding european journal of government and economics 1(2) 123 players. this, in turn, suggests that standard theory may be too restrictive to provide a proper understanding of legislative behavior, and that further experiments could be useful to guide future modelling. if subjects in a simple setup like ours deviate systematically from standard theory, one may also expect legislators operating in vastly more complex environments to deviate. references alchian, armen and harold demsetz (1972) ‘production, information costs, and economic organization’, american economic review 62(5): 777-795. alesina, alberto and sstephen e. spear (1988) ‘an overlapping generations model of electoral competition’, journal of public economics 37(3): 359-79. brañas-garza, pablo and maria paz espinosa (2011) ‘unraveling public good games’ games 2(4): 434-451. bender, bruce and john r. lott (1996) ‘legislator voting and shirking: a critical review of the literature’, public choice 87(1-2): 67-100. besley, timothy and valentino larcinese (2011) ‘working or shirking? expenses and attendance in the uk parliament’, public choice 146(3-4): 291-317. cadsby, charles bram and murray frank (1991) ‘experimental tests of ricardian equivalence’, economic inquiry 29(4): 645-664. charness, gajiy and mathew rabin (2000) ‘social preferences: simple tests and a new model’, economics working paper 441. department of economics and business. universitat pompeu fabra. crain, w. mark, donald r. leavens and robert d. tollison (1986) ‘final voting in legislatures’, american economic review 76(4): 833-41. cremer, jaques (1986) ‘cooperation in ongoing organizations’, quarterly journal of economics 101(1): 33-50. diermeir, daniel (1995) ‘commitment, deference, and legislative institutions’, american political science review 89(2): 344-55. engelman, dirk and martin strobel (2002) ‘inequality aversion, efficiency, and maximin preferences in simple distribution experiments’, american economic review 94(4): 857-69. fehr, ernst and urs fischbacher (2005) ‘the economics of strong reciprocity’, in herbert gintis, samuel bowles, robert boyd and ernst fehr (eds) moral sentiments and material interests: the foundations of cooperation in economic life, pp. 151-192. cambridge mass.: the mit-press. fehr, ernst, simon gächter and georg kirchsteiger (1997) ‘reciprocity as a contract enforcement devise’, econometrica 65(4): 833-60. fehr, ernst, michael naef and klaus m. schmidt (2005) ‘the role of equality and efficiency in social preferences‘, cepr discussion paper no. 5368. fehr, ernst, georg kirchsteiger and arno reidl (1993) ‘does fairness prevent market clearing?’, quarterly journal of economics 108(2): 437-59. fehr, ernst, georg kirchsteiger and arno reidl (1998) ‘gift exchange and reciprocity in competitive experimental markets’, european economic review 42(1): 1-34. fehr, ernst, alexander klein and klaus m. schmidt (2001) ‘fairness, incentives and contractual incompleteness’, working paper no. 72, insitute for empirical research in economics, university of zurich. european journal of government and economics 1(2) 124 fehr, ernst and urs fischbacher (2002) ‘why social preferences matter the impact of non-selfish motives on competition, cooperation and incentives’, the economic journal 112 (478): c1-c33. fischbacher, urs (1999) z-tree: zurich toolbox for readymade economic experiments. university of zurich. güth, werner, hartmut kliemt and axel ockenfels (2000) fairness vs efficiency. an experimental study of (mutual) gift giving. berlin: humbolt university (unpublished). hammond, peter (1975) ‘charity: altruism or cooperative egoism’, in edmund s. phelps (ed) altruism, morality, and economic theory. new york: sage. horn, murray j. (1995) the political economy of public administration. new york: cambridge university press. hsu, ming, cedric anen and steven r. quartz (2008) ‘the right and the good: distributive justice and neural encoding of equity and efficiency’, science 320(5879): 1092-1095. kandori, michi (1992) ‘social norms and community enforcement’, the review of economic studies 59(1): 63-80. kopecký, petr and gerardo scherlis (2008) ‘party patronage in contemporary europe’, european review 16(3):355-71. krehbiel, keith (1990): information and legislative organization. ann arbor: university of michigan press. lien, jon (2008) a politician who's quitting has already quit. m.sc. thesis (political economy). norwegian school of management bi. lim suk s., edward c. prescott and shyam sunder (1994) ‘stationary solution to the overlapping generations model of fiat money: experimental evidence’, empirical economics 19(2): 255-77. ledyard, john o. (1995) ‘public goods: a survey of experimental research’, in john h. kagel and alvin e. roth (eds) handbook of experimental economics, pp. 11194. princeton: princeton university press. lott, john r. (1990) ‘attendance rates, political shirking, and the effect of postelective office employment’, economic inquiry 28(1): 133-50. marimon, ramon and shyam sunder (1993) ‘indeterminacy of equilibria in a hyperinflationary world: experimental evidence’, econometrica 61(5): 1073-1107. mutho, abhinay and kenneth a. shepsle (2007) information, institutions and constitutional arrangements. university of harvard (unpublished). mutho, abhinay and kenneth a. shepsle (2004) agenda-setting power in organizations with overlapping generation of players. university of harvard (unpublished). offerman, theo, jan potters and harrie a. a. verbon (2001) ‘cooperation in an overlapping generations experiment’, games and economic behavior 36(2): 264275. pappas, takis s. (2009) ‘patrons against partisans: the politics of patronage in mass ideological parties’, party politics 15(3): 315-334. parker, glenn r. and stephen c. powers (2002) ‘searching for symptoms of political shirking: congressional foreign travel’, public choice 110(1-2): 173-191. parker, glenn r. (2004) self policing in politics. princeton, princeton university press. european journal of government and economics 1(2) 125 rothenberg, lawrence s. and mitchell s. sanders (2000) ‘serving the electoral connection: shirking in the contemporary congress’, american journal of political science 44(2): 316-325. salant, david j. (1991) ‘a repeated game with finitely lived overlapping generations of players’, games and economic behavior 3(2): 244-259. shepsle, kenneth a. (1986) ‘institutional equilibrium and equilibrium institutions’, in herbert f. weisberg (ed) political science: the science of politics, pp. 51-81. new york: agathon. shepsle, kenneth a., eric s. dickson and robert p. houweling (2002) bargaining in legislatures with overlapping generations of politicians. harvard university (unpublished). shepsle, kenneth a. and barry nalebuff (1990) ‘the commitment to seniority in self-governing groups’, journal of law, economics and organization 6(special issue): 45-72. schelling, thomas (1960): the strategy of conflict. cambridge mass.: harvard university press. van der heijden, eline c. m., jan nelissen, jan potters and harrie a. a. verbon (1998) ‘transfers and the effect of monitoring in an overlapping-generations experiment’, european economic review 42(7): 1363-1391. the effect of subsidising firms on voting behaviour: evidence from flemish elections european journal of government and economics volume 1, number 1 (june 2012) issn: 2254-7088 30 the effect of subsidising firms on voting behaviour: evidence from flemish elections caroline buts, vrije universiteit brussel, belgium* marc jegers, vrije universiteit brussel, belgium dimi jottier, vrije universiteit brussel, belgium abstract despite an international consensus on the importance to limit state aid spending, large amounts of resources are still devoted to a wide variety of subsidies to firms. a sizable literature studies the relationship between general government spending and the proximity of elections, mostly documenting a positive link. in addition, other studies verify whether this strategy of increasing government expenditure pays off in terms of number of votes. we focus on one type of government spending that can be quite vulnerable to becoming ‘targeted spending’, i.e. subsidies to firms. we empirically test the relationship between the amount of subsidies granted to firms at the local level and local support for incumbent parties in the regional government. to that end, we make use of subsidy data derived from financial statements on 2008 and flemish election results of 2004 and 2009. we find that the total amount of subsidies as well as subsidies per capita granted in 2008 positively correlate to support for incumbent parties 2009, meaning that voters appear to reward subsidy granting politicians. jel classification h25; d72; l50 keywords subsidies; voting behaviour; industrial policy * address for correspondence: caroline buts, vrije universiteit brussel, department of applied economics, pleinlaan 2, be-1050 brussel, belgium. e-mail: caroline.buts@vub.ac.be. doi: https://doi.org/10.17979/ejge.2012.1.1.4275 european journal of government and economics 1(1) 31 introduction governments spend a substantial amount of resources, allocating subsidies to private firms, seemingly in order to correct market failures or to support specific regions or sectors. however, there also seems to be a consensus that (taking into account a few exceptions) state aid should be forbidden as it can lead to costly subsidy races between countries and in the long run can create inefficient companies (hancher et al. 2006). furthermore, in practice, we see that these subsidies are often ineffective or inefficient in a sense that goals are not (fully) reached or that the benefits achieved do not justify the costs incurred. therefore, it is argued that mechanisms should be implemented to control or restrict wasteful spending on subsidies. dewatripont and seabright (2006) suggest that domestic political control mechanisms are not the best option. in theory, political accountability works as follows: citizens vote as a function of their evaluation of government performance. incumbent parties are electorally punished for a ‘bad’ performance, and a ‘good’ performance is rewarded with additional support in the ballot box. in doing so, incentives are created for politicians to implement a policy perceived to be ‘good’ in order to remain in office. as already mentioned, subsidies are not the most efficient way to allocate government’s resources, but are perceived by voters as “evidence of effort on the part of the politicians” (dewatripont and seabright, 2006, p. 514). if voters reward this effort with additional votes, the political control mechanism creates a perverse effect: it then becomes rational for politicians to spend money on wasteful subsidies as this would “improve their chances of re-election” (dewatripont and seabright, 2006, p. 514). the aim of this paper is to empirically test whether subsidies indeed are related to electoral outcome, a relationship which, to the best of our knowledge, has not yet been directly tested. it serves as a further specification of the broader literature on pork-barrel politics that studies politicians who direct spending in order to win additional votes. subsidies are chosen as they are probably more sensitive to become ‘targeted’ spending projects than many other types of government spending. the relationship is tested at the local level. the economic voting literature shows that electoral support for national governments is greater in local jurisdictions where economic prosperity is higher (pattie et al., 1997; johnston et al., 2000, 2002). hence, if subsidies have an electoral effect, the support for governments should be higher in local jurisdictions where the amount of allocated subsidies is higher. we compare the 2009 electoral results of the incumbent parties of the flemish government between the different electoral cantons and hypothesize that a high amount of subsidies allocated to a specific region increases local electoral support for the incumbent parties. the remainder of this paper is organized as follows: in section 1 the ineffectiveness of many state aid measures is discussed as well as the obvious international desire to diminish or even abolish this kind of spending. furthermore, we look at literature on the determinants of election outcome and the importance of a local context. the last part of this section reviews the literature on pork-barrel politics and thus discusses the relationship between the level of government spending and upcoming elections as well as the influence of this spending on vote shares of incumbent parties. the second section specifies our model and describes the data. the results of the empirical analysis are presented in section 3. a conclusion and discussion can be found in section 4. literature subsidies are granted for a variety of reasons such as support to r&d and aid to firms in difficulty. for many of these objectives, the effectiveness and efficiency of different measures has been studied, often resulting in quite negative findings. european journal of government and economics 1(1) 32 frequently, goals are not reached, or when they are, the costs are out of proportion. examples of this are the crowding out effect of r&d subsidies1, the ineffectiveness of many rescue and restructuring measures2 and the hugely costly aid to boost employment.3 next to this wide evidence of inefficiencies, an international consensus seems to exist among politicians and organizations about the desire to diminish state aid measures. this becomes evident in the european union’s state aid policy and the world trade organization’s agreement on subsidies and countervailing measures. from a theoretical point of view, collie (2000 and 2002) makes a well-reasoned plea for a serious reduction of state aid over time in the eu, based on welfare grounds. we then may ask why governments keep spending resources to the current extent in this often inefficient way. while searching for explanations, we find several possibilities. a first one is provided by baldwin and robert-nicoud (2007). governments support inefficient firms simply because these firms have a strong lobby. in a way, the process of integration in the european union stimulates this demand for government support. integration increases market performance but, at the same time, higher competition threatens to force non-efficient firms out of the market. these firms will then ask for aid from their national governments. furthermore, penalties for acting against eu state aid rules are not always severe enough, possibly leading to situations in which, despite welfare being reduced, granting aid is an equilibrium outcome (martin and valbonesi 2008). another explanation is offered by dewatripont and seabright (2006) and represents the theoretical foundation for our empirical analysis: subsidies are seen by voters as an effort of politicians. voters will reward politicians for this effort in the ballot box. in their turn, politicians are aware of this and grant subsidies to show their commitment and thereby gain support for future elections. normally, the mechanism of domestic political accountability would control politicians, but in this case it does not and it even provides an incentive for wasteful spending. political accountability builds on the ‘carrot and stick’ model: a good performance (as perceived by voters) leads to electoral gain for incumbent parties, contrary to a bad performance. this hypothesis has been intensively investigated in the economic voting literature. from this literature, we conclude that economic prosperity – generally measured by variables such as unemployment rate, inflation and economic growth has positive effects on electoral results for incumbent parties (for a review see mueller, 2003 and nannenstad and paldam, 1994). as wasteful spending is shown to be a sign of rather bad governance, we would expect the accountability mechanism to punish politicians for this. however, here the mechanism works in the opposite way. subsidies are not evaluated as being wasteful but as an effort of politicians which generates extra votes. this encourages politicians to continue subsidising. economic prosperity as measured in the voting literature can be perceived at different levels. voters can assess the performance of the national economy (voting socio-tropically) and they can look at the prosperity of their own household (voting ego-tropically). research has shown that both levels are important although voters seem to be more socio-tropic than ego-tropic (lewis-beck and paldam, 2000). furthermore, it has been suggested that the local context wherein these evaluations are made also plays an important role (books and prysby, 1999). 1 for a review on the econometric literature on crowding out of r&d subsidies, see david et al. (2000). in addition, lach (2002) and gelabert et al. (2009) find that these kinds of subsidies are only effective when granted to small firms or firms with low levels of appropriability. 2 london economics (2004), chindooroy (2007) and glowicka (2008). 3 bergström (1998) and tannenwald (2002). european journal of government and economics 1(1) 33 through the local media, social interaction and personal experiences, voters in states with low economic prosperity would be more confronted with negative information, which affects their assessment of the government’s economic performance. not only assessments but also voting behaviour is affected by the local context. controlling for national and personal assessment, pattie et al. (1997) find that, for the 1992 uk elections, voters who thought that their region was worse off were less likely to vote conservative (the incumbent party at that time) and more likely to vote labour (the main opposition party). their results are confirmed by johnston et al. (2000, 2002) who find that, controlling for national, personal and regional assessments, economic prosperity in a region (measured by the unemployment rate) increased support for incumbent parties in that specific region. however, berry and howell (2007) state that evidence of retrospective voting in national elections seems to be steadier than in local settings. from duch and stevenson (2006) we learn that economic voting is not found to influence all elections, but a clear reason for this has not yet been found. they find varying evidence for economic voting depending on national context and time. the accountability mechanism has also been tested with a wider range of variables. hagerty (2006) adds measures for the quality of life to this ‘responsibility hypothesis’ and finds that crime rates are important in predicting election outcome. however, the influence of crime rates tends to be smaller than that of the economic condition. finally, the relationship between government spending and vote shares has been discussed in the literature. there is evidence suggesting that governments indeed seem to believe that subsidies positively impact on voting behaviour. in addition, it has been discussed whether this kind of spending to win additional votes actually pays off. the phenomenon is referred to as pork-barrel politics. verdier (1995) explains that politicians grant subsidies in order to create a (stable) network of supporters. kwon (2005) documents the fact that the level of government expenditure in south korea depends on the proximity of future elections. also, it is known that in developing countries, white elephants (funded projects with a negative social surplus) are used in an attempt to influence election outcome. robinson and torvik (2005) show that certain politicians are able to derive very large political benefits from this kind of projects. keeping the above findings in mind, keefer and knack (2007) make an interesting addition. they find that public investment is higher in regimes with little political checks. however, they do say that further research is necessary to determine the cause: is it to compensate for incompetence of the regime or is the money spent on ‘preferential’ projects? also manzetti and wilson (2007) state that nations with weaker democratic regimes are more likely to have a high amount of targeted spending to win votes. important to note is also that smaller regions are believed to be more sensitive to pork-barrelling. hauk and wacziarg (2007) provide evidence that infrastructure spending is disproportionately allocated to smaller states in the us. previously, herron and shotts (2003) already documented that certain funds or projects which are claimed to attribute electoral benefits were allocated to the smaller states. thus, politicians seem to believe that ‘targeted’ spending might earn them extra votes and often act accordingly. it is off course important to study whether incumbents really benefit in the ballot-box of this kind of spending. several studies such as the ones by feldman and jondrow (1984) and stein and bickers (1994) failed to find general evidence. stein and bickers (1994) argue that redefining the general model is necessary and show that it are mainly politicians in office that do not feel confident about re-election that engage in this kind of spending, and more importantly that it are mostly politically attentive voters that respond to it. moreover, european journal of government and economics 1(1) 34 alvarez and saving (1997), levitt and snyder (1997) and leigh (2008) find that high spending results in more votes. the second team finds that it takes about $14 000 in federal funding to win an additional vote. the last author distinguishes between types of spending and finds the strongest relationship between funding and additional votes in the roads to recovery program, about the same amount that was found by levitt and snyder (1997). specification and data in the literature review, we find evidence that state aid is often ineffective and that it is accepted that it should mostly be forbidden. nevertheless, huge amounts of state aid are spent each year by national governments. for example, in 2010 the eu-27 spent 0.6% of aggregated gdp on state aid.4 this, together with the evidence from the literature on pork-barrelling, leads us to believe that in many cases there could be political motives to government spending. in this research we want to look at one specific form of spending, namely subsidies to firms as we believe that they could be especially sensitive to becoming ‘targeted’ spending in order to win votes. in a way, it boils down to the fact that politicians grant subsidies even though overall believe is that these subsidies should only be allowed in exceptional circumstances. therefore, keeping in mind that a link exists between subsidies and electoral competition, we think that a realistic explanation is presented by dewatripont and seabright (2006): politicians grant subsidies to show their commitment and in this way try to gain more support in next elections. in this theory subsidies are perceived by voters as an effort of politicians, signalling a caring government. this is rewarded with electoral support in next elections. the voting decisions are made on a point in time when voters are aware of the subsidy, but not of its effect, which in most cases can only be evaluated after a few years. in the present paper we want to empirically investigate whether subsidies indeed affect voting behaviour and hypothesize that differences in electoral support for incumbent parties between the different regions can partly be explained by differences in the amount of subsidies granted to firms located in these different regions. we thus start from the theoretical model presented by dewatripont and seabright (2006) as we believe that voters reward politicians for their effort. subsidies to firms are used to test this as they are a form of spending likely to be used for pork-barrel politics. people working in the aid receiving firms are aware of the fact that subsidies were granted to these firms. this may not always be true, but it will be in the majority of cases. some types of subsidies are namely more likely to be communicated towards the employees than others. for example, aid for r&d or rescue and restructuring aid will usually be known in the entire firm, whereas several smaller subsidies may be not. the employees of the receiving firm are likely to inform family and close friends and thus in this way make sure the information spreads. the more firms receive subsidies, the more inhabitants of a certain area are aware of the ‘caring’ incumbent politicians. also, as pointed out in the literature review, pork-barrelling is observed more in smaller states as described by hauk and wacziarg (2007). indeed, the effect of pork-barrelling will logically be larger were local communities are small and social interaction is often quite high. where there are many contacts with the neighbourhood and within recreational organizations, information will spread more easily. moreover, press in flanders is very local and helps spreading these facts by reporting all kinds of local news, including economic news that is locally relevant. for the empirical analysis, we use election data of the flemish region for 2004 and 2009. flanders represents a typical example of a western democracy, with a proportional multiparty system, located within the european union. it provides for a 4 european commission (2010). this percentage excludes support to railways and crisis measures. european journal of government and economics 1(1) 35 good test case as data on subsidies can be calculated per firm (see below). flanders is one of the three belgian regions besides wallonia and the (smaller) brussels region. the belgian regions have considerable autonomy. their major competences are in the field of economic policy and education. every five years, elections are held to choose representatives for the flemish parliament. within this parliament, a majority forms the government. after the 2004 elections, the government was formed out of three cartels: cd&v-n-va (cartel of the christian democrats and a nationalist party), vld-vivant (a liberal cartel) and sp.a-spirit (a left wing cartel). the next elections for the flemish government in 2009 resulted in a new coalition of cd&v, sp.a and n-va.5 flanders (13 682 km²) is divided into five electoral districts. each of these districts consists of several cantons (103 in total). in 2009 there lived, on average, 59 821 people in a canton with a minimum of 3 672 and a maximum of 490 492.6 to analyze the impact of subsidies, we study the electoral results of parties that made up the government between 2004 and 2009. these results differ substantially between cantons as can be seen in figure 1. vote share changes are calculated for every canton. on average, incumbent parties lost 1 percentage point of their electoral support. however, in one canton they lost 14 percentage points of their vote shares, whereas they also won 8 in another. figure 1: kernel density plot of vote change of incumbent parties (per canton) 0 .0 2 .0 4 .0 6 .0 8 .1 d en si ty -15 -10 -5 0 5 10 change vote share kernel density estimate 5 cartel partners cd&v and n-va decided to go separately to the 2009 elections. the two other cartels making up the flemish government after 2004, chose a new name for the 2009 elections. 6 the difference in size of the cantons implies that the perception of localness differs for voters within different cantons. a small canton is much more ‘local’ than a larger canton. johnston et al. (2000), however, demonstrate that the response of voters to local unemployment levels is scale invariant. whether unemployment was measured at a very small scale (around 500 people) or at the constituency scale (around 70 000 people on average) it still had a negative effect on the electoral support of incumbent parties. we see that including a dummy variable for cantons containing a big city does not impact on our results (see below). all in all, all cantons can be labeled ‘small’ compared to the small units of observations found in the literature. also, we include voting results of previous elections as an independent variable. local dimensions are in this way incorporated on the ‘independent’ side. and, despite the large variance in canton size, they all are small comparing to other countries. including a population variable in the analysis does not lead to significance. furthermore, it does not change the other results. european journal of government and economics 1(1) 36 empirical model we hypothesize that these differences in electoral results between cantons can be (partly) explained by differences in subsidies granted to firms located in these cantons. we expect that in cantons where private firms received more subsidies, the electoral support for incumbent parties will be higher than in other cantons. this leads to the following function, estimated at the canton level: votes2009 = α + β1 votes2004 + β2 subsidy + β3 x + ε where votes2009 stands for the vote share of all the incumbent parties at the 2009 elections, votes2004 is the vote share of these parties at the 2004 elections. subsidy is the amount of subsidies or in a second step the subsidies per capita granted to firms in the canton in 2008. x is a set of socio-economic parameters ranking all cantons. in a second step we also look at relative change in vote share. subsidies. we will focus on all subsidies to firms in flanders, excluding other measures such as loans or guarantees. in belgium, subsidies are granted by several levels of government and through different organizations under their supervision. the subsidies are linked to the elections of the regional governments, which grant the bulk of the subsidies. a small proportion of the subsidies is provided by the federal government (also directly going to the firms). however, from limosani and navarra (2001) we learn that voters can reward politicians for spending that occurs at a different level of government. we assume that voters do not really differentiate between resources coming from different levels of government. through personal experience, their social network or local press, they are informed about the subsidies and relate this to the politicians that are in office. most of the subsidy granting organizations are not willing to provide detailed information on this topic, the amount of subsidies they award and to whom, except for iwt (agency for innovation by science and technology). therefore, we estimated the subsidies per firm (and subsequently calculated subsidies per canton) from the firms’ annual accounts.7 we include two tests, one with the absolute level of subsidies for which the variable is scaled by a factor 100 000 and one with subsidies per capita. the method of calculation of subsidies leads to an inclusion of all capital grants to firms conditionally upon being subsequently invested by the firm. it concerns subsidies provided by all levels of government for a variety of objectives such as research and development, start-up aid and sustainable development. subsidies granted as a result of aid schemes8 as well as ad hoc types are included. the purpose is to investigate the joint influence of all these subsidies to firms. flemish firms received subsidies for a total amount of €829 million in 2008. this represents on average a subsidy of €80 per capita. however, there are quite some differences between the cantons. in one canton, we find no subsidies at all. in another canton, subsidies amount to €2 622 per capita. in our analysis we include only subsidies and subsidies per capita of 2008 instead of total subsidies over the four years between the two elections because, as was mentioned in the literature review, it has been shown that governments seem to increase their spending, including subsidies, in the year prior to elections. additionally, voters can be assumed to have short memories, rather recalling last years’ subsidies than last five year period’s subsidies. we expect that subsidies will positively impact on incumbent support following the mechanism explained above. 7 the method of calculation of subsidies can be found in appendix 1. 8 on one hand, aid schemes provide the possibility for multiple firms to apply for a subsidy. they usually have to fulfill specified criteria in order to receive the subsidy. they are however not available to all firms. they can be selective, for example, in a geographical or sectoral manner. on the other hand, there are also ad hoc types of aid where only one company receives an aid for a certain reason. european journal of government and economics 1(1) 37 socio-economic condition. we control for the socio-economic condition in a canton. as suggested in the literature, the economic condition and quality of life in a specific region influences the evaluation of government performance made by voters in that region. to control for these effects we use a socio-economic canton rank as proposed by sanderson and eggerickx (2010). a low value for the index represents a high ranking and thus a ‘better’ canton. this index is an overall performance indicator and includes a canton’s scores for different parameters, measuring the general well-being of inhabitants. to control for economic performance, income per capita and the unemployment level are taken into account. to have a broader picture, also measures of more general well-being are included such as quality of life, living environment and the availability of public services. these aim to quantify certain topics that are important in the daily life in local communities. the first two are based on quantifiable dimensions such as the quality of houses (measured amongst others by the number of bathrooms), built surface, proximity and surface of green and wooded areas, health, life expectancy and environmental concerns such as local emission rates. availability of public services will typically measure the accessibility of several kinds of services such as public schools, administrative services, public nursing homes, public health care institutions, doctors, and public transport. here, it is the proximity and quantity that is measured, as well as the availability of public buses. each time, the partial scores are added and cantons are ranked according to the total score. we expect better socio-economic conditions to result in better electoral results for incumbent parties. results by means of a classical ols regression, we explain the number of votes that incumbent parties of the 2004-2009 period receive at the flemish elections of 2009. regression results can be found in tables 1 and 2.9 in table 1 we look at the absolute value of subsidies as one of the independent variables. table 2 shows results when we change to subsidies per capita. the second column of each table explains the joint vote share of all incumbent parties. in column three the relative change in vote share each time is the dependent variable.10 first, it is clear that there is a very strong and positive link between the election outcome of 2004 and results of 2009. incumbents of the 2004-2009 period are more likely to have many votes in a canton at the 2009 elections when they had a high vote share at previous elections, despite some ‘regression’ in the sense that high shares in 2004 tend to be followed by a loss of votes. next, we take a look at subsidies and subsidies per capita going to firms. throughout the four estimations, we find that there is a significant positive relationship between subsidies and the vote share in 2009, meaning that cantons that receive high subsidies the year previous to the elections reward incumbent 9 including a dummy variable for cantons containing one of the five province capitals does not alter results significantly. one could argue that this kind of variable should be included as it is not unusual for firms to have their headquarters in a larger city, but their main activity somewhere else. also, cities typically host more workers living somewhere else. subsidies might enter the calculations of the bigger cantons, but employees are from another region. including a dummy variable for ‘undecided’ cantons, i.e. cantons where the largest party changed between the 1999 and 2004 elections or cantons were vote share for incumbents and opposition was close, does not alter the results.we also estimated our regressions applying robust regression techniques. the results do hardly differ from the ones presented (results can be obtained from the authors upon request). including a variable to control for size of the cantons does not change the results. the size variable does not reach significance. an overview of all variables included and their sources can be found in appendix 2. 10 by relative change in vote share we mean the percentage change in vote share of the incumbent parties between the 2009 and 2004 elections: (vote share 2009-vote share 2004) /vote share2004. european journal of government and economics 1(1) 38 parties with more votes than cantons that receive less subsidies. an increase of €1 subsidy per capita, increases vote share for incumbent parties with 0.0021 percentage points (table 2). €80 subsidy per capita is granted on average. for the canton with the highest subsidies, this represents a change in vote share of 6 percentage points, which is substantial given the number of parties participating at the elections. the found effect confirms expectations. the subsidies taken into account are the ones going to firms. these matter because they are observed by the voter public in the following way: many people work close to home or at least have many relatives and/or friends working in that region. if more subsidies are awarded to firms in a specific canton, we can expect that they are ‘experienced’ by more voters as more employees will be aware of the subsidies. voters are informed by personal experience, but can also learn about the subsidies through their social network or local press. then, we continue with the reasoning by dewatripont and seabright (2006) that voters indeed reward politicians for subsidies. the subsidies are perceived as a kind of effort of politicians. the economic impact of the spending decision can often only be evaluated after the elections. therefore, voting support is given on the basis of the spending decision itself and not on the actual effectiveness of the subsidy which can only be observed at a later moment in time and is then implicitly incorporated in the variable measuring general well-being.11 third, following the literature, we control for the socio-economic condition of the cantons. we expect that a good economic condition and quality of life would be rewarded with more votes. this is confirmed by the results for the variable measuring the socio-economic condition. the sign observed is negative, remembering that the variable is an ordinal variable, with higher values pointing at worse socio-economic conditions. cantons were inhabitants have a high feeling of general well-being, measured as economic condition, but as well as quality of life and availability of public services, award more votes to incumbent parties.12 table 1: results with subsidies as an absolute variable incumbents 2009 relative change in vote share incumbents 2004 0.5520*** -0.0067*** subsidies 0.0024** 0.000041** socio-economic index -0.0153*** -0.0002*** constant 33.9927*** 0.4974*** n 103 103 r² 0.6223 0.5308 *** significant at 1% level; ** significant at 5% level; * significant at 10% level 11 this paper focuses on a type of spending that is likely to be sensitive to becoming ‘targeted’ spending. building on existing literature, it could be interesting to look at the effect of more general types of public spending. this however falls outside the scope of this paper. following the method proposed by petrarca and padovano (2011), we also test whether there is an influence of the level of subsidies in neighbouring cantons. we calculate per canton the level of subsidies relative to the subsidies received in neighbouring cantons. this however does not seem to have an effect. 12 substituting the socio-economic index by the variables income and unemployment does not substantially change the found effect of subsidies on voting outcome. the variable income is positively correlated to voting outcome as is found in the literature. the variable unemployment shows the expected negative sign, but does not reach significance. european journal of government and economics 1(1) 39 table 2: results with subsidies per capita incumbents 2009 relative change in vote share incumbents 2004 0.5345*** -0.0070*** subsidy per capita 0.0021* 0.0001** socio-economic index -0.0143*** -0.0002*** constant 35.0201*** 0.5136*** n 103 103 r² 0.6122 0.5176 *** significant at 1% level; ** significant at 5% level; * significant at 10% level most usual regression diagnostics do not raise concerns. when checking multicollinearity, we find no variance inflation factors that exceed 2. as the normally used threshold is 10, we conclude that there are no concerns with regard to multicollinearity. the augmented partial residual plots do not raise concerns about non-linearity. when plotting residuals versus fitted values, we see that the data cloud becomes somewhat thinner towards the ending which might raise concerns about heteroskedasticity. these are however only minor as there is no real pattern to be found in the plot. however, subsidy per capita has a few outliers. in order to check whether these influence the results, we run regressions with and without them and find no significant differences. robust regression was also used, which provided similar results (see also footnote 9). finally, one might be critical about the assumed direction of causality between subsidies and election outcome. instead of higher subsidies leading to better election outcome, one could argue that more subsidies are awarded to regions where incumbent support was rather low in the past. theoretically, this cannot be a problem in the present case as election results of 2009 are explained by subsidies of 2008. nonetheless, we check endogeneity by means of a durbin-wu-hausman test (as one could say that subsidies are dependent on previous election results and that these are highly correlated to future election outcome), but conclude that there are no problems of that kind. in addition and perhaps most convincingly, when the opposite direction is tested, we find no evidence of a correlation between subsidies and 2004 vote share. considering these diagnostics, results can be confidently interpreted. conclusion and discussion we study the impact of subsidies to firms on election outcome in flanders (belgium). to that end, we look at the election outcome for the incumbent parties at the flemish level in the 2009 elections for all 103 cantons. we find a significant positive relationship between the amount of subsidies and subsidies per capita granted to a canton’s firms and the vote share of the incumbent parties, meaning that voters seem to reward politicians for spending resources on subsidies. the motive for this paper is to uncover a possible explanation for granting subsidies. apart from the overall consensus that state aid should be forbidden, literature provides evidence that many of the subsidies that are still in place for a variety of reasons are not efficient. the question then remains why governments keep on investing such large amounts of resources in different kinds of subsidies. following the theoretical work of dewatripont and seabright (2006) who explain that subsidies are a by-product of political accountability, we show empirically that voters seem to reward politicians for subsidies as higher amounts of subsidies result in more support for the incumbent parties. we do however not judge the effectiveness of the subsidies. we also provide a better understanding of voting behaviour in general, adding an index containing the usual control variables that we find in the existing literature. up to now, economic voting is more persistently found in research on national european journal of government and economics 1(1) 40 elections than on the local level, as is explained by berry and howell (2007). the present study shows that there can be retrospective voting at the local level. we add the possibility that the accountability mechanism can create undesired effects as well when voters make a ‘wrong’ assessment on a government decision under the limited information that they observe as will often be the case with regard to subsidies. the literature on pork-barrelling is specified further as we distinguish one specific type of spending, namely subsidies to firms. even in stable democracies and multi-member electorates, both usually less sensitive to porkbarrel spending, awarding subsidies can improve the incumbent’s chances of reelection. however, this is not always confirmed in existing literature. so, specific conditions can be defined further for this kind of spending to work. probably, the small size of the cantons and the organization thereof is an important factor. having relatively dense social networks and a truly local press, for example, helps politicians to spread information among the voter public. this makes it easier to bring the message across that resources were spent, and therefore, as dewatripont and seabright (2006) pointed out in their theoretical model, that incumbents ‘care’, resulting in extra support during upcoming elections. for future research it would be interesting to look at this type of effect in other countries and especially to verify whether the ideology of the standing government matters for the effect of subsidies on voting outcome. two remarks still need to be made: as we studied flemish election outcomes, one could argue that only flemish subsidies should have been taken into account. however, we included all subsidies received by firms. this is justified by the fact that first, most of these subsidies come from the flemish government or an organization under its responsibility. second, most voters do not distinguish between regional and federal funds in their evaluation of government performance. and even when they do, literature provides proof that it is possible for voters to reward politicians at one level of government for spending stemming from another level. a second remark, as mentioned earlier: flanders is a very specific case and further research is thus necessary to check the extent of the found relationship in other countries. quite substantial policy implications can be derived from the results obtained. as politicians are rewarded for subsidies by voters, they will probably put a rather high effort into granting subsidies. from this point of view, we can thus conclude that a control on state aid is absolutely essential. this control will need to be organized at the national as well as at the supranational levels. the european commission, for example, has the jurisdiction to decide on the authorization of state aid measures planned by eu member states. however, not all subsidies fall under the notification obligation for a variety of reasons such as no effect on trade between member states, block exemptions, and the de minimis regulation. for this rather considerable amount of financial resources that is spent on subsidies to firms, there should be an independent control system in place at the national level as well, to ensure that there are no harming consequences stemming from subsidies and to control for possible misallocation as a result of political motives. references alvarez, michael and janson l. saving (1997) ‘deficits, democrats and distributive benefits’, political research quarterly, 50(4): 809-831. baldwin, richard e. and frédéric robert-nicoud (2007) ‘entry and asymmetric lobbying: why governments pick losers’, journal of the european economic association, 5(5): 1064-1093. bergström, fredrik (1998) ‘do capital subsidies to firms increase employment?’ in eliason gunnar and nils karlson (eds.), 2001, the limits of government: on policy competence and economic growth, city university press:stockholm. european journal of government and economics 1(1) 41 berry, christopher r. and william g. howell (2007) ‘accountability and local elections: rethinking retrospective voting’, the journal of politics, 69(3): 844-858. books, john and charles prysby (1999) ‘contextual effects on retrospective economic evaluations, the impact of the state and local economy’, political behaviour, 21(1): 1-16. chindooroy, rohan, patrice muller and giovanni notaro (2007) ‘company survival following rescue and restructuring state aid’, european journal of law and economics, 24(2): 165-187. collie, david r. (2000) ‘state aid in the european union: the prohibition of subsidies in an integrated market’, international journal of industrial organization, 18(6): 867-884. collie, david r. (2002) ‘prohibiting state aid in an integrated market: cournot and bertrand oligopolies with differentiated products’, journal of industry, competition and trade, 2(3): 215-231. david, paul a., bronwyn h. hall and andrew a. toole (2000) ‘is public r&d a complement or substitute for private r&d? a review of the econometric evidence’, research policy, 29(4-5): 497-529. dewatripont, mathias and paul seabright (2006) ‘“wasteful” public spending and state aid control’, journal of the european economic association, 4 (2-3): 513-522. duch, raymond m. and randy stevenson (2006) ‘assessing the magnitude of the economic vote over time and across nations’, electoral studies, 25(3): 528-547. european commission (2010) ‘studies and reports: total state aid by member state as a percentage of gdp’, url (consulted on 12 dec. 2011): http://ec.europa.eu/competition/state_aid/studies_reports/expenditure.html. european union (2008) ‘consolidated versions of the treaty on european union and the treaty on the functioning of the european union’, official journal c 115 of 9 may 2008. feldman, paul and james jondrow (1984) ‘congressional elections and local federal spending’, american journal of political science, 28(1): 147-164. gelabert, liliana, andrea fosfuri and josep a.tribo (2009) ‘does the effect of public support for r&d depend on the degree of appropriability?’, the journal of industrial economics, 57(4): 736-767. glowicka, elzbieta (2008) state aid and competition policy: the case of bailouts in the european union, dissertation, humboldt-universität berlin. hagerty, michael r. (2006) ‘quality of life from the voting booth: the effect of crime rates and income on recent u.s. presidential elections’, social indicators research, 77(2): 197-210. hancher l., t. ottervanger t. and j. p. slot (2006) ec state aids. london: sweet and maxwell. hauk, willian r. and romain wacziarg (2007) ‘small states, big pork’, quarterly journal of political science, 2: 95-106. herron, m.c. and shotts k.w., 2003 ‘winners, losers and the congressional pork barrel’, working paper, northwestern university. jegers, marc and hilda theunisse (2007) ‘elementen van boekhouden en analyse van jaarrekeningen’ (elements of accounting and financial statement analysis). brussel: vubpress. johnston, ron, charles pattie, daniel dorling, iain macallister, helena tunstall and david rossiter (2000) ‘local context, retrospective economic evaluations, and european journal of government and economics 1(1) 42 voting: the 1997 general election in england and wales’, political behavior 22(2): 121-143 johnston, ron and charles pattie (2002) ‘geographical scale, the attribution of credit/blame, local economic circumstances, and retrospective economic voting in great britain 1997: an extension of the model’, environment and planning c: government and policy, 20(3): 421-438. keefer, philip and stephen knack (2007) ‘boondoggles, rent-seeking, and political checks and balances: public investment under unaccountable governments’, the review of economics and statistics, 89(3): 566-572. kwon, hyeok y. (2005) ‘targeting public spending in a new democracy: evidence from south korea’, british journal of political science, 35(2): 321-341. lach, saul (2002) ‘do r&d subsidies stimulate or displace private r&d? evidence from israel’, the journal of industrial economics, 50(4): 369-390. leigh, andrew (2008) ‘bringing home the bacon: an analysis of the extent and effects of pork-barreling in australian politics’, public choice, 137(1-2): 279-299. levitt, steven d. and james m. snyder (1997) ‘the impact of federal spending on house election outcomes’, the journal of political economy, 105(1): 30-53. lewis-beck, michael s. and martin paldam (2000) ‘economic voting: an introduction’, electoral studies,19: 113-121. limosani, michele and pietro navarra (2001) ‘local pork-barrel politics in national pre-election dates: the case of italy’, public choice, 106(3-4): 317-326. london economics (2004) ‘ex-post evaluation of the impact of rescue and restructuring aid on the international competitiveness of the sector(s) affected by such aid’, final report to the european commission, enterprise directorate general. manzetti, luigi and carole j. wilson (2007) ‘why do corrupt governments maintain public support?’, comparative political studies, 40(8): 949-970. martin, stephen and paola valbonesi (2008) ‘equilibrium state aid in integrated markets’, the be journal of economic analysis and policy, 8(3) article 33. mueller, dennis c. (2003) public choice iii. cambridge university press :cambridge. nannenstad, peter and martin paldam (1994) ‘the vp-function: a survey of the literature on vote and popularity functions after 25 years’, public choice, 79(3-4): 213-245. pattie, charles, daniel dorling and ron johnston (1997) ‘the electoral geography of recession: local economic conditions, public perceptions and the economic vote in the 1992 british general elections’, transactions of the institute of british geographers, new series, 22(2):147-161. pattie, charlie and ron johnston (2008) ‘positional issues, valence issues and the economic geography of voting in british elections’, journal of economic geography, 8(1):105-126. petrarca, ilaria and fabio padovano (2011) ‘from taxes to politics, from politics to taxes: evidence of yardstick competition in the italian municipalities’, condorcet center working paper. url (consulted 24 may 2012) : http://crem.univrennes1.fr/wp/2011/2011-01-ccr.pdf. robinson, james a. and ragnar torvik (2005) ‘white elephants’, journal of public economics, 89: 197-210. sanderson, jean-paul and thierry eggerickx (2010) ‘des outils d'aide à la décision : les indicateurs de condition de vie et de mixité socio-démographique des european journal of government and economics 1(1) 43 communes et des quartiers en belgique’, cahiers de démographie locale, 2: 115158. stein, robert m. and kenneth n. bickers (1994) ‘congressional elections and the pork barrel’, the journal of politics, 56(2): 377-399. tannenwald robert (2002) ‘are state and local revenue systems becoming obsolete?’, national tax journal, 55(3): 467-489. verdier, daniel (1995) ‘the politics of public aid to private industry-the role of policy networks’, comparative political studies, 28(1): 3-42. appendix 1: calculating the amount of subsidies received to calculate the amount of subsidies that each firm received, we follow jegers and theunisse (2007). they propose the following formula: investment subsidies = difference in book value of investment subsidies (∆15) + investment subsidies recognized as revenues (proportional to depreciation of subsidized asset) (9125) +difference in book value deferred taxes (∆168)-transfer to deferred taxes (680)+transfer from deferred taxes (780)13 as subsidies increase later taxes, part of the subsidy obtained is booked as deferred tax. the last three terms allow us to estimate this part. the first two terms calculate the net subsidy received. appendix 2: variable description and sources variable name description source incumbents 2009 independent variable 1: vote share of incumbent parties at the 2009 elections per canton (in%) http://www.binnenland.vlaander en.be/verkiezingen relative change independent variable 3: growth of vote share per canton own calculation from: http://www.binnenland.vlaander en.be/verkiezingen incumbents 2004 vote share of incumbent parties in the 2004 elections (in%) http://www.binnenland.vlaander en.be/verkiezingen inhabitants number of inhabitants per canton (only used to calculate per capita variables) http://statbel.fgov.be/nl/modules/ publications/statistiques/bevolki ng/ subsidy and subsidy per capita amount of subsidies and subsidies per capita going to each canton calculated as described in appendix 1 socio-economic index ranks municipalities according to their socio-economic performance. (http://knack.rnews.be/ last consulted: 25. march 2010) sanderson and eggerickx, 2010 13 numbers between brackets refer to the codes in the belgian annual accounts. european journal of government and economics volume 2, number 1 (june 2013) issn: 2254-7088 59 ceausescu’s population policy: a moral or an economic choice between compulsory and voluntary incentivised motherhood? florin s. soare, university of bucharest and institute for the investigation of communist crimes and the memory of the romanian exile (iiccmer), romania abstract the purpose of this paper is to explain why, in 1966, the romanian leadership adopted a wholly restrictive pronatalist policy, based on the strict limitation of abortion, instead of one based on socioeconomic incentives to families, as suggested by technocrats. previous literature shows disagreement on whether the choice was motivated by moralistic or economic considerations. in order to find an answer to this question, hundreds of pages of archival material unpublished so far have been analysed, including the minutes of the central committee of the communist party, statistics, documents identified in the ministry of health archive, and the technical reports that were on the table at the time of the decision. the conclusion of this study, drawn on the basis of these documents, indicates that at the time of 1966, regardless of the suggestions of the technocrats, a decision had already been taken by ceausescu himself. this decision was influenced directly by economic considerations, namely the wish to obtain the maximum pronatalist effect at a minimum budgetary cost. jel classification j10; j13; j18 keywords pronatalist policy; abortion; romanian communist party; birth rate european journal of government and economics 2(1) 60 introduction for almost two decades, the ceausescu regime applied a pronatalist policy, the character of which was greatly influenced by a limited vision of the problems of the population by the political leaders, who had a traditionalist vision of the family and its role, and a tendency to look for solutions in the stalinist model. in communist romania, the enactment of restrictive abortion legislation emerges after 1966 as a special kind of intervention of the state in the private life of the individual, as a way to bolster its control over the population. the ban on abortion, as an attempt to control fertility and influence the natural movement of the population, was a widely popular measure in (but not restricted to) totalitarian systems, being one of the lynchpins of population policy. in the majority of cases, this represented only one of the components of a demographic programme. in general, we can identify three pillars of a policy meant to increase the number of births: (1) the restriction of abortion, (2) the use of propaganda, and (3) economic and social measures designed to stimulate births. the prevalence of one of the three, or rather, of the first two, defines a regime as restrictive and coercive, centred on the drastic limitation of abortion. why did romania implement such an aggressive population policy? why did the prohibition of abortion constitute the central element of this policy? these questions have multiple potential answers, which will be analysed in this article, beginning with economic and ideological motivations and the way in which these influenced the restrictive character of the legislation adopted after 1966. the study thus addresses two different questions. the first tries to find explanations for the ambitious expansionary nature of romania’s demographic policy, in general. the second question tries to make sense of the overwhelmingly restrictive nature of the policies implemented. in order to answer these questions, the article presents an account of the way in which romania’s 1966 pronatalist legislation was adopted. it also discusses the general characteristics of this population policy, and presents some of the policy alternatives that were debated at the time. all these aspects are analysed on the basis of source material coming from public discourse and the internal debates of the romanian communist party (rcp), as well as consultations with medical specialists on the issues of demography and natalism. the article presents the two main pronatalist options that were considered at the time: on the one hand, the technocratic option, contained in two documents elaborated by the ministry of health; on the other hand, the political vision of the executive committee of the central committee of the rcp resulting from its two meetings held in august and september 1966. it also describes the consultation with the medical staff that took place on 20 september 1966 in order to ensure their consent and the legitimacy of the final decision. in contrast to the existing literature, which has focused on other aspects, such as the negative consequences of the policy, this article tries to explain the motivations of political leaders who were responsible for the decision, and their reasons for rejecting the technocratic proposals presented by medical specialists. the concluding part of the study highlights the role played by economic considerations in the population policy choice of the ceausescu regime, by comparing the budgetary impact in terms of expenditure and revenue of both the technocratic and political solutions. historiography of the subject the theme of pronatalist policies of contemporary states is still quite rife with controversy, being highly debated in romanian and western historiographies. scientific research devoted to the topic, both in romania and in the west, is soare ● ceausescu’s population policy 61 generous. a large number of studies and volumes by historians, sociologists, and demographers were published after 1989, the majority of which focused on the negative demographical, medical and social outcomes of the post-1966 pronatalist policy. they detail the high level of maternal and infant mortality, the physical and mental trauma of women, the abandonment and institutionalisation of children, the appearance of congenital diseases among them, and the appearance and proliferation of aids among institutionalised children (hord et al 1991; johnson 1991; kivu 1993; kligman 1992; şerbănescu et al 1995). raw data on the pronatalist policy, studies on social memory, oral history, gender studies are supplied by anton (2007; 2009), baban (1999; 2000), bărbulescu (1998), betea (2004), bodeanu (2002) and pop-eleches (2006, 2010). gail kligman (1998) offers the first complex synthetic work dedicated to the demographic policies of the ceausescu regime. kligman’s work constitutes a documentary analysis of the pronatalist policies of the regime from a number of perspectives: legislation and its application, communist propaganda, the manner in which the entire society was constrained to submit to the new rules, and the social problems which beset romania at the beginnings of the 1990s as a direct consequence of those measures. the work has a few caveats, the most important of which being related to the scarcity of archival documentation, which is easily explainable due to restricted access in the 1990s. in recent years, the debate in the field of historiography regarding the population policy of the ceausescu regime has become more nuanced due to the opening of the archives of some communist state institutions. the discussion has moved from consequences to causes, with a focus on understanding the implementation of pronatalist legislation. researchers such as pălăşan (2009), doboş (2012), jinga and soare (2011) have attempted to demonstrate the exceptional character of the post-1966 romanian population policy. the current study brings a new perspective on this problem, on the basis of sources obtained from the archive of the ministry of health and the romanian communist party, documents that greatly clarify the motivation of the political leadership to opt for a restrictive policy, as well as the arguments against the adoption of the technocratic initiatives. justification of romanian pronatalism economic arguments. in 1965, romania was in demographic decline, manifested especially through the decline in births, a situation which was similar to that in the majority of centraland eastern-european states. according to the census of 15 march 1966, romania had a population of only 19,105,056 inhabitants,1 and births were only at a rate of 14.6 percent, an indicator that ‘did not even ensure the simple reproduction of the population, that is one female child for each woman in her fertility period from 15 to 49 years of age’.2 1 comunicat cu privire la rezultatele preliminare ale recensământului populaţiei şi locuinţelor din 15 martie 1966 [statement on preliminary results of population and housing census of march 15, 1966] arhivele naţionale istorice centrale (a.n.i.c.), fond c.c. al p.c.r., secţia organizatorică, dosar 30/1966, f. 1. 2 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, f. 106. european journal of government and economics 2(1) 62 figure 1. total fertility rate (tfr) in romania, 1956-1966 source: national institute of statistics, romanian demographic yearbook, 2006: 120. demographic considerations notwithstanding, the new regime that ruled over bucharest was genuinely worried about the obstacles this would pose for the construction of socialism, for industrial development, due to the existence of a population ‘with a structure based on old-age groups’. even during the existence of the regime, mary ellen fischer (1985: 125) argued for the economic motivation for the adoption by the regime of a coercive policy designed to raise birth-rates. the economic argument rests in the stalinist dogma, according to which more people could generate a stronger economy. the need for work-force became more poignant in the 1950s, due to the ample industrialisation process of gheorghiu dej, intensified further after his refusal of the soviet economic integration plans.3 from an economic standpoint, the regime’s population policy was centred on the expansion of the work-force. this was realised in the 1950s through the expansion of the female work-force. once these reserves were exhausted, ceausescu’s plans became centred on the encouragement of population growth by means of pronatalist policies. thus, we may admit the fact that one of the main reasons behind the adoption of a pronatalist policy was the need to create and maintain a supplementary work-force, necessary for the industrialisation process. the regime’s propaganda constantly linked population policy to the economic development of romania, population policy being considered to be ‘an integral part of the economic and social policy of the state’ (pavelescu, 1977: 8). an interdependency of economic and demographic growth was invoked: ‘the country’s economic growth cannot be envisioned without demographic growth, without complex measures to raise birth rates and maintain them at a relatively high level’ (trebici, 1971: 26). 3 at the beginning of the 1960s, the soviet union proposed a project of economic organisation in eastern europe that sought the specialisation of the economies of these countries on the production of certain items (the e.b. valev plan). romania was to become a predominantly agricultural country. gheorghiu dej understood this plan as one that undermined his authority, and he rebuffed it, and intensified the industrialisation process even more by developing common projects with yugoslavia, and bought western technological developments. in this context, the need for a sufficient workforce became essential (see retegan, 2002). soare ● ceausescu’s population policy 63 figure 2. the dynamics of the labour movement in industry by gender in romania 1957-1965 (in thousands) source: national institute of statistics, romanian statistic yearbook, 1957-1966. in the planned economies of the communist states, control over the population was considered to be highly necessary. according to communist ideology, population has a triple role: work-force, a subject of income, and a consumer of the goods that were created, so the state ‘has to provide the policies for the population, in the same way it plans the national economy, in order to respond to the overall interests of society’ (trebici, 1971: 39). the discussion was permanently centred on the way in which this intervention should be done. in the report of the committee on the study of the measures designed to improve the national birth rates, presented in the session of the executive committee of the central committee of the rcp of august 1966 (the basis for the preliminary discussion on the adoption of pronatalist policies), it was stated that the entire set of measures destined to raise birth rates was determined in an objective manner by the need to ensure an adequate work force for the development of the romanian economy, especially after 1980.4 in the rcp manifesto of 1974, the sustained growth of the population was thought to be the essential factor for the dynamism and the productive strength of society. the target was the attainment of an active work-force of 11.5 million persons by the year 1990 (partidul comunist român, 1975: 92). to the above-mentioned economic considerations, we must add political and ideological arguments. referring to mcintosh (1982: 277), there are three types of state attitudes toward population: population perceived as being linked to national power, the role that population plays in the productive economy, and the role of the state in society. according to this scheme, we may affirm there is a close link between population and the power of the state, as the number and quality of people greatly determines the power of the state and the policies that it promotes at a national and international level. the same hypothesis was put forward by adrian cioroianu, who attempted to link nicolae ceausescu’s admiration for the french president, charles de gaulle, and the latter’s population policy, generated by his obsession for the growth of the population (cioroianu, 2012). a second model for the population policy of 4 studiu privind situaţia natalităţii din republica socialistă românia şi măsuri de redresare a natalităţii din ţara noastră, a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, f. 108. european journal of government and economics 2(1) 64 bucharest was undoubtedly the soviet model. the soviet answer to the birth rate issue is exceptionally important for understanding the measures adopted by the ceausescu regime. the stalinist model of the 1930s-40s was a source of inspiration, from its severe restriction of abortions and the limitations it imposed on divorces, to its large-scale use of pronatalist propaganda (goldman, 1993). in his volume of interviews entitled ‘i was a carver of chimeras’, dumitru popescu states that one of the main reasons why ceausescu opted for his population policy was his complex regarding ‘the withering of the territory he ruled’ (popescu, 1993: 307). certainly, ‘solving the population problem’, represented one of the major objectives of his 23 years of pronatalist policy. the reason was, as mentioned in the health ministry report above, the need to maintain a certain place of his country, among all the countries of the world.5 the promotion of a policy that would increase the number of inhabitants served, at the same time, the nationalist goals of the ceausescu regime, visible during the proceedings of the world population conference in bucharest in 1974, during which population was referred to as ‘a component of national power’, and ‘population policy’ became an ‘attribute of state sovereignty’ (ceausescu 1981). in 1966, the regime showed concern about the possibility of only limited growth of the population (21.5 million people projected for the year 2000), while at the same time, a 53 percent growth rate was recorded for the developed states of the world.6 the authors of the study entitled ‘…on the situation of natality…’ argued that, in order for the same rate to be achieved in romania, a 23 percent rise in births would be necessary so that the population would reach 29 million by the year 2000. in 1966, romania had a population numbering 19,083,443, and a natural population growth of 6.1 percent. the comparison to the developed states, which systematically appears in nicolae ceausescu’s discourse, was not the only argument, but, especially after 1968, it was coupled with the obsession of a foreign threat.7 demographic growth was beginning to be seen as a source of military power, as it was directly related to the number of persons that the state would be able to mobilise in the case of war. the problem was highlighted in the context of the elaboration of a new defence doctrine. ‘the doctrine of the fight of the entire people’ and the role of demography for the security of the nation were constantly espoused by ceausescu (1970: 300) as: ‘…an anti-imperialist war, a war of defence cannot be anything but a popular war, and the victory would be obtained not only on the battlefront but through the general engagement of the entire people.’ beyond this line of thought, which gained strength during the 1970s, in 1966 the political leadership did not view the problem of population in its complexity, but sought simple solutions that were cost-effective and produced quick results. this may be the reason for the moral arguments that were called for during the sessions of the executive committee of the party, prior to the adoption of the decree 770 of 1966. ‘moral traditionalism’, invoked during the session, cannot be excluded as one 5 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, f.108. 6 to date, i have been unable to identify the un study to which the document (studiu privind situaţia natalităţii) refers . therefore we cannot regard the data used by the political establishment in 1966 in bucharest as reliable when discussing the necessity of implementing a new programme for the increase of the birthrate. see studiu privind situaţia natalităţii din republica socialistă românia şi măsuri de redresare a natalităţii din ţara noastră, a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, f. 104. 7 several members belonging to ceausescu’s inner executive circle, among whom the ex-foreign minister between 1978-1985, stefan andrei, agree with the fact that the communist leader was genuinely concerned about a soviet military intervention, both during the year of 1968, and even after towards the end of the regime (betea 2011). soare ● ceausescu’s population policy 65 of the considerations for banning abortion, but this represented, alongside the medical argument, only the pretext, not the reason for implementing ceausescu’s pronatalist plan. one must not forget that the severe limitation of abortions was but one of the component elements of the global policy of the regime. it is difficult to state that moral arguments took precedence over concrete political and economic motivations. in 1966, romania was in full demographic transition, a process which marks the passing from a demographic regime marked by high numbers of births and mortality, to one defined by low birth rate and mortality. the phenomenon was due to a number of socio-economic factors (industrialisation, urbanisation, the integration of women into the workforce, the improvement of medical services), demographic factors (the drop in the number of marriages, the rise in the number of divorces, the rise in the number of abortions), cultural and educational factors (the rise of the duration of obligatory schooling, the rise in the quality of life, the drop in influence of the religious factor), legislative factors (the introduction of liberal legislation concerning divorces and abortions), etc. as far as the demographic transition is concerned, romania was a few steps behind even in relation to its eastern european neighbours, due to the lateness in the onset of modernity: the low degree of urbanisation, industrialisation, the improvement of living standards, the improvement of hygiene and health provisions, and so forth (trebici, 1981). all these elements were cast aside by those who decided romania’s population policy. the technocratic option: ‘the study on the situation of natality in the romanian socialist republic’ the regulation of abortions, as a central element of pronatalist policy, did not come about at once. the initiative did not come from the political leadership, as the specialist literature of the topic shows, but from a committee that studied fertility and sterility. this committee, functioning under the auspices of the health ministry, grouped specialists from the fields of medicine, demography, genetics, which in 1965 put together some material presenting romania’s demographic evolution, the reasons for the drop in birth rates (socio-economic, socio-cultural, demographic), and made a demographic projection up to the year 2000. it also proposed a number of socio-economic, cultural, and sanitary-educational measures for the improvement of the natural growth of the population.8 this document was the first one which put forward the need to limit the number of abortions through medical and legal measures. the proposals were based on the consultation of specialists of the ministry of health and the council for obstetrics and gynaecology. the material suggested the implementation of urgent measures for halting declining birth rates. socio-economic and cultural, educational and sanitary measures were proposed.9 it was for the first time since 1957 that a stricter regulation of abortions was proposed. the initiative came from within the ministry of health, and did not envision changing existing legislation, but the introduction of new instructions for applying decree 463 of 1957, which would limit abortions, with a generous list of exceptions (medical and socio-medical), and also the promotion of sexual and contraceptive education among the population. besides protecting the health of women, the measure had a pronounced demographic character, underlined by the 8 ministerul sănătăţii şi protecţiei sociale, unele probleme privind dinamica sporului natural al populaţiei din republica socialistă românia (brochure), 1965. arhiva ministerului sănătăţii, fond cabinet 1, dosar 69/1966, unnumbered. 9 a.m.s., fond cabinet 1, dosar 69/1966, unnumbered. european journal of government and economics 2(1) 66 subtitle of the chapter, which read: ‘proposals for measures concerning the improvement of the dynamics of natural population growth’. figure 3. infant mortality dynamics in romania, 1920-1966 source: united nations (1967: 292) for the period 1920-1945, and muresan (1999: 237-238) for the period 1946-1966. the beginning of the 1960s saw an important rise in the number of requested abortions, from 578,000 in 1959 to 1,115,000 in 1965.10 as infant mortality in this period was in a declining trend (see figure), limiting abortion started to be seen as the solution for increasing birth rates. the study recommended the setup of a committee of members of the state committee of planning, the ministry of finance, health and social provisions, justice, the committee for culture and art, the state committee for work and salaries, the central direction of statistics, the central council of unions, and the national committee of women, which would set up a plan of social, economic, and cultural measures for improving the natural growth of the population. the committee started its work in february 1966 and had the objective of issuing a new report, which would analyse the birth rate in romania and the measures that could be taken to improve it.11 the committee produced two reports: one on ‘the healthcare of the romanian socialist republic: methods of improvement’, and the other one entitled ‘study concerning the situation of natality in the romanian socialist republic, and the improvement of natality in our country’, with an annex on ‘the regulation of abortions’.12 the second one was more important, as it drew up a plan for the improvement of demographic indicators. the study was finalised and debated in the session of the executive political committee of the central committee of the rcp of 2 august 1966. the study on the situation of natality is of special importance, as it reflects the vision of specialists from the state administration concerning both the existing demographic situation and the measures for improving birth rates. the first part of this document consists of a demographic study, an overall presentation of the dynamics of the main indicators (natural growth of the population, general and infant mortality, birth rates), of their evolution in the period between 1938 and 1965, and a comparative study of the situation of 26 european states. the authors consider romanian abortion legislation ‘much too permissive’. 10 a.m.s., fond cabinet 1, dosar 86/1966, unnumbered. 11 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 102/1966, f. 23. 12 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, ff. 102-152. soare ● ceausescu’s population policy 67 at that time, the interruption of the natural course of a pregnancy could be made at the request of the pregnant woman without restrictions (anonymously, and without consent from the husband or family members, in the case of minors). identified as one of the causes of low birth rates, decree number 463 of 195713 is not, however, attacked. it is even justified at times, as an attempt to reduce the number of clandestine abortions in the second half of the 1950s. we can mention the fact that decree 463 of 1957 was the work of the minister for health, voinea marinescu, the same minister who was at the head of the committee of specialists drafting the above-mentioned study. criticism of the 1957 legislation would have meant selfincrimination. for these reasons, the authors of the study proposed ‘the introduction of restrictions on requested abortions’. in this way, the technocrats admitted the need to revise decree 463 of 30 september 1957 in order to introduce stricter procedures. women who requested abortions had to undergo examination by a specialised committee made up of a gynaecologist, a specialist in internal medicine, and a social worker, who could approve the request only for medical reasons (cases where the foetus or the patient were in danger of disease, genetic factors), or socio-medical reasons (physical and psychological deficiencies, social disapproval, rape, incest). if the applicant was under the age of 16, the approval of her parents was necessary, and the intervention could be requested also when the woman already had two live children or was over 40 years of age. according to a proposal of the state planning committee and the state committee for work and salaries mentioned in the study, the approval for abortions could be given in regard to the civil state of the woman, her material conditions or geographic location. it has to be mentioned that, among the socio-medical reasons, only rape, incest, post-45 age, and four children or births will be accepted in the final version, the other suggestions being ignored by the decree. we may also mention the total exclusion of the post-birth evolution of the mother from the discussion. the possibility the she would benefit from an abortion if she became pregnant in the few months following the birth of a child was not taken into consideration even by specialists. after the application of decree 770/1966, and not taking into consideration the numerous complaints that doctors filed to the ministry of health, even the possibility that these women would receive oral contraceptives was denied. i can also mention here that the administration of oral contraceptives was prohibited even for women who, for medical reasons, could benefit from legal abortions. the use of contraceptives was considered by the specialists the most appropriate and safe method of family planning. however, their scarce availability on the domestic market, and a lack of proper education for their utilisation, made them inaccessible for the majority of the population. under these conditions, abortion still remained the most widely spread means of birth control, with negative effects on women’s health. in the 1980s, the minister of health, victor ciobanu, responded to a formal complaint by a gynaecologist concerning the use of contraceptives for women who had medical reasons for abortion: ‘any medical or surgical procedure that suppresses the woman’s ability to procreate constitutes a violation of the party’s and the state’s vision concerning the stimulation of births and is against the law’.14 the authorities position concerning contraception oscillated between marginalisation and banning. in 1966, with the start of the pronatalist policies, the problem of contraception was totally marginalised, because of its contradiction with 13 ‘decretul 463/1957 pentru încuviinţarea întreruperilor de sarcină’ (decree 463/1957 for allowing abortion on request), în buletinul oficial al m.a.n. a r.p.r., nr.26 din 30 septembrie 1957. 14 a.m.s., fond d.a.m.-o.m.c.t., dosar 13/1986, unnumbered. european journal of government and economics 2(1) 68 the main objectives of the policies, but also because of the traditionalist mentality of nicolae ceausescu, and other political leaders, who were sceptic about the whole idea of birth control. in this way, the situation of contraception was maintained in an ambiguous state, not being banned, but at the same time, not being permitted. i am referring here to modern methods of contraception only, such as the contraceptive pill, as other mechanical methods, such as condoms, were available in the market during this period, both from domestic production and imported. in the 1980s, health ministers prohibited modern contraception altogether. the study ends with a number of suggestions regarding the measures which could be adopted for solving the problem of the birth rate in romania. these had a predominantly socio-economic character and projected giving out bonuses for births, an increase in the provision of childcare and the improvement of its functioning, the prolongation of maternity leave, the creation of advantages for employed mothers with respect to pensions, working hours, schedules, medical assistance, and increases in state allowances for children. the socio-economic measures envisioned by the committee for the study of measures regarding the improvement of birth rates would be mostly ignored or rejected for political reasons based on the idea that any profoundly pronatalist policy requires a severe restriction of abortions. because of this, decree 770 would be fundamentally different from the plan proposed by the committee, reflecting a more radical vision on abortions. session of the executive committee of the rcp of 2 august 1966: the political solution15 the session of the party executive committee of 2 august 1966 was important for understanding the vision of the political leadership on the population problem and the way in which demographic indicators could be controlled and regulated. dogmatism and superficiality, coupled with a pretension of omniscience regarding the issues of such a sensitive domain, all represent major sources of the failure of the pronatalist policy and its negative consequences. as i shall show, the main topic of the session was the issue of abortions, while the stimulation of birth rates through economic measures was marginalised and considered somewhat complementary. this point of view would impose itself in the end and would decisively shape the romanian policy as restrictive. an analysis of the discourses of the participants demonstrates the session’s nature as a pure formality: the decision was already taken beforehand. with only one exception, that of ion gheorge maurer, all positions converge toward the severe curtailment of abortions, the limitation of divorces, the utilisation of propaganda, and the economic stimulation of natality. the most hard-line speech belongs to alexandru draghici, former minister of the interior, and head of the securitate in the 1950s, who vehemently condemned the material put together by the committee. his line of argument shows a limited, schematic understanding of the problems of the population, and he directly correlates the rise in the number of abortions to the drop in birth rates. his categorical position reads: ‘we must at once get rid of this sinful decree, which does nothing but encourage loose morals, and family must remain family’. his position underlined once again the conception of the political leadership up to 1989 on the demographic issue, which consisted on the insistence on the psychological factor (comfort, loose morality) as a source of low birth rates. according to this view, the solution could never be the economic stimulation of the number of births, 15 see a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 102/1966. soare ● ceausescu’s population policy 69 but a restriction of abortions and the limitation on the number of divorces. the projected measures should be the most cost-effective as possible. the experience of other european states and their solutions consisting in giving bonuses, stipends, and material aid for families with many children, was considered demagogic: ‘this whole study is a cobbling together of foreign experiences, and there is much demagogy in it, especially concerning bonuses, as if the state should buy these children’. the lengthening of maternity leave and the rise in the provision of childcare were also discarded. the position of the political leadership denoted a lack of proper understanding of the problems involved in raising children. in romania, maternity leave was set in the 1950s at 112 days, 52 days of prenatal leave and 60 days postnatal leave. in comparison, other socialist states had a much higher number of days, such as the german democratic republic, with 182 days (doboş 2010: 276277; klinger 1991: 517), czechoslovakia, 154 days, hungary, 140 days, bulgaria, between 120 and 180 days, depending on the number of previous children.16 the relatively low number of days for maternity leave was justified by highlighting the need of reintegrating women in production and industry. virgil trofin, secretary of the central committee of the rcp, iosif banc, vice-president of the council of ministers, leonte rautu, member of the executive committee of the cc of the rcp, and chivu stoice, president of the state council, all discussed measures related to the drastic sanctioning of doctors, a substantial rise in fees for abortions and divorces, a sustained propaganda campaign, and the establishment of a differentiated tax for childless persons. the most balanced position was that of gheorghe maurer, the president of the council of ministers, who considered the problem to be one of major importance for social evolution, and who was aware that, for its solution, measures would be necessary that went beyond a simple limitation on abortions and divorces. his position is close to that of the specialists from the ministry of health. the limitation on abortions, in maurer’s vision, should not be introduced only for the purpose of raising birth rates, ‘natality can exist and abortions can influence it very slightly’, but to protect the health of women.17 he proved himself to be much more cautious than the rest when talking about limiting abortions. although he declared that ‘this law…was a big mistake and must be ended’, he highlighted the importance of the way in which the law would be applied, establishing a limit on state involvement in the lives of its citizens: ‘after all, the problem of the family is a problem which is solved more and more by the family itself…we must think that we cannot force the family to have more children than it wants to have’.18 the proposed solutions were centred on the modification of the legislation and the need to adopt long term educational programmes. nicolae ceausescu’s intervention at the closure of the debate did nothing else but to loosely draw up his demographic vision, which was to be translated into an aggressive population policy in the coming decades. ceausescu declared himself displeased with the existing situation: ”in my opinion, we have legalised prostitution through abortions and free divorces, with this decree… how is this possible, are we an institution for the encouragement of prostitution or do we have the responsibility to keep the health of the people, the natural growth of the people, to defend the morality of the people?... [t]he problem of natality is not a problem of the desires of one to have or not have children, but a social problem, each man has obligations toward society”.19 this fragment highlights ceausescu’s traditionalist, rural 16 notă privind legislaţia acordării concediului de maternitate în alte ţări, a.m.s., fond cabinet 1, dosar 52/1966, unnumbered. 17 a.n.i.c., fond c.c. al p.c.r. cancelarie, dosar 102/1966, f. 17. 18 a.n.i.c., fond c.c. al p.c.r. cancelarie, dosar 102/1966, f. 17. 19 a.n.i.c., fond c.c. al p.c.r. cancelarie, dosar 102/1966, f. 23. european journal of government and economics 2(1) 70 mentality. on the background of a process of state modernisation (the acceleration of industrialisation, the improvement of medical services, the growth of urbanisation) and the dissolution of the religious factor, the traditional family model became implicitly less popular, and new forms of civic behaviour (abortion, divorce, cohabitation) appeared, which were unacceptable to the regime. in these conditions, the conclusion was as expected: ‘i think we must immediately put an end to abortions’. ceausescu decided that in two weeks time a bill would be presented to regulate the situation of abortions, taking into account the examples of france, the united kingdom and the united states, and other countries with ‘an acceptable level of population growth’, and including the presentation of measures designed to drastically sanction those who break the new rules.20 it is interesting that the three countries are mentioned not as much for their ”acceptable” growth rates, but mainly due to their restrictive abortion legislation. the problem of birth control was absent from the bill, on his indications, ‘as it is not current’. in this manner, the objective of raising birth rates at any cost became evident. the position of medical experts as we can surmise, the decision of limiting abortions was taken before the session of the party executive committee on 2 august 1966, the discussions being purely formal and void of content. a political decision needed unanimous support from specialists, which would confer legitimacy to the initiative. to this end, on 20 september 1966, a consultation of the medical establishment took place in the session of the central committee of the rcp, dedicated to ‘problems regarding the growth of natality and the continuous improvement of the mother and child’, with the participation of ceausescu, paul niculescu-mizil, member of the political executive committee of the rcp, suzana gâdea, president of the national council of women, milică măgureanu, second head of the health section of the cc of the rcp, the new minister of health, aurel moga, and 32 other medical experts, renowned specialists in the fields of gynaecology, obstetrics, paediatrics, pathology, and neurology, as well as regional heads and directors from the ministry, the majority of those involved in the development of the technocratic option. the purely formal character of the session was highlighted by the quasi unanimity of the opinions expressed by the medical experts in favour of the regulation of abortions. the official reason for the organisation of the session was purely consultative, as ceausescu reminded even from the outset: ‘i insisted on having this consultation in order to discuss together the project for a measure which will appear as a decree regarding the issue of abortion, in medical terms, but which is a consequence of the need to ensuring natality and the growth of our population”.21 although the role of the meeting was a purely formal one, that of obtaining legitimacy from specialists for the proposed legislation, there were certain medical specialists that expressed reservations, if not for the measure itself, then for some of its details. the tone was however quite moderate in the confines of the decree itself. no participant disagreed with the severe restriction of abortion, but only with certain provisions of the decree, the most disputed point being that related to the age-limit for abortion. the attitude of ion moraru, secretary general of the ministry, was decisive as he stated that all proposed exceptions do nothing but weaken the efficiency of the decree: ‘if we introduce many, we reduce the content of this act’.22 he stated that 20 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, f. 3. 21 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966. 22 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 14. soare ● ceausescu’s population policy 71 an upper age limit should not be included in the decree due to the fact that the rate of births over the age of 40 is less then 1 percent. also, abortions below the age of 16 should be considered unacceptable, the solution to such problems being the institutionalisation of such children. the serious physical, psychological or sensorial disability of the pregnant women could not justify an abortion, because ‘pregnancy does not harmfully affect disability in the case of a woman who is blind, deaf or mentally disabled’.23 the lack of proper housing was not considered to be grounds for a couple not to have children, as ‘we all come from households with at least 5-6-7 children, from large families with modest conditions, but in which children could be raised, without child care, without subsidies’24 said suzana gâdea, the president of the national council of women. this attitude was shared by ceausescu, who had limited capability for understanding the socio-economic problems of large families, and the major differences related to the raising of children between the urban and rural backgrounds, problems related to living space, security. the large family was, in the past, the traditional, rural, modest family, in which the number of children had an important economic role. within such an environment, a family with 8, 9, or even 10 members was common. the members of the family worked a plot of land together, as it was their only resource of existence. the majority of the ruling party elite came from such traditional, large families, from a rural background. the references by ceausescu to his own experience were a constant: ‘i will tell you but one thing, comrades, in the past we did not have better living conditions but the number of births was higher… i have here a letter from târnăveni from a family which has 13 living children, and the mother gave birth to 17… all these 13 live children, among which a girl who is married and has a child of her own, live in one room… we also were 10 children and lived in one room.’25 the new minister of health, aurel moga, made a strong declaration: ‘we must revise certain formulae in order to disallow any liberalisation and to take as strong as possible measures concerning abortions.’26 ceausescu closed the session with a speech that synthesises his positioning, found later in his pronatalist discourse: closing the gap vis-à-vis european countries on the issue of birth rates,27 the defence of women’s health and the defence of morality, the care and responsibility for the human resources of the country. he concluded: ‘[t]he problem of natality is a duty for each citizen of the motherland, and freedom must be understood in the sense of responsibility that each one has toward the national future.’28 one observation regarding the lack of preparation of the population for the measure of limiting abortion was promptly answered: ’i think the decree is eagerly expected. it is known about for three months. those who wanted to make preparations for it have made them. on the contrary, among workers, peasants, intellectuals, the decree is quite expected. i feel that certain objections come mostly from the very doctors that occupy these offices (in which abortions take place) and create an unfavourable atmosphere 23 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 14. 24 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 31. 25 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 31. 26 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 38. 27 the statistic data reveal a sharp decline of birthrate in most of the european countries, thus such a statement in the particular context seems to be unreasonable. in all the talks dealing with the demographic issue, the communist leader always and obsessively brings about the gap between romania and the other european states in terms of birthrate. 28 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 40. european journal of government and economics 2(1) 72 towards this measure. but, since their number is small, we shall be able to convince them as well.’29 beyond these statements, the goal was clear, and the conclusion of the session reflected the importance and weight of the pronatalist policy as an essential part of the general state development plan: ‘… we should observe that all the measures we adopt today and will adopt in the future must take into consideration the insurance of the birth rate and the natural growth of the population of our country…all health and social considerations should start from that…’30 in 1966, the only major objective of limiting abortions was the rise of birth rates. the defence of the ”morality of the people” and the protection of the health of women are only arguments used to legitimate the restrictive legislation. we must not forget the fact that, after nicolae ceausescu came to power in the second half of the 1960s we saw a general process of liberalisation in romania, the limitation of abortions being somewhat atypical within this trend. the session of the executive committee of the cc of the rcp of 27 september 1966: the adoption of the political alternative protocol 26 of the session of the executive committee of the rcp of 2 august 1966 notes the decision to re-work the ‘study concerning the situation of natality in romania’, based on the observations of the session and the two issues, abortion and divorce. the latter were foreseen to become the object of discussion of a special plenary meeting of the central committee.31 the new material had to consider ‘the complex and realistic solution to the problem of birth rate growth’, in three main directions, the prioritising the limitation of abortions and the establishment of disciplinary measures for those who ‘break the norms of moral behaviour in society’, the use of wide-scale propaganda and a complementary measure, the organisation of child care in cities and villages. during the session of 27 september 1966, the final decision on the decrees 770 and 771 was taken, and they were adopted as they stood. decree 770/1966 for the regulation of abortions was much more intransigent when compared to the form proposed by the specialist committee, both in form and content. if, in the first form of the decree, article 1 stated that ‘abortions on request can be made only in specialised institutions in cases in which there are medical or socio-medical reasons’, the final version read: ‘abortions are banned’, their use is authorised ‘only in exceptional cases’, with article 2 enumerating such exceptions:32 (a) the pregnancy puts the life of the woman in danger, a danger which cannot be averted by other means; (b) one of the parents suffers from a serious disease, which is hereditary, or which determines severe congenital defects; (c) the pregnant woman is severely disabled physically, psychologically or in a sensorial manner; (d) the woman is over 45 years of age; (e) the woman gave birth to and has four children in her care; (f) the pregnancy is the outcome of rape or incest. many of these social conditions did not represent exception criteria in the new project for the decree, being admissible only in extreme circumstances.33 for 29 a.n.i.c., fond c.c. al p.c.r., secţia administrativ-politică, dosar 10/1966., f. 40. 30 a.n.i.c., fond c.c. al p.c.r. administrativ-politică, dosar 10/1966, f. 39. 31 a.n.i.c., fond c.c. al p.c.r. administrativ-politică, dosar 10/1966, f. 2. 32 ‘decret nr. 770/1966 pentru reglementarea întreruperii cursului sarcinii’, în buletinul oficial al r.s.r., nr. 60 din 1 octombrie 1966. 33 a.n.i.c., fond consiliul de stat decrete, dosar 770/1966, f. 1. soare ● ceausescu’s population policy 73 example, the age limit for a woman seeking abortion grew from 40 to 45, and the lower age limit of 16 was dropped. if the woman requesting an abortion thought that she fitted into one of the categories outlined above, she could go before a medical committee, appointed by the executive committee of the popular councils and chaired by an obstetrician-gynaecologist.34 the committee could request specialised clinical examinations, laboratory tests, functional investigations or social inquiries, and in the case the request for an abortion was approved, it chose the location and programmed the surgery, which happened out of working time. in the case in which the abortion was approved, it could be done in the first three months of pregnancy, or in exceptional cases, ‘when a serious pathological condition that endangered the life of the woman was ascertained, the period could be extended to six months.’35 perhaps the most controversial point of the decree 770 is the sixth one, which refers to the realisation of abortions in case of emergency. it left room for interpretation regarding the procedures to be applied, as the doctor was the one who had to evaluate if an abortion was provoked or occurred spontaneously.36 the doctor was constrained to report the case to the district attorney before the procedure, or if impossible, in 24 hours immediately after, in writing. the attorney then ruled, based on the advice of the coroner and other information available, if the intervention was indeed necessary. if not, the doctor risked being incarcerated for a period up to three months (article 482 of the penal code). during the session of the executive committee of 27 september, although he had not previously expressed any of his views, alexandru birladeanu, a member of the executive committee of the cc of the rcp, and first vice-president of the council of ministers, expressed scepticism toward certain aspects of the new legislation. he spoke to the minister of health, aurel moga, regarding the manner in which the doctor should decide or not to do the abortions in cases of extreme medical urgency and whether the pressure of the prosecutor would not lead the doctor to attempt other methods, putting the life of the woman in danger.37 ceausescu answered this question promptly: ‘such cases shall be analysed’, while ioan morary placed the entire responsibility on the shoulders of the doctor: ‘if the woman dies, the doctor will be accused of manslaughter’.38 this latter session was of special significance, due to it being the moment when the decree was finally adopted. the main pronatalist measures, as described above, which represented the political option, were linked to the limitation of abortions, the criminalisation of those responsible, and the limitation of divorces through the complication of divorce procedures. the majority of the elements that were meant to stimulate births socio-economically were dropped. only those which implied a minimal budgetary effort were retained. the extension of child care represented probably the only substantial measure unanimously accepted, due to its nature of being both pronatalist and supportive of the return of women to work. the number of places in child care was insufficient even in the conditions of total fulfilment of the projected numbers. in 1966, romania needed 69,000 places, and only had 11,800.39 up to 1970, 6,500 further places were created, but a large part 34 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966, f. 136. 35 ‘decret nr. 770/1966 pentru reglementarea întreruperii cursului sarcinii’, în buletinul oficial al r.s.r., nr. 60 din 1 octombrie 1966. 36 the spontaneous abortion occurs without and exterior interference mainly due to the mother’s health condition. 37 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 127/1966, f. 22. 38 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 127/1966, f. 22. 39 informare în legătură cu raportul întocmit de comitetul de stat al planificării privind ‘posibilităţi şi soluţii în vederea rezolvării problemei creşelor, a.n.i.c., fond c.c. al p.c.r., secţia administrativpolitică, dosar 8/1967, f.57. european journal of government and economics 2(1) 74 of these were not the result of constructing new facilities, but overcrowding the existing ones. the other measures did not manage to contribute to the significant improvement of the situation of the families with many children. when compared with the technocratic plan, the only benefits that stimulated births were: the rise in the number of places in child care, kindergartens and orphanages, special stipends for births and special aid for pregnancy after the third child, the modification of the working schedule for mothers with children under 7 –priority for dayshifts, priority for rest and treatment facilities for families and their children, priority for assignment of homes for families with children and pregnant women after the sixth month, a fee waiver for kindergartens during regular working hours, healthcare for families of unemployed people with at least three children, a revision of the family code, the code for civil procedure and the legislation concerning the award of orders and medals for women with many children.40 expenses and finances an analysis of state expenditure for pronatalist goals is important in order to understand their essence, which was profoundly simplistic, centred almost exclusively on coercive measures, and in which the stimulation of births had a secondary role, necessary mostly for propagandistic use. in the table below, we may observe the financial costs that were necessary to implement the two plans, the technocratic and the political alternatives, as well as their sources of funding. table 1. budgets of the two alternative pronatalist plans (in million lei) technocratic alternative (version proposed by the committee for the study of measures towards improving the natural growth rate) political alternative (revised version following the meeting from 2 august 1966) required costs (i) birth bonus 628 childbirth allowance 50 increasing the number of nurseries kindergartens and orphanages. 25 increasing the number of nurseries, kindergartens and orphanages. 19.8 additional maternity leave 78 confinement aid for employed mothers 5.5 additional leave for working mothers 25 free healthcare for mothers with children from agricultural cooperatives and non-salaried families. 17.5 child allowance shift from age 14 to 15 years 150 expenditures rest and treatment camps for children from families of employees. 5.3 child allowance granting 100 for rural workers 140 additional funds for orphans 30 total 1,076 total 98.1 sources of revenue (ii) additional tax for employees 830 additional tax for cooperative members 30 increased fee for legal abortions 75 increase fee for divorce 15 increase payroll taxes for unmarried employees and childless couples 210 total 950 total 210 difference between ii-i 126 difference between ii-i +111.9 source: a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 101/1966,f. 127 (pentru varianta comisiei…) şi a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 127/1966, f. 143 (pentru varianta revizuită). the technocratic plan foresaw a total cost of 1,076 million lei, most of this sum coming from a tax on childless employees, cooperating peasants and increased 40 hotărârea comitetului central al partidului comunist român şi a consiliului de miniştri al republicii socialiste românia cu privire la sprijinirea familiilor cu copii, promovarea natalităţii şi consolidarea familiei. a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 127/1966, ff. 161-165. soare ● ceausescu’s population policy 75 fees from abortions and divorces. the budgetary effort would have been in this case 126 million lei, which would have had to come from existing budgetary reserves. the sum is considerable since, for example, the total funding for sociocultural actions in 1965 amounted to 8990.8 million, of which only 158 million for social work (direcţia centrală de statistică 1982: 275). the second approach, the political alternative, foresaw a total expenditure of 98.1 million, ie 977.9 million less than the technocratic plan. it is interesting to note that, after applying the measures meant to stimulate the number of births, the state budget would not suffer, but instead would record a surplus of 111.9 million lei. this difference was estimated strictly on the basis of the increase of taxation on childless persons, the real sum being much larger, if we add the effect of the increase of the sale prices of alcoholic beverages, tobacco and derivatives from january 1967.41 these figures help us identify the sources of the failure of the pronatalist policies of the state, which were made up of poor information on the real needs of families and centred on the direct needs of the state, in a system which constrains to obtain. in the case of romania, the approach of the problem of birth rates was a profoundly ideological one, based on a traditionalist ideology, together with communist principles, which took on the shape of national stalinism. the demographic strategy implemented in romania after 1966 does not contain original elements, as each piece of the pronatalist policy had its counterpart in the stalinist plan of 1936. the benefits granted were great in number, but substantially insignificant. under these conditions, propaganda tended to overestimate the benefits offered by the state, and underplay the coercive measures, which were presented as a show of concern of the party toward the health of the people. conclusion on the basis of the documents analysed in the article we can distinguish between the need to adopt a pronatalist policy in itself, and the restrictive manner in which it was done. if, for the first question, the entire argumentation based on economic, political-ideological, and moral aspects was valid, for the second question, the policy choice was determined by the need to obtain quick results with a minimal economic cost. the debate within the executive committee of the rcp around the two policy alternatives, namely the political option consisting in the imposition of a strict ban on abortion, and the technocratic alternative resulting from consultations with specialist medical staff, was purely formal. the character of the policy adopted was decisively influenced by the political desire to achieve higher birth rates. the minutes of these sessions show a lack of a real debate on the set of pronatalist measures to be adopted. the restrictive and unpopular measures that finally carried their way only needed to be legitimated at that time. essentially, in 1966, the communist state did not show any interest in promoting births through economic incentives. in this paper i have presented an analysis of the budgetary cost of the two alternative approaches to the growth of birth rates. this comparative exercise undertaken by the government at the time when the 1966 decree was being debated points in the direction that the budgetary constraint was a deciding factor in favour of the political alternative based on a strict ban on abortion. the political plan projected a total expenditure of 98.1 million lei, ie 977.9 million less than the technocratic plan. the projected costs of the political option were to be entirely covered by the revenue from taxes imposed on childless employees. thus, the 41 a.n.i.c., fond c.c. al p.c.r., secţia cancelarie, dosar 127/1966, f. 164. european journal of government and economics 2(1) 76 regime focused on the propagandistic role of the measures, permanently present in the communist media, rather than on their direct effect on birth rates. the regulation of abortions represented, during the 23 years of application of the decree 770/1966, the central element of ceausescu’s pronatalist policy, and the one which gave his population policy its overall restrictive character. after 1966, romania singled itself out among the states of the socialist bloc, but even among the states of europe, as one of the states with the most severe population policy. the uniqueness of the romanian case is not due to the restriction of abortion, but the severity of the measure, via the imposition of a strict filter of medical and social exceptions, and, moreover, by the limitation, marginalisation, in a direct and indirect way, of modern contraception and sexual education as a means of family planning. all these aspects are unusual because romania had traversed a period of almost a decade of liberalisation concerning abortions. the entire vision on the growth of births is anachronistic. when the general european tendency was one of liberalisation, romania returned to the stalinist model of the 1930s. references anton, lorena (2007) ‘abortion during communist romania: from its official history and its memory’, in catherine baker, christopher j. gerry, barbara madaj, liz mellish, jana nahodilová (eds.), nation in formation: inclusion and exclusion in central & eastern europe. london: ucl-ssees, pp. 177-195. anton, lorena (2009) ‘abortion and the making of the socialist mother during communist romania’, in lisa bernstein, pamela monaco (eds.), (m)othering the nation: constructing and resisting regional and national allegories through the maternal body. cambridge: cambridge scholars press, pp. 49-61. baban, adriana (1999) ‘‘romania’ in h. p. david (ed.) from abortion to contraception. a resource to public policies and reproductive behavior in central and eastern europe from 1917 to the present. london: greenwood press, pp. 191222. baban, adriana (2000) ‘’women’s sexuality and reproductive behavior in postceausescu romania: a psychological approach’ in susan gal and gail kligman (eds.) reproducing gender. politics, publics, and everyday life under socialism. princeton: princeton university press, pp. 225-256. bărbulescu, elena (1998) ‘femeia şi avortul în perioada 1966-1989’, anuarul de istorie orală 1: 177-193. betea, lavinia (2004) ‘interzicerea avorturilor (1966-1989) ca fapt de memorie socială’ [‘forbidden abortions in the social memory (1966-1989)’], in adrian neculau (ed.) viaţa cotidiană în communism. iaşi: polirom. betea, lavinia (2011) i se spunea machiavelli. ştefan andrei în dialog cu lavinia betea. bucureşti: adevărul holding. bodeanu, denisa florentina (2002) ‘politica demografică a regimului comunist reflectată în revista „femeia” (1966-1989)’, in ghizela cozma, virgiliu ţârău (eds.) condiţia femeii în românia în secolul xx [the condition of women in romania in the twentieth century], vol. ii. cluj-napoca: presa universitară clujeană. ceausescu, nicolae (1970) românia pe drumul construirii societăţii socialiste multilateraldezvoltate, vol. 4. bucureşti: editura politică. ceausescu, nicolae (1981) populaţia şi forţa de muncă. bucureşti: editura politică. cioroianu, adrian (2012) ‘ceauşescu, decretul 770/1966 şi generaţia „decreţeilor”. generaţia şi decretul care au schimbat românia’, url (consulted jan. 2013): http://geopolitikon.wordpress.com. direcţia centrală de statistică (1982) anuarul statistic al r.s.r. 1982. soare ● ceausescu’s population policy 77 doboş corina, luciana m. jinga and florin s. soare (eds.) (2010) politica pronatalistă a regimului ceauşescu 1966-1989. o perspectivă comparativa. iaşi: polirom. fischer, mary ellen (1985) ‘women in romanian politics: elena ceauşescu, pronatalism and promotion of women’, in sharon wolchik and alfred meyer (eds.) women, state and party in eastern europe. durham: duke university press. goldman, wendy z. (1993) women, the state and revolution. soviet family policy and social life, 1917-1936. cambridge: cambridge university press. hord, charlotte, henry p. david, france donnay and meril wolf (1991) ‘reproductive health in romania: reversing the ceauşescu legacy’, studies in family planning 22(4): 231-240. jinga, luciana m. and florin s. soare (eds.), corina doboş and cristina roman (2011) politica pronatalistă a regimului ceauşescu 1966-1989. instituţii şi practici. iaşi: polirom. johnson, brooke, mihai horga and laurenţia andronache (1991) ‘contraception and abortion în romania’, lancet 341: 231-240. kivu, mircea (1993) ‘une retrospective: la politique demographique en roumanie 1945-1989’, annales de demographie historique: 107-125. kligman, gail (1992) ‘when abortion is banned: the politics of reproduction în ceauşescu’s romania, and after, national council for soviet and east european research’, url (consulted jan. 2013): http://www.ucis.pitt.edu/nceeer/1992-80514-kligman.pdf. kligman, gail (1998) the politics of duplicity: controlling reproduction in ceausescu's romania. berkeley, los angeles and london: university of california press. klinger, andràs (1991) ‘les politiques familiales en europe de l'est’, population 46(3): 511-526. mcintosh, alison c. (1983) population policy in western europe: responses to low fertility in france, sweden, and west germany. armonk, new york: m.e. sharpe. mureşan, cornelia (1999) evoluţia demografică a româniei. tendinţe vechi, schimbări recente, perspective (1870-2030). cluj-napoca: editura presa universitară clujeană. pavelescu, ion (1977) 25 de milioane. populaţia sursă a dinamismului economico-social. bucureşti: editura politică. pălăşan, corina (2009) ‘caracterul profund restrictiv al politicii nataliste româneşti’, in ruxandra ivan (ed.) transformarea socialistă. politici ale regimului comunist între ideologie şi administraţie. iaşi: editura polirom. partidul comunist roman (1975) programul partidului comunist român de făurire a societăţii socialiste multilateral dezvoltate si înaintare a româniei spre communism. bucureşti: editura politică. pop-eleches, cristian (2006) the impact of an abortion ban on socioeconomic outcomes of children: evidence from romania, journal of political economy 114(4): 744-773. pop-eleches, cristian, (2004) ‘the supply of birth control methods, education and fertility: evidence from romania’, journal of human resources 45(4): 971-997. popescu, dumitru, (1993) un fost lider comunist se destăinuie. am fost şi cioplitor de himere convorbire realizată cu ion tecşa, ziarist. bucureşti: editura expres. european journal of government and economics 2(1) 78 retegan, mihai (2002) război politic în blocul comunist. relaţii româno-sovietice în anii şaizeci. bucureşti: editura rao. şerbănescu, florina, leo morris, paul stupp and alin stănescu (1995) ‘the impact of recent policy changes on fertility, abortion, and contraceptive use in romania’, studies in family planning 26(2): 76-87. trebici, vladimir (1971) populaţia româniei şi creşterea economică, studii de demografie economică. bucureşti: editura politică. trebici, vladimir (1981) ‘la transition demographique dans la pays de l’europe de l’est: le cas de la roumanie’, atti del seminario su la transitione demografica. interrelazioni tra sviluppo demografico e sviluppo economico, società italiana d’economia demografica e statistica. united nations (1967) demographic yearbook, incl. special topic: mortality statistics ii. new york: united nations. microsoft word ejge_05_2016-010.docx the pe amid in ekaterina alyona art abstrac russia has soviet un cohabitatio trend throu this researc of entrance union is im aim, such m cohabitatio proportion grow into a cities have childrearing social insti implications keywor first matri sequence jel cla j12; j13; j euro erspecti ncreasi mitrofanova tamonova, n ct s long been ion collapse ons have be ugh the pers ch was to de e to non-ma mplicated in r methods as e on is not a of russians a marriage is e started to c g. it demonst itution which s for children rds imonial unio analysis. assificati 18; c23. opean jou ives of f ng coha a, national r national res n characteriz e, the aver come comm pective of th efine the natu arital unions recent dialog event history complete a s from variou on the rise. consider non trates that co h is a trend n’s well-being ons; life co on urnal of g volu 47 family p abitatio research uni search unive zed by early rage age of mon. many s he second d ure of cohab in order to gues about fa y analysis an alternative to us social stra young, non n-marital unio ohabitation is d of great c g. ourse; family overnmen ume 5, nu policy in on iversity, highe ersity, higher y and univer f marriage scholars exp demographic bitations in r discuss how amily policy. nd sequence o marriage ata for whom -religious, ed on appropria s close to be oncern to p y policy; e nt and ec mber 1 (ju issn: 2 n russi er school of school of ec rsal marriage has been plain the cau c transition. ussia, revea w and why in order to a e analysis w in russia y m cohabitatio ducated peop ate for childb ecoming an in policymakers vent history conomics une 2016) 2254-7088 ia economics conomics e. after the rising, and uses of this the aim of l the factors non-marital achieve the were used. yet, but the on does not ple from big bearing and ndependent s due to its y analysis; russian federation russian federation mitrofanova and artamonova ● perspectives of family policy in russia 48 introduction modern social demographic policy of russia is focused on family, which is founded upon the unit of the married couple. here is a quote from the concept of demographic policy of russia to 2025: the solution of strengthening the family institution, the revival of spiritual and moral values of family relations includes: <...> propaganda on the values of families with more than one child, different forms of family placement of children without parental care – with the aim of creating a positive image of family in society with stable registered marriage of spouses – and having several children or adopting children without parental care. the quote shows that the modern family – which is not always based on stable marriage – is changing; and these transformations are ignored by lawmakers. moreover, the changes in matrimonial behaviour are not accepted as a normal stage of development of the society, but are ignored and even demonized. more often there are such alarmist statements declaring family and family values to be in crisis, that there is an evident call to revive traditional values (although the meaning of these words is rarely explained). instead of listening to the people and equalizing the rights and opportunities of registered and non-registered couples, lawmakers offer to increase the importance of marriage and neglect cohabitation. for example, in april 2015 in rossiskaya gazeta1, the proposal of one lawyer was published; in it, he claimed that, after two years from the beginning of cohabitation, such a living arrangement should be recognized as a marriage. the intended goal was the protection of financial rights of partners holding property jointly. nevertheless, the selected way to achieve the goal is rather questionable. to explain why, the experience of countries with totalitarian regimes (that proclaimed their high birth rate) shows that the birth rate is falling rapidly after the regime change. the experiences of anti-alcohol campaigns (including russian ones) show that, as soon as anti-alcohol campaigns end, the death rate from alcoholic poisoning increases dramatically and life expectancy falls. these, and many other examples, show that any interventions into the private lives of people by the government and the imposition of specific behaviour models may have negative consequences in the long term. the initiative of the transformation of cohabitations into marriage after two years constitutes not only the invasion of privacy of individuals, but also the attempt to make responsible decisions for people. if people decide not to marry, then such a decision constitutes a personal choice that should be considered not only at the level of society (which has already been happening for 20 years), but also at the government level. this article provides empirical evidence that russia tends to have more cohabitations, and that these appendages of marriage turn into full-fledged matrimonial couples, in which children are born and raised. which social and psychological factors stand behind the choice of this living arrangement is a question for future research. this work aims to draw attention to the fact that such choices are made: the amount of cohabitations, which will not result in a marriage, but which are stable and fit for childbearing, is increasing. we should stop ignoring this phenomenon (i.e. recognizing only marriages) and give it due consideration in legislation – not to transform cohabitations into marriages forcibly, but to equate their statuses at the legal level. it is necessary to expand the conceptualization of the family, without ignoring those people who have chosen alternative marriage forms. 1 rossiskaya gazeta. cohabitation is going to become equal with marriage (http://www.rg.ru/2015/04/29/brak.html). european journal of government and economics 5(1) 49 marriages and cohabitations in russia the study of the transformation of matrimonial behaviour in a population is an area of sociological and demographic research, the results of which help to explain changes in society and take part in improving family and demographic policy. this study is conducted to trace the dynamics of marriage and partnership formation by using the example of russian generations which were socialized before the collapse of the soviet union (levada, 2005) calls these generations soviet) and after it (post-soviet or modern generations). after the collapse of the ussr, all spheres of russian society have undergone changes (although russia had relatively low marital age and strong institute of marriage) that have led, since 1993, to the rise of average marital age; there was an increase in the number of unregistered couples, and the average age of people in these couples was decreased. such changes in the culture of marital and partnership behaviour of the youth were regarded by demographers as the evidence of a second demographic transition occurrence in our country, where unmarried couples became an alternative to marriage. what are these unions – a preparatory stage before marriage or an independent social institution? what segments of a population prefer to live in unregistered relationships? for how long are such relationships stable and strong? are they suitable for birth and parenting? in this article we present the most relevant and empirically supported answers to these questions. but first we will address terminology. among all the variety of terms for the unregistered relations between people, there are correct terms (e.g. cohabitation, partnership, consensual marriage, an unregistered marriage, the actual marriage) and incorrect terms (e.g. common-law marriage). all of these terms (except partnership) are synonyms and presuppose that individuals are living together, but cohabitation is the most concise, intuitive term and therefore is often used in demographic studies. this term has not caught on in society, because the word cohabitant has a negative connotation; the term is, however, accepted in science, so we will use it in our study. the main theoretical framework of this study is the conception of the life course. in sociology, the life course is studied as a development process, culturally constructed life stages and roles related to them (elder and glen, 1975; mayer, 2009; mills, 2000). the life course is a chain of phases of life which acquire different meanings in the process of socialization of the individual. in this sense, sociologists are studying periods of the human life from childhood to old ages, dividing them into different spheres: family, education and work. this study is focused on family at the beginning of marital or partnership trajectory. in recent decades, the life courses of russians (which was standardized before the 1960s) were supplemented by characteristics such as the fundamental openness and diversity of individual maneuvers. matrimonial behaviour ceased to be independent from sexual and reproductive incentives, as it was earlier under the traditional model of demographic behaviour (zakharov, 2008). marriage has been replaced by alternative forms of coresidence. the most common is cohabitation (avdeev and monnier, 2000). in many western european countries, cohabitation ceased to be a form of deviant behaviour a long time ago; thus, it has been studied there far longer than in russia. there are a few theories explaining the spread of cohabitations, but the main one is the theory of the second demographic transition, developed by r. lesthaeghe and d. van de kaa (1986). the researchers believe that it much better describes the changes in matrimonial behaviour than other theories. this point of view, strengthened after the publication of the study by gerber and berman (2010) which proved empirically that, rather mitrofanova and artamonova ● perspectives of family policy in russia 50 than the lack of good economic conditions, the focus on self-development and independence contribute to the spread of cohabitation. the second demographic transition is bounded by fundamental shifts in the life course of the individual. the freedom of choosing a marriage partner and a living arrangement are more expansive, and approaches to the consequences of sexual relations are more responsible than earlier. the emergence of contraception and more careful life planning enhances individuals’ ability to control their destinies. a fixed sequence of events in individual biography was replaced by a variety of life courses (lesthaeghe and van de kaa, 1986). thinking about these changes on the value-normative level, giddens (2003: 66) describes family and marriage as shell institutions, pointing out that the names remained the same, but the main inner content has changed. in the traditional family, which was an economic unit, the married couple was only a part of the structure, with communication with children and other relatives being more important. this structure was replaced by a couple, not necessarily married, but is based on romantic love and sexual attraction; such a unit became the main component of family life. as giddens writes, the most correct current definition of what is happening in private life is the formation of couples and their break ups, rather than marriage and family creation. mills and blossfeld (2013) have supplemented the theory of the second demographic transition with the theory of globalization. this complex approach shows that cohabitation is a ‘combination of emotional and physical intimate relations that mean living together and existing without legal or religious sanctioning’. cohabitation is becoming an independent institution (common among young people) which is more oriented on acquiring education and selfdevelopment, but not on family creation in conditions of increasing uncertainty caused by globalization and world economic problems. pluralism of forms of marital and reproductive behaviour, coupled with a rising general level of education of women and the advent of modern contraception, has led to the postponement of first births at a later age or not having children. as a result, young people prefer to enter into cohabitation as a union, not always involving the legitimation of relations in the future, but providing some of the benefits of marriage, including the ability to have a common budget. in many relationships, the flexibility of unregistered unions responds to the need of reducing the high level of economic and political uncertainty. according to the research of a group of theorists working under the guidance of kierman, who compared matrimonial behaviours in a number of countries2, cohabitation during its diffusion goes through four ‘evolutionary stages’: innovation, popularization, legitimation, and habitualization (kiernan, 2002; sobotka and toulemon, 2008; matyasiak, 2009). in the first stage, non-marital unions are few and are found only among people with low income. in the second stage the practice of cohabitation becomes more common, and it can be found in almost all social layers, yet these unions remain short (marriages are still longer than cohabitations), are considered as the first step to marriage and regarded as inappropriate for childbirth and parenting. in the third stage, unregistered unions start to be considered as an alternative to marriage, the duration of which is equal to marriage and is quite suitable for parenting. in the last stage, cohabitations merge with marriages and efficiently perform all its functions. the structuring of the stages is presented in table 1. 2 albania, austria, belgium, bulgaria, finland, france, germany, hungary, italy, poland, romania, russia, slovakia, spain, sweden, the czech republic, the netherlands, ukraine, the united states, etc. european journal of government and economics 5(1) 51 table 1. the stages of the spread of cohabitation features stages 1. innovation (cohabitation as marginal phenomenon) 2. popularization (cohabitation as a prelude to marriage) 3. legitimization (cohabitation as a substitute for marriage) 4. habitualization (cohabitation and marriage are indistinguishable) diffusion is practiced by marginal groups of society spreads among persons from the various social strata permanency short duration or a short pre-marital experience lasts longer and is less likely to be converted into marriage cohabitation as a family arrangement pregnancy is a reason for marriage parenthood becomes more and more common among cohabiting couples russia, as a country of the eastern bloc, demonstrates demographic outcomes of the second demographic transition (sdt) in an atypical fashion. in terms of destandardized family forms, russians began to perceive cohabitation as an alternative to marriage (mills et al., 2013). growing cohabitation rates alongside declining marital rates emerged in the soviet union in the middle of the 1980s, years before the fall of socialism (gerber and berman, 2010). zakharov (2008) showed that the birth cohorts from the 1970s already started to postpone entry unions and demonstrate all features of sdt. mills clarified that non-marital union in russia in the long run is a signal of low education and limited financial capital that is more common for the united states than for europe (mills et al., 2013). the purpose of this study is to identify factors of people’s entry to cohabitations and to ascertain whether this type of union is an independent institution, which functions in the same way as marriage. we suppose that, in russia, cohabitation is on the transition stage from popularization to legitimation; what this means is that it is not only exclusive to people with low income (as was shown in previous research (zakharov, 2008)) and shorter than marriage, but also still inappropriate for childbirth and parenting. we hypothesize that: 1. cohabitations are spread mainly among those of young generations who are non-religious, are living in big cities and have a secondary education; 2. conception causes marriage as the first union or the transition from cohabitation to marriage; 3. the most common statuses for the modern generations are: single at the age of 15, cohabiting without children by the middle of the observed age window, married with at least one child by the age of 35 (was conceived being married). to test these hypotheses, we decided to apply methods used in life course sociology: event history analysis and sequence analysis. the use of these methods is possible only with longitudinal data and retrospective surveys. database as the empirical base of the study, the panel part of the russian part of the generations and gender survey (ggs) (2004, 2007, 2011) and data from the survey person, family, society (pfs) survey (2013) were selected.3 both surveys 3 the russian part of ggs is ‘parents and children, men and women in family and society’. the three waves of the survey were conducted by the independent institute for social policy (http://www.socpol.ru/eng/research_projects/proj12.shtml). mitrofanova and artamonova ● perspectives of family policy in russia 52 are retrospective and representative for russia. it should be noted that the difference in years of the surveys is not very big – we would otherwise be able to compare the results. at the same time, coverage generations in surveys still differ: this provides an opportunity to supplement the data of the ggs survey (mainly reflecting the behaviour of the soviet generation of russians) using the data from the pfs survey (the sample of this research consists predominantly of the younger generation representatives). general information about the databases can be found in table 1. table 1. the ggs and pfs databases features ggs pfs years of the surveys 3 waves: 2004, 2007, 2011 2013 number of respondents 4948 cases 3572 cases generations 1935-1984 years of birth 1975-1994 гг. years of birth age of respondents 24-85 18-44 in order to complete the tasks of this study, we have to impose some restrictions on the data. one of the restrictions relates to the sequence of matrimonial events. it is necessary to consider only the first unions for understanding the nature of cohabitation in russia. this localization allows for a better understanding of generational differences in the beginning of matrimonial and reproductive trajectories of respondents. young generations choose new models of behaviour associated with marital or partnership spheres: they are focused on the emotional sphere of relations, hedonistic context and, accordingly, are not in a hurry to start a family and have children. in addition, the majority of russians in their life are limited only by the first unions. another restriction concerns the lower and upper age limits of respondents, who fall into the sub-sample to build some models. 15 years – the lower age limit – was taken as the age of beginning of reproductive behaviour. in the ggs sample there are representatives of five generations, the youngest respondents at the time of the third survey wave were 25 years old. there is an even younger generation in the pfs. to understand the changes, which happened to the matrimonial behaviour of russians, one should equalize the chances of the representatives of all generations to entry to the first union. after considering several options for establishing the upper age limit, we decided on the age of 35 years, as this will help to avoid the influence of atypical ages of entry to first unions of older generation representatives, but at the same time maintain the age structure of the sample. analysis of the occurrence of matrimonial events the first group of models focuses on the identification and comparison of factors of entry to cohabitation and marriage as first unions (models 1 and 2). the second group of models describes how the conception is associated with the following events: entry into cohabitation and marriage as first unions (models 3 and 4) and entry into marriage after the experience of living together (model 5). the first group of models the beginning of matrimonial trajectories belonging to the first group of models, presented in figure 1. the pfs survey was conducted by the russian presidential academy of national economy and public administration (http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2559218). figure 1. m m cohabitat m marriag for the mo two options the first co marriage a as first unio specificatio table 2. ba element ev su dependent v pred stratifyin table 3 pre in the anal age structu 1945 to 19 young rus and wome obtained b women is 5 serious dis waves was result of r recalculate considered marital-partne model №1: tion as first unio odel №2: ge as first union odel which id s are suitable habitation an after the coha ons, it is nec on of models asic characte ts of model vent bject variable (time) dictors ng variable esents the re lysis. the ge ure of the po 975. the pf ssians. to a n, the variab by the russ 54 per cent splacement t s not made b respondent ed because d negligible. europ ership traject on dentifies fact e: (1) when nd not marrie abitation. fo cessary that s 1 and 2 are eristics of co 15 years months fro into elative numb enerational s pulation of r fs survey is ccount for t ble sex was sian nationa of the popu towards wom because the e attrition is their diver pean journal 53 tories for mod tors of entry the respond ed, and (2) w or the model a marriage w e presented i ox regression cox r № 1 s – 1st cohabitat e russian om 15 years to 1st cohabitation ber of observ structure of t russia, whic s aimed to id he differenc s included to al population ulation. the d men. recou estimation o not possibl rgence from of governm dels 1 and 2 into cohabit ent was thro when the ind , identifying was not prec n table 2. ns for models regressions for tion 15 entry into first un ns at the age of entry mont sex place of residen level of educat religiosity generations vations for a the ggs sur h is dominat dentify featu e in matrimo o the model n census 20 data array o unting of the f the displac e. the arra m the gener ent and econ tation as the oughout his o dividual had a the factors ceded by a c s 1 and 2 r models № 2 5 year – 1st direc nion 15+ years ths from 15 yea 1st marria nce ion ll the variabl rvey correspo ted by people res of the b onial behavi . according 010, the pr of the ggs p e array of al ement of the ay pfs dat ral populatio nomics 5(1) e first union, or her life in at least one of marriage cohabitation. ct marriage ars to entry into age les included onds to the e born from behaviour of iour of men to the data roportion of panel has a ll the panel e panel as a ta was not on can be mitrofanova and artamonova ● perspectives of family policy in russia 54 table 3. the relative number of observations for the predictors of the first group of models, % predictors of models categories of variables ggs pfs generations 1935-1944 16.8 1945-1954 21.4 1955-1964 26.8 1965-1974 21.4 1975-1984 13.5 48.1 1985-1994 51.9 sex male 32 49 female 68 51 place of residence urban area 56.8 40.1 rural area 43.2 59.9 level of education higher 22.3 33.8 secondary general 54.6 37.6 secondary vocational 23.1 28.6 religiosity religious 82.3 65.4 to control for the impact of the environment on respondents’ behaviour, we included the variable type of the settlement. in the case of the ggs panel, this figure was taken at the time of the third wave (2011). migration mobility of population in russia is low, so the use of the current settlement type can only slightly shift the overall estimations. ggs largely covers the urban population, while in the array of the pfs, the rural population accounts for more than half of the respondents. regarding the level of education, in the array of pfs surveys – which represents the young population of russia – in comparison with ggs there is a higher proportion of people with higher education and a lower share of people with secondary education. the binary variable religiosity is created in order to manage the religiousness of the respondents with the categories “confess one religion” (religious) and “do not profess any religion” (non-religious). in the pfs there is a much smaller percentage of people belonging to the believers than in ggs. all variables were subjected to a correlation analysis, which showed that the variables are independent and can be included in the model. results of the analysis of the first group of models basic characteristics of the models indicate that at least one of the variables explains the change of the dependent variable, which means that it is associated with the risk of an onset of cohabitation or marriage. the significance of almost all coefficients does not exceed 5 per cent, that is, they are statistically significant. the coefficients of model 1 are presented in table 4. c l ti h d f ti table 4. co after the ag predictors o gend level of ed place of res religio in accordan union is ab fact that w sexes to jo the respon the case o respondent group, that in russia, a rural areas cities (ggs cohabitatio those who 1.4 times ( religious pe figure 2. h c um ul at iv e h az ar d fu nc tio n months from the stratific represents representa pfs. the y ox regression ge of 15 f models c er ducation s sidence sity nce with mod bout 1.2 time women enter in a union of ndents with of the ggs ts) more like t is, those w as revealed are 0.7 time s). in the c on for the res o do not refe (ggs) and eople. hazard curve ggs m 15 years to st cohabitation cation for all the proba tives of each younger the europ ns for model ategories of var male female higher secondary gen secondary voca urban area rural area non-religiou religious del 1, the pro es higher tha unions at a f any type are professional s responden ely to enter ith higher ed by mills et a es less likely case of youn idents of rura r to themselv 1.1 times (p s for cohabit g tart the first l models of t ability functio h of the six g generation th pean journal 55 №1: the ris riables neral ational a us obability of w an of men of an earlier ag e aligned at and genera nts) and 1.1 into cohab ducation that al. (2013). if w y to enter into nger genera al and urban ves as the fo pfs) more li tation as first generations c um ul at iv e ha za rd fu nc tio n the study wa on of an generations f hat an individ of governm sk to enter in ggs: 1444 events models c r 1,18*** r 1,65*** 1,44*** r 0,73*** r 1,41*** * p = women to en f the same a ge than men the age of 40 al education 5 times (in itation than t display a ‘p we talk abou o cohabitatio tions (pfs), n areas are n ollowers of th kely to ente t union (strat months fro as the variab entry into for ggs and dual represe ent and econ to a first coh 192 coefficients exp( ref. group 1 ref. group 1 1 ref. group ref. group 1 0.10, ** p = 0.05 gage in an u ages. this is n. the chanc 0. are about 1 the case o those in the pattern of dis t the type of n than those , the likeliho ot significant he existing r r into cohab ification by g pfs om 15 years to cohabitation ble generatio marital unio the two gen ents, the mor nomics 5(1) habitation pfs: 25 events (b) 1,26*** 1,15*** 1,15*** 0,94 1,13*** 5, *** p = 0.01. unregistered s due to the ces of both .5 times (in of the pfs e reference sadvantage’ f settlement, e who live in ood to start tly different. religions are bitation than generation) start the first on. figure 2 on for the nerations for re he or she c l ti h d f ti is prone to of 35 years age; howev a first coha table 5. co marriage a predictors o gend level of e place of re religio model 2 wa experience into marria of the sam of the gg cohabitatio the situatio marry than less likely the probab indicator, respondent cohabitatio figure 3. p c um ul at iv e ha za rd fu nc tio n months fro in figure 3 higher cha representa mitrofanov enter into co s. the repres ver, we can abitation will i ox regression fter the age o of models der ducation esidence osity as constructe e of prelimina ge as the fir e age, 1.2 tim gs responde on is loosely on is differen n their peers to register a bility of marri which is ca ts are abou on than religio probability fun ggs om 15 years to star marriage 3 we see th ances to e tives of post va and artam ohabitation a sentatives of already assu increase for n for model № of 15 categories variables male female higher secondary ge secondary voca urban are rural area non-religio religious * p = 0.10 ed to identify ary coresiden st union is m mes more fo ents, the pr associated w nt: young ru with genera a relationship iage of rural alculated fo t 0.7 times ous people. nction of entr g rt the first hat the respo enter into m -soviet gene monova ● pe 56 as a first unio the younges ume that the them. №2: the pro s of s eneral ational ea a ous s 0, ** p = 0.05, ** y factors of e nce (table 5 more than the or ggs and robability to with the leve ussians with al education; p than their residents as or denizens less likely t ry into first m generations c f ondents from marriage wit erations. erspectives o on. the dotte st generation e cumulative obability of en ggs: 3135 events models c r 1,18*** r 0,96 1,10** r 1,22*** r 0,79*** ** p = 0.01. entry into a fi 5). the proba e same indic 1.5 times mo enter into el of educatio higher educ however, th peers with s the first uni of all gen o enter into marriage as fi c um ul at iv e ha za rd fu nc tio n months m soviet ge thout prior of family polic ed line show ns have not r probability t ntering into a 56 coefficients exp( ref. group 1 ref. group ref. group ref. group 0 rst marriage ability of wom cator calcula ore for pfs. marriage w on. in the ca cation are mo hey are abou professional on 1.2-fold e erations. n marriage w irst union pfs from 15 years to st marriage nerations ha cohabitation cy in russia ws the onset reached this to enter into a first pfs: 68 events (b) 1,55*** 0,78** 1,28** 1,19** 0,69*** e without the men to enter ated for men in the case without prior ase of pfs, ore likely to ut 1.3 times l education. exceeds this on-religious without prior start the first ave 3 times n than the the seco the purpos is suitable after they for this pu and marria cohabiting, analyzed u figure 4. m for models the indepe the fact of unions. for marked by in the case belonged t respondent figure), it marriage o the first con the child w thus conce because th childbirth, b the concep considered table 6. in elements of event subjec dependent varia predicto in the situ conception ond group se of the sec for parentin have found rpose, it was age as first where con nion (figure marital-partne 3, 4 and 5 endent varia f conception r models 3 a purple and g e of model to one or a t entered int is difficult to r cohabitatio nception and was at least t eption had s he marriage but to legitim ption, which o d as the stimu itial characte model 1 1s ct able (time) mo ors uation, when (marked europ p of model cond group o g or the res out about co s necessary t unions, a nception was 4 and table ership traject able was con occurrence and 4 the con green) did no 5, because another unio o cohabitatio o understand on. it was fou d marriage w three years o stimulated en was not has mize the right occurs more ulator of coha eristics of the №3 15 years – 1st cohabitation st cohabitation russians in the a onths from 15 years n the marria in yellow) pean journal 57 ls of models is spondents pr onception (t to compare as well as s an explana e 6). tories and oc nstructed se strictly befo nstruction of ot cause any of the lack on, in a situ on and then d whether co und that more were not less old at the tim ntry into coh sty, it is like ts of the child e than 45 mo abitation (mo e cox regress cox re №4 15 year – 1st marriage 1st direct marria age of 15+ years s to entry to 1st unio conceiving be age was aft happened of governm to understan refer to form he so-called the model o marriage af atory variab ccurrence of parately for re entry into f this variable y questions. of informatio ation where into marriag onception inc e than 80 pe s than 45 mo me of marria habitation in ly that it wa d and family onths before odel 3). sions for mo egressions for mode t 1s age 1st m russia on months f fore entry to analyz er the expe strictly be ent and econ nd whether c malize their re “marriage a of entry into c fter the exp le about ent conception each model o each of th e (the fact of on about wh e, after conc ge (marked in creases the r cent of cas onths, which age after livin stead of into s intended n at the legal the marriage dels 3, 4 and els №5 t cohabitation – 1st marriage after 1st co ans after entry to 1s rom 1st cohabitation zed union erience of c etween ma nomics 5(1) cohabitation elationships after baby”). cohabitation perience of try into the l to capture he analyzed f conception hich partner ception, the n red in the chances of ses between means that ng together, o marriage. not to cover level. thus, e, was then d 5 marriage ohabitation st cohabitation n to 1st marriage cohabitation, rriage and cohabitatio models 3, 4 the resu in order to unions, thr small num built for mo not given. in the form according t stimulated marriages according happened higher prob conception the signific conception figure 5. t types (ggs according 6.9 per cen and 6 per c in the surv of such coh experience the ggs sa proved itse the pfs sa who, in ca sample. th their partne mitrofanov on (even if i 4 and 5 are p ults of the o evaluate th ree cox reg ber of event odels 3, 4 an however, si of schemes to all three m the beginnin and 16 per to the reg have 4 times bability to en before entr cance of the , but other fa the role of co s) to the pfs s nt of cohabit cent of marr ey of stimula habitations in e of living ample, wher elf to be suita ample, that se of birth o his step could ers. va and artam t was the c presented in analysis o he role of co ressions we ts that matc d 5 turned ou gnificant res (figures 5 a models built f ng of 20 per cent of ma ression coe s higher likel nter into a m ry into any u e coefficient actors stimula onception in t survey (figu ations as firs riages after t ated marriag n pfs is alm together e more matu able for raisi represents y of children, h d act as a gu monova ● pe 58 cohabitation table 6. of the sec onception in ere construct ch the specif ut to be low sults were ag and 6). for ggs (fig r cent of firs arriages after efficients, th lihood to ent marriage as f union. in mo is much gr ate the entry the process ure 6), a con st union, 21 the experien ges is only 2 most 3 times is 2.6 tim ure individua ing children young gener had more of uarantee of t erspectives o of second cond group the process ted for each fied trajector in quality, so ggregated, a gure 5), it wa st cohabitatio r the experie ose couples ter into a coh first union th odel 3 (marri eater than 5 y into a such of entering in nception stim per cent of ce of living t per cent fro less, and tha mes less. t als were pres – almost on ration, there fficial relation the seriousn of family polic order). spec p of mode of forming h database. ry, the cox o the detailed analyzed and as found that ons, 23 per ence of livin s in which habitation an han the coup age after co 5 per cent, marriage. nto unions of mulated the b marriages as together. the om ggs. the at of marriag this means sented, coha a par with m are more r nships than i ess of young cy in russia cification of els matrimonial due to the regressions d results are d presented t conception cent of first ng together. conception nd a 3 times ples without ohabitation), so it is not f various beginning of s first union e difference e difference ges after the s that, in abitation has marriage. in respondents in the ggs g people for figure 6. t unions of v the aggr in the seq reproductiv different ev the typical the behavio in figure respondent ages of th have not generations axes the p shown. these chro solitude or marriage a end of the decreases place to th traditional f confirms th 4 we used th the role of co various types regated fa quence ana ve events of vents, we ad statuses of ours of repre 7, we see ts over gene e responden yet reached s we had to proportions o onograms sh r living apa nd marriage age window during the hose who ra form of matr he assumptio he r package t europ onception in t s (pfs) amily form lyses,4 we f both surve ded the sec life of the re esentatives o the distribu erations. on nts between d the upper work with ce of individuals how that sin art together with concep . the freque transition fro aise children rimonial beha on of de-instit raminer (gaba pean journal 59 the process mation traje constructed eys. in orde cond unions. espondents, of different ge ution of par the horizont n 15 and 35 age limit, ensored data s belonging glehood, to is the mos ptions are do ncy of these om soviet to n while singl aviour gives tutionalized l adinho et al. 201 of governm of entering in ectories chronogram er to trace t this allowed but also to enerations. rtnerships a tal axes of th years. the therefore, in a (indicated to each sta which can b st frequent s ominant statu statuses by o post-sovie le or in the way to the o life courses. 11). ent and econ nto ms for matri he transition d us not only notice the d nd fertility s he chronogra youngest r n the case in grey). on ate at a give be referred t state at the uses for rus the age of 3 et generation second ma other sequen nomics 5(1) monial and ns between y to capture difference in statuses of ams are the respondents of modern the vertical en age are to as actual age of 15. ssians at the 35 gradually ns, giving a arriage. the nces, which figure 7. d 1935-44 ggs: 833 interviews 1945-54 ggs: 1,058 interviews 1955-64 ggs: 1,328 interviews 1965-74 ggs: 1,058 interviews c s co single single, statuses s (cohabitatio mitrofanov distribution of “soviet union” ge censored single, no onceptions e, 1 conception , 2 conceptions stand out w on as a trial 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 17 18 19 the propo of cohabit people, especially at least 1 child, is ve small 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 17 18 19 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 17 18 19 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 17 18 19 2 va and artam f partnership enerations first coh first co first coh sec second c second c when marriag marriage), a 20 21 22 23 24 25 26 ortion ting with ery 20 21 22 23 24 25 26 20 21 22 23 24 25 26 20 21 22 23 24 25 26 monova ● pe 60 ps and fertility 1975ggs 670 intervi 1975pfs 1,71 intervi 1985pfs 1,85 intervi legend habitation, no conce ohabitation, 1 conce habitation, 2 conce cond cohabitation, n conceptions cohabitation, 1 con cohabitation, 2 conc ge is prece and also an 6 27 28 29 3 6 27 28 29 3 6 27 28 29 3 6 27 28 29 3 erspectives o y statuses by “mode 84 s: 0 ews -84 s: 17 ews -94 s: 55 ews eptions eption eptions no ception ceptions eded by one interlayer of 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 1 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 1 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0,9 1 15 16 1 the propo cohab peop child has b incre of family polic y generation ern” generations first marriage, n first marriage, first marriage, second ma concep second marriage second marriage e or two co f those who 17 18 19 20 21 22 23 17 18 19 20 21 22 23 17 18 19 20 21 22 23 ortion of biting le with ren been asing cy in russia no conceptions 1 conception 2 conceptions arriage, no ptions e, 1 conception e, 2 conceptions ohabitations have never 3 24 25 26 27 28 29 3 3 24 25 26 27 28 29 3 3 24 25 26 27 28 29 3 3 3 3 european journal of government and economics 5(1) 61 been married and had one or two conceptions in the first or second cohabitation (cohabitation as an independent union). such interlayers are becoming increasingly visible for each subsequent generation, and thus, the spread of cohabitation is gradually moving from the stage of promotion to the stage of legitimation. conclusions the aim of this study was to identify factors of entry into the first cohabitation, compare them with the factors of entry into the first marriage, determine the stage of the spread of cohabitation in russia and, finally, demonstrate that policymakers should stop ignoring this phenomenon. not all of our hypotheses were confirmed. 1. for the soviet generations (ggs data), women from 1965-74, those who received secondary or vocational education, urban residents and non-religious people are more likely to enter into cohabitation as a first union. the probability of entry into marriage as a first union is loosely associated with the level of education. the chance is higher for women, religious people and those who live in rural areas. for modern generations (pfs data), the probability of entering into unions of various types are associated only with gender (females are more likely to enter into any matrimonial unions), level of education (people with vocational education are more likely to engage in cohabitation than in marriage) and religiosity in a situation of marriage (the risk of marriage is higher for the religious than for the nonreligious). 2. conception often precedes cohabitation and marriage as first unions, but does not act as a strong incentive for transition from cohabitation to marriage. 3. there are respondents with children (had at least one conception) in cohabitation in the samples of the surveys. sequence analysis has revealed that the most frequent status for all generations by the age of 35 is “married, was conceived”. besides such states as “raising one or two children”, “alone” or “in cohabitation” are becoming more common for the representatives of the modern generations by the end of the observed age. in these cases cohabitation is an alternative to marriage. the modern generations begin their matrimonial biographies with cohabitation as a union, involving a lower degree of responsibility than marriage, but at the same time having the benefits of marriage (e.g. general budget, housing), which is especially important in a period of uncertainty. the changing of norms and values in the modern world leads to the fact that people consider religious, social and family traditions to be less important than before, starting to shape and fill their life courses based on their own priorities. therefore, young, urban, non-religious, lesseducated people are considered to be the engine of change in the matrimonial field in russia, confirming once more that cohabitations in russia are more associated with the less-educated as opposed to western european countries where the highly-educated are significantly more likely to follow long term cohabitation trajectories (mills et al. 2013). in modern russian society, forms of marital partnership and reproductive behaviour are becoming more diverse and unpredictable. a standardized sequence of events in individual biographies is replaced by a variety of life courses, in which cohabitation becomes one of the options: cohabitations in russia gradually have come to the stage of legitimation. policy-makers should treat the new institution as a decision of the people. an important question arises: why do people choose cohabitations instead of marriages? this question is for further studies, but what we know now for sure is mitrofanova and artamonova ● perspectives of family policy in russia 62 that cohabitations are gaining power and popularity, and we cannot ignore them anymore. acknowledgements the article was prepared within the framework of the academic fund program at the national research university higher school of economics (hse) in 2016 (grant № 16-05-0011) and supported within the framework of a subsidy granted to the hse by the government of the russian federation for the implementation of the global competitiveness program. the authors also want to thank some colleagues who contributed in this paper. heartfelt gratitude to sergey v. zakharov for a permanent scientific advising, insightful recommendations and openness to new ideas. sincere thanks to francesco c. billari and melinda mills for the revision of our work in life course methods. a lot of thanks to philippe blanchard, roxana leu, alla o. tyndik, and svetlana s. biryukova for their consultations on sequence analysis. finally, thanks to vladimir a. kozlov and thomas h. espy for their consulting on politics and inestimable help in the preparation of this paper. references avdeev, alexandre, monnier, alain (2000) ‘marriage in russia: a complex phenomenon poorly understood’, population: an english selection 12: 7–49. billari francesco c. (2001) ‘sequence analysis in demographic research. special issue on longitudinal methodology’, canadian studies in population 28(2): 439458. gabadinho, alexis, gilbert ritschard, nicholas s. müller and mathia studer (2011) ‘analyzing and visualizing state sequences in r with traminer’, journal of statistical software 40(4): 1-37. gerber, theodore p. and danielle berman (2010) ‘entry to marriage and cohabitation in russia, 1985–2000: trends, correlates, and implications for the second demographic transition: la mise en couple en russie, 1985–2000: tendances, facteurs associés et implications par rapport à la seconde transition démographique’, european journal of population-revue europeenne de demographie 26(1): 3–31. giddens anthony (2003) runaway world: how globalization is reshaping our lives, london: profile books ltd. kaa, dirk j. van de (1987) ‘europe’s second demographic transition’, population bulletin 42(1): 3–57. kaa, dirk j. van de and ron lesthaeghe (1986) ‘two demographic transitions?’, population: growth and decline 9–24. lesthaeghe, ron (1995) the second demographic transition in western countries: an interpretation, in: gender and family change in industrialized countries, oxford: clarendon press. lesthaeghe ron and karel neels (2002) ‘from the first to the second demographic transition: an interpretation of the spatial continuity of demographic innovation in france, belgium and switzerland european’, journal of population 18(4): 325-360. levada, yu. (2005) ‘generations of хх century: opportunities of studies’ in yu levada , т. shanin (eds) fathers and children: analysis of contemporary russian generations. moscow: novoe literaturnoe obozrenie, pp. 39-60. european journal of government and economics 5(1) 63 matysiak anna (2009) ‘is poland really ‘immune’ to spread of cohabitation?’, demographic research 21(8): 215-234. mayer, karl u. (2004) ‘whose lives? how history, societies, and institutions define and shape life courses’, research in human development 1(3): 161–187. mills, melinda (2004) ‘stability and change: the structuration of partnership histories in canada, the netherlands, and the russian federation’, european journal of population revue europeenne de demographie 20(2): 141–175. mills, melinda and blossfeld, hans p. (2013) ‘the second demographic transition meets globalization: a comprehensive theory to understand changes in family formation in an era of rising uncertainty’, life course research and social policies 1(1): 9-33. mills, melinda, laurent lesnard and gina potarca (2013) ‘family formation trajectories in romania, the russian federation and france: towards the second demographic transition?’, european journal of population 29(1): 69-101. mitrofanova, ekaterina s. (2013) ‘demographic behavior of russians: family and fertility patterns across generations’, macrotheme review 2(3): 71–80. philipov, dimiter and aiva jasilioniene (2008) ‘union formation and fertility in bulgaria and russia: a life table description of recent trends’, demographic research 19: 2057–2114. sobotka, tomas and laurent toulemon (2008) ‘overview chapter 4: changing family and partnership behaviour: common trends and persistent diversity across europe’, demographic research 19(6): 85-138. vikat, andrés, zsolt spéder, gijs beets, francesco c. billari, christoph bühler, aline désesquelles, tineke fokkema, jan m. hoem, alphonse macdonald and gerda neyer (2008) ‘generations and gender survey (ggs): towards a better understanding of relationships and processes in the life course’, demographic research 17: 389–440. vishnevsky, anatoly g. (2009) ‘a new stage of russian demographic development’, in vishnevsky, a. g. and s.n. bobilev (eds.) russia facing demographic challenges. national human development report russian federation, moscow, pp. 18–25. zakharov, sergei (2008) ‘russian federation: from the first to second demographic transition’, demographic research 19, 907–972. transparency and local government corruption: what does lack of transparency hide? vol.7 • no.2 2018 issn: 2254-7088 european journal of government and economics 7(2), december 2018. european journal of government and economics issn: 2254-7088 number 7, issue 2, december 2018 doi: https://doi.org/10.17979/ejge.2017.7.2 transparency and local government corruption: what does lack of transparency hide? 106-122 juan luis jiménez, daniel albalate 106-122 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real. 123-137 carlos fernández_herraiz, antonio javier prado domínguez, carlos pateiro-rodriguez, jesus m. garcia-iglesias path – dependence and european fisheries management 138-153 federico martín palmero, fernando gonzález laxe why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem 154-170 filipa figueira deciding on financial renegotiation in public-private partnership projects 171-199 wiston risso https://doi.org/10.17979/ejge.2017.7.2 european journal of government and economics 7(2), december 2018, 106-122. european journal of government and economics issn: 2254-7088 transparency and local government corruption: what does lack of transparency hide? juan luís jiméneza, *, daniel albalateb a universidad de las palmas de gran canaria, spain b universitat de barcelona, spain * corresponding author at: juanluis.jimenez@ulpgc.es article history. received 13 december 2017; first revision required 7 may 2018; accepted 10 july 2018. abstract. this paper empirically investigates the causal relationship between local government transparency and political corruption in a sample of spain's 110 largest municipalities. after implementing a two-stage probit estimation procedure, our evidence indicates that transparency is inversely related to the likelihood of local political corruption. this result supports the hypothesis that a lack of transparency conceals corrupt activities and that an unwillingness to provide information is a good proxy of the likelihood of corruption. keywords. political corruption, transparency, local government. doi. https://doi.org/10.17979/ejge.2018.7.2.4509 1. motivation corruption is the focus of a growing body of literature in many fields, concerned with understanding its determinants and consequences. corruption has obvious negative impacts on the political system and on citizen confidence in a country’s institutions. empirical evidence has shown the negative relationship between institutional trust and corruption (mishler and rose, 2001; anderson and tverdova, 2003) thus, corruption weakens the credibility of institutions through the erosion of confidence (bowler and karp, 2004). beyond the political sphere, the literature has extensively shown that corruption also becomes a major burden for any economy, given that it distorts the allocation of resources and damages the functioning of institutions (knack and keefer, 1995; williamson, 2000; acemoglu et al., 2002; kaufman and kraay, 2002). corruption has been shown to affect the development, prosperity and equity of an economy (see bardhan, 1997, aidt, 2009 for reviews). transparency is one of the main cornerstones of corruption. it has been widely considered in the literature, but rarely tested as a determinant of corruption (lindstedt and naurin 2010). for this reason, the aim of this study is to evaluate the impact of transparency on the probability of corrupt behavior at the local level. in the first instance we seek to determine the extent to which government openness can prevent corruption. our results indicate a negative relationship mailto:juanluis.jimenez@ulpgc.es https://doi.org/10.17979/ejge.2018.7.2.4509 j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 107 between the transparency index and the likelihood of corruption in the largest municipal governments of spain. as such, the paper provides evidence of the potential role of transparency in designing policies aimed at fighting local political corruption. the rest of the paper is organized as follows. after this introduction, next section reviews the empirical literature on this topic. then we describe the empirical strategy by presenting our main hypothesis, the data used and the empirical method implemented. we report our main results and interpret them. finally, the paper finishes with some concluding remarks and policy recommendations that can be derived from our empirical analysis. 2. literature review the pioneering work of mauro (1995) associated corruption with lower levels of investment and, hence, of economic growth. this negative impact was subsequently confirmed by mo (2001), ali and isse (2003), meon and sekkat (2005) and aidt (2009) among others. similarly, gupta et al. (2002) reported a positive relationship between corruption and economic inequality and poverty. institutional quality and better governance, are two of the main drivers of economic growth (olson et al. 2000; meon and weill 2008), and have become central to the study of corruption. in fact, meon and sekkat (2005), aidt (2009), blackburn and forges-puccio (2009) and swaleheen (2011), all situate institutions in the central area for economic growth. the study of corruption determinants has flourished over the last years. major determinants appear to be the low wages of public servants (fisman and gatti (2002b), the larger size of government expenditure (goel and nelson, 1998, 2010; del monte and papagni, 2007), low educated and low income constituencies (glaeser and sak, 2006), high income inequality (you and khagram, 2005), and more centralized political systems (fisman and gatti, 2002a). at micro level, mocan (2008) identifies the individual features that induce being offered briberies. in all, corruption seems to respond to standard economic incentive theory, although officials find alternative strategies to pursue rents and, by doing so, attenuate the effects of anti-corruption policies (olken and pande, 2012). there is evidence that audits may limit corruption, suggesting that traditional top-down monitoring can play an important role in reducing corruption, even in a highly corrupt environment (olken, 2007). together with audits, electoral and judicial accountability increase the perceived future probability of being exposed to punishment, leading to a political context less favorable to corruption (ferraz and finan, 2008). furthermore, the fight against corrupt activities have also moved in favor of higher levels of transparency and accountability understanding that the “increased flow of timely and reliable economic, social and political information, which is accessible to all relevant stakeholders”, as it is described in kaufmann and kraay (2002), is good prescription against corruption. in addition, transparency is an essential component of the political system of deliberative democracies (elster, 1998; grigorescu, 2003; bellver and kaufman, 2005; rosendorff and doces, 2006; naurin, 2007). j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 108 as such, transparency is directly connected to the process of accountability (meijer 2003) and provides better incentives that favor the efficient allocation of resources. as a result, transparency is expected to promote economic growth by minimizing the moral hazard and principal-agent problems, improving the allocation of resources (holmström 1979) and limiting private-rent seeking (baumol 1990; murphy et al., 1993; calderón and chong, 2006). in fact, information is a public good that improves the functioning of markets and, as a result, it is desirable (stigler 1961; stiglitz and weiss 1981; stiglitz 2000). among other impacts, transparency has been shown to improve public finance, by lowering the accumulation of debt and deficits (alt and lassen, 2006a), to lower financing costs for governments (glennerster and shin, 2008), to improve market valuation from rating institutions and to improve fiscal discipline of governments (hameed, 2005), and finally, to limit opportunistic behavior of governments in their fiscal policy according to the electoral cycle (alt and lassen, 2006b). on the other side, the main determinants of transparency are found to be the economic development (bellver and kaufmann, 2005; ibp, 2013), social and historical and cultural heritage (la porta et al. 1999), administrative heritage (wehner and de renzio, 2013), political competition and better fiscal heritage in terms of low debt stocks (alt et al 2006). although transparency has been widely considered in the literature (blumkin and gradstein 2002; eingen, 2002; peisakhin and pinto 2010), it has been rarely tested as a determinant of corruption (lindstedt and naurin 2010). the available but scarce literature offers evidence on the positive relationship between fiscal transparency and control of corruption (hameed, 2005; bellver and kaufmann, 2005; kolstad and wiig, 2009; whener and de renzio, 2012). indeed, many studies have been limited to a consideration of access to information laws (costa, 2012), or to information technologies, including digital access, internet connection or press coverage, but not to public disclosure (see brunetti and weder, 2003; gentzkow et al. 2006 or di rienzo et al. 2007). among others, one mechanism through which transparency mitigates corruption has to do with the principal-agent theory. voters (the principal) have only limited information on the government performance (the agent). they are only able to check indirectly the behavior of politicians by looking at some imperfect signals. politicians may pursue private interests only if the asymmetry of information between them and the principal is high enough. discretionary policies are only possible with large information asymmetries. however, this mechanism and causality relationships are far from being obvious. indeed, there may be underlying factors that explain both the extent of transparency and the level of corruption. selection bias and endogeneity concerns (governments that are more honest may tend to be more transparent) may hamper the identification of true causality links between both concepts. to illustrate this concern, the literature discusses and describes a number of factors that lead to corruption environments that can also separately explain transparency levels. see for instance the works by andvig and moene, 1990 and ali and isse (2003), del monte and papagni (2007), among others. according to kolstad and wiig (2009), the lack of transparency increases the bias and j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 109 variance of the signal used by the voters to check the performance of governments. it is in fact, a rational behavior to increase information rents in a way they may perpetuate the appropriation of large public rents (stiglitz, 2002). transparency acts as a barrier to this discretionary room of governments, by lowering the asymmetry of information between the principal and the agent. this may prevent rent-seeking strategies and improve the ability of voters to check closely the behavior and performance of elected governments. as a result, transparency minimizes the moral hazard problem by improving public control, which should be correlated with lower corruption. the likelihood of committing a crime depends on the risk of being caught, the possible gain (booty) and the probable punishment (becker, 1968). transparency works on the risk of getting caught, given that in improves public control and diminishes the asymmetry of information. however, recent literature has also shown that transparency alone cannot be the only solution to corruption (kolstad and wiig, 2009; lindstedt and naurin, 2010). the literature on local corruption and transparency is mainly recent in publication and makes few connections between the two issues. in the case of corruption, most studies have been concerned with estimating its subsequent impact at the ballot box (ferraz and finan 2008 and 2011 for brazil). the findings are that the impact is limited unless media coverage is broad (fernández-vázquez and rivero 2010; costas-pérez et al. 2012) and that the impact varies depending on whether the party revealed as being corrupt is on the left or right of the political ideological spectrum (jiménez and garcía, 2018). moreover, local corruption is contagious (gonzález et al, 2017). as for transparency, research has sought to identify the institutional, social and economic determinants of local government transparency (styles and tennyson, 2007; guillamón et al. 2011; esteller and polo-otero, 2012; albalate 2013). political competition, the size of population, the economic specialization of the city, and fiscal imbalances and debt accumulation are some of the factors indicated as determinants of local government transparency. however, the relationship between transparency and corruption for local governments has been surprisingly neglected. 3. empirical strategy 3.1 hypothesis the objective of our analysis is to evaluate the relationship between transparency and corruption at the local government level, drawing on data from spain. transparency is expected to limit corruption and, therefore, we expect local governments providing better information, or satisfying high transparency standards, to be less prone to corruption: first, because a greater availability of information increases the probability of being caught in private rent-seeking activities; and, second, because local governments that are more willing to make information j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 110 available are assumed to be more honest than municipalities that prefer to conceal information. if our expectations are met, then transparent local governments should suffer fewer cases of corruption. the main hypothesis being tested in this paper, therefore, can be stated as follows: h1. local government transparency is inversely related to corruption cases (ceteris paribus). thus, we evaluate the extent to which levels of transparency are informative, or it hinders the search for private rents, of the likelihood of corruption at the local level. if our results confirm the hypothesis, then the prescription of greater transparency would be justified. 3.2 data on corruption and transparency first, we constructed a new database by compiling data on local corruption cases affecting local public servants and politicians in spain. these data were obtained by examining both published and electronic newspapers in spain. at least two references to an impeached politician were needed before the case of corruption was added to the dataset (see jiménez and garcía, 2018).1 following this principle, we define an alleged offence as a case of corruption only when the accused had been impeached before the court. clearly, however, we need to distinguish between an accusation of corruption and a proven case of corruption. thus, in our database politicians are considered as impeached but not guilty if a judgment had not been passed during the period considered in our study. note that our corruption variable will be therefore built on real corrupt behavior, rather than being an index based on subjective perception of corruption, which has been a common strategy to deal with corruption measurement. however, subjective indexes have limitations because perception is not always perfectly correlated with real facts.2 fortunately, we follow the strategy of previous studies like fisman and gatti (2002b) and fisman and miguel (2007), which also compiled data on real corrupt activities. second, we use the transparency information made public by the ngo transparency international-spain in order to construct a variable that captures the degree of willingness on the part of the municipal government to be transparent. note that spain is one of the few european countries yet to have a transparency law and local governments are not, therefore, required to provide information on their activities or decision-making procedures. as a consequence, actual information disclosure is a voluntary act by governments, which enables us to identify differences across municipalities. given this situation of voluntary transparency, the ngo transparency international (ti 1 the database is compiled as in jiménez and garcía (2012). we first recorded all the electronic news items referring to local corruption in spain and then checked them to both national and regional newspapers. in fact, google publishes a “local corruption map” on which most of these cases are identified. our database contains a similar number of cases to those on other databases. 2 see tresman (2007) for a discussion on corruption indexes and their limitations. j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 111 spain) has been producing a transparency index for local governments in spain since 2008. this index contains data for the years 2008, 2009, 2010 and 2012 and is based on the collection of information and survey responses gathered from municipalities examining five areas of transparency: transparency of the municipal corporation, social transparency in relations with citizens, fiscal transparency, urban planning and procurement transparency, and transparency in the contracting of services.3 this index has been already used in guillamón et al. (2011) and albalate (2013). ti’s indexes range from 0 to 100, with 100 indicating a perfectly transparent government and 0 a government with no transparency at all. the transparency index is constructed with the affirmative and negative responses of municipalities to the 80 items considered and with information collected by ti directly from public sources. by using this index, we are able to exploit differences in the characteristics of the municipalities and the cases of corruption reported in these administrative areas. as transparency international does not cover all the municipalities in spain, our empirical strategy is limited to examining the relationship between corruption and transparency in a subset of the 110 largest local governments, that is, those included in the ti reports. table 1 reports the number of annual corruption cases since 2008 (we also include as informative previous cases from 2000 to 2007), identifying specifically those affecting the municipalities considered in the ti reports. in spite of the limitations produced by the available index, our sample includes all corruption cases reported in the 110 largest municipalities in the country. table 1. local corruption cases in spain by year period total number4 corruption cases included in our sample cases in sample as a percentage of total number of cases in spain 1999-2007 42 1 2 2008 19 3 16 2009 38 9 24 2010 54 11 20 2011 53 10 19 total 206 34 17 source: own elaboration from jiménez and garcía (2018). 3 for more information on the methodological process adopted in the conducting of this survey, see transparency international-spain’s website: http://www.transparencia.org.es/ 4 total cases of local corruption in municipalities with population higher than 1,000 inhabitants, which they represent more than 94 per cent of total population in spain. http://www.transparencia.org.es/ j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 112 3.3 variables following the literature on local government corruption, we introduce the following variables: dependent variable corrupt: binary variable that takes a value of 1 if there has been a case of corruption in municipality i in year t, between the period 2008 and 2012 – year 2011 is missing for the purpose of this study because transparency international did not compute the transparency index for that year. thus, we have four observations per municipality, one per year. source: authors’ own data (see previous paragraphs). explanatory variables transparency: a continuous variable that ranges from 0 to 100, where 100 is the value associated with a local government that satisfies all the transparency requisites included on the transparency international questionnaire; and 0 is the value associated with a local government that does not supply any of the information required on the questionnaire. as this variable may well be endogenous, given that corrupt politicians have considerable incentives to provide less transparency and honest politicians may be willing to provide more transparency, instruments are required. these instruments, which have been shown to be significant determinants of local government transparency in albalate (2013), are: i) the ideology of the mayor, taking a value of 1 if the mayor belongs to a left-wing party. ii) the size of the majority enjoyed by the mayor in the city council plenary. population: the population of municipality i in year t. this variable controls for the size of the municipality. other papers have found that the size is a factor that affects corruption (see jiménez, 2013, for example). source: la caixa municipal database. unemployment: the unemployment rate in municipality i in year t. this variable captures the economic situation of the municipality. source: la caixa municipal database. debt per capita: debt per capita of municipality i in year t. this variable controls for the financial accounts of the municipality. source: general database of local entities, spanish ministry of treasury. provincial_capital: binary variable taking a value of 1 if municipality i is a provincial capital and 0 otherwise. this variable controls for the greater degree of scrutiny to which the governments of provincial capital are exposed. source: la caixa municipal database. tourism: comparative index measuring the importance of the tourist sector, based on 2010 j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 113 information. the index is based on the tax on the economic activities included under the heading of tourism, which in turn is based on the tourist establishment category, number of rooms and annual occupancy (throughout the year or for just part of a year). as such it serves as a good indicator of tourism. the variable controls for differences in the intensity of tourism, a sector that accounts for the positive correlation between tourist activity and construction and commercial development. in the spanish coastal municipalities, especially, such development has been found to be a potential source of corruption at the local government level. in fact, jiménez et al (2014) have found a positive relationship among tourism municipalities and local corruption using a bigger sample than we use in our paper. source: la caixa municipal database. trend: a variable indicating the year of the observation. it is included to control for potentialtime effects in the data pool. we included the year to control for differences over time that could affect the probability of having a corruption case. we expect less cases of corruption over time as a general trend affecting all municipalities. the descriptive statistics of these variables are shown in table 2. the information is presented by non-uncovered and corrupt municipalities.5 the percentage of corrupt municipalities in the sample is 32%, while 68% did not report any cases of corruption between 2008 and 2012. note that the average transparency indicator is 3.6 per cent higher in nonuncovered municipalities (63.1 versus 60.93). we provide information on the mean, standard deviation, minimum and maximum values of all the variables included. an initial inspection of these descriptive statistics shows that corrupt municipalities have a slightly higher population, enjoy lower rates of unemployment and debt and tend to have higher intensities of tourist activities. table 2. descriptive statistics by municipality variables mean std. dev. minimum maximum c non-u c non-u c nonu c non-u transparency 60.93 63.1 20.76 21.18 12.5 17.5 100 100 population 179,419 170,763.9 0 159,526 207,154.2 0 54,600 35,03 7 7041,98 1,621,53 7 unemployment rate 10.1 10.8 3.4 3.6 3.4 3.7 17.3 20.9 debt per capita 61.57 73.05 47.81 57.49 0 7.52 440.2 542.95 provincial_capital 0.44 0.46 0.5 0.5 0 0 1 1 tourism 549.94 285.88 1369 477.4 9 3 7532 2973 note: c: corrupt municipality; non-u: non-uncovered municipality. 5 we refer to these municipalities as “uncovered” because we cannot confirm whether they are corrupt or not just with the absence of uncovered corruption scandals. j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 114 3.4 the method we specify the following model that is then estimated using an instrumental variable probit method. to solve the problem of endogeneity of transparency we use a two-stage estimation procedure, employing in the first stage of the estimation the instruments presented above. corruptedit = α i + ρ1transparencyit + β1populationit + β2unemploymentit + +β3debt _ pcit + β4 di capital + β5tourismi + β6yeart + εit [1] given this model, ρ1 is the coefficient estimating the impact of transparency on the probability that cases of corruption are uncovered in a given municipality. we expect a negative coefficient estimate indicating that high transparency indexes are signals of a low likelihood of corruption. 3. results table 3 shows our main results obtained from the instrumental variable probit models. all the statistical tests examining the strength and validity of the instruments used to treat the endogeneity of transparency give satisfactory results. the wald test shows there to be no endogeneity and the data treatment to be appropriate (chi2(1) = 2.40 prob>chi2 = 0.12), while the instruments can be considered strong and valid according to the kleibergen-paap rk lm statistic (chi2(2)=0.003) and the hansen j-statistic of overidentification (chi-sq(1) p-value = 0.32). columns (1) and (2) display results for the pooled models in which we estimate the relationship between transparency and corruption. the only difference is that standard errors account for all types of correlation within the same municipalities in model (2) by using municipality clusters. as expected, there is an inverse relationship between transparency and corruption, which implies that transparency prevents corruption or is, at least, a good proxy of good governance, which deters corruption. corruption also appears to be negatively related to the unemployment rate and a city’s status as a provincial capital, reflecting the higher degree of scrutiny to which a government is exposed by the media and citizenship. similarly, the size of a municipality’s population is statistically significant, having a positive impact on the probability of cases of corruption. this means there is more probability of cases of corruption being uncovered in large municipalities, ceteris paribus. these results serve to minimize any potential limitations in our sample, given that we have used data from spain’s largest municipalities. the same positive impact is found with the trend variable, indicating that a greater number of cases of corruption are uncovered over time. finally, we find no evidence of a relationship j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 115 between the intensity of tourism and local government corruption in our sample. although we report a positive coefficient, our model does not support the hypothesis that municipalities with highly developed tourist sectors (and which have experienced major construction and commercial developments and, consequently, more opportunities for private appropriation of rents) present a higher probability of suffering political corruption. as for the estimation procedure, it should be noted that the inclusion of clusters in the error estimations in model (2) does not seem to affect out main result, thus confirming its consistency. however, other coefficients are affected by the inclusion of these clusters. in model (2), the statistical significance of provincial capital falls, while in the case of population the significance disappears altogether. robustness check although there is a clearly negative correlation between transparency and the number of corruption cases, it could be argued that models (1) and (2) fail to deal appropriately with the timing of transparency measures and corruption cases, given that levels of transparency may well improve after cases of corruption have been reported. however, a review of changes in transparency following cases of corruption does not support this argument. to demonstrate the robustness of this analysis, we estimate model (3) taking into consideration only the data for 2008, the first year in our sample (i.e., the first year for which transparency indexes became available). in this specification [2], we replicate the estimation with all the observations for 2008 and provide estimates for a model in which the dependent variable takes a value of 1 when there is a corruption case after 2008 and 0 otherwise. corruptedi = α i + ρ1transparencyi,2008 + β1populationi,2008 + β2unemploymenti,2008 + +β3debt _ pci,2008 + β4 di capital + β5tourismi,2008 + εi,2008 [2] in this way, we are able to determine whether the transparency reported in the first year for which information was available might account for the cases of corruption that appeared over the short-term. column (3) displays these results and shows that the coefficient associated with transparency remains negative and statistically significant. this means that 2008 transparency is inversely related to the probability of cases of corruption occurring between 2008 and 2012. our results support the hypothesis that the lack of transparency might conceal corruption or, at least, account for the lower probability of corruption remaining hidden. j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 116 table 3. estimates from iv probit two stage procedure. explanatory variables corruption iv probit pooled 2008-2012 (1) corruption iv probit pooled 20082012 (2) corruption iv probit year 2008 (3) transparency -0.0401*** (0.0053) -0.0401*** (0.0086) -0.0495*** (0.0117) debt_pc -0.0019 (0.0019) -0.0019 (0.0020) -0.0009 (0.0021) unemployment -0.0858*** (0.0179) -0.0858*** (0.0284) -0.0924** (0.0404) population 1.00e-06*** (3.88e-07) 1.00e-06 (7.30e-07) 1.41e-06* (8.6e-07) province_capital -0.3679*** (0.1228) -0.3679* (0.2106) -0.2812 (0.2544) tourism 0.0001 (0.0001) 0.0001 (0.0001) 0.00004 (0.0001) trend 0.3881*** (0.0723) 0.3881*** (0.0827) clusters no yes no observations 407 407 95 wald chi2 130.14*** 53.12*** 29.98*** log pseudolikelihood -2032 -2033 -455.7 note: *** 1%, ** 5%, *10% significance test. robust to heteroskedasticity. standard errors shown in parenthesis. in model (2) we used clustered standard errors at the municipal level. 4. conclusions the relationship between transparency and corruption is coming under the increasing scrutiny of both academia and policy makers, but while recent studies have devoted first efforts to examining this relationship at national and international levels, there is still a lack of research at local government level. this paper offers early empirical evidence that reports an inverse relationship between transparency indexes (willingness to provide information) and political corruption in spain’s 110 largest local governments. after correcting for the endogeneity of transparency, our results suggest that local governments with low levels of transparency are more likely to suffer cases of political corruption than municipalities that provide greater amounts of information. in the absence of a specific transparency law providing for the compulsory provision of information to the public, it has been possible to conduct this test thanks to the existence of voluntary information disclosure at the local level in spain.6 6 a law on transparency has recently been passed in spain, although our database does not contain data after that change. ley 19/2013, 09 de diciembre, ley de transparencia, acceso a la información pública y buen gobierno. j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 117 it would seem that this result reflects the preventive action associated with increased transparency, which ensures greater scrutiny of political actions, or, alternatively, it might be the case that honest politicians are more willing to offer information. although we are unable to identify the specific mechanism or source accounting for this negative causality, our results clearly show that transparency indexes explain in part the probability of corruption. and for this reason transparency is a powerful tool in the hands of policy makers in their efforts to stamp out corruption. our study uncovers a clear negative relationship between transparency and real corruption activities and overcomes any of the problems of the research based on subjective corruption perceptions. however, a limitation of this study is that transparency might be correlated with better governance or other features of government performance that can be, in turn, linked to a more honest behavior of politicians and public servants. however, the availability of data does not offer the possibility to execute an experimental design able to fully identify the solely effect of transparency indexes. even in this case, our results show that transparency indexes are worth taking into account in the analysis of corruption, offering a contribution for further research on corruption determinants also at local government level. despite the statistical significance associated with transparency in our results, we cannot fully reject arguments that transparency on its own may be insufficient (kolstad and wiig 2009; lindstedt and naurin, 2010). transparency levels appear unable to offer unequivocal predictions of the probability of corruption and it seems that other mechanisms of support (sanctions, social embarrassment, education, broadcasting media, etc.) are required to control corruption. indeed, the promise of transparency is not satisfied with the mere existence of laws, but would seem to depend on institutional quality (fung et al., 2007; hood and heald, 2006). however, our results show the undeniable importance of transparency in predictions of the likelihood of corruption. indeed, the evidence reported here supports the implementation of policies aimed at promoting public information disclosure as a deterrent to political corruption and as a mechanism for enhancing accountability in local government. acknowledgements we thank comments and suggestions made by two anonymous referees. nevertheless, all errors are ours. references acemoglu, d., johnson, s., robinson, j.a. (2002) reversal of fortune: geography and institutions in the making of the modern world income distribution, quarterly journal of economics, 118, 1231-94. doi: https://doi.org/10.1162/003355302320935025 https://doi.org/10.1162/003355302320935025 j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 118 aidt, t. (2009) corruption, institutions, and economic development. oxford review of economic policy, 25(2), 2009, 271–291. doi: https://doi.org/10.1093/oxrep/grp012 albalate, d. 2013. the institutional, social and economic determinants of local government transparency. journal of economic policy reform, 16(1), 90-107. doi: https://doi.org/10.1080/17487870.2012.759422 ali, m. a & isse, h.s. (2003) determinants of economic corruption. cato journal 22(3), 449-466. alt, j. & lassen, d. (2006a) fiscal transparency, political parties and debt in oecd countries, european economic review, 50, pp. 1403-1439. alt, j. e., & lassen, d. (2006b). transparency, political polarization, and political budget cycles in oecd countries. american journal of political science, 50(3), 530–550. doi: https://doi.org/10.1016/j.euroecorev.2005.04.001 alt, j., lassen, d. & rose, s. (2006) the causes of fiscal transparency: evidence from the american states, imf staff papers, 53, pp. 30-57. doi: https://doi.org/10.1111/j.15405907.2006.00200.x anderson, c., tverdova, y. (2003) “corruption, political allegiances, and attitudes toward government in contemporary democracies”, american journal of political science, 47(1), 91-109. doi: https://doi.org/10.1111/1540-5907.00007 andvig, j.c. & moene, k.o. (1990) “how corruption may corrupt”, journal of economic behavior and organization, 13(1), 63-76. doi: https://doi.org/10.1016/0167-2681(90)90053-g bardhan, p. (1997) corruption and development: a review of issues. journal of economic literature, 35(3), 1320-1346. baumol, w. (1990) entrepreneurship: productive, unproductive, and destructive. journal of political economy 98, 893-921. doi: https://doi.org/10.1086/261712 becker, g. (1968) crime and punishment: an economic approach, journal of political economy, 76, pp. 169-217. doi: https://doi.org/10.1086/259394 bellver, a. & kaufmann, d (2005) transparenting transparency: initial empirics and policy applications (the world bank, washington dc). blackburn, k. and g.f. forgues-puccio, 2009. why is corruption less harmful in some countries than in others? journal of economic behaviour and organisation,72(3), 797-810. doi: https://doi.org/10.1016/j.jebo.2009.08.009 blumkin, t. & gradstein, m. (2002) transparency gloves for grabbing hands? politics and (mis)governance, cepr discussion paper no. 3668. brunetti, a., weder, b. (2003) a free press is bad news for corruption. journal of public economics 87(7-8), 1801-1824. doi: https://doi.org/10.1016/s0047-2727(01)00186-4 bowler s. & karp, j. (2004) politicians, scandals and trust. political behavior, 26(3), 271-287. doi: https://doi.org/10.1023/b:pobe.0000043456.87303.3a calderón, c. & chong, a. (2006) do democracies breed rent-seeking behavior? the journal of policy reform 9(4), 247-260. doi: https://doi.org/10.1080/13841280601079209 costa, s. (2012) do freedom of information laws decrease corruption? journal of law, economics, and organization advance access published may 18, 2012 https://doi.org/10.1093/oxrep/grp012 https://doi.org/10.1080/17487870.2012.759422 https://doi.org/10.1016/j.euroecorev.2005.04.001 https://doi.org/10.1111/j.1540-5907.2006.00200.x https://doi.org/10.1111/j.1540-5907.2006.00200.x https://doi.org/10.1111/1540-5907.00007 https://doi.org/10.1016/0167-2681(90)90053-g https://doi.org/10.1086/261712 https://doi.org/10.1086/259394 https://doi.org/10.1016/j.jebo.2009.08.009 https://doi.org/10.1016/s0047-2727(01)00186-4 https://doi.org/10.1023/b:pobe.0000043456.87303.3a https://doi.org/10.1080/13841280601079209 j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 119 doi: https://doi.org/10.1093/jleo/ews016 costas-pérez, e., solé-ollé, a. & sorribas-navarro, p. (2012) corruption scandals, voters reporting, and accountability. european journal of political economy, 28(4), 469-484. doi: https://doi.org/10.1016/j.ejpoleco.2012.05.007 del monte, a., papagni, e. (2007) “the determinants of corruption in italy: regional panel data analysis”, european journal of political economy 23, 379-396. doi: https://doi.org/10.1016/j.ejpoleco.2006.03.004 di rienzo, c., das, j., cort, k. & burbridge, j. (2007) corruption and the role of information. journal of international business studies 38, 320 -322. doi: https://doi.org/10.1057/palgrave.jibs.8400262 eingen, p. (2002) measuring and combating corruption, journal of economic policy reform, 5(4), pp.187-201. doi: https://doi.org/10.1080/1384128032000096805 elster, j. (1998) deliberation and constitution making, in jon elster (ed), deliberative democracy. cambridge: cambridge university press. doi: https://doi.org/10.1017/cbo9781139175005.006 esteller, a. & polo-otero, j. (2012) fiscal transparency: (why) does your local government respond? public management review, 14(8), 1153-1173. doi: https://doi.org/10.1080/14719037.2012.657839 ferraz, c. & finan, f. (2008) exposing corrupt politicians: the effects of brazil's publicly released audits on electoral outcomes. the quarterly journal of economics, 123(2), 703745. doi: https://doi.org/10.1162/qjec.2008.123.2.703 ferraz, c. & finan, f. (2011) electoral accountability and corruption: evidence from the audits of local governments. american economic review, 101(4), 1274-1311. doi: https://doi.org/10.1257/aer.101.4.1274 fernández-vázquez, p. & rivero, g. (2010) electoral consequences of corruption charges in andalucía (spain), 2003-2007. mimeo. fisman, r., gatti, r. (2002a). “decentralization and corruption: evidence across countries”. journal of public economics, 83, 325–345. doi: https://doi.org/10.1016/s00472727(00)00158-4 fisman, r., gatti, r. (2002b) “decentralization and corruption: evidence from u.s. federal transfer programs”, public choice, 113, 25-35. doi: https://doi.org/10.1023/a:1020311511787 fisman, r., miguel, e. (2007) “corruption, norms, and legal enforcement: evidence from diplomatic parking tickets”, journal of political economy, 115(6), 1020-1048. doi: https://doi.org/10.1086/527495 fung, a., graham, m., & weil, d., (2007) full disclosure: the perils and promise of transparency. cambridge, uk: cambridge university press. doi: https://doi.org/10.1017/cbo9780511510533 gentzkow m., glaeser e. & goldin c. (2006) the rise of the fourth estate: how newspapers became informative and why it mattered. in: glaeser e, goldin c (eds) corruption and reform: lessons from america’s economic history, pp. 187–230. nber and chicago https://doi.org/10.1093/jleo/ews016 https://doi.org/10.1016/j.ejpoleco.2012.05.007 https://doi.org/10.1016/j.ejpoleco.2006.03.004 https://doi.org/10.1057/palgrave.jibs.8400262 https://doi.org/10.1080/1384128032000096805 https://doi.org/10.1017/cbo9781139175005.006 https://doi.org/10.1080/14719037.2012.657839 https://doi.org/10.1162/qjec.2008.123.2.703 https://doi.org/10.1257/aer.101.4.1274 https://doi.org/10.1016/s0047-2727(00)00158-4 https://doi.org/10.1016/s0047-2727(00)00158-4 https://doi.org/10.1023/a:1020311511787 https://doi.org/10.1086/527495 https://doi.org/10.1017/cbo9780511510533 j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 120 university press, cambridge, ma and chicago, il. doi: https://doi.org/10.7208/chicago/9780226299594.003.0007 glaeser, e., sak, r. (2006) corruption in america, journal of public economics, 90, 10531072. https://doi.org/10.1016/j.jpubeco.2005.08.007 glennerster, r., and y. shin. 2008. does transparency pay? imf staff papers 55(1), 183–209. doi: https://doi.org/10.1057/palgrave.imfsp.9450028 gonzález, b., jiménez, j.l. & perdiguero, j. (2017). danger: local corruption is contagious. journal of policy modelling, 39, 790-808. https://doi.org/10.1016/j.jpolmod.2017.08.002 goel, r. k., nelson, m. a. (1998). corruption and government size: a disaggregated analysis. public choice, 97, 107–120. doi: https://doi.org/10.1023/a:1004900603583 goel, r. k., & nelson, m. a. (2010). “causes of corruption: history, geography and government”. journal of policy modeling, 32(4), 433-447. doi: https://doi.org/10.1016/j.jpolmod.2010.05.004 grigorescu, a. (2003) international organizations and government transparency: linking the international and domestic realms, international studies quarterly, 47(4), 643–667. doi: https://doi.org/10.1046/j.0020-8833.2003.04704003.x gupta, s., davoodi, h. & alonso-terme, r. (2002) does corruption affect income inequality and poverty? economics of governance, 3(1), 23-45. doi: https://doi.org/10.1007/s101010100039 guillamón, m.d., bastida, f. & benito, b. (2011) the determinants of local governments’ financial transparency. local government studies, 37(4), 391-406. doi: https://doi.org/10.1080/03003930.2011.588704 hameed, f. 2005. “fiscal transparency and economic outcomes.” imf working paper wp/05/225. washington: international monetary fund. holmström, b. (1979) moral hazard and observability. the bell journal of economics 19(1), 7491. doi: https://doi.org/10.2307/3003320 hood, c. & heald, d (2006) transparency. the key to better governance. oxford: oxford university press. doi: https://doi.org/10.5871/bacad/9780197263839.001.0001 ibp (international budget partnership) (2013) open budget survey 2012.washington. jiménez, j.l. & garcía, c. (2018) does local public corruption generate partisan effects on polls? crime, law and social change, 69, 3-23. doi: https://doi.org/10.1007/s10611-0169671-1 jiménez, jl. (2013) corrupción local en españa. información comercial española, revista de economía, 85, 23-41. jiménez, j.l., nombela, g. and suárez-alemán, a. (2014) tourism, political corruption and voting patterns in spanish municipalities, mimeo. kaufmann, d. & kraay, a. (2002) growth without governance, policy research working paper 2928, world bank. knack, s. & keefer, p. (1995) institutions and economic performance: cross country test using alternative institutional measures, economics and politics, 7(3), pp. 207-227. https://doi.org/10.7208/chicago/9780226299594.003.0007 https://doi.org/10.1016/j.jpubeco.2005.08.007 https://doi.org/10.1057/palgrave.imfsp.9450028 https://doi.org/10.1016/j.jpolmod.2017.08.002 https://doi.org/10.1023/a:1004900603583 https://doi.org/10.1016/j.jpolmod.2010.05.004 https://doi.org/10.1046/j.0020-8833.2003.04704003.x https://doi.org/10.1007/s101010100039 https://doi.org/10.1080/03003930.2011.588704 https://doi.org/10.2307/3003320 https://doi.org/10.5871/bacad/9780197263839.001.0001 https://doi.org/10.1007/s10611-016-9671-1 https://doi.org/10.1007/s10611-016-9671-1 j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 121 doi: https://doi.org/10.1111/j.1468-0343.1995.tb00111.x kolstad, i. & wiig, a., (2009) is transparency the key to reducing corruption in resource-rich countries? world development 37 (3), 521–532. doi: https://doi.org/10.1016/j.worlddev.2008.07.002 la porta, r., lopez-de-silanes, f., shleifer, a., vishny, r. (1999) “the quality of government.” journal of law, economics, and organization 15, no. 1 (april): 222–79. doi: https://doi.org/10.1093/jleo/15.1.222 lindstedt, c. & naurin, d., (2010) transparency is not enough: making transparency effective in reducing corruption. international political science review 31(3), 301–321. doi: https://doi.org/10.1177/0192512110377602 mauro, p. (1995) corruption and growth. quarterly journal of economics, 110 (3), 681-712. doi: https://doi.org/10.2307/2946696 meijer, a. j. (2003) transparent government: parliamentary and legal accountability in an information age. information polity, 8(1-2), 67-78. doi: https://doi.org/10.3233/ip-2003-0027 meon, p.-g. & weill, l. (2008) is corruption an efficient grease? bofit discussion papers 20/2008. bank of finland, institute for economies in transition. meon, p.-g. & sekkat, k. (2005) does corruption grease or sand the wheels of growth. public choice 122, 69-97. doi: https://doi.org/10.1007/s11127-005-3988-0 mishler, w., rose, r. (2001) “what are the origins of political trust? testing institutional and cultural theories in post-communist societies”, comparative political studies 34(1), 30-62. doi: https://doi.org/10.1177/0010414001034001002 mo, p. h. (2001), corruption and economic growth. journal of comparative economics, 29, 66–79. doi: https://doi.org/10.1006/jcec.2000.1703 mocan, n. (2008) “what determines corruption? international evidence from microdata”, economic inquiry, 46(4), 493-510. doi: https://doi.org/10.1111/j.1465-7295.2007.00107.x murphy. k, schliefer. a. & vishny. r, (1993) why is rent-seeking so costly to growth? american economic review, 83(2), 409-14. naurin, d. (2007) deliberation behind closed doors. transparency and lobbying in the european union. colchester: ecpr press. olken, b.a. (2007) “monitoring corruption: evidence from a field experiment in indonesia”, journal of political economy, 115(2), 200-249. doi: https://doi.org/10.1086/517935 olken, b.a. & pande, r. (2012) “corruption in developing countries” annual review of economics 4, 479-505. doi: https://doi.org/10.1146/annurev-economics-080511-110917 olson, m., sarna, n. & swamy, a. (2000) governance and growth: a simple hypothesis explaining cross-country differences in productivity growth. public choice 102(3-4), 341-364. doi: https://doi.org/10.1023/a:1005067115159 peisakhin, l. & pinto, p. (2010) is transparency an effective anti-corruption strategy? evidence from a field experiment in india. regulation and governance 4, 261-280. doi: https://doi.org/10.1111/j.1748-5991.2010.01081.x rosendorff, b.p. & doces. j. (2006) transparency and unfair eviction in democracies and https://doi.org/10.1111/j.1468-0343.1995.tb00111.x https://doi.org/10.1016/j.worlddev.2008.07.002 https://doi.org/10.1093/jleo/15.1.222 https://doi.org/10.1177/0192512110377602 https://doi.org/10.2307/2946696 https://doi.org/10.3233/ip-2003-0027 https://doi.org/10.1007/s11127-005-3988-0 https://doi.org/10.1177/0010414001034001002 https://doi.org/10.1006/jcec.2000.1703 https://doi.org/10.1111/j.1465-7295.2007.00107.x https://doi.org/10.1086/517935 https://doi.org/10.1146/annurev-economics-080511-110917 https://doi.org/10.1023/a:1005067115159 https://doi.org/10.1111/j.1748-5991.2010.01081.x j.l. jiménez and d. albalate / european journal of government and economics 7(2), 106-122. 122 autocracies, swiss political science review, 12(3), pp. 99. stigler, g. (1961) the economics of information. journal of political economy, 69, 213-225. doi: https://doi.org/10.1086/258464 stiglitz, j. (2000) the contributions of the economics of information to twentieth century economics. quarterly journal of economics, 115(4), 1441-1478. doi: https://doi.org/10.1162/003355300555015 stiglitz, j. (2002) “information and the change in the paradigm in economics. american economic review, 92(3), 460-501. doi: https://doi.org/10.1257/00028280260136363 stiglitz, j. & weiss, a. (1981) credit rationing in markets with imperfect information. american economics review, 71(3), 393-410. styles, a. & tennyson, m. (2007) the accessibility of financial reporting of us municipalities on the internet, journal of public budgeting, accounting and financial management, 19(1), pp. 56-92. swaleheen, m. (2011) "economic growth with endogenous corruption: an empirical study”, public choice, 146, 23-41. doi: https://doi.org/10.1007/s11127-009-9581-1 you, j., & khagram, s. (2005) “a comparative study of inequality and corruption”. american sociological review, 70, 136-157. doi: https://doi.org/10.1177/000312240507000107 wehner, j. & de renzio, p. (2013) citizens, legislators, and executive disclosure: the political determinants of fiscal transparency, world development, 41, 96-108. doi https://doi.org/10.1016/j.worlddev.2012.06.005. williamson, o.e (2000) the new institutional economics: taking stock; looking ahead, journal of economic literature, 38, pp. 595-613. doi: https://doi.org/10.1257/jel.38.3.595 https://doi.org/10.1086/258464 https://doi.org/10.1162/003355300555015 https://doi.org/10.1257/00028280260136363 https://doi.org/10.1007/s11127-009-9581-1 https://doi.org/10.1177/000312240507000107 https://doi.org/10.1016/j.worlddev.2012.06.005 https://doi.org/10.1257/jel.38.3.595 ejge_front_7_2 ejge oficial 7-2 content page 7-2-1 abstract. this paper empirically investigates the causal relationship between local government transparency and political corruption in a sample of spain's 110 largest municipalities. after implementing a two-stage probit estimation procedure, our evi... keywords. political corruption, transparency, local government. doi. https://doi.org/10.17979/ejge.2018.7.2.4509 3.2 data on corruption and transparency references do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey vol.9 • no.1 2020 issn: 2254-7088 european journal of government and economics 9(1), june 2020. european journal of government and economics issn: 2254-7088 number 9, issue 1, june 2020 doi: https://doi.org/10.17979/ejge.2020.9.1 the politics of renewable power in spain 5-25 doi: https://doi.org/10.17979/ejge.2020.9.1.5231 john s. duffield voting turnout in greece: expressive or instrumental? 26-45 doi: https://doi.org/10.17979/ejge.2020.9.1.5426 irene daskalopoulou the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach 46-73 doi: https://doi.org/10.17979/ejge.2020.9.1.5714 luzius stricker and moreno baruffini do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey 74-94 doi: https://doi.org/10.17979/ejge.2020.9.1.5948 esra n. kilci a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 95-114 doi: https://doi.org/10.17979/ejge.2020.9.1.5847 barbara pawełek https://doi.org/10.17979/ejge.2020.9.1 https://doi.org/10.17979/ejge.2020.9.1.5231 https://doi.org/10.17979/ejge.2020.9.1.5426 https://doi.org/10.17979/ejge.2020.9.1.5714 https://doi.org/10.17979/ejge.2020.9.1.5948 https://doi.org/10.17979/ejge.2020.9.1.5847 european journal of government and economics 9(1), june 2020, 74-94 74 european journal of government and economics issn: 2254-7088 do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey esra n kilcia* a istanbul arel university, department of international trade and finance, turkey * corresponding author at: esrakilci@arel.edu.tr article history1. received 8 january 2020; first revision required 28 february 2020; accepted 7 april 2020. abstract. the primary aim of this study is to analyze the impact of financial services and real sector confidence indexes on some macroeconomic and financial indicators such as industrial production, inflation, stock market index, foreign exchange rates and interest rates in turkey for the period from may 2012 to may 2019. in this study, the unit root properties of these series are tested by using the narayan and popp (2010) unit root test with two structural breaks and the enders and lee (2012) fourier adf unit root test with multiple structural breaks. we investigate the causal link between confidence indicators and macro-financial variables using the fourier toda yamamoto causality test proposed by nazlioglu et al. (2016). the results suggest a strong link between financial services and real sector confidence indexes on macro-financial indicators such as stock market index and inflation, supporting the evidence of the short-run impact of confidence indexes on these variables. keywords. confidence indicators, stock market index, inflation, structural breaks jel codes. c10, e70, g40 doi. https://doi.org/10.17979/ejge.2020.9.1.5948 1. introduction the turkish economy, which launched a strong recovery period towards minimizing the adverse effects of 2000-01 economic crisis by ensuring fiscal and financial stability, has experienced a substantial growth period with the contribution of the positive global economic atmosphere in the post-crisis period. it is obvious that confidence in the real economy and financial system has played a major role in this development. fiscal discipline, structural reforms aimed at achieving macroeconomic stability and tight monetary and fiscal policies have contributed to the significant increase in confidence and stability in the turkish economy. therefore, the fundamental indicators have shown a significant improvement in the post-crisis period. especially in the five-year-period following the year of 2002, high economic growth rates, gradually decreasing inflation rates and improving fiscal balances have been recorded. the 1 a summary of this study has been presented at the “y-bis 2019 conference: recent advances in data science and business analytics, september 25-28, 2019, istanbul. mailto:esrakilci@arel.edu.tr https://doi.org/10.17979/ejge.2020.9.1.5948 esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 75 structural reforms implemented after the 2000-01 economic crisis have strengthened the public finances and banking sector considerably. also, they helped the turkish economy to suffer less from the recent fluctuations of the 2008-09 global financial crisis and 2010-14 european sovereign debt crisis, which have had globally adverse effects. however, this positive trend has reversed in the last few years, and macro-financial indicators have started to signal negatively. given the importance of confidence for the real economy and financial markets, it is clear that the deterioration tendency of the confidence indicators adversely affects the macroeconomic and financial variables such as gdp growth, employment, foreign exchange rates and interest rates through production, spending and investment decisions of economic agents. simultaneous relationship between expectations and production-spending-investment decisions of economic agents seems not surprising from a theoretical point of view since good or bad expectations, in other words, optimism or pessimism might limit the expenditures of both households and businesses. according to behavioral finance, economic agents and investors do not behave rationally and decide by taking into consideration psychological factors like expectations about the future. therefore, future economic performance might be influenced by consumer and business sentiment. the importance of psychological factors as sentiments and expectations has been emphasized in the analyses by a wide range of economists such as keynes (1936) focusing on consumer and investor sentiments, katona (1951) stating the importance of psychological factors, cass and shell (1983), akerlof and shiller (2009) drawing attention to the government’s role on manipulating animal spirits, farmer (2013), bacchetta and van wincoop (2013), de grauwe and ji (2016) and acharya et al. (2017). they see the root of the macroeconomic fluctuations in strictly psychological waves of optimism-pessimism and sunspot-driven waves and assume that the actions following these waves produce changes in fundamentals making the initial boom or bust in confidence sentiments as expectations ultimately materialize. another group of economists like beaudry and portier (2006), barsky and sims (2012), blanchard et al. (2013) beaudry and portier (2014), furthermore, suppose that agents have access to a nonmeasurable source of incomplete information concerning future developments of the economy; a signal, which make them perform to meet the economy’s future demand today. in this framework, the economy is subject to recurrent booms and occasional busts (nowzohour and stracca, 2017). while forecasting high confidence brings with becoming optimistic about the future, low confidence expectations can lead to pessimism (akerlof and shiller, 2009). typically, the consumers with low expectations for the future might slow down their spending, and the businesses which are pessimistic concerning the economic outlook might postpone or cancel their planned investments (kuzmanović and sanfey, 2012). therefore, economic activity decreases, investment-production falls, and unemployment increases. based upon the basic idea which suggests that the interaction between high uncertainty and non-smooth adjustment frictions might lead businesses to react prudently and also indicates that these businesses, esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 76 facing a more uncertain environment, might stop hiring and investment and present the “wait and see” behaviour that result in a drop in the economic activity, bachman et al (2013) construct measures of time-varying uncertainty from business surveys and investigate their relationship with economic activity over the business cycle for both germany and u.s. they find out that an unexpected change in the the survey-based measures of uncertainty has a relationship with a significant decrease in production and employment in both germany and the u.s. moreoever, baumohl (2012) emphasizes that while the positive expectations of investors is mostly reflected as increases in stock markets, the adverse investor sentiment lead to a decreasing trend in the stock prices. on the other hand, since confidence cannot be observed or measured directly, any evaluation of confidence must depend upon indicators which are partial, qualitative and subject to several interpretations in many cases. in this regard, confidence indices, consistent measures of confidence, are implemented based on surveys to measure how economic decision-makers respond to the economic developments and to express the relationship between expectations and macroeconomic or financial variables. sentiment measures are acquired from surveys capturing assessments on past, current and expected economic developments. although the empirical relationship between sentiment indicators and economic variables is sometimes not obviously established as a result of the subjective nature of the responses, sentiment indicators have the capability to reflect movements in economic variables purely, and they can lead shifts in these variables. therefore, sentiment indicators are quite helpful for economic analysis and forecasting when they bring cyclical economic movements. confidence indicators seem to provide a good picture of serious cyclical changes in output, and they might help detect significant acceleration or deceleration in output growth when great changes in confidence are observed (santero and westerlund, 1996). oral et al. (2005) indicate that there is an increasing concern in terms of the confidence indicators to follow the economic developments and to provide the researchers with the early signals of the turning points in the economic activity, these indicators are used by both the government and the private sector decision-makers in checking their performance and planning their actions. in this context, countries improved their indicator systems by using indexes from surveys. the surveys on expectations are mainly designed to indicate several changes in economic activity and commonly used in macroeconomic estimations and forecasts. the advantage of benefiting from survey results is that these results are available quickly before the related quantitative measures, including the same types of economic activity, are made public. consequently, they are believed as complementary to the official statistics. the major objective of the surveys carried out in several ways is to help economic decision-makers get necessary information about the general tendency of the cyclical developments and future expectations as well. confidence surveys can show changes to the economic outlook and turning points in the economic cycle. according to the findings of surveys, confidence indexes as consumer, real sector and financial services confidence indexes are acquired. in this regard, having an idea about the esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 77 level of consumption, economic activity or financial services is highly important to interpret the current economic outlook. data concerning real economic activity is announced with a large time lag. therefore, timely data at higher frequencies become valuable to understand the current economic situation and anticipate future economic activity. this makes indicators with shorter publication lags quite popular for both policymakers and the public. in this context, the confidence indices, which are one of these timely indicators and released monthly, are cautiously followed by policymakers, analysts and forecasters, as well as media and their signals on private consumption and economic activity, are interpreted carefully. since they are widely considered to be closely linked to current consumption-investment dynamics and also viewed as an instrument to forecast the future direction of these, it is well-established that confidence indices are sensitive to developments in financial markets and also affected by financial volatility (karasoy, 2015; karasoy and yunculer, 2015). in this study, our question is, whether the confidence indexes, which are simply linked with the sentiments and expectations, help us predict the changes in the macro-financial indicators or not. in this context, we focus on the impact of real sector confidence index and financial services confidence index on some macro-financial indicators. according to the definition made by oecd (2019), business confidence indicator, reflecting the expectations in the real sector, give us some information about future developments, based upon opinion surveys on developments in orders-stocks of finished goods and production in the industry sector. this indicator can be used to watch output growth and to predict turning points in economic activity. numbers greater than 100 signal increased confidence in near future business performance and numbers less than 100 suppose pessimism towards future performance. the real sector confidence index is established to provide a key indicator of short-term business conditions for business managers and economic policymakers owing to the need of early warning indicators in order to anticipate financial and economic crises. the aim of the index is to forecast the expansion and contraction periods of economic activity. the real sector confidence index is a critical leading indicator in terms of the economic activity, and it signals to the business confidence based on the business tendency survey. the financial services confidence index, which is used as a second explanatory variable in this study, is based on the answers given to the financial services survey of the cbrt carried out with the institutions operating in the financial sector and which reflects the confidence to the financial services in the system (cbrt, 2019b: 4). the fsci, with monitoring of past assessments and future prospects by executives of financial institutions including bank, insurance companies, factoring, leasing companies and so on, has become a key benchmark that reflects trends and developments in the financial industry. the objective of this study is to evaluate the empirical validity of the real sector and financial services confidence indexes in anticipating the evolution of economic activity by considering monthly data from 2012:05 to 2019:05. in answering whether the confidence indexes have an impact on economic activity, after testing the unit root properties of the series by using the narayan and popp (2012) and enders and lee (2012) fourier adf unit root tests, the causal relationship from the confidence indices to these variables are examined by esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 78 employing the fourier toda yamamoto causality test proposed by nazlioglu et al. (2016). when reviewed the academic literature, it is seen that there are very few studies that concentrate on emerging market economies. to our knowledge, this paper is the first to systematically analyze the relationships from financial services and real sector confidence indexes to macro-financial variables in turkey by modelling the variables under the fourier framework. in this way, we appropriately take into consideration multiple structural breaks without a need of the number, form or date of these breaks through the fourier approach. the study is organised as follows. in the following section, we give brief information pointing to the factors leading to deterioration in confidence indicators and the current economic situation in turkey. after presenting the other studies focusing on confidence indicators, we try to analyze the causality relationship from the real sector and financial services indexes to the macro-financial indicators and report the estimation results acquired by employing fourier toda yamamoto causality test. finally, we conclude by summing up and giving some policy implications. 2. the factors leading to deterioration in confidence indicators in turkey turkey, in the aftermath of the 2000-01 crisis, has shown a healthy macro-economic outlook with the succesful trend of indicators such as growth rate, inflation rate and unemployment rate. confidence in the economy and financial markets has played a major role in this positive trend. however, in the last few years, this trend has reversed, and macroeconomic indicators have started to signal negatively. the turkish economy has been badly affected in the last period, which covers the period of 2016-2018 particularly. economic growth rate decreased to 1,6 percent in 2018:q3, and a negative growth rate is expected in 2019 by economists. in parallel, consumer, real sector and financial services confidence showed a drop in 2018. in a report concerning emerging market economies including turkey published in may 2019 by moody’s, a credit rating agency, weak or deteriorating macroeconomic environment and low economic growth have signaled to the instability of confidence (moody’s report, may 2019). in the report, the inflation rate and the foreign exchange rates are expected to keep their high levels in the following year, too. in another report published by fitch, global economic outlook report, it is emphasized that that low growth has been expected to be triggered by low confidence in 2019 (fitch global economic outlook, 2019). table 1 includes information about the macro-financial indicators in the period of 2012-2018. esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 79 table 1. macro-financial indicators: 2012-2018. indicators 2012 2013 2014 2015 2016 2017 2018q3 2018q4 economic growth rate 2,2 8,5 5,2 6,1 3,2 7,4 1,6 -3,0 inflation rate(cpi) 6,1 7,4 8,2 8,8 8,5 11,9 24,5 20,3 unemployment rate 9,2 9,7 9,9 10,3 10,9 10,9 13,1 13,5 current account deficit -5,5 -6,7 -4,7 -3,7 -3,8 -5,6 -0,8 -1,7 usd/try (average) 1,8 1,9 2,2 2,7 3,0 3,7 5,6 5,5 eur/try (average) 2,3 2,5 2,9 3,0 3,3 4,1 6,5 6,3 benchmark interest rate 9,7 10,0 8,0 10,8 10,7 13,4 25,9 19,7 industrial production index 91,6 100,9 108,5 116,9 117,5 129,9 114,7 117,2 bist 100 index 64,9 67,8 85,7 71,7 78,1 115,3 99,9 91,2 source: tusiad(2019), turkstat(2019), cbrt(2019a). figure 1 and 2 present the trends of fsci and rsci for the period from 2012:05 thorough 2019:05. as seen from the graphs, there has been a clear decline in rsci and fsci since the second half of 2018. most notably, the confidence indicators have continued its moderate trend in 2017.there has been a significant decline in both financial services and real sector confidence indicators by the end of 2018. in december 2018, the fsci rose to the lowest level, recorded as 118.10, since 2012:05. it is seen that it took its highest value, recorded as 187.30 in november 2014. when analyzed the values of rsci, it is seen that it took its lowest value, 87.60, in 2018:10 and it took its highest value, 115.40, in 2012:05. in this period, it is clear that the expectations regarding the general economic situation have become more pessimistic in the second half of 2018. figure 1. financial services confidence index: 2012:05-2019:05. source: cbrt evds (2019a). available at: https://evds2.tcmb.gov.tr/index.php?/evds/seriemarket access date: 15.06.2019. https://evds2.tcmb.gov.tr/index.php?/evds/seriemarket esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 80 figure 2. real sector confidence index: 2012:05-2019:05. source: cbrt evds (2019a). available at: https://evds2.tcmb.gov.tr/index.php?/evds/seriemarket access date: 15.06.2019. the issue of confidence is of great importance for both the real economy and the financial markets. to analyse in more detail what has been driving deterioration in confidence indicators in the recent period, it is necessary to look at the developments in the economy. in turkey, credit expansion in recent years, economic fragility has increased, the increase in the external debt and inflation made economic outlook of 2018 quite risky. in 2018, the capital outflows from emerging market economies, including turkey and some adverse political country-specific uncertainties made turkey confront a serious financial shock. the real economy continues to shrink due to this shock. the large external debt and the credit deposit rate rising to 150 percent have signaled that the economy needs a controlled slowdown. it is seen that foreign debt to gdp ratio and net foreign exchange position of the real sector to gdp ratio have reached 53 percent and 25 percent, respectively. the real sector has faced not only the exchange rate but also the interest rate shock. cash shortage and access to financial sources have become the most important issues as a result of increasing uncertainty. real sector confidence index has seen its lowest level since the 2008-09 crisis. compared to the last year, it is seen that since the usd/try and euro/try rates as well as market interest rates increased significantly, the real economy remains under this heavy financing and debt burden. given the fact that the inflation rate, which rose to 11,9 percent in 2017, was recorded as 20,30 percent as of 2018, the expectations worsen mainly due to the increases in the cost of imported goods. the contraction in the economy led to a rise in the unemployment rate, which was recorded as 12 percent, as well (tusiad, 2019). 3. literature review expectations and uncertainty about future economic and financial outlook, which are of key importance particularly according to the behavioral finance theory, are measured through confidence indexes such as consumer confidence index, real sector confidence index and financial services confidence index, which are regularly followed by central banks, governments and other private institutions. therefore, in the academic literature, the number of studies has https://evds2.tcmb.gov.tr/index.php?/evds/seriemarket esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 81 been increasing on their use in monitoring the current economic situation and predicting economic developments. on the other hand, it is seen that most of these studies focus on consumer confidence indices and some of these deal with business confidence. for instance, santero and westerlund (1996) find in their study, in which they analyze the relationship between confidence indicators based on consumer and business surveys of several oecd countries, that there is a statistically significant relationship between business confidence and gross domestic product, industrial production and real business investment. similarly, after testing the rationality of consumer expectations and assessing their usefulness in forecasting expenditure by using the household data of the michigan survey, souleles (2001), indicate that consumer is useful in predicting future consumption, even checking for lagged consumption and a group of macro-variables like stock prices. in the study carried out by jansen and nahuis (2003), which focus on the short-run relationship between the consumer confidence and stock market developments in a group of european countries in the period of 1986-2001, is found a positive correlation between the changes in consumer sentiment and stock returns for nine countries, except germany. in another study using vector autoregressions which include variables representing consumer confidence, utaka (2003) examines the impact of consumer confidence on the economic activity in japan empirically. based on confidence and gdp monthly, quarterly and semiannually data in the period of 19822000, it is found that consumer confidence does have a positive and significant effect on gdp fluctuations in the case of monthly and quarterly data, while it does have no impact on gdp in the case of semiannual data implying that consumer confidence has an effect on economic fluctuations in short-run. furthermore, ludvigson (2004), reviewing the university of michigan’s consumer sentiment index and the conference board’s consumer confidence index, which are the most widely followed measures of u.s. consumer confidence, focus on the relationship between consumer confidence and consumer spending. he concludes that such measures of consumer attitudes have a significant forecasting power. accordingly, measures of consumer attitudes appear to be directly related to future consumption growth, and confidence surveys can predict future changes in labor earnings and non-stock market wealth. vuchelen (2004), proposing a direct measure of expected economic conditions and uncertainty, test the average growth rate and the dispersion of forecasts in regression analyses for the consumer sentiment in belgium for the period of 1985-2000 by using quarterly data. in contrast with the findings of utaka (2003), his results suggest that a decrease in consumer confidence imply a decrease in the growth rate in belgium by using quarterly data for the period of 1985-2000. in another study, focusing on the relationship between confidence and economic fluctuations, afshar et al. (2007), investigate the link between three confidence measures belonged to businesses, consumers, investors and economic fluctuations by using quarterly data for the u.s. in the period from 1980 to 2005. their findings support the evidence of a causality relationship from confidence measures to gdp. moreover, forecast variance decompositions of gdp suggest that consumer confidence, stock return, and purchasing manger’s index, account for large variations in gdp, meaning that these three measures play esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 82 important roles in the economic fluctuations. in his study focusing on the relationship between the industrial confidence indicator and industrial production index for france, germany, italy and romania, gagea (2012) finds evidence supporting the relationship between the confidence index and industrial production index, even the relationship between the variables for germany and romania seems quite weak. van aarle and kappler (2012) evaluate the potential impact of economic sentiment in business cycle fluctuations by integrating the european commission economic sentiment indicator (esi) data into an empirical analysis of the euro-area business cycle. in this context, after examining the role of changes in economic sentiment indicator (esi) on unemployment, output and retail sales over the period of 1990-2011, they suggest that sentiment shocks have an impact on these macroeconomic variables. they also find out that economic sentiment is affected by economic conditions and shocks as well. in the same way as utaka (2003), utaka (2014) empirically test the effect of consumer confidence on changes in gdp for japanese economy bu using the quarterly data and adopted vector autoregression framework. in addition to utaka (2003), after dividing the total estimation period into two parts as pre-and post-bubble periods, analyze the transition of that effect through the ages by comparing their effects in each period. it is found that although consumer confidence does have a significant effect on gdp, this effect becomes greater in a post-bubble period than in a prebubble period. nowzohour and stracca (2017), focusing on the role of sentiment in the business cycle, review six existing measures of sentiment such as consumer confidence, economic policy uncertainty and stock market volatility. they use monthly panel data for 27 countries, including consumer and business confidence indexes in the period of 1985:01-2016:10. their results indicate that different measures are surprisingly low correlated on average in each country while these measures are highly positively correlated across countries, signaling to the existence of a global impact. they emphasize, of these sentiments, consumer confidence has the closest comovement with economic and financial variables, implying that economic sentiment is a real driver of activity. unfortunately, it is seen there are a few studies focusing on the relationship between confidence indicators and some macroeconomic variables or financial indicators in turkey. oral (2005), kandir (2006), ozsagir (2007), celik and ozerkek (2009), aktas and akdag (2013), arisoy (2012) and iskenderoglu and akdag (2017) empirically analyze the relationship between measures of confidence and macro-financial variables. of these, ozsagir (2007) in his study, after testing the relationship between real sector confidence index and economic growth rates for the period of 1988-2005, indicates that there is a significant positive relationship between rsci and economic growth rate, meaning that the confidence atmosphere does have a positive impact on economic growth. çelik and özerkek (2009) investigate the relationship between consumer confidence and some macroeconomic and financial variables such as personal consumption, stock market index, real exchange rates and interest rates not for turkey but for 9 european union countries. using panel data analysis in the period of 1997-2006, they detect the existence of a long-run relationship between consumer confidence and the relevant variables, particularly signaling that consumers have the capability to detect the early signals esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 83 about future economic growth as they contribute via the channel of consumption. arisoy (2012) analyzing the effect of the confidence indices on consumption spending, employment and stock market indices in the period of 2005:01-2012:01, conclude that real sector confidence index do have impact on both the stock market indices and industrial production index while consumer confidence index do have impact on consumer spending. directly focusing on the financial services confidence index, iskenderoglu and akdag (2017), examine the existence of a relationship between financial services confidence index announced by cbrt and the other variables as the stock market index (bist100) and cbrt net funding in the period beginning from 2015:05 to 2017:08. their findings support the evidence of the causality relationship from fsci to both bist100 index and cbrt net funding. 4. empirical analysis in this section, we study whether confidence indicators provide information about the macrofinancial outlook. following testing the stationary properties of variables with the help of narayan and popp (2012) and enders and lee (2012) fourier adf unit root tests, we then proceed to test if there is a causality relationship from the confidence indicators to each of the series by using the fourier toda yamamoto causality test proposed by nazlioglu et al. (2016). according to the findings, if the asymptotic p-values are below 0.05 as stated in the study of nazlioglu et al. (2016), we, therefore, can identify a causality relationship between the variables which signals that confidence indicators provide us information about the macro-financial outlook. it is emphasized that when fourier functions are used in the analysis, multiple structural breaks are taken into consideration, and the findings can significantly change. 4.1. data confidence based on expectations is one of the key factors determining agents’ decisions. in analyzing the impact of confidence indicators on some macroeconomic fundamentals such as industrial production index and inflation rate and some domestic volatility indicators of financial variables such as foreign exchange rates, interest rates and stock market index, we use monthly data belonged to the financial services confidence index and the real sector confidence index. in turkey, real sector and financial services confidence surveys have been carried out regularly by the central bank of the republic of turkey since 2007:05 and 2012:05, respectively. the surveys are carried out monthly via face-to-face interviews with 1.000 individuals that are chosen through stratified random sampling. economic activity, income, age, education and gender are the criteria which are taken into consideration in forming the strata. in this study, our data set covers the period 2012:05-2019:05. although our sample seems smaller than the samples of the studies which cover other developed economies like the u.s. and some euro-area countries including germany, u.k and france, our data set includes the available data because we have no confidence data belonged to financial services for turkey before 2012. despite this limitation, our data set seems sufficient to carry out our analysis. esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 84 table 2. list of variables. independent variables measure abbreviation confidence indicator 1 financial services confidence index fsci confidence indicator 2 real sector confidence index rsci dependent variables measure abbreviation expected relationship production industrial production index pmi (+) inflation rate consumer inflation index cpi (+) foreign exchange rate usd/try (average) usd (+) foreign exchange rate eur/try (average) eur (+) stock market index bist100 (average) bist (+) interest rates interbank o/n rate (average) o/n (+) variables appear in monthly in our analysis and in log and delta (first difference) in our regressions. as seen at table 2, in our analysis, we employ real sector and financial services confidence indexes as the independent variables while industrial production index, consumer inflation index, stock market index, interest rates and foreign exchange rates as dependent variables. 4.2. stationarity and causality in order to evaluate the impact of the confidence indicators, at first, we try to get adequate information on the stationarity properties of the variables being used in the analysis by employing narayan and popp (2010) as well as enders and lee (2012) fourier adf unit root tests. both tests are based on the augmented dickey-fuller type test. in this context, narayan and popp (2010) develop an augmented dickey-fuller-type test for unit roots which allows for two structural breaks while enders and lee (2012) propose a new unit-root test by using fourier function in the deterministic term in a dickey-fuller type regression framework that can complement the fourier lm and df-gls unit root tests and accounts for multiple structural breaks. it is seen that both these tests do have good size and power properties. furthermore, in the fourier adf (2012) unit root test, contrary to many other methods, it is not essential to know the number, form or date of the structural changes. narayan and popp (2010) think two different specifications by allowing for two breaks in the level of a trending data series and two breaks in the level and slope of a trending data series. in the model, the date of breaks is considered to be unknown. the dgp of a time series 𝑦𝑦𝑡𝑡, which they present, does have two components; (𝑑𝑑𝑡𝑡), deterministic component and (𝑢𝑢𝑡𝑡) stochastic component, as follows: 𝑦𝑦𝑡𝑡 = 𝑑𝑑𝑡𝑡 + 𝑢𝑢𝑡𝑡, (1) 𝑢𝑢𝑡𝑡 = ρut−1 + 𝜀𝜀𝑡𝑡, (2) 𝜀𝜀𝑡𝑡 = 𝜓𝜓∗(𝐿𝐿)𝑒𝑒𝑡𝑡 = 𝐴𝐴∗(𝐿𝐿)−1𝐵𝐵(𝐿𝐿)𝑒𝑒𝑡𝑡, (3) esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 85 with 𝑒𝑒𝑡𝑡~𝑖𝑖𝑖𝑖𝑑𝑑(0,𝜎𝜎𝑒𝑒2). 𝐴𝐴∗(𝐿𝐿) and b(𝐿𝐿), the roots of the lag polynomials, which are of order p and q, are assumed to lie outside the unit of circle. of two different specifications which they consider both for trending data, model 1 (m1) (level) allows for two breaks in level and the model 2 (m2) allows for two breaks in level as well as slope. the specifications of both models differ in how the deterministic component, 𝑑𝑑𝑡𝑡 , is defined as follows: 𝑑𝑑𝑡𝑡𝑀𝑀1 = 𝛼𝛼 + 𝛽𝛽𝛽𝛽 + 𝜓𝜓∗(𝐿𝐿)(𝜃𝜃1𝐷𝐷𝑈𝑈1,𝑡𝑡 | + 𝜃𝜃2𝐷𝐷𝑈𝑈2,𝑡𝑡 | ), (4) 𝑑𝑑𝑡𝑡𝑀𝑀2 = 𝛼𝛼 + 𝛽𝛽𝛽𝛽 + 𝜓𝜓∗(𝐿𝐿)(𝜃𝜃1𝐷𝐷𝑈𝑈1,𝑡𝑡 | + 𝜃𝜃2𝐷𝐷𝑈𝑈2,𝑡𝑡 | + 𝛾𝛾1𝐷𝐷𝑇𝑇1,𝑡𝑡 | + 𝛾𝛾2𝐷𝐷𝑇𝑇2,𝑡𝑡 | ), (5) 𝐷𝐷𝑈𝑈1,𝑡𝑡 | = 1(𝛽𝛽 > 𝑇𝑇𝐵𝐵,𝑖𝑖 | ), 𝐷𝐷𝑇𝑇1,𝑡𝑡 | = 1(𝛽𝛽 > 𝑇𝑇𝐵𝐵,𝑖𝑖 | )(𝛽𝛽 − 𝑇𝑇𝐵𝐵,𝑖𝑖 | ), i=1,2 (6) here, the true break dates are denoted by 𝑇𝑇𝐵𝐵,𝑖𝑖 | , 𝑖𝑖 = 1,2, . 𝜃𝜃1 and 𝛾𝛾𝑖𝑖, the parameters, show the magnitude of the level and slope breaks, respectively. the inclusion of 𝜓𝜓∗(𝐿𝐿) in equations (4) and (5) make breaks occur slowly over time. this process, which is known as the io model, is used. accordingly, the io-type test regressions to test for the unit root hypothesis for m1 and m2 could be derived by merging the structural model (1)-(5). the test equations for m1 and m2 do have the following forms, respectively: 𝑦𝑦𝑡𝑡𝑀𝑀1 = 𝜌𝜌𝑦𝑦𝑡𝑡−1 + 𝛼𝛼1 + 𝛽𝛽∗𝛽𝛽 + 𝜃𝜃1𝐷𝐷(𝑇𝑇𝐵𝐵 | )1,𝑡𝑡 + 𝜃𝜃2𝐷𝐷(𝑇𝑇𝐵𝐵 | )2,𝑡𝑡 + +𝛿𝛿1𝐷𝐷𝑈𝑈1,𝑡𝑡−1 | + 𝛿𝛿2𝐷𝐷𝑈𝑈2,𝑡𝑡−1 | + ∑ 𝐵𝐵𝑗𝑗𝑘𝑘 𝑗𝑗=1 ∆𝑦𝑦𝑡𝑡−𝑗𝑗 + 𝑒𝑒𝑡𝑡 (7) with 𝛼𝛼1 = 𝜓𝜓∗(1)−1[(1 − 𝜌𝜌)𝛼𝛼 + 𝜌𝜌𝛽𝛽] + 𝜓𝜓∗|(1)−1(1 − 𝜌𝜌)𝛽𝛽, the mean lag being 𝜓𝜓∗|(1)−1, 𝛽𝛽∗ = 𝜓𝜓∗(1)−1(1 − 𝜌𝜌)𝛽𝛽, 𝜙𝜙 = 𝜌𝜌 − 1, 𝛿𝛿𝑖𝑖 = −𝜙𝜙𝜃𝜃1 and 𝐷𝐷(𝑇𝑇𝐵𝐵 | )1,𝑡𝑡 = 1(𝛽𝛽 = 𝑇𝑇𝐵𝐵,𝑖𝑖 | + 1), i=1,2. 𝑦𝑦𝑡𝑡𝑀𝑀2 = 𝜌𝜌𝑦𝑦𝑡𝑡−1 + 𝛼𝛼∗ + 𝛽𝛽∗𝛽𝛽 + 𝜅𝜅1𝐷𝐷(𝑇𝑇𝐵𝐵 | )1,𝑡𝑡 + 𝜅𝜅2𝐷𝐷(𝑇𝑇𝐵𝐵 | )2,𝑡𝑡 + +𝛿𝛿1∗𝐷𝐷𝑈𝑈1,𝑡𝑡−1 | + 𝛿𝛿2∗𝐷𝐷𝑈𝑈2,𝑡𝑡−1 | + 𝛾𝛾1∗𝐷𝐷𝑈𝑈𝑇𝑇1,𝑡𝑡−1 | + 𝛾𝛾2∗𝐷𝐷𝑈𝑈𝑇𝑇2,𝑡𝑡−1 | + ∑ 𝐵𝐵𝑗𝑗𝑘𝑘 𝑗𝑗=1 ∆𝑦𝑦𝑡𝑡−𝑗𝑗 + 𝑒𝑒𝑡𝑡, (8) where 𝜅𝜅𝑖𝑖 = (𝜃𝜃𝑖𝑖 + 𝛾𝛾𝑖𝑖), 𝛿𝛿𝑖𝑖∗ = (𝛾𝛾𝑖𝑖 − 𝜙𝜙𝜃𝜃𝑖𝑖) and 𝛾𝛾𝑖𝑖∗ = −𝜙𝜙𝛾𝛾𝑖𝑖), i=1,2. we use the t-statistics of 𝜌𝜌�, denoted 𝛽𝛽𝜌𝜌�, in equations (7) and (8), to test the unit root null hypothesis of 𝜌𝜌 = 1 against the alternative hypothesis of 𝜌𝜌 < 1. here, 𝐷𝐷𝑈𝑈𝑖𝑖,𝑡𝑡 | and 𝐷𝐷𝑇𝑇𝑖𝑖,𝑡𝑡 | , the dummy variables are lagged in equations (7) and (8). https://www.tandfonline.com/doi/full/10.1080/02664760903039883#m0004-5 https://www.tandfonline.com/doi/full/10.1080/02664760903039883#m0004-5 https://www.tandfonline.com/doi/full/10.1080/02664760903039883#m0004-5 https://www.tandfonline.com/doi/full/10.1080/02664760903039883#m0007 https://www.tandfonline.com/doi/full/10.1080/02664760903039883#m0008 esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 86 table 3. narayan and popp (2010) adf unit root test results with two structural breaks. break in level(m1) break in level and trend(m2) test statistic tb1 tb2 test statistic tb1 tb2 fsci -6,004 2014:04 2016:04 -6,614 2014:04 2017:02 rsci -5,701 2016:12 2018:02 -8,485 2015:04 2017:11 pmi -8,548 2013:07 2016:08 -8,586 2014:11 2015:11 cpi -2,973 2016:10 2018:02 -6,891 2016:06 2017:12 usd -5,177 2014:02 2018:02 -6,430 2014:12 2017:10 eur -4,648 2014:07 2018:02 -5,318 2016:04 2018:02 lbist -4,487 2014:02 2016:11 -4,621 2016:03 2018:02 o/n -5,509 2016:05 2018:02 -6,836 2013:10 2017:08 notes: critical values for m1 = -4.922, -4.191, -3.823 at 1%, 5%, 10%, respectively. critical values for m2 = -5.380, -4.631, -4.251 at 1%, 5%, 10%, respectively. tb1 and tb2 are the dates of the structural breaks. the test based on 5000 replications. table 3 presents the findings of the narayan and popp (2010) unit root test. it is seen that in all cases, the test rejects the unit root null for m1 and m2 except cpi series for which the test can not reject the unit root for m1 while it also becomes stationary for m2, implying that all series are stationary under two structural breaks. the break dates are closely related to the critical period 2014-2018 in which the political risk increased significantly, and the confidence indicators demonstrated a deteriorating trend in turkey. after checking the stationary properties of series by using narayan and popp (2010) unit root test which allows for two structural breaks, we then proceed by employing enders and lee (2012) fourier adf unit root test. enders and lee (2012) consider the following dickey-fuller test in which the deterministic term is a time-dependent function specified by α(t): yt = α(t) + ρyt −1+ γt+εt , (9) where εt is a stationary disturbance with variance σε2 and α(t) is a deterministic function of t. enders and lee (2012) try to test the null hypothesis of a unit root (i.e., ρ = 1). any test for ρ = 1 is problematic if α(t) is misspecified, when the form of α(t) is unknown. as an approximation of the unknown functional form of α(t), they consider the fourier expansion: α(t) = α0 + � �α𝑘𝑘 sin 2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 + β𝑘𝑘 cos 2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 � 𝑛𝑛 𝑘𝑘=1 ; n ≤ t/2, (10) where n symbolizes the number of frequencies included in the approximation, k stands for a particular frequency, and t denotes the number of observations. obviously, the process is linear and the conventional non-stationary testing methodologies are appropriate, if α1 =β1 =···=αn =βn = 0. on the other hand, at least one fourier frequency must be present in the data-generating process, if there is a break or nonlinear trend. since it is not possible to use a large value of n in a regression framework and the use of many frequency components can lead to an overfitting problem, enders and lee (2012) try to choose the proper frequencies to include in equation (10), instead of positing the specific form of α(t). supposing esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 87 that they use only a single frequency k and consider the testing regression as follows: ∆yt = ρyt−1 + c1 + c2t + c3 sin(2πkt/t)+c4 cos(2πkt/t)+et (11) let τdf_t the t -statistic for the null hypothesis ρ = 0 in equation (11). the asymptotic characteristics of the df version tests do not differ from those of the lm version of the test and they decide not to present the asymptotic distribution. the important point is that the critical values for the null hypothesis of a unit root will rely solely on the frequency (k) and the sample size (t) just like in the other version tests. nevertheless, they do not rely on the coefficients of the fourier terms or other deterministic terms. thus, ender and jones (2012) can systematize critical values benefiting from simulations. critical values of τdf _t are shown in table 1(a) in their study. if the researcher wants to specify the value of k, the test could be carried out directly benefiting from these critical values. if the value of k is approximated, the test for a break could be implemented as follows: at the first step: they estimate equation (11) for all integer values of k such that 1 ≤ k ≤ 5. it is seen that the regression with the smallest sum of squared residuals (ssr) yields 𝑘𝑘�. if the residuals show serial correlation, augment (11) with lagged values of ∆yt . at the second step: they indicate that pretesting for nonlinearities could be conducted. for this purpose, they use the usual f-test for the null hypothesis: c3 = c4 = 0. when the unit-root null is imposed on the data-generating process (dgp), the distribution of the f-statistic is nonstandard. thus, they can use the critical values of f�𝑘𝑘�� shown in table 1(a) which was stated in their study. accordingly, if the sample value of f is less than the critical value, the null hypothesis of a linear trend cannot be rejected. under this circumstance, they suggest performing the usual linear augmented dickey-fuller test. table 4. adf and fourier adf unit root test results (t=85). variables frequency minssr fourier adf test-statistic adf test-statistic fourier adf f-statistic fsci 3 5895.048 -3.87 -3.23 3.24 rsci 2 954.5519 -3.55 -3.54 1.61 pmi 1 5905.716 -6.24 -1.82 9.42 inf 5 790.0850 3.82 3.05 3.01 difinf 1 773.5279 -6.72 -2.86 2.55 logbist 1 0.250497 -3.91 -2.72 3.77 eur 5 3.235448 1.33 1.18 1.92 difeur 5 2.979959 -6.39 -8.25 1.01 usd 5 2.384208 1.62 1.21 2.47 difusd 5 2.129045 -5.94 -4.24 1.10 intrate 5 140.5274 -0.06 -0.16 2.59 difintrate 5 115.8366 -5.52 -5.90 1.24 notes: k denotes the optimal frequency value with the smallest sum of squared residuals (ssr). critical values for fourier adf unit root test for k=1 and t=85; -4,42, -3,81, -3,49, k=2 and t=85; -3,97, -3,27, 2,91, k=3 and t=85; -3,77, -3,07, -2,71, k=5 and t=85; -3,58, -2,93, -2,60 at 1%, 5% and 10%, respectively. critical values for fourier adf f statistics for t=85; 10,35, 7,58, 6,35 at 1%, 5% and 10%, respectively. because we employ the model used without a linear trend, we take into consideration the critical values shown in table 1(b) in the study of enders and lee (2012). critical values for adf unit root test are -3,51, -2,89 and -2,58 at 1%, 5% and 10%, respectively. esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 88 as seen in table 4, fsci and rsci series seem stationary because the t-statistics are greater than the critical values stated in the study of enders and lee (2012). on the other hand, f-test results which are used to test the significance of the trigonometric terms, do not confirm that these series are stationary because it is seen that trigonometric terms for these variables are not significant when checked the f-statistics with those shown in the same study. so after employing the standard adf unit root test, we can confirm that fsci and rsci series are stationary. the pmi series is stationary because both the fourier adf t-statistics and fstatistics are significant while the logbist series is stationary after checking both fourier adf and standard adf t-statistics because trigonometric terms are not significant. as to the inf, eur, usd and intrate series, it is seen that the variables have unit-root at its level, they become stationary after their first difference. according to the f-test results which are used to test the significance of the trigonometric terms, it seems that trigonometric terms for these variables are not significant, the adf test-statistic values again are taken into consideration, and it is seen that they all become stationary after their first difference. in the second stage, we employ fourier toda yamamoto causality test proposed by nazlioglu et al. (2016) in order to investigate the causal linkages from fsci and rsci to the macro-financial indicators. since the linkages between the variables have been exposed to gradual shifts, and linear specifications are generally improper to capture the relationships, econometric examinations are not usually direct and simple. as a result, traditional procedures which look for sudden shifts become insufficient in capturing gradually emerging structural changes. nazlioglu et al. (2016) modify the toda-yamamoto (1995) granger causality approach by implanting a fourier approximation to be able to explain gradual or smooth structural shifts. there is no need for a prior knowledge concerning the number, dates and form of breaks when used the fourier approximation. their study based on the analysis proposed by ender and jones (2016) in which a fourier approximation is employed by using a limited number of low-frequency components in an effort to clarify the determination of the form of breaks and estimation of the number and dates of shifts in a var framework. the causality approach which nazlioglu et al. (2016) utilize to examine the causal relationship between oil prices and real estate investment trusts (reits) in their study is based on considering a var(p + d) model in which p denotes lag length and d shows the maximum integration degree of the variables. the var(p+ d) model can be written as: 𝑦𝑦𝑡𝑡 = 𝛼𝛼 + 𝛽𝛽1𝑦𝑦𝑡𝑡−1 + ⋯+ 𝛽𝛽𝑝𝑝+𝑑𝑑𝑦𝑦𝑡𝑡−(𝑝𝑝+𝑑𝑑) + 𝜖𝜖𝑡𝑡 (12) where yt consists of k endogenous variables, α is a vector of intercept terms, β are coefficient matrices and ϵt are white noise residuals. the null hypothesis of granger non-causality is based on zero restriction on first p parameters (h0:β1=…=βp=0) of the kth element of yt. wald statistic for this hypothesis does have an asymptotic χ2 distribution with p degrees of freedom. here, yt in equation (12) is presumed not to have any structural break by the presumption that the esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 89 intercept terms α are constant over a period of time. in a var specification, since a break in one variable might cause shifts in the other variables, managing structural breaks and specifying the original source of breaks become hard. the standard granger causality test has a plausible size and power attributes if the shifts are sharp and the test functions much better if the shifts are gradual. it is also sensitive to the unit root and co-integration characteristics of the var model and makes testing unit root and co-integration for causal inferences necessary because wald test does have a non-standard distribution when the variables in var model are integrated or co-integrated, and it also depends on nuisance parameters. by solving such problems, the toda-yamamoto approach seems strong to unit root and co-integration characteristics of the var system. by expanding the toda–yamamoto framework with gradual structural breaks by embedding a fourier approximation, nazlioglu et al. (2016) introduce a new and simple methodology to capture breaks in granger causality analysis. in order to take into account the structural shifts, they ease the assumption of the intercept terms α being constant over time and modify the var model in equation (12) as: 𝑦𝑦𝑡𝑡 = 𝛼𝛼(𝛽𝛽) + 𝛽𝛽1𝑦𝑦𝑡𝑡−1 + ⋯+ 𝛽𝛽𝑝𝑝+𝑑𝑑𝑦𝑦𝑡𝑡−(𝑝𝑝+𝑑𝑑) + 𝜖𝜖𝑡𝑡 (13) where the intercept terms α(t) imply the functions of time and indicate any structural shifts in yt. to be able to capture structural shifts as a gradual process with an unknown date, number and form of breaks, the fourier expansion is specified by 𝛼𝛼(𝛽𝛽) = 𝛼𝛼0 + ∑ 𝛾𝛾1𝑘𝑘𝑛𝑛 𝑘𝑘=1 sin �2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 � + ∑ 𝛾𝛾2𝑘𝑘𝑛𝑛 𝑘𝑘=1 cos (2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 ), (14) where n shows the number of frequencies, γ1k and γ2k mensurate the amplitude and displacement of the frequency, respectively. by the way, a large value of n is most likely to be connected with a stochastic parameter variation, leading to a decrease in freedom and creating an over-fitting problem. as stated by becker et al. (2006), a single fourier frequency produces a variety of shifts in deterministic components without taking into account date, number, and form of breaks. hence, nazlioglu et al. (2016) use a single frequency component and define α(t) as: 𝛼𝛼(𝛽𝛽) = 𝛼𝛼0 + 𝛾𝛾1 sin �2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 � + 𝛾𝛾2cos (2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 ) (15) where k indicates the frequency for the approximation. by replacing equation (15) in equation (13), they acquire the equation as: 𝑦𝑦𝑡𝑡 = 𝛼𝛼0 + 𝛾𝛾1 sin �2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 � + 𝛾𝛾2 cos �2𝜋𝜋𝑘𝑘𝑡𝑡 𝑇𝑇 � + 𝛽𝛽1𝑦𝑦𝑡𝑡−1 + ⋯+ 𝛽𝛽𝑝𝑝+𝑑𝑑𝑦𝑦𝑡𝑡−(𝑝𝑝+𝑑𝑑) + 𝜖𝜖𝑡𝑡 (16) esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 90 in this identification, examining the null hypothesis of granger non-causality is the same as it is in equation (12), and the hypothesis could be tested by using the wald statistic. when the fourier terms are used to check for breaks, the causality results vary from those documented before in several ways, and stronger relations, as well as richer sets of interactions between the variables, are found. table 5. fourier toda yamamato causality test results. relationship frequency wald-stat asymptotic* p-value bootstrap p-value fsci→ bist 2 17.122 0.017** 0.017 fsci→ intrate 3 2.194 0.334 0.336 fsci→ usd 3 0.911 0.823 0.828 fsci→ eur 3 2.070 0.558 0.563 rsci→ pmi 1 0.338 0.561 0.575 rsci→ inf 1 36.676 0.000*** 0.003 rsci→ intrate 1 11.609 0.312 0.340 notes: → denotes to causality. optimal k (frequency) and p (lag) are determined by akaike information criterion. bootstrap p-values are based on 1000 replications. ***, **, and * denote %1, %5, and %10 levels of statistical significance, respectively. because n>50 in this study, we will take asymptotic p-value in comparison. table 5 shows the results of fourier toda-yamamoto causality test. according to the results, when checked the asymptotic p-value regarding the variables, there seem to be causality relationships both from fsci to bist and rsci to inf. it means that both the changes fsci and rsci have impacts on bist and inf, respectively. on the other hand, we could not find any causality from fsci to intrate, usd and eur or from rsci to pmi, inf and intrate. the findings supporting the impact of fsci on bist and the impact of rsci on inf, signal that confidence indicators are associated with changes in macro-financial indicators such as inflation and stock market index. 5. conclusions and policy implications after the economic crisis of 2000-01, turkey launched a process of a strong recovery in the financial system, particularly in the banking sector. confidence has played a key role in the recovery of the macro-financial outlook of turkey during the post-crisis period. in spite of the financial fluctuations in u.s. and euro-area in recent years, the sound financial system has been the main cause of the good economic performance of turkey. parallel to this, the real sector has recorded a good performance as a result of positive expectations concerning the turkish economy and financial system until the last few-year-period in which there has been felt deterioration in confidence indicators. this study aims to examine the impact of confidence indicators on explaining some macro-financial variables, in other words, to assess the usefulness of fsci and rsci in forecasting economic and financial indicators in turkey. in this context, analyzing the causality relationship from confidence indicators on some macroeconomic and financial variables in turkey for the period of 2012:05-2019:05, we look for the answer of this question: do real sector and financial services confidence indexes have esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 91 impact on industrial production index, inflation rate, stock market index, interest rates and foreign exchange rates. in answering this question, after testing the unit root properties of the series by using narayan and popp (2012) and enders and lee (2012) fourier adf unit root tests, the causality relationships from the confidence indexes to these variables are investigated by employing fourier toda yamamoto causality test proposed by nazlioglu et al. (2016). the results support the impact of confidence indexes on macro-financial indicators as the stock market index and inflation, indicating that confidence indexes have explanatory power on the macro-financial outlook of turkey in the period of 2012:05-2019:05. our findings are in line confidence with most of the studies in the literature, which indicate that the confidence indexes do have impacts on macro-financial outlook. therefore, it could be said that fsci and rsci can be used to predict some of the macro-financial indicators like stock market index and inflation rate in turkey. as the literature on the impact of fsci and rsci on indicators like stock market index, interest rates and foreign exchange rates in turkey is quite limited, our study is expected to be quite useful. it is seen that the studies dealing with this issue is not enough because of the data limitation, but as more data belonged to these variables, the opportunities for comprehensive analyses could increase in the future. to emphasize, there is a strong relationship betwen financial stability and confidence in an economy. in the countries including turkey, which implement floating exchange rate regimes, capital inflows accelerate with the increasing confidence to the country and policies carried out, supporting economic growth. in case the fiscal discipline is ignored, and also the growth is based on external borrowing and imports, a deterioration in confidence indicators will lead to sudden capital outflows and severe fluctuations in exchange rates, causing uncertainty increase in both the real economy and financial markets. as a result of decreasing confidence, capital outflows will accelerate even more, making the financial vulnerability increase to a great extent. in this regard, to achieve and maintain financial stability is of great importance since it is a sign of sustainable confidence in both the real economy and the financial system. as emphasized in a monthly bulletin of european central bank (january 2013), confidence indicators gain importance and predictive power during the period of financial stress. accordingly, the link between confidence indicators and economic performance is not straightforward. in the normal periods, changes in confidence might reflect misperceptions regarding economic activity or basically be following real developments. therefore, confidence indices might have weak leading properties. nevertheless, a significant deterioration in confidence indices helps us predict future economic developments since confidence indices reflect critical changes in economic agents’ behaviour in such circumstances. given the fact that the tension increases gradually in turkey recently, it is obvious that the number of studies will rise in the future. overall, this study, in which we try to analyze whether confidence indicators have impacts on macroeconomic and financial performance in turkey, partly confirms the usefulness of fsci and rsci to forecast changes in macro-financial indicators like inflation and stock market index. in future studies, more detailed analyses could be carried out to reveal the esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 92 level of these impacts by employing other recent empirical approaches. furthermore, by using a wide range of data and increasing the macroeconomic or financial variables, the scope of the studies could be expanded. references acharya, s., benhabib, j., and huo, z. (2017). the anatomy of sentiment-driven fluctuations, nber working paper series, no: 23136, 1-39. https://doi.org/10.3386/w23136 afshar t., arabian, g., and zomorrodian, r. (2007). stock return, consumer confidence, purchasing manager’s index and economic fluctuations, journal of business & economics research, 5(8), 97-106. https://doi.org/10.19030/jber.v5i8.2575 akerlof, g. a., and shiller, r. j. (2009). animal spirits: how human psychology drives the economy, and why it matters for global capitalism, princeton university press, isbn 978-0691-14233-3. arisoy, i. (2012). türkiye ekonomisinde i̇ktisadi güven endeksleri ve seçilmiş makro değişkenler arasındaki i̇lişkilerin var analizi, maliye dergisi, 16, 304-315. bacchetta, p., and van wincoop, e. (2013). sudden spikes in global risk, journal of international economics, 89(2), 511-521. https://doi.org/10.1016/j.jinteco.2012.07.006 bachmann, r., elstner, s., and sims, e. r. (2013). uncertainty and economic activity: evidence from business survey data. american economic journal: macroeconomics, 5 (2). 217-249. https://doi.org/10.1257/mac.5.2.217 barsky, r. b., and sims, e. r. (2012). information, animal spirits, and the meaning of innovations in consumer confidence, the american economic review, 102(4), 1343-1377. https://doi.org/10.1257/aer.102.4.1343 baumohl, b. (2012). the secrets of economic indicators: hidden clues to future economic trends and investment opportunities, new jersey: financial times press, isbn-9780132932073. beaudry, p., and portier, f. (2006). stock prices, news and economic fluctuations. american economic review, 96(4), 1293-1307. https://doi.org/10.1257/aer.96.4.1293 beaudry, p., and portier, f. (2014). news-driven business cycles: insights and challenges, journal of economic literature, 52(4), 993-1074. https://doi.org/10.1257/jel.52.4.993 blanchard, o. j., l’huillier, j. p., and lorenzoni, g. (2013). news, noise, and fluctuations: an empirical exploration, american economic review, 103(7), 3045-3070. https://doi.org/10.1257/aer.103.7.3045 cass, d. & shell, k. (1983). do sunspots matter?, journal of political economy, 91(2), 193-227. https://doi.org/10.1086/261139 central bank of the republic of turkey (2019a). electronic data delivery system (evds), available at: https://evds2.tcmb.gov.tr/index.php?/evds/seriemarket. access date: 15.06.2019. central bank of the republic of turkey (2019b). financial services statistics and financial services confidence index. available at: https://www.tcmb.gov.tr/wps/wcm /connect/9d2d0544-b3bd-4641-aed9-cf82621a905b/fss-reportint.pdf?mod=ajperes&cacheid=rootworkspace-9d2d0544-b3bd-4641-aed9cf82621a905b-mcbazhv. access date: 13.05.2019. celik, s., and ozerkek, y. (2009). panel cointegration analysis of consumer confidence and personal consumption in the european union, journal of business economics and management, 10 (2), 161-168. https://doi.org/10.3846/1611-1699.2009.10.161-168 de grauwe, p., and ji, y. (2016). animal spirits and the international transmission of business cycles, cesifo working paper series, no: 5810. enders, w., and lee, j. (2012). the flexible fourier form and the dickey-fuller type unit root tests, economics letters, 117, 196-199. https://doi.org/10.1016/j.econlet.2012.04.081 https://doi.org/10.3386/w23136 https://doi.org/10.19030/jber.v5i8.2575 https://doi.org/10.1016/j.jinteco.2012.07.006 https://doi.org/10.1257/mac.5.2.217 https://doi.org/10.1257/aer.102.4.1343 https://doi.org/10.1257/aer.96.4.1293 https://doi.org/10.1257/jel.52.4.993 https://doi.org/10.1257/aer.103.7.3045 https://doi.org/10.1086/261139 https://evds2.tcmb.gov.tr/index.php?/evds/seriemarket https://www.tcmb.gov.tr/wps/wcm/connect/9d2d0544-b3bd-4641-aed9-cf82621a905b/fss-report-int.pdf?mod=ajperes&cacheid=rootworkspace-9d2d0544-b3bd-4641-aed9-cf82621a905b-mcbazhv https://www.tcmb.gov.tr/wps/wcm/connect/9d2d0544-b3bd-4641-aed9-cf82621a905b/fss-report-int.pdf?mod=ajperes&cacheid=rootworkspace-9d2d0544-b3bd-4641-aed9-cf82621a905b-mcbazhv https://www.tcmb.gov.tr/wps/wcm/connect/9d2d0544-b3bd-4641-aed9-cf82621a905b/fss-report-int.pdf?mod=ajperes&cacheid=rootworkspace-9d2d0544-b3bd-4641-aed9-cf82621a905b-mcbazhv https://www.tcmb.gov.tr/wps/wcm/connect/9d2d0544-b3bd-4641-aed9-cf82621a905b/fss-report-int.pdf?mod=ajperes&cacheid=rootworkspace-9d2d0544-b3bd-4641-aed9-cf82621a905b-mcbazhv https://doi.org/10.3846/1611-1699.2009.10.161-168 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2764806## https://doi.org/10.1016/j.econlet.2012.04.081 esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 93 enders, w., and lee, j. (2016). grain prices, oil prices, and multiple smooth breaks in a var, studies in nonlinear dynamics & econometrics, 20(4), 399-419. https://doi.org/10.1515/snde-2014-0101 european central bank (2013), monthly bulletin, january 2013), available at: https://www.ecb.europa.eu/pub/pdf/other/art1_mb201301en_pp45-58en.pdf. access date: 21.09.2019 farmer, r. e. (2013). animal spirits, financial crises and persistent unemployment, the economic journal, 123(568), 317-340. https://doi.org/10.1111/ecoj.12028 fitch (2019). turkey march 2019 global economic outlook forecast. available at: https://www.fitchratings.com/site/re/10067275. access date: 15.06.2019 gagea, m. (2012). the contribution of business confidence indicators in short-term forecasting of economic development, annals of faculty of economics, university of oradea, 1(1), 617-623. iskenderoglu, o., and akdag, s. (2017). investigation of the validity of financial services confidence index: the case of turkey, uluslararası ekonomik araştırmalar dergisi, 3(4), 625-633. jansen, w. j., and nahuis, n. (2003). the stock market and consumer confidence: european evidence, economics letters, 79(1), 89-98. https://doi.org/10.1016/s0165-1765(02)00292-6 kandir, s. y. (2006). tüketici güveni ve hisse senedi getirileri i̇lişkisi: imkb mali sektör şirketleri üzerinde bir uygulama, çukurova üniversitesi sosyal bilimler enstitüsü dergisi, 15(2), 217-230. karasoy, h. g (2015). consumer confidence indices and financial volatility, cbrt research notes in economics, no: 2015-16. karasoy, h. g., and yunculer, c. (2015). the explanatory power and the forecast performance of consumer confidence indices for private consumption growth in turkey, cbrt working papers, no: 15/19. katona, g. (1951). psychological analysis of economic behavior, new york: mcgraw-hill press. keynes, j. m. (1936). the general theory of interest, employment and money, macmillan cambridge university press. kuzmanović, m., and sanfey, p. (2012). can consumer confidence data predict real variables? evidence from croatia, european bank for reconstruction and development, working papers, no: 151. ludvigson s. c (2004). consumer confidence and consumer spending, journal of economic perspectives, 18(2):29-50. https://doi.org/10.1257/0895330041371222 moody’s (2019). cross-sector-emerging markets: global economic stress weighing unevenly on business conditions in 2019-20, may, 2019. available at: https://www.moodys.com/login?lang=en&cy=global&returnurl=https%3a%2f%2fwww.mood ys.com%2fresearchdocumentcontentpage.aspx%3fdocid%3dpbc_1161041. access date: 15.06.2019. nazlioglu, s., gormus, a., and soytas, u. (2016). oil prices and real estate investment trusts (reits): gradual-shift causality and volatility transmission analysis, energy economics, (60), 168-175. https://doi.org/10.1016/j.eneco.2016.09.009 narayan, p. k., and popp, s. (2010). a new unit root test with two structural breaks in level and slope at unknown time, journal of applied statistics, 37, 14251438. https://doi.org/10.1080/02664760903039883 nowzohour, l., and stracca, l. (2017). more than a feeling: confidence, uncertainty and macroeconomic fluctuations, ecb working paper series, no: 2100. oecd (2019), business confidence index (bci) (indicator). accessed date: 25.05.2019. https://doi.org/10.1787/3092dc4f-en oral, e., ece, d., and hamsici, t. (2005). building up a real sector confidence index for turkey. central bank review, 1, 23-54. ozsagir, a. (2007). ekonomide güven faktörü. elektronik sosyal bilimler dergisi, 6 (20), 46-62. santero, t. and n. westerlund (1996). confidence indicators and their relationship to changes in economic activity", oecd economics department working papers, no: 170. https://doi.org/10.1515/snde-2014-0101 https://www.ecb.europa.eu/pub/pdf/other/art1_mb201301en_pp45-58en.pdf https://doi.org/10.1111/ecoj.12028 https://www.fitchratings.com/site/re/10067275 https://ideas.repec.org/a/ora/journl/v1y2012i1p617-623.html https://ideas.repec.org/a/ora/journl/v1y2012i1p617-623.html https://ideas.repec.org/s/ora/journl.html https://econpapers.repec.org/article/eeeecolet/ https://doi.org/10.1016/s0165-1765(02)00292-6 https://doi.org/10.1257/0895330041371222 https://www.moodys.com/login?lang=en&cy=global&returnurl=https%3a%2f%2fwww.moodys.com%2fresearchdocumentcontentpage.aspx%3fdocid%3dpbc_1161041 https://www.moodys.com/login?lang=en&cy=global&returnurl=https%3a%2f%2fwww.moodys.com%2fresearchdocumentcontentpage.aspx%3fdocid%3dpbc_1161041 https://doi.org/10.1016/j.eneco.2016.09.009 https://doi.org/10.1080/02664760903039883 https://doi.org/10.1787/3092dc4f-en esra n. kilci. / european journal of government and economics 9(1), june 2020, 74-94 94 souleles, n. s. (2001). consumer sentiment: its rationality and usefulness in forecasting expenditure evidence from the michigan micro data, nber working papers, no: w8410. https://doi.org/10.3386/w8410 turkstat (2019). statistical databases, available at: http://www.turkstat.gov.tr/ pretabloarama.do?metod=search&aratype=vt. access date: 13.05.2019. tusiad (2019). 2019 yılına girerken türkiye ve dünya ekonomisi, raporlar, available at: https://tusiad.org/tr/basin-bultenleri/item/10236-tusiad-2019-yilna-girerken-turkiye-ve-dunyaekonomisi-raporu. access date: 10.05.2019. utaka, a. (2003). confidence and the real economy the japanese case, applied economics 35, 337-342. https://doi.org/10.1080/00036840210135205 utaka, a. (2014). consumer confidence and the japanese economy comparison of preand post-bubble period-, economics bulletin, 34 (2), 1165-1173. van aarle, b. and kappler, m. (2012). economic sentiment shocks and fluctuations in economic activity in the euro area and the usa, intereconomics, 47(1), 44-51. https://doi.org/10.1007/s10272-012-0405-z vuchelen, j. (2004). consumer sentiment and macroeconomic forecasts, journal of economic psychology, elsevier, 25(4), 493-506. https://doi.org/10.1016/s0167-4870(03)00031-x https://doi.org/10.3386/w8410 http://www.turkstat.gov.tr/pretabloarama.do?metod=search&aratype=vt http://www.turkstat.gov.tr/pretabloarama.do?metod=search&aratype=vt https://tusiad.org/tr/basin-bultenleri/item/10236-tusiad-2019-yilna-girerken-turkiye-ve-dunya-ekonomisi-raporu https://tusiad.org/tr/basin-bultenleri/item/10236-tusiad-2019-yilna-girerken-turkiye-ve-dunya-ekonomisi-raporu https://doi.org/10.1080/00036840210135205 https://doi.org/10.1007/s10272-012-0405-z https://ideas.repec.org/a/eee/joepsy/v25y2004i4p493-506.html https://ideas.repec.org/s/eee/joepsy.html https://ideas.repec.org/s/eee/joepsy.html https://doi.org/10.1016/s0167-4870(03)00031-x number 9, issue 1, june 2020 do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey 1. introduction 2. the factors leading to deterioration in confidence indicators in turkey 3. literature review 4. empirical analysis 5. conclusions and policy implications references stock prices, uncertainty and risks: evidence from developing and advanced economies european journal of government and economics 9(3), december 2020, 265-279 european journal of government and economics issn: 2254-7088 stock prices, uncertainty and risks: evidence from developing and advanced economies mary elena sánchez-gabarre a* a university of a coruña, spain * corresponding author at: mary.e.sanchezg@udc.es abstract. this paper studies the relationship between stock prices and three types of uncertainty: economic policy uncertainty, stock market volatility, and geopolitical risks. in particular, our aim is to determine whether these forms of uncertainty play the same role in developed and developing countries. with this purpose, we take spain and brazil as representative cases. in order to provide new insights into the abovementioned relationship, a cointegration approach is applied, specifically an ardl model, using monthly data from the period january 2006-december 2019 for a series of financial and macroeconomic variables. the results obtained reveal that there is no uniform effect of uncertainty in stock markets of developing and developed countries. first, in spain, there is a high perception of uncertainty in economic policy and stock market volatility, which impact negatively in share prices, both in the short and long term. regarding brazil, the global uncertainty in the stock markets has effects on share prices, in both time horizons. by contrast, geopolitical risks do not show any significant impact on brazilian and spanish share returns. keywords. economic policy uncertainty; geopolitics; stock markets; uncertainty; volatility jel codes. c22; g12; g18 doi. https://doi.org/10.17979/ejge.2020.9.3.6999 1. introduction the main objective of this paper is to determine the relationship between uncertainty and the performance of stock market indices in the long term, to measure its effects and evaluate the possible differential impacts depending on the type of economy. according to literature, one would expect an increase in uncertainty to be associated in the long term with a decrease in stock market indices (malkiel & xu, 2006; durnev, 2010; caldara & iacoviello, 2019), in other words, the uncertainty-price ratio would be negative. therefore, the initial hypothesis we analyse is to what extent uncertainty (considered according to three perspectives that are explained afterwards) could affect stock prices. in addition, we study whether those different perspectives of uncertainty have different effects depending on the type of economy. likewise, the level of the short-term impact of an uncertainty shock on the stock market is analysed, as well as the adjustment time required to return to the previous situation in the stock markets. to study this relationship, as a novel aspect in the analysis, as far as uncertainty is concerned, we use three representative indicators at a world level, rarely taken into account together: economic policy uncertainty, stock markets volatility and geopolitical risks. mailto:mary.e.sanchezg@udc.es https://doi.org/10.17979/ejge.2020.9.3.6999 mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 266 this way, we can study the impact of this uncertainty at a global level on two economies with different features (a developed country and a developing country) and evaluate the possible differential impacts depending on the type of economy. uncertainty in the markets has often been seen as a factor affecting the functioning of the economy as a whole. among the main drivers of this uncertainty both market volatility and policymakers' decisions are usually considered (carney, 2016). there is a strong tendency towards financial instability, which has made globalisation a phenomenon highly determined by uncertainty. in order to fill the gap, this paper analyses the importance of uncertainty in the evolution of stock market indices, more specifically the possible long-term relationship between uncertainty and two relevant indices, corresponding to a representative developing country and a representative developed country, respectively: bovespa [from brazil, which has the largest capitalization of latin american stock markets (coleman, leone & medeiro, 2018; oecd, 2019)] and ibex 35, which is the main stock index of spain. thus, we can compare the results obtained for these two cases and determine whether uncertainty has different effects in both stock markets, corresponding to two different economies in terms of degree of development. this paper is organized as follows. in section 2, to contextualise the object of study, we begin by considering the relationship between uncertainty and stock markets. in section 3, we present and justify the data, variables and econometric methods used, and we also show and discuss the results of our analysis. finally, the main conclusions of this paper are summarised. 2. literature review the relationship between uncertainty and the economy is not a new issue to be studied. uncertainty has intensified in the wake of the global financial crisis, crises in the euro area and partisan political disputes. evidence of this is provided by the research of the international monetary fund (imf, 2012 and 2013) and the federal open market committee (2009), which highlight that uncertainty about fiscal, regulatory and monetary policies in the united states and europe contributed to a sharp economic decline in the years of the financial crisis in 2008 and a slowdown in the subsequent economic recovery. due to the negative impact of economic policy uncertainty on the stock markets, it is important to include it in this section. both in the us and in europe it has been found that it leads to a worsening of macroeconomic performance, in addition to effects on the volatility of share prices (baker, bloom & david, 2016). furthermore, carney (2016) includes geopolitical risk, together with the economic and political uncertainty, in an "uncertainty trinity", as these three factors could have significant adverse economic effects. in recent years, the european central bank in the economic bulletin -, the international monetary fund and the world bank in the world economic outlook have systematically highlighted and monitored these risks to the outlook posed by geopolitical uncertainties. mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 267 in this respect, authors such as erb, harvey and viskanta (1996) find a relationship between political risk, as measured by the international country risk guide, and future stock returns. it is also important to note the abnormally bullish stock market in the weeks leading up to major elections, especially those characterised by high levels of uncertainty (pantzalis, stangeland & turtle, 2000; li & born, 2006). this evidence is consistent with a positive relationship between risk premium and political uncertainty. in this sense, we must expect a positive relationship between the risk premium and their measure of economic policy based on uncertainty in an international scenario (brogaard & detzel, 2015) and relate the risk premium to political cycles (santa-clara & valkanov, 2003; belo, gala & li, 2013). in addition, some authors (bittlingmayer, 1998; voth, 2003; boutchkova, doshi, durnev & molchanov, 2012) have studied the relationship between political uncertainty and volatility in stock markets. therefore, three perspectives are taken into account together throughout this paper: economic policy uncertainty, geopolitical risk and stock market volatility. firstly, economic policy uncertainty provide additional information on the stock market effects. this economic policy uncertainty comprises the actions and decisions that the authorities of each country take within the scope of the economy. focusing explicitly on the existing literature on policy uncertainty, we find, for example, friedman (1968), rodrik (1991), higgs (1997) and hassett and metcalf (1999), who consider the detrimental economic effects of uncertainty on monetary, fiscal and regulatory policies. more recently, born and pfeifer (2014) and fernández-villaverde, guerrón-quintana, kuester and rubio-ramírez (2015) study policy uncertainty by means of dsge (dynamic stochastic general equilibrium) models, finding moderately negative effects, whereas pastor and veronesi (2012, 2013) model the theoretical links between their fluctuations, policy uncertainty and stock market volatility. accordingly, some recent papers studied the negative relationship between economic policy uncertainty and economic activity (bloom, 2009; giavazzi & mcmahon, 2012; julio & yook, 2012; bloom, 2014; leduc & liu, 2015; koijen, philipson & uhlig, 2016; scotti, 2016; basu & bundick, 2017). secondly, geopolitical risks owe its relevance to the fact that entrepreneurs, market participants and central banks have in recent years considered geopolitical risks to be key determinants of investment decisions and stock market dynamics. investors express concern about the economic impact of the various military and diplomatic conflicts around the world, even indicating increased concern in this area with regard to political and economic uncertainty (carney, 2016). the study of the influence of the geopolitical environment on the economy has not been widely analised because of the limitation resulting from the absence of an indicator of geopolitical risk that is consistent with the perception of press, public, investors and policy makers. these issues were taken into account by caldara and iacoviello (2019), who constructed a geopolitical risk index from 1985 and show that a negative geopolitical shock induces persistent declines in mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 268 investment, employment, consumer confidence and stock market price returns. the decline in equity prices is slightly larger than that in investment, suggesting that risk premia are increasing, although only to a limited extent. these results are supported by theoretical models such as those of ilut and schneider (2014) and others in which high levels of uncertainty lead to declines in employment and investment, e.g. dixit and pindyck (1994) and bloom, bond and van reenen (2007). thirdly, stock market volatility is a measure of uncertainty in the equity markets. volatility is one way of measuring this price risk, probably the most widely used. in fact, this dispersion of the prices on an asset has long been considered an indicator of risk since markowitz (1959). in this line and taking the capm model as a reference, some authors (merton, 1973; campbell, 1993, 1996; chen, 2003; ang, hodrick, xing & zhang, 2006 and 2009; malkiel & xu, 2006; chen, ghysels & wang, 2015; farago & tédongap, 2018; hollstein & prokopczuk, 2018; kaeck, 2018) argue that an increase in volatility can be interpreted as a worsening of overall investment opportunities and this predicts a negative relationship. although there is a growing body of literature regarding the adverse influence of uncertainty on stock markets, to the best of our knowledge, there is not a joint analysis of the three most studied uncertainty forms. furthermore, it is interesting to compare whether the influence of uncertainty behaves in the same way in developed and developing countries. 3. empirical analysis 3.1. data and variables in the study of the literature, it was possible to clearly see the relationship of the stock markets with macroeconomic variables, and several necessary control variables were taken into account. in this research, such variables are represented by the gross domestic product at market prices (gdp), the consumer price index (cpi), the interest rate of the 12-month national bond (ir) and the real effective exchange rate index (er) of brazil and spain respectively. in addition, uncertainty is represented by the following variables, respectively: the epu or economic policy uncertainty (baker et al., 2016), the gpr or geopolitical risk index (caldara & iacovello, 2019) and the vix or chicago board options exchange market volatility index (cboe, 2019). the choice of the epu for our study is due to the fact that, on the one hand, the macroeconomic factor increasingly seeks to anticipate investors' expectations regarding the evolution of fundamental economic variables. it is therefore not surprising that this variable associated with macroeconomic fundamentals has been the main determinant of european stock market dynamics over the last two decades (caixabank research, 2016). furthermore, although the gpr is correlated with the epu, it shows a remarkable amount of additional and independent variation that predicts a lower percentage of economic activity. in relation to existing proxy indicators of uncertainty that tend to increase during recessions, this index points to episodes that are independent of the economic cycle. and we have to take into mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 269 account the benchmark indicator in the major stock markets around the world, the vix. although it is linked to the s&p 500 index, its use is increasing due to the importance of the us index in the rest of the world's stock markets. therefore, it should be pointed out that in both global situations, it seems plausible to argue that the direction of causality ranges from geopolitical events to stock market volatility and political uncertainty. they do not all have to behave in the same way in the face of geopolitical risks. for example, the invasion of iraq by the united states in 2003 appears to cause increased economic policy uncertainty, although it does not induce financial volatility. furthermore, the three indices also show a large number of independent variations, as the gpr index does not vary much during periods of economic and financial difficulties, as can be seen in the period of the dot-com bubble (1997-2001) and during the world financial crisis in 2007, when the vix and the epu showed increases. the gpr index also does not move around the presidential elections, periods characterised by high political uncertainty. therefore, compared to the vix and the epu, the gpr index captures those events that are most likely to be exogenous to business and financial cycles, and could lead to greater financial volatility and policy uncertainty. the monthly time series used in our empirical analysis covers the period from january 2006 to december 2019 and thus consists of 156 data for each variable. this is due to the availability of consistent data for all variables. a descriptive statistic of the data, as well as their sources, is presented in table 1. it gives an initial idea of the data to be analysed. first and secondly, appear the minimum and maximum value that the variables have had throughout the period of study. in third and fourth place, respectively, are the mean and standard deviation of the variables over the 14 years observed. as we can see, the variables with the greatest standard deviations and extreme values are those of uncertainty. it is also worth mentioning the values of the interest rate for the case of spain, since in certain periods it has a very marked variation. table 1. descriptive statistics and data sources. variables max. min. mean sd source epu 57.099 -39.112 1.5366 17.192 https://www.policyuncertainty.com/ vix 90.751 -38.490 1.8067 21.725 eikon gpr 120.71 -58.516 3.4295 30.581 https://www.matteoiacoviello.com/gpr.htm brazil bovespa 24.918 -26.636 0.7316 8.7593 eikon gdp 2.6903 -2.5557 0.7255 1.2277 https://www.ibge.gov.br/ ir 12.337 -15.391 -0.6538 4.5457 eikon er 10.167 -13.901 0.0038 3.1427 https://ec.europa.eu/eurostat/data/database cpi 1.3472 -0.3339 0.4412 0.2947 https://ec.europa.eu/eurostat/data/database spain ibex 16.625 -17.033 0.0612 5.4932 eikon gdp 3.1493 -3.5299 0.1912 1.8530 https://ec.europa.eu/eurostat/data/database ir 1347.4 -1340.0 -0.6800 154.27 eikon er 2.6182 -3.1091 -0.0012 0.8454 https://ec.europa.eu/eurostat/data/database cpi 1.4106 -1.9070 0.1298 0.6060 https://ine.es/ https://www.policyuncertainty.com/ https://www.matteoiacoviello.com/gpr.htm https://www.ibge.gov.br/ https://ec.europa.eu/eurostat/data/database https://ec.europa.eu/eurostat/data/database https://ec.europa.eu/eurostat/data/database https://ec.europa.eu/eurostat/data/database https://ine.es/ mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 270 3.2. methodology as we have been pointing out, in this study, an approach based on econometric techniques of cointegration was adopted. in particular, we used the ardl (or pesaran-shin-smith) method, since as we will see in our case, it is preferable to other methodologies, such as that of engle and granger (1987) or that of gregory and hansen (1996). in this method, a bound test is first carried out to determine the direction of co-integration [which variable(s) could act as a dependent], and, once the optimum equation has been selected, the estimation of the equilibrium model is carried out in both the short and long term. since the possibilities of using the different methods of co-integration depend on the characteristics of the variables in relation to their stationarity or non-stationarity, before proceeding with the application of such methods, we must determine these characteristics, a task that we will address in the following section. 3.2.1. unit root test to try to make our analysis more robust, we opted to apply different methods, such as the dickeyfuller test (adf) and kwiatkowski-phillips-schmidt-shin test (kpss), to level values. in this case, an analysis with constant has been used and 12 lags have been selected due to the monthly periodicity of the data; it should also be noted that these tests have been carried out using akaike information criterion (aic). table 2. results of unit root tests. variables adf levels kpss levels epu i(0) i(0) vix i(0) i(0) gpr i(0) i(0) brazil bovespa i(0) i(0) gdp i(1) i(1) ir i(0) i(0) er i(0) i(0) cpi i(1) i(0) spain ibex i(0) i(0) gdp i(1) i(0) ir i(0) i(0) er i(0) i(0) cpi i(1) i(0) considering all the information provided by the unit root tests (table 2), it can be seen that the results do not coincide optimally between both tests. the only non-stationary variable in all the tests would be safely the gdp in the brazil case. therefore, we see contradictory outcomes with respect to cpi (brazil and spain) and gdp (spain). when carrying out the co-integration analysis, these results should be taken into account, as different methods could be applied depending on the stationary properties of our variables. the ardl (autoregressive distributed delays) method, also called pss (pesaran-shinsmith) test (pesaran & shin, 1999; pesaran, shin & smith, 2001), differs from other methods in mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 271 that it allows a more in-depth study, in a less restrictive way. for this reason, in the present study, an empirical analysis will be carried out applying the ardl method, among other reasons because the engle-granger and gregory-hansen methods, require that to carry out the co-integration analysis absolutely all the variables considered (dependent and regressors) are integrated of order 1, i(1), that is, they have a unitary root. 3.2.2. optimal number of lags before applying the ardl method, the number of lags with which to work must be established. a priori, according to pesaran and pesaran (2009), as these are monthly data, the recommended number of lags would be 12, which for our model would imply that the number of regressions required would be no less than 815.73 million by each country, which would be unmanageable in practice. therefore, we decided to use the procedure followed by khan and khan (2018), in order to know the number of optimum lags. according to this procedure, for this case the appropriate number of lags would be 12 according to aic and r2 criteria, whereas according to fpe (final prediction error) and hq (hannan-quinn information criterion) criteria this number would be 1. given that, as we have seen above, in our model it is very complicated in practice to work with 12 lags, according to these results we opted to establish a maximum of 1 lag. 3.2.3. bound test as the unit-root analysis was already carried out previously, we can now continue with the next phase of the estimation of the ardl models, which is the bound test by means of an f-statistic. this test analyses the causality of the variables, that is, whether and how (dependent/explanatory variable) each of the variables considered should be included in the model. its null hypothesis is the non-existence of cointegration. we must point out that when analysing the results of the corresponding statistics, there are problems due to their non-standard distribution, a situation that toda and yamamoto (1995) have tried to justify and for which pesaran et al. (2001) indicate a special method to carry out the test with two theoretical values, a lower limit (li) and an upper limit (ls). both limits must be sought in the tables provided by pesaran and pesaran (2009, p. 300), the search criteria being the number of variables of the model to be studied (in our case, k=8: bovespa or ibex, epu, vix, gpr, gdp, ir, er and cpi), the characteristics of the model (intercept, trend) and the significance level we want to apply. in addition to contrasting the f-statistic, whose null hypothesis would indicate the nonexistence of joint significance of the first lag of the variables in levels used in the analysis, a complementary test could be made in the event that the fstatistic does not give conclusive results: a t-test, which would contrast the individual significance of the first lag of the variables. since we are contrasting whether or not there is a long-term relationship between the variables, it must be taken into account that if we reject the null hypothesis, it would imply the existence of co-integration. mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 272 table 3. pss test. dependent variable f-test t-test regressors bovespa 17.550 140.403 epu, vix, gpr, gdp, ir, er, cpi significance level i(0) i(1) i(0) i(1) 5% 2.3907 3.6049 19.1256 28.8395 10% 2.0812 3.2004 16.6500 25.6036 dependent variable f-test t-test regressors ibex 18.732 149.855 epu, vix, gpr, gdp, ir, er, cpi significance level i(0) i(1) i(0) i(1) 5% 2.3907 3.6049 19.1256 28.8395 10% 2.0812 3.2004 16.6500 25.6036 the result of the bound test for our models can be seen in table 3, where the bovespa and ibex indexes are the dependent variable (the regressors of the model being the variables epu, vix, gpr, gdp, ir, er and cpi). the critical values of the intervals in the contrast of f and t come from the tables of pesaran et al. (2001) for unrestricted constant and without trend (k = 8). in addition, it should be noted that the results reflected in table 3 point out that there is no evidence against co-integration with a significance level of 5%, so we can proceed to the analysis of the ardl models (castellanos-garcía, pérez-díaz-del-río & sánchez-santos, 2014). 3.2.4. model estimation once having applied the bound test, in which we have contrasted the relationship of our variables over time, the estimation of the unrestricted error correction model will be carried out. in this way, the ardl model for the case of brazil, could be expressed as follows: 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝑡𝑡 = 𝛼𝛼0 + �𝜔𝜔𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝐵𝑡𝑡−𝑖𝑖 + �𝛿𝛿𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝐵𝐵𝐸𝐸𝑡𝑡−𝑖𝑖 + �𝛽𝛽𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝑉𝑉𝑉𝑉𝑡𝑡−𝑖𝑖 + �𝜂𝜂𝑖𝑖 1 𝑖𝑖=0 𝐺𝐺𝐵𝐵𝐺𝐺𝑡𝑡−𝑖𝑖 + �𝛾𝛾𝑖𝑖 1 𝑖𝑖=0 𝐺𝐺𝐺𝐺𝐵𝐵𝑡𝑡−𝑖𝑖 + �𝜓𝜓𝑖𝑖 1 𝑖𝑖=0 𝑉𝑉𝐺𝐺𝑡𝑡−𝑖𝑖 + �𝜑𝜑𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝐺𝐺𝑡𝑡−𝑖𝑖 + �𝜛𝜛𝑖𝑖 1 𝑖𝑖=0 𝐶𝐶𝐵𝐵𝑉𝑉𝑡𝑡−𝑖𝑖 + 𝜀𝜀𝑡𝑡 where the variables have already been defined, εt would correspond to the random disturbance, α0 to the independent term of the equation and ωi, δi, βi, ηi, γi, ψi, φi and ϖi to the regressor coefficients. once the estimation of the band test is done, we obtain the estimation of the ardl method, being the optimal model for the case of brazil [ardl(1, 0, 0, 1, 1, 0, 0)]. we will replicate the model for the case of spain, which could be expressed as follows: 𝑉𝑉𝐵𝐵𝐵𝐵𝑉𝑉𝑡𝑡 = 𝛼𝛼0 + �𝜔𝜔𝑖𝑖 1 𝑖𝑖=0 𝑉𝑉𝐵𝐵𝐵𝐵𝑉𝑉𝑡𝑡−𝑖𝑖 + �𝛿𝛿𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝐵𝐵𝐸𝐸𝑡𝑡−𝑖𝑖 + �𝛽𝛽𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝑉𝑉𝑉𝑉𝑡𝑡−𝑖𝑖 + �η𝑖𝑖 1 𝑖𝑖=0 𝐺𝐺𝐵𝐵𝐺𝐺𝑡𝑡−𝑖𝑖 + �γ𝑖𝑖 1 𝑖𝑖=0 𝐺𝐺𝐺𝐺𝐵𝐵𝑡𝑡−𝑖𝑖 + �ψ𝑖𝑖 1 𝑖𝑖=0 𝑉𝑉𝐺𝐺𝑡𝑡−𝑖𝑖 + �φ𝑖𝑖 1 𝑖𝑖=0 𝐵𝐵𝐺𝐺𝑡𝑡−𝑖𝑖 + �ϖ𝑖𝑖 1 𝑖𝑖=0 𝐶𝐶𝐵𝐵𝑉𝑉𝑡𝑡−𝑖𝑖 + 𝜀𝜀𝑡𝑡 mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 273 where all variables have been previously defined: εt would correspond to the random disturbance, α0 to the independent term of the equation and ωi, δi, βi, ηi, γi, ψi, φi and ϖi to the regressor coefficients. knowing the optimal ardl model, it is carried out a diagnostic of autocorrelation, adequate specification, normality and heteroscedasticity of the residuals is carried out below, using the akaike criterion (aic). once the different tests of the ardl model have been calculated, there are two versions (lm and f), the first being suitable for large samples, as is the case of this study. the results are shown in table 4 below. table 4. results of the different tests of the ardl model. statistical tests lm version f version brazil lagrange test chsq (12) = 10.978 [.531] f(12, 141) = .84296 [.606] reset test chsq (1) = 3.6438 [.056] f(1, 152) = 3.4539 [.065] heteroscedasticity test chsq (1) = 0.4605 [.497] f(1, 162) = .45618 [.500] normality test chsq (2) = 1.4133 [.493] not applicable spain lagrange test chsq (12) = 10.056 [.611] f(12, 141) = 0.7676 [.683] reset test chsq (1) = 1.0724 [.300] f(1, 152) = 1.0005 [.319] heteroscedasticity test chsq (1) = 0.4724 [.492] f(1, 162) = 0.4680 [.495] normality test chsq (2) = 3.3103 [.191] not applicable note. in brackets, p-values. in the lagrange test, the null hypothesis that the residuals are not correlated is contrasted. in this case, it is observed that, at the 5% significance level, h0 is not rejected so both models would be valid from the point of view of the existence of autocorrelation, that is, there would be no autocorrelation. ramsey's test contrasts the null hypothesis that the model is well specified. here it turns out that, at the 5% significance level, h0 is not rejected, which would indicate that both models are well specified. the heteroscedasticity test contrasts the null hypothesis of homoscedasticity; in this case, h0 is not rejected, so both models are homoscedastic at the 5% significance level. going on to develop the study of co-integration itself, below we estimate, on the one hand, by ordinary least squares (ols), the long-term model in which the variables are included in levels. as its own name points out, this model indicates us the equilibrium relationship in the long term between the variables. on the other hand, we estimate, also by ols, an error correction model (ecm), now using the first differences of the variables; this model represents the dynamics of the short-term relationships between the variables studied. • long-term model the main issue studied in this paper is the co-integration (that is, the long-term relationship) between the uncertainty variables and the returns of the quotations of both stock market indexes. the long-term model coefficients obtained for each of the estimates are shown in table 5. mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 274 table 5. long-term ardl model coefficients. regressor coefficient t-statistic p-value brazil c 2.2630 2.4744 0.014 epu -0.0101 -0.4103 0.682 vix -0.0782 -4.0090 0.000 gpr -0.0059 -0.2689 0.788 gdp -0.3375 -0.8521 0.396 ir -0.0947 -0.9916 0.323 er 1.5795 11.6093 0.000 cpi -2.5474 -1.7327 0.085 spain c 0.4929 1.1792 0.240 epu -0.0851 -3.6358 0.000 vix -0.1312 -6.1142 0.000 gpr 0.0328 1.6703 0.097 gdp -0.0906 -0.3615 0.718 ir 0.0004 0.1594 0.874 er 0.3983 0.6880 0.492 cpi -1.5811 -1.2630 0.209 table 5 shows that the global uncertainty generated in the stock markets (vix) has a negative influence on both economies, but this influence would have a greater weight in the second country. specifically, in the face of a unitary increase in such uncertainty, in the long term, such yields would decrease by 0.0782 units in the case of brazil, as opposed to the 0.1312 decrease in the yields of the spanish index. uncertainty in economic policy would have a long-term negative influence on ibex 35 returns, whereas in the case of bovespa it would not be significant. in the case of spain, specifically, if there were a unitary increase in this economic policy uncertainty, in the long term, these yields would decrease by 0.0851 units. it should be noted that volatility in the stock markets would have a greater long-term effect on the national index than the uncertainty derived from economic policy. in contrast, the third indicator of uncertainty, the gpr, is not at all significant in either case. with respect to the control variables, at 5% only the er is significant in the case of brazil, with a positive sign. specifically, in the face of a unit increase in the brazilian real effective exchange rate, such yields would increase by 1.5795 units in the case of brazil. this currency appreciation makes investors in international stock markets very aware of the currency in which they will be investing and how it is expected to behave over the investment horizon. • short-term model finally, the results of the error correction model (ecm) are shown to determine the short-term relationship of the variables included in the analysis. this ecm allows to analyse the impact of a shock on the model variables as well as the adjustment time needed to return to the initial equilibrium situation. first, it should be verified that in the estimation results of this model the error correction coefficient (which shows the speed of adjustment toward equilibrium after the short-term impact) is statistically significant and is negative. as we can see in table 6, both requirements are met, and the speed of adjustment is very fast in the both cases. mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 275 table 6. short-term ardl model coefficients (ecm). regressor coefficient t-statistic probability brazil depu -0.0121 0,0247 0.682 dvix -0.0934 0,0205 0.000 dgpr 0.0244 0,0132 0.137 dgdp 0.7045 0,4903 0.161 dir -0.1131 0.1134 0.319 der 1.8870 0.1804 0.000 dcpi -3.0434 1,7528 0.087 ecm(-1) -1.1947 -19.278 0.000 spain depu -0.0809 -3.5634 0.000 dvix -0.1248 -7.3291 0.000 dgpr 0.0165 1.3705 0.172 dgdp 0.2819 0.8464 0.399 dir 0.0004 0.1595 0.874 der 0.3787 0.6869 0.493 dcpi -1.5033 -1.2853 0.201 ecm(-1) -0.9508 -13.945 0.000 observing the results for the uncertainty variables, we can indicate that, as in the previous analysis, the vix has a significantly negative relationship with the returns of bovespa and ibex quotations, being higher in the latter. the epu is significant for the spanish case and has a negative sign. thirdly, the geopolitical risk would not be significant, so we could not indicate a clear inverse relationship between this uncertainty and stock market prices. the variables of uncertainty considered are therefore inversely related in the short term to stock market returns, i.e., increases in uncertainty from an economic policy and financial market point of view are associated with decreases in the quotations of the brazilian and spanish reference indices, bovespa and ibex 35. 4. conclusions in this paper, we explore the relevance of economic policy, geopolitical decisions and stock markets volatility in the stock markets returns, comparing the results obtained for a developed and developing country. the main conclusions of this paper can be summarized as follows. in the long term, we conclude that uncertainty would cause decreases in the quotations of bovespa (brazilian index) and ibex 35 (spanish index). more specifically, uncertainty in economic policy would negatively influence stock returns in the long term only for the case of the spanish index. moreover, the uncertainty stemming from the volatility of the stock markets exerts a negative influence on both stock market indexes. therefore, in the long term, when faced with an increase in the vix, the returns on the bovespa and ibex 35 share prices decrease. furthermore, it should be noted that volatility in the stock markets would have a greater long-term effect on the spanish national index than on the brazilian index. our findings also show that for the spanish case, the uncertainty derived from economic policy is less than that produced by fluctuations in the financial markets. mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 276 finally, the third indicator taken into account in terms of measuring uncertainty in the global geopolitical sphere (gpr) is not at all significant in any case. this situation could be due to the low long-term influence of events that generate global geopolitical risks on these stock market indexes, since the countries in which they are listed would not be involved in such events. however, this risk could cause relevant shocks in very short periods, where the monthly periodicity of our data implies that its influence cannot be appreciated. in the short-term, we have obtained empirical evidence suggesting the existence of a significant relationship between economic policy uncertainty and volatility in the financial equity markets with respect to the ibex 35 and between financial market volatility with respect to bovespa stock prices. in other words, they have a negative relationship in the short term, so increases in uncertainty would translate into decreases in stock price returns, with the speed of adjustment of the uncertainty shock to return to the initial equilibrium situation within a period of approximately one month. these results may be useful for investors when analysing their financial planning, as they must adopt optimal investment strategies that take into account these uncertainties, since their capital would be affected by fluctuations in these areas at a global level. in addition, they must take into account in a complementary manner that in developed countries, this global uncertainty affects to a greater extent than in emerging or developing countries. it should be noted that policymakers must take into account their decisions in the field of economic policy, since in countries like spain, these decisions will have an effect on stock market performance over time. references ang, a., hodrick, r.j., xing, y. and zhang, x. (2006). the cross-section of volatility and expected returns. journal of finance 61(1), 259-299. https://doi.org/10.1111/j.15406261.2006.00836.x ang, a., hodrick, r.j., xing, y. and zhang, x. (2009). high idiosyncratic volatility and low returns: international and further u.s. evidence. journal of financial economics 91(1), 1-23. https://doi.org/10.1016/j.jfineco.2007.12.005 baker, s.r., bloom, n. and davis, s.j. (2016). measuring economic policy uncertainty. the quarterly journal of economics 131(4), 1593-1636. https://doi.org/10.1093/qje/qjw024 basu, s. and bundick, b. (2017). uncertainty shocks in a model of effective demand. econometrica 85(3), 937-958. https://doi.org/10.3982/ecta13960 belo, f., gala, v.d. and li, j. (2013). government spending, political cycles, and the cross section of stock returns. journal of financial economics 107(2), 305-324. https://doi.org/10.1016/j.jfineco.2012.08.016 bittlingmayer, g. (1998). output, stock volatility, and political uncertainty in a natural experiment: germany, 1880–1940. the journal of finance 53(6), 2243-2257. https://doi.org/10.1111/0022-1082.00090 bloom, n. (2009). the impact of uncertainty shocks. econometrica 77(3), 623-685. https://doi.org/10.3982/ecta6248 bloom, n. (2014). fluctuations in uncertainty. journal of economic perspectives 28(2), 153-76. https://doi.org/10.1257/jep.28.2.153 https://doi.org/10.1111/j.1540-6261.2006.00836.x https://doi.org/10.1111/j.1540-6261.2006.00836.x https://doi.org/10.1016/j.jfineco.2007.12.005 https://doi.org/10.1093/qje/qjw024 https://doi.org/10.3982/ecta13960 https://doi.org/10.1016/j.jfineco.2012.08.016 https://doi.org/10.1111/0022-1082.00090 https://doi.org/10.3982/ecta6248 https://doi.org/10.1257/jep.28.2.153 mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 277 bloom, n., bond, s. and van-reenen, j. (2007). uncertainty and investment dynamics. the review of economic studies 74(2), 391-415. https://doi.org/10.1111/j.1467937x.2007.00426.x born, b. and pfeifer, j. (2014). policy risk and the business cycle. journal of monetary economics 68, 68-85. https://doi.org/10.1016/j.jmoneco.2014.07.012 boutchkova, m., doshi, h., durnev, a. and molchanov, a. (2012). precarious politics and return volatility. the review of financial studies 25(4), 1111-1154. https://doi.org/10.1093/ rfs/hhr100 brogaard, j. and detzel, a.l. (2015). the asset pricing implications of government economic policy uncertainty. management science 61(1), 3-18. https://doi.org/10.1287/mnsc.2014 .2044 caixabank research (2016). incertidumbre y mercado bursátil. im09, september 2016, p. 9. retrieved from http://www.caixabankresearch.com caldara, d. and iacoviello, m. (2019). measuring geopolitical risk. international finance discussion papers, board of governors of the federal reserve system (eeuu), working paper no. 1222. https://doi.org/10.17016/ifdp.2018.1222 campbell, j. (1993). intertemporal asset pricing without consumption data. american economic review 83(3), 487-512. https://doi.org/10.3386/w3989 campbell, j. (1996). understanding risk and return. the journal of political economy 104(2), 298-345. https://doi.org/10.1086/262026 carney, m. (2016). uncertainty, the economy and policy. speech given by mark carney, governor of the bank of england and chairman of the financial stability board, at the bank of england, london, 30 june 2016. retrieved from https://www.bankofengland.co.uk/speech/2016/ uncertainty-the-economy-and-policy castellanos-garcía, p., pérez-del-río, i. and sánchez-santos, j.m. (2014). european journal of government and economics 3(2), 148-161. https://doi.org/10.17979/ejge.2014.3.2.4303 cboe (2019). white paper. cboe volatility index. retrieved of https://cdn.cboe.com/resources/ futures/vixwhite.pdf chen, j. (2003). intertemporal capm and the cross section of stock returns. university of california davis, unpublished paper. https://doi.org/10.2139/ssrn.301918 chen, x., ghysels, e. and wang, f. (2015). hybrid-garch: a generic class of models for volatility predictions using high frequency data. statistica sinica 25(2), 759-786. https://doi.org/ 10.5705/ss.2012.283 coleman, s., leone, v. and medeiros, o.r. (2018). latin american stock market dynamics and comovement. international journal of finance and economics 24(3), 1109-1129. dixit, a.k. and pindyck, r.s. (1994). investment under uncertainty (pp. 11-14). princeton university press, nueva jersey. https://doi.org/10.1515/9781400830176 durnev, a. (2010). the real effects of political uncertainty: elections and investment sensitivity to stock prices. mcgill university, working paper 22 december 2010. https://doi.org/ 10.2139/ssrn.1695382 engle, r.f. and granger, c.w.j. (1987). cointegration and error correction: representation, estimation and testing. econometrica 55(2), 251-276. https://doi.org/10.2307/1913236 erb, c.b., harvey, c.r. and viskanta, t.e. (1996). expected returns and volatility in 135 countries. the journal of portfolio management 22(3), 46-58. https://doi.org/10.2139/ ssrn.871253 farago, a. and tédongap, r. (2018). downside risks and the cross-section of asset returns. journal of financial economics 129(1), 69-86. https://doi.org/10.1016/j.jfineco.2018.03.010 federal open market committee. meeting diciembre 2009. retrieved from http://www.federalreserve.gov/monetarypolicy/fomcminutes20091216.htm fernández-villaverde, j., guerrón-quintana, p., kuester, k. and rubio-ramírez, j. (2015). fiscal volatility shocks and economic activity. american economic review 105(11), 3352-3384. https://doi.org/10.1257/aer.20121236 friedman, m. (1968). the role of monetary policy. the american economic review 58(1), 1-17. recuperado de https://www.jstor.org/stable/1831652?origin=jstor-pdf&seq=1 giavazzi, f. and mcmahon, m. (2012). policy uncertainty and household savings. the review of economics and statistics 94(2), 517-531. https://doi.org/10.1162/rest_a_00158 https://doi.org/10.1111/j.1467-937x.2007.00426.x https://doi.org/10.1111/j.1467-937x.2007.00426.x https://doi.org/10.1016/j.jmoneco.2014.07.012 https://doi.org/10.1093/rfs/hhr100 https://doi.org/10.1093/rfs/hhr100 https://doi.org/10.1287/mnsc.2014.2044 https://doi.org/10.1287/mnsc.2014.2044 http://www.caixabankresearch.com/ https://doi.org/10.17016/ifdp.2018.1222 https://doi.org/10.3386/w3989 https://doi.org/10.1086/262026 https://www.bankofengland.co.uk/speech/2016/uncertainty-the-economy-and-policy https://www.bankofengland.co.uk/speech/2016/uncertainty-the-economy-and-policy https://doi.org/10.17979/ejge.2014.3.2.4303 https://cdn.cboe.com/resources/futures/vixwhite.pdf https://cdn.cboe.com/resources/futures/vixwhite.pdf https://doi.org/10.2139/ssrn.301918 https://doi.org/10.5705/ss.2012.283 https://doi.org/10.5705/ss.2012.283 https://doi.org/10.1515/9781400830176 https://doi.org/10.2139/ssrn.1695382 https://doi.org/10.2139/ssrn.1695382 https://doi.org/10.2307/1913236 https://doi.org/10.2139/ssrn.871253 https://doi.org/10.2139/ssrn.871253 https://doi.org/10.1016/j.jfineco.2018.03.010 http://www.federalreserve.gov/monetarypolicy/fomcminutes20091216.htm https://doi.org/10.1257/aer.20121236 https://www.jstor.org/stable/1831652?origin=jstor-pdf&seq=1 https://doi.org/10.1162/rest_a_00158 mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 278 gregory, w.a. and hansen, e.b. (1996). residual-based test for cointegration in models with regime shifts. journal of econometrics 70(1), 99-126. https://doi.org/10.1016/0304-4076(69) 41685-7 hassett, k.a. and metcalf, g.e. (1999). investment with uncertain tax policy: does random tax policy discourage investment? the economic journal 109(july), 372–393. https://doi.org/10.1111/1468-0297.00453 higgs, r. (1997). regime uncertainty: why the great depression lasted so long and why prosperity resumed after the war. independent review 1(4), 561-590. retrieved from https://www.jstor.org/stable/24560785?seq=1 hollstein, f. and prokopczuk, m. (2018). how aggregate volatility-of-volatility affects stock returns. the review of asset pricing studies 8(2), 253-292. https://doi.org/10.1093/rapstu/ rax019 ilut, c.l. and schneider, m. (2014). ambiguous business cycles. american economic review 104(8), 2368-99. https://doi.org/10.1257/aer.104.8.2368 imf. world economic outlook: coping with high debt and sluggish growth. imf press, october 2012. imf. world economic outlook: hopes, realities, risks. imf press, april 2013. julio, b. and yook, y. (2012). political uncertainty and corporate investment cycles. journal of finance 67(1), 45-83. https://doi.org/10.1111/j.1540-6261.2011.01707.x kaeck, a. (2018). variance-of-variance risk premium, review of finance. european finance association 22(4), 1549-1579. https://doi.org/10.1093/rof/rfx008 khan, j. and khan, i. (2018). the impact of macroeconomic variables on stock prices: a case study of karachi stock exchange. business and economics journal 9(3), 1-8. https://doi.org/10.4172/2151-6219.1000365 koijen, r.s.j., philipson, t.j. and uhlig, h. (2016). financial health economics. econometrica 84(1), 195-242. https://doi.org/10.3982/ecta11182 leduc, s. and liu, z. (2015). uncertainty shocks are aggregate demand shocks. federal reserve bank of san francisco, working paper no. 2012-10. https://doi.org/10.24148/ wp2012-10 li, j. and born, j.a. (2006). presidential election uncertainty and common stock returns in the united states. the journal of financial research 29(4), 609-22. https://doi.org/10.1111/ j.1475-6803.2006.00197.x malkiel, b.g. and xu, y. (2006). idiosyncratic risk and security returns. annual meetings of the american finance association. the econometrics society conference, working paper. retrieved from https://personal.utdallas.edu/~yexiaoxu/ivot_h.pdf markowitz, h. (1959). portfolio selection: efficient diversification of investments. yale university press. retrieved from https://www.jstor.org/stable/j.ctt1bh4c8h merton, r.c. (1973). theory of rational option pricing. the bell journal of economics and management science 4(1), 141-183. https://doi.org/10.2307/3003143 oecd (2019). equity market development in latin america: enhancing access to corporate finance. pantzalis, c., stangeland, d.a. and turtle, h.j. (2000). political elections and the resolution of uncertainty: the international evidence. journal of banking and finance 24(10), 1575-1604. https://doi.org/10.1016/s0378-4266(99)00093-x pastor, l. and veronesi, p. (2012). uncertainty about government policy and stock prices. journal of finance 67(4), 1219-1264. https://doi.org/10.1111/j.1540-6261.2012.01746.x pastor, l. and veronesi, p. (2013). political uncertainty and risk premia. journal of financial economics 110(3), 520-545. https://doi.org/10.1016/j.jfineco.2013.08.007 pesaran, m.h. and shin, y. (1999). an autoregressive distributed lag modelling approach to cointegration analysis. econometrics and economic theory in the 20th century: the ragnar frisch centennial symposium, strom, s. (ed.) cambridge university press, cambrige. https://doi.org/10.1017/ccol521633230.011 pesaran, b. and pesaran, h. (2009). time series econometrics using microfit 5.0. oxford, england: oxford university press. https://doi.org/10.1016/0304-4076(69)41685-7 https://doi.org/10.1016/0304-4076(69)41685-7 https://doi.org/10.1111/1468-0297.00453 https://www.jstor.org/stable/24560785?seq=1 https://doi.org/10.1093/rapstu/rax019 https://doi.org/10.1093/rapstu/rax019 https://doi.org/10.1257/aer.104.8.2368 https://doi.org/10.1111/j.1540-6261.2011.01707.x https://doi.org/10.1093/rof/rfx008 https://doi.org/10.4172/2151-6219.1000365 https://doi.org/10.3982/ecta11182 https://doi.org/10.24148/wp2012-10 https://doi.org/10.24148/wp2012-10 https://doi.org/10.1111/j.1475-6803.2006.00197.x https://doi.org/10.1111/j.1475-6803.2006.00197.x https://personal.utdallas.edu/%7eyexiaoxu/ivot_h.pdf https://www.jstor.org/stable/j.ctt1bh4c8h https://doi.org/10.2307/3003143 https://doi.org/10.1016/s0378-4266(99)00093-x https://doi.org/10.1111/j.1540-6261.2012.01746.x https://doi.org/10.1016/j.jfineco.2013.08.007 https://doi.org/10.1017/ccol521633230.011 mary elena sánchez-gabarre / european journal of government and economics 9(3), december 2020, 265-279 279 pesaran, m.h., shin, y. and smith, r.j. (2001). bounds testing approaches to the analysis of level relationships. journal of applied economics 16(3), 289-326. https://doi.org/10.1002/ jae.616 rodrik, d. (1991). policy uncertainty and private investment, journal of development economics. 36 (2), 229-242. https://doi.org/10.1016/0304-3878(91)90034-s santa-clara, p. and valkanov, r. (2003). the presidential puzzle: political cycles and the stock market. the journal of finance 58(5), 1841-1872. https://doi.org/10.1111/1540-6261.00590 scotti, c. (2016). surprise and uncertainty indexes: real-time aggregation of real-activity macro-surprises. journal of monetary economics 82, 1-19. https://doi.org/10.1016/ j.jmoneco.2016.06.002 toda, h. and yamamoto, t. (1995). statistical inference in vector autoregressions with possibly integrated processes. journal of econometrics 66(1-2), 225-250. https://doi.org/10.1016/ 0304-4076(94)01616-8 voth, h.j. (2003). stock price volatility and political uncertainty: evidence from the interwar period. mit department of economics. working paper no. 02-09. https://doi.org/10.2139/ ssrn.342641 https://doi.org/10.1002/jae.616 https://doi.org/10.1002/jae.616 https://doi.org/10.1016/0304-3878(91)90034-s https://doi.org/10.1111/1540-6261.00590 https://doi.org/10.1016/j.jmoneco.2016.06.002 https://doi.org/10.1016/j.jmoneco.2016.06.002 https://doi.org/10.1016/0304-4076(94)01616-8 https://doi.org/10.1016/0304-4076(94)01616-8 https://doi.org/10.2139/ssrn.342641 https://doi.org/10.2139/ssrn.342641 governance and domestic investment in africa vol.8 • no.1 2019 issn: 2254-7088 european journal of government and economics 8(1), june 2019. european journal of government and economics issn: 2254-7088 number 8, issue 1, june 2019 doi: https://doi.org/10.17979/ejge.2018.8.1 the evaluation index improvement for the relocation of research equipment 6-29 donghun yoon residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) 30-47 josé ramón cardona, daniela álvarez bassi, maría dolores sánchez-fernández social capital and banking system profitability: a survey of european union countries 48-62 arash nayebyazdi governance and domestic investment in africa 63-80 chimere okechukwu iheonu economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt 81-95 marwa shibl biltagy give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia 96-109 jaroslav mihálik, peter horváth, martin švikruha https://doi.org/10.17979/ejge.2018.8.1 european journal of government and economics 8(1), june 2019, 63-80. european journal of government and economics issn: 2254-7088 governance and domestic investment in africa chimere okechukwu iheonua, * a university of nigeria nsukka, nigeria * corresponding author at: iheonuchimere@yahoo.com article history. received 5 october 2018; first revision required 24 october 2018; accepted 18 march 2019. abstract. the study empirically examined the impact of governance on domestic investment in 16 african countries with a balanced panel data set, between the years 2002 and 2015. the study employed six unbundled governance indicators from the world bank, world governance indicators and constructed three bundled governance indicators using the principal component analysis. the driscoll and kraay fixed effects model which accounts for serial correlation, groupwise heteroskedasticity and cross-sectional dependence were employed with empirical results revealing that all the indicators of governance positively and significantly influence domestic investment in africa, except for government effectiveness which happens to be insignificant. also, voice/accountability and the control of corruption exert more influence on domestic investment as indicated by their coefficient values. furthermore, economic growth is also an important factor in explaining domestic investment in africa. policy recommendations are discussed. keywords. governance; domestic investment; africa; principal component analysis; fixed effects model. jel codes: c1, e2, r5 doi. https://10.17979/ejge.2019.8.1.4565 1. introduction investment has mostly been described as a significant component of growth in economic discourse. it has been deemed essential for economies in order to enhance economic growth, increase employment and reduce poverty. investment has been considered to be essential to the african region as it enables job creation and feed its teeming population. according to ouedraogo and kouaman (2014), an investment can improve production capacity through the acquisition of new equipment which incorporates technical progress and thus increases labour productivity. investment can also increase productivity through the increase in aggregate demand as well as increasing employment and wages. economic theory posits that a significant factor influencing investment is the real cost of borrowing, which adjusts for the rate of inflation, likewise economic growth. investment is essential for economies whose priority is a consistent movement towards economic growth and development. hamuda et al. (2013) point out that countries that have accumulated a high level of long-term investment mailto:iheonuchimere@yahoo.com https://10.0.70.59/ejge.2019.8.1.4565 chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 64 belong to the cadre of developed nations. investment in machinery and buildings not only create jobs but also contributes to the current demand for capital goods and thus increases domestic expenditure. an increase in investment also prompts an increase in aggregate supply, which helps in relaxing stagflationary tendencies. according to lim (2014), between 1980 and 2010, gross capital formation rate ranged from 1 to 90 per cent of production in the world. this loose gap in investment activity has been connected to diverse varieties of frictions existing in many economies and has hindered the normalisation of the proceeds from investment undertakings across countries (chuku, onye and ajah, 2015). in comparison to other regions of the globe, the level of domestic investment in africa stands low, as reported in figure 1.1. the figure shows that sub-saharan africa, which of course is a subset of africa, has the lowest level of domestic investment in the world between the years 2000 and 2016 as indicated by the region’s gross fixed capital formation. europe and central asia (eca) had the highest level of domestic investment in the world. this is not farfetched as the region is made up of the most advanced countries in the world. latin america and the caribbean (lac) comes just behind the eca. in fact, the level of domestic investment in lac is more than six times the level of domestic investment in sub-saharan africa and about twice the level of domestic investment in the middle east and north africa (mena) in 2000. the substantial disparity between domestic investment undertaken in sub-saharan africa and that of lac and eca is also apparent across the time structure of the data observation. in subsaharan africa, though the lowest among other regions, the level of domestic investment in the region has continuously increased. figure 1.1. gross fixed capital formation, constant us$ (2000-2016) source: authors computation from world development indicator, wdi (2017) note: ssa is sub saharan africa, mena is the middle east and north africa, eca is europe and central asia, sa is south asia, and lac is latin america and the caribbean in 2005, domestic investment in sub-saharan africa increased to more than us$ 154 billion from about us$ 112 billion in 2000 based on wdi (2017) data. domestic investment in sub0 1e+12 2e+12 3e+12 4e+12 5e+12 6e+12 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 ssa mena eca sa lac chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 65 saharan africa increased to more than us$ 269 billion in 2010 and further increased to more than us$ 332 billion in 2015, according to the data. the quality of governance is pointed out as one of the factors that significantly affect the variations in investment activities across countries and regions. according to khan (2007), governance has been identified as a critical factor explaining the difference in economic performance across developing countries. according to akanbi (2010), the low quality of governance, which is mirrored by the insecure political atmosphere in most african nations has remained a key hindrance to the growth of domestic investment over time. this then has made it imperative to model investment determinants by incorporating the quality of governance (ajide, 2013). fayissa and nsiah (2013) argue that good governance or the absence of sound governance adds to the gap in income per capita amongst richer and poorer african nations. chauvet and collier (2004) posit that countries experiencing poor governance are associated with an average of 2.3 percentage points less gdp per year relative to other developing countries. in fact, there is a current stream of african development literature which suggests that governance is essential in driving private investment (asongu et al., 2015), growth (asongu, 2016a), inclusive development (asongu and nwachukwu, 2016) and the fight against policy syndromes such as capital flight (asongu and nwachukwu, 2017) and terrorism (asongu et al., 2017). in the empirical literature, various studies have examined the determinants of investment across countries. studies by aysan, gaobo and marie-ange (2005), bader and malawi (2010), eregha (2010), majed and ahmad (2010) show that interest rate negatively affects investment. these conclusions support the keynesian theory of investment where investment choices are made by relating the marginal efficiency of capital (mec) to the real interest rate. the investment would be made if the mec is higher than the real interest rate and as such, the real interest rate is negatively associated with investment decisions. studies by batina (1998), pereira (2000) and pereira (2001) have also revealed that growth spurs investment, which supports the accelerator theory of investment. this theory asserts that increasing output drives investment. according to gordon (2009), temporary changes in output could lead to changes in investment spending. the modified version of this theory introduced a time lag between the increase in output and the subsequent increase in investment. tobin (1969) developed an investment theory known as the tobin q theory. tobin q can be defined simply as the proportion of the market value of a unit of capital to its replacement cost. q, according to romer (2012) is said to summarise complete information about the future that is important to a firm’s investment choice. q reveals in what way an extra dollar increase in capital affects the present value of profits and so when q is high, firms want to increase their capital stock, and when q is low, firms reduce it. inquiries on the impact of governance on domestic investment are rare as most related studies focus more on foreign direct investment (fdi), notably: morisset (2000), globerman and shapiro (2002), asiedu (2005), samini and ariani (2010) and mengistu and adhikary (2011). firms invest when the investment climate is favourable, and governance quality is a chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 66 subset of the investment climate in africa. among the few studies on the role that governance plays in the domestic investment decision is that of ouedrago and kouaman (2014), which examines the role of governance in explaining private investment in sub-saharan africa. the study employed 38 sub-saharan african countries from 2006 to 2011 and made use of the generalised method of moment (gmm) estimation technique. the results revealed that heavy regulations disturb private investment, whereas the business environment improves investment in sub-saharan africa. aysan, nabli and veganzones (2011) also studied the impact of governance on private investment in the middle east and north africa (mena) region. their empirical findings revealed that corruption control, bureaucratic quality, investment-friendly administrations, law and order and a stable political environment play significant roles in explaining private investment decisions. ngov (2008), using intra-group regression investigated the impact of governance on fdi as well as promoting domestic investment along with growth performance in three different income sets of countries which include low-income countries, middle-income countries and high-income countries. the result revealed that governance is positively related to per capita growth rate in both the middle and high-income groups but not in the low-income group. the result further revealed that governance has a positive impact on total investment ratio, which is a combination of domestic investment and fdi. in empirical examining the impact of governance on domestic investment in africa, this study employs a balanced panel data of 16 african countries as reported by world development indicators (wdi, 2017) and the world governance indicators (wgi, 2017) of the world bank for the year 2002 to 2015. data availability constraints strictly guided the choice of countries. the remainder of this research is organised as follows: section 2 presents the methodology and model specification adopted in the study. section 3 presents the econometric results. section 4 concludes the research with relevant policy recommendations. 2. methodology and model specification the study employed descriptive statistics, the correlation matrix and panel linear models comprising of the pooled ordinary least square (ols), the fixed effects (fe) model and the random effects (re) model. the descriptive statistics employed the mean of each of the variables within the countries employed in the study. this would enable us to understand the distinct uniqueness of the economic and governance fundamentals employed in the model. the correlation matrix helps to understand the degrees of substitution of relationships among the variables in the model, which are particularly important in order to avoid the problems of multicollinearity and biased estimates. also, the study employed nine bundled and unbundled governance indicators sourced from wgi. the bundled governance indicators were constructed by exploiting principal component analysis (pca) to reduce six governance variables. results permitted to obtain the following constructs: (1) political governance, which comprises political stability and voice/accountability, (2) economic governance, which is composed of government effectiveness and regulatory chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 67 quality, and (3) institutional governance, which includes the control of corruption and the rule of law. according to asongu et al. (2017), the principal component analysis involves reducing a set of strongly correlated indices into an uncorrelated set of small variables known as principal components (pc). according to tchamyou (2017), the pcs are said to account for most of the information in the original data set. in the pca, it is required that only common factors that have an eigenvalue greater than one or the mean should be retained (kaiser, 1974; jollife, 2002). among the three alternative estimation techniques, the pooled ols which is an ols technique that is run in panel data assumes that there is no heterogeneity across cross sections and stated as; 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝛽𝛽𝑋𝑋𝑖𝑖𝑖𝑖 + 𝜃𝜃𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝜖𝜖𝑖𝑖𝑖𝑖 [1] where 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 is the natural logarithm of gross fixed capital formation which proxy’s for domestic investment, 𝛼𝛼 is the common intercept, 𝑋𝑋 is a vector of control variables which comprises of the natural logarithm of gross domestic product in constant us$, which proxy’s for real economic growth, the natural logarithm of exchange rate for each of the individual countries’ currency employed in the model to the us$, and the real interest rate which signifies the real cost of borrowing. 𝐺𝐺𝐺𝐺𝐺𝐺 is composed of the governance indicators (both bundled and unbundled). they include political stability, voice/accountability, political governance, government effectiveness, regulatory quality, economic governance, control of corruption, rule of law and institutional governance. 𝜖𝜖 is the error term while i denotes the cross-sectional index, 𝑡𝑡 denotes the time index. the fe model which controls for heterogeneity across countries in the intercept parameters is expressed as; 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑖𝑖 + 𝛽𝛽𝑋𝑋𝑖𝑖𝑖𝑖 + 𝜃𝜃𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝜖𝜖𝑖𝑖𝑖𝑖 [2] where 𝛼𝛼𝑖𝑖 is the regional specific parameter which denotes the fixed effect. according to algieri and mannarino (2013), the 𝛼𝛼𝑖𝑖 is said to represent ignorance about every other systematic feature that predict the dependent variables other than 𝑋𝑋 and 𝐺𝐺𝐺𝐺𝐺𝐺. the basic insight into the fe model is that 𝛼𝛼𝑖𝑖 does not change over time. hence any variations in the outcome variable must be due to stimuluses other than these fixed individualities (stock & watson, 2008). the fe model is commonly used when examining the influence of variables that change with time as it controls for fixed individual characteristics of the countries in the model. the re model, on the other hand, treats the heterogeneity across cross sections as random components and is stated as; 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝛽𝛽𝑋𝑋𝑖𝑖𝑖𝑖 + 𝜃𝜃𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝜇𝜇𝑖𝑖𝑖𝑖 + 𝜖𝜖𝑖𝑖𝑖𝑖 [3] 𝜇𝜇𝑖𝑖𝑖𝑖 is the distinct specific error, also known as the between-entity error. the variations across units are presumed to be random and uncorrelated with the independent variables in the model. in the re model, it is assumed that the entity error term is uncorrelated with the independent variables in the model and thus allowing time-invariant variables play a role as explanatory variables. chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 68 the choice of the study’s baseline model is based on several tests. in choosing between the fe and the pooled ols, the study applies the f-test which confirms if there is omitted variable bias in the model. a p-value of less than 5 per cent signifies that there are important country effects, which means that overlooking unobserved heterogeneity in the model can lead to estimation bias and inconsistency. the study also tests between the pooled ols and the re model using the breusch-pagan (bp)-langragian multiplier (lm) test. the null hypothesis of the bp-lm test is that there is no substantial variance across regions. a less than 5 per cent probability value for the bp-lm test indicates that the re model is appropriate, and the pooled ols is not appropriate. the hausman 𝜒𝜒2 test is also performed in selecting between the fe model and the re model. according to algieri and mannarino (2013), the hausman 𝜒𝜒2 test is intended to identify a violation of the re modelling assumption that the regressors are orthogonal to unit effects. this means that there is no correlation between the independent variables and the unit effect. this further implies that the estimates of the fe model should be analogous to the estimates of the re model. while the alternative hypothesis is that the fe model is preferable to the re model, the null hypothesis is that both the fe and the re produce similar coefficients. according to baek and yang (2010), the hausman test allows us to determine whether the appropriate error terms are fixed or drawn from a random distribution. a p-value greater than 5 per cent denotes that both the fe model and the re model are reliable but the re model is more efficient because it utilises a lesser degree of freedom. the study also tests for serial correlation, which biases the standard errors in linear panel data models and makes findings less efficient according to drukker (2003). the study applies the wooldridge (2002) test, which hypothesises a null indicating the absence of autocorrelation in the model. we also test for heteroskedasticity in the fe model using the modified wald test developed by lasker and king (1997). the null hypothesis of this test is that the variance of the error is similar for all countries (amaz, gaume and lefevre, 2012). the study also tests for crosssectional dependence in the model employing the pesaran (2004) cd test and the pesaran (2004) scaled lm test. according to baltagi, kao and peng (2016), cross-sectional dependence could arise as a result of unknown common shocks, spatial effects or interactions within social networks. overlooking cross-sectional dependence could result to estimation bias. a reason for crosssectional dependence, according to de hoyos and sarafidis (2006), may be due to the growing economic and financial integration of countries and financial bodies, which suggests solid interdependencies amongst cross sectional entities. countries involved in this study include algeria, egypt, gambia, kenya, lesotho, liberia, malawi, mauritius, mozambique, nigeria, rwanda, sierra leone, south africa, swaziland, tanzania and uganda. the time span for this study is between the years 2002 to 2015. chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 69 3. presentation and discussion of empirical results this section begins with a simple descriptive statistic of the variables in the model employing the mean for each cross section. results from table 3.1 reveal that in its natural logarithm, domestic investments are quite similar across the countries employed in the model. while liberia has the lowest mean value of 19.18, south africa has the highest level of domestic investment with a mean value of 24.93 for the years n view. also, in the third column, economic growth likewise exhibits similarities across the countries in its natural logarithm. while south africa can be seen to have the highest level of economic growth, gambia has the lowest level of economic growth. on the average, egypt has the strongest currency among the countries employed in the study, with an average exchange rate of 1.78 in its natural logarithm to the us$ while sierra leone has the weakest currency of 8.13 in natural logarithm to the us$. the real interest rate signifying the real borrowing cost is highest in gambia and lowest in algeria. table 3.1. mean of variables across countries, 2002-2015 country domestic investment economic growth exchange rate real interest rate cc ge ps rl rq va algeria 24.64 25.76 4.33 2.02 -0.55 -0.55 -1.27 -0.69 -0.90 -0.96 egypt 24.25 25.99 1.78 2.11 -0.57 -0.50 -0.99 -0.22 -0.42 -1.07 gambia 19.21 20.57 3.37 21.43 -0.62 -0.64 0.12 -0.39 -0.40 -0.99 kenya 22.58 24.34 4.38 6.36 -0.99 -0.53 -1.25 -0.86 -0.22 -0.27 lesotho 20.11 21.53 2.11 5.79 0.01 -0.35 0.07 -0.20 -0.55 -0.21 liberia 19.18 20.92 4.21 7.43 -0.78 -1.36 -1.10 -1.18 -1.29 -0.53 malawi 20.41 22.53 5.13 13.71 -0.61 -0.60 0.001 -0.22 -0.58 -0.28 mauritius 21.46 22.95 3.41 9.68 0.49 0.79 0.87 0.95 0.76 0.85 mozambique 21.28 22.93 3.31 12.25 -0.56 -0.56 0.14 -0.66 -0.46 -0.14 nigeria 24.39 26.48 4.96 4.19 -1.13 -1.02 -1.93 -1.22 -0.87 -0.72 rwanda 20.74 22.36 6.39 8.65 0.11 -0.28 -0.56 -0.47 -0.36 -1.25 sierra leone 19.69 21.63 8.13 10.42 -0.90 -1.23 -0.34 -1.01 -0.92 -0.34 south africa 24.93 26.60 2.11 4.14 0.17 0.49 -0.08 0.10 0.50 0.61 swaziland 20.29 22.17 2.11 4.58 -0.33 -0.71 -0.21 -0.58 -0.52 -1.28 tanzania 22.76 24.08 7.19 4.63 -0.62 -0.52 -0.32 -0.42 -0.42 -0.24 uganda 22.18 23.60 7.66 12.39 -0.90 -0.50 -1.09 -0.43 -0.16 -0.56 source: author’s computation from wdi and wgi data. note: cc is the control of corruption, ge is government effectiveness, ps is political stability, rl is rule of law, rq is regulatory quality, va is voice/accountability. in terms of governance indicators, of the 16 african countries employed in the model, only lesotho, mauritius, rwanda and south africa have a positive value for the control of corruption which denotes healthier control of corruption in these countries. among these four countries, mauritius controls corruption better as indicated by its higher value with respect to the other countries. also, nigeria can be seen to have the lowest level of corruption control amongst the chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 70 countries in our model with a value of -1.13. of course, this means that corruption is prevalent in the country. furthermore, the table reveals that south africa has a more effective government compared to other countries in our model while liberia has the lowest government effectiveness. also, table 3.1 shows that mauritius is more politically stable than the other countries in our model. gambia, lesotho, malawi and mozambique are the other countries that enjoy a relatively stable political environment as indicated by their positive values. mauritius and south africa have a strong rule of law while nigeria has the weakest rule of law as indicated by their values. in terms of regulatory quality and voice/accountability, mauritius also has the best of both while liberia has the lowest, in terms of regulatory quality as an indicator of quality governance, swaziland has the lowest level of voice/accountability. the quality of governance in mauritius might as well be a good reason why the country has a very low rate of poverty with less than 1 per cent of its population living on $1 a day or less (see, borgenproject.org). table 3.2. correlation matrix di gdp exc ri cc ge ps rq rl va di 1.0000 gdp 0.9804 1.0000 exc -0.1120 -0.1201 1.0000 ri -0.3368 -0.3489 0.1281 1.0000 cc -0.0073 -0.0608 -0.3887 -0.0427 1.0000 ge 0.3368 0.2715 -0.3290 -0.0384 0.7790 1.0000 ps -0.3827 -0.4279 -0.2285 0.2201 0.6503 0.5355 1.0000 rq 0.2689 0.2330 -0.2203 0.0142 0.6351 0.8803 0.5100 1.0000 rl 0.1046 0.0537 -0.3098 0.0652 0.7635 0.8785 0.6971 0.8117 1.0000 va 0.1107 0.0977 -0.0877 -0.0105 0.4102 0.5775 0.4942 0.5659 0.5547 1.0000 source: author’s computation. note: di is domestic investment, gdp is economic growth, exc is exchange rate, ri is real interest rate, cc is the control of corruption, ge is government effectiveness, ps is political stability, rq is regulatory quality, rl is rule of law, va is voice/accountability. table 3.2 is the correlation analysis, which shows the correlation between the variables in our model. the table reveals that domestic investment and economic growth have a strong positive correlation to the value of 0.98. the table also shows that exchange rate, real interest rate, the control of corruption and political stability all have negative correlations with domestic investment. furthermore, our explanatory variables do not have a strong correlation with each other, which then solves the issue of multicollinearity apart from the governance indicators which we would include in separate regressions. the result from the principal component analysis in table 3.3 reveals that the study retains the first pc for the three constructed governance indices since their eigenvalues are greater than one and represent 75 per cent of the information in political governance, 94 per cent of the chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 71 information in economic governance and 88 per cent of the information in institutional governance. table 3.3. principal component analysis of governance indicators principal component component matrix(loadings) proportion cumulative eigenvalue ps va ge rq cc rl political gov first pc 0.7071 0.7071 0.7471 0.7471 1.4942 second pc 0.7071 -0.7071 0.2529 1.0000 0.5057 economic gov first pc 0.7071 0.7071 0.9401 0.9401 1.8803 second pc 0.7071 -0.7071 0.0599 1.0000 0.1197 institutional gov first pc 0.7071 0.7071 0.8818 0.8818 1.7635 second pc -0.7071 0.7071 0.1182 1.0000 0.2364 source: author’s computation. note: pc denotes principal component. gov denotes governance. cc is the control of corruption, ge is government effectiveness, ps is political stability, rq is regulatory quality, rl is rule of law, va is voice/accountability. tables 3.4 to 3.6 show the empirical results on the nexus between governance and domestic investment in africa. though the results across the various econometric techniques show that governance generally has a positive influence on domestic investment, for policy inference, the study needs to decide on the appropriate baseline model. the study estimated the f-test to choose between the pooled ols and fe model. a significant f-test probability value indicates that the fe model is more suitable than the pooled ols. the choice between the re model and the pooled ols employing the bp lm test also suggests that the re model is more suitable while the hausman test suggests that the fe model is more suitable in comparison to the re model. our baseline model then becomes the fe model. the study then proceeds to test for heteroskedasticity with results suggesting that our models (tables 3.4 to 3.6) all suffer from non-constant variance in the error term. results from the serial correlation test of wooldridge also show that all our models suffer from serial correlation while the pesaran cd and pesaran scaled lm tests for cross-sectional dependence show contrary results at 5 per cent level of statistical significance. while the pesaran cd test indicates the absence of cross-sectional dependence, the pesaran scaled lm test indicates its presence. however, the study account for these biases in the fe model employing the driscoll and kraay (1998) standard errors which correct for heteroskedasticity, serial correlation and cross-sectional dependence using a nonparametric covariance matrix estimator (see hoechle, 2007). chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 72 table 3.4. political governance and domestic investment in africa political stability voice/accountability political governance pooled ols fixed effects random effects driscoll kraay pooled ols fixed effects random effects driscoll kraay pooled ols fixed effects random effects driscoll kraay constant -2.4394a (0.000) 17.4543a (0.000) -6.9798a (0.000) 17.4543a (0.000) -1.8616a (0.000) 18.0391 a (0.000) -6.4212a (0.000) 18.0391a (0.000) -1.8838a (0.000) 18.1185a (0.000) -6.5165a (0.000) 18.1185a (0.000) gdp 1.0329a (0.000) 1.6622a (0.000) 1.2081a (0.000) 1.6622a (0.000) 1.0085a (0.000) 1.6978a (0.000) 1.1860a (0.000) 1.6978a (0.000) 1.0085a (0.000) 1.6978a (0.000) 1.1861a (0.000) 1.6978a (0.000) exc 0.0203 (0.138) 0.0841 (0.395) 0.1295a (0.000) 0.0841 (0.678) 0.0065 (0.628) 0.0313 (0.750) 0.1148a (0.005) 0.0313 (0.869) 0.0065 (0.628) 0.0313 (0.872) 0.1148a (0.005) 0.0314 (0.869) ri -0.0005 (0.986) -0.0002 (0.932) 0.0004 (0.835) -0.0002 (0.942) 0.0011 (0.729) 0.0003 (0.872) 0.0011 (0.624) 0.0003 (0.877) 0.0011 (0.729) 0.1089c (0.058) 0.0011 (0.624) 0.0003 (0.877) ps 0.1363a (0.001) 0.1248b (0.019) 0.2263a (0.000) 0.1248b (0.037) va 0.0495 (0.249) 0.1769c (0.058) 0.2121b (0.015) 0.1769b (0.028) polgov 0.0306 (0.249) 0.1089c (0.058) 0.1307b (0.015) 0.1089b (0.028) r2 overall 0.9632 0.9614 0.9553 0.9614 0.9610 0.9519 0.9614 0.9610 0.9519 r2 between 0.9784 0.9706 0.9789 0.9682 0.9789 0.9682 r2 within 0.7626 0.7508 0.7626 0.7603 0.7561 0.7603 0.7603 0.7561 0.7603 f-stat 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 f-test 0.0000 0.0000 0.0000 wald 0.0000 0.0000 0.0000 hausman 0.0350 0.0000 0.0000 b-p lm 0.0000 0.0000 0.0000 m.wald 0.0000 0.0000 0.0000 wooldridge 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 pesaran cd 0.2969 0.0000 0.2969 0.0000 0.1216 0.0000 pesaran scaled lm 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 observations 224 224 224 224 224 224 224 224 224 224 224 224 source: author’s computation. note: dependent variable: domestic investment. gdp is economic growth, ri is the real interest rate, cc is the control of corruption, rl is rule of law, polgov is political governance, b-p lm is breusch pagan langragian multiplier test, m.wald is the modified wald test for heteroskedasticity. probability values of coefficients are in parenthesis. diagnosis and post-estimation test results presented are probability values. a, b and c denote statistical significance at 1, 5 and 10%, respectively. results in table 3.4 on the relationship between political governance and domestic investment along with its elements reveals that an improvement in political stability increases domestic investment in africa by 0.125 percentage point and this relationship is significant at 5 per cent statistical level. similarly, voice/accountability increases domestic investment by 0.177 percentage point while an improvement in political governance increases domestic investment by 0.109 percentage point significantly. chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 73 the f-statistics probability values which are less than 5 per cent indicates that all the variables in the model together have a significant influence on domestic investment in africa. further results revealed that economic growth significantly spurs domestic investment while the exchange rate and the real interest rate are insignificant to domestic investment in africa. table 3.5. economic governance and domestic investment in africa government effectiveness regulatory quantity economic governance pooled ols fixed effects random effects driscoll kraay pooled ols fixed effects random effects driscoll kraay pooled ols fixed effects random effects driscoll kraay constant -1.5102a (0.000) 18.5526a (0.000) -6.0490a (0.000) 18.5526a (0.000) -1.5957a (0.000) 17.4008a (0.000) -5.7341a (0.000) 17.4008a (0.000) -1.6685a (0.000) 17.4535a (0.000) -5.8515a (0.000) 17.4535a (0.000) gdp 0.9944a (0.000) 1.7126a (0.000) 1.1685a (0.000) 1.7126a (0.000) 0.9982a (0.000) 1.6612a (0.000) 1.1563a (0.000) 1.6612a (0.000) 0.9982a (0.000) 1.6612a (0.000) 1.1563a (0.000) 1.6612a (0.000) exc 0.0227c (0.094) 0.0529 (0.595) 0.1149a (0.004) 0.0529 (0.792) 0.0145 (0.274) 0.0750 (0.453) 0.1221a (0.002) 0.0752 (0.683) 0.0146 (0.274) 0.0752 (0.453) 0.1221a (0.002) 0.0752 (0.683) ri 0.00001 (0.997) -0.0001 (0.952) -0.0007 (0.743) -0.0001 (0.858) -0.0001 (0.975) 0.0001 (0.980) 0.0012 (0.613) 0.0001 (0.946) -0.0001 (0.975) 0.0001 (0.947) 0.0012 (0.613) 0.0001 (0.946) ge 0.2086a (0.000) 0.0079 (0.938) 0.1091 (0.274) 0.0079 (0.942) rq 0.1708a (0.001) 0.1239 (0.145) 0.2758a (0.001) 0.1238c (0.082) ecogov 0.0905a (0.001) 0.0656 (0.145) 0.1460a (0.001) 0.0656c (0.082) r2 overall 0.9639 0.9607 0.9556 0.9624 0.9612 0.9549 0.9632 0.9612 0.9549 r2 between 0.9788 0.9727 0.9787 0.9708 0.9787 0.9708 r2 within 0.7561 0.7520 0.7561 0.7586 0.7488 0.7586 0.7586 0.7488 0.7586 f-stat 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 f-test 0.0000 0.0000 0.0000 wald 0.0000 0.0000 0.0000 hausman 0.0000 0.0000 0.0000 b-p lm 0.0000 0.0000 0.0000 m.wald 0.0000 0.0000 0.0000 wooldridge 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 pesaran cd 0.1675 0.0000 0.1675 0.0000 0.2106 0.0000 pesaran scaled lm 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 observations 224 224 224 224 224 224 224 224 224 224 224 224 source: author’s computation. note: dependent variable: domestic investment. gdp is economic growth, ri is the real interest rate, cc is the control of corruption, rl is rule of law, ecogov is economic governance, b-p lm is breusch pagan langragian multiplier test, m.wald is the modified wald test for heteroskedasticity. probability values of coefficients are in parenthesis. diagnosis and post-estimation test results presented are probability values. a, b and c denote statistical significance at 1, 5 and 10%, respectively. on the relationship between economic governance (and its components) and domestic investment, empirical results, as indicated in table 3.5 reveal that government effectiveness chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 74 has a positive but insignificant relationship with domestic investment in africa. regulatory quality has a positive and significant relationship with domestic investment in africa at 10 per cent statistical level of significance and economic governance also has a positive and significant influence on domestic investment. while regulatory quality spurs domestic investment by 0.124 percentage point, economic governance enhances domestic investment by 0.066 percentage point. the result also discloses similar findings from table 3.4 that economic growth significantly spurs domestic investment in africa while the exchange rate and the real interest rate are insignificant in driving domestic investment. table 3.6. institutional governance and domestic investment in africa control of corruption rule of law institutional governance pooled ols fixed effects random effects driscoll kraay pooled ols fixed effects random effects driscoll kraay pooled ols fixed effects random effects driscoll kraay constant -2.0952a (0.000) -17.3095a (0.000) -6.0694a (0.000) -17.3095a (0.000) -1.7981a (0.000) -17.6059a (0.000) -5.8223a (0.000) -17.6059a (0.000) -1.9006a (0.000) -17.6739a (0.000) -5.9708a (0.000) -17.6739a (0.000) gdp 1.0182a (0.000) 1.6554a (0.000) 1.1699a (0.000) 1.6554a (0.000) 1.0065a (0.000) 1.6730a (0.000) 1.1611a (0.000) 1.6730a (0.000) 1.0065a (0.000) 1.6730a (0.000) 1.1611a (0.000) 1.6730a (0.000) exc 0.0309b (0.026) 0.0906 (0.368) 0.1351a (0.000) 0.0905 (0.648) 0.0240c (0.075) 0.0631 (0.542) 0.1251a (0.000) 0.0631 (0.739) 0.0240c (0.075) 0.0632 (0.524) 0.1251a (0.001) 0.0632 (0.739) ri 0.0017 (0.573) -0.0002 (0.926) 0.0006 (0.840) -0.0002 (0.919) -0.0005 (0.879) -0.0003 (0.894) -0.0002 (0.917) -0.0003 (0.890) -0.0005 (0.879) -0.0003 (0.894) 0.0002 (0.917) -0.0003 (0.890) cc 0.2651a (0.000) 0.1554c (0.056) 0.3218a (0.000) 0.1554b (0.034) rl 0.2184a (0.000) 0.1447 (0.113) 0.3164a (0.000) 0.1448b (0.039) instgov 0.1179a (0.000) 0.0781 (0.113) 0.1707a (0.000) 0.0781b (0.039) r2 overall 0.9648 0.9616 0.9574 0.9644 0.9626 0.9576 0.9644 0.9626 0.9576 r2 between 0.9789 0.9731 0.9802 0.9735 0.9802 0.9735 r2 within 0.7604 0.7476 0.7604 0.7591 0.7475 0.7591 0.7591 0.7475 0.7591 f-stat 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 f-test 0.0000 0.0000 0.0000 wald 0.0000 0.0000 0.0000 hausman 0.0000 0.0000 0.0000 b-p lm 0.0000 0.0000 0.0000 m.wald 0.0000 0.0000 0.0000 wooldridge 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 pesaran cd 0.3692 0.0000 0.0590 0.0000 0.0590 0.0000 pesaran scaled lm 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 observations 224 224 224 224 224 224 224 224 224 224 224 224 source: author’s computation. note: dependent variable: domestic investment. gdp is economic growth, ri is the real interest rate, cc is the control of corruption, rl is rule of law, instgov is institutional governance, b-p lm is breusch pagan langragian multiplier test, m.wald is the modified wald test for heteroskedasticity. probability values of coefficients are in parenthesis. diagnosis chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 75 and post-estimation test results presented, are probability values. a, b and c denote statistical significance at 1, 5 and 10%, respectively. finally, from table 3.6 on the relationship between institutional governance and its constituents on domestic investment, empirical outcomes reveal that the control of corruption positively and significantly drives domestic investment in africa, improvements in controlling corruption in africa increases domestic investment by 0.155 percentage point and this relationship is statistically significant at 5 per cent level. rule of law also positively and significantly drive domestic investment by 0.145 percentage point while institutional governance positively and significantly increases domestic by 0.078 percentage point. economic growth from table 3.6 remains a driving factor for domestic investment in africa, while the exchange rate and real interest rate remain insignificant. 4. conclusions and recommendations the quality of governance in africa has been generally understood to be poor as political crises, corruption, absence of the rule of law abound in most countries in the region. it has also been argued that governance performs a central role in explaining the dynamics of macroeconomic aggregates both in developed and developing economies. this study thereby accessed the impact of governance on domestic investment in africa employing unbundled and bundled governance indicators for 16 african countries between 2002 and 2015 and adopting the driscoll and kraay fe model which accounts for most biases in panel econometric modelling. the results revealed that all the indicators of governance employed in this study have positive and statistically significant influences on domestic investment in africa except for government effectiveness which has a positive but insignificant influence on the outcome variable. the results further revealed that voice/accountability and the control of corruption have the most significant impact on domestic investment in africa, as indicated by their coefficient values. it can thus be said that the quality of governance is a significant factor in explaining the level of domestic investment in africa. also, economic growth has a strongly positive and significant impact on domestic investment in the continent. the recommendations of this study are straightforward: african leaders should as a matter of urgency endeavour to improve the quality of governance by ensuring political stability. political stability can be achieved through engagement, empowerment and education of the youths in order to curb any potential civil unrest. youths are specifically mentioned here due to the ‘youth bulge’ existing on the african continent. poverty also has a strong potential for generating political instability. a reduction in povertyis necessary to ensure a more stable political environment in africa. inclusive growth policies must be adopted in order for the poor to participate in the growth process. there also have to be development and implementation of economic development policies in the areas of health, education, nutrition and sanitation, which ultimately leads to the reduction in poverty. moreover, for there to be an increase in domestic chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 76 investment in africa, citizens should be able to partake in choosing their governments as well as enjoy the liberty to express themselves. if citizens can choose the leaders they want, devoid of electoral malpractices, this does not only lead to a more stable political environment but also a more stable economic environment which ensures domestic investment activities. furthermore, african governments should implement sound monetary and fiscal policies that are in line with current economic realities to promote the private sector and increase domestic investment. again, there is a need to curb corruption in africa. this can be achieved through sanctions, i.e. the punishment of corrupt political office holders. public sector reforms are also necessary to improve financial management in the public sector. citizens as well have a role to play by holding their governments accountable. there has to be an improvement in the quality of contract enforcement, property rights and a reduction in crime and violence for domestic investment to increase in africa. economic policies should also be targeted towards improving economic growth in africa, which will precipitate an increase in domestic investment. in conclusion, future studies can assess how the established findings withstand scrutiny within the comparative framework of the washington consensus versus the beijing model. it is important to note that governance indicators employed are prioritised differently by the two contemporary paradigms of development. while the washington consensus prioritises political governance, the beijing model emphasizes economic governance. more insights into the dynamics of the paradigms are apparent in recent literature on the subject (asongu and ssozi, 2016; asongu, 2016b). references ajide, k. (2013). the role of governance on private investment in nigeria: a preliminary analysis. central bank of nigeria economic and financial review, 51(1), 93-119. akanbi, o. (2010). role of governance in explaining domestic investment in nigeria. university of pretoria, working paper 168. algieri, b., & mannarino, l. (2013). the role of credit conditions and local financial development on export performances: a focus on the italian regions, universita della calabria, dipartimento di economia e statistica. retrieved from http://www.ecostat.unical.it/wp2/wp/10_a%20panel%20analysis%20for%20the%20sectoral %20manufacturing%20exports.pdf amaz, c., gaume, v., & lefevre, m. (2012). the impact of training on firm scrap rate: a study of panel data. retrieved from http://wwwperso.gate.cnrs.fr/polome/pages2012_13/panel/lefevre_gaume_amaz_rapport.pdf asiedu, e. (2003). foreign direct investment in africa: the role of government policy, governance and political instability. university of kansas working paper. asongu, s. (2012). on the effect of foreign aid on corruption. economics bulletin, 32(3), 21742180. doi: https://doi.org/10.2139/ssrn.2493289 http://www.ecostat.unical.it/wp2/wp/10_a%20panel%20analysis%20for%20the%20sectoral%20manufacturing%20exports.pdf http://www.ecostat.unical.it/wp2/wp/10_a%20panel%20analysis%20for%20the%20sectoral%20manufacturing%20exports.pdf http://www-perso.gate.cnrs.fr/polome/pages2012_13/panel/lefevre_gaume_amaz_rapport.pdf http://www-perso.gate.cnrs.fr/polome/pages2012_13/panel/lefevre_gaume_amaz_rapport.pdf https://doi.org/10.2139/ssrn.2493289 chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 77 asongu, s. (2016). determinants of growth in fast-developing countries: evidence from bundling and unbundling institutions. politics & policy, 44(1), 97-134. doi: https://doi.org/10.2139/ssrn.2493289 asongu, s. (2016). sino-african relations: a review and reconciliation of dominant schools of thought. politics and policy, 44(2). doi: https://doi.org/10.1111/polp.12152 asongu, s., & nwachukwu, j. (2016). the role of governance in mobile phones for inclusive human development in sub-saharan africa. technovation, 1-13. doi: https://doi.org/10.1016/j.technovation.2016.04.002 asongu, s., & nwanchukwu, j. (2017). fighting capital flight in africa: evidence from bundling and unbundling governance. journal of industry, competition and trade, 17(3), 305-323. doi: https://doi.org/10.1007/s10842-016-0240-1 asongu, s., & ssozi, j. (2016). sino-african relations: some solutions and strategies to the policy syndromes. journal of african business, 17(1), 33-51. doi: https://doi.org/10.1080/15228916.2015.1089614 asongu, s., batuo, e., & tchamyou, v. (2015). bundling governance: finance versus institutions in private investment promotion. african development institute working paper 15/051. doi: https://doi.org/10.2139/ssrn.2713087 asongu, s., tchamyou, v., asongu, n., & tchamyou, n. (2017). fighting terrorism in africa: evidence from bundling and unbundling institutions. empirical economics, 1-51. doi: https://doi.org/10.2139/ssrn.2885742 aysan, a., gaobo, p., & marie-ange, v. (2005). how to boost private investment in the mena countries: the role of economic reforms. topics in middle eastern and north african economics, meea, online journal, 7, 1-15. aysan, a., nabli, m., & veganzones, m. (2011). governance and private investment in the middle east and north africa. hal archives. retrieved from https://halshs.archivesouvertes.fr/halschs-00557250 bader, m., & malawi, a. (2010). the impact of interest rate on investment in jordan: a cointegration analysis. journal of king abdul aziz university: economics and administration, 24(1), 199-209. doi: https://doi.org/10.4197/eco.24-1.6 baek, s., & yang, d. (2010). institutional quality, capital flight and capital flows. korean economic review, 26(1), 121-155. baltagi, b., kao, c., & peng, b. (2016). testing cross-sectional correlation in large panel data models with serial correlation. econometrics, 1-24. doi: https://doi.org/10.3390/econometrics4040044 batina, r. (1998). on the long run effect of public capital and disaggregated public capital on aggregate output. international tax and public finance, 5, 263-281. doi: https://doi.org/10.1023/a:1008626025932 bjuggren, p., dzansi, j., & shukur, g. (2010). remittances and investment. centre for excellence for science and innovation studies working paper 216. caporale, m., raul, c., sova, r., & sova, a. (2009). financial development and economic https://doi.org/10.2139/ssrn.2493289 https://doi.org/10.1111/polp.12152 https://doi.org/10.1016/j.technovation.2016.04.002 https://doi.org/10.1007/s10842-016-0240-1 https://doi.org/10.1080/15228916.2015.1089614 https://doi.org/10.2139/ssrn.2713087 https://doi.org/10.2139/ssrn.2885742 https://halshs.archives-ouvertes.fr/halschs-00557250 https://halshs.archives-ouvertes.fr/halschs-00557250 https://doi.org/10.4197/eco.24-1.6 https://doi.org/10.3390/econometrics4040044 https://doi.org/10.1023/a:1008626025932 chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 78 growth: evidence from ten eu members. diw berlin discussion paper 940. doi: https://doi.org/10.2139/ssrn.1499786 chauvet, l., & collier, p. (2004). development effectiveness in fragile states: spillovers and turnarounds. centre for the study of african economies. department of economics, oxford university (mimeo). chuku, c., onye, k., & ajah, h. (2015). structural and institutional determinants of investment activity in africa. mpra paper 68163. doi: https://doi.org/10.1007/978-3-319-44787-2_2 de hoyos, r., & sarafidis, v. (2006). testing for cross-sectional dependence in panel data models . the stata journal, 6(4), 482-496. doi: https://doi.org/10.1177/1536867x0600600403 driscoll, j., & kraay, a. (1998). consistent covariance matrix estimation with spatial dependent data. review of economic and statistics, 80, 549-560. doi: https://doi.org/10.1162/003465398557825 drukker, d. (2003). testing for serial correlation in linear panel data models. the stata journal, 3(2), 168-177. doi: https://doi.org/10.1177/1536867x0300300206 eregha, p. (2010). interest rate variation and investment determination in nigeria. international business management, 4(2), 41-46. doi: https://doi.org/10.3923/ibm.2010.41.46 fayissa, b., & nsiah, c. (2010). the impact of governance on economic growth: further evidence from africa. journal of developing areas, 47(1), 91-108. doi: https://doi.org/10.1353/jda.2013.0009 globerman, s., & shapiro, d. (2002). global foreign direct investment flows: the role of governance infrastructure. world development, 30(11), 1899-1919. doi: https://doi.org/10.1016/s0305-750x(02)00110-9 hamuda, a., sulikova, v., gazda, v., & horvath, d. (2013). ardl investment model in tunisia. theoretical and applied economics, 20(2), 57-68. hoechle, d. (2007). robust standard errors for panel regressions with cross-sectional dependence. the stata journal, 7(3), 281-312. doi: https://doi.org/10.1177/1536867x0700700301 iheonu, c., & nwakeze, n. (2016). investment, output and real interest rate in nigeria: an ardl analysis. journal of economics and allied research, 1(1), 72-86. jolliffe, i. (2002). principal component analysis (2nd ed.). new york: springer. kaiser, p. (1974). an index of factorial simplicity. psychometrika, 51(4), 804-821. doi: https://doi.org/10.1007/bf02291575 khan, m. (2007). governance, economic growth and development since the 1960s. desa working paper 54. lasker, m., & king, m. (1997). modified wald test for regression disturbances. economic letters, 56(1), 5-11. doi: https://doi.org/10.1016/s0165-1765(97)00128-6 lim, j. (2014). institutional and structural determinants of investment worldwide. journal of macroeconomics, 41, 160-177. doi: https://doi.org/10.1016/j.jmacro.2014.05.007 https://doi.org/10.2139/ssrn.1499786 https://doi.org/10.1007/978-3-319-44787-2_2 https://doi.org/10.1177/1536867x0600600403 https://doi.org/10.1162/003465398557825 https://doi.org/10.1177/1536867x0300300206 https://doi.org/10.3923/ibm.2010.41.46 https://doi.org/10.1353/jda.2013.0009 https://doi.org/10.1016/s0305-750x(02)00110-9 https://doi.org/10.1177/1536867x0700700301 https://doi.org/10.1007/bf02291575 https://doi.org/10.1016/s0165-1765(97)00128-6 https://doi.org/10.1016/j.jmacro.2014.05.007 chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 79 mengistu, a., & adhikary, b. (2011). does good governance matter for fdi inflows? evidence from asian economies. asia pacific business review, 17, 281-299. doi: https://doi.org/10.1080/13602381003755765 morisset, j. (2000). foreign direct investment in africa-policies also matter. transnational corporation, 9(2), 107-125. ngov, p. (2008). governance, foreign direct investment and economic growth. forum of international development studies, 36, 255-278. ouedrago, i., & kouaman, p. (2014). governance and private investment in sub-saharan africa. international journal of african development, 2(1), 5-25. pereira, a. (2000). is all public capital created equal? the review of economic and statistics, 82(3), 513-518. doi: https://doi.org/10.1162/rest.2000.82.3.513 pereira, a. (2001). public capital formation and private investment: what crowds in what? public finance review, 29(1), 3-25. doi: https://doi.org/10.1177/109114210102900101 pesaran, m. (2004). general diagnostic test for cross section dependence in panels. university of cambridge, faculty of economics working papers 0435. romer, d. (2012). advanced macroeconomics (fourth ed.). new york: mcgraw-hill. samani, a., & ariani, f. (2010). governance and fdi in mena region. australian journal of basic and applied sciences, 4, 4880-4882. stock, j., & watson, m. (2008). heteroskedasticity-robust standard errors for fixed effects regression. econometrica, 76, 155-174. doi: https://doi.org/10.1111/j.00129682.2008.00821.x tchamyou, s. (2017). the role of knowledge economy in african business. journal of the knowledge economy, 8(4), 1189-1228. doi: https://doi.org/10.1007/s13132-016-0417-1 the borgen project. (2017). addressing five causes of poverty in mauritius. retrieved from http://borgenproject.org/causes-of-poverty-in-mauritius/ tobin, j. (1969). a general equilibrium approach to monetary theory. journal of money, credit and banking, 1(1), 15-29. doi: https://doi.org/10.2307/1991374 wooldridge, j. (2012). econometric analysis of cross section and panel data. cambridge: mit press. https://doi.org/10.1080/13602381003755765 https://doi.org/10.1162/rest.2000.82.3.513 https://doi.org/10.1177/109114210102900101 https://doi.org/10.1111/j.0012-9682.2008.00821.x https://doi.org/10.1111/j.0012-9682.2008.00821.x https://doi.org/10.1007/s13132-016-0417http://borgenproject.org/causes-of-poverty-in-mauritius/ https://doi.org/10.2307/1991374 chimere okechukwu iheonu / european journal of government and economics 8(1), june 2019, 63-80. 80 appendix table a1: definition of governance variables variable functional definition control of corruption according to the world governance indicator (2017), the control of corruption is the viewpoint of the degree to which public power is used for private gain, comprising both minor and grand forms of corruption, as well as the capture of the state by elites and private interests. government effectiveness according to the world governance indicator (2017), government effectiveness captures perceptions of the quality of public services, the quality of the civil service and the extent of its independence from political forces, the quality of policy design and execution, and the reliability of the government's commitment to such policies. political stability/absence of violence and terrorism according to the world governance indicator (2017), this captures perceptions of the chances of political instability and/or politically-motivated violence, including terrorism. rule of law according to the world governance indicator (2017), the rule of law reveals perceptions of the extent to which agents have confidence in and abide by the guidelines of society, and specifically, the quality of contract enforcement, property rights, courts and the police, as well as the chances of crime and violence. regulatory quality regulatory quality captures perceptions of the capability of the government to design and execute sound policies and rules that allow and promote private sector development (world governance indicator, 2017). voice/accountability according to the world governance indicator (2017), voice and accountability capture insights to the extent to which a country's citizens are able to partake in choosing their government, as well as liberty to express, freedom of association, and a free media. political governance according to asongu et al (2017), this is defined as the election and replacement of political leaders. this index is measured with two indicators, which are political stability/no violence and voice/accountability. economic governance according to asongu et al (2017), economic governance is defined as the formulation and implementation of policies that deliver public commodities. this index is measured with two indicators which are regulatory quality and government effectiveness. institutional governance according to asongu et al (2017), institution governance is defined as the respect by the state and citizens of institutions that govern interactions between them. this index is measured with two variables: corruption-control and the rule of law front81 ejge oficial 8-1 contents81 number 8, issue 1, june 2019 4565 abstract. the study empirically examined the impact of governance on domestic investment in 16 african countries with a balanced panel data set, between the years 2002 and 2015. the study employed six unbundled governance indicators from the world ban... keywords. governance; domestic investment; africa; principal component analysis; fixed effects model. jel codes: c1, e2, r5 doi. https://10.17979/ejge.2019.8.1.4565 institutional quality and the wealth of autocrats european journal of government and economics 6(2), december 2017, 106-125. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 institutional quality and the wealth of autocrats christopher j. boudreauxa, *, randall holcombeb a texas a&m international university, u.s.a. b department of economics, florida state university, u.s.a. * corresponding author at: texas a&m international university, laredo, tx 78041. chrisboudreaux1@gmail.com article history. received 14 december 2015; first revision required 9 march 2016; accepted 5 july 2017. abstract. one frequently given explanation for why autocrats maintain corrupt and inefficient institutions is that the autocrats benefit personally even though the citizens of their countries are worse off. the empirical evidence does not support this hypothesis. autocrats in countries with low-quality institutions do tend to be wealthy, but typically, they were wealthy before they assumed power. a plausible explanation, consistent with the data, is that wealthy individuals in countries with inefficient and corrupt institutions face the threat of having their wealth appropriated by government, so have the incentive to use some of their wealth to seek political power to protect the rest of their wealth from confiscation. while autocrats may use government institutions to increase their wealth, autocrats in countries with low-quality institutions tend to be wealthy when they assume power, because wealthy individuals have the incentive to use their wealth to acquire political power to protect themselves from a potentially predatory government. keywords. institutional quality; wealth; autocrats jel classification. p51; o50; h11 1. introduction an important question in economic development is why autocrats in poor countries maintain low-quality economic institutions when a substantial body of literature concludes that specific institutional improvements can produce higher economic growth rates and higher per capita incomes. one common answer is that while the general population may be better off with higher-quality economic institutions, the political elite is better off maintaining more oppressive economic institutions that benefits it, even though those institutions make their countries poorer. referring to the poor states of africa, ayittey (2008: 168) says, “african despots are loath to relinquish control or power. they would rather destroy their economies and countries than give up economic and political power. this power allows them to allocate or extract resources to build personal fortunes and to dispense patronage to buy political support.” acemoglu and robinson (2006: 115) say it is “…because superior institutions and technologies may reduce their political power and make it more likely that they will be replaced.” acemoglu and robinson (2013) emphasize the importance of taking into account both economic and political factors http://www.ejge.org/ mailto:chrisboudreaux1@gmail.com 107 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 when looking at the potential for economic reform, and offer examples that illustrate the peril of not doing so. if autocrats benefit from maintaining low-quality economic institutions, an important question is what benefit autocrats actually receive. the few systematic empirical studies that have been done have not found evidence that the political elite receive any benefit from low-quality economic institutions. this paper uses a small dataset that looks directly at autocrat wealth to see if it is positively impacted, and finds no evidence that it is. the evidence indicates that if anything, autocrats benefit from overseeing institutional improvements in their countries, and one would conjecture that if this research is correct, autocrats themselves would be in a good position to recognize the benefits they could reap from improving institutions. a plausible explanation for why they do not improve institutions in their countries is that institutional change is difficult to implement and cannot be accomplished by a single individual, even if that individual wants to improve institutional quality and has substantial political power. the casual observation that poor countries tend to be ruled by wealthy autocrats may have led researchers to consider the wrong question. wealthy autocrats tend to be wealthy when they assume power, so a more appropriate question is why wealthy individuals in countries with inefficient and corrupt institutions seek political power when the wealthy do not typically do so in countries with higher-quality institutions. the wealthy in the united states, western europe, and japan, for example, tend to focus on their economic activities rather than seek political office.1 the observation of wealthy autocrats in poor countries is consistent with the hypothesis that when institutional quality is poor, the wealthy use some of their wealth to seek political power to protect the rest of their wealth from confiscation by the political elite. wealthy autocrats hold political power in poor counties not because those institutions enable them to enhance their wealth, but rather because those governments tend to be more predatory, so the wealthy seek political power to protect their wealth from government predation. 2. evidence on the benefits to autocrats despite the conventional wisdom that autocrats receive personal benefits from maintaining lowquality economic institutions at the expense of the well-being of their citizens, there is relatively little evidence to support that conclusion. as noted in the introduction, ayittey (2008) offers anecdotal evidence regarding african dictatorships. acemoglu and robinson (2006) develop a theoretical framework in which policies that can produce economic development for the general population can also impose costs on political elites, causing them to choose to inhibit economic development for their own benefit. they offer historical examples of czarist russia and the nineteenth century austro-hungarian empire to support their case, but also show, using the united states, germany, and britain as examples, that political elites sometimes implement development-enhancing institutions. acemoglu and robinson (2013) look at destabilizing 1 donald trump is a notable exception here, but trump is different from most other wealthy americans in that he was a television personality prior to running for office, a profession that rewards a high-profile public image. unlike president trump, most wealthy individuals try to keep a low profile and remain out of the public eye rather than seeking personal recognition, as trump did well before he ran for public office. 108 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 economic reforms in ghana and sierra leone to show how improving economic institutions can work to the detriment of political elites. this evidence is anecdotal: a theoretical framework with case studies that provide examples consistent with the model. anecdotes can also be found that lean the other way. both augusto pinochet in chile and park chung-hee in south korea were autocrats who held power for 16 years in their countries and implemented substantial economic reform, which provided benefits to both the autocrats and their citizens.2 examples can be cited both for countries in which autocrats ruled countries that remained poor and with low-quality institutions and cases in which autocrats improved institutional quality and oversaw institutional improvement and increases in income. if better institutions create wealth, there should be a way to distribute that wealth so that everyone benefits, although rodrik (2014) notes that it can be difficult to design institutional reforms that guarantee that everyone will share in the gains. anecdotes can be chosen to support various conclusions. evidence on the issue is mostly anecdotal and we are aware of only two studies that look at a cross-section of countries for evidence that political elites benefit personally from maintaining low-quality economic institutions that reduce the standard of living for their citizens. holcombe and rodet (2012) use government spending as a proxy for benefits to political elites under the assumption that while those elites cannot appropriate all that government spends, they control who gets the benefits of that spending. they look at a set of more than 100 countries with all types of governments – not just autocracies – to see if lower-quality economic institutions increase the size of government, both in real per capita terms and as a share of total income. they find that better economic institutions are associated with a larger government share, which suggests that political elites could benefit from improved economic institutions. political elites have control over more resources when economic institutions are improved, calling into question the conventional wisdom that political elites benefit from maintaining low-quality economic institutions. looking at the claim of acemoglu and johnson (2006) that institutional reforms may destabilize the incumbent regime, holcombe and boudreaux (2013) examine a dataset of 80 autocrats and find that the greater the increase in the quality of economic institutions under their tenure, the longer their tenure, indicating that autocrats can increase their hold on power by improving institutional quality. the few studies that analyze a dataset of multiple countries rather than anecdotal evidence and case studies conclude that autocrats can improve their control over resources and their hold on power by implementing higher-quality economic institutions, calling into question the hypothesis that autocrats maintain poor institutions because they benefit at the expense of their citizens. 2 another possibly relevant example is mikheil saakashvili’s presidency of the republic of georgia from 2004-2013. although saakashvili was elected, georgia was poor and its government was corrupt and inefficient when he came to power. substantial economic reforms led georgia to prosper relative to its economic status prior to his presidency and when compared with other former soviet republics in the caucuses. as president, saakashvili was also accused of appropriating property without compensation and cronyism during his term in office. this is an example of a political leader who improved economic institutions both for the benefit of his citizens and himself and his cronies as an elected leader. 109 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 this study examines additional evidence on the relationship between autocrat wealth and the quality of economic institutions. the evidence here is more direct than in previous studies, although it is from a limited dataset on the wealth of 35 autocrats. there is no comprehensive data available on the wealth of autocrats, so this study relies on data collected from various sources on that wealth. recognizing the limitations of the data, the findings here are conclusive. rather, they offer some additional evidence that calls into question the hypothesis that autocrats benefit personally from maintaining low-quality economic institutions. the results that follow are consistent with the hypothesis that in countries with low-quality institutions, wealthy individuals are attracted to political power, perhaps to protect their wealth from predation that can occur in countries that have a poor record of protecting property rights and establishing rule of law. 3. data on the wealth of autocrats and institutional quality table 1 lists 35 autocrats along with their countries, the time periods in which they were in power, and their wealth. data sources are listed in appendix 1. the dataset is limited because these were the only autocrats for which we were able to find estimates of their wealth. wealth information was found for leaders of countries who ruled in 2010. in a few cases, leaders from previous years were also included in the analysis. khaleda zia of bangladesh exited office in 2006, mobutu sese seko of the democratic republic of congo exited in 1997, and mohamed suharto of indonesia exited in 1998. for these leaders, wealth estimates are adjusted for inflation and stated in 2010 u.s. dollars. included in the list are several monarchs who have limited power over their governments but who have retained power that is largely symbolic. while they would not ordinarily be classified as autocrats, they make a good comparison group as wealthy heads of state in countries that have relatively high-quality economic and political institutions. this allows a comparison of the effects of institutional quality that includes a wider range of institutions, rather than just lowquality institutions. the quality of economic institutions was measured using the fraser institute’s economic freedom of the world (efw) index, updated annually by gwartney, lawson, and hall (2012). the efw index is frequently used as a measure of the quality of economic institutions. it is designed to quantify the level of economic freedom that exists in a country, and is deliberately designed to quantify the quality of economic institutions but to leave out measures of political freedom, such as civil liberties or democratic government. thus, it is a good measure of institutional quality to use for the examination of the hypothesis that autocrats can increase their wealth by imposing economically unproductive institutions. 110 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 table 1. list of leaders. country leader years $ wealth † angola jose eduardo dos santos 1979-present 31 billion azerbaijan ilham aliyev 2003-present 500 million bahrain hamad ibn isa al khalifa 2002-present 5.44 billion bangladesh khaleda zia 1991-1996; 2001-2006 220 million cameroon paul biya 1982-present 200 million china wen jiabao 2003-present 2.7 billion colombia juan manuel santos 2010-present 215 million dem. rep. of congo mobutu sese seko 1970-1997 6.9 billion egypt hosni mubarek 1981-2011 40 billion ethiopia girma wolde-giorgis 2001-present 1.5 million gabon ali bongo ondimba 2009-present 1 billion indonesia mohamed suharto 1967-1998 21.8 billion iran ali khamenei 1981-present 30 billion kazakhstan nursultan nazarbayev 1991-present 1.11 billion kuwait h.h sheikh sabah al-ahmad al-jaber al-sabah 2006-present 14.23 billion montenegro milo đukanović 2003-2006; 2008-2010 15 million morocco mohammad iv 1999-present 2.5 billion myanmar than shwe 1992-2011 4 billion netherlands beatrix 1980-2013 200 million nigeria goodluck jonathan 2010-present 2.8 million norway harald v 1991-present 17.3 million oman qaboos bin said al said 1970-present 9.33 billion pakistan asif ali zardari 2008-present 1.8 billion philippines benigno noynoy aquino iii 2010-present 1.3 million qatar hamad bin khalifa 1995-present 10.88 billion saudi arabia abdullah bin abdulaziz 2005-present 21 billion spain juan carlos i 1975-2014 5 billion sweden carl xvi gustaf 1973-present 42 million syria bashar al-assad 2000-present 1 billion thailand bhumibol adulyadej 1946-present 30 billion tunisia zine el abidine ben ali 1987-2011 7.8 billion u.a.e. sheikh khalifa bin zayed al nahyan 2004-present 21 billion uganda yoweri musevini 1986-present 1.7 billion united kingdom elizabeth ii 1952-present 450 million zimbabwe robert mugabe 1987-2017 10 billion note † wealth amounts ending before 2010 are converted into 2010 dollars. 111 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 literature reviews by berggren (2003) and de haan, lundstrom, and sturn (2006) show that countries with higher-quality economic institutions as measured by the efw index, have higher per capita incomes, and countries that improve their economic freedom as measured by the index have higher rates of economic growth. subsequent studies, such as faria and montesinos (2009), have reaffirmed the positive impact of market institutions. countries with higher quality institutions, as measured by the efw index, are more prosperous. this makes the efw index the ideal measure of the quality of economic institutions for present purposes, because of the many studies that have shown it is positively correlated with per capita income and income growth. while there is little doubt that higher-quality economic institutions improve the economic well-being of a country, the question examined here is whether autocrats who maintain low-quality institutions to the detriment of their citizens receive personal benefits in the form of higher personal wealth. it is a more direct measure of personal benefit than was used in any of the previous studies. 4. institutional quality and the wealth of autocrats the hypothesis that autocrats maintain low-quality institutions to increase their wealth is tested directly in an ols model by looking at the correlation between autocrat wealth and institutional quality. the dependent variable is the log of the wealth of each political leader, w, and the major independent variables are measures of the quality of economic and political institutions at the beginning of the leaders’ tenure and the average annual change in institutional quality during their tenures. economic institutions are measured by the efw score at the beginning of the leaders’ tenures, initialefw, and the average annual change in the efw measure, efwrate, over the leaders’ tenures. efwrate is calculated by subtracting initialefw from the final efw value and dividing by the number of years the leader held power. four different measures are used for the quality of political institutions, which are measured the same way. initial political rights are represented as initialpl, and the change over the autocrats’ tenures is plrate. the measures used are the polity 2 index, the freedom house index of political freedom, and the two individual components of the freedom house index: political rights, and civil rights. initialefw and initialpr are variables outside the control of the autocrat, but they could influence the autocrat’s wealth. efwrate and plrate are variables that, according to the hypothesis being examined, are under the autocrat’s control. the hypothesis predicts that autocrats choose poor institutions for their own benefit, but they could not choose those institutions if institutional quality is outside their control. the hypothesis being examined assumes a direction of causation that runs from efwrate and plrate to autocrat wealth. other factors could also affect autocrat wealth. most obviously, people in wealthier countries tend to be wealthier, so per capita income when the autocrat comes to power, initialpci, and the change in per capita income over the autocrat’s tenure, pcirate, are included. some scholars, 112 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 such as diamond (1997) and sachs (2001), have suggested that natural resource endowments can affect incomes and institutional quality, so a measure of resource rents, r, is included in the empirical model. easterly (2006) and coyne (2013) have suggested foreign aid is often counterproductive to the countries that receive it, although it might benefit a country’s political elite, prompting the inclusion of foreign aid, f, in the empirical model. another geographic factor often associated with a state’s economic well-being is distance from the equator, d. following holcombe and boudreaux (2013), the length of tenure, t, of an autocrat might be related to the autocrat’s wealth, and a binary variable, m, is included which is 1 for autocrats who are monarchs and 0 otherwise.3 appendix 2 shows variable definitions and sources. the empirical specification is w = α + β1initialefw + β2efwrate + β3initialpci + β4pcirate + β5r + β6f + β7d + β8t + + β9m + β10initialpi + β11pirate + ε. the empirical results for eight different empirical specifications are reported in table 2. the specifications differ only in that four of them include only efwrate, not, initialefw, and different measures are used to measure the differences in political institutions. in the eight different specifications, initialefw is never statistically significant at the 10 percent level or better, while efwrate is positive and statistically significant at the 10 percent level in three of the four specifications which omit initialefw. the results in table 2 indicate that the quality of economic institutions at the time an autocrat takes office are unrelated to autocrat wealth, but there is weak evidence that wealth is higher when the quality of economic institutions improves over the autocrat’s tenure. consistent with previous studies, those results show that if anything, autocrats benefit from improvements in the quality of economic institutions during their tenures. initial per capita income is statistically significant at the 10 percent level in half of the regressions, and the autocrat’s length of tenure is statistically significant in all of them. autocrats who rule over higher-income countries tend to have higher levels of wealth, and autocrats who retain power longer tend to have higher wealth. the direction of causation of the tenure variable is uncertain. wealthier autocrats might have an advantage in maintaining power, but it also may be that longer tenures allow more time to accumulate wealth. the wealth variable measures wealth at the end of the autocrat’s tenure, and data limitations do not allow calculating the change in the autocrat’s wealth over the autocrat’s tenure. perhaps the tenure variable is significant because autocrats with longer tenures tend to come to power at a younger age. to check for this, the age that autocrats assumed power was substituted for tenure, and the age variable was significant. however, when both age and tenure were included in the same regressions, tenure remained significant but age did not. thus, we report the specifications than include only the tenure variable. 3 all specifications were also run without the monarchy variable and the institutional variables were qualitatively the same, so for space considerations the paper does not report the results without the monarchy variable. 113 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 table 2. effect of economic and political institutions on the wealth of leaders. log wealth (1) (2) (3) (4) (5) (6) (7) (8) initial efw -0.357 -0.275 -0.510 -0.536 (-1.08) (-0.69) (-1.33) (-1.27) efw rate 4.866 1.753 8.139+ 5.578 9.924+ 5.138 9.713+ 4.501 (1.19) (0.42) (1.74) (0.98) (1.80) (0.88) (1.86) (0.74) initial pci† 0.0220 0.0410 0.0688* 0.0817* 0.0668 0.0912* 0.0548 0.0813+ (0.86) (1.32) (2.17) (2.13) (1.70) (2.09) (1.38) (1.79) pci rate† -0.0864 -0.0647 -0.128 -0.0992 -0.206 -0.168 0.0907 0.0834 (-0.27) (-0.19) (-0.51) (-0.36) (-0.60) (-0.46) (0.22) (0.19) resource rents 0.0317 0.0314 0.00305 0.00442 0.0104 0.0120 0.0184 0.0215 (1.30) (1.31) (0.15) (0.20) (0.41) (0.46) (0.64) (0.72) foreign aid -8.823 -11.30 -9.652 -11.33 -8.410 -12.12 -9.650 -13.18 (-1.23) (-1.47) (-0.81) (-0.88) (-0.77) (-1.02) (-0.77) (-0.98) distance† 0.101 0.111 0.0573 0.0722 0.0280 0.0496 -0.0300 -0.0121 (0.47) (0.50) (0.29) (0.35) (0.13) (0.22) (-0.14) (-0.05) tenure 0.0433+ 0.0425+ 0.0685** 0.0670** 0.0591* 0.0565* 0.0941** 0.0879** (2.00) (1.98) (2.89) (2.99) (2.39) (2.49) (3.10) (3.18) initial polity2 -0.181* -0.177* (-2.49) (-2.37) polity2 rate 7.728** 7.206* (2.80) (2.57) monarchy 1.064 1.026 0.521 0.545 0.650 0.655 0.702 0.754 (1.38) (1.33) (0.50) (0.52) (0.68) (0.69) (0.59) (0.66) initial freedom house -1.140*** -1.109*** (-5.62) (5.01) freedom rate -11.90+ -10.49 (2.03) (1.57) initial political rights -0.926*** -0.879** (3.78) (3.38) political rights rate 1.906 1.511 (-0.19) (-0.15) initial civil rights -1.024** -0.942* (3.07) (2.61) civil rights rate -7.338 -7.686 (-1.11) (-1.24) constant -2.836* -0.691 -7.579*** -5.864* -6.809** -3.572 -7.552** -3.972 (-2.38) (-0.35) (-5.02) (-2.26) (-3.26) (-1.25) (-3.47) (-1.16) adj. r2 0.56 0.56 0.55 0.54 0.46 0.47 0.37 0.37 note. n=35 observations. t statistics in parentheses. + p<0.10* p<0.05 ** p<0.01 *** p<0.001 114 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 one possible area of concern in the list of autocrats used here is that they represent governments with vastly different government structures. the list includes the queen elizabeth ii of great britain and northern ireland and former president robert mugabe of zimbabwe, for example. the inclusion of these measures of political institutions should mitigate those differences by taking into account the political structure. as a robustness check, the regressions were run without elizabeth ii, and without assad of syria, and the results are essentially unaffected.4 monarchs were also separated out with a binary variable, and table 2 shows that the monarchy variable was never statistically significant. the quality of political institutions is controlled for in the same way as economic institutions. the first two regressions use the polity2 index to control for political institutions. the index rates the degree of democratic control of government, with values ranging from -10 for the least democratic institutions to 10 for the most democratic. in the first two regressions, the negative sign on initial polity2 shows that the less democratic are the initial political institutions, the higher is the wealth of the autocrat. the positive sign on polity2 rate shows that movements toward more democratic institutions during the autocrat’s tenure are associated with higher autocrat wealth. while the quality of economic institutions appears to have weak effects on autocrat wealth, as political institutions do. the less democratic a state’s institutions when the autocrat assumes power, the more wealthy the autocrat, and the more the country moves toward democratic institutions during the autocrat’s tenure, the higher the wealth of the autocrat. this suggests a benefit to the autocrat from moving toward more democratic institutions. the remaining six regressions use the freedom house index of political freedom as a robustness check on this result. the freedom house index measures both civil rights and political rights—the degree to which citizens have control over their governments—as components of political freedom. regressions 3 and 4 replace the polity2 index with the freedom house index of political freedom, and while the initial level of political freedom is negative and significant, consistent with the polity2 results, freedom rate, which is the average annual change in political freedom over the autocrat’s tenure is not statistically significant. the freedom house index looks at political institutions more broadly than polity2, and the political rights component is more consistent with the democratic control measured by polity2 than is the entire index. regressions 5 and 6 include only the political rights component of the freedom house index. the results are similar to those using the entire index. initial political rights show up negative and significant, as with the entire index, the political rights rate is not statistically significant. regressions 7 and 8 include only the civil rights component of the freedom house index, and yield the same qualitative result. 4 assad was deleted as a robustness check because there is some question about the actual amount of his wealth. 115 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 table 3. effect of economic and political institutions on the wealth of leaders (ordered probit model). wealth (1) (2) (3) (4) (5) (6) (7) (8) initial efw -0.0967 0.404 -0.217 0.319 -0.203 0.390 -0.227 0.277 (-0.54) (1.56) (-1.14) (1.21) (-1.14) (1.29) (-1.21) (1.03) efw rate 20.64** 22.08*** 26.95*** 22.08*** (2.93) (3.68) (3.32) (3.58) initial pci† 0.0380+ 0.0135 0.0870** 0.0681* 0.0716** 0.0712* 0.0774** 0.0543* (1.95) (0.50) (2.95) (2.33) (2.60) (2.48) (2.92) (1.97) pci rate† -0.302 -0.125 -0.313 -0.174 -0.449 -0.061 -0.431 -0.164 (-0.91) (-0.38) (-0.93) (-0.54) (-1.12) (-1.02) (-1.14) (-0.48) resource rents 0.0102 0.0291 0.00521 0.0208 0.00248 0.0202 0.0112 0.0252 (0.62) (1.57) (0.34) (1.28) (0.16) (1.25) (0.67) (1.53) foreign aid -8.657+ -7.829 -8.866+ -8.276 -9.496+ -9.218 -9.047+ -9.018 (-1.78) (-1.26) (-1.66) (-1.19) (-1.73) (-1.23) (-1.73) (-1.25) distance† -0.00798 -0.149 -0.0330 -0.150 -0.0255 -0.185 -0.0744 -0.180 (-0.07) (-1.07) (-0.25) (-1.08) (-0.21) (-1.28) (-0.58) (-1.45) tenure 0.0150 0.0221 0.0380** 0.0458** 0.0212+ 0.0313* 0.0309* 0.0452** (1.19) (1.60) (2.66) (2.63) (1.65) (2.00) (2.16) (2.95) initial polity2 -0.0900+ -0.0982* (-1.94) (-2.12) polity2 rate 2.528 3.189* (1.58) (2.03) monarchy 1.006* 1.482* 1.164** 1.410** 0.895* 1.390** 1.156** 1.323** (2.26) (2.51) (2.61) (2.92) (2.21) (3.05) (2.71) (2.83) initial freedom house -0.567*** -0.672*** (-3.45) (4.30) freedom rate 5.073 2.386 (-1.47) (-0.71) initial political rights -0.392** -0.572*** (2.81) (4.06) political rights rate 1.323 3.715 (-0.32) (-0.70) initial civil rights -0.409* -0.547** (2.32) (3.06) civil rights rate -8.781* -6.811 (-2.40) (-1.24) lr χ2 27.94 40.18 32.10 44.36 25.92 43.57 25.22 34.25 note. n=35 observations. t statistics in parentheses. + p<0.10 * p<0.05 ** p<0.01 *** p<0.001. † in thousands. dependent variable is ordered in four categories. 116 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 in all regressions, whether the polity2 index or variants of the freedom house index are used to quantify the quality of political institutions, lower-quality political institutions are associated with higher levels of autocrat wealth. autocrat wealth tends to be higher when autocrats come to power in countries that are less democratic, where citizens have fewer political rights and lower levels of civil rights. assuming that political freedom and more democratic control of government are good things, the lower the quality of political institutions when an autocrat assumes power, the higher will be the autocrat’s wealth. regardless of how the quality of those political institutions is measured, the coefficients are always significant at the 1 percent level. as a robustness check on these results, autocrat wealth was respecified into four categories: 1: those with wealth under $100 million; 2: wealth between $100 million and $1 billion; 3: wealth between $1 billion and $10 billion; and 4: wealth greater than $10 billion. ordered probits were run using the same eight specifications as in table 2, and those results appear in table 3. the results are mostly unchanged, but with two differences worth noting. the polity2 rate is not significant in the ordered probit, but more noteworthy, efwrate is strongly significant and positive in all four specifications in which it is included. autocrats are wealthier in countries that have improving economic institutions under their rule. consistent with holcombe and rodet (2012) and holcombe and boudreaux (2013), this suggests that an increase in the quality of economic institutions makes autocrats better off, which calls into question the conventional wisdom that autocrats benefit from maintaining lowquality institutions. consistent with the results in table 2, the measures of initial political institutions all remain negative and highly statistically significant. also note that, in contrast to table 2, the monarchy variable is positive and statistically significant in all of the regressions in table 3. while this indicates that monarchs tend to be wealthier than other autocrats, the institutional variables are qualitatively the same in both tables. table 4 shows a correlation matrix with all of the economic and political institutions variables. the table shows a positive, although weak, correlation between the initial efw and all of the initial political variables.5 because the measured quality of economic institutions and the measured quality of political institutions are positively correlated, what appears to be the effect of economic institutions may actually be the result of political institutions. people have observed wealthy autocrats overseeing countries with low-quality economic institutions and inferred that those autocrats benefit from maintaining low-quality institutions, but this evidence points to a different conclusion: countries with lower-quality political institutions tend to attract wealthier autocrats. the positive correlation between the quality of political and economic institutions means that those wealthy autocrats will also be ruling countries with low-quality economic institutions. autocrat wealth is negatively correlated with the quality of political institutions when they assume power. 5 also note that there tends to be a negative correlation between the initial variables and the rate variables. this may indicate convergence in institutional quality, although the correlation coefficients are small enough in magnitude that one would hesitate to draw firm conclusions from them. 117 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 table 4. correlation matrix. initial efw efw rate initial polity 2 polity2 rate initial freedom house freedom rate initial civil rights civil rights rate initial political rights political rights rate initial efw 1 efw rate -0.530 1 initial polity 2 0.179 -0.140 1 polity2 rate -0.435 0.257 -0.277 1 initial freedom house 0.294 -0.154 0.784 -0.216 1 freedom rate 0.125 0.156 0.246 0.021 -0.072 1 initial civil rights 0.314 -0.167 0.752 -0.165 0.932 0.031 1 civil rights rate -0.181 0.285 0.214 0.236 0.048 0.020 -0.066 1 initial political rights 0.272 -0.044 0.828 -0.234 0.932 0.138 0.924 -0.059 1 political rights rate -0.043 0.023 -0.041 0.339 -0.122 0.121 -0.091 0.54 -0.253 1 if there is any correlation between their wealth and a change in the quality of economic institutions, it is that improving economic institutions during an autocrat’s rule is associated with higher autocrat wealth. these results are inconsistent with the hypothesis that autocrats maintain low-quality institutions because they benefit from them. the empirical results have implications with regard to both economic institutions and political institutions, which are discussed in the next two sections. 5. autocrat wealth and economic institutions the paper’s introduction notes the conventional wisdom that autocrats in countries with lowquality economic institutions do not improve those institutions because the political elite benefit from the low-quality institutions even though those institutions lower the economic well-being of the general population. empirical evidence for that hypothesis is anecdotal, and an important question is: if the political elite benefit from maintaining low-quality economic institutions, what benefit does it receive? the only articles that look at a cross-section of countries to evaluate that hypothesis do not support it. holcombe and rodet (2012) find that the political elite can increase the resources under its control by improving the quality of economic institutions, and holcombe and boudreaux (2013) find that improving the quality of economic institutions lengthens the tenure of autocrats and stabilizes their political power. this paper finds that if anything, improving the quality of economic institutions is associated with higher autocrat wealth, and there is no evidence in any of these studies of any benefit to the political elite from maintaining low-quality economic institutions. the evidence in this paper is based on a limited dataset, and certainly is not conclusive, but it does contribute some additional evidence to the literature that finds no benefit to autocrats from maintaining low-quality institutions, and finds that if anything, autocrats can improve their 118 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 well-being in several dimensions by overseeing institutional improvements. applied narrowly to the conventional wisdom, the question raised by this literature is, what is the benefit the political elite receive from maintaining low-quality institutions? perhaps different data or empirical methods will reveal some benefit, but at this point, the literature has identified no benefit to the political elite from maintaining low-quality economic institutions. there is room for additional research on this question, either to identify the benefit the political elite gains from maintaining low-quality economic institutions, or to provide additional evidence supporting the studies that, at this point, all indicate that the political elite do not benefit from keeping their countries poor. this empirical result raises the question of why, if autocrats would benefit personally from improving institutional quality, they do not make those improvements. one would think that autocrats would know their own interests at least as well as academics. one reason might be that institutional change is difficult to implement, so even if autocrats see the benefit from doing so, they may not be able to actually accomplish institutional reforms. some autocrats—augusto pinochet and park chung-hee were mentioned earlier—have done so, while others have not. the idea that autocrats see advantages to institutional reform but are unable to accomplish it is a conjecture that goes beyond the analysis in this paper, so is a topic for future research. with regard to economic institutions, the major question this paper prompts is what benefit, if any, autocrats receive from maintaining low-quality institutions. thus far, the literature has provided no systematic evidence that there are any benefits. 6. autocrat wealth and political institutions the empirical results in this paper consistently find a strong negative correlation between the quality of political institutions at the time an autocrat assumes power and the wealth of autocrats. wealthier autocrats tend to come to power in those countries in which the political institutions are poorest. a possible explanation for this finding is that in countries with low-quality political institutions, wealthy individuals seek political power to protect their wealth. winters (2011: 7) notes, “in systems where property is reliably defended externally (especially by an armed state through institutions and strong property rights and norms), oligarchs have no compelling need to be armed or engaged directly in political roles.” bill gates and warren buffett do not need to seek direct political power (although they do lobby to attempt to turn political outcomes their way) because political institutions in the united states protect their wealth, but in countries with lower-quality political institutions, winters argues that the economic elite has more of an incentive to secure political power directly to prevent the political elite from appropriating its wealth.6 this suggests that the conventional wisdom may be based on a mistaken interpretation of the observation that political leaders in countries with low-quality institutions tend to be wealthy. 6 winters (2011: 6) defines oligarchs as “…actors who command and control massive concentrations of material resources that can be deployed to defend or enhance their personal wealth and exclusive social position.” an example that illustrates the benefit of the wealthy assuming political power to protect their wealth is mikhail khodorkovsky of russia, whose oil wealth at one time made him russia’s wealthiest individual, but was jailed for 10 years by putin and had his assets appropriated by the state. 119 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 the observation of wealthy autocrats ruling over poorer countries suggests the conventional wisdom that those autocrats benefit from the low-quality institutions, but another possibility is that the wealthy seek political power to protect their wealth from predation by the political elite in countries that have a poor record of protecting property rights and maintaining rule of law. the elite does not benefit from maintaining low-quality institutions, but rather seeks political power to protect itself from being the victim of political predation under those institutions. where political institutions are weak, wealth is more at risk to be confiscated by the political elite, so the wealthy have an incentive to convert some of its wealth into political power, to protect the rest. we do not doubt that once it gains political power, the political elite will use that power to further its own interests, but examples of corrupt autocrats using political power to their advantage do not speak directly to the conventional wisdom. the question is whether the political elite would benefit more from maintaining low-quality institutions or from initiating institutional improvements, and all of the evidence so far—including the evidence in this paper— indicates that the political elite would be better off by overseeing improvements in institutional quality. 7. conclusions a substantial literature demonstrates that states with economic institutions that protect property rights and support market exchange prosper, while those with poor institutions that inhibit market activity remain poor. mokyr (1990) and landes (1998) give persuasive historical evidence, and olson (1996) notes that profit opportunities do not remain unexploited for long. poor countries remain that way because their economic institutions prevent innovative and entrepreneurial acts from being profitable. entrepreneurial individuals are led to predatory rather than productive activity in countries with low-quality economic institutions, baumol (1990) notes. gwartney, lawson, and hall (2012) identify and quantify those institutions that lead to prosperity. if economists know what economic institutions lay a foundation for growth and prosperity, an important question for world economic development is why poor countries do not adopt those institutions. a commonly-given answer is that even though most of the population would benefit from institutional reform, the political elite benefits from maintaining those lowquality institutions, even as they impose costs on most people in their countries. this paper has two main conclusions. first, it offers support for a small empirical literature which finds that autocrats do not benefit from maintaining low-quality institutions. autocrats who maintain low-quality institutions do not increase their wealth as a result, and this finding is consistent with an existing literature finding that if anything, autocrats will be better off if they oversee improvements in institutional quality. the empirical evidence indicates that the hypothesis that autocrats in poor countries maintain low-quality political institutions because they benefit is incorrect. second, the paper finds that there is a negative correlation between the quality of political institutions and the wealth of autocrats at the time when they assume power. rather than asking 120 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 why autocrats maintain low-quality institutions, this result suggests a different question, which is why countries with lower-quality political institutions attract wealthier autocrats. a possible explanation for this is that high-quality political institutions protect individual wealth, so wealthy individuals in countries with high-quality institutions do not need to assume positions of political power to retain their economic power. wealthy autocrats tend to assume power in countries with low-quality political institutions. they can use their wealth to try to buy political power, and then use their political power to try to protect and enhance their wealth. an important question is whether maintaining low-quality economic institutions furthers those goals, and this paper supports a recent literature that says it does not. the results in this paper go a good distance toward understanding why even though wealthy autocrats rule countries with poor institutions, they do not benefit from maintaining low-quality institutions. when wealthy autocrats gain political power, they will use it to enhance and protect their wealth. we do not disagree with the conventional wisdom on this point. the question is whether it is to their advantage to maintain low-quality economic institutions, and this paper provides additional evidence to support the existing literature which says they do not. political elites tend to be wealthier in countries with low-quality institutions not because they benefit from keeping institutional quality low, but because where institutional quality is low the wealthy have a greater incentive to seek political power. any empirical evidence on a hypothesis is always tentative. the questions addressed here are important for economic development and to date, the small amount of empirical research has not found evidence that low-quality institutions benefit the political elite. more research would be welcome, either to reinforce the tentative conclusion that the political elite does not benefit from maintaining low-quality economic institutions, or to provide evidence that when other data are examined or other methods are used, there is indeed evidence to support the hypothesis that autocrats receive personal benefits from maintaining these institutions. it appears that autocrats in countries with low-quality institutions tend to be wealthier because the wealthy have a greater incentive to seek political power in those countries, to protect their wealth, but this tentative conclusion would benefit from further research. important implications for development policy turn on the question. one would be hardpressed to argue that autocrats do not understand what is in their own interest. if the conclusion that autocrats do not benefit from maintaining low-quality economic institutions holds up to further scrutiny, this would suggest that outside agencies and foreign governments can look for ways to help autocrats implement institutional reform. if the conventional wisdom is correct, then there is more of an argument for pushing for regime change and perhaps broader political reforms. further research on this issue would be very worthwhile, regardless of whether it supports our finding or, after further study, finds contrary evidence. 121 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 references acemoglu, d., & robinson, j.a. (2006). economic backwardness in political perspective. american political science review, 100(1), 115-131. https://doi.org/10.1017/s0003055406062046 acemoglu, d., & robinson, j.a. (2013). economics versus politics: pitfalls of policy advice. journal of economic perspectives, 27(2), 173-192. https://doi.org/10.1257/jep.27.2.173 ayittey, g.b.n. (2008). the african development conundrum. in b. powell (ed.), making poor nations rich: entrepreneurship and the process of economic development (pp. 137-188). stanford: stanford business and finance. baumol, w.j. (1990). entrepreneurship: productive, unproductive, and destructive. journal of political economy, 98(5), 893-921. https://doi.org/10.1086/261712 berggren, n. (2003). the benefits of economic freedom: a survey. independent review, 8(2), 193-211. https://doi.org/10.4337/9781784718237.00009 coyne, c.j. (2013). doing bad by doing good: why humanitarian action fails. stanford: stanford university press. de haan, j., lundstrom, s., & sturm, j. (2006). market-oriented institutions and economic growth: a critical survey. journal of economic surveys, 20(2), 157-191. https://doi.org/10.1111/j.0950-0804.2006.00278.x diamond, j. (1997). guns, germs, and steel. new york: w.w. norton & company. easterly, w. (1997). the white man’s burden: why the west’s efforts to aid the rest have done so much ill and so little good. new york: penguin. faria, h.j., & montesinos, h.m. (2009). does economic freedom cause prosperity?: an iv approach. public choice, 141(1-2), 103-127. https://doi.org/10.1007/s11127-009-9440-0 gwartney, j., lawson, r., & hall, j. (2012). economic freedom of the world, 2012 report. vancouver, bc: fraser institute. holcombe, r.g., & boudreaux, c.j. (2013). institutional quality and the tenure of autocrats. public choice, 156(3-4), 409-421. https://doi.org/10.1007/s11127-013-0089-3 holcombe, r.g., & rodet, c.s. (2012). rule of law and the size of government. journal of institutional economics, 8(1), 49-69. https://doi.org/10.1017/s1744137411000348 landes, d.s. (1998). the wealth and poverty of nations: why some are so rich, and some so poor. new york: w.w. norton. mokyr, j. (1990). the lever of riches. oxford: oxford university press. olson, m., jr. (1982). the rise and decline of nations. new haven: yale university press. olson, m., jr. (1996). big bills left on the sidewalk: why some nations are rich, and others poor. journal of economic perspectives, 10(2), 3-24. https://doi.org/10.1093/0198294905.003.0002 rodrik, d. (2014). when ideas trump interests: preferences, worldviews, and policy innovations. journal of economic perspectives, 28(1): 189-208. https://doi.org/10.1257/jep.28.1.189 https://doi.org/10.1017/s0003055406062046 https://doi.org/10.1257/jep.27.2.173 https://doi.org/10.1086/261712 https://doi.org/10.4337/9781784718237.00009 https://doi.org/10.1111/j.0950-0804.2006.00278.x https://doi.org/10.1007/s11127-009-9440-0 https://doi.org/10.1007/s11127-013-0089-3 https://doi.org/10.1017/s1744137411000348 https://doi.org/10.1093/0198294905.003.0002 https://doi.org/10.1257/jep.28.1.189 122 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 sachs, j.d. (2001). tropical underdevelopment. nber working paper 8119. https://doi.org/10.3386/w8119 winters, j.a. (2011). oligarchy. cambridge: cambridge university press. https://doi.org/10.3386/w8119 european journal of government and economics 6(2), december 2017, 106-125. appendix 1. sources of leaders’ wealth leader wealth source jose eduardo dos santos http://docteurmilando.ivoire-blog.com/archive/2011/01/14/angola-jose-eduardo-dos-santos-possede-une-fortune-de-31-mil.html ilham aliyev http://www.celebritynetworth.com/richest-politicians/presidents/ilham-aliyev-net-worth/ hamad ibn isa al khalifa http://www.almanachdegotha.org/id229.html khaleda zia http://www.thedailystar.net/newdesign/news-details.php?nid=71084%3cbr%20/%3e paul biya http://www.therichest.org/nation/richest-person-in-cameroon/ wen jiabao http://www.guardian.co.uk/news/blog/2012/oct/26/who-are-the-wealthiest-world-leaders juan manuel santos http://en.mediamass.net/people/juan-manuel-santos/highest-paid.html mobutu sese seko http://www.ghanafilla.net/archives/33753#.ubxy6fnqkfv hosni mubarek http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/hosni-mubarak-photo-1348740984.html girma wolde-giorgis http://www.celebritynetworth.com/richest-politicians/presidents/girma-woldegiorgis-net-worth/ ali bongo ondimba http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/ali-bongo-ondimba-photo-1348740985.html mohamed suharto http://www.cbv.ns.ca/dictator/suharto.html ali khamenei http://money.ca.msn.com/savings-debt/gallery/dictators-and-their-wealth?page=5 uhuru kenyatta http://www.nairobiwire.com/2013/04/uhuru-kenyatta-is-africas-richest.html h.h sheikh sabah iv al-ahmad al-jaber al-sabah http://www.almanachdegotha.org/id229.html filip vujanović https://reportingproject.net/underground/index.php?option=com_content&view=article&id=6:ukanovis-montenegro-a-familybusiness&catid=3:stories&itemid=19 http://docteurmilando.ivoire-blog.com/archive/2011/01/14/angola-jose-eduardo-dos-santos-possede-une-fortune-de-31-mil.html http://www.celebritynetworth.com/richest-politicians/presidents/ilham-aliyev-net-worth/ http://www.almanachdegotha.org/id229.html http://www.thedailystar.net/newdesign/news-details.php?nid=71084%3cbr%20/%3e http://www.therichest.org/nation/richest-person-in-cameroon/ http://www.guardian.co.uk/news/blog/2012/oct/26/who-are-the-wealthiest-world-leaders http://en.mediamass.net/people/juan-manuel-santos/highest-paid.html http://www.ghanafilla.net/archives/33753#.ubxy6fnqkfv http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/hosni-mubarak-photo-1348740984.html http://www.celebritynetworth.com/richest-politicians/presidents/girma-woldegiorgis-net-worth/ http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/ali-bongo-ondimba-photo-1348740985.html http://www.cbv.ns.ca/dictator/suharto.html http://money.ca.msn.com/savings-debt/gallery/dictators-and-their-wealth?page=5 http://www.nairobiwire.com/2013/04/uhuru-kenyatta-is-africas-richest.html http://www.almanachdegotha.org/id229.html https://reportingproject.net/underground/index.php?option=com_content&view=article&id=6:ukanovis-montenegro-a-family-business&catid=3:stories&itemid=19 https://reportingproject.net/underground/index.php?option=com_content&view=article&id=6:ukanovis-montenegro-a-family-business&catid=3:stories&itemid=19 c. boudreaux and r. holcombe / european journal of government and economics 6(2), 106-125 leader wealth source mohammad iv http://www.therichest.com/celebnetworth/politician/royal/king-mohammed-iv-net-worth/ beatrix http://www.forbes.com/sites/investopedia/2011/04/29/the-worlds-richest-royals/ goodluck jonathan http://nigeriavillagesquare.com/articles/ogaga-ifowodo/vice-president-jonathans-qmodestq-fortune.html harald v http://e24.no/makro-og-politikk/kongens-formue-under-100-mill/3603222 qaboos bin said al said http://www.almanachdegotha.org/id229.html asif ali zardari http://www.daily.pk/president-asif-ali-zardari-2nd-most-richest-man-of-pakistan-6666/ benigno "noynoy" aquino iii http://www.affordablecebu.com/load/finance_wealth/president_benigno_quot_noynoy_quot_aquino_iii_net_worth/34-1-0-2692 hamad bin khalifa http://www.almanachdegotha.org/id229.html abdullah bin abdulaziz http://www.forbes.com/2006/05/03/cz_forbes_0522_royals_slide.html?thisspeed=35000 carl xvi gustaf http://exploredia.com/richest-politicians-in-the-world/ bashar al-assad http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/bashar-al-assad-photo-1348740985.html bhumibol adulyadej http://www.almanachdegotha.org/id229.html zine el abidine ben ali http://money.ca.msn.com/savings-debt/gallery/dictators-and-their-wealth?page=3 sheikh khalifa bin zayed al nahyan http://www.forbes.com/2006/05/03/cz_forbes_0522_royals_slide_3.html?thisspeed=35000 yoweri musevini http://ddungu.wordpress.com/2012/04/13/president-yoweri-mueveni-s-accumulation-of-wealth-surprises-ugandans%e2%80%8f/ elizabeth ii http://www.forbes.com/sites/investopedia/2011/04/29/the-worlds-richest-royals/ robert mugabe http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/robert-mugabe-photo-1348740999.html http://www.therichest.com/celebnetworth/politician/royal/king-mohammed-iv-net-worth/ http://www.forbes.com/sites/investopedia/2011/04/29/the-worlds-richest-royals/ http://nigeriavillagesquare.com/articles/ogaga-ifowodo/vice-president-jonathans-qmodestq-fortune.html http://e24.no/makro-og-politikk/kongens-formue-under-100-mill/3603222 http://www.almanachdegotha.org/id229.html http://en.wikipedia.org/wiki/asif_ali_zardari http://www.daily.pk/president-asif-ali-zardari-2nd-most-richest-man-of-pakistan-6666/ http://www.affordablecebu.com/load/finance_wealth/president_benigno_quot_noynoy_quot_aquino_iii_net_worth/34-1-0-2692 http://www.almanachdegotha.org/id229.html http://www.forbes.com/2006/05/03/cz_forbes_0522_royals_slide.html?thisspeed=35000 http://exploredia.com/richest-politicians-in-the-world/ http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/bashar-al-assad-photo-1348740985.html http://www.almanachdegotha.org/id229.html http://money.ca.msn.com/savings-debt/gallery/dictators-and-their-wealth?page=3 http://www.forbes.com/2006/05/03/cz_forbes_0522_royals_slide_3.html?thisspeed=35000 http://ddungu.wordpress.com/2012/04/13/president-yoweri-mueveni-s-accumulation-of-wealth-surprises-ugandans%e2%80%8f/ http://www.forbes.com/sites/investopedia/2011/04/29/the-worlds-richest-royals/ http://in.finance.yahoo.com/photos/8-of-the-richest-dictators-in-history-slideshow/robert-mugabe-photo-1348740999.html european journal of government and economics 6(2), december 2017, 106-125. 2. definition of variables variable definition source log wealth natural log of real wealth see appendix table 1 efw rate (efw at end efw at beginning )/ tenure economic freedom of the world 2012 initial efw efw at beginning of tenure economic freedom of the world 2012 pci rate (pci at end pci at beginning) / tenure gdp per capita, world bank, ppp initial pci pci at beginning of tenure gdp per capita, world bank, ppp freedom rate (freedom at end freedom at beginning)/ tenure freedom house, freedom rating initial freedom freedom at beginning of tenure freedom house, freedom rating civil rights rate (civil rights at end civil rights at beginning)/ tenure freedom house, civil liberties initial civil rights civil rights at beginning of tenure freedom house, civil liberties political rights rate (political rights at end political rights at beginning)/ tenure freedom house, political rights initial political rights political rights at beginning of tenure freedom house, political rights polity2 rate (polity2 at end polity2 at beginning)/ tenure polity iv initial polity2 polity2 at beginning of tenure polity iv distance distance from the equator http://lab.lmnixon.org/4th/worldcapitals.html; http://www.eaae-astronomy.org/eratosthenes/ index.php?option=com_content&view=article &id=47&item&id-68 resource rents natural resources (% of gdp) total natural resources, world bank oda official development assistance (% of gdp) indexmundi.com, net official development assistance and official aid received tenure length of tenure (in years) abstract. one frequently given explanation for why autocrats maintain corrupt and inefficient institutions is that the autocrats benefit personally even though the citizens of their countries are worse off. the empirical evidence does not support this... keywords. institutional quality; wealth; autocrats jel classification. p51; o50; h11 references the debate on the economic effects of minimum wage legislation european journal of government and economics 6(2), december 2017, 171-190. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 the debate on the economic effects of minimum wage legislation santos miguel ruesga-benito a, fernando gonzález-laxeb, jose picatostea, * a universidad autónoma de madrid, 28049 madrid, spain. b universidade da coruña, campus de elviña, 15071 a coruña, spain * corresponding author at: facultad de ciencias económicas y empresariales, c/ francisco tomás y valiente 5, universidad autónoma de madrid, 28049 madrid, spain. jose.picatoste@uam.es article history. received 3 august 2017; first revision required 3 october 2017; accepted 20 november 2017. abstract. the minimum wage establishment has its origin in the first third of the last century. since its creation has been a focus of continuing controversy and an unfinished debate on economics field. this work reviews the effects of the minimum wage on employment and other macroeconomic variables, from both theoretical and empirical perspectives. the method is based on the revision of the literature and the main economic indicators. the central contribution of this paper is providing a general reflection on theoretical and empirical analysis about the debate on minimum wage and its effects. the results showed that some labor policies are taking account the effects of austerity strategies, shifting the attention towards the implementation of minimum wages or their updating, in order to reduce the growing inequalities in the distribution of income, and even poverty levels. keywords. political economy; minimum wage; employment; economic inequality jel classification. j38 “raising the federal minimum wage, expanding the earned income tax credit for workers without dependent children, limiting tax breaks for high-income households, preventing colleges from pricing out hardworking students, and ensuring men and women get equal pay for equal work would help to move us in the right direction too” (barack obama, 2016: “the way ahead”, the economist, october, 8th-14th) 1. introduction in the context of the great recession years (2007-2014), there were evident effects on the rising of economic and social inequalities for most developed countries. those effects made increased the interest in the figure of the “minimum wage” (mw). this interest was either for its implementation in those countries where it did not exist yet, or for its revaluation where it already existed, in order to establish a minimum level of income for the wage-earning population, with derived effects for the value of different social benefits, in many cases. http://www.ejge.org/ mailto:jose.picatoste@uam.es 172 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 in this perspective is, for example, the debate unfolded during the final years of president obama's administration in the united states. furthermore, in most european union countries and in other oecd1 countries here are debates about this issue. that is why around the aforementioned has been reactivating an economic debate, already with a long trajectory in time, which started in the first third of the last century xx, when the minimum wages began to be implemented in some countries (united states, among others) under different minimum wage formulas, in the context of the effects of the great depression. it is a long debate, with positions founded both in the field of economic science, and in the large and multiply empirical studies, which have tried to determine the effects of this labor institution in the macro and microeconomic field of relationships economic and labor, for almost a century2. it is possible to find hundreds of examples specified by the spanish central bank, for warning about the negative effects of increasing minimum wages on competitiveness and employment in the spanish economy. “the implementation a system for annually revising the inter-professional minimum wage would be inappropriate, particularly if it were introduced in the workers' statute law”(central bank of spain, 2005). environmental context, in fact, greatly condition the results of the empirical analysis, so that beyond the methodologies applied and the theoretical approaches taken. the presence of enormous disparity of elements, which intervene in its interpretation, condition the effects of the application or expansion of the minimum wage mechanism. the presence or absence of competitive markets for goods and services, the situations of greater or less monopsonic intensity in labor, for example, will determine that those effects induced by the existence or revision of the minimum wage act in one direction or another on the level of employment, prices or, ultimately, the competitiveness of enterprises. it is also relevant to assess the direction and level of the effects of the minimum wages, taking into account their relative amount, on the average wage levels in force in the economic and labor space analyzed3. ultimately, a large number of what we mentioned in the previous paragraphs should be understood in relation to the diversity of economic and labor institutions that accompany the presence of the minimum wage and its revaluation. this scenario gives rise to a specific labor framework, which could be more or less rigid. in addition, this scenario will determine possibilities of action and reaction of the economic agents in relation to the minimum-wage institution (mwi). particularly with the labor market and its characteristics, which will determine 1 precisely, in september 2016 i attended an international seminar on minimum wages, organized by "action against poverty", with the participation of various political and civic authorities. it was this event that motivated me to develop the work presented here. 2 “to paraphrase an old joke, if there were only two economists left in the world, they would disagree about the minimum wage. does it cost jobs or create jobs? it’s a valid question, but the answer depends on who you ask. opponents of minimum wages contend that increased labor costs force businesses to cut staff, costing jobs. that sounds reasonable. proponents of minimum wages argue that giving workers more disposable income puts money back into the economy, which in turn creates jobs. that makes sense, too. so what’s the answer?” mejeur (2014). 3 this is a fundamental issue, for example in cases such as mexico, where the minimum wage level is around 100 euros, a level much lower than the estimated average income in the so-called informal economy. in the event referenced in note 1, we discussed this issue. 173 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 the different elasticities for both supply and demand, and which will influence in a singular way the concretion of the effects of legislation on the minimum wage, both in conditions of competitive markets as in monopsony situations, or in other markets with asymmetric information. in addition, we also should be remember that some particular aspects, as the preeminent theoretical approach in scientific publications (reflecting preferences of journal editors) or, the methodological matters, strongly condition the results of the empirical analysis of the effects of the minimum wage. one example is the remoteness of the spaces in comparison, when the analysis of data of panel of different geographical spaces is used. we could numbered a long list of conditioning factors on the amount and direction of the obtained results. finally, we should emphasize the existence of profuse literature (theoretical and empirical) on the effects of minimum wage legislation on employment and its evolution. this is not the same situation for trying to relate this regulatory instrument (mwi) to other economic variables like prices, productivity or labor income inequality, such as on-the-job training, or social, such as schooling or poverty rates, among other considered aspects. 2. the minimum wage theory: analyzing different approaches empirical results from the theoretical discourse are sometimes abused, and/or empirical results are extrapolated to confirm positions, which are rather opinions, on the effects of the minimum wage on employment and other economic variables (see figure 1)4. however, the debate on the subject is long, deep and, in our opinion, still unfinished. in the forties of the last century, it began the debate on the effects of an increase of the minimum wage on employment5. from that point on, and always with the same and uncontestable assumption that competitive labor markets exist, the apparent consensus in economic science has been that a minimum wage increase affected employment in a negative way6. in textbooks, the minimum wage usually appears as the paradigm of an institutional rigidity that causes unemployment by interfering with the invisible hand that empties markets. many of those who pontificate in this direction has not passed this stage of manual. 4 see, for example, how the spanish central bank intrepetes in a biased way this phenomenon, on the quotation above (spanish central bank, 2005). 5 see for example lester (1946), pigou, (1920) and stigler (1946) 6 without being exhaustive, you can see information on this line: abowd, kramarz & margolis, 2000, baker, benjamin & stanger, (1999), bazen & skourias (1997), burkhauser, couch & wittenburg (2000), currie & fallick (1996), deere, murphy & welch, (1995), deere, murphy & welch, 1996, kim & taylor, (1995), meyer & wise (1983a), meyer & wise, (1983b), neumark & wascher, (1992), neumark & wascher, (2000), neumark & wascher (2003), orazem & mattila, (2002). 174 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 figure 1: hypothesis on the economic-labor incidence of minimum wages legislation 2.1 the classical-neoclassical approach if we start from the hypothesis of a perfectly competitive labor market, the result is the expected one. under this approach, it is clear: when the minimum wage exceeds the equilibrium one, which firms hire without limitations, there will be a substitution effect, because of the reaction of entrepreneurs for facing that increase externally induced in the market. this situation will reduce the hired work (falling employment) and will incorporate more capital resources, now cheaper in relative terms7. the main assumption on which the inverse relationship between the minimum wage and employment relays on is completely discredited: the existence of perfect competition in labor markets, even when considered plural. the distribution of the information is not equally among all the participants and the good that is distributed is far from homogeneous, neither by supply nor by demand. faced with a starting market with low levels of competition, the minimum wage 7 this perspective is well illustrated in the following statements by milton friedman: “the minimum wage law is most properly described as a law saying that employers must discriminate against people who have low skills. that’s what the law says. the law says that here’s a man who has a skill that would justify a wage of $5 or $6 per hour (adjusted for today), but you may not employ him, it’s illegal, because if you employ him you must pay him $9 per hour. so what’s the result? to employ him at $9 per hour is to engage in charity. there’s nothing wrong with charity. but most employers are not in the position to engage in that kind of charity. thus, the consequences of minimum wage laws have been almost wholly bad. we have increased unemployment and increased poverty” (from http://blog.1worldonline.com/miltonfriedman-minimum-wage-myths/, visited at 2016, 11, 11th). production prices other wage issues minimum wage legislation resources allocation (labour and capital) employment productivity labor or capital productivity increase / decrease qualified or unskilled employment salary gaps labor / capital goods prices business benefits 175 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 breaks monopsony demand, improves the circulation of information and the expectations of the actors. ultimately, raising the minimum wage improves the overall efficiency of the labor market. from this perspective, a major branch of economic science, among the three hundred studies that have already been used to test the meaning and size of the relationship, supports the thesis that minimum wage growth has no or negligible effect on the employment (brown, 1999 freeman, 1994 and freeman 1996). summarizing, reality tells us that labor markets move away from this formal model, showing rather situations of limited or imperfect competition. it may be on the supply side, because, for example, workers are not very mobile geographically, which limits one of the basic conditions of a competitive market (absolute mobility of factors). or it may be on the demand side, by concentrating on the company (in various ways) the power of decision on wage determination. in this context, especially on monopsony conditions8, the effects of the minimum wage are ambiguous, if not contrary to what the theory predicts in the presence of perfect competitive conditions. in this case, the employer is able to fix wages, below or above the marginal product of labor, so that the appearance of the minimum wage does not have to necessarily lead to a fall in employment. on the contrary, the imposition of a minimum wage can increase employment. the positive effect on employment of the minimum wage can be maximized in the level of the wages corresponding to the equilibrium wage in a competitive labor market. with minimum wage levels above the equilibrium wage (of the competitive labor market) these gains in employment are reversed and employment may fall below its original level. but under monopsony conditions, a greater elasticity of labor supply and / or a lower elasticity of labor demand imply a larger range in which minimum wages can be increased without generating job losses (oecd 1998). 2.2 the efficiency wage approach another interpretation, link up with a more recent theoretical approach, within the neoclassical current, denominated of “efficiency wages”. according to this line of thinking, under conditions of imperfect competition, employers can set their employees' wages above the level of market equilibrium in order to increase worker productivity and reduce absenteeism and job turnover. in this context, a rise in minimum wages may result in an increase in employment. however, as in the case of the monopsony model, beyond a certain level, an increase in the minimum wage could have negative effects on the level of employment. rebitzer and taylor (1995) show that, within a model of wage efficiency, an increase in the minimum wage will increase employment in the short term. but, in the long run, however, profits may be lost, depending on the position of firms along the profit curve and on subsequent changes in product prices and the number of firms operating in the market (oecd, 1998). 8 see footnote 5 176 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 2.3 the endogenous development approach and human capital other models admit the possibility of positive effects in the employment of the minimum wage raise take into account the endogenous growth linked to the decisions of investment in human capital. the key is that minimum wages gave incentives for low productivity workers in order to invest more in capacitation or education, which will result in productivity growth. the increase in human capital has a positive impact on growth and, consequently, employment. cahuc and michel (1996) show that a decrease in the minimum wage may even reduce growth. cubitt and hargreaves-heap (1999) argue that the net loss of employment resulting from an increase in the minimum wage may be zero for a given range of wage values, since firms' investments in physical capital and human capital by the workers. acemoglu and pischke (1998) also show that minimum wages can increase the training provided by the firm to low skilled employees, with positive effects on labor productivity (oecd 1998). 2.4 employment search models minimum wages has also been analyzed using the theoretical framework based on “job search” models. within this framework, the effects on employment depend on the level of the minimum wage and its impact on the intensity of job search, the level of accepted wages and the probability of the offer of a job. swinnerton (1996) presents a search equilibrium model in which firms show work demand curves with downward slopes, labor productivity varies from one firm to another, and the unemployed have imperfect information and job search at random and sequentially. the author shows that, because of an increase in average labor productivity, the positive welfare effects can still arise even in those cases where there are negative effects on employment. 2.5 segmented market models9 minimum wages would only affect some of the segments of the labor market, located on secondary market space. those in which wages were below that minimum wage. depending on the size of the segment or segments, the degree of informality of it and the position of its productive supply in the value chain, the impact would be significant on aggregate employment or not, as well as on prices. we could expect some effects on work mobility between some and other segments of the labor market, which do not give the overall volume of labor available and labor employed as a whole10, but, in any case, its quality. 9 ruesga et al (2014) show a detailed and illustrated explanation of how the labor market is working, taking account those theoretical approaches 10adding formal and informal sector. in this perspective you can be framed, for example, monica jimenez doctoral thesis recently presented at the university of alcala de henares before a committee, which i chaired. in its conclusions it states “it is confirmed that the minimum wage (mw) is presented as a determinant of employment conditions in the medium and large businesses and also in the public sector. in fact, the increasing on mw is linked to an increase in the bad quality work positions and in the informal workplaces, besides to the reduction of the good quality jobs. this implies that the quality of employment in 177 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 then, until cover almost the entire spectrum of theoretical interpretations on the functioning of labor markets and the role of institutions in the same and in the whole of economic system. after almost a century of theoretical debate, the positions has not substantially changed. neither the arguments wielded in one or other direction, arguments that we can observe again in the most recent debates about changes in the legislation or amount of the minimum wage. for example, in the united states11. 3. empirical analysis of the effects of the minimum wage to focus on the wide range of empirical analysis carried out on the impact of minimum wage legislation on different functions or aspects of economic activity, the following chart 1 and 2, have been elaborated, summarizing the arguments in favor and against order to do some kind of action about the minimum wage (creation, increase, etc.). then the social cohesion and sustainable development is also crucial (novo-corti et al, 2015). chart 1. reasons in pro of legislating on minimum wage (or increases) the most cited arguments, according to mejeur (2014), in favor of legislation on the minimum wage, specifically its growth, include among others:  places more money in the pocket of low-income workers.12  it reduces the gap between low and high wages, reducing inequalities, both within companies as in the whole of economy.  it introduces more money into the economy; insofar low-income workers are more likely to spend their wages at a greater rate than their better-paid counterparts, who are more likely to save. these companies and for the public sector is a key point to the scope and coverage of labor institutions and social security element, as it is in this specific case mw. therefore, the compliance and adverse effects of mw are evaluated when the labor market, to which it is applied, such as that of large and medium-sized enterprises as well as in the public sector, has different employment conditions and is segmented. therefore, the non-compliance of the mw is not ultimately due to the successive increases introduced at its level but to the context [segmented markets] in which they occurred” (jiménez 2016). 11 virtually all prominent economists worldwide have advocated some sort of argumentation about the effects of minimum wage in the economy and the labor market. some of them are on www.projectsyndicate.org. moreover, there is a synthesis or the supported arguments in the recent debates in united states of america, in favor to the rise of mw in komlos (2015) and another against in the text quoted in note 10 (blog.1worldonline.com). 12 according to a 2013 congressional research service report from the us congress, a father with two children and who works full time with the current minimum wage should be earning about $15,000 and living at 76 percent of the federal poverty line. if the minimum wage rises from the current 7.25 hours to $9 the same family could be placed at 94 percent of the poverty line.  income bottom, avoiding company abuses. expanded salary effects on workers, over minimum salaries.  improve social cohesion and stable and sustainable economic growth.  minimum social benefits threshold. incomes increase for marginal social groups.  replacement unskill by skill labour  decrease jobs rotation and searching costs  downturn salary gap  productivity improvement  boosts business efficiency  i+d+i improvement 178 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190  it decrease mobility among low-wage workers, which can have a positive effect on business insofar as higher mobility raises training costs and reduces productivity. chart 2. reasons against legislation on minimum wage (or increases). in turn, the arguments against the growth of minimum wages are (mejeur, 2014):  reduces employment. labor costs are the substantial part of the budget of many companies, which means that entrepreneurs will be driven to reduce jobs or hours of work to keep their production floor and profits. this could lead to the disappearance or reduction of incomes for workers with low wages  there are better ways to face poverty, such as income tax credits for low incomes, or fiscal policies that encourage the development of financial assets and savings for low income families  increases in labor costs can be passed on to consumers through prices. higher prices lead to declines in demand may lead to depressive effects on the economy  the increase in labor costs leads to lower profits for companies, which means that entrepreneurs have fewer monetary resources to invest in their companies to create jobs and expand their business as we discussed above, the extent of theoretical interpretation on the neoclassical and derivative slopes, the empirical contributions to this theory are somewhat more modest, in number and in the size of the results13 even in low-wage sectors. and they are strongly focused on the analysis of the relationship between minimum wage legislation and employment. for example, machin, manning and rahman (2003) find a negative effect on the use of minimum wage introduction in the united kingdom. but this effect is very modest even considering that a third of the sector analyzed, the domestic service, had previous salaries below the established minimum. this idea of classical manual is, in a way, an anachronism that has been confronted with an abundant economic literature, empirical, with consistent evidence on the zero or even positive effect on employment, in certain circumstances, derivative on the growth of the minimum wage14. 13 for example, they shown a mean elasticity around –0,1 for the relation por ciento between minimum wages and employment (brown, gilroy and kohen, 1982). 14 in this perspective, with nuances as to its conclusions, a long list of references can be consulted, among others: azam, (1992), bell, (1997), bernstein & schmitt (1998), bhaskar & ton (1999), boadway & cuff (2001), bruno & cazes (1997), card (1992a), card & krueger (1994), card & krueger (1995), card, (1992b), dickens, machin & manning, (1999), hyslop & stillman, 2004, islam & nazara, 2000, jones, (1987), katz & krueger (1992), klerman (1992), lemos (2003), lemos (2004a y b), machin & manning (1997), machin & manning, (1994), machin & manning, (1996), manning (1995), shepherd (2000), stewart (2001) & wellington (1991). some of the most outstanding works in this field, developed in the  jobs losses (in factory and macrolevel).  changes of labour relative prices regards to skill levels.  some labour intensive and unskill jobs activities way out to informality..  competitive companies losses.  higher inflation  not reduce poverty, but rather increases. 179 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 the sense and amount of the effect of the growth of the minimum wage would not be unidirectional. it vary depending on three variables that, in general, can be annulled until they resolve in zero consequences: the elasticity of the labor demand in the sector covered by the minimum wage, the elasticity of demand in the uncovered sector and the size of the growth of the minimum wage (fields, 1994). thus, in the less qualified and less paid jobs, increases in the minimum wage would have a minimal impact on employment. under this premise, in spain, where the minimum wage is over the average level of the developed countries (see table 1 and figure 2), the possibilities of influencing employment would be remarkably low. table 1. minimum wage as a proportion of the average wage (in%), 1976, 2000, 2009 and 2015 in oecd countries. country 1976 2000 2009 2015 australia 65 58 54 39 belgium 58 53 52 45 canada 52 41 42 48 chile .. 64 71 61 chech republic .. 32 36 66 estonia .. 34 40 59 france 58 56 62 53 germany .. .. .. 36 greece 69 52 49 57 hungary .. 37 48 85 ireland .. 68 47 52 israel .. .. 57 38 japan 29 32 36 38 korea .. 29 45 40 luxembourg 41 54 55 42 mexico .. .. 37 31 netherlands 64 51 47 44 new zeland 57 50 59 54 poland .. 40 46 94 portugal 48 49 54 61 slovak republic .. 42 45 78 spain 48 42 42 46 united kingdom .. 41 46 34 united states 47 36 37 33 source: own elaboration from neumark & wascher (2003) and datos-macro.com in this regard joliet (2015) notes that: “i have collected 138 empirical papers that deal with the minimum wage and its impact on employment. these studies date from 1957 to 2011. they test the effects of minimum wages in more than 21 countries. my survey found that 64% of research in this time period resulted in negative effects on employment, 19% found no effect, 12% found mixed effects, and 5% found positive effects. when the period of time is reduced to 1992-2011, 52% of research found negative effects on employment, 25% found no effect, 16% found mixed effects, and 7% found positive effects. this survey gives us an estimate of whether field of empirical analysis, began in the early nineties of the last century in response to the dominant postulates of the neoclassical current, those of card, krueger and katz. in this regard, the blog quoted above (blog.1worldonline.com) states that: “now, the clinton administration is advancing the novel economic theory that modest increases in the minimum wage will have no impact whatsoever on employment. this proposition is based entirely on the work of three economists: david card and alan krueger of princeton, and lawrence katz of harvard. their studies of increases in the minimum wage in california, texas and new jersey apparently found no loss of jobs among fast food restaurants that were surveyed before and after the increase” (see footnote 9, from http://blog.1worldonline.com/miltonfriedman-minimum-wage-myths/, visited at 2016, 11, 11th) 180 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 or not the theory of competitive markets is correct or whether the theory of monopsony is in theory the opposite. what we see is that the vast majority of research supports the theory of competitive markets or the idea that minimum wage is not well structured. fortunately, workers with a minimum wage constitute a very small portion of the total labor force. due to this fact, these global negative effects are not so great“ [sic]. figure 2. minimum wages in ocde countries. source: own elaboration from neumark & wascher (2003), oecd (2004) and datos-macro.com if indeed the growth of the minimum wage has imperceptible effects on employment, the limited growth of labor costs entailed, then, in a general equilibrium model, would have to be paid either by the entrepreneurs, subtracting it from their profit margins (card & a. krueger, 1995), or by consumers through the inflation they assume (aaronson, 1997, macdonald & aaronson, 2000. l'horty & raults, 2004, katz & krueger, 1992 & lemos, 2004a). however, the very limited literature on these eventualities shows that both effects are very small. most workers who hover around the minimum wage are in very competitive sectors (in the domestic economy, but not competing with the outside) with little tendency to transfer any wage growth to prices. if there is no effect on inflation15 or employment or even on benefits, what the economic literature sees is that raising the minimum wage could lead to a structural transformation of supply to adjust to new labor costs. in this sense, the sectoral distribution of workers who got earnings equal or above the minimum wage is also important, since it affects their real effects on the competitiveness of the national entrepreneurship network. for example, in accordance with the previous paragraph, 15 werner & sell (2015) stated that “in contrast, we cannot find any significant price reaction for west germany. this suggests that implemented minimum wage in west germany is too low in comparison to the predominantly paid wages and is hence not binding”. -40,0 -20,0 0,0 20,0 40,0 60,0 80,0 100,0 au st ra lia be lg iu m ca na da ch ile cz ec h r. es to ni a fr an ce gr ee ce hu ng ar y ire la nd ja pa n ko re a lu xe m bo ur g m ex ic o n et he rla nd s n ew z ea la nd po la nd po rt ug al sl ov ak r . sl ov en ia sp ai n tu rk ey u . k in gd om u . s ta te s real increasing mw 2007/2000 (%) real increasing mw 2015/2007 (%) % mw/average wage 2015 (right sca.) % 181 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 komlos (2015) points out that minimum wage earners do not generally work in the export sector16, which would imply that changes in the minimum wage would have little effect on the external competitiveness of workers. on the contrary, raising minimum wages could stimulate some productivity adjustments through training (public or private) (acemoglu & pischke 1999, arulampalam, booth & bryan 2004, cubitt & hargreaves-heap 1999 & cunningham 1991), or by means of the greater permanence in employment (fraja, 1996, nichell & layard, 1997, portugal & cardoso, 2001 & teulings, 2000) or even by promoting investment in capital goods, in case they were able to catch the effects of (askenazy, 2001, gordon, 1995, & lucas, 1988). much more important is the first effect, because the majority of minimum wage earners are in branches of services that are not exposed to international competition, as has been said. in the export sector, minimum wages are an anecdote, so the contribution of their rise to international competitiveness is practically nil, this happens practically anywhere in the world, but particularly in the developed countries (standing, sender & weeks, 1996). in this sense, the minimum wage and its growth could be interpreted as a welfare program without side effects, besides contributing to the improvement of the country's productivity. this is due to the absence of incontestable evidence that there is a dragging effect of minimum wages on the rest of the wage pyramid, but rather reduces the dispersion among the different wages (dinardo, fortin & lemieux, 1996, lee, 1999, dickens & manning, 2002 & smicht, 2015). reducing wage disparities also contributes to increasing domestic demand17 by rebalancing the demand-import elasticity that increases with wage dispersion. that these growths in domestic demand do not translate into inflation will not be the responsibility of the recipients of rising minimum wages, but rather of policies on supply in general and market liberalization in particular. under this scenario we would be facing one of the few examples of virtuous circles in economics, within reasonable limits of growth. regarding the indexation of the minimum wage to some kind of reference, be it productivity or inflation, it is interesting the declaration that “we cannot get any conclusion on which of the effects of the indexation of the minimum wage, if they are any, will differ from the current effects of the minimum wage“ (h.i. grossman, 1981). the evidence is that in ireland and the united kingdom, the last countries to introduce minimum wages, at levels substantially higher than those discussed in spain, this fact has not translated into negative impacts on employment. both countries have unemployment rates among the lowest in the european union. in observing the differences in the field of empirical studies, however, it would be useful to distinguish between effects on the volume of labor or on unemployment, on two sectors of analysis: employment covered and not covered by the institution of the minimum wage (mwi), 16 he concludes: “do you have a chinese mcdonald’s in your neighborhood? i don’t think so. in fact, most of the people who work for under $10 an hour are working as cashiers at grocery and department stores (1.4 million), retail salespeople (1.1 million), cooks (1 million) and janitors, cleaners, waiters or waitresses (1.5 million) — none of whom work in the export sector. raising their minimum wage would not hurt our exports at all. their wages have nothing to do with competing with the rest of the world”. (komlos, 2015). 17 “there is no evidence about the adverse effect of this rent and there is evidence on the positive effect on consumption” (garcía vega, 2013). 182 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 as mincer (1976) did18. in this regard, it is important to consider the relative importance of the minimum wage (its actual coverage, which in most developed countries is often reduced19), to differentiate between the covered and uncovered sectors by the presence or growth of the minimum wage. in any case, the empirical evidence does not seem to be conclusive due to the effects of modifying the minimum wage nor in any one variable (active population / unemployment) as shown below, in a simple way, in figures 3 and 4. figure 3. real increase of the minimum wage (2015/2000 and 2015/2007) and the unemployment rate in oecd countries. source: own elaboration based on oecd data. there is no relationship between the growth of the minimum wage (both before and including the analysis of the great depression) and the unemployment rate (in 2016). the countries with 18 what, curiously, could lead us to interpretations close to those that arise from the "theory of segmentation", understanding that the labor market is articulated in different segments, on which the modification in the institution of the minimum wage would impact in a way very different manner (with labor movements included, as noted above, in note 11). although mincer is actually referring to another type of situation, of separation between those who have wages above or below the minimum wage, considering those as not covered, those not directly affected by the mwi; however, the different effects on unemployment or activity rates could be explained by the fact that part of the labor displacement effect due to the raising of the minimum wage could be directed linked not to unemployment but to inactivity (part of which, could fall within the scope of the informal employment). 19 for example, for the united states, based on the bureau of labor statistics, it has been pointed out that in 2012:  around 3.6 million (or 4.8 percent) of the 75 million workers paid on an hourly basis earn $7.25 an hour or less.  more than half of minimum wage workers are under the age of 25.  six percent of women and 3 percent of men earn minimum wages.  the leisure and hospitality sector has the highest proportion of minimum wage workers.  louisiana, oklahoma, texas and idaho have the highest percentage of minimum wage workers.  alaska, oregon, california, montana and washington have the lowest percentage of minimum wage workers. (mejeur, 2014). -20 -10 0 10 20 30 40 50 60 -20 0 20 40 60 80 100 120 140 160 0 5 10 15 20 25 re al in cr ea se m w 2 01 5/ 20 7 re al in cr ea se o f m w 2 01 5/ 20 00 unemployment rate (%) real increase mw 2015/2000 (%) real increase mw 2015/2007 (%) (right sca.) 183 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 the highest unemployment rate in 2016 (greece and spain) recorded very low, if not negative, growth of the minimum wage in the two periods considered. however, for the places with lower unemployment rates, there is a great dispersion in minimum wage levels. figure 4. real increase of the minimum wage (2015/2000 and 2015/2007) and activity rate (%)in oecd countries. source: own elaboration in basis to oecd data. similarly, the possible relationship of increase in the minimum wage and activity rate (data 2015), analyzed by means of panel data for several oecd countries, can be considered not significant. 4. concluding remarks in the current debate on the incidence of the minimum wage on employment or, more broadly, on the functioning of the economic system and the labor market, there is much more of theoretical discourse (with its ideological connotations) than of consistent and unquestionable empirical evidence. this debate has become more pronounced in recent decades, increasing pressures for the increase in the ims, given the deregulation of european labor markets and subsequent growth in income inequality (functional and salary), in an economic recessive context, boosted by austerity policies20 implemented in some european countries (díaz-roldan, 2017). this has led to understand the minimum wage as a restrainer of inequalities and even reducing poverty levels. when making empirical estimates, first should be taken into account, on the one hand, the structural and institutional characteristics of labor markets analyzed and, on the other hand, the comparisons or causal relations. in order to appreciate such differences, it is necessary to rely on analytical models that simultaneously contemplate the structural and institutional characteristics of these markets (affecting the elasticities of the variables of the model, the 20 see garcía and ruesga benito, 2014 -30,0 -20,0 -10,0 0,0 10,0 20,0 30,0 40,0 50,0 60,0 -20 0 20 40 60 80 100 120 140 160 50 52 54 56 58 60 62 64 66 68 70 re al in cr ea se m w " 01 7/ 20 07 re al in cr ea se o f m w 2 01 5/ 20 00 activity rate (%) real increase mw 2015/2000 (%) real increase mw 2015/2007 (%) (right sca.) 184 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 supply and demand of labor, the occupational and sectoral structure, the level and characteristics of informality in employment, etc., as well as the disparity of institutions that regulate the operation of markets), the singularities of minimum wage legislation in each area (relative amount, intensity of coverage, etc.) and the actual situation of the labor market (salary level and structure, etc.). with this background, an abundant literature of empirical analysis has been generated, generally based on inevitably simplifying models, which offer a huge disparity of uneven results. in general, there is not enough evidence to ratify without any discussion one of the theoretical approaches contemplated. different reasons are pointed out: “the lack of understanding of the functioning of labor markets, our inability to get rational legislation and the small size of the workers affected by the minimum wage are factors that explain how it is a myth the fact that it is possible to manipulate the labor market to achieve meaningful results” (joliet, 2015). and yet a good part of the developed countries and most europeans have had legislation on minimum wage for decades, plus a few others that have recently been incorporating it, such as ireland, the united kingdom and germany. it could therefore be concluded that the redistributive and, by extension, social stability aspects are more concerned than the possible adverse effects on employment or other relevant variables for the purposes of economic equilibrium and corporate competitiveness (or, alternatively, evidence of other positive effects, such as those on labor productivity). in general, there is no significant political rejection of increases in mwi, except in strongly ideologized business or political environments (in the neoliberal perspective) because of the hypothetical job losses that would result from it (and derivative effects that would decline welfare). in many cases, the revaluation of minimum wages is indexed to some macro variable (usually the evolution of consumer prices) and its application does not usually generate extreme controversy in the political world. it is therefore appropriate, before establishing a priori conclusions about the effects of the minimum wage on any relevant variable characteristic of the equilibrium or economic growth, to take into account, for a particular space:  a diagnosis of the labor market situation: imbalances, informality, wage level, functioning of labor institutions.  the institutional landscape of the minimum wage. existence, scope of application, interaction of different levels (geographical, occupational, etc.) revaluation and management, etc.).  the definition of the pursued objectives with the implementation / increase of the amount of minimum wage. all the above mentioned should be taken account for actuate on the next issues: legislating on minimum wage minimizing adverse effects and promoting pro-positive results: definition of objectives (hierarchy), basic parameters of the minimum wage or its increase: scope, social participation, etc. continuous monitoring on these possible effects for evaluation: 185 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190  microeconomic effects at the enterprise level: job search, productivity, differential wages, etc.  and of the observable effects at the macroeconomic level: on aggregate employment, wage developments, production supply, aggregate demand, price evolution, economic growth, impacts on income distribution and asset concentration , on poverty, etc. to such effects, and in general, it can be concluded, following belman and wolfson (2014: 401), that “modest increases in the minimum wage raise the wages of the working poor without substantially affecting employment or working hours, and provide. the minimum wage is constituted as a policy instrument which, when used correctly and in combination with other policies and programs, can improve the standard of living of the working poor “(ibídem)21 but, of course, theoretical reflection and empirical applied research on the effects of legislation on the minimum wage continues. references aaronson, d. (1997). price pass-through and minimum wages. federal reserve bank of chicago. abowd, j. m., kramarz, f. & margolis, d. n. (2000). minimum wages and youth employment in france and the united states. in d.g. blanchflower & r.b. freeman (eds.), youth employment and joblessness in advanced countries. national bureau of economic research. acemoglu, d. & pischke, j. s. (1999). the structure of wages and investment in general training. journal of political economy, 107(3), 539-72. doi: 10.1086/250071 arulampalam, w., booth, a.l., & bryan, m.l. (2004). training and the new minimum wage. economic journal, 114, 87-94. doi: 10.1111/j.0013-0133.2003.00197.x askenazy, p. (2001). minimum wage, exports and growth. centre d'etudes prospectives d’economie mathematique appliquees a la planification, ecole normale supérieure. azam, j. p. (1992). the agricultural minimum wage and wheat production in morocco (197189). journal of african studies, 1(2), 171-191. baker, m., benjamin, d., & stanger, s. (1999). the highs and lows of the minimum wage effect. a time-series cross-section study of canadian law. journal of labor economics, 17(2), 31850. doi: 10.1086/209923 banco de españa, (2005). boletín económico, enero 2005. banco de españa. 21 these conclusions are obtained through a meta-analysis of more than 200 academic articles related to the minimum wage published since 1991 (focusing mainly on articles published since 2000), in which they try to clarify existing research and explain the effects of raising the minimum wage on various variables including income, education, employment, inequality and poverty. three models exist in the literature in order to analyze the populations affected by the variation in the minimum wage, the temporal sequence and the magnitude of its effects. although primarily focused on the united states, they also consider research conducted in canada, the united kingdom, and several other developed countries to perform micro and macroeconomic analysis. 186 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 bazen, s., & skourias, n. (1997). is there a negative effect of minimum wages in france? european economic review, 57(3-5), 723-732. doi: 10.1016/s0014-2921(97)00004-4 bell, l.a. (1997). the impact of minimum wages in mexico and colombia. journal of labor economics, 15(3), 103-135. doi: 10.1086/209878 belman d., & wolfson, p. j. (2014). what does the minimum wage do? kalamazoo, mi: w.e. upjohn institute for employment research. doi: 10.17848/9780880994583. bernstein, j., & schmitt, j. (1998). making work pay: the impact of the 1996-97 minimum wage increase, economic policy institute. bhaskar, v., & to, t. (1999). minimum wages for ronald mcdonald monopsonies: a theory of monopsonistic competition. economic journal, 109(455), 190-203. doi: 10.1111/14680297.00427 boadway, r., & cuff, k. (2001). a minimum wage can be welfare-improving and employmentenhancing. european economic review, 45(3), 553-76. doi: 10.1016/s0014-2921(00)000660 brown, c. (1999). minimum wages, employment, and the distribution of income. in o. ashenfelter & d. card (eds.), handbook of labor economics, north holland press. brown, c., gilroy c., & kohen, a. (1982). the effect of the minimum wage on employment and unemployment. journal of economic literature, 20(2), 487-528. bruno c., & cazes, s. (1997). le chômage des jeunes en france: un état des lieux. revue de l'ofce, 62, 75-107. burkhauser, r., v., couch, k.a., & wittenburg, d.c. (2000). a reassessment of the new economics of the minimum wage literature with monthly data from the current population survey. journal of labor economics, 18(4), 653-80. doi: 10.1086/209972 cahuc, p., & michel, p. (1996). minimun wage: unemployment and growth. european economic review, 40(7), 1463-1482. card, d. (1992a). using regional variation in wages to measure the effects of the federal minimum wage. industrial and labor relations review, 46(1), 22-37. doi: 10.2307/2524736 card, d. (1992b). do minimum wages reduce employment? a case study of california, 19871989. industrial and labor relations review, 46(1), 38-54. doi: 10.2307/2524737 card, d., & krueger, a.b. (1994). minimum wages and employment: a case study of the fastfood industry in new jersey and pennsylvania. american economic review, 84(4), 772-93. card, d., & krueger, a. (1995). myth and measurement: the new economics of the minimum wage, new jersey: princeton university press. cubit, r.p., & hargreaves-heap, s. p. (1999). minimum wage legislation, investment and human capital. scottish journal of political economy 46(2), 135–157. doi: 10.1111/14679485.00125 cunningham, j. (1991). the impact of minimum wages on youth employment, hours of work, and school attendance: cross-sectional evidence from the 1960 and 1970 censuses. in s. rottenberg (ed.). the economics of legal minimum wages, american enterprise institute. currie, j., & fallick, b. c. (1996). the minimum wage and the employment of youth. journal of http://dx.doi.org/10.17848/9780880994583 187 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 human resources, 31(2), 404-28. doi: 10.2307/146069 deere, d. k., murphy, m., & welch, f. (1995). employment and the 1990-1991 minimum-wage hike. american economic review, 85(2), 232-37. deere, d. k., murphy m., & welch, f. (1996). examining the evidence on minimum wages and employment. in m. kosters (ed.), the effects of the minimum wage on employment, american enterprise institute. díaz-roldán, c. (2017). fiscal performance in monetary unions: how much austerity should be allowed?. panoeconomicus, 64(1), 61. doi: 10.2298/pan140730021d dickens, r., machin, s., & a. manning, (1999). the effects of minimum wages on employment: theory and evidence from britain. journal of labor economics, 17(1), 1-22. doi: 10.1086/209911 dickens, r., & manning, a. (2002). has the national minimum wage reduced uk wage inequality? london school of economics. dinardo, j., fortin, n. m., & lemieux, t. (1996). labor market institutions and the distribution of wages, 1973-1992: a semiparametric approach. econometrica, 64(5), 1001-1044. doi: 10.2307/2171954 fields, g.s. (1994). the unemployment effects of minimum wages. international journal of manpower, 15(2), 74-81. doi: 10.1108/01437729410059323 de fraja, g. (1996). minimum wage legislation, work conditions and employment. cepr discussion paper, 1524. freeman, r. (1994). minimum wages: again!. international journal of manpower, 15(2), 8-25. doi: 10.1108/01437729410059305 freeman, r. (1996). the minimum wage as a redistributive tool. economic journal, 106(436), 639-649. doi: 10.2307/2235571 garcía vega, m. a. (2013, december 28). ¿destruye empleo el salario mínimo?, el país. garcía, n., & ruesga benito, s. m. (2014). ¿qué ha pasado con la economía española? la gran recesión 2.0 (2008 – 2013), editorial pirámide. gordon, r. j. (1995). is there a trade off between unemployment and productivity growth?. nber working paper, 5081. hyslop, d., & stillman, s. (2004). youth minimum wage reform and the labour market, new zealand treasury working paper. islam, i., & nazara, s. (2000). minimum wage and the welfare of indonesian workers. occasional discussion paper series, 3, international labour organization. jiménez, m. (2016). la calidad del empleo y la segmentación laboral en el mercado de trabajo de las medianas y grandes empresas y del sector público de argentina, (phd thesis). retrieved from universidad de alcalá de henares. joliet, a. (2015). the myth of manipulation. the economics of minimum wage. retrieved from https://es.scribd.com/document/54970861/the-myth-of-manipulation-the-economics-ofminimum-wage jones, s. r. g. (1987). minimum wage legislation in a dual labor market. european economic https://es.scribd.com/document/54970861/the-myth-of-manipulation-the-economics-of-minimum-wage https://es.scribd.com/document/54970861/the-myth-of-manipulation-the-economics-of-minimum-wage 188 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 review, 31(6), 1229-46. doi: 10.1016/s0014-2921(87)80015-6 katz, l., & krueger, a. (1992). the effect of the minimum wage on the fast-food industry. industrial and labor relations review, 46(1), 6-21. doi: 10.2307/2524735 kim, t., & taylor, l. (1995). the employment effect in retail trade of california’s 1988 minimum wage increase. journal of business and economic statistics, 13(2), 175-82. doi: 10.1177/001979390906200305 klerman, j. a. (1992). employment effect of mandated health benefits. in health benefits and the workforce, u.s. department of labor, pension and benfits administration, u.s. government printing office. komlos, j. (2015, december 15). why raising the minimum wage is good economics. pbs. retrieved from http://www.pbs.org/newshour/making-sense/why-raising-the-minimum-wageis-good-economics/ l’horty, y., & raults, c. (2004). inflation, minimum wage and other wages: an econometric study on french macroeconomic data. applied economics, 36(4), 277-290. doi: 10.1080/00036840410001674213 lee, d. s. (1999). wage inequalities in the u.s. during the 1980s: rising dispersion or falling minimum wage?. quarterly journal of economics, 114(3), 941-1023. doi: 10.1162/003355399556197 lemos, s. (2003). political variables as instruments for the minimum wage. university college london. retrieved from http://discovery.ucl.ac.uk/2552/ lemos, s. (2004a). are wage and employment effects robust to alternative minimum wage variables? university of leicester. retrieved from https://www.le.ac.uk/economics/research/repec/lec/leecon/dp04-4.pdf?uol_r=d307e306 lemos, s. (2004b). the effect of the minimum wage on prices in brazil, university of leicester. retrived from https://www.le.ac.uk/economics/research/repec/lec/leecon/dp044.pdf?uol_r=d307e306 lester, r. (1946). shortcomings of marginal analysis for wage-employment problems. american economic review, 36, 62-82. lucas, r. (1988). on the mechanics of economic development. journal of monetary economics, 22, 3-42. doi: 10.1016/0304-3932(88)90168-7 macdonald, j. m., & aaronson, d. (2000). how do retail prices react to minimum wage increases? federal reserve bank of chicago. machin, s., & manning, a. (1994). the effects of minimum wages on wage dispersion and employment: evidence from the uk wages councils. industrial and labor relations review, 47(2), 319-29. doi: 10.1177/001979399404700210 machin, s., & manning, a. (1996). employment and the introduction of a minimum wage in britain. economic journal, 106(436), 667-76. doi: 10.2307/2235574 machin, s., & manning, a. (1997). minimum wages and economic outcomes in europe. european economic review, 41(3-5), 733-42. doi: 10.1016/s0014-2921(97)00032-9 machin, s., manning, a., & rahman, l. (2003). where the minimum wage bites hard: 189 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 introduction of minimum wages to a low wage sector. journal of the european economic association, 1(1), 154-180. doi: 10.1162/154247603322256792 manning, a. (1995). how do we know that real wages are too high?. quarterly journal of economics, 110(4), 1111-25. doi: 10.2307/2946650 mejeur, j. (2014 march). maximum divided on minimum wage. state legislatures magazine, march, p.14-17. retrieved from: http://www.ncsl.org/portals/1/documents/magazine/articles/2014/sl_0314-workforce.pdf meyer, r., & wise, d. (1983a). the effects of the minimum wage on the employment and earnings of youth. journal of labor economics, 1, 66-100. meyer, r. & wise, d. (1983b). discontinuous distributions and missing persons: the minimum wage and unemployed youth. econometrica, 51(6), 1677-98. doi: 10.2307/1912112 neumark, d., & wascher, w. (1992). employment effects of minimum and subminimum wages: panel data on state minimum wage laws. industrial and labor relations review, 46(1), 5581. doi: 10.2307/2524738 neumark, d., & wascher, w. (2000). minimum wages and employment: a case study of the fast-food industry in new jersey and pennsylvania: comment. american economic review, 90(5), 1362-96. doi: 10.1257/aer.90.5.1362 neumark, d., & wascher, w. (2003). minimum wages, labor market institutions, and youth employment: a cross-national analysis. feds working paper, 23, national bureau of economic research. nichell, s., & layard, r. (1997). the labour market consequences of technical and structural change. oxford university discussion paper series. novo-corti, i., gonzález-laxe, f., & pociovalisteanu, d. m. (2015). the economic analysis of maritime catastrophes in sensitive areas: the assessment and calculation of damages in the environment and population's way of life. journal of cleaner production, 107, 267-278. doi: 10.1016/j.jclepro.2015.04.084 oecd (1998). making the most of the minimum: statutory minimum wages, employment, and poverty. in employment outlook, (oecd), paris: oecd publications oecd (2004). benefits and wages: oecd indicators. paris: organisation for economic cooperation and development. orazem, p.f., & mattila, j. p. (2002). minimum wage effects on hours, employment and number of firms: the iowa case. journal of labor research, vol. 23(1), 3-23. doi: 10.1007/s12122002-1014-6 pigou, a.c. (1920). the economics of welfare (4th edition). london: macmillan. portugal, p. & cardoso, a. r. (2001). disentangling the minimum wage puzzle: an analysis of job accessions and separations from a longitudinal matched employer-employee data set. cepr disscucion paper, 2844. rebitzer, j. & taylor, l. (1995). the consequences of minimum wage laws: some new theoretical ideas. journal of public economics, 56(2), 245-255. doi: 10.1016/00472727(93)01411-3 190 s. m. ruesga-benito et al. / european journal of government and economics 6(2), 171-190 ruesga, s. m. et al. (2014). economía del trabajo y política laboral (2nd edition). editorial pirámide. schmitt, j. (2015). explaining the small employment effects of the minimum wage in the united states. industrial relations, 54(4), 547-581. doi: 10.1111/irel.12106 shepherd, a. r. (2000). minimum wages and the card–krueger paradox. southern economic journal, 67(2), 469-78. doi: 10.2307/1061482 standing, g., sender, j., & weeks, j. (1996). restructuring the labour market: the south african challenge (1st edition). international labour office. stewart, m. (2001). the impact of the introduction of the uk minimum wage on the employment probabilities of low wage workers. warwick economic research papers, 630. stigler, g. j. (1946). the economics of minimum wage legislation. american economic review, 36, 358-65. swinnerton, k. a. (1996). minimum wages in an equilibrium search model with diminishing returns to labor in production. journal of labor economics, 14(2), 340-355. doi: 10.1086/209814 teulings, c.n. (2000). bridging the gap between ‘joe sixpack’ and ‘bill gates’: on the efficiency of institutions for redistribution. de economist, 148(5), 603-24. doi: 10.1023/a:1004105100439 barack obama (2016, october 8th-14th). the way ahead. the economist. vincer, j. (1976). unemployment effects of minimum wages. journal of political economy, vol. 84(4), s87-104. wellington, a. (1991). effects of the minimum wage on the employment status of youths: an update. journal of human resources, 26(1), 27-46. doi: 10.2307/145715 werner, t., & sell, f. l. (2015). price effects of the minimum wage: a survey data analysis for the german construction sector. labour, 29(3), 310-326. doi: 10.1111/labr.12058 wihbey, j. (2016, july). minimum wage: updated research roundup on the effects of increasing pay, journalist’s resource. retrieved from (http://journalistsresource.org/studies/economics/inequality/the-effects-of-raising-theminimum-wage) abstract. the minimum wage establishment has its origin in the first third of the last century. since its creation has been a focus of continuing controversy and an unfinished debate on economics field. this work reviews the effects of the minimum wag... keywords. political economy; minimum wage; employment; economic inequality jel classification. j38 2. the minimum wage theory: analyzing different approaches 2.2 the efficiency wage approach 2.3 the endogenous development approach and human capital 2.4 employment search models 2.5 segmented market models8f 3. empirical analysis of the effects of the minimum wage 4. concluding remarks references clarifying the concept of social capital through its three perspectives: individualistic, communitarian and macro-social european journal of government and economics 6(2), december 2017, 146-170. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 clarifying the concept of social capital through its three perspectives: individualistic, communitarian and macro-social matias membiela-pollán a, *, josé-atilano pena-lópez a a universidade da coruña, faculty of economics and business, campus de elviña, 15071 a coruña, spain * corresponding author at: department of business, univeridade da coruña, campus elviña, 15071 a coruña, spain. matias.membiela@udc.es article history. received 9 may 2017; first revision required 2 november 2017; accepted 20 november 2017. abstract. the concept of social capital has received increasing attention in recent years. the complexity and multidimensionality that accompany the variable of social capital have caused confusion and ambiguity. this article presents a synthesis of social capital in three perspectives. from the individualistic or micro-social perspective, social capital is an "individual resource" that consists of the networks of relations of the focal subject that bring it a set of instrumental and expressive resources. for the communitarian or meso-social perspective, social capital is a "community resource" or set of attributes and properties present in the social structure (shared norms and values, private trust, closure ...) that facilitate its functioning and collective action. finally, for the macrosocial perspective, social capital is a "macro-social and macro-institutional resource" resting on aspects such as civic-mindedness, general trust and social cohesion, which favors the functioning of the economy and society in general. keywords. social capital; individual social capital; communitarian perspective of social capital; general trust jel classification. a14; z13 1. introduction since the 1990s there has been an exponential growth in the research of social capital. the concept has come to inform a large number of subjects and theories, such as economics, sociology, politics, education, family, health, crime, organizational theory and even technological innovation (tsai and ghoshal, 1998, p. 446; adler and kwon, 2002, p. 17; castiglione et al., 2008a, p. 1; woolcock and radin, 2008; guiso, sapienza and zingales, 2011, p. 418). however, as ahn and ostrom (2008, p. 92) express, criticism is part of the take-off of all theory and paradigm, especially when it is played in the field of new and abstract concepts (castiglione, 2008). in this sense, the main reproach that falls on social capital alludes to its ambiguity as a variable. it is argued that social capital is an umbrella concept because it brings together aspects as diverse as norms and values, rules, social networks and trust (sobel, 2002, p. 145; sánchez-santos and pena-lópez, 2005, p. 142). http://www.ejge.org/ mailto:matias.membiela@udc.es 147 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 despite the seductive appearance of the idea, this evidence results in the construct being ambiguous and difficult to outline (dasgupta, 2005), which is one of the reasons why it is strongly criticized by those who do not recognize its own personality. it is called "a wonderfully elastic term" (adler and kwon, 2002, p. 18), "a notion that means many things to many people", a term without demarcation (dasgupta and serageldin, 2000, pp. xi), a "vague and imprecise concept" (roche, 2004, p. 107) and an "ambiguous if not incoherent concept and valid for everything" (fine, 1999). the term "umbrella concept" is justified; however, it has a lot to do with the "multidimensional" nature of this construct. it is evident that "social" is a broad and complex entity. however, this peculiarity does not negate the relevance it has in economic fact and productivity in general. this theory should go deeper into the analysis of how cultural aspects (norms, values, trust) are linked to structural aspects (social networks); and how the micro and macro dimensions of the concept are linked (portes, 2000). it must also rely on the validity of the indicators used for its empirical measurement, even though the nature of the construct makes it difficult to have a single quantification system or a system of consolidated indicators. this article aims to concretize the perspectives that coexist within the theory of social capital. this point is central because it helps to specify what it is and what the use of this variableparadigm is. in what follows the concept of social capital will be introduced and we shall show some attempts of conceptual classification carried out by some investigators. next, we shall present three great perspectives that coexist within the theory and that encompass in their totality what social capital is and what it is used for. each of these three approaches places social capital within a social level (micro-, meso-, or macro-), which in turn is a resource for such a network. a) from the individualistic or micro-social perspective, social capital is an "individual resource" that consists of the networks of relations of the focal subject that bring him a set of instrumental and expressive resources (bordieu, 1986; lin, 1999a, 2001). b) for the communitarian or meso-social perspective, social capital is a "community resource"; a set of attributes and properties present in the social structure (shared norms and values, particular trust, closure) that facilitate its functioning and collective action (coleman, 1988; bowles and gintis, 2002). c) finally, for the macrosocial perspective, social capital is a "macrosocial and macroinstitutional resource" resting on aspects such as civic-mindedness, general trust and social cohesion that favors the performance of the economy and society in general (putnam, 1993; knack and keefer, 1997). as we have pointed out, the lack of consensus and the ambiguity about the precise meaning of social capital naturally affects the agreement about its measurement system (bjørnskov and svendsen, 2003; pérez garcía et al., 2005; castiglione et al., 2008, p. 6; van deth, 2008, pp. 151-153). in this research we review the empirical methodology that accompanies each of the three perspectives. 148 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 2. the concept of social capital the concept of capital in the economy refers to the installed productive capacity involved in the execution of productive processes, which is strictly speaking called "physical capital" (penalópez and sánchez-santos, 2011, p. 125). another conventional form of capital, commonly cited, is "natural capital", which refers to the natural resources of the environment. however, none of the above reflects the uniqueness of the human being, his capacities and relationships. for this reason, the concepts of "human capital", referring to capacities and knowledge, and "social capital", referring to relational structures and other social attributes, have come to complete the analysis of endowments that presents the economic fact (sánchez-santos and pena-lópez, 2005, pp. 138-139). social capital is an ambitious concept that serves to agglutinate the problem of embeddedness, that is to say, the question of the overlap between social and economic (penalópez and sánchez-santos, 2013). more concisely, it is very useful when studying the contribution of the "social dimension" to the economic fact (membiela, 2013, 2016); with a special focus on "the active value of social networks" (conill, 2004). the reasons for its success lie in its capacity for synthesis and in that it allows, from the supporters who defend social institutions to those of third parties, participatory ways to politically unite, including those who are against regulatory interventions and who favor free social organization as those who are uncomfortable with the free play of market forces (sobel, 2002). it is a new and disputed concept, with a potential that has not yet been fully covered (paldam, 2000). depending on the individual or social approach that is adopted, we focus on seemingly unconnected facts (portes, 2000). as a whole, social capital research offers a complex and controversial conceptual framework that allows the integration of different lines of research from sociology, political science and economics (adler and kwon, 2002). history of the concept the systematic study of "social capital" is recent but several authors point out that the concept has a diversity of roots traceable to the eighteenth and nineteenth centuries (adam and roncevic, 2003). we could say that social capital is a general concept that links with the work of thinkers such as tocqueville (associative activity), john stuart mill (civic engagement), tönnies (community value), durkheim (forms of solidarity), weber (hierarchy and power), locke (civil society), rousseau (network of relations), simmel (group identity) and marx (social class) (bankston and zhou, 2002; brewer, 2003; lazega and pattison 2001; portes and sensenbrenner 1993; putnam, 1995; knack, 2002), and finds antecedents in the work of other authors like aristotle and machiavelli, who studied "civic virtue", and adam smith that approached the grounds of relationality, moral feelings and trust (brewer, 2003); see figure 1. 149 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 figure 1. historical evolution of the concept of social capital. castiglione, van deth and wolleb (2008, p. 2) indicate that in the economic literature of the nineteenth century it is possible to find this term, although without consistency in its use and with a different meaning. robert putnam writes that the first person to refer to social capital as we now understand it, was lyda j. hanifan (1920). this rural educator of the early twentieth century understood that "the community as a whole will benefit from the cooperation of all its parts, while the individual will find in his associations the advantages of the help, sympathy, and social capital background aristotle, machiavelli tocqueville, tönnies, a. smith treatment of similar concepts: civic virtue, civic participation, ring of trust... first real approximation lyda j. hanifan reference to social capital as we know it today: community and cooperation. the concept is focused jane jacobs, glen loury they explain its personality and peculiarities: social network resources founders of the theory of social capital pierre bordieu, james coleman, robert putnam in-depth study of the mechanisms that operate in social capital. "micro" (networks) and "macro" (cultural attributes) approaches. two major orientations: "individualist" and "culturalist" burt, portes, fukuyama, dasgupta, lin, knack, ostrom, glaeser, woolcock ... 150 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 fellowship of his neighbors" (putnam, 2000, p. 19). recently, jane jacobs (1961) and the economist glen loury (1977) have appeared as the first to explicit social capital with a personality and peculiarities accepted by the common researchers, even though the concept seems to require a more systematic elaboration/development. these authors expressed the importance of community ties because they contain the social resources that are useful for the development of individuals and that contribute to give vitality to cities and society (woolcock 1998; tsai and ghoshal, 1998; glaeser et al., 2002, p. 440; durlauf, 2008, p. 602). however, it was the two sociologists, pierre bordieu (1986) and james coleman (1988, 1990), who made the two most relevant contributions with their methodical and detailed study of social capital in the 1980s. we shall return to the work developed by these authors when we deal in depth with the three perspectives of social capital. bordieu, in his classic article "the forms of capital" (1986), understands social capital as the resources available to a subject that are a function of belonging to a group. from a marxist perspective he tries to outline a general theory of social reproduction and develops an instrumental vision in which the network of relationships is the product of "investment strategies, individual or collective, consciously or unconsciously aimed at establishing or reproducing social relationships that are directly usable in the short or long term" (bordieu, 1986, p. 52). for james coleman, social capital corresponds to a variety of aspects of social structures, information channels, obligations and expectations, sets of rules and systems of sanction that facilitate or inhibit the actions of actors (coleman, 1988, p. 98). from the methodological individualism, this american sociologist used the field of education to study the implications of social capital and in his popular article "social capital in the creation of human capital" (1988) he analyzed the relation and the degree of influence with the human capital. in spite of the fact that the roots of the concept go back to the authors mentioned above, it is the writings of robert putnam, making democracy work (1993) and bowling alone (2000), which introduce social capital with clarity in public debate (castiglione et al., 2008, p. 3). these works spurred the interest of scholars and a number of publications on the subject.1 contrary to previous individualistic or class reproduction perspectives, this author develops a more culturalist view of social capital. in the first of the cited works putnam establishes the importance of social capital for the development of democratic institutions. in particular, putnam (1993) understands this concept as "features of social organization, such as trust, norms, and networks that can improve the efficiency of society by facilitating coordinated actions" and points out that the different degree of civic engagement between regions of northern and southern italy had played a relevant role in the greater economic success and superior institutional performance of the north facing south. this author claims a rich associative activity, since, he points out, associations "instill in 1 in these years diverse authors begin to publish articles and essays following the wake of bordieu, coleman and putnam, becoming independent or no longer acknowledging them and acquiring wide relevance; we can use the example of francis fukuyama with his well-known work trust: the social virtues and the creation of prosperity (1995). 151 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 their members habits of cooperation, solidarity and public spirit" (putnam, 1993). putnam's other major work, bowling alone (2000), talks about the causes that have led to the loss of social capital, and focuses on the case of the united states. the author pays attention to the generation of prosocial behaviors and the characteristics of social organizations that facilitate cooperation for mutual benefit (sánchez-santos and pena-lópez, 2005, p. 139). among other names that illustrate articles, books and studies considered "key" in the broad theory of social capital, we must mention the works of mark granovetter, ronald burt, alexander portes, francis fukuyama, patha dasgupta, nan lin, s. keefer, norman uphoff, elinor ostrom, glaeser and michael woolcock. these researchers fall into the two main orientations of social capital mentioned above: the individualist and the culturalist, and, in a more systematic way and following the classification that we propose below, they fall into one of these three approaches: individualistic, communitarian and macrosocial. 3. the three perspectives of social capital as we pointed out earlier, social capital is an ambiguous and disputed concept. the term "umbrella concept" is justified by the coexistence of multiple definitions and measurement systems (dasgupta and serageldin, 2000). robinson et al. (2002) indicate that definitions of social capital are grouped and depend on whether they focus on substance, sources (causes) or effects (consequences). adler and kwon (2002, pp. 19-20) classify the families of definitions based on the relationship that an actor maintains with other agents (bridging views) or based on the structure recreated by actors within the community (bonding views). herreros and de francisco (2001) differentiate "structural definitions" —where social capital is conceived as a set of available resources derived from the participation of the individual in social networks)— and "cultural definitions" —in which social capital carries more than relations because it is a subjective phenomenon that is composed of attitudes and values—. paldam (2000) talks about three families of concepts within the theory of social capital; the one that has "trust"as its axis, which is based on the "ease of cooperation" and the one that identifies it with the "social network". finally, millán and gordon (2004) link the different streams of social capital with the thought of the following authors: james coleman, robert putnam and nan lin. we propose that classifications can be systematized as follows: the first understands social capital as an "individual resource" that consists of the networks of relations of the focal subject that bring him a set of instrumental and expressive resources. the second sees social capital as a "community resource", i.e. as a set of attributes and properties present in a social structure that facilitate its operation and collective action. and the third, conceives social capital as a "macrosocial and macroinstitutional resource" that rests on aspects such as civic-mindedness, social cohesion and general trust that favors the functioning of the economy and society in general (see figure 2). 152 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 individualist or micro-social perspective social capital is an "individual resource" that consists in the networks of relations of the focal subject that bring him a set of instrumental and expressive resources. authors: bordieu (1986), portes (1998), lin (2001). communitarian or meso-social perspective social capital is a "community resource", i.e. a set of attributes and properties present in the social structure that facilitate the cooperation and collective action. authors: coleman (1988), uphoff (1999), bowles and gintis (2002). macro-social and macro-institutional perspective social capital is a "macrosocial and macroinstitutional resource" resting on aspects such as civic-mindedness and social trust that favors the functioning of the economy and society in general authors: putnam (1993), knack and keefer (1997), inglehart (1997). figure 2. the three perspectives of social capital. this division into three approaches corresponds to the triple cataloging of social capital as "individual good", as "collective good" and as "public good" (putnam, 1993, 2000; adler and kwon, 2002, p. 22; dasgupta, 2005; sánchez-santos and pena-lópez, 2005). that is, social capital generates returns for the individual, for the group and for the whole of society. figure 3. micro, meso and macro measurements of social capital. social trust civic-mindedness civic engagement individual social network macro measurements of social capital micro measurements of social capital meso measurements of social capital community attributes three perspectives of social capital 153 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 in what follows we shall analyze each one of the perspectives presented and we shall present the empirical methodology that accompanies them (figure 3). the latter diverges on the basis of four attributes: (1) the level of analysis of social capital: micro, meso or macro. (2) the main study variable: structural (focusing on social networks) or cultural (focusing on norms, values, trust). (3) the method of measurement used: surveys, polls and statistical indicators, community studies and observations, projects and experiments. (4) the type of study: quantitative, qualitative or comparative (membiela, 2016). 3.1 individualistic or micro-social perspective. the "micro" perspective, also known as "networks approach", is very widespread and focuses on the analysis of social capital as an individual resource, where the actor uses his network of relationships2 in the achievement of personal goals, both instrumental (income, status, power...) and expressive (welfare, health, recognition, mutual aid...). the most outstanding reference within this expository line is nan lin. however, one of the initiators of the theory of social capital, pierre bordieu, in his pioneering article the forms of capital (1986) already defined social capital as the "the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance and recognition". from lin's point of view, it could be interpreted as an instrumental action from the perspective of class position defense, in which "the network of relationships is the product of investment strategies, individual or collective, consciously or unconsciously aimed at establishing or reproducing social relationships that are directly usable in the short or long term" (bordieu, 1986, p. 52). in bordieu, useful relationships serve to obtain material and symbolic resources; and thus, the social capital possessed by the agent depends on the network of connections he can mobilize and on the volume of capital (economic, cultural or symbolic) that networks have. "social capital as an individual resource" is reflected in numerous definitions, as shown in table 1. these definitions are grouped within the same approach because they rest on two identifiable elements: the individual as the focal subject of the benefit and the social network as the entity from which the benefit is extracted. however, in their reading matter it can be observed that the authors differ when specifying what social capital is, since some identify it with the amount of networks of relations previously constructed by the person that facilitates the access to a pool of resources (burt, 1992; boxman, de graaf and flap 1991; belliveau, o'reilly and wade 1996; pena-lópez and sánchez-santos, 2013), while for others social capital consists of "resources" that the individual obtains from such networks (bordieu, 1986; lin, 2008; baker, 1990; nahapiet and goshal, 1998). therefore, we can talk about the prominence of networks or resources. 2 these networks are made up of very different bonds (family, community, work, associations...; what is called bonding social capital or bridging social capital depending on the strength of the ties) and have a different origin because they could have been inherited or created by the main actor. 154 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 table 1. definitions of social capital in the individualistic perspective. authors individualistic perspective pierre bordieu (1986) "the aggregate of the actual or potential resources which are linked to possession of a durable network of more or less institutionalized relationships of mutual acquaintance or recognition." nan lin (2001, cap. 2; 2008, p. 51) "resources embedded in one's social networks, resources that can be accessed or mobilized through ties in the networks." baker (1990, p. 619) "a resource that actors derive from specific social structures and then use to pursue their interests; it is created by changes in the relationship among actors." burt (1992, p. 9) "friends, colleagues, and more general contacts through whom you receive opportunities to use your financial and human capital." nahapiet and goshal (1998, p. 243) "the sum of the actual and potential resources embedded within, available through, and ghoshal derived from the network of relationships possessed by an individual or social unit. social capital thus comprises both the network and the assets that may be mobilized through that network." portes (1998, p. 6) "the ability of actors to secure benefits by virtue of membership in social networks or other social structures." knoke (1999, p. 18) "the process by which social actors create and mobilize their network connections within and between organizations to gain access to other social actors' resources." boxman, de graaf and flap (1991, p. 52) "the number of people who can be expected to provide support and the resources those people have at their disposal." belliveau, o´reilly and wade (1996, p. 1572) "an individual's personal network and elite institutional affiliations." from our perspective, the first option is more appropriate because the networks, with their attributes, constitute the "social fact" from which the resources available to the principal actor emerge a posteriori. in this sense, it can be said that networks are instrumental and polymorphic and allow a multitude of resources to be obtained. lin's perspective (1999b, 2008) integrates both proposals. for him, social capital is the set of "resources embedded in one’s social networks, resources that can be accessed or mobilized through ties in the networks." that is, it integrates both material and immaterial resources and networks. now, for this author "social capital" and "social networks" are not equivalent or interchangeable terms; the networks provide the necessary condition to access and use the resources rooted in them but to equate both notions is incorrect (lin, 2008, pp. 58-59). lin (1999b) understands that actors find in their network (of strong and weak ties) a place of exchange where interests (or preferences) are able to coordinate naturally and generate equilibria, just as the market can regulate prices (millán and gordon, 2004). individuals invest in social relationships with the expectation of obtaining benefits and therefore interact with other subjects, assuming that the responses are the consequence of the expectation of a reward or return. this investment in relationships facilitates the flow of information and opportunities and access to differentiated resources, both instrumental and expressive. in lin's model, "network" is understood as the structure that organizes material and immaterial resources and the strategic positions of individuals. for lin, "social interaction" is a means of access to these resources and of approach between positions; and "action" is conceived as the orientation of the behavior destined to obtain returns (millán and gordon, 155 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 2004, p. 741). as millán and gordon (2004, p. 743) point out, in this perspective "the network is the element that organizes resources, and the action aimed at capitalizing them is what makes us talk about social capital." measurement in the individualistic or micro-social perspective. as we have just seen, in the individualistic perspective social capital is perceived as a set of resources which an individual has access to (actively or potentially) as a result of his membership of a social network. to measure social capital in this micro and structural approach, we analyze the extension of the individual network and the resources that it implies. to do this, a survey is usually used to provide information about his access to nodes and the capacity to mobilize them when some resurce is needed. as an example: someone who knows an official in the local administration has access to a number of resources (information, procedures ...); whoever belongs to an extended family has access to greater family support; and whoever knows one or several subjects in the banking sector has better financial advice. nan lin (2008, pp. 54-56) points to two methods: the "name generator" and the "position generator".3 in the name generator, a list of contacts proposed by the focal subject is created for each of the proposed resources (pena-lópez and sánchez-santos, 2013). in the "position generator" a list of socially useful positions for the purposes of the focal subject is presented to a respondent and then asked about the first individual who could give access to that resource, additionally considering the level of relationship between them (family, friend, acquaintance) (erickson, 1996; pena-lópez and sánchez-santos, 2017). both methods have advantages and disadvantages. lin points out that the "name-generator" method is appropriate for testing the depth of close nodes; but is limited in the universe of contacts that the individual (usually three to five)4 proposes (and/or remembers). the "position generator" facilitates the study of the individual's access to the different levels of the social hierarchy (e.g. different occupations and strata in society)5 and shows high degrees of validity and reliability. however, this relational structure-based methodology is less developed and provide less precise information about strong links (lin, 2008, pp. 56-57; see also van der gagg and snijders, 2004, 2005). among the studies that have followed each methodology, campbell et al. (1986) examined the association of individual network resources and socioeconomic status, based on information from the detroit area for the years 1965-1966. following the "name-generator" method, they found that the composition of network resources had a significant association with the status 3 also, van der gaag and snijders (2005) propose a third methodology called resource generator. 4 in this sense, the resources tend to be homogeneous and the relations related to the subject, "homophilic". 5 it is therefore a more favorable methodology when evaluating "heterophilic" relationships and weak ties of the focal subject. 156 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 reached, as measured by occupational prestige and family income (lin, 1999, p. 477). volker and flap (1996), in their study of east germany, used the "position generator" methodology to ask the respondents to identify, among 33 occupations, if they knew someone in each of them, and, if so, to specify what kind of relationship they maintained (family, friends, acquaintances). for the occupational scenario of 1989, the effect of the highest status reached was positive and significant (volker and flap, 1996, p. 478). 3.2 communitarian or mesosocial perspective “gal oya was said by engineers and officials to be the most deteriorated and disorganized irrigation system in the country. yet it became one of the most efficient and cooperatively managed systems, even fairly quickly, once approached with an effective plan for engaging farmers in joint system management. the efficiency of water use was doubled within two years, even before the planned physical rehabilitation was completed, through the introduction of ‘social infrastructure’." (uphoff, 2000, p. 232).6 the social capital approach as a "community resource" highlights the benefit that it generates in the community, compared to the previous one that conceives it as a mere individual good. its axis of study is the "social organization", which contains a set of resources that in addition to serving the individual subject favors cooperation and global functioning. in this approach, however, the term "resource" is not equivalent to its use in the individualistic perspective of social capital. a rule or a sanction constitutes in this approach a resource for the subject and for the collective; while in the previous approach the word resources refers to material goods, influence, or promotion, for example. the definition of social capital provided by james coleman (1988, p. 98; 1990, p. 302) fits this way of understanding social capital, although the author's epistemological approach is based on the idea of individual who behaves with criteria of rationality. coleman (1988) points out that "social capital is defined by its function. it is not a single entity, but a variety of different entities, having two characteristics in common: they all consist of some aspect of a social structure, and they facilitate certain actions of individuals who are within the structure". the resource-entities referred to by this author, which mediate interactions between individuals and favor individual and community productivity, are: obligations and expectations, potential information, norms and effective sanctions, relations of authority, closure and appropriate social organization (coleman, 1998, p. 98),7 all integrable into the concepts of networks and community. 6 the author refers to his experience in a development project in sri lanka that began in 1980. 7 closure means the existence of sufficient ties between a certain number of people to guarantee the observance of norms (portes, 1998). coleman (1988) points out that the emergence of norms and their fulfillment depends not only on the negative external effects of "punishable" actions, but on how closed or open the structure is. while the appropriability of the organization (appropriability) expresses that social ties of one kind (e.g., friendship) often can be used for different purposes (e.g., moral and material support, work and non-work advice) (adler and kwon, 2002, p. 18). 157 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 table 2. definitions of social capital in the communitarian perspective. authors communitarian perspective m. schiff (1992) "the set of elements of the social structure that affects relations among people and are inputs or arguments of the production and/or utility function." fukuyama (1995) "the ability of people to work together for common purposes in groups and organizations." brehm and rahn (1997) "the web of cooperative relationships between citizens that facilitate resolution of collective action problems." fukuyama (1997) "social capital can be defined simply as the existence of a certain set of informal values or norms shared among members of a group that permits cooperation among them." woolcock (1998) "the information, trust, and norms of reciprocity inherent in one's social networks." nahapiet and goshal (1998) "the sum of the current and potential resources embedded within, available through, and derived from the network of relationships possessed by an individual or social unit. social capital thus comprises both the network and the assets that may be mobilized through that network." bowles and gintis (2002) "social capital generally refers to trust, concern for one´s associates, a willingness to live by the norms of one´s community and to punish those who do not." these entities are valued as resources by the actors, employed to achieve certain interests or meet certain needs (millán and gordon, 2004). the norms of reciprocity or closure are appropriable by the subjects and constitute a resource for them but in turn they are an input to the overall functioning of the structure. other authors have proposed some definitions that also bring us closer to the idea of social capital as a "community resource", collecting certain cognitive and structural assets (uphoff, 2000) that favor collective action (see table 2). francis fukuyama (1997) identifies social capital as shared norms and values (within the organization) that favor cooperation; woolcock (1998) and bowles and gintis (2002) point out that it consists of information, trust and norms of reciprocity inherent to the social network (which also favor cooperation); and brehm and rahn (1997) talk about the "network of cooperative relations" (see table 2). let us look at three examples or models that provide insight into the way in which social capital is conceived as a "community resource". i) the presence of a high intra-group social capital in a company that manifests itself in shared goals and values, an adequate work environment and high trust and interpersonal interaction, favors "group human capital" (gui, 2001, p. 158), the global productivity, and therefore the final result that the business seeks. ii) the relations of trust and reciprocity established between the neighbors of a residential estate help them to act more effectively in the face of problems that require collective action, such as prowling thieves or a fire; benefiting even those members who barely interact with other residents (putnam, 2000). iii) the new york diamond market, notes coleman (1988, pp. 98-99), is mainly played by the hyper-laced jewish ethnicity. ties (their closure) provide informal insurance in facilitating transactions in the market, making the merchants belonging to this group feel "compelled" and indebted to the norm avoiding any disaffection influencing the collective good. for example, if one of the individuals of this ethnicity affronts another leading to a significant deception (opportunistic behavior), he must expect a sanction that ranges from warning to expulsion. 158 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 however, it should be emphasized that these forms of social capital do not have to be positive. the same elements (obligations and expectations, norms and effective sanctions, potential information, relations of authority, closure, appropriability and trust) that are rooted in the organization and which are constitutive of social capital, understood as a community resource, can be used for negative results. in these cases, they may favor intra-group objectives, but they are adverse in their contribution to the aggregate social capital due to the (negative) sign of their externality (portes, 1998; paldam, 2000, p. 635; adler and kwon 2002; durlauf, 2008). this is the case of "bad social capital" often exemplified by organizations such as the mafia and the ku klux klan. measurement in the communitarian or meso-social perspective. in this perspective, social capital is considered as a collective property that emerges from the same community, benefiting the totality as an aggregate and each one of its members. as we have seen, coleman (1990, p. 302) points out that "social capital is defined by its function. it is not a single entity, but a variety of different entities having two characteristics in common: they all consist of some aspect of social structure, and they facilitate certain actions of individuals who are within the structure". consequently, coleman (1988) uses as indicators of social capital: (a) "aspects of the social structure" that (b) "facilitate actions of individuals within that structure". regarding the family and the community he focuses on the strength and frequency of interaction and more concretely on whether it is made up of a father and a mother or it is singleparent; on parents’ expectations regarding what the future holds for their children; on residential mobility; on whether or not the mother works; in the relationship between parents and children; on the type of school attended by the student: state school, private or private confessional — that is a question that involves differences in collegial community ties—; and on the existence of links between the students’ parents. in order to explain their impact on academic achievement and school drop-outs, coleman uses the structural/cultural indicators cited because he considers them determinants in the construction of the entire system of obligations and expectations, reliability, channels of information, norms and effective sanctions, closure and "appropriability" of social organization. these are elements that together make up social capital and give it a greater guarantee when they seek the stated objectives; in this case, to favor the creation of human capital. as this researcher points out, a greater number of links between the students’ parents (more community) impels closure in the social network and therefore the capacity of mutual monitoring on the children (and their studies), by facilitating the flow of information and the imposition of sanctions; and by promoting expectations of mutual behavior. other researchers develop case studies examining attributes such as norms and values, cohesion, roles and rules, trust and interpersonal interaction; all of those cognitive, structural, 159 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 and relational assets of social capital that favor collective action and the achievement of the group's objectives. as an example, tsai and ghoshal (1998) used data collected from a multiplicity of respondents belonging to the clusterscomplex business units of a multinational electronics company. the purpose of their work was to examine the relationships that are established between the structural, relational and cognitive dimension of social capital and between these dimensions and the different patterns of resource exchange and product innovations in the company. they observed in this process the positive and significant effect of social interaction (structural dimension), trust (relational dimension) and shared vision (cognitive dimension); concluding that investment in informal social capital and formal social arrangements drive the exchange and combination of productive resources, and promote innovation. on the other hand, portes (1995) and light and karageorgis (1994) analyzed the economic welfare of different communities of immigrants in the united states. they demonstrated that certain groups (koreans in los angeles and chinese in san francisco) thrive better than others (mexicans in san diego and dominicans in new york) because of the social structure in which new immigrants are integrated. successful communities are able to offer assistance to newcomers through informal sources of credit, insurance, migrant family support, language training, and job referrals. less successful communities sometimes show only a short-term commitment to the host country and are almost unable to stabilize their newly arrived members. 3.3 macro-social perspective previous visions of social capital do not enter into the analysis of its effects beyond the individual and/or the group. the third "classical" social capital approach, culturally oriented, sees it as a resource that benefits the broad socio-economic aggregate. without abandoning the benefits that it generates at the individual and group levels (microand meso-), it also conceives it as a "macrosocial" and "macroinstitutional" asset (fukuyama, 1999; sobel, 2002; woolcock, 2001, p. 70; castiglione et al., 2008, p. 7; warren, 2008, p. 123; van deth, 2008, pp. 200-202). in this sense, this proposal argues that social capital is important for the functioning of democracy, for institutional performance in general, for market articulation and economic development, and for social cohesion and the progress of civil society. the work of robert putnam making democracy work: civic traditions in modern italy (1993) is an exponent of this approach. in it, the author offers one of the most generalized definitions of social capital, identifying it with those "features of social organization, such as trust, norms and networks that can improve the efficiency of society by facilitating coordinated actions" (putnam, 1993, p. 167). putnam expresses himself in "macro" terms when referring to "improve the efficiency of society"; a posture which other authors have manifested with posterity. as table 3 shows, knack and keefer (1997) and knack (2002) equate social capital with the degree of civicmindedness of a society. sánchez-santos and pena-lópez (2005, p. 137) point to the importance of trust and cooperative and prosocial behavior for the proper functioning of the 160 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 economy and society, and serageldin (1996) and the world bank8 note the analogy of social capital with "social cohesion". in general and in this line, social capital is identified with different cultural and attitudinal aspects such as civic commitment, social trust, generalized reciprocity, cooperative norms and civic-mindedness, which facilitate social, economic and institutional functioning. within this approach, the researchers differ in the relevance attributed to the aspects that weave the social capital and in the process that links them. the most common divergence is between putnam's position, which understands networks of association as a prior step to the extension of civic-mindedness and social trust, and that of a number of critics who find no significant correlation between trust and associationism. for putnam, social capital is an asset that increases the efficiency of society and favors institutional and economic performance. specifically, and in his definition, it consists of the following three features of social organization: trust, norms and networks. these three elements, which appear in other definitions, tend to establish circular and mutually reinforcing relationships (putnam, 1993, p 177). putnam, like other scholars (e.g. fukuyama, 1995; la porta et al., 2000; dasgupta, 2005; paldam, 2000, pp. 629-630), attaches great importance to "trust" as a "core component of the social capital" that acts as lubricant in the functioning of society and also mediates in any relationship or exchange, with its economic reflection on the reduction of transaction costs (wolleb, 2008, p. 378). table 3. definitions of social capital in the macrosocial perspective. authors macrosocial perspective putnam (1993, p. 167) "features of social organisation, such as trust, norms and networks, which can improve the efficiency of society by facilitating coordinated actions." serageldin (1996) "the glue that holds societies together." ingleheart (1997, p. 188) "a culture of trust and tolerance, in which extensive networks of voluntary associations emerge." thomas (1996, p. 11) "those voluntary means and processes developed within civil society which promote development for the collective whole." turner (1999, p. 53) "those forces that increase the potential economic development of a society by creating and maintaining relationships and patterns of social organization." grootaert and thierry van bastelaer (2002) "the social capital of a society includes the institutions, the relationships, the attitudes and values that govern interactions among people and contribute to economic and social development." pena and santos (2005, p. 137) "social capital represents the idea that trust and the internalization of rules conducive to cooperative and prosocial behavior are the basis for the better functioning of the economy and of society in general." the world bank "social capital refers to the institutions, relationships and norms that make up the quality and quantity of the social interactions of a society. numerous studies show that social cohesion is a critical factor for societies to thrive economically and for development to be sustainable. social capital is not only the sum of the institutions that make up a society, but also the stuff that holds them together." 8 http://web.worldbank.org, under the title "what social capital is". 161 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 at the same time, trust is closely linked to "generalized reciprocity" (putnam, 1993, p 172; putnam, 2000; szreter, 2002; millan and gordon, 2004, p. 726; uslaner, 2008, p. 116). the frequency in interaction, typical of dense social networks, favors the birth and diffusion of norms of reciprocity, fosters trust and contributes to cooperation. putnam refers to civic engagement networks (neighborhood societies, purchasing associations, cooperatives, football clubs) characterized by a horizontal type of interaction9 in which there is a certain public and not merely a private interest that tends to favor generalized reciprocity and to extend the cooperative arrangement beyond the group. for this researcher, associations have a "didactic" effect on the culture of the population (millán and gordon, 2004, p. 734) and are a source of social commitment by fostering internal habits such as cooperation, solidarity, public spirit (putnam, 1993, pp. 89-90), which are then projected to other associations and generalized trust. putnam's perspective is in the line with alexis de tocqueville10 and assumes that institutional and democratic performance is linked to the characteristics of civic life in their common purposes. in his 1993 work, he compares northern and southern italy, finding that the greater civic commitment and participation of civil society and the highest level of trust in the northern regions correlates with increased government quality and institutional performance, and with more satisfactory economic development. this author goes so far as to say that good governance in italy is a byproduct of choral societies and football clubs (putnam, 1993, p. 176). despite being the most widespread approach, it has been very much challenged. it has been argued that it does not take into account aspects such as the size of the network, the sociability within it, the intensity of the contact between its components or the weight of its benignity (since there are violent, racist, criminal organizations, rent seekers...) (paldam 2000; glaeser et al., 2002, p. 444). in addition, researchers debate the role of associations in building trust, generalized reciprocity and cooperation. paxton (2007) expresses that civic participation generates social trust when the associations are interconnected and do not remain isolated. stolle and lewis (2001) point out that there is no empirical evidence to show that associations work as schools of democracy, and levi (1996) doubts that belonging to one type of association leads to overcoming problems of free riding in another; this same researcher warns us furthermore, that putnam does not take into account the structure and philosophy of the italian political system when it comes to establishing the differences between the north and the south. in this way, a new approach emerges within the perspective that considers social capital as a "macrosocial" and "macroinstitutional" resource. social capital is equated with trust and civic 9 a vertical network, however dense and important it may be for its participants, cannot maintain trust and cooperation, since "vertical flows of information are often less reliable than horizontal" (putnam, 1993, p. 174). 10 in tocqueville (1987 [1835-1840]) there is a virtuous link between civic associations, common good, and democracy. in this sense, the author writes: "if men are not linked in a solid and permanent way, it is difficult they act in common" (1987, p. 477). "political parties can be considered as great free schools, where all citizens come to learn the general theory of associations" (1987, p. 481 [1835, 1840]). "in all the countries where political associations are prohibited, civil associations are rare [...]" (tocqueville, 1987; taken from millán and gordon, 2004, pp. 731-732). 162 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 cooperation, but such aspects do not correlate with associationism-civic engagement in the way putnam puts it. this is the line followed by knack and keefer (1997) and knack (2002) when they criticize the american sociologist on the basis of olson's (1982) idea that the nature of associations is an element that must be taken into account since because of this the effect of associationism on trust and economic growth can be neutral. knack (2002, p. 778) sets the example of those interest groups whose objectives undermine efficiency and the general interest. this criticism goes back to the "olson-putnam controversy." because not all types of associations generate trust, the positive effect per se of greater social connectivity cannot be affirmed (sánchez-santos and pena-lópez, 2005, pp. 138, 146). olson (1982) noted that horizontal associations could damage growth by acting as lobbies pursuing their own interests in a preferential policy that imposes a cost on society (knack and keefer, 1997, p. 1271). for these authors, trust and civic cooperation (social capital) are actually the two faces of the same mirror (knack and keefer, 1997, p. 1258) because of the logical and intense relationship between them. an atmosphere of high trust acts by reducing the costs associated with economic activity (in general all transaction costs), it encourages innovation and investment, and creates less dependence of society towards formal institutions. while civic cooperation impacts on the economic outcome by restricting through its (civic) norms the behavior based on the self-interest of the subjects, thus reducing opportunism and leading individuals to favor the public and private good.11 likewise, it promotes the improvement of the political channel, as well as the reduction of bureaucracy and corruption (knack and keefer, 1997, pp. 1252-1254). sánchez-santos and pena-lópez (2005) express themselves in a similar line since they understand the social capital as "moral capital"; while guiso, sapienza and zingales (2011) conceive it as "civic capital". the concept of "moral capital" focuses on the behaviors that underlie the formal and informal relationships that are established within civil society and the ethical attitudes that predispose to general reciprocity and cooperation. while the notion of "civic capital" focuses on the values and beliefs that internalize the common good, fostering cooperative behavior and economic development. social capital ("moral capital" and "civic capital" in their case) benefits the "macro" functioning of the economy and society. however, unlike putnam, they also point out that "although the tautological relationship between trust and social capital is evident, it is not the relationship raised by the sociological perspective, that is, the relation between associationism and social trust" (sánchez-santos and pena-lópez, 2005, pp. 148-149). measurement in the macrosocial perspective. in presenting the third perspective of social capital, we have already indicated that there are two approaches. the most significant exponent of the first one is robert putnam who understands 11 by promoting such rules, it facilitates cooperative solutions to numerous problems of collective action. in this sense, civic norms solve the "prisoner's dilemmas" without imposing external costs while increasing allocative efficiency. 163 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 that associative activity and civic engagement favor institutional performance and economic development by promoting civic-mindedness and social trust. the second approach, which includes authors such as knack and keefer (1997), sánchez-santos and pena-lópez (2005), uslaner (2008) or guiso, sapienza and zingales (2011), equates social capital with civicmindedness and social trust and agrees on the high relevance of these attributes for economic performance; rejecting, however, its necessary correlation with the degree of associative activity. from this perspective —macro and cultural— two measurement options fit. putnam employs the density of voluntary organizations as a tool (paldam, 2000) by measuring civic commitment to the per capita number of groups and associations (church groups, labor unions, sports groups, academic or professional societies, political, and fraternal organizations) which the residents of each region or state belong to (kawachi et al., 1997, p. 1493). nevertheless, several authors criticize the vision of putnam and with this, the instrument used to evaluate the social capital. for martin paldam (2000) there are three main problems: the definition of voluntary organization, the intensity of contacts and the problem of the weight of benignity (for example, criminal and racist associations); while glaeser et al. (2002, p. 445) criticize the fact that the size of the network within the organization and the sociability of the organization are not taken into account. in the second approach, the measurement of social capital is carried out using trust as the base indicator. in this type of research and in these trials the questions of the world values survey (wvs) are posed: "do you think most people would try to take advantage of you if they got a chance, or would they try to be fair?" (item for the study of the perception of probability of opportunism) and " generally speaking, would you say that most people can be trusted or that you can't be too careful in dealing with people?" (item for the study of social trust-distrust). the indicator (trust) is the percentage of respondents who states that "most people can be trusted" (paldam, 2000; kawachi et al., 1997, p. 1492; knack and keefer, 1997, p. 1256). to measure trust in a general sense, the so-called "wallet test" has also been used (paldam, 2000). an experiment is performed in which n wallets are lost in public spaces, and the test analyzes how many of them are returned. knack and keefer (1997, p. 1257) point out that, in their study, the measure of lost wallets is correlated by 0.67 with the level of social trust. we have indicated that social trust and civic-mindedness bear a strong correlation. however, there are systems that are primarily aimed at measuring civic capital. this is the case of guiso, sapienza and zingales (2011, p. 430) who propose to focus on those values and beliefs that foster cooperative behavior. respondents are asked about their opinion on issues such as paying taxes, abuse of public goods, and always negative actions of: unfairly changing their position in a queue, littering the streets, and other similar behaviors; scores range from 1 = never justifiable to 10 = always justifiable. it is suggested that these items with their peers can be good indicators of the prevalence of moral actions and the desire of people to internalize the common good. in a similar sense, knack and keefer (1997, p. 1256), who equate social capital with "civic164 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 mindedness", point out that the strength of norms and civic cooperation is assessed on the basis of justification (justified, never justified or something justified) attributed to the following reasons: a) claiming government benefits to which one is not entitled, b) avoid paying the ticket on public transport, c) lying when paying taxes, d) saving money which has been found at random, and e) covering up the damage done to a parked vehicle. however, these authors point out that those measures are not exempt from contradictions and it is more than likely that respondents will be reluctant to admit their judgments. in this case, it is often used parallel questions (trap questions) that indicate the truthfulness with which the individual responds to the survey. 4. conclusions we started this article with an introduction to the concept and history of social capital, an issue that is subject to increasing attention in the fields of economics, society and politics. as we indicated, the term of "umbrella concept" (social capital is many things for many people) makes it difficult to recognize its utility as a variable that favors socioeconomic and institutional functioning. in this sense, our article has sought to clarify, based on the status questions of the theory, the three great perspectives of the concept, also reviewing the empirical methodology used in each of them. social capital consists of the set of social networks that gives the individual material and immaterial resources (individualistic or micro-social perspective). social capital also lies in the attributes and properties of the social structure that facilitate its operation and collective action (community or meso-social perspective). equally, social capital resides in general cultural aspects such as civic-mindedness and social trust that favor the global functioning of the economy and society (macro-social perspective). social capital is therefore a multidimensional concept with great potential in its application. for this purpose, it is necessary to clarify what it is and what it is for, an issue aided by this presentation of the three great approaches within the theory. after presenting this conceptual clarification, future lines of research will focus on different functions developed by social capital in each of the proposed approaches, as well as the concrete measures that policymakers can take to enhance this factor of development. references adam, f., & roncevic, b. (2003). social capital: recent debates and research trends. social science information 42(2), 155-183. https://doi.org/10.1177/0539018403042002001 adler, p. a., & kwon, s. w., (2002). social capital: prospects for a new concept. the academy of management review 27(1), 17-40. https://doi.org/10.5465/amr.2002.5922314 ahn, t. k., & ostrom, e. (2008). the meaning of social capital and collective action. in d. https://doi.org/10.1177/0539018403042002001 https://doi.org/10.5465/amr.2002.5922314 165 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 1735). new york: oxford university press. baker, w. e. (1990). market networks and corporate behavior. american journal of sociology 96(3), 589-625. https://doi.org/10.1086/229573 bankston, carl l., & zhou, m. (2002). social capital as process: the meanings and problems of a theoretical metaphor? sociological inquiry 72(2), 285-317. https://doi.org/10.1111/1475682x.00017 belliveau, m. a., o'reilly, c.-a., & wade, j.-b. (1996). social capital at the top: effects of social similarity and status on ceo compensation. academy of management journal 39(6), 1568-1593. https://doi.org/10.2307/257069 bjørnskov, ch., & svendsen, g. t. (2003). measuring social capital – is there a single underlying explanation? working paper, n.º 5, department of economics. university of aarhus, aarhus school of business. bordieu, p. (1986). the forms of capital. in j. richardson (ed.), handbook of theory and research for the sociology of education (pp. 241-258). new york: greenwood. bowles, s., & gintis, h. (2002). social capital and community governance. the economic journal 112 (483), 419-436. https://doi.org/10.4337/9781781002841.00006 boxman, e. a. w., de graaf, p. m., & flap, h. d. (1991). the impact of social and human capital on the income attainment of dutch managers. social networks 13(1), 51-73. https://doi.org/10.1016/0378-8733(91)90013-j brehm, j., & rahn, w. (1997). individual-level evidence for the causes and consequences of social capital. american journal of political science 41(3), 999-1023. https://doi.org/10.2307/2111684 brewer, g. a. (2003). building social capital: civic attitudes and behavior of public servants. journal of public administration research and theory 13(1), 5-26. https://doi.org/10.1093/jopart/mug011 burt, r. (1992). structural holds: the social structure if competition. cambridge, ma: harvard university press. campbell, k. e., marsden, p. v., & hurlbert, j.-s. (1986). social resources and socioeconomic status. social networks, 8(1), 97-117. https://doi.org/10.1016/s0378-8733(86)80017-x castiglione, d., van deth, j. w., & wolleb, g (eds.) (2008a). the handbook of social capital. new york: oxford university press. castiglione, d., van deth, j. w. , & wolleb, g (eds.), (2008b). social capital's fortune: an introduction. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 1-12). new york: oxford university press. coleman, j. s. (1988). social capital in the creation of human capital. the american journal of sociology 94, 95-120. https://doi.org/10.1086/228943 coleman, j. s. (1990). foundations of social theory. cambridge: harvard university press. conill sancho, j. (2004). horizontes de economía ética. madrid: editorial tecnos. dasgupta, p. (2005). the economics of social capital. the economic record 81(s1), s58-s70. https://doi.org/10.1086/229573 https://doi.org/10.1111/1475-682x.00017 https://doi.org/10.1111/1475-682x.00017 https://doi.org/10.2307/257069 https://ideas.repec.org/p/hhs/aareco/2003_005.html https://ideas.repec.org/p/hhs/aareco/2003_005.html https://ideas.repec.org/s/hhs/aareco.html https://doi.org/10.4337/9781781002841.00006 https://doi.org/10.1016/0378-8733(91)90013-j https://doi.org/10.2307/2111684 https://doi.org/10.1093/jopart/mug011 https://doi.org/10.1016/s0378-8733(86)80017-x https://doi.org/10.1086/228943 166 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 https://doi.org/10.1111/j.1475-4932.2005.00245.x dasgupta, p., & serageldin. i (eds.) (2000). social capital: a multifaceted perspective. washington, d. c.: world bank. durlauf, s. n. (2008). membership and inequality. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 594-623). new york: oxford university press. erickson, b. (1996). culture, class and connections. american journal of sociology 102(1), 217-51. https://doi.org/10.1086/230912 fine, b. (1999). the developmental state is dead long life social capital? development and change 30(1), 1-19. https://doi.org/10.1111/1467-7660.00105 fukuyama, f. (1995). trust: the social virtues and the creation of prosperity. new york: free press. fukuyama, f. (1997). social capital and the modern capitalist economy: creating a high trust workplace. stern business magazine 4(1), 4-17. fukuyama, f. (1999). social capital and civil society. virginia: the institute of public policy, george mason university. glaeser, e., laibson, d., & sacerdote, b. (2002). an economic approach to social capital. the economic journal 112(483), 437-458. https://doi.org/10.1111/1468-0297.00078 granovetter, m. s. (1973). the strength of weak ties. american journal of sociology 78(6), 1360-1380. https://doi.org/10.1086/225469 grootaert, ch., narayan, d., nyhan, v., & woolcock, m. (2004). measuring social capital: an integrated questionnaire. world bank working paper 18. https://doi.org/10.1596/0-82135661-5 grootaert, ch., & van banstelaer, t. (2002). understanding and measuring social capital. analysis 1(320). https://doi.org/10.1596/0-8213-5068-4 gui, b. (2001). organizaciones productivas con fines ideales y realización de la persona: relaciones interpersonales y horizontes de sentido. in l. bruni, economía de comunión (pp. 149-168). madrid: editorial ciudad nueva. guiso, l., sapienza, p., & l. zingales (2011). civic capital as the missing link. in j. benhabib, m. o. jackson and a. bisin, handbook of social economics (vol. 1a., pp. 417-480). san diego: elsevier. https://doi.org/10.1016/b978-0-444-53187-2.00010-3 hanifan, lyda j. (1920). the community center. boston: silver, burdett and company. herreros, f., & de francisco, a (2001). introducción al capital social como programa de investigación. zona abierta 94/95, 1-46. hirsch, p. m. and levin, d. z. (1999). umbrella advocates versus validity police: a life-cycle model. organization science 10(2), 199-212. https://doi.org/10.1287/orsc.10.2.199 homans, g. c. (1961). social behavior: its elementary forms. new york: harcourt, brace and world. inglehart, r. (1997). modernization and post-modernization: cultural, economic, and political change in 43 societies. princeton, nj: princeton university press. https://doi.org/10.1111/j.1475-4932.2005.00245.x https://doi.org/10.1086/230912 https://doi.org/10.1111/1467-7660.00105 https://doi.org/10.1111/1468-0297.00078 https://doi.org/10.1086/225469 https://doi.org/10.1596/0-8213-5661-5 https://doi.org/10.1596/0-8213-5661-5 https://doi.org/10.1596/0-8213-5068-4 http://www.google.es/search?hl=es&tbo=p&tbm=bks&q=inauthor:%22jess+benhabib%22 http://www.google.es/search?hl=es&tbo=p&tbm=bks&q=inauthor:%22matthew+o.+jackson%22 http://www.google.es/search?hl=es&tbo=p&tbm=bks&q=inauthor:%22alberto+bisin%22 https://doi.org/10.1016/b978-0-444-53187-2.00010-3 https://doi.org/10.1287/orsc.10.2.199 167 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 jacobs, j. (1961). the life and death of great american cities. new york: random house. kawachi, i., kennedy, b. p., lochner, k., & prothrow-stith, d. (1997). social capital, income inequality, and mortality. american journal of public health 87(9), 1491-1497. https://doi.org/10.2105/ajph.87.9.1491 knack, s. (2002). social capital and the quality of government: evidence from the states. american journal of political science 46(4), 772-785. https://doi.org/10.2307/3088433 knack, s., & keefer, p. (1997). does social capital have an economic payoff? the quaterly journal of economics 112(4), 1252-1288. https://doi.org/10.1162/003355300555475 knoke, d. (1999). organizational networks and corporate social capital. in r. leenders and s. m. gabbay (eds.),, corporate social capital and liability (pp. 17-42). boston: kluwer. https://doi.org/10.1007/978-1-4615-5027-3_2 la porta, r., lópez-de-silanes, f., schleifer, a., & vishny, r. w. (2000). trust in large organizations. in p. dasgupta, p., and i. serageldin (eds.), social capital: a multifaceted perspective (pp. 310-324). washington, d. c.: world bank. lazega, e., & pattison, p. e. (2001). social capital as social mechanisms and collective assets: the example of status auctions among colleagues. in n. lin, k. c. cook and r. s. burt (eds.),. social capital: theory and research (pp. 185-208). new york: aldine de gruyter. levi, m. (1996). social and unsocial capital: a review essay of robert putnam’s making democracy work. politics and society 24(1), 45-55. https://doi.org/10.1177/0032329296024001005 light, i., & karageorgis, s. (1994). the ethnic economy. in n. j. smelser and r. swedberg (eds.),, the handbook of economic sociology (pp. 650–677). princeton and nueva york: princeton university press. https://doi.org/10.1515/9781400835584.650 lin, n. (1999a). building a network theory of social capital. connections 22(1), 28-51. lin, n. (1999b). social networks and status attainment. american review of sociology 25(1), 467-487. https://doi.org/10.1146/annurev.soc.25.1.467 lin, n., k. s cook and r. s. burt (2001). social capital: theory and research. new jersey: transaction publishers. lin, n. (2008). a network theory of social capital. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 50-69). new york: oxford university press. loury, g, (1977). women, minorities, and employment discrimination. in p. a. wallace and a. le mund (eds.),. a dynamic theory of racial income differences, lexington.: lexington books. membiela pollán, m. (2013). capital social: glosario. el valor económico de la dimensión social. a coruña: editorial camiño do faro. membiela pollán, m. (2016). la teoría del capital social. a coruña: editorial camiño do faro. millán, r., & gordon, s. (2004). capital social: una lectura de tres perspectivas clásicas. revista mexicana de sociología 66(4), 711-747. https://doi.org/10.2307/3541414 https://doi.org/10.2105/ajph.87.9.1491 https://doi.org/10.2307/3088433 https://doi.org/10.1162/003355300555475 https://doi.org/10.1007/978-1-4615-5027-3_2 https://doi.org/10.1177/0032329296024001005 https://doi.org/10.1515/9781400835584.650 https://doi.org/10.1146/annurev.soc.25.1.467 https://doi.org/10.2307/3541414 168 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 nahapiet, j., & ghoshal, s. (1998). social capital, intellectual capital and the organizational advantage. academy of management review 23(2), 242-266. https://doi.org/10.5465/amr.1998.533225 olson, m. (1982). the rise and decline of nations. new haven: yale university press. paldam, m. (2000). social capital, one or many? definition and measurement. journal of economic surveys 14(5), 629-653. https://doi.org/10.1111/1467-6419.00127 paxton, pamela (2007). association memberships and generalized trust: a multilevel model across 31 countries. social forces 86(1), 47-76. https://doi.org/10.1353/sof.2007.0107 pena lópez, j. a., & sánchez santos, j. m. (2011). el capital social como recurso de la empresa familiar: la familiness. a coruña: netbiblo. pena lópez, j. a., & sánchez santos, j. m. (2013). el capital social individual: lo micro y lo macro en las relaciones sociales. in j. l. veira veira, desigualdad y capital social en españa (pp. 11-31). oleiros: netbiblo. pena lópez, j. a., & sánchez santos, j. m. (2017). individual social capital: accessibility and mobilization of resources embedded in social networks. social networks 49, 1-11. https://doi.org/10.1016/j.socnet.2016.11.003 pérez garcía, f. (dir.), v. montesinos, l. serrano, & fernández de guevara, j. (2005). la medición del capital social: una aproximación económica. madrid: bbva. portes, a. (1995). the economic sociology and sociology of inmigration: a conceptual review. in a. portes (ed.), essays on networks, ethnicity and entrepreneurship. nueva york: rusell sage. portes, a. (1998). social capital: its origins and applications in modern sociology. annual review of sociology 24(1), 1-24. https://doi.org/10.1146/annurev.soc.24.1.1 portes, a. (2000): the two meanings of social capital. sociological forum 15(1), 1-12. portes, a., & sensenbrenner, j. (1993). embeddedness and immigration: notes on the social determinants of economic action. american journal of sociology 98(6), 1320-1350. https://doi.org/10.1086/230191 putnam, r. (1995). bowling alone: america's declining social capital. journal of democracy 6(1), 65-78. https://doi.org/10.1353/jod.1995.0002 putnam, r. (2000). bowling alone. the collapse and revival of american community. new york: simon and schuster. putnam, r., leonardi, r., & nanetti, r. (1993). making democracy work: civic traditions in modern italy. new jersey: princeton university press. robison, l., schmid, l. j., & siles, m. e. (2002). is social capital really capital? review of social economy 60(1), 1-21. https://doi.org/10.2139/ssrn.280892 roche, m. (2004). social policy and social capital: a clear case of putting merit before method? social policy and society 3(2), 97-111. sánchez santos, j. m., & pena lópez, j. a. (2005). actividad asociativa, confianza y generación de capital social: evidencia empírica. ekonomiaz: revista vasca de economía 59, 136-159. https://doi.org/10.5465/amr.1998.533225 https://doi.org/10.1111/1467-6419.00127 https://doi.org/10.1353/sof.2007.0107 http://dialnet.unirioja.es/servlet/libro?codigo=529270 http://dialnet.unirioja.es/servlet/libro?codigo=529270 https://doi.org/10.1016/j.socnet.2016.11.003 https://doi.org/10.1146/annurev.soc.24.1.1 https://doi.org/10.1086/230191 https://doi.org/10.1353/jod.1995.0002 https://doi.org/10.2139/ssrn.280892 169 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 schiff, m. (1992). social capital, labor mobility, and welfare: the impact of uniting states. rationality and society 4(2), 157-175. https://doi.org/10.1177/1043463192004002003 serageldin, i. (1996). sustainability as opportunity and the problem of social capital. brown journal of world affairs 3(2), 187-203. sison, a. (2003). the moral capital of leaders: why virtue matters. cheltenham/northampton: edward elgar publishing. sobel, joel (2002). can we trust in social capital? journal of economic literature 40(1), 139154. https://doi.org/10.1257/0022051027001 stolle, d., & lewis, j. (2001). social capital – an emerging concept. in b. hobson, j. lewis and b. siim (eds.),, key concepts in gender and european social politics, cheltenham: edward elgar press. https://doi.org/10.4337/9781781950340.00012 szreter, s. (2002). the state of social capital: bringing back in power, politics and history. theory and society 31(5), 573-621. https://doi.org/10.1023/a:1021300217590 thomas, c. y. (1996). capital markets, financial markets and social capital. social and economic studies 45(23), 1-23. tsai, w., & ghoshal, s. (1998). social capital and value creation: the role of intrafirm networks. the academy of management journal 41(4), 464-476. https://doi.org/10.2307/257085 tocqueville, a. (1987). la democracia en américa. méxico: fondo de cultura económica. turner, j. h. (2000). the formation of social capital. in p. dasgupta and i. serageldin (eds.),. social capital: a multifaceted perspective (pp. 94–146). washington, dc: the world bank uphoff, n. (2000). understanding social capital: learning from the analysis and experience on participation. in p. dasgupta, and i. serageldin (eds.), social capital: a multifaceted perspective (pp. 215-252). washington, d. c.: world bank. uslaner, e. m. (2008). trust as a moral value. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 101-121). new york: oxford university press. van der gaag, m., & snijders, t. (2004). proposals for the measurement of individual social capital. in h. d. flap and b. völker (eds.),. creation and returns of social capital, (pp. 199218). london: routledge. van der gaag, m., & snijders, t. (2005). the resource generator: measurement of individual social capital with concrete items. social networks 27(1), 1-29. https://doi.org/10.1016/j.socnet.2004.10.001 van deth, j. w. (2008). measuring social capital. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 150-176). new york: oxford university press. volker b., & flap, h. (1996). getting ahead in the gdr: human capital and social capital in the status attainment process under communism. acta sociologica 42(1), 17-34. https://doi.org/10.1080/00016999950079925 warren, m. e. (2008). the nature and logic of bad social capital. in in d. castiglione, j. w. https://doi.org/10.1177/1043463192004002003 https://doi.org/10.1257/0022051027001 https://doi.org/10.4337/9781781950340.00012 https://doi.org/10.1023/a:1021300217590 https://doi.org/10.2307/257085 https://doi.org/10.1016/j.socnet.2004.10.001 https://doi.org/10.1080/00016999950079925 170 m. membiela-pollán and a. pena-lópez / european journal of government and economics 6(2), 146-170 van deth, and g. wolleb (eds.), the handbook of social capital (pp. 122-149). new york: oxford university press. winter, i. (2001). toward a theorized understanding of family life and social capital. working paper 21. melbourne university. wolleb, g. (2008). introduction: social capital and economic development. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 373-385). new york: oxford university press. woolcock, m. (1998). social capital and economic development: toward a theoretical synthesis and policy framework. theory and society 27(2), 151-208. https://doi.org/10.1023/a:1006884930135 woolcock, m. (2001). the place of social capital in understanding social and economic outcomes. canadian journal of policy research 2(1), 1492-1525. woolcock, m., & e. radin (2008). a relational approach to the theory and practices of economic development. in in d. castiglione, j. w. van deth, and g. wolleb (eds.), the handbook of social capital (pp. 411-437). new york: oxford university press. world bank. (21/03/2017). overview: social capital. retrieved from http://www.worldbank.org/en/webarchives/archive/ https://doi.org/10.1023/a:1006884930135 http://www.worldbank.org/en/webarchives/archive/ abstract. the concept of social capital has received increasing attention in recent years. the complexity and multidimensionality that accompany the variable of social capital have caused confusion and ambiguity. this article presents a synthesis of s... keywords. social capital; individual social capital; communitarian perspective of social capital; general trust jel classification. a14; z13 2. the concept of social capital references microsoft word ejge_05_2016-008.docx european journal of government and economics volume 5, number 1 (june 2016) issn: 2254-7088 64 urban population and economic growth: south asia perspective sandip sarker, independent researcher, bangladesh arifuzzaman khan, bangladesh bank, bangladesh mehdad mamur mannan, independent researcher, bangladesh abstract previously economic growth was generally discussed in terms of foreign direct investment (fdi), educational growth, savings, investments, inflation as well as trade openness of a nation. very recently it has been identified that population is one of the major determinants of economic growth of a nation. in the recent years, the study of urbanization has gained a matter of concern in developing countries as it has been recognized as part of a larger process of economic development which is affecting developing countries. south asian countries are one of the emerging economics and growing at a faster rate over the past few years. at the same time, population of south asia is growing at a significant rate. therefore the study has attempted to identify the causal relationship between urban population and economic growth in south asia using a panel data analysis. the study makes use of the augmented dickey-fuller (adf) and phillips-perron (pp), pesaran as well as fisher methods for panel unit root test. the panel pedroni cointegration test suggests that there is long run relationship between the variables. the further panel vector error correction model (vecm) suggests that there is long run causality running from urban population growth to economic growth in south asia. the study concludes that the growth of urban population can have significant impact on economic growth in south asia in the long run. keywords urban population; economic growth; south asia; vector error correction model; panel data. jel classification o18; c32; j10; r11. european journal of government and economics 5(1) 65 1. introduction urbanization can be defined as ‘the demographic process whereby an increasing share of the national population lives within urban settlements.’ urbanization is closely linked to economic development of a society. as an economy develops, relative and absolute changes in demand increase the relative and absolute importance of the manufacturing and service sectors. according to united nations projections, 54 per cent of the world’s population used to reside in urban areas in 2014. many economists welcomed this continued increase in the share of the population living in urban areas in recent decades as they believe urbanization is a positive achievement on the track toward wealth and prosperity. according to this view, urbanization reinforces and improves economic growth and therefore increases the wealth of nations in the long run. according to the urban economists, urbanization is a complex phenomenon. they have also identified some positive effects of urbanization. at first small cities can rapidly develop and new economic and political structures emerge. successful segments within the city attract additional investment, generate increased demand for labour force and elicit migration to the city as a further instrument of urban growth. as the quality of life increases, those cities may become a major attractor for poor rural populations, which will lead to creation of large urban unemployment. the evidence of a positive link between cities (urban areas) and economic development is prodigious. friedberg and hu nt (1995) mentioned that population growth and urbanization go together, and economic development is closely related with urbanization. additionally, bloom, canning and fink (2008), found that there is no evidence that urbanization level affects economic growth rate. this research has highlighted the importance of reassessing the relationship between urbanization and economic growth, and makes us rethink profoundly the popular ideas and practice of accelerated urbanization in developing countries. very recently, turok and mcgranahan (2013) have also argued that it is not urbanization or the city size per se that induces economic growth; rather it is infrastructure and institutional settings of the country. the population of south asia is projected to cross 2 billion at the end of 2020. the world population by 2050 will see 2.3 billion additional people are being added to the world population. no doubt, population is going to be a major challenge for developing countries, especially for countries in south asia. at the same time there has been a tendency of people moving towards urban areas for better opportunities. there are people who are engaged in different economic activities and contributing to our economic development as well. given its impact on the majority of the world’s population and the sustainable development of the global economy, the relationship between urbanization and economic growth is of remarkable scientific and societal importance. therefore our study aims to identify the relationship between urban population and economic growth of south asia from a panel data analysis. the findings can provide important policy suggestion regarding the expected economic benefit obtaining from rapid urbanization and achieve sustainable development. we consider our paper to a complement of the existing line of research that relies heavily on the long run relationship. the rest of the paper has been organized as follows. section 2 summarizes the literatures on relationship between urban population growth and economic growth of the developed and developing economies. section 3 describes the data and methodology of the study. section 4 discusses the econometric procedures and empirical results. section 5 concludes the give policy recommendations. sarker, khan and mannan ● urban population and economic growth 66 2. literature review there are varieties of ways through which urbanization can affect economic growth of a country and the majority of studies suggest that urbanization should have a positive impact on economic growth of a nation. mills and becker (1986) found a positive relationship between the urban population growth and gnp per capita and a negative one between per cent urban and the agricultural share. kasman and duman (2015) found a short-run unidirectional panel causality running from urbanization to gross domestic product (gdp). however, the impact of urbanization on economic growth is diverse across countries based on their level of annual income and development. moomaw and shatter (1993) applied different measures of urbanization and urban concentration on growth and find that metropolitan concentration has a positive impact while urban primacy, defined as concentration of urban population in the largest city, has a negative impact. rosenthal and strange (2003) argued that doubling the size of cities can lead to an increase in productivity in between 3 to 8 per cent that is urban areas generate 85 per cent of gdp in high-income countries. at the same time glaeser, rosenthal and strange (2010) stressed the role of urbanization to promote entrepreneurship. according to them urban populations access to finance helps to promote their ideas and to some extent enable them to do business in their own locality. bacolod, blum and strange (2010) examined the effect of urbanization of economic growth of a nation. they believe that urban concentration causes interactions and engenders spillovers of knowledge and skills which enable people to upgrade their skills and knowledge more efficiently. this raises productivity in urban areas which ultimately affects economic growth of a nation. in the study of quigley (2007) it is evident that mexico city accounts for 20.8 per cent of mexico's total population and 34.3 per cent of gdp, which indicates urbanization contributes significantly to the national gdp. chen et al. (2014) in their work on global pattern of urbanization and economic growth from the last three decades found close links between urbanization levels and gdp per capita. however they found no correlation between urbanization speed and economic growth rate at the global level. however some studies also found negative relationships between urbanization and economic growth. alam et al. (2007) argued that rapid urbanization can negatively impact the economy via its effect on damaging infrastructures. shabu (2010) in her study on urbanization and economic development in developing countries found that there is weak relationship between urban growth and economic development in developing countries. however in the case of developed countries, urbanization has a positive impact on economic growth. castells (2011) believes that the consequence of urbanization on economic growth is a complex phenomenon and depends on several factors such as level of development, stage of urbanization, and nature of main economic activities. there are several studies that focus on population changes and its impact on economic growth. however only a few of those focused on urban population and its economic impact. our study is contributed to the existing literature on population and growth by adding an outcome from a cross-country analysis using panel data in south asia. 3. urbanization and growth in south asia according to the world bank statistics 561,129,658 people lived in urban areas in south asia in 2014. at the same time total gdp is estimated nearly $1,938 billion in south asia in 2014. however this region seems to fail in utilizing its huge urban population. south asia is made up of afghanistan, bangladesh, bhutan, nepal, india, maldives, pakistan and sri lanka which produced just 9 per cent of global gdp in 2013, despite having roughly 14 per cent of the world's urban population. by contrast, east asia includes china, hong kong, japan, mongolia, north and south korea and taiwan fared much better, producing 29 per cent of gdp, and european journal of government and economics 5(1) 67 with 32 per cent of the urban population, in the same period. as the population of south asia is expected to grow by 250 million over the next 15 years, the region could be left further behind if it is unable to harness the benefits of urbanization. between 2000 and 2012, average real gdp per capita increased by almost 56 per cent, from $2,556 to $3,999, with annual gdp per capita growth rates of more than 4.5 percent a year in all countries except nepal and pakistan. and while half of south asians were living on less than $1.25 a day in 1999, fewer than a third were doing so by 2010 (world urbanization prospects database, 2011). since 2000, the report found, south asia has made good strides in achieving greater prosperity with the increase in productivity linked with the growing number of people living in the region’s towns and cities. still, south asia’s share of the global economy remains extremely low relative to its share of the world’s urban population, and, in general, urbanization in the region remains underleveraged. the share of the region’s population officially classified as living in urban settlements increased only marginally from 27.4 per cent in 2000 to 30.9 per cent in 2011, an annual growth rate of 1.1 per cent. according to peter ellis, a lead urban economist with the world bank’s social, urban, rural and resilience global practice, a big reason why south asia is not fully realizing the potential of its cities for prosperity and livability is that its urbanization has been messy and hidden. moreover according to sri mulyani indrawati, managing director and chief operating officer of the world bank, if managed well urbanization could lead to sustainable growth by increasing productivity, allowing innovation and new ideas to emerge in south asia. at the same time un report (2014) explained that south asia has experienced a long period of robust economic growth and it has been among the fastest growing in the world. growth is projected to steadily increase from 7 per cent in 2015 to 7.6 per cent by 2017 through maintaining strong consumption and increasing investment. being the world’s largest working-age population, a quarter of the world’s middle-class consumers, the largest number of poor and undernourished in the world, south asia will play an important role in the global development story as it takes its place in the asian countries. figure 1. growth (%) statistics of south asia source: world development indicator, 2015 (data on afghanistan and maldives are unavailable) sarker, khan and mannan ● urban population and economic growth 68 from figure 1 we can see that bangladesh has been able to maintain a steady growth of 6 per cent over the past few years. india made significant progress in the last two years followed by pakistan as well. at the same time nepal also made a rapid progress in the last three years. 4. data, model and strategy 4.1. data tables 1 and 2 provide the descriptive statistics of the variables used in this study. in this study our dependent variable is gdp (constant 2005 us$) and independent variable is urban population (% of total). here urban population refers to people living in urban areas as defined by national statistical offices. it is calculated using world bank population estimates and urban ratios from the united nations world urbanization prospects (wdi, 2015). on the other hand gdp at purchasers’ prices is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. it is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources. here the study chose gdp at constant price rather than gdp growth (%) because we want to measure the actual changes that take place in the economy of south asian countries. data are in constant 2005 in united states (us) dollars (wdi, 2015). data has been collected from world development indicators of world bank data sheet from 1980 to 2014. we have used panel data of six south asian countries e.g. bangladesh, india, sri lanka, nepal, bhutan & pakistan. the study didn’t use the panel data of other two countries e.g. afghanistan & maldives because of unavailability of long term panel data. eviews 9 has been used to analyze the data. table 1: descriptive statistics of gdp (constant 2005 in billion us$) in south asia country mean median cv maximum minimum bangladesh 54.98 46.76 0.5 118.89 23.15 india 638.88 502.14 0.63 1598.32 203.97 sri lanka 19.66 17.36 0.51 42.49 8.02 pakistan 82.75 77.33 0.42 150.57 31.71 nepal 6.35 6.02 0.43 12.01 2.66 bhutan 0.62 0.46 0.69 1.57 0.13 european journal of government and economics 5(1) 69 table 2: descriptive statistics of urban population (% of total) in south asia country mean median cv maximum minimum bangladesh 23.22 22.43 0.22 33.51 14.85 india 23.32 27.02 0.11 32.37 23.09 sri lanka 18.48 18.48 0.006 18.77 18.29 pakistan 32.64 32.34 0.09 38.3 28.06 nepal 11.94 11.85 0.32 18.24 6.09 bhutan 23.18 22.4 0.38 37.89 10.13 4.2. econometric modeling to analyze the relationship between urban population and economic growth in an analytical manner the study has employed an econometric modeling which consists of panel unit root, panel co-integration and panel causality. at first we have employed panel unit root test to check the stationarity of panel data used. the power of panel unit root tests is considered to be higher compared to individual unit root tests since the information in the time series is enhanced by that contained in the cross-section data. however due to the lack of power of conventional unit root tests, panel unit root tests have been developed quickly in recent twenty years. the traditional augmented dickey fuller (adf) (dickey and fuller, 1979) test is generally considered weak in testing stationarity for panel data since they contain both time and cross section dimensions. studies on panel unit root tests include hadri (2000), im et al. (2003), levin et al. (2002) and maddala and wu (1999). the most popular tests in recent applications are levin et al. (2002) and im et al. (2003). in this study we have checked the unit root of our data by all the methods stated above. if the series are individually integrated of the same order, then we can run co-integration to determine their long run relationship. traditional co-integration tests (johansen, 1988) have been used to identify the existence of long-run relationships between integrated variables in time series data. however as the time passes, new methods have been developed e.g. pedroni (1999), kao (1999). pedroni (1999) proposed a method for panel data cointegration test that can be considered as an extension of the traditional johansen cointegration test (johansen, 1988). in our study we used all the tests to ensure the accuracy of results. however, the presence of a cointegration relationship cannot explain the direction of causality among the variables. in order to analyze the direction of causality, a panel-based vector error correction model (vecm) should be performed bruno (2005). now if there exists a cointegration then the direction of causality can be estimated by using the vecm. through vector error correction model, we can determine their long run as well as short run relationship. in the short-run, then a panel vector error-correction model allows for the interaction of short-run dynamics across cross-sections. the two-equation vecm short run model can be written as follows: ∆y t = c0 + ∑ βi ∆ yt-i + ∑ αi ∆xt-i + ρi ectt-i + ut (1) ∆y t = c0 + ∑ γi ∆xt-i + ∑ ζ i ∆ yt-i + ηi ectt-i + t (2) where ∆ is the difference operator; k, is the numbers of lags, αi and ζ i are parameters to be estimated, ectt-i represents the error terms derived from the long-run cointegration relationship. in each equation, the change in the dependent variable is caused not only by the lag, but also by the previous period’s disequilibrium level. sarker, khan and mannan ● urban population and economic growth 70 5. results and discussion in this study we used four methods that have been used to check the accuracy of stationarity of the panel data. it includes levin, lin and chu, im. pesaran and shin, augmented dickey fuller test and phillips prennon test. table 3 states the unit root test results of dependent variable gdp. here all the four results state that gdp variable becomes stationary in their level and first difference. however they become stationary in their second difference. table 3: unit root test (gdp) method level 1st difference second difference probabilities probabilities probabilities levin, lin and chu t* 1.0000 0.7119 0.0000 im, pesaran and shin wstat 1.0000 0.8801 0.0000 adf fisher chi-square 1.0000 0.3372 0.0000 pp fisher chi-square 1.0000 0.1043 0.0000 similarly, our independent variable, urban population has become non-stationary at its level and first difference (table 4). but they become stationary in its second difference. now as all the variables are integrated of the same order, we can run co-integration to determine their long run relationship. table 4: unit root test (urban population) method level 1st difference second difference probabilities probabilities probabilities levin, lin and chu t* 0.9999 1.0000 0.0000 im, pesaran and shin wstat 1.0000 1.0000 0.0000 adf fisher chi-square 0.9936 0.9837 0.0000 pp fisher chi-square 0.8553 0.0000 0.0000 in the co-integration tests we used all the methods to check the accuracy of the results. table 5 shows the pedroni residual cointegration test results. here all the probability statistics suggest that there exist long run relationships between the variables. here lag has been selected based on the aic value. in general the rule is lower the aic better the model. european journal of government and economics 5(1) 71 table 5: pedroni residual cointegration test trend assumption: deterministic intercept and trend alternative hypothesis: common ar coefs. (within-dimension) weighted statistic prob. statistic prob. panel v-statistic -3.201272 0.9993 -3.000223 0.9987 panel rho-statistic -7.283809 0.0000 -7.765958 0.0000 panel pp-statistic -47.59777 0.0000 -41.87995 0.0000 panel adf-statistic -13.06838 0.0000 -12.02557 0.0000 alternative hypothesis: individual ar coefs. (between-dimension) statistic prob. group rho-statistic -5.572512 0.0000 group pp-statistic -49.11905 0.0000 group adf-statistic -13.19536 0.0000 table 6 portrays the results of kao residual cointegration test. results states that there is also long run relationship between the variables used in this study. or we can interpret that all the variable move together in the long run. table 6: kao residual cointegration test t-statistic prob. adf 1.699912 0.0446 residual variance 1.546520 hac variance 9.057518 finally table 7 shows the traditional johansen fisher panel cointegration test results. it also indicates that there is at least one co-integration between the variables. so all the results suggest that there exist cointegration between the variables used in this study. table 7: johansen fisher panel cointegration test unrestricted cointegration rank test (trace and maximum eigenvalue) hypothesized fisher stat.* fisher stat.* no. of ce(s) (from trace test) prob. (from max-eigen test) prob. none 126.4 0.0000 107.0 0.0000 at most 1 55.98 0.0000 55.98 0.0000 as all variables are cointegrated of the same order, we can run the vecm. table 8 shows the long run causality model. here c (1) is the error correction term which represents the speed of the adjustments towards long run equilibrium. c (1) is the residuals of the one period lag of the cointegrating vector between gdp growth and urban population growth. we see that our c (1) is negative and is also significant which indicates that there is long run causality running from urban population growth to gdp. further we can interpret urban population growth has influence on our dependent variable gdp in the long run. sarker, khan and mannan ● urban population and economic growth 72 table 8: long run causality model coefficient std. error t-statistic prob. c(1) -2.437739 0.179264 -13.59859 0.0000 c(2) 0.894293 0.129008 6.932092 0.0000 c(3) 0.294118 0.076479 3.845758 0.0002 c(4) -8.493409 1.035410 -0.827385 0.4092 c(5) -4.151309 7.905609 -0.525356 0.6000 c(6) 51048906 5.005408 0.102069 0.9188 r-squared 0.731261 mean dependent var 1.185408 adjusted r-squared 0.723538 s.d. dependent var 1.277756 s.e. of regression 6.709609 akaike info criterion 48.12079 sum squared resid 7.815421 schwarz criterion 48.22722 log likelihood -4324.871 hannan-quinn criter. 48.16394 f-statistic 94.69346 durbin-watson stat 1.921217 prob(f-statistic) 0.000000 now we want to estimate whether there is any short run causality between the variables. to do this we used wald statistics test results. table 10 shows short run causality results between urban population (percentage of total) and gdp. the result suggests that there is no short run causality between urban population (percentage of total) and gdp since chi-square value is less than 5 per cent. table 9: short run causality between gdp and urban population growth wald test statistic value df probability f-statistic 0.383257 (2, 174) 0.6822 chi-square 0.766514 2 0.6816 moreover, figure 2 shows that residuals of our model are normally distributed. further our r2 value is 73 per cent which is high. also our f value is found to be significant at 5 per cent level of significance. therefore we can conclude that model used in this study is good fit. figure 2: normality test 0 4 8 12 16 20 24 -0.8 -0.6 -0.4 -0.2 0.0 0.2 0.4 0.6 0.8 series: residuals sample 1981 2014 observations 204 mean 0.006262 median 0.005419 maximum 0.883115 minimum -0.765186 std. dev. 0.302364 skewness 0.175789 kurtosis 3.615320 jarque-bera 4.268924 probability 0.118308 european journal of government and economics 5(1) 73 6. conclusion and policy recommendation so far our study has found long run causal relationship between urban population and economic growth in south asia. however in the short run, our study found no relationship between the variables. therefore it can be said that urban population growth does not immediately affect the economic growth in south asia. rather in the long run, urban population growth significantly affects the economic growth in south asia. our results are very much consistent with the findings of fay and opal (2000); and polese (2005). in south asia, urban population is growing at a higher rate. therefore south asian countries should develop aggressive programs of accelerated urbanization designed to spur economic growth. as the world continues to urbanize, sustainable development challenges will be increasingly concentrated in cities, particularly in the regions like south asia where the speed of urbanization is one of the quickest. governments must implement policies to ensure that the benefits of urban growth are shared equitably and sustainably. sustainability requires cities that generate better income and employment opportunities in order to expand the necessary infrastructure for water and sanitation, energy, transportation as well as information and communications. successful sustainable urbanization requires competent, responsive as well as accountable governments, charged with the management of cities and urban expansion. it also requires appropriate use of information and communication technologies for more efficient service delivery. urbanization is a complex issue that must be assessed not only in terms of urbanization speed or effects of economic growth. in order to increase the quality of the urbanization process, the forward conditions and backward effects must also be explored. it is suggested that the evaluation of urbanization can be improved from the following two aspects. first, the forward conditions of urbanization can be analyzed, such as the number of non-farm jobs, infrastructure level and the supply capacity of public services. second, the forward effects of urbanization should be comprehensively evaluated, including economic, social, and environmental sustainability. policymakers in south asian countries should seek multiple ways of enabling forms of urbanization that contribute to economic growth, an increase in jobs, environmental sustainability, and so on, rather than pursuing accelerated urbanization. our study has attempted to identify the long run relationship between urban population and economic growth in south asia using panel data analysis. we have found long run causal relationship running from urbanization to economic growth in south asia. south asian economies are growing at a faster rate over the last few years. therefore the challenge is how to utilize those rising urban inhabitants to accelerate economic growth in south asia. in the long term, successful urbanization is accompanied by the convergence of living standards between urban and rural areas as economic and social benefits spill beyond urban boundaries. our empirical findings suggest several policy implications. south asian countries should focus on building institutional capacities and applying integrated approaches so as to attain urban sustainability. at the same time in designing urban policies and reforms, it is essential to pay attention to the political economy of urbanization. urban infrastructure service delivery and urban land management need to be strengthened in order to hold the economic growth in south asia. although progress since 2000 has been impressive, the majority of south asia’s cities remain characterized by high levels of poverty, bad housing conditions, and generally poor livability for many of their inhabitants. therefore it is highly recommended that policymakers need to pay special attention to utilize the population portfolio of south asia in order to accelerate the economic growth of that region. however the study has used only two variables for a panel analysis. therefore the study recommends that future research should be conducted by adding more variables e.g. education, corruption index, governance indicator index and so on to interpret the relationship with economic growth in south asia. sarker, khan and mannan ● urban population and economic growth 74 references alam, shaista, ambreen fatima, and muhammad s. butt (2007) ‘sustainable development in pakistan in the context of energy consumption demand and environmental degradation’, journal of asian economics 18(5): 825-837. bacolod, marigee, bernardo s. blum, and william c. strange (2010) ‘elements of skill: traits, intelligences, education, and agglomeration’, journal of regional science 50(1): 245-280. becker charles m. (2007) ‘urbanization and rural-urban migration’ in amitava dutt, and jaime ros, (eds). international handbook of development economics. cheltenham: edward elgar, pp. 516-532. bloom, david e., david canning, and günther fink (2008) ‘urbanization and the wealth of nations’, science 319(5864): 772-775. bruno, giovanni s. f. (2005) ‘approximating the bias of the lsdv estimator for dynamic unbalanced panel data models’, economics letters 87(3): 361-366. castells-quintana, david (2011) ‘agglomeration, inequality and economic growth: cross-section and panel data analysis’ research institute of applied economics working paper 2011/14. chen, mingxing, hua zhang, weidong liu, and wenzhong zhang (2014) ‘the global pattern of urbanization and economic growth: evidence from the last three decades’, plos one 9(8): e103799. dickey, david a., and wayne a. fuller (1979) ‘distribution of the estimators for autoregressive time series with a unit root’, journal of the american statistical association 74(366a): 427-431. ellis. peter (2015) south asia not taking full economic advantage of urbanization. world bank press release. access from: http://www.worldbank.org/en/news/press-release/2015/09/24/south-asia-not-takingfull-economic-advantage-urbanization fay, marianne, and charlotte opal (2000) urbanization without growth: a not so uncommon phenomenon. washington dc: world bank publication. friedberg, rachel m., and jennifer hunt (1995) ‘the impact of immigrants on host country wages, employment and growth (in symposia: immigration)’, the journal of economic perspectives 9(2): 23-44. glaeser, edward l., stuart s. rosenthal, and william c. strange (2010) ‘urban economics and entrepreneurship’, journal of urban economics 67 (1): 01-14. hadri, kaddour (2000) ‘testing for stationarity in heterogeneous panel data’, econometrics journal 3 (2): 148-161. im, kyung so, m. hashem pesaran, and yongcheol shin (2003) ‘testing for unit roots in heterogeneous panels’, journal of econometrics 115(1): 53-74. indrawati, sri mulyani (2015) who sets the rules of the game in asia? east asia forum. access from http://blogs.worldbank.org/voices/who-sets-the-rules-of-thegame-in-asia johansen, søren (1988) ‘statistical analysis of cointegration vectors’, journal of economics dynamic and control 12(2-3): 231-254. kao, chihwa (1999) ‘spurious regression and residual-based tests for cointegration in panel data’, journal of econometrics 90(1): 01–44. kasman, adnan, and yavuz selman duman (2015) ‘co2 emissions, economic growth, energy consumption, trade and urbanization in new eu member and candidate countries: a panel data analysis’, economic modelling 44(1): 97-103. european journal of government and economics 5(1) 75 levin, andrew, chien-fu lin, and chia-shang james chu (2002) ‘unit root tests in panel data: asymptotic and finite-sample properties’, journal of econometrics 108(1): 01-24. maddala, gangadharrao s., and shaowen wu (1999) ‘a comparative study of unit root tests with panel data and a new simple test’, oxford bulletin of economics and statistics 61 (1): 631-652. mills, edwin s., charles m. becker, and satyendra verma (1986) studies in indian urban development. oxford: oxford univ. press pedroni, peter (1999) ‘critical values for cointegration tests in heterogeneous panels with multiple regressors’, oxford bulletin of economics and statistics special issue. polèse, mario (2005) ‘cites and national economic growth: a reappraisal’, urban studies 42(8): 1429-1451. quigley john m. (2007) urbanization, agglomeration and economic development. paper presented at the world bank seminar on growth and development, , washington, d.c. rosenthal, stuart s. and william c. strange (2003) ‘geography, industrial organization, and agglomeration’, review of economics and statistics 85(2): 377393. shabu. terwase (2010) ‘the relationship between urbanization and economic development in developing countries’, international journal of economic development research and investment 1(2): 30-36. the world bank, world development indicators (2015). [api_ny.gdp.mktp.kd.zg_ds2_en_v2]. retrieved from: http://data.worldbank.org/indicator/ny.gdp.mktp.kd turok, ivan, and gordon mcgranahan (2013) ‘urbanization and economic growth: the arguments and evidence for africa and asia’, environment and urbanization 25(1), 465-482. united nations (2005), world population prospects: the 2004 revision. cd-rom edition extended dataset. united nations (2014) world urbanization prospects. united nations, department of economic and social affairs. the link between quality of governance and stock market performance: international level evidence european journal of government and economics 6(1), june 2017, 78-101. 78 european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 the link between quality of governance and stock market performance: international level evidence isaac boadi a,*, hayford amegbe b a school of management, open university, heelen, the netherlands. b university of nairobi, school of business, kenya. * corresponding author at: school of management, open university, valkenburgerweg 177, 6419 at heerlen, netherlands. email: isaac.boadi@ou.nl article history. received 5 january 2017; first revision required 22 may 2017; accepted 16 june 2017. abstract. the present study investigates the link between quality of governance and stock market performance within the context of international markets. the study employed the fixed effect model using 23 countries with complete relevant data for the period spanning from 1996 to 2014. the study reveals that, quality of governance as captured by voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption significantly affect stock market performance. varying effects are produced when the countries are decomposed into income classifications. what is more, the findings and suggestions of this study suggest that quality of government significantly affect foreign direct investment and could have interesting policy implications. the main value of this paper is to examine the link between quality of governance and stock market performance within the context of international markets. keywords. stocks market; quality of governance; fixed effect; country-level jel classification. g13; g18; g19. 1. introduction what is the link between quality of governance framework and stock market performance? a number of cross-country studies for example, klapper and love (2004), durnev and kim (2005), bruno and claessens (2010), among others, have demonstrated that effective functioning of any investment activity hinges on good corporate governance mechanisms which in turn depend on the quality of governance framework of a country. this is because firms do not operate in a vacuum as they are affected by the governance systems in which they operate. empirical evidence has however shown that governance and stock market performance are somewhat inextricable. the united states (us) house of representatives on october 29, 2008 voted down the bailout bill proposed by the treasury and the federal reserve in order to http://www.ejge.org/ boadi & amegbe / european journal of government and economics 6(1), 78-101 79 provide extra liquidity to the troubled us financial markets. global stock markets1 and chicago board options exchange volatility index quickly reacted with an increased by 17 per cent within two hours of the announcement. dow jones industrial average index dropped 778 points a day after indicating clearly that, the uncertainty about the outcome of a critical vote was reflected by both domestic and global stocks. this seems to suggest that financial markets do not operate in a vacuum as they are affected by the governance systems in which they exist. studies by hail and leuz, (2006), hooper et al., (2009) chen et al., (2009) giannetti and koskinen (2010), chiou et al., (2010) among others have opined that quality of a country’s governance is known to be affecting the operation of financial and capital markets. dooley (1998), mckinnon and pill (1997) confirmed that governments are responsible for financial volatility and financial excesses. the novelty of this study over the previous related studies stems from the following grounds: first, although several studies had been conducted in the past, the primarily focus had been on firm-specific corporate governance and stock market performance. the present study beams a searchlight on the country-level governance environment under which firm-specific corporate governance is implemented. second, literature on stock market performance has focused mainly on non-governmental factors such as sovereign spreads (gendreau and heckman, 2003), valuation ratios (campbell and shiller., 1998; maroney et al., 2004; ciaessens et al., 1998; groot and verschoor, 2002), population demographics (bakshi and chen, 1994; bekaert et al, 1998), exchange rates (bailey and chung, 1995; harvey, 1995), and inflation rates (erb et al, 1995; hooker, 2004) as playing a contributory role on stock market performance. as a result, this study adds new empirical evidence to the existing stock of knowledge. third, the sample employed in this study comprises of 23 countries with complete relevant data for the period from 1996 to 2014. the high frequency dataset will ensure that more robust policy recommendations are made. further, country-level governance structures have remarkable informative power related to firm-level measures in explaining stock market performance (krishnamurti et al., 2005; doidge et al., 2007; claessens and yurtoglu, 2013). finally, country heterogeneity is considered as a key relevance of this study. the researchers are of the view that quality of governance is possible to differ from country to country, rendering any evidence on return predictability country-specific. hence, this study seeks to develop country-level governance indices of 23 countries sampled from high income, upper middle income and lower middle income countries, as shown in table 1, to examine whether country-level governance indicators can predict stock market performance and, if so, whether this has implications for investors. the study is structured in five sections. section 2 reviews related literature in the study. section 3 presents the data source and governance indicators. the next section presents the study methodology. section 5 presents the empirical results and section 6 discusses the results and recommendations of the study. 1 see gray and wood (financial times, september 30, 2008, p. b7). boadi & amegbe / european journal of government and economics 6(1), 78-101 80 table 1. list of countries included in the sample by income classification. classification countries region sample period high income(oecd) australia east asia & pacific 1996-2014 belgium europe & central asia 1996-2014 canada north america 1996-2014 germany europe & central asia 1996-2014 chile latin america & caribbean 1996-2014 israel the middle east & north africa 1996-2014 united states north america 1996-2014 united kingdom europe & central asia 1996-2014 japan east asia & pacific 1996-2014 lower middle income ghana sub-saharan africa 1996-2014 india south asia 1996-2014 nigeria sub-saharan africa 1996-2014 morocco the middle east & north africa 1996-2014 philippines east asia & pacific 1996-2014 ukraine europe & central asia 1996-2014 pakistan south asia 1996-2014 upper middle income south africa sub-saharan africa 1996-2014 thailand east asia & pacific 1996-2014 tunisia the middle east & north africa 1996-2014 turkey europe & central asia 1996-2014 brazil latin america & caribbean 1996-2014 china east asia & pacific 1996-2014 mexico latin america & caribbean 1996-2014 source: world bank data 2. literature review the literature on the quality of governance in influencing a country’s stock market performance is large and growing in recent times. recent literature in the early 2000s (see, e.g., la porta et al., 1997, 1998, 2000; ball et al., 2000; gul and qui, 2002; shleifer and wolfenson, 2002) have shifted the focus from firm-specific corporate governance to country-level governance environments. it is an undeniable fact that country-level governance has now become an important policy issue in many countries. in developed countries for instance usa, milyo (2012) reports that stock markets react to governance indicators and events. bechtel (2009) argued that a stable political situation has a systematic investment risk and encourages growth, capital investment and improves overall economy’s performance. jorion and geotzmann (1999) derived that political events had an interruption in the market transactions. chiu et al. (2005) proved that political elections in south korea changed the behavior of foreign investors in financial markets. beaulieu et al. (2006), aktas and oncu (2006), bailey et al. (2005) and frey and waldenstrom (2004) argued that political events had a strong effect on the returns and trading volume of the financial markets. low et al, (2011) found a negative relation between boadi & amegbe / european journal of government and economics 6(1), 78-101 81 governance quality and equity return when they examined the link between country-level governance and global stock market performance. a study by munteanu and brezeanu (2014) which employs a prais-winsten regression that allows for both autocorrelation and heteroskedasticity confirms that government effectiveness and control of corruption present significant positive effects on bank performance across emerging european economies. hooper et al, (2009) reveal a significant and positive association between quality of governance and stock market performance. lombardo and pagano (2000) employ a crosssection of national stock market indices from both developed and emerging markets and confirm the link between quality of governance and the return on equity. simplice (2011) study reveals a direct association between stock market returns and the quality of government institutions. hail & leuz (2003) concur a significant relationship between the strong legal institution's cost of capital. governments have also been blamed for financial volatility and financial excesses (see, for example, dooley 1998, mckinnon and pill 1997). albuquerque and wang (2008) find that high investments are often necessitated by poor investor protection. such results parallel those of harvey (1995), and of giannetti and koskinen (2010) with special reference to emerging markets. the trustworthiness of governments, the reliability of courts, and the full disclosures of accounting standards in the countries of the common law system significantly affect stock market returns (djankov et al., 2003). li and filer (2007) concur that countries which attract more equity investors often practices unbiased and transparent legal systems. the effects of political risk have been found to be statistically significant in emerging stock markets (see, e.g., erb et al., 1996a, diamonte et al., 1996; perotti and van oijen, 2001). lehkonen and heimonen (2015) employ 49 emerging markets panel data to investigate how stock markets respond to changes in democracy and political risk. the study finds evidence to support that stock markets respond significantly to changes in democracy and politics. their results reveal that decline in political risk leads to higher returns. evidence on the negative effects of democracy on the volatility of growth is provided by mobarak (2005). empirical works by bittlingmayer (1998), henry (2000), bekaert and harvey (2000) and bailey and chung (1995) confirm that political uncertainty significantly affects market volatility. research work on the us civil war by willard et al., (1996) discovered that the turning points during the civil war reflected the price of the greenbacks. bailey et al. (2005) examined the iraqi invasion of kuwait in august 1990 and concluded that the impact of political events on the returns affects us-based international equity mutual funds. frey and waldenstrom (2004) studied the fluctuation in the value of government bonds of germany and belgium traded in the zurich and stockholm markets during second world war, confirmed the relationship between political event and stock market performance. in 1995, a referendum conducted in quebec on the separation from the canada federation had a positive impact on the stock market performance. this was good news for financial markets because quebec will remain a part of canadian federation (beaulieu et al., 2006). research by ferguson (2006), which examined the behavior of the london bond market during the first world war, revealed a more significant effect on the international bond yield performance. boadi & amegbe / european journal of government and economics 6(1), 78-101 82 ismail and suhardjo (2001) examined the effect of domestic political events on the jakarta stock market performance. they concluded that the whole market and the overall industry did not show any significant response to all events. chiu et al. (2005) studied the behavior of foreign investors in the four elections of south korea. the results showed that negative relationship exists between kospi 200 index return and the volume of both future and option contracts. onder and simga-mugan (2006) evaluated the impact of economic and political news on the emerging markets; the study took a case of two markets the buenos aires stock exchange (base) in argentina and istanbul stock exchange (ise) in turkey. they examined political and economic news and financial markets from january 1995 to december 1997. the results showed that both economic and political news affects the stock markets. a pioneering study by javed and ahmed (1999) on the impact of the studies on two nuclear tests in pakistan and india in 1998 and 1999 respectively on karachi stock exchange on trading volume, volatility and average return 1995 to 1999 by using the arch model. the study reports that, whereas indian nuclear tests had a significant negative impact on the average rate of returns, trading volume and volatility level increased at kse, pakistani nuclear tests did not affect the average rate of returns significantly. they did, however, increase the volatility and trade volume. masood and sergi (2008), who used bayesian modeling and markov chain monte carlo techniques to examine major political events in pakistan from 1947 to 2006, which had an effect on the stock market, find that the pakistan’s political uncertainty has a risk premium of 7.5 to 12 percent. some researchers like robock (1971), haendal et al. (1975), kobrin (1979) and feils (2000) have examined the impact of political risk on the volatility of investment and observed both negative and positive effects. however overwhelmingly the literature on the link between quality of governance framework and stock market performance has advanced our knowledge, the empirical results have been mixed and contradictory, allowing the present study to add new empirical evidence to the existing stock of knowledge. first, the previous works have focused mainly on the developed and emerging countries for example us and europe where the impact may differ. the study extends the existing literature by examining the link between quality of governance framework and stock market performance by grouping these countries into three income classifications: lower middle income, upper middle income and high income. second, unlike the previous studies where few countries have been sampled and selected, the present study samples 23 countries with complete relevant data for the period from 1996 to 2014 to examine the of impact quality of governance framework on stock market performance. finally, previous studies have shown that the impact of quality of governance framework on stock market performance is inconclusive. whereas some studies have revealed a positive impact on the quality of governance and stock market performance, others have confirmed an inverse relationship. the inconclusiveness of the previous studies indicates that this problem deserves new research. the study therefore hypothesizes the following relationship. h1: quality of government significantly affects stock market performance boadi & amegbe / european journal of government and economics 6(1), 78-101 83 3. data source and governance indicators 3.1 stock returns data the global equity indices were obtained from the world bank development indicators data stream and are broadly representative of each country’s market composition. the equity returns indicating market performance were selected from january 1996 to december 2014, which correspond to the years that governance data are available. panel a of table 2 and panel a, b,c of table 3 reports both the summary statistics of stock returns and summary statistics of stock returns per income classifications from 1996 to 2014. table 2: summary statistics of stock returns from 1996 to 2014. countries credit rating political risk index(pri) with global index as 73 mean s.d. min max australia aaa 88 9.412 2.651 -5.408 7.237 belgium aa 81 9.542 3.132 -6.557 6.383 brazil a− 70 1.035 2.564 -4.904 5.753 canada aaa 93 1.127 2.696 -4.28 6.403 chile aa− 84 8.848 3.511 -4.124 8.399 china aa− 70 9.269 3.138 -3.468 7.376 germany aaa 83 8.347 1.874 -3.848 3.101 ghana b 70 5.337 2.026 -4.952 3.525 india bbb− 70 4.205 2.708 -3.486 5.672 israel aa− 82 4.541 3.901 -5.09 9.416 japan aa− 85 1.75 4.398 -6.414 9.414 mexico a− 80 1.238 3.582 -3.54 1.083 morocco bbb 70 6.851 2.612 -1.914 7.853 nigeria bb59 9.229 3.796 -6.16 7.152 pakistan b− 54 1.67 6.368 -8.225 1.703 philippines bb+ 73 1.825 3.931 -6.192 1.12 south africa a− 69 9.598 3.077 -4.171 5.61 thailand bbb+ 76 1.19 5.127 -7.876 1.472 tunisia bb69 1.911 2.19 -4.702 4.793 turkey bb+ 71 3.029 7.848 -6.24 2.545 ukraine b56 1.717 5.054 -5.717 1.251 united kingdom aaa 86 1.471 4.53 -5.27 1.022 united states aa+ 84 1.559 3.375 -4.507 7.85 boadi & amegbe / european journal of government and economics 6(1), 78-101 84 table 3. summary statistics of stock returns per income classifications from 1996 to 2014. market region mean s.d. min max panel a: lower middle income ghana sub-saharan africa 4.541 3.901 -5.09 9.416 india south asia 1.75 4.398 -6.414 9.414 nigeria sub-saharan africa 1.238 3.582 -3.54 1.083 morocco middle east & north africa 6.851 2.612 -1.914 7.853 philippines east asia & pacific 9.229 3.796 -6.16 7.152 ukraine europe & central asia 1.67 6.368 -8.225 1.703 pakistan south asia 1.825 3.931 -6.192 1.12 panel b: upper middle income south africa sub-saharan africa 9.598 3.077 -4.171 5.61 thailand east asia & pacific 1.19 5.127 -7.876 1.472 tunisia middle east & north africa 1.911 2.19 -4.702 4.793 turkey europe & central asia 3.029 7.848 -6.24 2.545 brazil latin america & caribbean 1.717 5.054 -5.717 1.251 china east asia & pacific 1.471 4.53 -5.27 1.022 mexico latin america & caribbean 1.559 3.375 -4.507 7.85 panel c: high income australia east asia & pacific 9.412 2.651 -5.408 7.237 belgium europe & central asia 9.542 3.132 -6.557 6.383 canada north america 1.035 2.564 -4.904 5.753 germany europe & central asia 1.127 2.696 -4.28 6.403 chile latin america & caribbean 8.848 3.511 -4.124 8.399 israel middle east & north africa 9.269 3.138 -3.468 7.376 united states north america 8.347 1.874 -3.848 3.101 united kingdom europe & central asia 5.337 2.026 -4.952 3.525 japan east asia & pacific 4.205 2.708 -3.486 5.672 3.2 global risk factors and governance data whereas the country’s credit ratings were collected from the international country risk guide (icrg), quality of governance (qg) indicators were also obtained from political risk services inc. (prs) published by the country risk services inc. (crs). the quality of governance (qg) gives a measure of the host country’s political environments. the measure includes many macro-assessments such as government stability, socioeconomic conditions, external and internal conflicts, corruption, law and order, military in politics, religious and ethnic tensions, democratic accountability and bureaucracy quality. to compare and contrast among countries with similar stages of economic development, the study divides the sample into three panels. the first panel comprises lower middle income (ghana, india, nigeria, morocco, philippines, pakistan and ukraine). the second panel includes (south africa, thailand, tunisia, turkey, brazil, china and mexico). the last panel considers high income countries (australia, belgium, boadi & amegbe / european journal of government and economics 6(1), 78-101 85 canada, germany, chile, israel, united states, united kingdom and japan). the present study seeks to develop country-level governance indices namely voice and accountability (democracy and military in politics), political stability and absence of violence (government stability and internal conflict), government effectiveness (bureaucratic quality), regulatory quality (investment profile), rule of law (law and order) and control of corruption (corruption). the choice and justification of country selections were motivated by two main criteria. the first of these is the number of firms for each country reflects the capital market size with a higher number allocated to a country with large capital market size. the second justification is that firms included had available and valid data for the analysis of future performance. 3.3 governance indicators the study employs six government indicators and these are categorised to measure different aspects of governance. the literature on country-level governance indicators as measured by voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption by low, kew & tee (2011) would be discussed in turns: indicator 1: voice and accountability voice and accountability describe how individuals who manage government institutions are selected and the stability of their positions in these organizations. voice and accountability as measured by democracy is not only a complex political and social phenomenon but a subject which needs more attention in developing countries and whether democracy can affect the behavior of the stock markets still remains unexplored. however, regardless of the connection between economic growth and stock market performance, it is possible that democracy and political stability might continue to have a direct impact on stock market performance over and above their impact on economic growth. indicator 2: political stability and absence of violence political stability and absence of violence as measured by government stability and internal conflict although are considered as events that do not have any direct relationship with stock markets but they are considered as one of the main factors that may affect the stock market’s performance. empirical works by bittlingmayer (1998), henry (2000), bekaert and harvey (2000) and bailey and chung (1995) confirm that political uncertainty significantly affects market volatility. boadi & amegbe / european journal of government and economics 6(1), 78-101 86 indicator 3: government effectiveness government effectiveness as a measure of bureaucratic quality concerns perceptions of the quality of public services, the quality of the bureaucracy and the reliability of the government's responsibility to such guidelines. it considers the ability of the government to formulate, initiate and implement sound policies. this index measures the ability of governments to produce and implement good policies and deliver public goods. the expanding and improving stock markets in developing countries demonstrate an important concern of how government frameworks affect stock market performance. governance quality has been adopted by an international organization to measure the state of developing countries. indicator 4: regulatory quality index regulatory quality index which measures the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development kaufmann et al. (2009). regulatory quality looks at the instances of market-unfavorable guidelines such as price controls or inadequate bank supervision, as well as perceptions of the burdens imposed by excessive regulation in areas. low et al, (2011) examine the link between country-level governance and global stock market returns and find the regulatory quality is positively and significantly related to stock market returns. indicator 5: rule of law rule of law selected as our fifth indicator of the study which measures the law and order reflect the extent to which citizens of a country has confidence in the courts, the police, the level of contract administration and the tendency of crime and violence. rule of law is an assessment of the law and order tradition in the country. it summarizes in broad terms the respect of citizens and the state for the institutions that govern their interactions. rule of law considers the effectiveness and predictability of the judiciary, and, more importantly, the enforceability of contracts and proprietary rights. this indicator is a proxy for the success of a society in developing an environment in which fair and predictable rules form the basis for economic and social interactions. the raw of law indicator can be considered as a measure of investor protection arising from the enforcement of equitable principles. chiou et al, (2010) using data on 4916 stocks from 37 countries, confirm equities found in countries practicing english common law often have higher risk premium than equities found in countries practicing civil law. the qualities of judicial system, legal protection of investors' rights, and the social/political environment in a state have significant association on return and risk. various research studies have confirmed the association between performance of financial systems and comprehensive legal protection and an efficient legal system both at the macroeconomic and firm levels and notable among these studies are la porta et al. (1998; 2000). boadi & amegbe / european journal of government and economics 6(1), 78-101 87 indicator 6: corruption corruption is the extent to which public power is exercised for private interest. corruption is not just about bribery. instead, corruption extends beyond bribery to include other exercises of discretionary power in the public sector. in the academic literature, corruption is often defined as the misuse of public office for private gains (shleifer and vishny, 1993; klitgaard, 1991; transparency international, 1995). the world bank calls corruption ‘‘the single greatest obstacle to economic and social development. it undermines development by distorting the rule of law and weakening the institutional foundation on which economic growth depends’’. corruption is a serious social problem that affects all facets of a society (qing et al, 2015). lee and ng (2004) document the empirical relationship between the level of corruption within a country and the valuation of its corporations to shareholders. they find that firms from more corrupt countries trade at significantly lower market multiples, after controlling for other factors. they document that corruption significantly decreases equity values after controlling for many other firmsand country-level control factors. gelos and wei (2006) show that lower country transparency is associated with lower investment from international funds. they also find that during financial crises, international funds flee non-transparent countries by a greater amount than their transparent counterparts. given the link between secrecy and corruption mentioned earlier, it seems that corrupted countries will receive less investment from foreign investors. 4. methodology to estimate the relationship between governance quality and stock market performance, a model for the empirical investigation takes the following form: 𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝐸𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑖𝑖𝑖𝑖 + 𝛽𝛽2𝑃𝑃𝑃𝑃𝑉𝑉𝑉𝑉𝑖𝑖𝑖𝑖 + 𝛽𝛽3𝐺𝐺𝐸𝐸𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛽𝛽4𝑅𝑅𝐸𝐸𝑅𝑅𝑉𝑉𝑖𝑖𝑖𝑖 + [1] + 𝛽𝛽5𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖𝑖𝑖 + 𝛽𝛽6𝑉𝑉𝑅𝑅𝑉𝑉𝑖𝑖𝑖𝑖 + 𝛿𝛿1𝐸𝐸𝐸𝐸𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛿𝛿2𝐺𝐺𝐸𝐸𝑃𝑃𝑖𝑖𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖, where equity index (eqind) being the dependent variable, the study regressed all six explanatory variables namely voice and accountability (vacc), political stability and absence of violence (psav), government effectiveness (geff), regulatory quality (rqua), rule of law (rol) and control of corruption (coc) on the dependent variable. other control variables include inflation(inf) and gross domestic product(gdp). the model was also specified separately for the various income classes (high income, upper middle income and lower middle income). fixed effect estimation is used in this study in order to include the country fixed effects that are largely unobserved in standard econometric models such as the strength of democracy. the fixed effect model has chosen ahead of a pooled ordinary least squares (ols) regression. both fixed effect and random effect estimations were done for all the models, after which a hausman specification test was conducted. the null hypothesis that individual effects are not boadi & amegbe / european journal of government and economics 6(1), 78-101 88 correlated with any of the model’s regressors was rejected (hausman, 1978). thus with systematic differences in the coefficients, a fixed effect model was adjudged more appropriate, hence the choice of fixed effect estimation. fixed effect estimation is used in this study in order to include the country fixed effects that are largely unobserved in standard econometric models such as the strength of democracy. the fixed effect model has chosen ahead of a pooled ols regression. both fixed effect and random effect estimations were done for all the models, after which a hausman specification test was conducted. the null hypothesis that individual effects are not correlated with any of the model’s regressors was rejected (hausman, 1978). thus with systematic differences in the coefficients, a fixed effect model was adjudged more appropriate, hence the choice of fixed effect estimation. 5. empirical results 5.1 descriptive statistics table 2 presents summary statistics of the equity indices for each country from 1996 through to 2014. five countries, namely belgium, china, australia, nigeria and south africa have the highest positive equity indices (>9%) with the lowest being brazil (1.035) during the sample period and correspond with a political risk index of each country exceeding 50 per cent. these results contradict the findings of bekaert and harvey (2000, 2003) and henry (2000) using ifc indices, and found poor stock performance. there are several causes of the better performances in these countries. first, these countries have now come out of a long-term economic recession and depreciation in the domestic currencies cause a positive return in equity markets. this is particularly true for some market like china. second, the selection of sample period excludes major market crashes in a number of states such like “tequila crisis” in 1994, “asian flu” in 1997, and “russian virus” in 1998. third, in recent years, an increase in integration of the global financial market and financial liberation has increased the abnormal returns particularly in less developed markets. the countries of the highest stock return volatility are turkey (7.848) and pakistan (6.368). this could be attributed to the recent turmoil in the region. table 3 presents summary statistics of the equity indices for each country, categorized by sub regions and income classifications mainly lower middle countries (panel a), upper middle income panel (b) and high income (panel c). of the eight countries sampled within the lower middle countries, philippians showed the highest, positive equity indices and exceeds 9 per cent with less than 2 per cent for countries such as india, ukraine, pakistan and nigeria. within the upper middle-income countries, south africa records the highest equity indices with turkey recording the highest stock return volatility. boadi & amegbe / european journal of government and economics 6(1), 78-101 89 table 4. definition of variables and summary statistics. variable definition obs. mean s.d. min max eqind equity indices 437 11.43 38.59 -82.25 254.5 vacc voice and accountability 437 0.62 0.33 0 1 psav political stability and absence of violence 437 0.59 0.28 0 0.936 geff government effectiveness 437 0.59 0.35 0 1 rqua regulatory quality 437 0.60 0.32 0 1 rol rule of law 437 0.57 0.32 0 1 coc control of corruption 437 0.42 0.26 0 1 inf inflation 437 17.86 7.42 8.58 40.5 gdp gdp growth rate 437 6.36 2.54 3.7 15.00 table 4 provides variable definition and summary descriptive statistics for the equity indices indicating each countries stock market performance and governance indicators for the entire study period of 1996 through to 2014. the data employed had a total of 437 observations. the mean of the governance indicators should by definition be zero due to the standardization process in their construction. however, the sample of countries selected based on the availability of stock market data results in a positive mean for each of the governance indicators. on government indicators, voice and accountability (vacc) had the highest positive average score, followed by regulatory quality (rqua), political stability and absence of violence (psav), government effectiveness (geff), rule of law (rol), and control of corruption (coc). overall, with the exception of eqind which had its standard deviation higher than its mean, all the variables had their means higher than their standard deviation. this depicts close spread and high quality of the data. the two macro level control variables exhibited good data qualities by showing the low spread in the distribution (with very low standard deviation compared to their means). as shown in table 5, the results of unit root tests indicated that the three tests employed (harris-tzavalis (1999), breitung (2000) and im-pesaran-shin (2003)) all rejected the null hypothesis of the presence of unit roots in all panels at 1 percent. all the variables used are therefore stationary and appropriate carrying out the panel estimation. boadi & amegbe / european journal of government and economics 6(1), 78-101 90 table 5. unit root tests. variable test harris tzavalis breitung im-pesaran shin eqind -0.1951*** -12.6534*** -10.6018*** vacc -0.0693*** -10.6741*** -9.6484*** psav -0.3124*** -11.6705*** -12.2565*** geff -0.1848*** -9.9423*** -11.2918*** rqua -0.1376*** -9.5214*** -8.1813*** rol -0.3114*** -10.2534*** -12.3119*** coc -0.2047*** -10.0288*** -11.5573*** inf -0.2373*** -10.7468*** -11.3411*** gdp -0.1822*** -9.4428*** -10.2082*** ho: all panels contain unit roots ha: at least one panel is stationary table 6 summarizes the statistics of quality governance indicators from 1996 to 2014. quality of governance (qg) indicators used in this study, which have been obtained from political risk services inc. (prs) and published by the country risk services inc. (crs), includes voice and accountability (democracy and military in politics), political stability and absence of violence (government stability and internal conflict), government effectiveness (bureaucratic quality), regulatory quality (investment profile), rule of law (law and order), and control of corruption (corruption). 5.2 model results from table 7, voice and accountability (vacc) had a positive impact on eqind, and the impact was significant at the 1 per cent level. this implies that the countries with higher voice and accountability rates are likely to have increased equity indices, and the opposite also holds. stated differently, the improvements in democracy lead to higher returns. whereas the findings confirm the results of lehkonen & heimonen, (2015), they contradict the results of (low et al, 2011). regulatory quality (rqua) considers instances of market-unfavorable guidelines i.e. weak bank oversight and surveillance had a positive and significant impact on eqind at the 5 per cent level. this shows that an improvement in regulatory quality results in an increased equity index, and vice versa. such results parallel those of albuquerque and wang (2008) findings suggest that high investments are often necessitated by poor investor protection and support those of harvey (1995), giannetti and koskinen (2010) with special reference to emerging markets. european journal of government and economics 6(1), june 2017, 78-101. 91 table 6. summary statistics of quality governance indicators. voice & accountability political stability & absence of violence government effectiveness market mean s.d. min max mean s.d. min max mean s.d. min max panel a: lower middle income ghana 0.695 0.310 0.000 0.830 0.485 0.220 0.000 0.670 0.632 0.281 0.000 0.750 india 0.365 0.178 0.000 0.500 0.477 0.223 0.000 0.740 0.211 0.125 0.000 0.500 nigeria 0.365 0.178 0.000 0.500 0.477 0.223 0.000 0.740 0.211 0.125 0.000 0.500 morocco 0.586 0.269 0.000 0.750 0.633 0.289 0.000 0.840 0.421 0.187 0.000 0.500 philippines 0.598 0.269 0.000 0.830 0.601 0.276 0.000 0.870 0.618 0.281 0.000 0.750 ukraine 0.618 0.344 0.000 0.880 0.560 0.300 0.000 0.840 0.197 0.105 0.000 0.250 pakistan 0.246 0.215 0.000 0.710 0.473 0.243 0.000 0.780 0.434 0.201 0.000 0.750 panel b: upper middle income south africa 0.680 0.305 0.000 0.830 0.601 0.269 0.000 0.790 0.434 0.201 0.000 0.750 thailand 0.526 0.245 0.000 0.710 0.572 0.278 0.000 0.910 0.434 0.201 0.000 0.750 tunisia 0.430 0.198 0.000 0.710 0.705 0.322 0.000 0.900 0.421 0.187 0.000 0.500 turkey 0.500 0.264 0.000 0.830 0.501 0.231 0.000 0.720 0.421 0.205 0.000 0.750 brazil 0.614 0.277 0.000 0.750 0.610 0.273 0.000 0.780 0.434 0.201 0.000 0.750 china 0.288 0.135 0.000 0.380 0.657 0.303 0.000 0.900 0.421 0.187 0.000 0.500 mexico 0.686 0.313 0.000 0.880 0.580 0.262 0.000 0.820 0.612 0.279 0.000 0.750 panel c:high income australia 0.842 0.375 0.000 1.000 0.647 0.292 0.000 0.940 0.842 0.375 0.000 1.000 belgium 0.834 0.372 0.000 1.000 0.646 0.291 0.000 0.890 0.842 0.375 0.000 1.000 canada 0.838 0.373 0.000 1.000 0.641 0.287 0.000 0.820 0.641 0.287 0.000 0.820 germany 0.819 0.366 0.000 1.000 0.666 0.302 0.000 0.910 0.842 0.375 0.000 1.000 chile 0.618 0.286 0.000 0.790 0.647 0.295 0.000 0.930 0.632 0.281 0.000 0.750 israel 0.613 0.275 0.000 0.830 0.408 0.187 0.000 0.570 0.836 0.373 0.000 1.000 united states 0.727 0.328 0.000 1.000 0.648 0.292 0.000 0.840 0.842 0.375 0.000 1.000 united kingdom 0.838 0.373 0.000 1.000 0.591 0.270 0.000 0.840 0.842 0.375 0.000 1.000 japan 0.727 0.327 0.000 1.000 0.683 0.310 0.000 0.910 0.836 0.373 0.000 1.000 boadi & amegbe / european journal of government and economics 6(1), 78-101 92 table 6 (cont.). summary statistics of quality governance indicators. regulatory quality rule of law control of corruption market mean s.d. min max mean s.d. min max mean s.d. min max panel a: lower middle income ghana 0.518 0.256 0.000 0.770 0.564 0.251 0.000 0.670 0.344 0.167 0.000 0.500 india 0.361 0.181 0.000 0.550 0.292 0.150 0.000 0.500 0.194 0.096 0.000 0.330 nigeria 0.361 0.181 0.000 0.550 0.292 0.150 0.000 0.500 0.194 0.096 0.000 0.330 morocco 0.599 0.281 0.000 0.770 0.722 0.332 0.000 1.000 0.395 0.184 0.000 0.500 philippines 0.590 0.274 0.000 0.820 0.361 0.186 0.000 0.670 0.310 0.164 0.000 0.670 ukraine 0.375 0.225 0.000 0.640 0.529 0.281 0.000 0.670 0.232 0.145 0.000 0.500 pakistan 0.401 0.234 0.000 0.730 0.443 0.204 0.000 0.670 0.266 0.131 0.000 0.500 panel b: upper middle income south africa 0.646 0.311 0.000 0.910 0.348 0.175 0.000 0.670 0.378 0.197 0.000 0.830 thailand 0.528 0.250 0.000 0.730 0.440 0.258 0.000 0.830 0.262 0.130 0.000 0.500 tunisia 0.545 0.263 0.000 0.820 0.699 0.311 0.000 0.830 0.337 0.166 0.000 0.500 turkey 0.498 0.241 0.000 0.820 0.571 0.264 0.000 0.750 0.335 0.158 0.000 0.500 brazil 0.476 0.224 0.000 0.640 0.293 0.143 0.000 0.500 0.382 0.202 0.000 0.670 china 0.480 0.227 0.000 0.820 0.596 0.277 0.000 0.830 0.271 0.138 0.000 0.420 mexico 0.673 0.315 0.000 0.950 0.319 0.165 0.000 0.500 0.310 0.164 0.000 0.670 panel c:high income australia 0.746 0.360 0.000 1.000 0.800 0.358 0.000 1.000 0.661 0.296 0.000 0.830 belgium 0.674 0.326 0.000 0.950 0.708 0.317 0.000 1.000 0.587 0.279 0.000 0.830 canada 0.780 0.372 0.000 1.000 0.817 0.365 0.000 1.000 0.709 0.327 0.000 1.000 germany 0.765 0.358 0.000 1.000 0.717 0.323 0.000 1.000 0.674 0.304 0.000 0.830 chile 0.749 0.342 0.000 0.950 0.678 0.304 0.000 0.830 0.580 0.277 0.000 0.750 israel 0.647 0.305 0.000 0.820 0.699 0.311 0.000 0.830 0.468 0.224 0.000 0.830 united states 0.808 0.366 0.000 1.000 0.726 0.329 0.000 1.000 0.581 0.265 0.000 0.830 united kingdom 0.762 0.353 0.000 1.000 0.777 0.351 0.000 1.000 0.615 0.279 0.000 0.830 japan 0.737 0.362 0.000 1.000 0.717 0.323 0.000 1.000 0.516 0.261 0.000 0.830 european journal of government and economics 6(1), june 2017, 78-101. 93 rule of law (rol), which reflects the extent to which citizens of a country have confidence in the courts, the police, the level of contract administration and the tendency of crime and violence, interestingly is seen to affect eqind rather negatively at 5 per cent level of significance. thus, countries with higher ratings of rule of law were seen to have lower equity indices. the trustworthiness of governments, the reliability of courts, and the full disclosures of accounting standards in the countries of the common law system significantly affect stock market returns (djankov et al., 2003). la porta et al. (1998, 1999) report that the countries with english common law origin provide the strongest legal protection to investors. our empirical results confirm that risk and performance of a financial asset are related to the tradition of commercial law in a country. the stocks in the countries with french/spanish civil law origin are the most volatile. the result validates the study hypothesis. control of corruption (coc) has a negative relation with eqind at 10 per cent level of significance. this implies that the more countries focused on reducing or controlling corruption, the more they scored in terms of their equity index. various studies which support this result include the work of mauro (1995) which affirms that corruption leads to lower levels of investment and growth. wei (1997) finds that corrupted countries attract less foreign direct investment. the presence of corruption reduces investors’ confidence in the rules that guide their businesses and thus boost investors’ risks of dealing in such financial market. (ng, 2006). table 7. regression results from fixed effects estimation. variables equity indices vacc 1.3737*** (0.2984) psav 3.0881 (2.8966) geff -2.0337 (2.1716) rqua 0.6989** (.2911) rol -3.7122*** (0.4282) coc 0.6552* (0.2992) inf -4.4678 (4.2992) gdpg 1.6062*** (0.0454) constant 5.8404*** (1.5331) observations 437 adj. r-squared 0.2392 hausman 61.33 prob > f 0.0000 f(8, 430) 144.25 prob > chi2 0.0000 note. *** p<0.01; ** p<0.05; * p<0.1 table 8 presents the results of fixed effect estimation, by grouping the observations into three: higher income, upper middle income and lower middle income countries. voice and boadi & amegbe / european journal of government and economics 6(1), 78-101 94 accountability (vacc) had a negative impact on eqind in high income and lower middle income countries, and these impacts were significant at the 1 per cent and 10 per cent levels respectively. the implication is that, for both high income and lower middle income countries, an increase in voice and accountability would result in reduced equity indices. for the high income states, political stability and absence of violence (psav) had a positive impact on eqind at the 5 per cent significance level. however, it was not significant for the other income classes. interestingly, government effectiveness (geff) had a significantly negative impact on eqind among the high income countries at 5 per cent level of significance, but also had a 5 per cent significant positive impact among the lower middle income class of countries. thus, an improvement in government effectiveness would reduce equity indices in high income states and increase equity indices in lower middle income states. although regulatory quality (rqua) had no significant impact on eqind among both high and upper middle income countries, the impact was positive and significant among the lower middle income countries at the 5 per cent level. thus, lower income countries would benefit significantly from improvement in regulatory quality. rol had a significant positive impact on eqind among the upper middle income and high income countries at 5 per cent and 10 per cent respective levels of significance, but also had a 1 per cent significant negative impact among the lower middle income class of countries. thus, an improvement in government effectiveness would increase equity indices in high income states but will reduce equity indices in lower middle income states. coc had a significant positive impact on eqind among the upper middle income countries at 1 per cent level of significance. table 8. regression results from fixed effects estimations. income class variables high income upper-middle income lower-middle income vacc -0.7635*** -0.8864 -0.8250*** (0.1217) (0.5352) (0.0259) psav 3.2525* 2.1666 1.6656 (1.5180) (5.2712) (3.1173) geff 0.8569** -0.8249 -0.9343** (0.3622) (0.4556) (0.3931) rqua 2.9788 0.7375 0.8054*** (2.0091) (0.9843) (0.2248) rol 0.6156*** 1.0025** 0.7268* (0.2055) (0.5118) (0.3505) coc 0.7360 0.4252*** 0.5524 (0.5684) (0.0575) 0.3979 inf 0.0347 0.6248 0.0257 (0.3468) (0.4972) (0.1685) gdp 0.5256* 0.6787** 0.4674*** (0.2566) (0.2088) (0.0467) constant -1.7959 -1.0238 -0.9171 (1.7588) (1.0737) (0.5846) observations 171 133 133 adj. r-squared 0.1754 0.2255 0.2712 hausman 77.0535 77.0535 77.0535 prob > f 0.0000 0.0000 0.0000 f-statistic 95.47 100.85 108.43 prob > f 0.0000 0.0000 0.0000 note. *** p<0.01,;** p<0.05; * p<0.1 boadi & amegbe / european journal of government and economics 6(1), 78-101 95 6. discussion and implications the sample employed in this study comprised of 23 countries with complete relevant data for the period from 1996 and 2014. the data is collected from different sources. the global equity indices were obtained from the world bank development indicators datastream and are broadly representative of each country’s market composition to investigate the relation between quality of governance and stock market performance within the context of international markets using a fixed effect model. the study reveals that quality of governance as captured by voice and accountability, political stability and absence of violence, government effectiveness, regulatory quality, rule of law and control of corruption significantly affect stock market performance. varying effects are produced when the countries are decomposed into income classifications. what is more, the findings and suggestions of this study suggest that quality of government significantly affect foreign direct investment and could have interesting policy implications. such examination of the relation between quality of governance and stock market performance using most recent data is a contribution to empirical literature. from the findings of the study, the authors recommend the strategic managerial and policy implications that follow. managerial implication the results of this study offer some strategic implications for security and exchange commission (sec), financial institutions and financial consulting firms. first and foremost the results demonstrate that quality of governance is statistically significant with stock market performance, consistent with hooper et al. (2009). this indicates that strong stock market performance is largely a result of an efficient institutional environment. besides, investors who are not risk lovers would like to invest in countries with mean-variance efficiency. this shows that the quality of governance lowers both transaction and agency costs and creates value for shareholders. the result of this paper incorporates various positions of the world business literature from different perspectives i.e. the call for institutional reforms, standardized rules and regulation (clark, 2003), especially a revitalization of regulation (ngugi, 2003), since a tight regulation will lead to greater market efficiency and low volatility (mutenheri and green, 2003). furthermore, corruption remains dire in the continent and represents a significant risk to financial market development. therefore as a policy recommendation to the governments of the sampled countries especially maintain sound regulation quality and respect for the rule of law (bartels et al., 2009; toumi, 2011; darley, 2012). measures should be put in place in african countries to avoid violence and political instability. policy implication the results of` this study have some policy implications for governments of various markets and other regulators. many stock markets found in the lower middle income countries within the boadi & amegbe / european journal of government and economics 6(1), 78-101 96 sub-saharan african particularly in the french speaking countries are taking too long to pick-up. regulatory environment and institutional arrangements significantly influence stock market development. unfortunately, these unique arrangements have been discounted; therefore, policies that improve the condition of the political environment of a country should be pursued moderately since it has an important impact on the equity market. the findings of this study highlight the importance of the political dimension and thus imply that political reform deserves urgent policy attention in countries with weak political structures. these surely deserve attention in future research. references aggarwal, r., klapper, r., & wysocki, p.d. (2002). portfolio preferences of foreign institutional investors. mit sloan working paper 4328-03. aktas, h., & oncu, s. (2006). the stock market reaction to extreme events: the evidence from turkey. international research journal of finance and economics 6(6), 78-85. albuquerque, r., & wang, n. (2008). agency conflicts, investment, and asset prices. journal of finance 63, 1-40. https://doi.org/10.1111/j.1540-6261.2008.01309.x bailey, b. a., heck, j. l., & wikens, k. a. (2005). international mutual fund performance and political risk. review of pacific basin financial markets and policies 8(1), 167184. https://doi.org/10.1142/s0219091505000312 bailey, w., & chung, y. p. (1995). exchange rate fluctuations, political risk and stock returns: some evidence from an emerging market. journal of financial and quantitative analysis 30(4), 541–562. https://doi.org/10.2307/2331276 ball, r., kothari, s. p., & robin, a. (2000). the effect of international institutional factors on properties of accounting earnings. journal of accounting and economics 29(1), 1– 51. https://doi.org/10.1016/s0165-4101(00)00012-4 bartels, f. l., alladina, s. n., & lederer, s. (2009). foreign direct investment in sub-saharan africa: motivating factors and policy issues. journal of african business 10(2), 141162. https://doi.org/10.1080/15228910903187791 beaulieu, m. c., cosset, j. c., & essaddam, n. (2006). political uncertainty and stock market returns: evidence from the 1995 quebec referendum. the canadian journal of economics/revue canadienne d'économique 39(2), 621-641. https://doi.org/10.1111/j.00084085.2006.00363.x bechtel, m. m. (2009). the political sources of systematic investment risk: lessons from a consensus democracy. the journal of politics 71(2), 661677. https://doi.org/10.1017/s0022381609090525 bekaert, g., & harvey, c. (2000). foreign speculators and emerging equity markets. journal of finance 55(2), 565—613. https://doi.org/10.1111/0022-1082.00220 bekaert, g., & harvey, c. r. (2003). emerging markets finance. journal of empirical finance https://doi.org/10.1111/j.1540-6261.2008.01309.x https://doi.org/10.1142/s0219091505000312 https://doi.org/10.2307/2331276 https://doi.org/10.1016/s0165-4101(00)00012-4 https://doi.org/10.1080/15228910903187791 https://doi.org/10.1111/j.0008-4085.2006.00363.x https://doi.org/10.1111/j.0008-4085.2006.00363.x https://doi.org/10.1017/s0022381609090525 https://doi.org/10.1111/0022-1082.00220 boadi & amegbe / european journal of government and economics 6(1), 78-101 97 10(1), 3−56. https://doi.org/10.1016/s0927-5398(02)00054-3 bittlingmayer, g. (1998). output, stock volatility, and political uncertainty in a natural experiment: germany 1880–1940. journal of finance 53(6), 2243– 2258. https://doi.org/10.1111/0022-1082.00090 breitung, j. (2001). the local power of some unit root tests for panel data. in b. h. baltagi, t. b. fomby, & r. carter hill (eds.), nonstationary panels, panel cointegration, and dynamic panels (advances in econometrics 15, pp. 161-177). emerald group publishing limited. bruno, v, & claessens, s. (2010). corporate governance and regulation: can there be too much of a good thing? journal of financial intermediation 19(4), 61– 82. https://doi.org/10.1016/j.jfi.2009.10.001 chen, c. w., chen, z., & wei, j. k. c. (2009). legal protection of investors, corporate governance, and the cost of equity capital. journal of corporate finance 15(3), 273289. https://doi.org/10.1016/j.jcorpfin.2009.01.001 chiou, w. j. p., lee, a. c., & and lee, c. f. (2010). stock return, risk, and legal environment around the world. international review of economics and finance 19(1), 95– 105. https://doi.org/10.1016/j.iref.2009.05.001 chiu, c. l., chen, c. d., & tang, w. w. (2005). political elections and foreign investor trading in south korea's financial markets. applied economics letters 12(11), 673677. https://doi.org/10.1080/13504850500190097 claessens, s., & laeven, l. (2003). financial development, property rights, and growth. journal of finance 58(6), 2401–2436. https://doi.org/10.1046/j.1540-6261.2003.00610.x claessens, s., & yurtoglu, b. b. (2013). corporate governance in emerging markets: a survey. emerging market review 15(1), 1–33. https://doi.org/10.1016/j.ememar.2012.03.002 clark, r. a. (2003). african securities market designs. journal of african business 4(3), 83102. https://doi.org/10.1300/j156v04n03_05 darley, w. k. (2012). increasing sub-saharan africa's share of foreign direct investment: public policy challenges, strategies, and implications. journal of african business 13(1), 6269. https://doi.org/10.1080/15228916.2012.657981 diamonte, l. r., liew, m. j., & stevens, l. r. (1996). political risk in emerging and developed markets. journal of portfolio management 52(3), 7176. https://doi.org/10.2469/faj.v52.n3.1998 djankov, s., la porta, r., lopez-de-silanes, f., & shleifer, a. (2003). courts. quarterly journal of economics 118, 453−517. https://doi.org/10.1162/003355303321675437 djankov, s., la porta, r., lopez-de-silanes, f., & shleifer, a., (2008). the law and economics of self-dealing. journal of financial economics 88(3), 430– 465. https://doi.org/10.1016/j.jfineco.2007.02.007 doidge, c., karolyi, a. g., & stulz, r. m. (2007). why do countries matter so much for corporate governance? journal of financial economics 86(1), 139. https://doi.org/10.1016/j.jfineco.2006.09.002 dooley, m. (1998). a model of crises in emerging markets. nber working paper 6300. https://doi.org/10.1016/s0927-5398(02)00054-3 https://doi.org/10.1111/0022-1082.00090 https://doi.org/10.1016/j.jfi.2009.10.001 https://doi.org/10.1016/j.jcorpfin.2009.01.001 https://doi.org/10.1016/j.iref.2009.05.001 https://doi.org/10.1080/13504850500190097 https://doi.org/10.1046/j.1540-6261.2003.00610.x https://doi.org/10.1016/j.ememar.2012.03.002 https://doi.org/10.1300/j156v04n03_05 https://doi.org/10.1080/15228916.2012.657981 https://doi.org/10.2469/faj.v52.n3.1998 https://doi.org/10.1162/003355303321675437 https://doi.org/10.1016/j.jfineco.2007.02.007 https://doi.org/10.1016/j.jfineco.2006.09.002 boadi & amegbe / european journal of government and economics 6(1), 78-101 98 cambridge, mass.: national bureau of economic research. durnev, a, & han kim, e. (2005). to steal or not to steal: firm attributes, legal environment, and valuation. journal of finance 60, 61–93. https://doi.org/10.1111/j.15406261.2005.00767.x erb, c., harvey, c., & viskanta, t. (1996). expected returns and volatility in 135 countries. journal of portfolio management 25, 46–58. https://doi.org/10.3905/jpm.1996.409554 feils, d.a. (2000). the impact of political risk on the foreign direct investment decision: a capital budgeting analysis. emerging economist 45, 129143. https://doi.org/10.1080/00137910008967541 ferguson, n. (2006). political risk and the international bond market between the 1848 revolution and the outbreak of the first world war. economic history review 59, 70112. https://doi.org/10.1111/j.1468-0289.2005.00335.x fisman, r. (1991). estimating the value of political connections. american economic review 94, 1095-102. frey, b. s., & waldenstrom, d. (2004). markets work in war: worldwar ii reflected in the zurich and stockholm bond markets. financial history review 11, 5167. https://doi.org/10.1017/s0968565004000046 gelos, r. g., & wei, s.-j. (2006). transparency and international investor behavior. imf working paper 02/174. giannetti, m., & koskinen, y. (2010). investor protection, equity returns, and financial globalization. journal of financial and quantitative analysis 45, 135168. https://doi.org/10.1017/s0022109009990524 gul, f., & qui, h. (2002). legal protection, corporate governance and information asymmetry in emerging financial markets. ssrn working paper. https://doi.org/10.2139/ssrn.298169 gupta, s., davoodi, h., & tiongson, e. (2001). corruption and the provision of health care and education services. in a. k. jain (ed.), the political economy of corruption (pp. 111-141), ny: routledge. haendel, d., west, g. t., & meadow, r. g. (1975). overseas investment and political risk (no. 21). foreign policy research institute. hail, l., & leuz, c. (2006). international differences in the cost of equity capital: do legal institutions and securities regulation matter? journal of accounting research 44, 485 531. https://doi.org/10.1111/j.1475-679x.2006.00209.x hail, l., & leuz, c. (2003). international differences in the cost of equity capital: do legal institutions and securities regulation matter?" ecgi—law working paper 15/2003, brussels. harris, r. d., & tzavalis, e. (1999). inference for unit roots in dynamic panels where the time dimension is fixed. journal of econometrics 91(2), 201-226. https://doi.org/10.1016/s03044076(98)00076-1 harvey, c. (1995). predictable risk and returns in emerging markets. review of financial studies 8(3), 773-816. https://doi.org/10.1093/rfs/8.3.773 hausman, j. a. (1978), "specification tests in econometrics". econometrica 46(6), 1251– https://doi.org/10.1111/j.1540-6261.2005.00767.x https://doi.org/10.1111/j.1540-6261.2005.00767.x https://doi.org/10.3905/jpm.1996.409554 https://doi.org/10.1080/00137910008967541 https://doi.org/10.1111/j.1468-0289.2005.00335.x https://doi.org/10.1017/s0968565004000046 https://doi.org/10.1017/s0022109009990524 https://doi.org/10.2139/ssrn.298169 https://doi.org/10.1111/j.1475-679x.2006.00209.x https://doi.org/10.1016/s0304-4076(98)00076-1 https://doi.org/10.1016/s0304-4076(98)00076-1 https://doi.org/10.1093/rfs/8.3.773 boadi & amegbe / european journal of government and economics 6(1), 78-101 99 1272. https://doi.org/10.2307/1913827 henry, p. b. (2000), "stock market liberalization, economic reform, and emerging market equity prices", journal of finance 55(2), 529−564. https://doi.org/10.1111/0022-1082.00219 hooper, v., sim, a. b., & uppal, a. (2009). governance and stock market performance. economic systems 33(2), 93–116. https://doi.org/10.1016/j.ecosys.2009.03.001 im, k. s., m. h. pesaran, & shin, y. (2003). testing for unit roots in heterogeneous panels". journal of econometrics 115(1), 53-74. https://doi.org/10.1016/s0304-4076(03)00092-7 ismail, i., & suhardjo, h. (2001). the impact of domestic political events on an emerging stock market: the case of indonesia. proceedings of asia pacific management conference, 235262. javed, a. y., & ahmed, a. (1999). the response of karachi stock exchange to nuclear detonation. the pakistan development review 38(4), 777-786. johnson, s., kaufmann, d., & zoido-lobaton, p. (1998). regulatory discretion and the unofficial economy. american economic review 88(1), 387-92. jorion, p., & geotzmann, w. n. (1999). global stock markets in the twentieth century. journal of finance 54(3), 953-980. https://doi.org/10.1111/0022-1082.00133 kaufmann, d., kraay, a., & mastruzzi, m. (2009). governance matters viii: aggregate and individual governance indicators 1996–2008. world bank policy research working paper no. 4978. https://doi.org/10.1596/1813-9450-4978 klapper, l., & love, i. (2004). corporate governance, investor protection, and performance in emerging markets. journal of corporate finance 10(5), 703723. https://doi.org/10.1016/s0929-1199(03)00046-4 klitgaard, r. (1991). gifts and bribes. in r. j. zeckhauser (ed.), strategy and choice, cambridge, ma: the mit press, kobrin, s. (1979). political risk: a review and reconsideration. journal of international business and studies 10(1), 67-80. https://doi.org/10.1057/palgrave.jibs.8490631 krishnamurti, c., sevic, a., & sevic, z. (2005). legal environment, firm-level corporate governance and expropriation of minority shareholders in asia. economic change and restructuring 38(8), 85–111. https://doi.org/10.1007/s10644-005-4524-4 la porta, r., lopez-de-silanes, f., shleifer, a., & vishny, r. w. (1997). legal determinants of external finance. journal of finance 52(3), 1131-1150. https://doi.org/10.1111/j.15406261.1997.tb02727.x la porta, r., lopez-de-silanes, f., shleifer, a., & vishny, r. w. (1998). law and finance. journal of political economy 106(6), 1113-1155. https://doi.org/10.1086/250042 la porta, r., lopez-de-silanes, f., shleifer, a., & vishny, r. w. (1999). the quality of government. journal of law, economics, and organization 15(1), 22279. https://doi.org/10.1093/jleo/15.1.222 la porta, r., lopez‐de‐silanes, f., shleifer, a., & vishny, r. w. (2000). agency problems and dividend policies around the world. the journal of finance 55(1), 133. https://doi.org/10.1111/0022-1082.00199 https://doi.org/10.2307/1913827 https://doi.org/10.1111/0022-1082.00219 https://doi.org/10.1016/j.ecosys.2009.03.001 https://doi.org/10.1016/s0304-4076(03)00092-7 https://doi.org/10.1111/0022-1082.00133 https://doi.org/10.1596/1813-9450-4978 https://doi.org/10.1016/s0929-1199(03)00046-4 https://doi.org/10.1057/palgrave.jibs.8490631 https://doi.org/10.1007/s10644-005-4524-4 https://doi.org/10.1111/j.1540-6261.1997.tb02727.x https://doi.org/10.1111/j.1540-6261.1997.tb02727.x https://doi.org/10.1086/250042 https://doi.org/10.1093/jleo/15.1.222 https://doi.org/10.1111/0022-1082.00199 boadi & amegbe / european journal of government and economics 6(1), 78-101 100 la porta, r., lopez-de-silanes, f., shleifer, a., & vishny, r. w. (2002). investor protection and corporate valuation. journal of finance 57(3), 1147-1170. https://doi.org/10.1111/15406261.00457 lee, c., & ng, d. (2004). corruption and international valuation: does virtue-pay. cornell university working paper. lehkonen, h., & heimonen, k. (2015). democracy, political risks and stock market performance. journal of international money and finance 59(1), 77– 99. https://doi.org/10.1016/j.jimonfin.2015.06.002 li, s., & filer, l. (2007). the effects of the governance environment on the choice of investment mode and the strategic implications. journal of world business 42(1), 80 98. https://doi.org/10.1016/j.jwb.2006.11.006 liu, q., lu, r., & ma, x. (2015). corruption, financial resources and exports. review of international economics 23(5), 1023-1043. https://doi.org/10.1111/roie.12194 lombardo, d., & pagano, m. (2000). legal determinants of the return on equity. csef discussion paper 24, naples. https://doi.org/10.2139/ssrn.209310 low, s.-w., kew, s.-r., & tee, l.-t. (2011). international evidence on the link between quality of governance and stock market performance. global economic review 40(3), 361– 384. https://doi.org/10.1080/1226508x.2011.601646 masood, o., & sergi, b. s. (2008). how political risks and events have influenced pakistan's stock markets from 1947 to present. economic policy in emerging economies 1(4), 427444. https://doi.org/10.1504/ijepee.2008.021285 mauro, p. (1995). corruption and growth. quarterly journal of economics 110(3), 681712. https://doi.org/10.2307/2946696 mckinnon, r., & pill, h. (1997). credible economic liberalizations and over borrowing. american economic review 87(2), 189–93. milyo, j. (2012). stock market reactions to political events: what can we learn about the efficacy of political connection? working paper no. 12-15. washington dc: mercatus center, george mason university. mobarak, a. m. (2005). democracy, volatility, and economic development. review of economic and statistics 87(2), 348–361. https://doi.org/10.1162/0034653053970302 munteanu, a., & brezeanu, p. (2014). government effectiveness and value creation: the case of emerging european listed banks. transylvanian review of administrative sciences 10(42), 140–155. mutenheri, e., & green, c. j. (2003). financial reform and financing decisions of listed firms in zimbabwe. journal of african business 4(2), 155170. https://doi.org/10.1300/j156v04n02_08 ng, d. (2006). the impact of corruption on financial markets. managerial finance 32(10), 822– 836. https://doi.org/10.1108/03074350710688314 ngugi, r. w., murinde, v., & green, c. j. (2003). how have the emerging stock exchanges in africa responded to market reforms? journal of african business 4(2), 89-127. https://doi.org/10.1111/1540-6261.00457 https://doi.org/10.1111/1540-6261.00457 https://doi.org/10.1016/j.jimonfin.2015.06.002 https://doi.org/10.1016/j.jwb.2006.11.006 https://doi.org/10.1111/roie.12194 https://doi.org/10.2139/ssrn.209310 https://doi.org/10.1080/1226508x.2011.601646 https://doi.org/10.1504/ijepee.2008.021285 https://doi.org/10.2307/2946696 https://doi.org/10.1162/0034653053970302 https://doi.org/10.1300/j156v04n02_08 https://doi.org/10.1108/03074350710688314 boadi & amegbe / european journal of government and economics 6(1), 78-101 101 https://doi.org/10.1300/j156v04n02_06 onder, z., & simga-mugan, m. c. (2006). how do political and economic news affect emerging markets? evidence from argentina and turkey. emerging markets finance and trade 42(4), 50-77. https://doi.org/10.2753/ree1540-496x420403 perotti, e.c., & van oijen, p. (2001). privatization, political risk and stock market development in emerging markets. journal of international money and finance 20(1), 43– 69. https://doi.org/10.1016/s0261-5606(00)00032-2 rajan, r. g., & zingales, l. (2003). the great reversals: the politics of financial development in the twentieth century. journal of financial economic 69(1), 5– 50. https://doi.org/10.1016/s0304-405x(03)00125-9 robock, s. (1971). political risk: identification and assessment. columbia journal of world business 6(4), 6-20. shleifer, a., & vishny, r. w. (1993). corruption. quarterly journal of economics 108(3), 599617. https://doi.org/10.2307/2118402 simplice a. (2011). government quality determinants of stock market performance in african countries. agdi working paper no. wp/11/019 tanzi, v., & davoodi, h. (1997). corruption, public investment, and growth. working paper 97/139. washington, dc: international monetary fund. toumi, k. (2011). the role of the investment climate and tax incentives in the foreign direct investment decision: evidence from south africa. journal of african business 12(1), 133 147. https://doi.org/10.1080/15228916.2011.555279 transparency international (1995). building a global coalition against corruption. annual report. berlin: transparency international. wei, s. j. (1997). how taxing is corruption on international investors? working paper 6030.washington, dc: nber. https://doi.org/10.3386/w6030 willard, k., guinnane, t., & rosen, h. (1996). turning points in the civil war: views from the greenback market. american economic review 86(4), 1001–18. https://doi.org/10.2753/ree1540-496x420403 https://doi.org/10.1016/s0261-5606(00)00032-2 https://doi.org/10.1016/s0304-405x(03)00125-9 https://doi.org/10.2307/2118402 https://doi.org/10.1080/15228916.2011.555279 https://doi.org/10.3386/w6030 abstract. the present study investigates the link between quality of governance and stock market performance within the context of international markets. the study employed the fixed effect model using 23 countries with complete relevant data for the ... keywords. stocks market; quality of governance; fixed effect; country-level jel classification. g13; g18; g19. political business cycle and fiscal discipline in sub-saharan africa european journal of government and economics 10(2), december 2021, 185-205 european journal of government and economics issn: 2254-7088 political business cycle and fiscal discipline in sub-saharan africa funsho idowu obakemi a*, hammed adesola adebowalea, babatunde nageri yusufa, and timothy terwase nevb a department of economics, university of ilorin, nigeria b institute of food security, university of agriculture, makurdi, nigeria * corresponding author at: obakemifunsho@gmail.com abstract. we tested the political business cycle theory in sub-sahara africa. to provide an empirical explanation for this nexus, this paper used unbalanced panel data from thirty-six (36) sub-saharan african countries between 1990 and 2018. the system generalized method of moment (gmm) developed by arrelano-bover/blundell-bond was employed to analyze the collected data. the results of the system gmm revealed that the fiscal deficit is significantly large in election years and the deficit spending spills into the year after the election, though not as high as in the election year. we could not, however, find a significant effect in the pre-election year. in addition, we found evidence suggesting that though democracy significantly lowers the fiscal deficit, it promotes higher deficit spending in the election year and the year after the election. hence, the study established the existence of a political business cycle in sub-saharan african countries. the study thus recommends that sound economic policies should be put in place to reduce the persistent deficit in ssa so as to maintain sustainable fiscal health, as well as the sustainability of macroeconomics, particularly enhanced industrialization, as the study found that countries' fiscal deficits are lower in more industrialized countries in the region. keywords. democracy; election; fiscal discipline; politics. jel codes. d72; h30; h60. doi. https://doi.org/10.17979/ejge.2021.10.1.6787 1. introduction the role of the fiscal deficit in developed and developing countries has re-emerged in public debates. on one hand, the fiscal implications in industrialized and developing countries appear to be vastly different. existing literature, on the other hand, shows that fiscal outcomes and traditional macroeconomic factors are closely linked. there are several studies that examine the impact of budget deficits on macroeconomic variables such as inflation rates and unemployment, including anderikopolous (2004), grier (2008), krause (2005) and batool and sieg (2009). according to these studies, while there are several points of divergence on this nexus, the central point of convergence is that the fiscal deficit is an essential expansionary policy used by the government to shape the outcome of macroeconomic objectives (drazen and eslava, 2010). however, the position that the country's fiscal plan takes is a function of the government's macroeconomic plans during and after a specific fiscal year. as a result, there are arguments for and against the effects of fiscal deficits on economies. surplus budgets rarely drive economic growth, according to the development literature. thus, nayab (2015)'s submission demonstrates that relying on yearly fiscal revenue does not support long-term development plans and cannot guarantee desired macroeconomic outcomes. this is premised on the reality that the government, https://doi.org/10.17979/ejge.2021.10.1.6787 funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 186 like other economic agents, does not have all of the resources required to meet unlimited wants. thus, while fiscal deficits are difficult, if not impossible, to prevent in some cases, their persistence has been characterized as fiscal indiscipline (aidt, asatryan, badalyan, & heinemann, 2015). because of its long-term effects on the exchange rate, inflation, and interest rates, as well as an unfavorable current account balance, a persistent budget deficit has proved detrimental to development. it is evidenced by the trend of fiscal deficit that many economies, especially in ssa countries, ignore the developmental and welfare implications of persistent fiscal indiscipline. this informed a concern about whether the influence of political variables on fiscal indiscipline pulled more weight than economic variables. thus, there are occasions when the incumbent government might not be rational in deploying fiscal policy because of selfish goals, especially in the build-up to the forthcoming election. this birthed the political business cycle (pbc), and according to nordhaus (1975), incumbent governments tend to retain office and therefore use every apparatus to achieve this goal. voters, on the other hand, are misled by economic prosperity prior to the election period, without taking into cognizance the tax or inflationary consequences. this is exacerbated when the country's fiscal and political institutions are weak, allowing politicians to pursue their unscrupulous goals at the expense of people's well-being (alesina and paradisi, 2017). thus, the emergence of political-institutions space within the context of macroeconomic variables appears to have altered the narrative of factors that shape macroeconomic variables. the emergence of democratic regimes in most developing countries has piqued the interest of academics and policymakers. against the backdrop of the political-institution-macroeconomic nexus, scholars hypothesized a link between democratic rule and the outcome of the fiscal component. among these scholars are north (1990) and acemoglue (2001). notably, since the return of ssa countries to democratic rule, the region's fiscal deficit has had a significant negative impact on the economies, worsening public debt, crowding out investment, and slowing economic development. and, over time, it's worth considering whether politically timed macroeconomic interventions actually explain these economies' deficit spending. during the years of presidential/parliamentary or state/provincial gubernatorial elections, politicians possess political power and incentives to increase government expenditures (without raising taxes) to improve their chances of re-election. such manipulation has a strong likelihood of leading to political business cycles (pbcs). theories in line with these assertions predicted that budget deficits could disappear during a more prosperous period, but in ssa they persist (obinyeluaku, 2006; bhatia, 2003). the trajectory of the consistent fiscal deficit of ssa countries over the last three decades has worsened the performance of macroeconomic outcomes. for instance, inflation targeting policies, exchange rate and interest rate policies aimed at stabilizing the economy usually fall short of expectations, and in most cases fail. this has raised concern among researchers about whether the consistent fiscal deficit in ssa economies has a link with the transition to democratic rule or if the quality of democracy in this region is below the threshold. for instance, during 2000-2015, the ssa region recorded an annual average of a 9 percent fiscal funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 187 deficit. on the other hand, as revealed by mosley and chripanhura (2016) in several african countries, there are always cases of an incumbent from a coalition of parties or, in some cases, a dominant party, which has an edge over the opposition or rival parties at the poll. this may allay the fear of stiff competition and, consequently, reduce the likelihood of incumbents indulging in electioninduced spending. thus, the political business cycle may be at a minimum or even non-existent. these possible scenarios have created assumptions and uncertainty regarding the bedrock of fiscal indiscipline in africa. and some scholars seem to attribute the fiscal deficit to what was bequeathed to the region by the colonial masters (acamoglue, 2000). most countries in the ssa region inherited budgetary frameworks which empowered spending outside the approved budget. these brought about a rapid expansion of overspending and extra-budgetary spending. the former emanates from an excess of approved contingency votes, while the latter from an excess of off-budget resources (adeyi, 2014). in light of these political and economic concerns, researchers are concerned and suspicious about whether the problem of fiscal indiscipline in sub-saharan africa (ssa) is political or coincidental. this apprehension informed why voters and political actors campaigned for the rotation of political offices such as president, governor, and legislative, among others, to reflect fairness. this is particularly typical in multiethnic countries like nigeria, where ethnic supremacy is a serious concern. as a result, the political structure in place in ssa countries encourages mediocrity rather than merit and sound ideas (obinyeluaku, 2006 and arowolo, 2011). consequently, the region's electoral system has been marred by tumultuous elections, widespread fraud, vote-buying, and corruption (adeyi, 2014; de haan & klomp, 2013; hassan and hassan, 2011). given the pervasiveness of fiscal indiscipline in sub-saharan africa, it is critical to continue to investigate the reasons for the indiscipline and how it might be alleviated in the region. because of its implications for the region's development, this is critical. against this backdrop we investigated whether ssa has a political business cycle. and the purpose of this research article is to add to the existing empirical literature by answering the following research questions: is there a difference between election years and non-election years in terms of fiscal discipline? what impact does democracy have on budgetary discipline in sub-saharan africa? the rest of the paper is arranged as follows. section two discusses the literature review which includes theoretical and empirical literature. section three presents the methodology. the presentation and discussion of results is presented in section four, while section five is the conclusion. 2. literature review there is a plethora of literature on pbc and fiscal deficits. most theoretical studies define fiscal discipline as avoiding or reducing fiscal deficits to the bare minimum. however, following nordhaus (1975), studies discovered a mixed relationship between certain macroeconomic funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 188 variables and the budget deficit (odhiambo, momanyi, othuon, and fredrick, 2013; buscemi and yallwe, 2012). the majority of studies for developing and emerging economies support nordhaus' theory, whereas similar studies for developed economies were found to be insignificant. we divided this part into two sections: theoretical literature and empirical findings. 2.1. theoretical framework we based this study on the opportunistic political business cycle theory as pioneered by nordhaus (1975), and we took cognizance of the modified version by royoff (1990). according to the pioneer version of pbc, re-election-minded incumbent executives knew of the trade-off between unemployment and inflation as encapsulated in the phillips curve, its macroeconomic implications, and the asymmetric knowledge of the voters. the incumbent politicians indulge in fiscal deficits to boost economic prosperity. consequently, it results in the welfare illusion of economic agents, and boosts the chances of the incumbent being re-elected. hibbs (1977) and brender and drazen (2008) refuted opportunistic theory. the former is premised on the heterogeneity of the political parties’ ideologies. the argument of hibbs (1977) was premised on the assumption that there are variations in political parties’ ideology. the latter is premised on the likelihood of voters having learnt from the previous pbc. thus, brender and drazen (2008) assumed a rational expectation hypothesis, which is different from the adaptive hypothesis of nordhaus (1975). the following factors influenced the choice of nordhaus' pbc theory for this study: first, the high level of corruption in ssa, combined with the prevalence of multidimensional poverty, has over time ruled out the possibility of a "learning effect," as proposed by brender and drazen (2008). second, the multiethnic nature of ssa countries, as well as religious bigotry rather than national allegiance, may facilitate cases of pbc in the region. in order to gain large voter support for the next election, incumbents often engage in pre-election spending for the benefit of dominant ethnic groups. in most situations, however, the godfathers are the beneficiaries of election or post-election-induced expenditure, while the voters face the financial burden. the 2015 and 2019 general elections in nigeria, for example, exemplify this point. finally, block (2001) claims that the quantum of pbc decreases over time. however, no research has established a time frame for the diminishing return. ssa countries are, above all, young democracies undergoing both economic and political reforms. 2.2. empirical studies the interaction of macroeconomic and political-institutional variables has been studied in the past. however, researchers and policymakers have recently become more interested in the major impact of both political-institution and macroeconomic variables on fiscal deficits. scholarly disputes have resulted in mixed views and opinions over the years, particularly on the politicalbudget nexus. different opinions formed are centered on the pioneered political business cycle developed by nordhaus (1975). these opinions either refute or validate the pbc theory. thus, funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 189 we reviewed empirical studies on both macroeconomic and political-institutional issues that influence fiscal discipline. many academics have argued that fiscal deficits can be used to accelerate growth by investing in public infrastructure and human capital. some argue that it provides a platform for influencing macroeconomic outcomes, such as economic growth, unemployment, and inflation (odhiambo, momanyi, othuon, and fredrick, 2013; nayab, 2015; cinar, ilhan, and baki, 2014). according to hussain and haque (2017), the outcome of these variables is dependent on the allocation of expenditure to productive projects. on the contrary, numerous studies have found that fiscal deficits reduce economic growth, crowd out investment, and worsen inflationary pressures (arjomanda, emamib and salimic 2016; mohanty, 2012; navaratnam and mayandy, 2016; rana and wahid, 2016; odhiambo et al. 2013; haider, shakil, fatema & rezaul, 2016). however, according to these studies, the impact of fiscal deficits on targeted outcomes is largely a function of the economic and political environment where the policies operate. aside from the fact that politically motivated fiscal spending has both positive and negative consequences, there is the possibility of endogenuity between the fiscal deficit and the relevant macroeconomic variables. among these are unemployment, economic growth, and foreign aid among others. for example, arjomanda, emamib, and salimic (2016) discovered that economic growth has a positive impact on the fiscal deficit. sucharita and sethi's (2012) submission was contradictory. moschovis (2010), on the other hand, attributed the disparities in findings to institution quality. and according to political economists, the impact of fiscal policy and other economic policies on the target or outcome variable is largely determined by the efficiency of the intermediate (political) variable. according to empirical studies, politically motivated fiscal deficit spending revolves around the age of democracy, pre-election, during-election, and post-election year spending, election competition, the party system, and party dominance, corruption, the government system, and institutions, among other things. however, the central question is whether the fiscal deficit in ssa can be traced back to pre-election or during-election spending by the incumbent to increase the likelihood of re-election. the relationship established by nordhaus (1975) between pre-election or during-election spending and the likelihood of incumbents being re-elected has been validated by alesina and paradisi (2017), balaguer-coll, brun-martos, forte, and tortosa-ausina (2015), chortareas, logothetis, and papandreou (2016), lilly (2012), wyplosz and kostrup (2010), drazen and eslava (2010) and maltritz (2015). however, the findings of studies conducted in developed and developing countries differ significantly. a number of empirical investigations revealed that nordhaus' opportunistic model was extremely unusual in developed economies, and that those rare instances gradually faded away over time. for instance, lewid-beck (1990) and drazen (2000) found no convincing evidence of pbc using a panel dataset. in addition, there was no evidence of a significant positive impact of incumbent governments' pre-election or election-year expenditure on economic growth and employment in industrialized economies. in furtherance of that, kneebone and mckenzie (2001) used data from canadian provinces, while alesina et al. (1997) used data from the united states funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 190 from 1947 to 1994. the former found evidence of opportunistic political cycles, while the latter supported partisan theories of the political system as a whole. according to alesina et al. (1997), the oecd and virtually all developed democracies, particularly those with a two-party system, tend to align with partisan models. however, persson and tabellini (2003) claimed that the quantum of pbc is a decreasing function of institutional quality, which is, of course, a concern in developing democracies, and ssa in particular. khmedov and zhuravskaya (2004) and aidt et al. (2011), for example, discovered a high degree of time elasticity of magnitude of pbc in russia and portugal, respectively. the rate and magnitude of decline with time, however, have not been addressed in the literature. such a decreasing function of time was attributed to brender and drazen's "learning effect" (2008). on the contrary, cases and quantum of the fiscal deficit and pbc are common in developing economies. but were still mixed findings in developing countries and even in the ssa region. shi and svenssion (2006) studied both developed and developing economies, and their findings were consistent with rogoff's (1988) version of the opportunistic model. however, it was higher in developing countries, but african countries were underrepresented in the study. other studies in developing countries that supported nordhaus' version of the opportunistic model are barberia and avelino (2011), ebeke and lçer (2013), and koksal, ahmet, and kan (2012). barberia and avelino (2011) sampled latin american democracies, while ebeke and ölçer (2013) and koksal, ahmet and kan (2012) sampled developing countries. also, jones et al. (2012) collected 136 cases of elections for three argentine cities (buenos aires, corrientes and tierra del fuego), whereas gonzalez (2002) sampled mexican data. their findings suggested the occurrence of pbc in these countries. and on a greater dimension across this research, pre-election year spending showed a favorable impact on budget deficits. also, klomp and de haan (2012) discovered evidence of pbc in 65 democratic countries, but in a modest dimension. klomp and de haan (2012b) further divided democratic countries into old and young democracies, with evidence of pbc increasing in young democracies but decreasing in more mature democracies. both studies supported the findings of akhmedov and zhuravskaya (2004) and aidt et al (2011). following the return of most african countries to democratic rule, several countries experienced various levels of political instability before embarking on political and economic reforms. as a result, the pace of democratic maturation has slowed and retarded in some countries, particularly in sub-saharan african countries. however, there aren't many studies on this topic in african political and economic literature. and, for the most part, the few studies on africa, particularly in sub-saharan africa, supported the opportunistic model. however, the political and economic experiences of these countries following these studies may have changed the narrative. the pbc model was tested in 51 african countries by mosley and chiripanhura (2016), who used fixed effect estimation (fee). they found existence of variation in the quantum of pbc among african countries, and was lower in countries with dominant-party systems, where the incumbent's re-election is less threatened by the opposition. while analyzing 36 nations from 1980 to 2010 using arellano-bond dynamic panel data, neumann and ssozi (2016) explored the relevance of political business cycles, as well as the magnitude of fiscal deficits' effects on money funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 191 growth and inflation in sub-saharan africa. however, there was evidence that the fiscal deficit positively and significantly affected the inflationary level. but there was no substantial evidence to support the existence of pbc in the region. block (2002) and block, ferree, and singh (2003), for example, used the gmm estimation technique to test for the presence of pbcs in 44 sub-saharan african countries from 1980 to 1995. the former focused on both fiscal and monetary pro-cycles, whereas the latter focused on fiscal pro-cycles. the findings revealed evidence of pbc in ssa, and even higher in the region's young emerging democracies. the inclusion of multiparty competition by block, ferree, and singh (2003) made it unique, and the findings were strongly aligned with pbc theory and the foundingelections hypothesis. similarly, ifere and okoi (2017) used descriptive statistics and the ordinary least square regression technique to investigate the political economy of nigeria's fiscal deficit. the findings revealed that fiscal policy in nigeria is influenced more by political factors than by economic factors. the study, however, did not test for pbc. furthermore, brender and drazen (2005) demonstrated a distinction between authoritarian and democratic fiscal discipline trends in developing countries in their argument about the transition to democracy. brender et al. (2005) discovered a link between democracy and fiscal discipline, while the years of democracy were a decreasing function of the pbc. higashijima (2016) and bohn (2018) found high cases of pbc in new and young democracies. in addition, higashijima (2016) found a non-linear result, and was higher in authoritarian governments. bohn (2018) convincingly attributed pbc cases to in-transparencies. de haan (2012) and klomp and de haan (2012b) found an increase in the fiscal deficit and rising cases of pbc in young democracies, but less in old democracies and none in matured democracies. studies by udoh, joshua, and etok (2012) and fabrizio and mody (2006), showed that nigeria recorded higher fiscal deficits during the democratic dispensation than during the military regime. these studies backed up alesina and tabellini's claims (1990). also, yaru, mobalaji, kilishi, and yakubu (2014) supported udoh et al. (2012) and fabrizio et al. (2006); the study found that the fiscal deficit defiled budgetary institutions in nigeria. but that does not negate the efficacy of budgetary institutions. rising cases of election-induced fiscal deficits in developing economies may be difficult to decouple from the prevalence of weak fiscal rules and electoral manipulation, combined with institutional weakness, which gives incumbent governments more power to veto policies that are not welfare-driven. these political deficiencies are practically common in developing countries (acemoglue, 2001). as a result, it is important to consider the impact of these inefficient variables on the fiscal deficit and pbc cases. more importantly, brender et al. (2005), de haan (2012) and klomp and de haan (2012b) suggested that fiscal deficits and pbc are common in young democracies. block et al. (2003) and higashijima (2016) found fewer or no cases of fiscal deficits in democracies with highly competitive elections. according to mosely and chiripanhura (2016), less competitive elections in a one-party democracy reduce the likelihood of cases of election-induced fiscal spending. according to mosely and chiripanhura (2016), in a one-party democracy, less funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 192 competitive elections reduce the likelihood of election-induced fiscal spending. mosely et al. (2016), on the other hand, found african pbc heterogeneous. thus, it is myopic to generalize previous studies of pbc in africa. north (2005) and acemoglue (acemoglu, robinson, james, and thaicharoen, 2003) demonstrated that institutions are inextricably linked to macroeconomic variables. furthermore, empirical studies have shown that weak fiscal institutions account for a large portion of the indiscriminate fiscal deficit in developing countries. for example, fatas and milhov (2006) and onye and okon (2017) investigated the impact of strict fiscal constraints on fiscal policy. the findings revealed that adhering to a deficit benchmark reduces the volatility of the fiscal deficit. however, it could be very damaging to the economy if the situation necessitates a spontaneous response of fiscal policy adjustment to deal with economic shocks. similarly, manasse (2006) discovered that fiscal rules reduce pbc cases, but not as effectively as institutional quality. similarly, we were concerned about whether the types of government could affect the cases and quantum of the opportunistic model, particularly in the ssa region. according to mandon and cazals (2019), the increase in fiscal deficit caused by pre-election spending is less in a presidential system of government than in a parliamentary system. although there was evidence of fairly pbc in the former, a shift in expenditure composition in favor of recurrent expenditure was recorded, as opposed to the expansionary aggregate spending found in other studies. this suggests that the incumbent's re-election campaign strategy could include a trade-off between recurring and capital projects. on the expenditure trade-off between recurrent and capital spending, mandon and cazals (2019) agreed with enkelman et al. (2013) and drazen and eslava (2010). furthermore, drazen and eslava (2005), potrafke (2010), klomp and de haan (2012b), and chortareas et al. (2016) found evidence of a shift in expenditure towards road construction, health, agriculture, and street works. this suggests that the incumbent may decide not to engage in election-induced fiscal deficits, but rather use an expenditure-switching strategy to increase their chances of retaining political office. we paid equal attention to the party system, being a pillar of democracy. chortareas et al. (2016) and jochimsen & nuscheler (2011) discovered evidence in favor of pbc in a multi-party system. it was stronger in the former, and aligned with the common pool hypothesis. it was found weak in the latter, resulting in massive debt, as is commonly seen in the case of a single-party system. klompa and de haan (2013), on the other hand, supported the party system but ignored the government system. according to klompa et al. (2013), the existence and the quantum of a pbc is determined by the degree of transparency, development, and democracy in the government, the country's political system, and the degree of political polarization. this appears to be related to the maturity of democracy advocated by brender and drazen (2005). the reviewed literature revealed extant of studies on fiscal discipline and the existence of opportunistic model, particularly in developing countries. although, there are relatively few studies (neumann and ssozi, 2016; block, 2002; and block, ferree, and singh, 2003) that actually focused on testing for pbc in africa, and in particular, ssa, there are evidences of pbc aggravated by ethnicity and vote buying (collier and vicente, 2012; van de wallance, 2003; funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 193 jensen and justesen, 2014; vincente, 2014). however, the period covered by these studies raised concerned whether the generalizations deduced are still relevant to current reality in ssa. for instance, neumann and ssozi (2016) studied the years 1980-2010, while block (2002) and block et al (2003) both studied the years 1980-1995. given the time frame and the fact that the ssa region is undergoing both political and economic reform, it may be naive to assume that the findings of these studies still represent the current reality in the region. we equally gave a consideration to examining the quality of democracy as suggested by akhmedov and zhuravskaya (2004) and aidt et al (2011). their findings in russian and portugal respectively, shown pbc to be a decreasing function of quality of democracy which is also a function of time. furthermore, during the time period covered by neumann and ssozi (2016), block (2002) and block et al (2003) among others, elections in africa, and particularly in ssa, were less competitive. as a result, there is a possibility that the narrative of pbc in the region might have changed. and perhaps, brender and drazen’s (2005) “learning effect” might be gaining momentum in the region. thus, this study tends to fill this gaps. 3. methodology model specification this study is rooted on political business cycle theory pioneered by nordhaus (1975). therefore, following udoh et al. (2012); and neumann and ssozi, (2016) the model specification is as follows: 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖 = 𝜏𝜏 + 𝛽𝛽𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖−1 + 𝛼𝛼𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛾𝛾𝑂𝑂𝐹𝐹𝑂𝑂𝑖𝑖𝑖𝑖 + 𝛿𝛿𝑈𝑈𝑈𝑈𝐺𝐺𝑈𝑈𝑈𝑈𝑖𝑖𝑖𝑖 + 𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝑖𝑖𝑖𝑖 + 𝜑𝜑𝐹𝐹𝜃𝜃𝑈𝑈𝑖𝑖𝑖𝑖 + 𝜔𝜔𝑂𝑂𝐺𝐺𝜔𝜔𝑖𝑖𝑖𝑖 + 𝜇𝜇𝑖𝑖𝑖𝑖 [1] where 𝜇𝜇𝑖𝑖𝑖𝑖 = 𝜌𝜌𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖 [2] if (2) is true, equation (1) becomes 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖 = 𝜏𝜏 + 𝛽𝛽𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖−1 + 𝛼𝛼𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛾𝛾𝑂𝑂𝐹𝐹𝑂𝑂𝑖𝑖𝑖𝑖 + 𝛿𝛿𝑈𝑈𝑈𝑈𝐺𝐺𝑈𝑈𝑈𝑈𝑖𝑖𝑖𝑖 + 𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝑖𝑖𝑖𝑖 + + 𝜑𝜑𝐹𝐹𝜃𝜃𝑈𝑈𝑖𝑖𝑖𝑖 + 𝜔𝜔𝑂𝑂𝐺𝐺𝜔𝜔𝑖𝑖𝑖𝑖 + 𝜌𝜌𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖 [3] the specification above is the baseline model of the study, which is a dynamic panel model. where 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖 is fiscal deficit, 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖−1 is a year lagged of fiscal deficit, 𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 is the real gdp growth, 𝑂𝑂𝐹𝐹𝑂𝑂𝑖𝑖𝑖𝑖 is the official development assistant received, 𝑈𝑈𝑈𝑈𝐺𝐺𝑖𝑖𝑖𝑖 is the unemployment rate, 𝑂𝑂𝐺𝐺𝜔𝜔𝑖𝑖𝑖𝑖 is trade openness, 𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝑖𝑖𝑖𝑖 is the election year and dem is the level of democracy all in country i at time t. while 𝜌𝜌𝑖𝑖represents individual country unobservable time invariant countryspecific characteristics and 𝜀𝜀𝑖𝑖𝑖𝑖is white noise error term which is independent and identically distributed. for further analysis, we checked the behavior of fiscal deficit in the pre-election year, election year and post-election year within a single model given as funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 194 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖 = 𝜏𝜏 + 𝛽𝛽𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖𝑖𝑖−1 + 𝛼𝛼𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛾𝛾𝑂𝑂𝐹𝐹𝑂𝑂𝑖𝑖𝑖𝑖 + 𝛿𝛿𝑈𝑈𝑈𝑈𝐺𝐺𝑈𝑈𝑈𝑈𝑖𝑖𝑖𝑖 + 𝜋𝜋𝐺𝐺𝜋𝜋𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃 + 𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝑖𝑖𝑖𝑖 + 𝜎𝜎𝐺𝐺𝜎𝜎𝜃𝜃𝜃𝜃𝜃𝜃𝜃𝜃 + 𝜑𝜑𝐹𝐹𝜃𝜃𝑈𝑈𝑖𝑖𝑖𝑖 + 𝜔𝜔𝑂𝑂𝐺𝐺𝜔𝜔𝑖𝑖𝑖𝑖 + 𝜌𝜌𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖 [4] a common problem that characterized a dynamic specification is the possibility of an endogeniety problem, since the lag of the dependent variable will be correlated with the disturbance term (𝜇𝜇𝑖𝑖𝑖𝑖). the second problem associated with dynamic model specification is the presence of country fixed effects which may be correlated with the explanatory variables. we treated these problems using the instrumental variable approach. following arrelano and bond (1991) and arrelano and bover/ blundell and bond (1998), we employed the generalized method of moment (gmm) technique under which the endogenous variable is instrumented by the further lag of the variables. generally, the strength of gmm lies in the robustness estimator it offers because it does not require information about the exact distribution of the disturbances. it selects parameter estimates such that the sample correlations between the instruments and parameters are as close as possible to zero. second, the technique requires small period (t) and large–n panels, and it is more efficient than the difference gmm (baltagi, 2008 and roodman, 2009). data and variable measurement the data used in the study is annual data generated from various secondary sources. the data on fiscal deficit (scaled by gdp) and unemployment as a percentage of the labour force were sourced from the world economic outlook on the imf database for various years. real gdp growth (measured by gdp per capita growth), aid received (measured by net official development assistance and official aid received scaled by gdp) and trade openness (measured by trade as a percentage of gdp) were sourced from the world development indicators data base for various years. election data was sourced from the african election database for thirty-six (36) ssa countries. regarding the measurement of election data, we used a binary measure of 1 for election years and 0 otherwise. also, the pre-election year assumed 1 and 0 otherwise, and ditto for the post-election year. democracy is measured using polity 2 from the polity iv data base, where the democracy level in a country is measured on a scale of -10 to 10, with a higher score indicating a higher level of democracy. the data scope ranges from 1990 to 2018 based on the availability of data. the selected ssa countries include angola, benin, burkina faso, botswana, cote’divore, cameroon, congo brazzaville, ethiopia, cape verde, eritrea, gabon, ghana, gambia, guinea, guinea-bissau, equatorial guinea, kenya, madagascar, mali, mozambique, mauritania, mauritius, malawi, namibia, niger, nigeria, rwanda, senegal, sierra leone, swaziland, chad, togo, tanzania, uganda, south africa and zimbabwe. funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 195 4. results and discussions 4.1. descriptive analysis the results of descriptive analysis of the variables used in the study are presented in table 4.1. from the results, the estimated average fiscal balance of the sampled countries over the sampled period is -4.398 with a minimum and maximum of -557.499% and 31.045% respectively. the estimated average gdp growth over the period of analysis is 4.764 with a minimum of -50.248 and a maximum of 149.973. the estimated average trade openness as a percentage of gdp is 74.763%, while its minimum is 20.723% and maximum is 531.737%. in addition, the estimated average unemployment rate is 35.832 with a minimum and maximum of 10.24 and 63.54 respectively. the average foreign aid is 64.588% with a minimum and maximum of -11.967 and 691.925 percent respectively. for the democratic level in the ssa, the average score recorded by the sampled countries is 1.22, with a minimum and maximum of -9 and 10 respectively, implying that ssa countries' democratic practice is at a low level. table 1. descriptive statistics. variable obs mean std.dev. min max fsd 944 -4.398 27.533 -557.499 31.045 gdpg 944 4.764 7.85 -50.248 149.973 opn 912 74.763 47.668 20.723 531.737 ump 915 35.832 13.115 10.24 63.54 oda 944 64.588 72.992 -11.967 691.925 prelec 944 .183 .387 0 1 elec 944 .176 .381 0 1 pselec 944 .177 .382 0 1 dem 915 1.22 5.459 -9 10 source. author’s computation, 2021. table 2. estimated correlation coefficients among variables. variables (1) (2) (3) (4) (5) (6) (7) (8) (9) (1) fsd 1.000 (2) gdpg -0.078 1.000 (3) opn -0.192 0.396 1.000 (4) ump1 -0.026 -0.018 0.359 1.000 (5) oda -0.085 0.042 0.105 0.151 1.000 (6) prelec 0.025 0.002 -0.046 -0.018 0.020 1.000 (7) elec 0.028 -0.014 -0.027 -0.027 -0.028 -0.219 1.000 (8) pselec 0.041 0.026 -0.022 -0.010 -0.003 -0.204 -0.214 1.000 (9) polity2 0.089 -0.021 -0.163 0.077 0.220 0.054 0.115 0.117 1.000 source. author’s computation, 2021. funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 196 the results from correlation analysis of the variables are presented in table 4.2. from the results, a weak negative relationship exists between fiscal balance and gdp growth given the estimated correlation coefficient of -0.078. a similar weak negative relationship was also found between trade openness, unemployment, foreign aid, and the fiscal deficit in ssa. on the contrary, the estimated correlation coefficient suggests that the pre-election year has a weak positive relationship with the fiscal balance. similarly, both election years and post-election years have a weak positive relationship with fiscal balance. in addition, we found a weak positive relationship between democracy and fiscal balance. 4.2. panel regression analysis the estimated results of the baseline model of the study using the arrelano-bover/blundell-bond system gmm estimation technique are presented in table 2. the results presented in the first column of the study are obtained without controlling for the democracy level in the sampled countries, while the results in column 2 are estimated by controlling for the democracy level in the countries that made up the sample of the study. from the results in column 1, the estimated coefficient of -8.025 for the election with its corresponding p value of 0.079 implies that the election year is negatively and significantly associated with lower primary fiscal balance or higher fiscal deficit. in line with the results, we find evidence suggesting that the fiscal deficit is higher in election years by 8.025 percent than in years when elections did not take place. this aligns with the theoretical proposition that governments in power who face election uncertainty may engage in more deficit financing in order to lure voters to their side through financial enticement such as vote buying. the results may also be linked to the huge electoral body budget, especially where institutions in the country give room for contract inflation. in the results presented in column 2, we control for democracy and the results reveal that democracy improves fiscal balance given its estimated coefficient and p value of 9.027 and 0.004 respectively. this may be linked to a number of reasons, including that good democratic institutions can check the excesses of the incumbent and allow voters to punish the incumbent when they suspect frivolous spending, thereby restricting the spending behavior of the government. on the impact of the election on fiscal balance after controlling for democracy, the study still found a negative coefficient with improved significance on fiscal balance. the estimated coefficient of -13.080 indicates that the impact of elections on spending becomes more pronounced after controlling for democracy. this result may be due to the fact that elections tend to be more competitive under democracy, and thus, the incumbent may spend a huge part of the country’s budget on frivolous projects whose proceeds are used to engage in vote buying to ensure victory in the election. this is possible because of the prevalence of poverty in most developing countries. nigeria is a case study. improvements in nigeria's elections conducted through improved independent electoral bodies, adoption of technology such as card readers, and pressure from international funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 197 organizations for free and fair elections have drastically reduced the incidence of physical election rigging in most parts of the country, especially in the south. this has, however, emboldened the culture of vote buying and several cases abound. one such is "dibo ko sebe in local parlance" (vote and cook a pot of soup) under which individual voters are offered a sum of 5000 nigerian naira to 10000 naira (10 to 20 us dollars) to vote for a certain candidate. this aligns with the findings of jensen and justesen (2014) who reported in a survey of 18 ssa democracies that the incidence of vote buying is higher in competitive elections in ssa countries. since competitive elections are a key attribute of good democracy, their results imply that vote buying is more prevalent in ssa countries with good democratic institutions. thus, our findings indicate that good democracy can limit deficit budget financing because of good monitoring mechanisms by the voters, but it increases spending during election years. table 3. estimated system gmm dynamic panel regression results with election year. (1) (2) variables 1 1 l.fsb -0.204*** -0.128** (0.001) (0.013) gdpg 0.073 -0.142 (0.632) (0.301) opn 0.762*** 0.858*** (0.000) (0.000) ump 1.580 0.972 (0.380) (0.397) oda -2.804*** -2.099*** (0.000) (0.001) elec -8.025* -13.080*** (0.079) (0.002) dem 9.027*** (0.004) constant 63.207 28.250 (0.285) (0.639) observations 848 848 number of pid 36 36 chi2p 0.000 0.000 instrument 10 11 ar1 p value 0.0685 0.0790 ar2 pvalue 0.979 0.800 hansen p value 0.318 0.444 note. pval in parentheses *** p<0.01, ** p<0.05, * p<0.1.source. author's computation 2021. funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 198 our findings on the significant negative influence of the election year align with many other studies on the political budget cycle. among the studies are those of barberia and avelino (2011), and koksal et al. (2012) in developing countries, klomp and de haan (2021) in sample selected countries, and jones et al. (2012) in argentina. the coefficient of the lagged values of the fiscal deficit is negative and significant in models 1 and 2, which implies that the dynamic specification in this study is justified. in addition, each of the values that are less than 1 implies stability in the dynamic system. the outcome is consistent with the findings of tujula & wolswijk (2007) and maltritz (2015) and may be explained in part by the fact that the current budget reflects correct past budgetary imbalances due to debt service obligations. the estimated coefficient of real gdp growth (an indicator of the overall economic situation) of 0.762 with a p value of 0.632 indicates that gdp growth has an insignificant impact on the fiscal balance in sub-saharan african countries. similar results have been documented in some other studies, including those of tujula and wolswijk (2007) and matritz (2015). even though we expect growth to increase the fiscal balance, the over dependency of sub-saharan african countries on loans may be a candidate for the cause. this assertion is justified by the significant negative impact of aid with a coefficient and p value of -2.009 and p value of 0.001, which indicates that countries that rely more on aid would record more deficit spending and thus a lower fiscal balance. our finding on official aid received contradicts that of neumann & ssozi (2016) who found a positive but insignificant relationship between fiscal balance and aid received. the findings here could be due to the fact that countries that receive aid, especially when it is not captured in the budget, engage in unguarded spending, which will eventually promote fiscal indiscipline in terms of higher fiscal deficits. in addition, the study found a very significant impact of trade openness on fiscal balance with its estimated coefficient and p value of 0.858 and 0.000 respectively, implying that liberalization policy is a key policy instrument that can be deployed by sub-saharan african countries to drive fiscal balance and reduce fiscal indiscipline in the region. this result agrees with udoh, et al. (2011). as for unemployment, we could not establish a significant impact of unemployment on the fiscal balance. this is contrary to maltritz (2015) who reported a negative but significant impact of the unemployment rate on the fiscal balance. post estimation diagnostic test results are presented in the lower part of table 3. from the table, the number of instruments used is 10 and 11 respectively for the results in column 1 and 2. since the number of instruments in each case is below the number of groups, which is 36, one of the requirements for the validity of instruments in dynamic panels is satisfied. we also checked for the validity of the instrument using the hansen test. the results of the test suggest that the instruments used in both columns 1 and 2 are valid since the null hypothesis of hansen in both cases cannot be rejected given their respective p values of 0.318 and 0.444. in addition, the ar (1) p values of 0.0685 and 0.0790 for the results obtained in columns 1 and 2 confirm the presence of first-order autocorrelation in the results in both models. this is expected, since the lag of the dependent variable is part of the explanatory variables. the ar (2) test p value of 0.979 and 0.800 funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 199 in each of the results in columns 1 and 2 fails to reject the null hypothesis of no second order auto correlation. we conducted further analysis of the results by considering pre and post-election years, the results of which are presented in table 4.4. the results in column 1 of table 4.4 are obtained for only the pre-election year and the results reveal that the fiscal balance is higher in the year before the election with an estimated coefficient and p value of 12.11 and 0.039 respectively. this may be explained by the deceptive antics of the incumbent, who often tries to impress voters in the year before the election so as to gain popularity and support during the election by running a surplus budget. this is also helped by good democratic institutions which limit the ability of the incumbent to spend frivolous in the year before the election because they may be punished in the election should the vote buying option fail. similar results have been documented elsewhere (mosley & chiripanhura, 2016). table 4. estimated system gmm dynamic panel regression results with pre and post-election years. (1) (2) (3) variables 1 2 3 l.fsdg -0.106 -0.114 -0.131 (0.191) (0.148) (0.138) gdpg -0.126 -0.072 -0.100 (0.787) (0.875) (0.828) opn 0.838*** 0.831*** 0.855*** (0.000) (0.000) (0.000) ump1 1.007 0.937 0.950 (0.442) (0.464) (0.472) oda -2.051** -2.063** -2.123** (0.018) (0.023) (0.025) prelec 12.110** 4.372 (0.039) (0.507) elec -17.965** (0.041) pselec -7.082 -15.048** (0.101) (0.027) polity2 8.904** 8.486** 9.138** (0.031) (0.040) (0.035) constant 21.251 27.512 32.558 (0.751) (0.682) (0.645) observations 848 848 848 number of pid 36 36 36 chi2p 0 0 0 j 11 11 13 ar1p 0.109 0.112 0.118 ar2p 0.890 0.845 0.923 hansenp 0.442 0.486 0.474 note. pval in parentheses *** p<0.01, ** p<0.05, * p<0.1. source. author's computation 2021. funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 200 in the second column of table 4.4, we found a negative impact of the post-election year, albeit not significant given the estimated value of -7.082 and p value of 0.101. in the results presented in column 3 of the table, it is, however, found that the post-election year with an estimated coefficient and p value of -15.048 and 0.027 respectively, records a significant negative impact on the fiscal balance, while the estimated coefficient and p value of -17.965 and 0.041 indicate that the fiscal balance is significantly lower in the election year. on the contrary, the estimated coefficient of 4.372 with a p value of 0.507 implies that the fiscal balance increased in the preelection though the impact is now insignificant. hence, the results of the full model show that the fiscal deficit is highest during an election year. it is also higher in the year following the election, but not as much as in the election year. during election years, countries' spending is highest because of the expenditure on the electoral process and the spending on white elephant projects with the intent of amassing proceeds which can then be used to buy votes during elections. also, powerful "cabals" mostly contribute to the financing of candidates’ elections in developing countries, including ssa. thus, once the election is gone, the elected administration would strive to reward their sponsors (cabals) by awarding contracts to them at a cost beyond the actual cost, the proceeds of which are the reward for sponsoring their election. thus, the possibility of deficit spending a year immediately after the election is high and thus justifies our results. the results obtained in table 4.4 satisfy various post estimation diagnostic tests. in each of the results in columns 1 to 4, the null hypothesis of the hansen test of over identifying restrictions could not be rejected. hence, the instrument used for the analysis is valid. in addition, all the results of the ar (1) and ar (2) tests obtained suggest that there is no evidence of either first order or second order serial correlation. 4.3. policy implications the results of our study suggest that fiscal indiscipline is highest in the election year, while this deficit spending persists till a year after the election, though the magnitude of the spending is lower compared to the election year. we further found that this fiscal indiscipline in election years is more pronounced when the democracy level is higher, which depicts the do or die attitude of politicians through vote buying, in line with the hypothesis that vote buying is prevalent when elections are competitive. the implication of this result is that indiscipline in the fiscal process is affected by politics in the region. this perfectly aligns with the theory of pbc, as politicians increase spending in election years as a means to boost their chances of being re-elected. however, due to the nature of poverty and poor press freedom in sub-saharan africa, they are not likely to be punished for deficit spending by the voters. rational expectations have failed in the region because of two factors. one, ethnicity allegiance is highly prevalent in the region. voters prefer voting for corrupt politicians to promote ethnic dominance at the detriment of social welfare. second, religion affiliation is an equally strong weapon that opportunistic leaders employ to enforce being reelected. the study also found that other control variables impact significantly funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 201 on the fiscal deficit, especially the level of industrialization, whose impact on the fiscal balance is positive and significant through the models. the implication of this is that, while trying to address the problem of fiscal indiscipline in the region, in addition to political variables proxy here by preelection year, election year, and year after election, other variables such as democracy, gdp growth, and industrialization must be taken into cognizance. 5. conclusions the nexus between political business cycle and fiscal discipline proxy by election year and fiscal deficit respectively, are examined using a panel of thirty-six (36) ssa countries over the period of 1990-2018. the generalized method of moments (gmm) estimation technique was used to estimate a dynamic panel model with lagged fiscal discipline. the result shows that government spending in the election year and the year after the election has a significant effect on the fiscal deficit, which implies that ssa countries expend more resources in the election year and the year immediately after the election. based on the findings, constitutional reform is urgent in the ssa region to curtail the frivolous spending of politicians, and necessary sanctions should be enforced to reduce persistent deficits. in addition, the incident of vote-buying, which is always prevalent in developing countries with strong democratic practices, should be addressed head on as part of strategies aimed at controlling political business cycles in ssa countries. references acemoglu, d., johnson, j., robinson, a., and thaicharoen, y. (2003). institutional causes, macroeconomics symptoms: volatility, crises and growth. journal of monetary economics 50, 49-123. https://doi.org/10.1016/s0304-3932(02)00208-8 acemoglu, d., johnson, s and robinson, j. a. (2001). the colonial origins of comparative development: an empirical investigation. american economic review 91, 1369-1401. https://doi.org/10.1257/aer.91.5.1369 adeyi, v. a. (2014) money in political parties and democracy in nigeria. kuru. national institute of policy and strategic studies 2(5), 289-313. african elections data base (2016). elections in nigeria. retrieved from http://www.africanelections.org aidt, t., f., & veiga, l. (2011). ‘election results and opportunistic policies: a new test of the rational political business cycle model’. public choice 148, 21–44. https://doi.org/10.1007/s11127-010-9644-3 aidt t., asatryan, z., badalyan l., & heinemann, f. (2015). vote buying or political business cycles as usual? zew discussion paper no.15-017. https://doi.org/10.2139/ssrn.2580781 akhmedov, a. & zhuravskaya, e. (2004). ‘opportunistic political cycles: test in a young democracy setting. quarterly journal of economics 119, 1301–1338. https://doi.org/10.1162/0033553042476206 alesina, a., & tabellini. g. (1990). “voting on the budget deficit.” american economic review 80(1), 37-49 alesina, a., & paradisi, m. (2017). political budget cycles: evidence from italian cities. economics & politics 29(2), 157-177. https://doi.org/10.1111/ecpo.12091 https://doi.org/10.1016/s0304-3932(02)00208-8 https://doi.org/10.1257/aer.91.5.1369 http://www.africanelections.org/ https://doi.org/10.1007/s11127-010-9644-3 https://doi.org/10.2139/ssrn.2580781 https://doi.org/10.1162/0033553042476206 https://doi.org/10.1111/ecpo.12091 funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 202 alesina, a., & perotti, d. (1999). budget deficits and budget institutions, nber chapters, fiscal institutions and fiscal performance 2, 13–36. alesina, a., & tabellini. 1990. “voting on the budget deficit,” american economic review 80(1), 37-49. arjomanda, m., emamib, k., & salimic, f. (2016). growth and productivity: the role of budget deficit in the mena selected countries. procedia economics and finance 36, 345-352. https://doi.org/10.1016/s2212-5671(16)30046-6 arowolo, d. (2011) fiscal federalism in nigeria: theory and dimensions. afro asian journal of social sciences 2(22), 1-22. arrelano, m., & bond, s. (1991). “some tests of specification for panel data: monte carlo evidence and an application to employment equations. review of economic studies, 58, 277-279. https://doi.org/10.2307/2297968 blundell, r. & bond, s. (1998). initial conditions and moment restrictions in dynamic panel data models. journal of econometrics 87, 115-143. https://doi.org/10.1016/s0304-4076(98)000098 balaguer-coll, m. t., brun-martos, m. i., forte, a., & tortosa-ausina, e. (2015). local governments' re-election and its determinants: new evidence based on a bayesian approach. european journal of political economy 39, 94-108. https://doi.org/10.1016 /j.ejpoleco.2015.04.004 baltagi, b.h. (2008). econometric analysis of panel data. 6th edition, wiley, chichester. barberia, l. g. & avelino, g. (2011). ‘do political budget cycles differ in latin american democracies? economia 11, 101–134. https://doi.org/10.1353/eco.2011.0001 batool, i., & sieg, g. (2009). pakistan, politics and political business cycles. working paper series, no. 7, technische universität braunschweig, institut für volkswirtschaftslehre, braunschweig bhatia, h.l. (2003). public finance: vikas publishing house pvt limited. block, s. a. (2002). political business cycles, democratization and economic reform: the case of africa. journal of development economics 67(2), 5–28. https://doi.org/10.1016/s03043878(01)00184-5 block, s. a., ferree, k., & singh, s. (2003). multiparty competition, founding elections and political business cycles in africa. journal of african economies 12(3), 444–68. https://doi.org/10.1093/jae/12.3.444 blundell, r., & bond, s. (1998). initial conditions and moment restrictions in dynamic panel data models. journal of econometrics 87, 115-143. https://doi.org/10.1016/s0304-4076(98)000098 bohn, f. (2018). political budget cycles, incumbency advantage, and propaganda. economics & politics. 2, 1-28. https://doi.org/10.1111/ecpo.12122 brender, a., & drazen, a. (2008). “how do budget deficits and economic growth affect re-election prospects?” evidence from a large cross-section of countries. american economic review 98, 2203–2220. https://doi.org/10.1257/aer.98.5.2203 buscemi, a., & yallwe, a. h. (2012). fiscal deficit, national saving and sustainability of economic growth in emerging economies: a dynamic gmm panel data approach. international journal of economics and financial issues, 2(2), 126-140. cinar, m., ̇ ilhan, e.g., & baki, d. (2014). examining the role of budget deficit policies in economic growth from a keynesian perspective. international journal of economics and finance 6(3), 23-41. https://doi.org/10.5539/ijef.v6n10p191 chortareas, g., logothetis, v., & papandreou, a. a. (2016). political budget cycles and re-election prospects in greece's municipalities. european journal of political economy 43, 1-13. https://doi.org/10.1016/j.ejpoleco.2016.02.001 collier, p., & vicente, p. c. (2010). violence, bribery, and fraud: the political economy of elections in sub-saharan africa. oxford university working paper. de haan, j., & klomp, j. (2013). conditional political budget cycles: a review of recent evidence. public choice 157, 387–410. https://doi.org/10.1007/s11127-013-0106-6 drazen, a., & eslava m. (2010). electoral manipulation via voter friendly spending: theory and evidence. journal of development economics 92(1), 39-52. https://doi.org/10.1016/ j.jdeveco.2009.01.001 https://doi.org/10.1016/s2212-5671(16)30046-6 https://doi.org/10.2307/2297968 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1016%0b/j.ejpoleco.2015.04.004 https://doi.org/10.1016%0b/j.ejpoleco.2015.04.004 https://doi.org/10.1353/eco.2011.0001 https://doi.org/10.1016/s0304-3878(01)00184-5 https://doi.org/10.1016/s0304-3878(01)00184-5 https://doi.org/10.1093/jae/12.3.444 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1111/ecpo.12122 https://doi.org/10.1257/aer.98.5.2203 https://doi.org/10.5539/ijef.v6n10p191 https://doi.org/10.1016/j.ejpoleco.2016.02.001 https://doi.org/10.1007/s11127-013-0106-6 https://doi.org/10.1016/j.jdeveco.2009.01.001 https://doi.org/10.1016/j.jdeveco.2009.01.001 funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 203 drazen, a., & eslava, m. (2005). electoral manipulation via expenditure composition: theory and evidence, working paper 11085, cambridge, ma: national bureau of economic research. https://doi.org/10.3386/w11085 ebeke, c., & ölçer, d. (2013). fiscal policy over the election cycle in low-income countries, imf working paper wp/13/153. https://doi.org/10.5089/9781475588514.001 enkelmann, s., & leibrecht, m. (2013). political expenditure cycles and election outcomes: evidence from disaggregation of public expenditures by economic functions. economics letters 121(1), 128–132. https://doi.org/10.1016/j.econlet.2013.07.015 fabrizio, s., & mody, a. (2006). can budget institutions counteract political indiscipline? european economic review 54(2), 269–278. fatás, a., & mihov, i. (2006). the macroeconomic effects of fiscal rules in the us states. journal of publiceconomics 90(1-2), 101-1 17. https://doi.org/10.1016/j.jpubeco.2005.02.005 ferree, k. e., gibson, c. c., & long, j. d. (2014). voting behavior and electoral irregularities in kenya’s 2013 election. journal of east african studies 8(1), 153–172. https://doi.org/10.1080/17531055.2013.871182 grier, k. (2008). presidential elections and real gdp growth in the usa, 1961 – 2004. public choice 135(34), 337-352. https://doi.org/10.1007/s11127-007-9266-6 higashijima, m. (2016). political business cycles in dictatorships. wias discussion paper no.2016-002 haider, a. s. m., shakil, s., fatema, s., & rezaul, k. m. (2016). impact of budget deficit on growth: an empirical case study on bangladesh. available online: https://www.researchgate.net/publication/298971557 (accessed on 31 may 2017). hassan, b. s., & hassan, a. o. (2011). ethnic politics and its implications for the survival of democracy in nigeria. journal of public administration and policy research 3(2), 28-33. hibbs, d. (1977) political parties and macroeconomic policy. american political science review 71(14), 67-87. https://doi.org/10.1017/s0003055400269712 hussain, m.e., & haque, m. (2017). fiscal deficit and its impact on economic growth: evidence from bangladesh. economies 5(37), 1-19. https://doi.org/10.3390/economies5040037 ifere, e. o., & okoi, o. b. (2018). political economy of fiscal deficits in a democracy. economia 19(1), 12–23. https://doi.org/10.1016/j.econ.2017.10.002 jensen, p. s. & justesen, m. k. (2014). poverty and vote buying: survey-based evidence from africa. electoral studies 33(2), 220–232. https://doi.org/10.1016/j.electstud.2013.07.020 jochimsen, b., & nuscheler, r. (2011). the political economy of the german länder deficits: weak governments meet strong finance ministers. applied economics 43(19), 2399-2415. https://doi.org/10.1080/00036840903194246 jones, m. p., meloni, o., & tommasi, m. (2012). voters as fiscal liberals: incentives and accountability in federal systems. economics & politics, 24(2), 135–156. https://doi.org/10.1111/j.1468-0343.2012.00395.x klomp, j., & de haan, j. (2012a). political budget cycles and election outcomes. public choice 2(1), 1-23. https://doi.org/10.1007/s11127-012-9943-y klomp, j., & de haan, j. (2012b). conditional election and partisan cycles in government support to the agricultural sector: an empirical analysis. american journal of agricultural economics 95(4), 793–818. https://doi.org/10.1093/ajae/aat007 klomp, j. & de haan, j. (2013). do political budget cycles really exist? applied economics 45(3), 329-341. https://doi.org/10.1080/00036846.2011.599787 kneebone, r. & mckenzie, k. (2001). “electoral and partisan cycles in fiscal policy: an examination of canadian provinces. international tax and public finance 8, 753-774. https://doi.org/10.1023/a:1012895211073 koksal, b., ahmet c., & kan, a. (2012). political business cycles and partisan politics: evidence from a developing economy. economic and politic 2(24), 182-199. https://doi.org/10.1111/j.1468-0343.2012.00397.x krause, g. a. (2005). electoral incentives, political business cycles and macroeconomic performance: empirical evidence from post-war us personal income growth. british journal of political science 35(1),77-101. https://doi.org/10.1017/s0007123405000049 lewis-beck, m. s. (1990). economics and elections: the major western democracies, ann arbor, https://doi.org/10.3386/w11085 https://doi.org/10.5089/9781475588514.001 https://doi.org/10.1016/j.econlet.2013.07.015 https://doi.org/10.1016/j.jpubeco.2005.02.005 https://doi.org/10.1080/17531055.2013.871182 https://doi.org/10.1007/s11127-007-9266-6 https://www.researchgate.net/publication/298971557 https://doi.org/10.1017/s0003055400269712 https://doi.org/10.3390/economies5040037 https://doi.org/10.1016/j.econ.2017.10.002 https://doi.org/10.1016/j.electstud.2013.07.020 https://doi.org/10.1080/00036840903194246 https://doi.org/10.1111/j.1468-0343.2012.00395.x https://doi.org/10.1007/s11127-012-9943-y https://doi.org/10.1093/ajae/aat007 https://doi.org/10.1080/00036846.2011.599787 https://doi.org/10.1023/a:1012895211073 https://doi.org/10.1111/j.1468-0343.2012.00397.x https://doi.org/10.1017/s0007123405000049 funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 204 mi, university of michigan press. https://doi.org/10.3998/mpub.8153 lilly, b. (2012). the canadian political business cycle. the economic and business history society 5, 1-38. maltritz, d. (2015). determinants of budget deficits in europe: the role and relation of fiscal rules, fiscal councils, creative accounting and the euro. economic modelling 48, 222-23. https://doi.org/10.1016/j.econmod.2014.12.001 manasse, p. (2006). pro-cyclical fiscal policy: shocks, rules, and institutions-a view from mars. imf working paper wp/06/27. https://doi.org/10.5089/9781451862874.001 mandon, p., & cazals, a. (2019). political budget cycles: manipulation by leaders versus manipulation by researchers? evidence from a meta-regression analysis. journal of economic surveys 33(1), 274–308. https://doi.org/10.1111/joes.12263 mohanty, r. k. (2012). fiscal deficit-economic growth nexus in india: a cointegration analysis. ph.d. dissertation. centre for economic studies & planning, school of social sciences, jawaharlal nehru university, new-delhi, india. moschovis, g. (2010). public spending allocation, fiscal performance and corruption. economic papers, 29(1), 64–79. https://doi.org/10.1111/j.1759-3441.2010.00048.x mosley, p. & chiripanhura, b. (2016). the african political business cycle: varieties of experience. journal of development studies 52 (7), 917-932. https://doi.org/10.1080/00220388 .2015.1113263 nayab, h. (2015). the relationship between budget deficit and economic growth of pakistan. journal of economics and sustainable development 6, 85–90. navaratnam, r. and mayandy, k. (2016). causal nexus between fiscal deficit and economic growth: empirical evidence from south asia. international journal for innovation education and research 4(8), 1-20. https://doi.org/10.31686/ijier.vol4.iss8.575 neuman, r., & ssozi, j. (2016). political influence on fiscal and monetary policy in sub-saharan africa. journal of african economies 25(1), 55-109. nordhaus, w. (1975). the political business cycle. review of economic studies 42, 169-90. https://doi.org/10.2307/2296528 north, d. c. (1990a). institutions, institutional change, and economic performance. cambridge university press, cambridge. north d. c. (2005b). institutions and the performance of economics over time. in: me´nard c, shirley mm (eds) handbook of new institutional economics. springer, dordrecht, 21–30. https://doi.org/10.1007/0-387-25092-1_2 obinyeluaku, m. i. (2006). testing the fiscal theory of price level in nigeria, university of kwazulunatal discussion papers series no. 58. odhiambo, s. o., momanyi, o., othuon, l., & fredrick, o. a. (2013). the relationship between fiscal deficits and economic growth in kenya: an empirical investigation. greener journal of social sciences 3, 306–23. https://doi.org/10.15580/gjss.2013.6.051113603 onye, k u., & okon, j.u. (2017). fiscal policy and monetary integration in wamz. an aerc thematic paper, wps. persson, t., & tabellini, g. e. (2003). “do electoral cycles differ across political systems?” igier working paper no. 232: 1-19. https://doi.org/10.2139/ssrn.392643 persson, t., tabellini, g. e. (1990). macroeconomic policy, credibility and politics. new york: harwood academic publishers. university of notre dame press. potrafke, n. (2010). the growth of public health expenditures in oecd countries: do government ideology and electoral motives matter? journal of health economics 29, 797–810. https://doi.org/10.1016/j.jhealeco.2010.07.008 rana, e. a., & wahid, a. n. m. (2016). fiscal deficit and economic growth in bangladesh: a timeseries analysis. the american economist 62, 31–42. https://doi.org/10.1177/056943 4516672778 rogoff, k. (1990). "equilibrium political budget cycles," american economic review 80, 21-36. rogoff, k., & sibert, a. (1988). elections and macroeconomic policy cycles. the review of economic studies 55, 1-16. https://doi.org/10.2307/2297526 roodman, d. (2009). “how to do xtabond2: an introduction to difference and system gmm in stata.” the stata journal 9(1), 86-136. https://doi.org/10.1177/1536867x0900900106 https://doi.org/10.3998/mpub.8153 https://doi.org/10.1016/j.econmod.2014.12.001 https://doi.org/10.5089/9781451862874.001 https://doi.org/10.1111/joes.12263 https://doi.org/10.1111/j.1759-3441.2010.00048.x https://doi.org/10.1080/00220388.2015.1113263 https://doi.org/10.1080/00220388.2015.1113263 https://doi.org/10.31686/ijier.vol4.iss8.575 https://doi.org/10.2307/2296528 https://doi.org/10.1007/0-387-25092-1_2 https://doi.org/10.15580/gjss.2013.6.051113603 https://doi.org/10.2139/ssrn.392643 https://doi.org/10.1016/j.jhealeco.2010.07.008 https://doi.org/10.1177/0569434516672778 https://doi.org/10.1177/0569434516672778 https://doi.org/10.2307/2297526 https://doi.org/10.1177/1536867x0900900106 funsho idowu obakemi et al. / european journal of government and economics 10(2), december 2021, 185-205 205 shi, m and svensson, j. (2002). 'conditional political budget cycles'. london, centre for economic policy research. https://cepr.org/active/publications/discussion_papers/dp.php? dpno=3352 sucharita, s. & sethi, n. (2012). fiscal discipline in india. romanian journal of fiscal policy 2(1), 1-23. alesina, a. g., & tabellini. (1990). “a positive theory of fiscal deficits and government debt. review of economic studies 57(3), 403-414. https://doi.org/10.2307/2298021 tujula, m., & wolswijk, g. (2007). what determines fiscal balances? an empirical investigation in determinants of changes in oecd countries. european working paper series 1, 22-45. udoh, e., joshua, n., & etok, e. a. (2012). political regime type, liberalization and fiscal deficits in developing countries: evidence from nigeria. european scientific journal 8(1), 22-38. van de walle, n. (2003). presidentialism and clientelism in africa’s emerging party systems. the journal of modern african studies 41(2), 297–321. https://doi.org/10.1017/ s0022278x03004269 vicente, c. p. (2014). is vote buying effective? evidence from a field experiment in west africa. economic journal 124 (574), 356–387. https://doi.org/10.1111/ecoj.12086 wyplosz, c., & kostrup, s. (2010). “a common pool theory of supranational deficit ceilings”. journal of development economics 73(1), 125-154. yaru, m.a., mobolaji, h. i., kilishi, a.a., & yakubu, a.t. (2014). budgetary institutions and fiscal discipline in nigeria. annual conference of the nigerian economic society. https://cepr.org/active/publications/discussion_papers/dp.php?dpno=3352 https://cepr.org/active/publications/discussion_papers/dp.php?dpno=3352 https://doi.org/10.2307/2298021 https://doi.org/10.1017/s0022278x03004269 https://doi.org/10.1017/s0022278x03004269 https://doi.org/10.1111/ecoj.12086 political business cycle and fiscal discipline in sub-saharan africa 1. introduction 2. literature review 3. methodology 4. results and discussions 5. conclusions references path – dependence and european fisheries management vol.7 • no.2 2018 issn: 2254-7088 european journal of government and economics 7(2), december 2018. european journal of government and economics issn: 2254-7088 number 7, issue 2, december 2018 doi: https://doi.org/10.17979/ejge.2017.7.2 transparency and local government corruption: what does lack of transparency hide? 106-122 juan luis jiménez, daniel albalate 106-122 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real. 123-137 carlos fernández_herraiz, antonio javier prado domínguez, carlos pateiro-rodriguez, jesus m. garcia-iglesias path – dependence and european fisheries management 138-153 federico martín palmero, fernando gonzález laxe why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem 154-170 filipa figueira deciding on financial renegotiation in public-private partnership projects 171-199 wiston risso https://doi.org/10.17979/ejge.2017.7.2 european journal of government and economics 7(2), december 2018, 138-153. european journal of government and economics issn: 2254-7088 path – dependence and european fisheries management federico martín palmero a, *, fernando gonzález laxea a universidade da coruña, spain * corresponding author at: fgmartin@udc.es article history. received 2 october 2018; first revision required 8 november 2018; accepted 3 december 2018. abstract. for a long time there has been an attempt to explain the current crisis in the fisheries sector in terms of a confrontation between those defending “fish” and those defending “fishermen”. however, the real problem concerns the governance of fisheries and how common resources are assigned; it is not just a crisis of resources per se. therefore, an insightful understanding of the scenario leading to a satisfactory solution is more complex than it is often believed since there is a need to tackle problems related to the state of stocks, fishermen’s strategies and ecosystems. the fishing sector is not exclusively concerned with production activities as some analysts would have us believe. rather, it is an area that integrates a number of important features and requires different approaches dealing with the industrial aspects of the sector, distribution and consumption. the fishing sector is characterised by a high level of public intervention, in terms of regulation, finance and state subsidies. the plethora of norms has become such that, currently, the main areas of debate are those concerning how best to preserve resources and ecosystems (by managing and sustaining certain economic levels for example), the welfare of those who make their living from fishing, and the social impact on coastal communities among others. the main focus of debate used to be the conditions of access to fishing and fisheries. nowadays, however, since early 2000s, efforts have concentrated on the limits of biological safety in order to guarantee sustainable and efficient fishing.this work carries out a dual analysis of the objectives of fisheries management. the first focuses on path dependence and the second on a debate among the three main players and their changing views. this approach allows us to clarify the different interests as regards policy-making, as well as to clearly define the different management implementation. keywords. fisheries management; common fishing policy; path dependence theory. doi. https://doi.org/10.17979/ejge.2018.7.2.4411 1. introduction anthony charles (1992) suggested that the analysis of fisheries management should be a mutual commitment on the part of the three main players to harmonize their conflicting points of view. he called these perspectives the conservation, rationalization and socio-community paradigms. each of these holds certain objectives to be more important than others when it comes to fisheries management and regulation and each prioritizes certain kinds of instruments in benefit of a particular socio-economic group. the terms of the commitment may be either more or less near each of the triangles of the paradigms. mailto:fgmartin@udc.es https://doi.org/10.17979/ejge.2018.7.2.4411 f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 139 in light of the history of the common fishing policy (cfp), many variables of the said paradigms are seemingly incompatible (holden, 1994, symes, 1997, 2007, lequesne, 2001, griève, 2001, gonzález-laxe, 2002, gray & hatchard, 2003). this disagreement has become so profound that the conservation measures recommended by experts and the decisions taken by policy-makers are blatantly at odds. also, thanks to this disagreement, the failure of attempts to palliate the effects of more intensive fishing by rebuilding fishing stocks is also evident. in addition, the impact wrought by the use of instruments clearly counter-productive in an economic and biological sense is both patent and verifiable. at present, after the passing of the 2013 reforms, it becomes highly pertinent to refocus upon the idea of path-dependence. further, on revisiting the thesis put forward by charles (boncoeur & mesnil, 1999) and his proposed paradigm, it becomes possible to demonstrate that there is a new “triangulation of conflicts and decisions”. these will be classified using different perspectives. the most important may be stated thus: is fishing policy en european policy or just the result of a set of common rules for certain aspects of fishing policy? is the combined set of principles of subsidiarity and relative stability well defined? are the governments of each member state the main interlocutors of their fishermen? or, in contrast, are the economic agents the interlocutors of the commission? these ideas lead to a new analysis, one which highlights the novel power relationships among the actors, the tensions and pressures upon the agents, and the preferences of both with respect to how to define the bases and principles of fisheries management and regulation. it is evident that with an analysis dealing with how the cfp works and its current status, it will be impossible to reach either a methodological consensus or a universally acceptable solution. despite the principles of rigour and coherence adopted by different researchers, it is clear that there is a “methodological pluralism” which adapts itself to new and successive methods of analysis. our study does not involve highlighting clashing methodological approaches, but instead it attempts to utilize methods that explain the current status of agents and how these behave. that is, this analysis is based upon a constructivist paradigm, with an objective review of the situation, combined with a critical evaluation. by so doing, problems become conceptualized within the dynamic of social patterns that underpin the current dynamic. in short, each actor is subjective as regards perceptions and proposals. contributions to fisheries analysis from anthropology and sociology are more and more usual. these contributions provide a wide view about both the establishment and the explanation of stable collaborative relationships. they avoid short-term particularism by trying to integrate local habits of population as the essential tool to manage and protect fisheries, mostly traditional and artisanal fisheries (berkes, 2012). there are, therefore, a wide range of objectives linked to certain public policies. as a consequence, it becomes necessary to apply techniques for multi-objective programming. this is far from easy, given the qualitative heterogeneity of fisheries management. that is why the f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 140 study deals with concepts that may be deemed to be satisfactory or acceptable, in other words pre-established objective or limit variables. logically, the optimization involves minimizing the weighted sum of the deviations with respect to these objective variables. by examining the resolutions passed by the european parliament, we can see that the main objectives of fisheries management in the eyes of public administrations focus on the conservation of commercial stocks, the minimization of conflicts among fishermen and safety on board. the remaining objectives are considered to be less important. for producers, the two main objectives are the conservation of commercial stocks and profitability. for the scientific community the objectives focus upon the conservation of stocks and the minimization of negative impacts on ecosystems; and, for some producers organizations the objective is the implementation of protected maritime areas to be complementary with fisheries management measures. does the cfp have the power to incorporate each group’s preferences? the short answer is no, although the final result is the sum of small alternatives with diverse periods of transition together with binary rather than total agreements. in short, this means that the views of each group are highly differentiated. figure 1. fishery management objectives. source: own elaboration. in short, anthony charles’ paradigm, which highlighted conservation (linked to scientists), rationalization (to producers), and the socio-community paradigm (to institutions), has slightly changed. nowadays, the relative stances of each agent are more clearly influenced by a worldwide vision rather than a trenchant defence of each group’s specific interests. in this line, it is pertinent to site four salient elements, the so-called “big problems” within the field of fisheries management. by common consent these four axes are as follows: a) sustainability; b) criteria about maximum sustainable yield (msy) and its implementation; c) the implementation of tacs and the individual transferable quota system (itq); and d) actions to reduce overfishing. sustainable management conservation maintaining commercial stocks at biologically reasonable levels minimize the effort oriented to the fishing of non-commercial species and ecosystems social and economic efficiency improve fishing profitability improve safety and onboard working conditions promotion of employment in fishing and in fishing dependent areas prevention of conflicts among fishermen groups among fishermen and other groups f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 141 first and foremost, it is important to highlight the fact that fishing sustainability cannot be gauged or analyzed in absolute terms (daw & gray, 2005). lessons from previous mistakes should be learnt and assimilated and, by so doing, resources should be managed in a healthier and more ecologically way. sustainable development is the final objective of any policy, so we should be prepared to rectify mistakes rapidly once the relationships between mankind and nature begin to collapse. therefore, key decisions should not be burdened by an increase in red tape or driven by the desires of a single lobby, either multinationals or coastal communities. sustainability depends more on the correct choice of a viable underlying policy, the levels of democracy of policy-makers, and on how to apply models to monitor resources. hence, it should not be forgotten that the political processes to best control natural resources depend upon those communities that live on fishing and that decision making in the heart of the fishing community should be as participative, open and democratic as possible. the second consideration refers to criteria establishing maximum sustainable yield (msy). the european commission is backing both the absolute concept of sustainability and the fixing of an msy for a specific date (2015). for some scientists, this is not a viable strategy (mace, 2001; mesnil, 2012). history records that, on occasions, it has taken a long time to restore overexploited stocks; or that the exact msy is not known for all stocks or for specific species. also, the wide annual variation in stocks and the complex interrelations among the different species make it very difficult to achieve the msy in 2015 (froese & proelss, 2010). in short, there are many factors that affect the fishing mortality. these include pollution, climate change, and mistakes in scientific and fisheries management. all these factors, among others, make it problematical to reach the objective. the third bone of contention lies on the implementation of fishing quotas, the application of individual quotas of transferable fishing (itqs), and the privatization of resources. the european commission is fighting to achieve the illusive msy as fast as possible, essentially basing its approach upon adjustments in the total allowable catch (tac) and the quotas of transferable fishing. this approach is by no means the only solution and falls short of guaranteeing a healthy outcome for its components and actors. it might be the appropriate measure for certain highly specific stocks and well-known stocks which have been evaluated for decades in certain areas. in contrast, this is probably not the optimum solution for multi-specific fisheries; perhaps an attempt to limit efforts and a variety of selective controls might allow for greater flexibility when it comes to fishing ordination and, by extension better results. a second line of criticism attempts to show that when the tac system and quotas come into force, this will irreversibly lead to privatization and an increase in catch-costs that encourage overfishing. if this were true, the leverage of financial institutions with respect to fishing would be reinforced; capital would become more concentrated, the number of fishing vessels would decrease, and the cost of market entry for new fishing vessels would rise. as a direct consequence, more expensive quotas would lead to more intensive fishing. in short, we might be on the eve of a great transformation in fishing and fisheries brought about by the imposition of quotas. this would accelerate the ascendency of financial criteria which would weaken the sustainability of f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 142 fishing systems and place the whole idea of sustainable fishing at risk. the fourth and final key factor to be taken into consideration is overfishing. according to european reports, the countries that make up the ec hold overfishing to be the prime cause of the crisis. as a consequence, the revealed aim of cfp reforms, right from the outset, has been to eliminate between two thirds and one half of fishermen and vessels in order to rapidly achieve the msy. the coming into force of systems such as the tac and fishing quotas will allow this to be carried out much faster whilst incurring minimal costs. the possible sale or leasing of quotas by artisanal fishing to the large groups would allow the financial sector to eliminate artisans without the need for public aid. the free concession of quotas would make operations profitable for financial groups and, by so doing, guaranteeing good profits. similarly, all of the actions carried out, including modernization, adaptation, construction and the formalization of joint-ventures, etc. have been developed with the help of state aid and, without exception, there has been no connection with managerial measures. hence, on certain occasions, there has been an increase in fishing potential but only at the expense of an excessive pressure upon resources. as a consequence, “preservation” is now the key-word, when it is necessary to apply severe and urgent measures. in order to understand the triangulation of positions in the cfp debate, it is evident that these elements must be given serious consideration. other specialists argue that it is impossible to apply administrative measures without taking into consideration the markets. it should also be highlighted that, under the banner of trade liberalization, there are no foreseeable safeguards for the protection of local production. in the same line, it becomes hard to mobilize producers against contingency measures, price falls or the introduction of eco-labels for certain species. 2. the concept of path dependence the concept of path-dependence is prevalent in the social sciences; recently it has become the focus of analyses in the fields of law and economics. the concept itself seeks to answer a series of questions. why to change the current scenario proving to be so difficult? why are certain social processes held back by inertia rather than moving towards change? and, finally, why is it impossible to replace some institutions? without a doubt, the answers to these quandaries are complex since they must include dynamic and multidisciplinary elements which must be analyzed from different levels. normally, the evolution of economic systems is carried out by using developmental/evolutionist theories, and focusing on the reach of the impact of the adopted measures. in this sense, when dosi & nelson (1994) explain the bases of evolutionary theory in economics they cite two characteristics. the first is an explanation of movement or change over time or an explanation of why certain things are as they are at a given time while emphasising f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 143 the causes that have led them to that juncture (the dynamic proposal). the second involves giving opportune explanations in order to contextualize both random elements (which allow us to distinguish what is being generated and renewed) and those that explain any substantial variation. path-dependence is based on the idea of a dependent trajectory, which is both inherent and key to diffusion processes. this explains why path-dependence has had little traction in economics. david (1998) believed that path-dependence was the property of a contingent irreversible dynamic process, which includes a wide set of social and biological processes which might be accurately described as evolutionary. that is, “history matters” and “observable results are a function of history itself”. the same author elucidates by stating that path-dependence is made up of “a dynamic property related to the idea of a history composed of divisible, irreversible processes”. as a consequence, certain historical phases can generate inefficient situations which, on occasion, can be foreseen. hence, path-dependence includes historical factors and biological and economic analyses, which are, in turn, broadened by neoclassical analyses. path-dependence can be defined as a theory that states that “the present is the result of choices made in the past”. therefore, in essence path-dependence is a form of analysis that attempts to discover the remote, historical causes of the social structures and processes of the present. this theory substitutes the evolutionist hypothesis based upon the idea that it is possible to explain movements over the long term or how and why certain factors change, and what the causes of these changes are. in this sense, the theory of path-dependence seeks to explain factors such as how to interpret the fact that certain contingent events are capable of producing far reaching social results; or how these social results are capable of replicating themselves without the existence of the original forces that brought them to life; or finally, the characteristics of those reproduction processes that prevent or impede change. the bases of path-dependence include what are known as self-reinforcing sequences which underline the long-term formation and reproduction of an institutional model; that is, a standard which, over time, becomes more difficult to change or substitute for other more readily available options. this facet of path-dependence means that it is ever more difficult to change the initial state given the undeniable supremacy of the original factors. path-dependence therefore, is made up of reactive sequences which are chains of events, chronologically ordered and connected causally. each event within a sequence is a reaction to an event which occurred previously. this type of sequence is important when it comes to explaining the final results because a small change in one of the events is capable of generating considerable differences in the final result of the sequence of events. f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 144 3. path-dependence and common fishing policy common fishing policy is a rule; it is the product of a decision. a different option might have been chosen. however, cfp, as an approved judicial rule, is gradually becoming a contingent event; that is, a point of confluence from which the institutional standard is reproduced. this has nothing to do with the fact that the political powers or actors have changed over time. it is evident that from the origins of the cfp agents have changed together with society and its interests. nevertheless, the rule has remained unbending and, more surprisingly, it will be unlikely to change. on analyzing the causes of this phenomenon we find that the analytical structure of pathdependence includes the following six factors: a) the starting point is a balanced situation. that is, ad initio there is a wide number of viable alternatives from which to develop the policy in question, or for the development of the institutions being examined. b) different random events might exercise a substantial function upon the establishment of a given policy or upon a specific institutional type. c) the conditions via which the systems of path-dependence reproduce their forms or condition their sequences must be specified. d) policies or institutions must generate feedback mechanisms that create inertia or possible returns in order to avoid the emergence of rival policy ideas or new sets of interest. e) once the trajectory has been established, this same trajectory will tend to generate an inertia whereby established individual and cultural interests will have a high opportunity cost when it comes to changing the system. f) finally, there will be a mechanism that allows for changes in the system of path-dependence, changes which are located in the interaction of cultural spheres, structural elements and human activity. the analytical structure may be explained graphically as follows, in table 1. table 1. analytical structure of path dependence background analysis of the current situation structural persistence relative sequences results historical factors that define the viable options and mould the selection processes the selection of one option among various alternatives the production and reproduction of institutional or traditional structural patterns reactions and counter-reactions to institutional or traditional structural patterns the solution of conflicts generated by reactions and counter-reactions source: own elaboration. f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 145 the main objectives of cfp, which officially came into being in 1983, were as follows: to prevent overfishing, to guarantee that fishermen had the means of production that were both permanent and lasting; to provide a regular supply to consumers and those involved in processing such that both volume and price remain relatively constant; to improve conservation and the management of resources; and to ensure balanced territorial development. after the initial postulates, cfp underwent two reforms in 1992 and 2002. the former was characterized by an attempt to better harmonize fishing fleets and stock management; by the actions and measures that tended to reduce the number of vessels; and by the instruments that facilitated the financial aid to attenuate the social consequences of the adjustments paralysations of fleets. the main axes of the first reform were the progressive implantation of fishing licenses among producers in order to reduce overfishing. the second reform is consistent with the first. the measures aimed at tackling overcapacity are now the main concern of the second reform and, as a result, the concept of “fishing effort” is applied in reaction to and as a response to the constant decline in stocks and biomass. the instruments used in order to do this were; the generalized establishment of the tac, according to species and fishing areas; the reinforcement of controls, the provision of aid to restructure the fleet, and a firm commitment to radically reduce the capacity of the fleet. the results of these decisions meant that public aid for the construction of vessels was now subject to control, a dynamic which is clearly expressed in the recent european court of auditors’ report (2011). further, aid aimed at modernization was only authorized when this did not involve an increase in fishing capacity. however, in spite of the reduction in the capacity of the fleet, levels of biomass and stocks and the health of these stocks are still below safety levels. the present discussions are intensifying scrutiny as to the judiciousness of the postulates and measures adopted right from the birth of the cfp. warnings heavily underline the inadequacy of measures and actions to preserve, conserve and manage resources. what is also stressed is a lack of connection among the mechanisms that should have balanced and harmonized levels of fishing exploitation in line with rational tenets that aimed to address the competition and rivalry among fishermen and european states. there is, in effect, an historical vicious circle and a whole set of blatant contradictions and inconsistencies. there is not solution for the triangulation of the problems: biological, environmental, and those related to ecosystems; economic and financial; and finally, social problems in terms of participation and responsibility. in short, the same objectives are mooted, while scarcely taking into account far reaching changes and transformations. further, the specifics, as stipulated in the new reform, are beginning to fall by the wayside as those involved argue that said commitments require additional time. this means that the period of implantation needs to be longer. for example: a) the definition of maximum sustainable yield (msy) for all of the species after 2015; b) the total prohibition of discards after 2015; c) limiting transferable quotas, until a more suitable juncture; d) the establishment of multi-annual recovery plans for fish stocks; e) the adaptation of governance my means of increased regionalization in a future reform; f) or the harmonizing of f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 146 international stances with respect to fishing activity whenever possible. various researchers have been confirming the failures and mistakes inherent in the cfp for some time (khalilian et al., 2010, gonzález-laxe, 2010). the question arises whether the new reforms should give preference and reinforce preceding measures, in the hope that they turn out to be more efficient; or, in contrast, try and learn from what happened in the past, thus facilitating the emergence of new actors and the assimilation of new types of innovative instruments and tools. table 2 shows the different positions in order to help to understand these controversies. table 2. synopsys of the negotiation of the cfp reform subject initial ec proposal agreements and disagreements sustainable management maximum sustainable yield eco-systemic approach fishing plans aims to ensure that, by 2015, all stocks are being exploited in a sustainable way via the application of fishing quotas and a measure of maximum sustainable yield. the eco-systemic approach will apply to the multi-annual plans with the aim of taking into account the links between fishing and eco-systems. mono-specific fishing plans should be replaced by plans based on fisheries. more flexible. in 2015 if possible. establishment of a new date: 2020. current plans are not based on this principle due to the difficulty of creating and implementing highly reliable mechanisms. failure of the attempt to substitute the tac mechanism and quotas with a system based on fishing effort. no deadlines or time limits are established and there are no new specific administrative measures. discards there is a commitment to eliminate all discards from 2016 onwards. the date has been moved back and 5% of total captures may still be rejected. the closest to a total discard will be in 2019. relative stability the key to the distribution of quotas by fishing area according to country is not subject to discussion by member states. relative stability is nonnegotiable. hence, there continues to be historical discrimination with respect to certain fleets. fishing consessions from 2014 onwards there will be a new system for the concession of fishing rights for vessels of over 12 metres in length. this will be mandatory with the aim of adjusting fleet capacity to resources. an attempt to boost the economic viability of vessels. the introduction of individual concessions of transferable fishing quotas will be voluntary and will be established by each member state and limited for each of them. the fleet adjustment programmes will continue, but without public aid. european fisheries fund there will be a new european fund for the fishing sector that will take in the period 2014-2020. there is a demand for greater clarification of contents and powers. illegal fishing attempts to prevent the rise of a parallel market. a profound reform of the control and execution system will be requested so that measures are applied uniformly throughout the eu. the remit of the european fisheries control agency will be broadened. all countries are in agreement. however, the legislation is different with respect to sanctions and fines. f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 147 subject initial ec proposal agreements and disagreements external dimension of fishing bilateral fishing agreements are being ignored since they are considered to be too expensive. it must be a priority and should manifest itself in the defence of the fishing company in all international fishing organisms and institutions and with respect to all bilateral agreements. public aid there is a move towards a reduction in public aid and subsidies. aid for scrapping vessels and for temporary stoppages is intended to continue. artisanal fishing not contemplated as a separate branch of the industry. there is a demand for a differentiated system since artisanal fishing and shellfishing are 2 singular types of fishing, and are specific to specific territories. common organization of markets they maintain the classical structure derived from the time when cfp came into existence and based on formulas established in the 1980’s. there is a demand for better market information and norms that establish “fair trade”. moreover these should be accompanied by a more transparent and competitive international framework which incorporates a greater level of equality and transparency with respect to the interchange of products imported and exported. participation and regionalism the commission wants to maintain its power in the decision making process that affects the cfp. so the regional advisory councils (rac) will be able to elevate their opinions to the commission but without these being binding. the aim, therefore, is to increase the degree of compliance and that the rac promotes dialogue among the interested parties. the resolutions must be in consonance with the instruments of governance, with a bottom-up approach and there must be a consensus among the agents. the role of the rac must be strengthened and this aspect is only guaranteed if there are ample financial resources and if the resolution proposals emitted by the rac are backed. source, own elaboration an initial impression suggests that the thesis and general principles underlying path dependence are perfectly apt for explaining the history and workings of the cfp. in short, there is great resistance to change; forms are always the same; there is a high degree of inertia and a considerable weight of historical and cultural conditions, as well as an administration that reinforces specific and private interests. 4. analysis of institutional and contractual factors the three main actors involved in fishing reforms (producers, institutions and researchers) have highlighted an opportunity to clearly set out what fishing activity involves and to formulate new adjustment and sustainability proposals. further, they have underlined that there are other external factors that might significantly influence the evolution of stocks. these include, climate change, the warming of the seas, the new distribution of species, pollution, etc. f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 148 table 3. behavioural code according to objectives institutional level scientific area sectoral plan resource management positive evaluation with respect to the recuperation of stocks, and positive expectations with regard to the future. the adoption of planning and management measures, but, with a delay and inconstancy in their implementation. the delays in remedying the imbalances are criticised. scientists state that the institutions charged with applying the measures do not follow their recommendations. they usually demand more specifics when it comes to defining the planning measures. highly critical of decisions. besides taking into consideration environmental factors, the reform should also consider social and economic conditions; that is, how employment and the areas themselves are affected. maximum sustainable performance establishing medium-term goals (7 years), a grace period for harmonization planning among member states, with respect to the method for assigning fishing options. the debate on planning mechanisms remains open: specifically, with respect to msy and the precautionary principle. certain scepticism. the application of measures depends upon the role assigned to scientific opinion (executive competence or consultation); and on budgets for their subsequent application. very critical. elements of uncertainty should be taken into account such as those relative to the size and productivity of populations, reference levels and the distribution of mortality among other indices. discards acceptance of the objective, but not its immediate application. in favour of its immediate implementation. they talk of its necessity immediate implementation is not possible. it is argued that scientists are out of touch with the underlying reality of fishing and underline that there is a disconnection among fishing institutions individual fishing quotas (ifqs) favourable, but conscious of the difficulty and complexity of their distribution favourable, but conscious of the need to implant them in line with the needs of the fishery it’s necessary to distinguish between types of fishing. there is no homogenization. artisanal fishing emphasises its relevance and are aware of its role. imprecision of definitions. they believe it is fundamental boosting artisanal fishing is essential. aid and support, whether this be via compensation or price controls sustainability feasible commitment rigorous control mandatory reference regionalization listed as a high institutional cost. difficulties with a definition and derived policies: renationalization and decentralization they emphasise the need to clearly define what the “conditional transfers” are in each area or fishery. international dimension defined as a highly important objective. however, budget restrictions are the reason why fishing agreements have not been renewed. disparity of criteria among member states plausible objective they underline its importance and that it requires greater institutional commitment source: own elaboration. f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 149 in the same line, it has been stated that since the origin of the cfp there have been many changes. in resume, the process has not been a total failure; there has been a certain progress to the extent that various member states organize the management of shared resources together by following a decision making process which, in the main, is characterised by a qualified majority. in the light of these considerations, it is easy to see that european council constantly gives in to pressure from the productive sector and regularly fails to follow the recommendations of scientists (villasante et al., 2011). this is normally justified by arguing that, on analyzing the evolution of stocks, it is interpreted that the objective of the cfp-2002 is not to achieve an msy, but rather to guarantee levels of stocks within reasonable biological limits and with sustainable exploitation (article 3, regulation 2371/2002). therefore, not having achieved the objectives cannot be a reason of reproval. as a consequence, it is possible to present an analysis of the economic motives derived from the behaviour of the actors. that is, a theoretical-practical description in the form of a code with economic pretentions. this is not intended to be an exhaustive analysis of the different actors’ positions. any attempt to include all casuistic reasoning or relevant specifics would be overwhelmed by a surfeit of information. the exercise we are carrying out is merely to illustrate the different stances (in some cases entailing permanent conflict) among the various perspectives. thus, the analysis includes both features directly connected to tradable and non-tradale issues, in order to contextualize and broaden the possible spill-over effects of the crisis. finally, the conflicts and disputes between compensatory and competitive policies and actions are highlighted. in short, we conclude that there are no restrictive variables or variables which are, a priori, more important than others; rather they are based upon the combined behaviour of a large number of agents. secondly, there is no obsession to achieve any kind of balance. third, the short-term objectives are constantly changing due to the heterogeneity of the agents and their own particular circumstances. 5. concluding remarks regulatory fishing policy has gained great importance in recent decades (arnason, 2001, clark, 2006, carciofi & azqueta, 2012) and the european union has been involved in much of it. the reason for the new push is that stocks have been placed under intense pressure. the main reasons for this are the oversized world fleet and, by extension, excess capacity, the fact that the technology for detecting and capturing resources evolves very quickly and exerts strong pressure on captures, and finally, the difficulty to obtain planning or management instruments sufficient to guarantee that objectives are reached, that stocks are re-built and there will be profitability for producers in the very near future. the actions carried out are interdisciplinary in nature and this tends to magnify complexity while offering different partial and independent f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 150 solutions. fisheries are controlled by situations where human interactions are combined, where fishermen follow and keep a group of rules in order to be able to coordinate their interactions with the others. this group of rules allow to restrict decisions and also to give form to individual preferences. in this sense, fishermen follow very specific patterns: a) the agents (fishermen) can take into account the consequences of their decisions according to their individual preferences, affecting both themselves and those with whom they interact; b) the agents (fishermen) are adaptive, that is, they follow the established rules. this means they can reduce those costs derived from limitations; and c) the agents (fishermen) behave according to the context; they can determine if this context is appropriate or not, bringing about an information process and generating incentives in the decision-making. in short, friction is transmitted by various basic channels. the first of these resides in the role of national interest and how that of one country or a group of countries clashes with the needs and perspectives of the rest. this prevents from updating the map of those regions which are highly dependent on fishing since this involves overlaying different kinds of plans and maps. the first layer includes spatial or territorial maps; specific and segmented maps according to fleet and also depending on the final markets where the product is eventually delivered. the second layer of mapping refers to the relative balance between the changing potential in the volume of captures and fleet capacity. in this regard, the tried processes of restructuring and modernization become to be highly relevant in terms of economic aid and a commitment to create mixed or joint-venture firms. the third set of plans includes the dynamics inherent in the process of liberalization taking place on the demand side of fishing markets; these affect the reductions in tariffs, contingencies, health conditions, labelling and the normalization of products, as well as the concept of responsible fishing. the fourth facet of the composite route map deals with adjusting the planning models to the reality of fishing. in this respect, there is a need to establish whether or not the fishery in question is multi-species and if the scenario is one in which vessels are capturing one particular species rather than several. fifth and finally, competition within some elements of the fleet impedes sustainable growth because there emerge distorting factors and variables, such as the concept of relative stability, and the dynamics of market liberalization which thwart, hinder and limit the development and potential benefits that inputs might derive. likewise, fisheries sector presents negative externalities coming from the no-consideration of the effects originated by the other fishermen and their own effects on fishing reserves. these externalities appear when these two conditions are met: a) fishing activity of an agent f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 151 (producer) causes the other producers to lose their well-being; and b) this loss of well-being is not compensated. the externalities in the fisheries sector are a consequence of market failures and this brings about serious consequences as regards fishing management policies. hence, there is a significant discrepancy between short-term and long-term objectives. if there are no incentives so that preservation effects can be taken into account, the short-term objectives of fishermen can prevail over long-term objectives. this derives in the so-called “social trick” (seijo et al. 1997), where market mechanisms are not able to achieve the optimum by themselves, but they give rise to the over-investment and over-exploitation of resources. to sum up, in view of these divergences, it may be observed that the european parliament and commission are unwilling to renounce immediate or short-term objectives. also, the council search for greater flexibility in the conceptual framework of the new cfp. the cfp is a wonderful source for analysing the path-dependence as it allows us to compare public and private efficacies, as well as the stable and new implemented actions and the framework of reference. in this case, we confirm that north’s thesis (1990) was right when defending that these actions and frameworks of reference were unstable. in this line, the study and analysis focusing on organizations and on institutional changes and transfers have become a broad field of research used evermore frequently by economists. not only are institutional efficiencies highlighted, and these exist at every level of the decision making process, but also they can be persistent. this leads to great inertia, mainly because of the innate limitations of human beings to reorganize themselves more flexibly and to be more adaptable the focus of those studying social change. acknowledgements. this paper was supported financially by the ministry of economy and competitiveness, government of spain [project reference number: cso2016-78122-r] references arnason, r. (2001). on applied fisheries economics. eafe, salerno, 18-20, april. berkes (2012). implementing ecosystem–based management: evolution or revolution. fish and fisheries. 13, 465-476. doi: https://doi.org/10.1111/j.1467-2979.2011.00452.x boncoeur, j., mesnil, b. (1999). surexploitation des stocks et confits dans le secteur des pêches. une discussion du «triangle des paradigmes» de a.charles, dans le contexte européen. informations et commentaires, 107, 10-17. carciofi, i., azqueta, d. (2012). territorio, desarrollo tecnológico y gestión de recursos naturales renovables: el caso de la pesca. investigaciones regionales, 23. https://doi.org/10.1111/j.1467-2979.2011.00452.x f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 152 charles, a.t. (1992). fishery conflicts: a midfield framework. marine policy. 16(5), 379-393. doi: https://doi.org/10.1016/0308-597x(92)90006-b clark, c. (2006). the worldwide crisis in fisheries. cambridge university press. uk. daw, t., gray t.s. (2005). fisheries science and sustainability in international policy: a study of failure in the eu common fisheries policy. marine policy. 29(3), 189-197. doi: https://doi.org/10.1016/j.marpol.2004.03.003 david, p.a. (1998). path dependence: its critics and the quest for historical. oxford stanford university. working paper. dosi, g., nelson, r.r. (1994). an introduction to evolutionary theories in economics. evolutionary economics. 4,153-172. doi: https://doi.org/10.1007/bf01236366 european court of auditors (eca) (2011). have eu measures contributed to adapting the capacity of the fishing fleets to available fishing opportunities? special report nº 12. luxembourg. froese, r., proelss a. (2010). rebuilding fish stocks no later than 2015: will europe meet the deadline? fish and fisheries. 11(2), 194-202. doi: https://doi.org/10.1111/j.14672979.2009.00349.x gonzález-laxe, f. (2002). desarrollo y estrategia de la pesca europea. ed.netbiblo. a coruña. doi: https://doi.org/10.4272/978-84-9745-031-7 gonzález-laxe, f. (2010). dysfunctions in common fishing regulations. marine policy. 34(1), 182-188. doi: https://doi.org/10.1016/j.marpol.2009.06.003 gray, t., hatchard, j. (2003). the 2002 reform of the cfp system of governance-rhetoric or reality? marine policy. 27, pg. 545-554. doi: https://doi.org/10.1016/s0308-597x(03)00066-6 griève, ch. (2001). reviewing the cfp. eu fisheries management for the 21st century. institute european environmental policy. london holden, m. (1994). the common fisheries policy. origin, evolution and future. the buckland foundation. fishing news books. oxford. khalilian, s., froese, r., proelss, a., requate, t. (2010). designed for failure. a critique of the cfp of the european union. marine policy 34, 1178-1182. doi: https://doi.org/10.1016/j.marpol.2010.04.001 lequesne, cl. (2001). l´europe bleue. a quoi sert la politique communautaire de la pêche. presses de sciences po. paris. mace, pm. (2001). new role for msy in single species and ecosystem approaches to fisheries stocks assessment and management. fish and fisheries 2(1), 2-32. doi: https://doi.org/10.1046/j.1467-2979.2001.00033.x mesnil, b. (2012). the hesitant emergence of msy in fisheries policies in europe. marine policy, 36, 473-478. doi: https://doi.org/10.1016/j.marpol.2011.08.006 north, d. (1990). institutions, institutional change, and economic performance. symes. d. (1997). the european community common fisheries policy. ocean and coastal management, 35(2-3), 137-155. doi: https://doi.org/10.1016/s0964-5691(97)00030-6 seijo, j.c., defeo, o., salas, s. (1997). bio-economía pesquera. teoría, modelación y manejo,. https://doi.org/10.1016/0308-597x(92)90006-b https://doi.org/10.1016/j.marpol.2004.03.003 https://doi.org/10.1007/bf01236366 https://doi.org/10.1111/j.1467-2979.2009.00349.x https://doi.org/10.1111/j.1467-2979.2009.00349.x https://doi.org/10.4272/978-84-9745-031-7 https://doi.org/10.1016/j.marpol.2009.06.003 https://doi.org/10.1016/s0308-597x(03)00066-6 https://doi.org/10.1016/j.marpol.2010.04.001 https://doi.org/10.1046/j.1467-2979.2001.00033.x https://doi.org/10.1016/j.marpol.2011.08.006 https://doi.org/10.1016/s0964-5691(97)00030-6 f. martín palmero and f. gonzález laxe / european journal of government and economics 7(2), 138-153. 153 fao. doc. tec. pesca, nº 368. roma symes, d. (2007). fisheries management and institutional reform: a european perspective. ices journal of marine science, 28, 1-7. doi: https://doi.org/10.1093/icesjms/fsm007 villasante, c.s., garcía-negro, mª c., gonzález-laxe, f., rodríguez, g. (2011). overfishing and the cfp: (un)successful results from tac regulation? fish and fisheries, 12(1), 34-50. doi: https://doi.org/10.1111/j.1467-2979.2010.00373.x https://doi.org/10.1093/icesjms/fsm007 https://doi.org/10.1111/j.1467-2979.2010.00373.x ejge_front_7_2 ejge oficial 7-2 content page 7-2-3 abstract. for a long time there has been an attempt to explain the current crisis in the fisheries sector in terms of a confrontation between those defending “fish” and those defending “fishermen”. however, the real problem concerns the governance of ... keywords. fisheries management; common fishing policy; path dependence theory. doi. https://doi.org/10.17979/ejge.2018.7.2.4411 references varieties of capitalism and labour market opportunities for the youth european journal of government and economics 9(3), december 2020, 232-251 233 european journal of government and economics issn: 2254-7088 varieties of capitalism and labour market opportunities for the youth: a comparison of attitudes towards skill formation felix hörisch a*; jale tosunb; julian erhardtc; william maloneyd a university of applied sciences, saarbrücken, germany b heidelberg university, germany c university of bern, switzerland d newcastle university, united kingdom * corresponding author at: felix.hoerisch@htwsaar.de abstract. in this study, we examine the extent to which socio-economic institutions shape young people’s perceptions of labour market opportunity structures and their employment attitudes (i.e. skills and retraining). building on the varieties of capitalism approach, we expect young people (aged 18–35) in coordinated market economies (cmes) with encompassing welfare states to regard firmand industry-specific skills as more important than their peers in liberal market economies (lmes). to assess this proposition, we draw on original survey data and compare young people’s employment attitudes in five european countries: the united kingdom (uk), which represents a typical liberal market economy, and austria, denmark, germany and switzerland as representatives of coordinated market economies. to what extent do different training regimes in cmes and lmes shape individual attitudes towards skill formation? the empirical analysis shows that young people’s attitudes with regard to the specificity of skills and the willingness to undertake retraining differ systematically between cme and lme countries and supports our argument that the specific socio-economic institutions matter. keywords. country-comparative survey; skill formation; skill specificity; varieties of capitalism; labour market. jel codes. j24; p16; p5; p51; m53 doi. https://doi.org/10.17979/ejge.2020.9.3.5966 1. introduction the varieties of capitalism (voc) literature as established by hall and soskice (2001) stresses that different training systems and skill formation structures are key distinguishing characteristics of discrete economic systems (see also estevez-abe et al., 2001; amable, 2003; busemeyer, 2009; busemeyer and trampusch, 2011, 2012). this literature differentiates between liberal market economies (lmes), such as the united states and the united kingdom (uk), and coordinated market economies (cmes), such as germany and austria. these two types of capitalism differ according to their training systems, which in turn condition a range of political choices (culpepper, 2007, 611). cmes tend to stress firm-based training, and the skills developed are at a much higher level of specialisation than the comparatively more general skills inculcated in lmes. in cmes, the existence of unemployment insurance and high levels of replacement rates mailto:felix.hoerisch@htwsaar.de https://doi.org/10.17979/ejge.2020.9.3.5966 felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 234 for unemployment make it more attractive for employees to invest in firmand industry-specific skills (rhodes, 2005; schröder, 2013, 70–74). hence this fundamental difference. voc is a firm-level approach, which at first glance does not seem to lend itself to the microlevel analysis of individuals’ attitudes and perceptions (but see, e.g., busemeyer and jensen, 2012). however, the institutional contexts in which individuals are embedded is central to both their perception of risks and opportunities and plays a key role in shaping their attitudes (e.g. vráblíková, 2014; walter, 2017). accordingly, we pose the following question: to what extent do different training regimes in cmes and lmes shape individual attitudes towards skill formation? educational decisions and skill formation are processes that mainly take place during adolescence and early adulthood. as estévez-abe et al. (2001, 145) argue, ‘young people are less likely to invest in specific skills if the risk of loss of employment opportunities that require those specific skills is high’. accordingly, to answer our research question, we draw on an original and bespoke survey dataset explicitly designed to capture attitudes and work values of young people (18–35 years) in europe gathered in the context of the large-scale collaborative cupesse project (tosun et al., 2019). we compare the attitudes of young people towards skill formation in the uk, an archetypical liberal market economy, with their counterparts in four coordinated market economies: austria, denmark, germany and switzerland. in the remainder of this article, we outline our main theoretical argument, methodological approach and research design. we then report and interpret our empirical results, which reveal that attitudes towards skill formation differ systematically between the uk and the four cmes – meaning they align with our theoretical argument. lastly, we offer some concluding remarks. 2. theoretical reasoning and hypotheses in its most basic form, the voc approach claims that the success of a political economy depends on firms and companies solving their coordination problems in various economic subsystems such as industrial relations or education and vocational training (hall and soskice 2001). following this main line of reasoning, lmes and cmes represent the extremes of a spectrum between pure market powers and the strong coordination of market forces. it is also worth noting that more recent contributions to the voc literature have identified mixed-market economies (molina and rhodes, 2007) and dependent market economies (nölke and vliegenthart, 2009). as their respective names suggest, in lmes strong market forces dominate, while in cmes coordination mechanisms between the various market participants play a major role. in lmes, companies coordinate their activities in all subsystems of the economy primarily by focusing on price signals, free competition in the market and marginal cost. in cmes, non-market based relations play a crucial role in the development of corporate core competencies. hall and soskice (2001) find that cmes produce goods that stem from incrementally innovative production, which requires especially industryand firm-specific human capital, more efficiently than lmes. one of the reasons for this is the nature of intercompany relations giving rise to joint research, product differentiation and niche production, rather than direct product competition. felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 235 another reason is based on the human capital argument. in cmes, employees are provided with sophisticated training schemes and apprenticeships that rely on sufficiently specialised knowledge to propose and implement changes in the production line (see, e.g., estevez-abe et al., 2001; busemeyer, 2009; trampusch, 2010; busemeyer and trampusch, 2012). this requires a ‘mix of company-specific and more general technical skills’ in the workforce (hall and soskice 2001, 36). furthermore, the likelihood of becoming unemployed is lower in cmes, because firms are more diversified, which makes investments in firm-specific skills more attractive. the differences between cmes and lmes has a direct effect on the policy-making process and ultimately on policy design (see, estevez-abe, 2005; bonoli and reber, 2010; jensen, 2011; hörisch and weber, 2014; marques and hörisch, 2019). several contributions to the political economy literature claim that the voc approach ‘brought the question of skill formation from the periphery to the centre of comparative welfare state research and political economy’ (busemeyer and trampusch, 2011, 424; dobbins and busemeyer, 2015). regarding the systems of skill creation, there are several reasons to assume that different institutional contexts in varying types of capitalism shape individuals’ attitudes towards skill formation. along these lines, we argue that the existing socioeconomic institutions shape the attitudes of young people in two ways. first, they alter the strategic behaviour of young people by heightening their (rational) perception of risk; and second, they encourage normative adaption to the existing institutions through processes of socialisation. the skill creation systems of lmes and cmes generate different types of skills. substantial parts of these differences stem from the rational adaptation of variations in welfare state arrangements by (young) adults. for example, estévez-abe argues that ‘welfare states … affect the decisions of workers and employers about the skills in which to invest’ (2005, 191; see also estévez-abe et al., 2001). in cmes, employers invest in the formation of industryand firmspecific skills1 by means of strong vocational training and education systems, while enjoying the support of generous welfare state policies that incentivise (young) employees to invest in those specific skills (estévez-abe et al., 2001; iversen and soskice, 2001; mares 2003; busemeyer and trampusch, 2011; schröder, 2013; busemeyer and schlicht-schmälzle, 2014). furthermore, the higher degree of diversification of firms between different sectors in cmes makes it more attractive for (young) adults to invest in firm-specific skills, as the risk of becoming unemployed is lower in case of sector-specific downturns. in contrast, the more competitive environment of lmes severely disincentivises skill investments by the firms. instead, skill formation primarily takes place in formalised education institutions as well as on-the-job training, resulting in general rather than firm-specific skills (busemeyer and trampusch, 2011; coulter, 2018). this institutional arrangement might go hand-in-hand with the skill formation preferences rationally developed by (young) adults. 1 several skill typologies can be found in the literature. for example, estévez-abe et al. (2001) distinguish between firmspecific skills, industry-specific skills and general skills. even though some strands of the literature criticise the differentiation between general and specific skills as too broad categories (streeck 2011), we rely on the difference between general and specific skills here, subsuming firm-specific and industry-specific skills into the latter category, because we are interested in general differences as well as broad patterns between varieties of capitalism and skill formation preferences. felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 236 besides these more rational adaptation procedures, socialisation processes and social norms also play a major role in the development of attitudes towards skill formation. political economies are socially constructed entities comprising both formalised structures (such as rules) and informal codes and expectations (see craig, 2015). the process that leads to the shared understanding of formal and informal constructions in political economies is socialisation (see mendoza, 2007). we argue that socialisation processes in political economies shape and structure individuals’ attitudes towards skill formation and will engender attitudinal differences in different varieties of capitalism, while controlling for an extensive battery of alternative explanations. this argument aligns with newer contributions to the literature on political economy that demonstrate how macro-level phenomena affect individual-level perceptions (e.g. walter, 2017). herein we argue that young people will have clear attitudinal positions on skill formation for two main reasons. first, they are either currently in education and training or have relatively recently made decisions concerning education and training programmes. as a result, issues around education, training and skill formation are at the forefront of their decision-making in this area. second, in many european countries, young people do not experience linear transitions into the labour market (côté, 2014). their experience is more likely to be elongated by delays, difficulties and periods of unemployment that make the issue of skill formation salient for them. given our theoretical proposition that the institutional context – i.e. growing up and living in an lme or a cme – is crucial to the pattern of attitudes and beliefs, we expect young people socialised in these different milieus to have varying views. hypothesis 1: young people in lmes will consider a general education to be more important than job-specific skills compared to their counterparts in cmes. firms in lmes do not require the same type of specialist skills as cmes and tend to oppose initiatives to introduce far-reaching insurance mechanisms. the consequence is that there are a less-specialist training and a more flexible workforce in lme production lines. this enables companies to produce their goods relatively cheaply and to innovate quickly through the firing of current and hiring of new employees. along this line, streeck (2011) argued that it is important to note that general skills do not go hand-in-hand with high skills, and specific skills do not automatically accompany low skills. from his viewpoint, the difference between general and specific skills relates to labour market mobility and the connectivity to the firm. the latter is usually substantially higher in cmes because of their strong employment protection, codetermination rules and collective bargaining agreements. this also leads workers to rationally develop more general skill sets that are transferable between sectors and companies to remain employable and competitive in the labour market. as lloyd and payne (2002) contend, in lmes such as the united kingdom, ‘… a viable high skills project in the uk is fundamentally incompatible with the model of british capitalism as currently configured’ (see also coulter, 2018). furthermore, finegold and soskice (1988) have shown that the failure of several attempts by british governments to establish felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 237 a dual apprenticeship training system can be explained by the fundamental differences in the socioeconomic institutional framework, especially when compared to cmes such as germany. accordingly, we include a second question in our analysis, assessing the extent to which young people would be willing to retrain to get a new job. we expect strategic behaviour as well as the socialisation processes in different types of capitalism to affect young people’s willingness to consider retraining. in particular, we argue that long-term employment and the focus on industryspecific skills that is typical of cmes provide young people with a stronger incentive to learn new skills that are also beneficial to their current occupation, while at the same time disincentivising them from retraining completely. in contrast, the fluid labour market and the focus on general skills we observe in lmes should make retraining more attractive: hypothesis 2: young people in lmes are more willing to retrain compared to young people in cmes. while the overall differences between lmes and cmes concerning skill formation are clear from a theoretical viewpoint, the ordering within the group of cmes is less distinct. within the voc literature there is an ongoing debate related to the hybridisation of cmes that sees several cmes moving closer to lmes. this process of hybridisation has been discussed in the german (e.g. lütz and eberle, 2008), danish (campbell and pedersen, 2007; dobbins and busemeyer, 2015), swiss (armingeon et al., 2004; börsch, 2007; afonso and mach, 2011) and austrian contexts (afonso and mach, 2011; schröder, 2013). table 1 provides an overview of the voc coordination index scores for the five countries included in our analyses based on four well-established voc coordination indices. the four indices vary across time (1998 to 2016), and while there are slight differences, the overall pattern is similar. according to all four indices, austria is the most coordinated economy, followed by germany and denmark and then switzerland. our selection of cmes is close to optimal, since all four selected cmes rely on collective skill formation systems (cf. busemeyer and trampusch, 2012) and there are no ambiguous cases – e.g. the swedish statist skill formation system. in contrast, the uk is the closest to the ideal-type lme with an exceptionally low degree of coordination. in our analyses, we will assess the extent to which these coordination differences are reflected in the skill formation attitudes of the young adults in our sample. table 1. varieties of capitalism coordination index scores of the five countries compared. index witt and jackson (2016) hall and gingerich (2009) hall and gingerich (2004) hicks and kenworthy (1998) labour relations corporate governance austria 0.82 1.00 1.00 1.00 0.96 denmark 0.51 0.58 0.65 0.70 0.72 germany 0.74 0.92 0.95 0.95 0.80 switzerland 0.52 0.48 0.44 0.51 0.55 uk 0.06 0.04 0.14 0.07 0.10 range 0 = no coordination 1 = fully coordinated 0 = no coordination 1 = fully coordinated 0 = no coordination 1 = fully coordinated 0 = no coordination 1 = fully coordinated felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 238 3. clarifications on data and methods in this study, we draw on a unique dataset that was produced in the cupesse project (tosun et al., 2019). the project was funded by the european commission within the seventh framework programme and ran from 2014 to early 2018. the survey instrument was designed to determine young adults’ attitudes and values towards work and education, their current financial and labour market situations and their expectations about the future. cupesse involved social scientists at academic institutions in eleven countries: austria, czech republic, denmark, germany, greece, hungary, italy, spain, switzerland, turkey and the uk. given the theoretical and empirical focus of this article, we only use the data gathered for respondents in austria, denmark, germany, switzerland and the united kingdom. ideally, we would have liked to have included at least one additional lme. however, since no lme-type country, other than the uk, was included in the cupesse survey, our comparative assessment contrasts the uk with the four other european cmes. within the group of cmes, austria, germany and switzerland represent conservative coordinated market economies, while denmark can be characterised as a social democratic coordinated one (schröder, 2013). the country surveys were conducted online with the assistance of leading commercial polling companies. consistent sampling frames were used and the survey companies provided probability samples of individuals between 18 and 35 years old, giving their employment status (e.g. employed; self-employed; unemployed; in education/training), nuts-2 region, age, education, and migration background/minority group membership. the dataset consists of 1,648 respondents in austria, 1,142 in denmark, 3,279 in germany, 1,002 in switzerland and 3,004 in the uk. the variation in the number of individuals surveyed was mostly due to population size, though practical issues related to survey costs and the different procedures followed by the respective polling firms also played a role. the data were collected throughout 2016. the cupesse survey contains two questions that are central to our analysis. the first one addresses respondents’ attitudinal positions on the importance of a good general education, as opposed to job-specific skills. the second and broader question asks respondents about their willingness to learn completely new skills or to retrain to secure a new and different job, as opposed to learning new skills that may help secure a job in their current industry. these questions investigate differences related to the skill sets that firms tend to support in lmes and cmes. in lmes, firms usually focus on general skills that are transferable between sectors and companies, i.e. a good general education, whereas in cmes the focus is on occupational or sector-specific skills. in addition, they address differences in the employment conditions of lmes and cmes: employment in lmes is more fluid, making retraining and general skills more attractive, while employment in cmes is more long-term, increasing the benefits of occupationspecific skills and of learning new skills that will also be beneficial to the current occupation. the exact wording of the question and the corresponding response categories are: felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 239 question 1: what skills and qualities do you think are important to finding a good job in [your country]? a) a good general education. b) occupation-specific or job-specific skills. response: 1 (very unimportant), 2 (rather unimportant), 3 (rather important), 4 (very important). question 2: what changes would you be willing to make to get a new job? a) i would be willing to learn new skills such as a new language, computer programmes. b) i would be willing to learn completely new skills or retrain to get a job. response: 1 (no), 2 (maybe), 3 (yes). table 2. descriptive statistics. uk (n=1,768) switzerland (n=585i) denmark (n=754) germany (n=2,508) austria (n=1,312) variable min max mean / share mean / share mean / share mean / share mean / share comparison of skill specificity both equally important to find a job (ties) 0 1 0.503 0.529 0.499 0.530 0.585 a good general education is more important 0 1 0.289 0.210 0.243 0.243 0.150 job-specific skills are more important 0 1 0.208 0.261 0.259 0.227 0.265 comparison of willingness to retrain vs. learning new skills both equally willing to do for a job/better job (ties) 0 1 0.724 0.748 0.613 0.670 0.687 more willing to learn new skills 0 1 0.191 0.179 0.363 0.257 0.275 more willing to retrain 0 1 0.085 0.073 0.024 0.073 0.038 age 18 35 27.2 27.8 28.1 27.6 26.0 employed 0 1 0.73 0.76 0.59 0.69 0.50 self-employed 0 1 0.05 0.03 0.02 0.03 0.04 unemployed 0 1 0.04 0.03 0.09 0.04 0.06 highest level of education a 1 7 4.9 5.6 4.7 4.7 4.6 sex b 0 1 0.55 0.41 0.60 0.52 0.46 political affiliation c 0 10 4.7 5.0 4.5 4.8 4.3 income d 1 10 4.7 5.1 4.0 5.2 4.6 held job e 0 1 0.86 0.89 0.89 0.83 0.73 unemployment experience f 0 1 0.34 0.19 0.32 0.24 0.26 migration background g 0 1 0.31 0.33 0.11 0.25 0.28 income dependency h 1 3 1.4 1.4 1.7 1.6 1.8 note. numbers only include respondents for which all control variables were available: a) 1: less than lower secondary; 2: lower secondary; 3: lower tier upper secondary; 4: upper tier upper secondary; 5: advanced vocational; 6: lower tertiary; 7: higher tertiary; b) 0: female; 1: male; c) measured on a left-right scale from 0 (left) to 10 (right); d) personal monthly total net income measured in income groups from 1 (lowest) to 10 (highest); e) dummy for whether the respondent has ever had a job for one year or more; f) dummy for whether the respondent has ever been unemployed for a period longer than six months; g) dummy for whether the respondent or their parents were born in a different country; h) 1: independent income; 2: partially dependent income; 3: dependent income; i) due to missings, the number of respondents in switzerland is only 582 for the comparison of skill specificity and 575 for the comparison of willingness to retrain. felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 240 we test our hypotheses through seven multinomial regression models: a base model with only the country dummies, five-country models with control variables and a full model with both the country dummies and control variables. control variables include the age and sex of the respondent, their employment status, level of educational attainment, income, political affiliation, previous employment and unemployment experiences, migration background, and income dependency (see, e.g., baartman and de bruijn, 2011; chen, 2011). the summary statistics of the dependent and independent variables are shown in table 2. 4. empirical findings we begin the discussion of the empirical findings with the question on skill specificity. as we see in table 3, 95 per cent of our respondents consider a good general education to be either rather or very important – the same goes for job-specific skills. the response patterns show that there is indicative support for hypothesis 1, because the percentage of respondents in cmes who consider job-specific skills as important is higher than in the uk (96.2 per cent vs. 94.4 per cent; the difference is significant at the 1 per cent level). at the same time, more respondents in the uk consider a good general education to be important (96.0 per cent vs 95.3 per cent; however, the difference is not significant). table 3. responses to the importance of a good general education as opposed to job-specific skills. q13a: what skills and qualities do you think are important to finding a good job in [your country]? – a good general education. categories united kingdom switzerland denmark germany austria frequency percent frequency percent frequency percent frequency percent frequency percent very unimportant 8 0.5 2 0.3 6 0.8 14 0.6 13 1.0 rather unimportant 60 3.4 15 2.6 51 6.8 113 4.5 74 5.6 rather important 693 39.2 251 43.1 368 48.8 796 31.7 430 32.8 very important 1,007 57.0 314 54.0 329 43.6 1,585 63.2 795 60.6 total 1,768 100.0 582 100.0 754 100.0 2,508 100.0 1,312 100.0 categories lme (uk) cmes combined frequency percent frequency percent very/rather unimportant 68 3.9 288 5.6 very/rather important 1,700 96.2 4,868 94.4 total 1,768 100.0 5,156 100.0 note. numbers only include respondents included in the regression models (i.e. without any missings in any of the variables). a chi-squared test indicates a significant difference between the lme and the cmes at the 1%-level. felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 241 q13b: what skills and qualities do you think are important to finding a good job in [your country]? – occupation-specific or job-specific skills. categories united kingdom switzerland denmark germany austria frequency percent frequency percent frequency percent frequency percent frequency percent very unimportant 8 0.5 3 0.5 0 0.0 5 0.2 3 0.2 rather unimportant 75 4.2 18 3.1 50 6.6 86 3.4 40 3.1 rather important 813 46.0 214 36.8 366 48.5 879 35.1 334 25.5 very important 872 49.3 347 59.6 338 44.8 1,538 61.3 935 71.3 total 1,768 100.0 582 100.0 754 100.0 2,508 100.0 1,312 100.0 categories lme (uk) cmes combined frequency percent frequency percent very/rather unimportant 83 4.7 205 4.0 very/rather important 1,685 95.3 4,951 96.0 total 1,768 100.0 5,156 100.0 note. numbers only include respondents included in the regression models (i.e. without any missings in any of the variables). a chi-squared test indicates no significant differences between the lme and the cmes. since the response patterns vary across these two items and are unevenly distributed, we directly compare the responses to the two questions. we do this by examining whether the respondents consider a good general education and job-specific skills to be equally important to finding a good job or whether they prioritise one over the other. this approach aligns with the empirical strategy adopted by busemeyer et al. (2017) in their study on attitudes towards education policy. the authors argue that surveys on preferences regarding education policies and skills that do not force participants to prioritise the relative importance of the individual education goals and skills are likely to blur general attitudes and specific preferences. when not forcing respondents to choose, surveys are likely to capture ‘cheap talk’ rather than an expression of preferences (busemeyer et al., 2017, 38). from this perspective, the survey approach we followed in the cupesse project corresponds to the latest developments in the relevant literature on survey methodology. our dependent variable includes the following possible response categories: ‘a good general education is more important than job-specific skills to find a job’, ‘job-specific skills are more important than a good general education to find a job’, and ‘both a good general education and job-specific skills are equally important to find a job (ties)’. of the respondents who rated one statement higher than the other, a majority considers job-specific skills to be more important than a good general education in switzerland, denmark and austria. although a majority considers a good general education more important in both the uk and germany, the margin is substantially higher in the uk. the uk has the highest share of respondents judging a good education to be more important (28.9 per cent vs 15.0 per cent – 24.3 per cent) and the lowest share of respondents judging job-specific skills to be more important (20.8 per cent vs 22.7 per cent – 26.5 per cent). the overall pattern thus supports hypothesis 1. felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 242 table 4. multinomial logistic models for the comparison of skill specificity. note. odds ratios, robust standard errors in parenthesis, significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001. given the measurement level of the dependent variable, ordered logit regression would initially appear the most appropriate method. however, since the assumption of parallel slopes, on which ordered logit models are based, does not hold, this method is not suitable for the analyses we conduct. therefore, we analyse the data via multinomial logistic models. the conclusions we arrive at, however, would be the same for ordered logit regression models. table 4 presents the results of our multinomial logistic regression models. the coefficients are displayed as odd-ratios with robust standard errors. table 4 shows the coefficients for comparing the base outcome (job-specific skills are more important than a good general education) with the two other outcomes (a good general education is more important than job-specific skills, and both are equally important). our main interest is in the comparison between the perception of job-specific skills or a good general education being more important. consequently, we omitted the comparison with both being equally important. the full table can be seen in table 4 in the appendix. both the baseline model and the full model show that the differences between the uk and the cmes are significant at a 1 per cent or a 0.1 per cent significance level. in substantive terms, the predicted probabilities of considering a good general education more important are 29.0 per cent in the uk, but they range from 14.8 per cent (austria) to 24.8 per cent (denmark) in the four cmes. in contrast, the predicted probabilities of considering job-specific skills to be more important are only 20.3 per cent in the uk, but range from 23.2 per baseline model uk sui den ger aut full model base outcome: job-specific skills more important a good general education more important sui 0.58*** (0.08) 0.54*** (0.08) den 0.67** (0.08) 0.69** (0.09) ger 0.77** (0.07) 0.72*** (0.07) aut 0.41*** (0.05) 0.40*** (0.05) age 0.99 (0.02) 0.98 (0.03) 1.03 (0.03) 1.00 (0.01) 0.99 (0.02) 1.00 (0.01) employed 1.26 (0.33) 1.07 (0.43) 1.25 (0.45) 0.83 (0.15) 0.82 (0.21) 0.99 (0.11) selfemployed 1.02 (0.39) 1.25 (0.96) 0.00*** (0.00) 1.27 (0.48) 0.31* (0.17) 0.77 (0.16) unemployed 0.69 (0.28) 1.30 (0.95) 1.15 (0.52) 0.62 (0.22) 1.19 (0.50) 0.84 (0.16) education level 1.02 (0.05) 1.04 (0.09) 0.76*** (0.05) 0.96 (0.04) 1.05 (0.07) 0.97 (0.02) sex 0.78 (0.11) 0.82 (0.21) 0.81 (0.19) 0.73** (0.09) 0.73 (0.14) 0.76*** (0.06) political affiliation 1.13*** (0.03) 1.01 (0.05) 1.03 (0.04) 1.05 (0.03) 1.07 (0.05) 1.06*** (0.02) income 1.00 (0.04) 1.01 (0.05) 1.03 (0.05) 1.02 (0.02) 1.10** (0.04) 1.03 (0.01) held job 0.58* (0.14) 1.30 (0.54) 0.77 (0.29) 1.04 (0.19) 1.72* (0.44) 0.98 (0.11) unemployme nt experience 0.87 (0.14) 1.01 (0.31) 0.81 (0.19) 0.97 (0.15) 0.82 (0.18) 0.90 (0.08) migration background 0.90 (0.14) 0.72 (0.20) 0.77 (0.28) 0.97 (0.14) 1.20 (0.25) 0.97 (0.08) partially dep. income 1.08 (0.20) 1.09 (0.32) 0.75 (0.22) 0.97 (0.13) 1.03 (0.24) 1.00 (0.09) dependent income 1.41 (0.43) 1.43 (0.82) 0.67 (0.28) 1.05 (0.25) 0.85 (0.28) 1.02 (0.14) both equally important (ties) [omitted, see table 4 in the appendix] n 6924 1768 582 754 2508 1312 6924 log pseudolikelihood -6994.12 -1794.97 -571.96 -750.81 -2510.04 -1219.54 -6924.94 felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 243 cent (germany) to 27.5 per cent (switzerland) for the cmes.2 in line with our theoretical argument, the distance between the uk and austria is the largest, showing that the country with the highest degree of coordination (austria) is also the country in which respondents more strongly favour job-specific skills over a good general education (cf. afonso and mach, 2011). when turning to the control variables, we observe that significantly more men in the full model than women consider job-specific skills to be more important, while people with right-leaning ideological views consider a good general education to be more important than job-specific skills. table 5. responses to the willingness to retrain vs. learning new skills. q12c: what changes would you be willing to make to get a new job? – i would be willing to learn new skills such as a new language, computer programmes. categories united kingdom switzerland denmark germany austria frequency percent frequency percent frequency percent frequency percent frequency percent no 63 3.6 8 1.4 9 1.2 68 2.7 24 1.8 maybe 372 21.0 69 12.0 75 10.0 502 20.0 186 14.2 yes 1,333 75.4 498 86.6 670 88.9 1,938 77.3 1,102 84.0 total 1,768 100.0 575 100.0 754 100.0 2,508 100.0 1,312 100.0 categories lme (uk) cmes combined frequency percent frequency percent no/maybe 435 24.6 941 18.3 yes 1,333 75.4 4,208 81.7 total 1,768 100.0 5,149 100.0 note. numbers only include respondents included in the regression models (i.e. without any missings in any of the variables). a chi-squared test indicates a significant difference between the lme and the cmes at the 0.1%-level q12d: what changes would you be willing to make to get a new job? – i would be willing to learn completely new skills or retrain to get a job. categories united kingdom switzerland denmark germany austria frequenc y percent frequenc y percent frequenc y percent frequenc y percent frequenc y percent no 104 5.9 11 1.9 68 9.0 183 7.3 99 7.6 maybe 505 28.6 128 22.3 255 33.8 802 32.0 394 30.0 yes 1,159 65.6 436 75.8 421 57.2 1,523 60.7 819 62.4 total 1,768 100.0 575 100.0 754 100.0 2,508 100.0 1,312 100.0 categories lme (uk) cmes combined frequency percent frequency percent no/maybe 609 34.5 1,940 37.7 yes 1,159 65.6 3,209 62.3 total 1,768 100.0 5,149 100.0 note. numbers only include respondents included in the regression models (i.e. without any missings in any of the variables). 2 average predicted probabilities across observations were calculated for the different countries. all other variables were treated as observed. felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 244 turning to our second dependent variable, the willingness to retrain, we observe that less than 4 per cent of respondents indicate that they are unwilling to learn new skills in all of the countries in our sample (see table 5). the percentage of respondents who are unwilling to retrain varies between the countries but is consistently below 10 per cent. looking at the data, there is again some indication that the theoretical argument holds. compared to the cmes, the willingness to learn new skills is lower in the uk (75.4 per cent vs. 81.7 per cent; the difference is significant at the 0.1 per cent-level), whereas the willingness to retrain is higher (65.6 per cent vs. 62.3 per cent; the difference is significant at the 5 per cent-level). in contrast to our expectation, however, the willingness to retrain is actually higher in switzerland (75.8 per cent) than in the uk (65.6 per cent). a chi-squared test indicates a significant difference between the lme and the cmes at the 5%-level. our data present us with an uneven distribution of responses, so we directly compare the responses of each individual as we did with skill specificity. our response categories are as follows: ‘more willing to learn new skills than to retrain to get a new job’, ‘more willing to retrain than to learn new skills to get a new job’, and ‘equally willing to both learn new skills and retrain (tied)’. the frequency of the tied response is markedly higher than it was for skill specificity. furthermore, there is also greater variance in the frequency of the tied response. while only 61.3 per cent of danish respondents gave a tied answer, 74.8 per cent of the swiss respondents are equally willing to learn new skills and retrain. in all countries, respondents are on average more willing to learn new skills than to retrain. nevertheless, we can still observe patterns as predicted by the voc approach. in the uk, the percentage of respondents more willing to retrain is higher than in the four cmes (8.5 per cent vs 2.4 per cent – 7.3 per cent). as for the willingness to learn new skills, respondents in the uk are less willing to learn new skills than respondents in three of the cmes (germany, austria and denmark; 19.1 per cent vs 25.7 per cent – 36.3 per cent). the results for switzerland, however, contradict expectations, as the percentage of swiss respondents more willing to learn new skills is below that of the uk (17.9 per cent). as with skill specificity, we conduct multinomial logistic regression models to test whether the difference is significant and whether it holds when adding control variables. table 6 presents the results. respondents who are more willing to learn new skills are compared to respondents who are more willing to retrain, as well as to respondents who are equally willing to do both. as before, we omitted those equally willing to do both because the former comparison is key to our theoretical implications. the proportion of respondents who are more willing to retrain is significantly higher in the uk than in denmark, germany and austria, though not in switzerland. the predicted probability of being more willing to retrain is significantly higher in the uk (8.3 per cent) than in denmark and austria (2.2 per cent and 4.1 per cent), but only marginally and insignificantly lower in switzerland (8.0 per cent) and germany (7.1 per cent). at the same time, the predicted probability of being more willing to learn new skills is lower in switzerland than in the uk (17.8 per cent vs. 19.3 per cent), but it is significantly higher in austria, germany and denmark (25.8 per cent – 36.2 per cent). as already indicated by our descriptive results, the felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 245 regression models confirm our expectations for only three of the four cmes. with respect to the control variables, respondents who are older, have lower educational attainment levels, are in employment or are ideologically right-leaning are more willing to retrain. following the hybridisation hypothesis, it is plausible that the country with the strongest degree of hybridisation, namely switzerland, shows the weakest effects. although it is typically classified as a cme, switzerland is often seen as incorporating a mixture of elements from both liberal and coordinated market economies. as a less regulated labour market than other cmes (schröder, 2013, 124–126), switzerland incentivises retraining and employment attitudes more closely follow the pattern of lmes. rather surprisingly, however, the share of respondents who are willing to retrain is actually lowest in denmark, a country known for its flexicurity approach (emmenegger, 2010). despite its flexible labour market with low levels of job security regulations and pronounced active labour market policies, these institutions do not translate to corresponding individual attitudes. table 6. multinomial logistic models for the comparison of willingness to retrain vs. willingness to learn new skills. note. odds ratios, robust standard errors in parenthesis, significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001. baseline model uk sui den ger aut full model base outcome: job-specific skills more important sui 0.91 (0.19) 1.05 (0.23) den 0.15*** (0.04) 0.13*** (0.04) ger 0.63*** (0.08) 0.62*** (0.08) aut 0.31*** (0.06) 0.36*** (0.07) age 1.02 (0.02) 1.11* (0.05) 1.18* (0.08) 1.01 (0.02) 1.11* (0.05) 1.04*** (0.01) employed 2.03 (0.81) 1.72 (1.12) 1.77 (1.45) 2.20* (0.69) 1.89 (1.02) 1.97*** (0.39) self-employed 2.44 (1.34) 2.31 (3.74) 0.00*** (0.00) 1.22 (0.83) 1.52 (1.36) 1.64 (0.56) unemployed 2.78 (1.61) 1.33 (1.22) 2.15 (2.01) 0.80 (0.45) 1.02 (0.72) 1.25 (0.35) education level 0.77*** (0.05) 0.74* (0.09) 0.50*** (0.08) 0.74*** (0.04) 0.87 (0.10) 0.75*** (0.03) sex 0.86 (0.18) 0.83 (0.33) 0.95 (0.53) 1.12 (0.20) 0.76 (0.25) 0.96 (0.11) political affiliation 1.14** (0.05) 1.06 (0.09) 1.02 (0.10) 1.17*** (0.05) 1.26** (0.09) 1.13*** (0.03) income 0.93 (0.05) 0.92 (0.08) 1.09 (0.13) 0.99 (0.03) 0.85** (0.05) 0.96 (0.02) held job 2.31* (0.86) 0.61 (0.39) 2.36 (2.82) 1.61 (0.47) 0.65 (0.30) 1.42 (0.27) unemployme nt experience 1.09 (0.25) 3.47* (1.70) 0.24 (0.18) 1.27 (0.28) 0.84 (0.31) 1.18 (0.16) migration background 0.82 (0.18) 2.20 (0.97) 2.74 (1.78) 1.12 (0.23) 1.16 (0.39) 1.13 (0.14) partially dep. income 1.03 (0.28) 0.89 (0.41) 1.24 (1.00) 0.91 (0.19) 0.85 (0.35) 0.97 (0.13) dependent income 1.25 (0.55) 0.32 (0.33) 3.72 (3.54) 1.60 (0.62) 1.45 (1.01) 1.31 (0.32) both equally important (ties) [omitted, see appendix, table 6] n 6,917 1,768 575 754 2,508 1,312 6,917 log pseudolikelihood -5322.31 -1297.81 -386.53 -530.81 -1962.03 -932.68 -5178.29 felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 246 5. conclusions voc theory contends that lmes and cmes differ by the degree of coordination between firms and other actors. the different coordination mechanisms produce distinct advantages in innovation and affect policymaking. in this article we have sought to advance the voc literature by illustrating how it can be combined with the microlevel, thereby offering insights into how individuals living and working in the different political economies internalise their structural characteristics and adjust their attitudes and arguably their behavioural patterns accordingly. we argued that the institutional differences in the structure of the labour market are reflected in young people’s attitudes towards skill formation. using the unique cupesse survey data, we compared young people’s attitudinal positions on skills and retraining in austria, denmark, germany and switzerland, and the uk. we found statistically significant differences between the respondents in the individual countries for both types of employment attitudes. respondents in the uk considered general skills as opposed to job-specific skills to be more important than respondents in austria, denmark, germany and switzerland. young british citizens are also more willing to retrain in order to get a new job rather than simply learn new skills when compared to all of the cmes except for switzerland. these findings are in line with our expectations of how the respective skill creation systems in political economies shape the employment attitudes of the young. our findings are robust and are reinforced by the inclusion of various control variables. finally, drawing on hybridisation arguments within the voc literature, we were also able to make some sense of the ordering of the countries along the cme-lme spectrum. nevertheless, some of the results for the within-cme variation are more mixed. further research could thus investigate in greater depth the interplay between hybridisation tendencies and the development of attitudes towards skill formation. the findings reported in this study are important for increasing the academic impact of political economy research, as they connect it to subdisciplines in political science (and disciplines elsewhere – e.g. sociology and business and management) that are primarily interested in the attitudes and behaviour of individuals. despite the richness of the voc literature, we argue that it can benefit from incorporating microlevel factors. including the microlevel in analyses offers new insights, for example, into the impact of austerity measures implemented in different types of political economies on individuals’ perceptions of the problem-solving capacity of the state. the same is true of individuals’ trust in state institutions. despite the importance of demand-side factors for the labour market participation of individuals, education and career choices are certainly the most important determinants on the supply side and therefore, it is worth paying greater attention to them. evidence from southern europe demonstrates that making poorly informed choices regarding education and training can lead to a mismatch between the qualifications of jobseekers (supply-side) and the qualifications employers seek (demand-side) (e.g. pompei and selezneva, 2017). in this respect, our study presents positive news: it shows that the attitudes of employees towards education, training and retraining generally align well with the institutional context they live in. lmes and cmes tend to felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 247 reproduce themselves on the microlevel: institutional configurations that rely more heavily on general skills and a flexible workforce tend to produce employees that are more willing to learn general skills and undergo complete retraining – and vice versa. a good match between the institutional context and the attitudes of young people should improve their chances of finding a suitable job in the labour market. we fully recognise that our research is open to challenge. most importantly, it rests on the assumption that the differences we find in the employment attitudes of the younger generation in the five countries can be traced back to the different skill creation systems of the two varieties of capitalism. an alternative perspective might argue that the differences are not inherent to the varieties of capitalism approach, and there may be other explanatory factors at play. while we accept that other factors will have an influence on the outcomes, the response patterns within the countries we examined are consistent enough to suggest that they are the outcome of established differences between these countries (see iversen and stephens, 2008). from a theoretical viewpoint, it is plausible and credible to argue that the different skill creation regimes engender different employment attitudes towards the process of skill formation. our empirical findings support this. further research could build on the results presented above by testing whether they also hold for the older workforce, in a larger set of countries with additional lmes, and for the other subsystems of the economy defined by varieties of capitalism. moving forward, an additional avenue for future research would be to take into account the more recently identified types of mixed-market economies (molina and rhodes, 2007) and dependent market economies (nölke and vliegenthart, 2009). while our objective was to empirically test the classic differentiation between lmes and cmes, we believe it would be worth replicating our analysis using these more refined categorisations of vocs. references afonso, a., and mach, a. (2011). coming together but staying apart: continuity and change in the austrian and swiss varieties of capitalism. in becker, u. (ed.), the changing political economies of small west european countries, (pp. 99 – 124). amsterdam, netherlands: amsterdam university press. https://doi.org/10.1515/9789048514540-005 amable, b. (2003). the diversity of modern capitalism. oxford, united kingdom: oxford university press. https://doi.org/10.1093/019926113x.001.0001 armingeon, k., bertozzi, f. and bonoli, g. (2004). swiss worlds of welfare. west european politics, 27(1), 20–44. https://doi.org/10.1080/01402380412331280793 baartman, l.k.j., and de bruijn, e. (2011). integrating knowledge, skills and attitudes: conceptualising learning processes towards vocational competence. educational research review, 6(2), 125–134. https://doi.org/10.1016/j.edurev.2011.03.001 bonoli, giuliano, and frank reber. (2010). the political economy of childcare in oecd countries: explaining cross‐national variation in spending and coverage rates. european journal of political research, 49(1), 97–118. https://doi.org/10.1111/j.1475-6765.2009.01884.x https://doi.org/10.1515/9789048514540-005 https://doi.org/10.1093/019926113x.001.0001 https://doi.org/10.1016/j.edurev.2011.03.001 https://doi.org/10.1111/j.1475-6765.2009.01884.x felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 248 börsch, a. (2007). institutional variation and coordination patterns in cmes: swiss and german corporate governance in comparison. in hancké, b. (ed.), beyond varieties of capitalism: conflict, contradictions, and complementarities in the european economy (pp. 173–194). oxford, united kingdom: oxford university press. https://doi.org/10.1093/acprof:oso /9780199206483.003.0006 busemeyer, m.r. (2009). asset specificity, institutional complementarities and the variety of skill regimes in coordinated market economies. socio-economic review, 7(3), 375–406. https://doi.org/10.1093/ser/mwp009 busemeyer, m.r., and jensen, c. (2012). the impact of economic coordination and educational institutions on individual-level preferences for academic and vocational education. socioeconomic review, 10(3), 525–547. https://doi.org/10.1093/ser/mws002 busemeyer, m.r., and schlicht-schmälzle, r. (2014). partisan power, economic coordination and variations in vocational training systems in europe. european journal of industrial relations, 20(1), 55–71. https://doi.org/10.1177/0959680113512731 busemeyer, m.r., and trampusch, c. (2011). review article: comparative political science and the study of education. british journal of political science, 41(2), 414–443. https://doi.org/10.1017/s0007123410000517 busemeyer, m.r., and trampusch, c. (2012). the political economy of collective skill formation. oxford, united kingdom: oxford university press. https://doi.org/10.1093/acprof:oso /9780199599431.001.0001 busemeyer, m.r., garritzmann, j.l., neimanns, e., and nezi, r. (2017). investing in education in europe: evidence from a new survey of public opinion. journal of european social policy, 28(1), 34–54. https://doi.org/10.1177/0958928717700562 campbell, j.l., and pedersen, o.k. (2007). the varieties of capitalism and hybrid success: denmark in the global economy. comparative political studies, 40(3), 307–332. https://doi.org/10.1177/0010414006286542 chen, y.-w. (2011). once a neet always a neet? experiences of employment and unemployment among youth in a job training programme in taiwan. international journal of social welfare, 20(1), 33–42. https://doi.org/10.1111/j.1468-2397.2009.00711.x côté, j.e. (2014). towards a new political economy of youth. journal of youth studies 17(4), 527– 543. https://doi.org/10.1080/13676261.2013.836592 coulter, s. (2018). skill formation, immigration and european integration: the politics of the uk growth model, new political economy, 23(2), 208–222. https://doi.org/10.1080/ 13563467.2017.1370446 craig, m. (2015). post-2008 british industrial policy and constructivist political economy: new directions and new tensions. new political economy, 20(1), 107–125. https://doi.org/10.1080/13563467.2014.908176 culpepper, p.d. (2007). small states and skill specificity: austria, switzerland, and interemployer cleavages in coordinated capitalism. comparative political studies, 40(6), 611–637. https://doi.org/10.1177/0010414006295927 dobbins, m., and busemeyer, m.r. (2014). socio-economic institutions, organized interests and partisan politics: the development of vocational education in denmark and sweden. socioeconomic review, 13(2), 259–284. https://doi.org/10.1093/ser/mwu002 emmenegger, p. (2010). the long road to flexicurity: the development of job security regulations in denmark and sweden. scandinavian political studies, 33(3), 271–294. https://doi.org/10.1111/j.1467-9477.2010.00251.x estevez-abe, m. (2005). gender bias in skills and social policies: the varieties of capitalism perspective on sex segregation. social politics: international studies in gender, state & society, 12(2), 180–215. https://doi.org/10.1093/sp/jxi011 estevez-abe, m., iversen, t., and soskice, d. 2001. social protection and the formation of skills: a reinterpretation of the welfare state. in hall, p.a., soskice, d. (ed.), varieties of capitalism: the institutional foundations of comparative advantage, (pp. 145 – 183). new york, ny: oxford university press. https://doi.org/10.1093/0199247757.003.0004 finegold, d., soskice, d. (1998). the failure of training in britain: analysis and prescription. oxford review of economic policy, 4(3), 21–53. https://doi.org/10.1093/oxrep/4.3.21 https://doi.org/10.1093/acprof:oso/9780199206483.003.0006 https://doi.org/10.1093/acprof:oso/9780199206483.003.0006 https://doi.org/10.1093/ser/mwp009 https://doi.org/10.1093/ser/mws002 https://doi.org/10.1177/0959680113512731 https://doi.org/10.1017/s0007123410000517 https://doi.org/10.1093/acprof:oso/9780199599431.001.0001 https://doi.org/10.1093/acprof:oso/9780199599431.001.0001 https://doi.org/10.1177/0958928717700562 https://doi.org/10.1177/0010414006286542 https://doi.org/10.1111/j.1468-2397.2009.00711.x https://doi.org/10.1080/13676261.2013.836592 https://doi.org/10.1080/13563467.2017.1370446 https://doi.org/10.1080/13563467.2017.1370446 https://doi.org/10.1080/13563467.2014.908176 https://doi.org/10.1177/0010414006295927 https://doi.org/10.1093/ser/mwu002 https://doi.org/10.1111/j.1467-9477.2010.00251.x https://doi.org/10.1093/sp/jxi011 https://doi.org/10.1093/0199247757.003.0004 https://doi.org/10.1093/oxrep/4.3.21 felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 249 hall, p.a., and soskice, d. (2001). varieties of capitalism: the institutional foundations of comparative advantage. new york, ny: oxford university press. https://doi.org/10.1093/ 0199247757.001.0001 hall, p. a., and gingerich, d.w. (2004). varieties of capitalism and institutional complementarities in the macroeconomy. max-planck-institut für gesellschaftsforschung (mpifg) discussion paper no. 04/5. hall, p.a., and gingerich, d.w. (2009). varieties of capitalism and institutional complementarities in the political economy: an empirical analysis. british journal of political science, 39(3), 449482. https://doi.org/10.1017/s0007123409000672 hicks, a., and kenworthy, l. (1998). cooperation and political economic performance in affluent democratic capitalism. american journal of sociology, 103(6), 631–672. https://doi.org/10.1086/231403 hörisch, f., and weber, j. (2014). capitalizing the crisis? explanatory factors for the design of short-time work across organisation for economic co-operation and development countries. social policy & administration, 48(7), 799-825. https://doi.org/10.1111/spol.12047 iversen, t., and soskice, d. (2001). an asset theory of social policy preferences. american political science review 95(4), 875–893. https://doi.org/10.1017/s0003055400400079 iversen, t., and stephens, j.d. 2008. partisan politics, the welfare state, and three worlds of human capital formation. comparative political studies, 41(4-5), 600–637. https://doi.org/10.1177/0010414007313117 jensen, c. (2011). conditional contraction: globalisation and capitalist systems. european journal of political research, 50(2), 168–189. https://doi.org/10.1111/j.1475-6765 .2010.01925.x lloyd, c., and payne, j. (2002). on the 'political economy of skill': assessing the possibilities for a viable high skills project in the united kingdom. new political economy, 7(3), 367–395. https://doi.org/10.1080/1356346022000018739 lütz, s., and eberle, d. (2008). varieties of change in german capitalism: transforming the rules of corporate control. new political economy, 13(4), 377–395. https://doi.org/10.1080/13563460802436541 mares, i. (2003). the politics of social risk: business and welfare state development. cambridge, united kingdom: cambridge university press. marques, p. and hörisch, f. (2019). promoting workplace-based training to fight youth unemployment in three eu countries: different strategies, different results? international journal of social welfare, 28(4), 380-393. https://doi.org/10.1111/ijsw.12381 mendoza, p. (2007). academic capitalism and doctoral student socialization: a case study. the journal of higher education, 78(1), 71–96. https://doi.org/10.1353/jhe.2007.0004 molina, o., and rhodes, m. (2007). the political economy of adjustment in mixed market economies: a study of spain and italy. in hancké, b., rhodes, m., thatcher, m. (ed.) beyond varieties of capitalism: conflict, contradictions and complementarities in the european economy (pp. 223 – 252). oxford, united kingdom: oxford university press. https://doi.org/10.1093/acprof:oso/9780199206483.003.0008 nölke, a., and vliegenthart, a. (2009). enlarging the varieties of capitalism: the emergence of dependent market economies in east central europe. world politics, 61(4), 670–702. https://doi.org/10.1017/s0043887109990098 pompei, f., and selezneva, e. (2017). education mismatch and youth labour force status in europe. in: caroleo, f.e., demidova, o., marelli, e., signorelli, m. (ed.) young people and the labour market: a comparative perspective, (pp. 41 – 74). london, united kingdom: routledge. https://doi.org/10.4324/9781315178424-3 rhodes, m. (2005). 'varieties of capitalism' and the political economy of european welfare states. new political economy, 10(3), 363–370. https://doi.org/10.1080/13563460500204266 schröder, m. (2013). integrating varieties of capitalism and welfare state research: a unified typology of capitalisms. new york, ny: palgrave macmillan. https://doi.org/10.1057/ 9781137310309 streeck, w. (2011). skills and politics. general and specific. mpifg discussion paper 2011/1. https://doi.org/10.2139/ssrn.1781042 https://doi.org/10.1093/0199247757.001.0001 https://doi.org/10.1093/0199247757.001.0001 https://doi.org/10.1017/s0007123409000672 https://doi.org/10.1086/231403 https://doi.org/10.1111/spol.12047 https://doi.org/10.1017/s0003055400400079 https://doi.org/10.1177/0010414007313117 https://doi.org/10.1111/j.1475-6765.2010.01925.x https://doi.org/10.1111/j.1475-6765.2010.01925.x https://doi.org/10.1080/1356346022000018739 https://doi.org/10.1080/13563460802436541 https://doi.org/10.1111/ijsw.12381 https://doi.org/10.1353/jhe.2007.0004 https://doi.org/10.1093/acprof:oso/9780199206483.003.0008 https://doi.org/10.1017/s0043887109990098 https://doi.org/10.4324/9781315178424-3 https://doi.org/10.1080/13563460500204266 https://doi.org/10.1057/9781137310309 https://doi.org/10.1057/9781137310309 https://doi.org/10.2139/ssrn.1781042 felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 250 tosun, j., arco-tirado, j., caserta, m., cemalcilar, z., freitag, m., hörisch, f., jensen, c., kittel, b., littvay, l., lukes, m., maloney, w., mühlböck, m., rainsford, e., rapp, c., schuck, b., shore, j., steiber, n., sümer, n., tsakloglou, p., vancea, m., and vegetti, f. (2019). perceived economic self-sufficiency: a countryand generation-comparative approach. european political science, 18(3), 510-531. https://doi.org/10.1057/s41304-018-0186-3 trampusch, c. (2010). co-evolution of skills and welfare in coordinated market economies? a comparative historical analysis of denmark, the netherlands and switzerland. european journal of industrial relations, 16(3), 197–220. https://doi.org/10.1177/0959680110375129 vráblíková, k. (2014). how context matters? mobilization, political opportunity structures, and nonelectoral political participation in old and new democracies. comparative political studies, 47(2), 203–229. https://doi.org/10.1177/0010414013488538 walter, s. (2017). globalization and the demand-side of politics: how globalization shapes labor market risk perceptions and policy preferences. political science research and methods, 5(1), 55–80. https://doi.org/10.1017/psrm.2015.64 witt, m.a., and jackson, g. (2016). varieties of capitalism and institutional comparative advantage: a test and reinterpretation. journal of international business studies, 47, 778– 806. https://doi.org/10.1057/s41267-016-0001-8 https://doi.org/10.1057/s41304-018-0186-3 https://doi.org/10.1177/0959680110375129 https://doi.org/10.1177/0010414013488538 https://doi.org/10.1017/psrm.2015.64 https://doi.org/10.1057/s41267-016-0001-8 felix hörisch et al. / european journal of government and economics 9(3), december 2020, 232-251 251 appendix table 4. multinomial logistic models for the comparison of skill specificity. baseline model uk sui den ger aut full model base outcome: job-specific skills more important a good general education more important sui 0.58*** (0.08) 0.54*** (0.08) den 0.67** (0.08) 0.69** (0.09) ger 0.77** (0.07) 0.72*** (0.07) aut 0.41*** (0.05) 0.40*** (0.05) age 0.99 (0.02) 0.98 (0.03) 1.03 (0.03) 1.00 (0.01) 0.99 (0.02) 1.00 (0.01) employed 1.26 (0.33) 1.07 (0.43) 1.25 (0.45) 0.83 (0.15) 0.82 (0.21) 0.99 (0.11) self-employed 1.02 (0.39) 1.25 (0.96) 0.00*** (0.00) 1.27 (0.48) 0.31* (0.17) 0.77 (0.16) unemployed 0.69 (0.28) 1.30 (0.95) 1.15 (0.52) 0.62 (0.22) 1.19 (0.50) 0.84 (0.16) education level 1.02 (0.05) 1.04 (0.09) 0.76*** (0.05) 0.96 (0.04) 1.05 (0.07) 0.97 (0.02) sex 0.78 (0.11) 0.82 (0.21) 0.81 (0.19) 0.73** (0.09) 0.73 (0.14) 0.76*** (0.06) political affiliation 1.13*** (0.03) 1.01 (0.05) 1.03 (0.04) 1.05 (0.03) 1.07 (0.05) 1.06*** (0.02) income 1.00 (0.04) 1.01 (0.05) 1.03 (0.05) 1.02 (0.02) 1.10** (0.04) 1.03 (0.01) held job 0.58* (0.14) 1.30 (0.54) 0.77 (0.29) 1.04 (0.19) 1.72* (0.44) 0.98 (0.11) unemployment experience 0.87 (0.14) 1.01 (0.31) 0.81 (0.19) 0.97 (0.15) 0.82 (0.18) 0.90 (0.08) migration background 0.90 (0.14) 0.72 (0.20) 0.77 (0.28) 0.97 (0.14) 1.20 (0.25) 0.97 (0.08) partially dep. income 1.08 (0.20) 1.09 (0.32) 0.75 (0.22) 0.97 (0.13) 1.03 (0.24) 1.00 (0.09) dependent income 1.41 (0.43) 1.43 (0.82) 0.67 (0.28) 1.05 (0.25) 0.85 (0.28) 1.02 (0.14) both equally important (ties) sui 0.84 (0.10) 0.75* (0.09) den 0.80* (0.09) 0.81 (0.09) ger 0.96 (0.08) 0.91 (0.07) aut 0.91 (0.08) 0.91 (0.09) age 1.02 (0.01) 1.06* (0.03) 1.08** (0.03) 1.03* (0.01) 1.03 (0.02) 1.03*** (0.01) employed 0.93 (0.22) 0.50* (0.17) 1.14 (0.34) 0.78 (0.13) 1.12 (0.21) 0.88 (0.08) self-employed 1.00 (0.34) 0.50 (0.31) 0.68 (0.40) 0.75 (0.26) 0.54 (0.17) 0.73 (0.12) unemployed 0.79 (0.29) 0.93 (0.55) 1.34 (0.47) 0.87 (0.25) 1.52 (0.47) 1.08 (0.16) education level 1.00 (0.04) 0.94 (0.07) 0.83** (0.05) 0.96 (0.03) 1.03 (0.05) 0.96* (0.02) sex 0.71** (0.09) 0.75 (0.16) 0.54** (0.10) 0.55*** (0.06) 0.82 (0.11) 0.65*** (0.04) political affiliation 1.14*** (0.03) 1.05 (0.05) 1.02 (0.04) 1.00 (0.02) 1.03 (0.03) 1.05*** (0.01) income 1.00 (0.03) 1.03 (0.04) 0.98 (0.05) 1.03 (0.02) 1.06* (0.03) 1.02 (0.01) held job 0.64* (0.14) 1.42 (0.55) 0.66 (0.21) 1.15 (0.19) 1.21 (0.22) 1.00 (0.09) unemployment experience 0.75* (0.11) 0.54* (0.14) 0.59** (0.12) 1.18 (0.16) 0.63** (0.10) 0.80** (0.06) migration background 0.87 (0.12) 1.15 (0.26) 1.39 (0.39) 1.24 (0.15) 1.49** (0.23) 1.18* (0.08) partially dep. income 0.92 (0.16) 1.08 (0.27) 0.87 (0.21) 0.78* (0.09) 1.05 (0.18) 0.90 (0.07) dependent income 1.23 (0.35) 0.64 (0.33) 1.16 (0.40) 0.70 (0.15) 1.06 (0.26) 0.92 (0.11) n 6924 1768 582 754 2508 1312 6924 log pseudolikelihood -6994.12 -1794.97 -571.96 -750.81 -2510.04 -1219.54 -6924.94 note. odds ratios, robust standard errors in parenthesis, significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001. table 6. multinomial logistic models for the comparison of willingness to retrain vs. willingness to learn new skills. baseline model uk sui den ger aut full model base outcome: more willing to learn new skills more willing to retrain sui 0.91 (0.19) 1.05 (0.23) den 0.15*** (0.04) 0.13*** (0.04) ger 0.63*** (0.08) 0.62*** (0.08) aut 0.31*** (0.06) 0.36*** (0.07) age 1.02 (0.02) 1.11* (0.05) 1.18* (0.08) 1.01 (0.02) 1.11* (0.05) 1.04*** (0.01) employed 2.03 (0.81) 1.72 (1.12) 1.77 (1.45) 2.20* (0.69) 1.89 (1.02) 1.97*** (0.39) self-employed 2.44 (1.34) 2.31 (3.74) 0.00*** (0.00) 1.22 (0.83) 1.52 (1.36) 1.64 (0.56) unemployed 2.78 (1.61) 1.33 (1.22) 2.15 (2.01) 0.80 (0.45) 1.02 (0.72) 1.25 (0.35) education level 0.77*** (0.05) 0.74* (0.09) 0.50*** (0.08) 0.74*** (0.04) 0.87 (0.10) 0.75*** (0.03) sex 0.86 (0.18) 0.83 (0.33) 0.95 (0.53) 1.12 (0.20) 0.76 (0.25) 0.96 (0.11) political affiliation 1.14** (0.05) 1.06 (0.09) 1.02 (0.10) 1.17*** (0.05) 1.26** (0.09) 1.13*** (0.03) income 0.93 (0.05) 0.92 (0.08) 1.09 (0.13) 0.99 (0.03) 0.85** (0.05) 0.96 (0.02) held job 2.31* (0.86) 0.61 (0.39) 2.36 (2.82) 1.61 (0.47) 0.65 (0.30) 1.42 (0.27) unemployment experience 1.09 (0.25) 3.47* (1.70) 0.24 (0.18) 1.27 (0.28) 0.84 (0.31) 1.18 (0.16) migration background 0.82 (0.18) 2.20 (0.97) 2.74 (1.78) 1.12 (0.23) 1.16 (0.39) 1.13 (0.14) partially dep. income 1.03 (0.28) 0.89 (0.41) 1.24 (1.00) 0.91 (0.19) 0.85 (0.35) 0.97 (0.13) dependent income 1.25 (0.55) 0.32 (0.33) 3.72 (3.54) 1.60 (0.62) 1.45 (1.01) 1.31 (0.32) both equally willing (ties) sui 1.10 (0.14) 1.12 (0.14) den 0.44*** (0.04) 0.44*** (0.04) ger 0.69*** (0.05) 0.66*** (0.05) aut 0.66*** (0.06) 0.72*** (0.07) age 1.05*** (0.02) 1.02 (0.03) 1.09*** (0.02) 1.04** (0.01) 1.06*** (0.02) 1.05*** (0.01) employed 1.34 (0.28) 1.09 (0.34) 1.88* (0.51) 1.36* (0.19) 1.22 (0.21) 1.34*** (0.11) self-employed 1.42 (0.48) 4.21 (4.54) 0.87 (0.49) 1.10 (0.33) 1.27 (0.44) 1.27 (0.22) unemployed 2.09 (0.87) 0.39 (0.26) 0.75 (0.22) 1.62 (0.45) 1.50 (0.45) 1.27 (0.18) education level 0.87** (0.04) 0.93 (0.08) 0.78*** (0.04) 0.86*** (0.03) 0.88** (0.04) 0.85*** (0.02) sex 0.72* (0.09) 0.91 (0.23) 0.69* (0.11) 0.80* (0.08) 0.87 (0.12) 0.79*** (0.05) political affiliation 1.01 (0.03) 1.01 (0.05) 1.03 (0.03) 1.03 (0.02) 1.02 (0.03) 1.02 (0.01) income 0.93* (0.03) 1.01 (0.04) 1.01 (0.04) 1.03 (0.02) 1.04 (0.02) 1.02 (0.01) held job 1.44 (0.27) 1.94 (0.77) 1.28 (0.34) 0.92 (0.13) 1.19 (0.20) 1.15 (0.10) unemployment experience 0.93 (0.13) 1.47 (0.51) 1.17 (0.22) 1.16 (0.15) 1.07 (0.17) 1.09 (0.08) migration background 0.77 (0.10) 2.66*** (0.73) 1.08 (0.27) 1.45** (0.17) 1.18 (0.17) 1.17* (0.08) partially dep. income 0.87 (0.14) 1.38 (0.37) 1.35 (0.32) 0.95 (0.10) 1.00 (0.17) 0.99 (0.07) dependent income 0.82 (0.21) 1.12 (0.60) 1.12 (0.34) 0.75 (0.14) 0.87 (0.20) 0.85 (0.09) n 6,917 1,768 575 754 2,508 1,312 6,917 log pseudolikelihood -5322.31 -1297.81 -386.53 -530.81 -1962.03 -932.68 -5178.29 note. odds ratios, robust standard errors in parenthesis, significance levels: * p < 0.05, ** p < 0.01, *** p < 0.001. vol.8 • no.2 2019 issn: 2254-7088 european journal of government and economics 8(2), december 2019. european journal of government and economics    issn: 2254‐7088  number 8, issue 2, december 2019 doi: https://doi.org/10.17979/ejge.2018.8.2 why regions fail (or succeed). the role of government institutions in the long-run 114-144 doi: https://doi.org/10.17979/ejge.2019.8.2.4989 filippo bonanno restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case 145-160 doi: https://10.17979/ejge.2019.8.2.5478 joão ricardo catarino and ricardo de moraes e soares fdi in selected developing countries: evidence from bundling and unbundling governance 161-188 doi: https://doi.org/10.17979/ejge.2019.8.2.4970 simplice asongu the relationship between corporate tax rate and economic growth during the global financial crisis: evidence from a panel var governanceand domestic investment in africa 189-202 doi: https://10.17979/ejge.2019.8.2.5074 gamze oz-yalaman the impact of crd iv on bank lending 203-217 doi: https://doi.org/10.17979/ejge.2019.8.2.4656 matias huhtilainen european journal of government and economics 8(2), december 2019, 161-188 european journal of government and economics issn: 2254-7088 fdi in selected developing countries: evidence from bundling and unbundling governance simplice asongu*a a african governance and development institute, yaoundé, cameroon. * corresponding author at: asongus@afridev.org article history. received 17 january 2019; first revision required 10 august 2019; accepted 11 september 2019. abstract. the objective of this study is to assess governance drivers of fdi in a panel of brics and mint countries for the period 2001-2011. we bundle and unbundle governance determinants using a battery of contemporary and noncontemporary estimation techniques. our findings reveal the following: firstly, for both contemporary and noncontemporary specifications, while the majority of our governance determinants of gross fdi are significant, they are overwhelmingly insignificant for net fdi. secondly, the significance of the governance dynamics in increasing order of magnitude are general governance, political governance, economic governance, political stability, regulation quality and government effectiveness. thirdly, for non-contemporary specifications, the significance of governance variables is as follows in ascending order of magnitude: economic governance, institutional governance, general governance, corruption-control, political governance and political stability. the importance of combining governance indicators is captured by the effects of political governance, economic governance and institutional governance. the results indicate that the simultaneous implementation of the various components of governance clarifies a country’s attractiveness for fdi location. policy implications are discussed with particular emphasis on the timing of fdi and its targeting. keywords. foreign direct investment, emerging countries, governance jel codes. c52; f21; f23; p37; p39 doi. https:// 10.17979/ejge.2019.8.2.4970 1. introduction foreign-owned investments have been in existence since the colonial era in many parts of the globe. after a substantial drop in these investments in the 1980s, the need for security in food, energy and water is pushing many countries to adopt this new strategy of investment, especially in the aftermath of the 2008 food crisis (arezki et al., 2015). in essence, policies favouring restrictions to trade and capital that were predominant in developing nations in the 1970s and 1980s were considerably eased after these same countries suffered from declining economic prosperity and foreign investment (rodrik, 1998). hence, some domestic industries which these policies were initially meant to protect, bore much of the brunt of diminishing social and private returns (de mello, 1997; dupasquier & osakwe, 2006; unescap, 20001; akpanet al., 2014). hence, the policies known as ‘structural adjustment’ were fundamentally meant to address the capital scarcity in developing countries, while at the same time enabling multinational 1 united nations economic and social commission for asia and the pacific. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 162  corporations from the more developed world to benefit from the cheap labour in less developed nations (asongu, 2013a, 2014a; unctad2, 2013)3. with the current trend of land grab in the world, there is a growing strand in the literature focusing on foreign land acquisitions (fla) in developing countries (osabuohien, 2014, 2015)4. this rush for foreign direct investment (fdi) and fla extends well beyond african, asian and latin american countries in the south of the globe to ukraine, russia, and australia. two types of foreign investors have been documented: a european private sector characterised for the most part by investment banks and hedge funds, and asian investment of private and public origin (un, 2010). reasons advanced for motivating this fdi/fla range from debates to more fundamental poverty alleviation goals. consistent with world bank (2007), lipton (2009) and arezki et al. (2015), the underlying rush needs to provide some guarantee for initiatives favouring smallholder structures of agriculture which are friendlier towards poverty alleviation. the intuition for this line of narrative is the asian experience of relatively higher poverty reduction which has been substantially driven by small scale agriculture (loayza & raddatz 2010; de janvry & sadoulet, 2010; asongu & nguena, 2015). collier (2008) also follows this argument for sub-saharan africa because the region has been substantially affected by the 2008 global food price changes. in light of the literature discussed in section 2, fdi is mutually beneficial to both investment corporations and domestic economies. some advantages for host nations are finance, employment and positive externalities like managerial experience, technology & skills transfer and corporate governance. the benefits of the investing company are, inter alia: cheap labour, market access, natural resource availability and appealing externalities from bilateral and multilateral trade policies (akpan et al., 2014). in accordance with the narrative, as of 2012, fdi in developing nations soared substantially over the past decades to about 52% of global flows (unctad, 2013). among these recipients, a set of countries have accounted for most of the fdi flowing into developing economies: the brics (brazil, russia, india, china & south africa) and mint (mexico, indonesia, nigeria & turkey). according to the world bank (2013), 2united nations conference on trade and development. 3it is also interesting to note that not all privatisation policies delivered the much needed fdi. for instance, as shown by rolfe & woodward (2004), this has been the experience of zambia. the need for fdi has been further highlighted in recent literature on population studies which show that unless other sources of investment are channelled into developing countries (especially in africa), only public investment would be used to settle issues arising from the burgeoning population growth and resulting unemployment(asongu, 2013b). 4the mention of fla literature is meant to clarify how the positioning of the study on fdi departs from previous studies. the focus of the study is not on fla but on fdi. from intuition, fdi from the rest of the world to every sampled country (i.e. including china) includes fdi from other sampled countries. it is important to clarify the distinction between fdi and fla. the latter (i.e. fla) is the process by which foreign investors acquire large chunks of land. such foreign investors could be a corporation, an individual or a government agency acquiring land from individuals or the state in accordance with laws, inter alia: the payment of fixed government compensation in order to cover for losses that are incurred by local owners of the land. on the other hand, fdi is an investment that is made by an individual or a firm in one country for business interests that are located in another country. accordingly, fdi occurs when an investor acquires foreign business assets or foreign business operations which generally entail ownership or control of interest related to the foreign business. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 163  these countries account for most the fdi in their respective regions5: mexico in central america, nigeria in africa, india in southern asia, indonesia in south-eastern asia, turkey in west asia, brazil in south america and china in east asia. despite the increasing importance of the nine nations in attracting fdi and influencing the shape of the global economy, as far as we have reviewed, the fdi literature on these countries is scant. fdi determinants in the brics have been examined by vijayakumar et al. (2010) on panel data for the period 1975-2007 to conclude that, whereas the impact of trade and inflation is not insignificant, factorssuch as market size, labour cost, infrastructure and capital formation are more favourable to fdi inflows. jadhav (2012) concludes that fdi is fundamentally marketoriented since ‘natural resources’ have a negative impact, while a positive effect is found in trade, market size and the rule of law. jadhav & katti (2012) use the same periodicity to conclude that regulation quality and government effectiveness have positive effects, while voice & accountability, corruption-control and political instability have negative effects. akpan et al. (2014), a study in the literature closest to the present line of inquiry, assessed both the brics and mint economies with data from 2001 to 2011. the authors established that, whereas the quality of institutions and resources have insignificant effects, the impact of trade openness, infrastructure and market size are positive for fdi. these studies leave room for improvement in at least four areas: control for endogeneity,the contemporaneous nature of the relationships, complementing the brics with the mint, and articulating the essence of governance. in light of the above, our contribution to the literature is fourfold. first, contingent on the hausman test for endogeneity, we use panel fixed-effects (fe) to control for unobserved heterogeneity in terms of countryand time-effects. second, we introduce contemporary and non-contemporary specifications to assess whether determinants are contingent on their contemporary features. third, but for akpan et al. (2014), the underlying literature has been limited to the brics. hence, we complement existing literature by providing evidence from both the brics and mint economies. fourth, we have found that the effects of governance may be insignificant (akpan et al., 2014) or limited to the rule of law (jadhav, 2012) and economic governance (jadhav & katti, 2012). we extend the dimension of institutions by bundling and unbundling governance dynamics. in essence, we use ten governance indicators, notably: institutional governance, economic governance, political governance, general governance, corruption-control, rule of law, regulation quality, government effectiveness, voice & accountability and political stability/no voilence6. the bundling and unbundling of governance variables is through principal component analysis, and the bundled governance indicators represent principal components of constituents variables in the principal component analysis. the derived principal components represent composite or bundled indicators which reflect common information in the constituent indicators. 5geographic regions are consistent with the unctad classification. 6governance and institutions are used interchangeably throughout the paper. the latter concept of institutions is different from institutional governance which is measured by corruption-control and rule of law. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 164  contemporary regressions entail a process where-by the contemporary outcome variable is regressed on contemporary independent variables while non-contemporary regressions entail the regression of the contemporary outcome variable on non-contemporary independent variables. while contemporary variables are in level series, non-contemporary variables are lagged by one year. the intuition for articulating the quality of institutions draws on a recent stream of interesting literature focusing on bundling and unbundling institutions for development outcomes. oluwatobi et al. (2015) investigated the effect of various governance components on innovation in africa and concluded that economic governance (regulation quality and government effectiveness) isthe most important. andrés & asongu (2013) have investigated how various governance dynamics affect the fight against software piracy, and found corruption-control to be the most effective tool. andrés et al. (2014) employ the same governance mechanisms to access how upholding intellectual property rights (ipr) treaties affect the knowledge economy (ke) and concluded that formal institutions are a necessary, but not sufficient condition for ke in africa. asongu & kodila-tedika (2016) followed the same strategy employed by the two preceding studies in assessing which governance channels are most effective in the fight against african conflicts and crimes. they conclude that corruption-control is the most effective institutional weapon. drawing on the above, asongu & nwachukwu (2016a) bundled and unbundled institutions in predicting the arab spring. this process has also been employed for the measurement of political governance (voice &accountability and political stability/no violence) to show the effect of lifelong learning on political stability and non-violence in africa (asongu & nwachukwu, 2016b). in light of the above, the objective of this study is to assess governance drivers of fdi in a panel of brics and mint countries for the period 2001-2011. assessing governance drivers is relevant to both scholars and policymakers. while the gap it fills in the scholarly literature has been discussed in the preceding paragraph, the policy importance of bundling and unbundling governance indicators is to improve policy options regarding how a plethora of governance drivers can independently and collectively affect fdi. to make the assessment, the study bundles and unbundles governance determinants using a battery of contemporary and noncontemporary estimation techniques. our findings reveal the following: firstly, for both contemporary and non-contemporary specifications, while the majority of our governance determinants of gross fdi are significant, they are overwhelmingly insignificant for net fdi. secondly, the significance of the governance dynamics in increasing order of magnitude are general governance, political governance, economic governance, political stability, regulation quality and government effectiveness. thirdly, for non-contemporary specifications, the significance of governance variables is as follows in ascending order of magnitude: economic governance, institutional governance, general governance, corruption-control, political governance and political stability. the importance of combining governance indicators is captured by the effects of political governance, economic governance and institutional governance. the results indicate that the simultaneous implementation of the various s. asongu / european journal of government and economics 8(2), december 2019, 161-188 165  components of governance clarifies a country’s attractiveness for fdi location. policy implications are discussed with particular emphasis on the timing of fdi and its targeting. the rest of the paper is organised in the following manner. the theoretical underpinnings, empirical literature and stylized facts are covered in section 2. data and methodology are covered in section 3. section 4presents the empirical analysis and discussion of results. we conclude in section 5. 2. theoretical underpinnings, empirical literature and stylized facts 2.1. theoretical underpinnings this first section on theoretical underpinnings of fdi/fla location substantially draws on vernon (1966), who also documented a product life cycle which articulates four main stages: introduction, growth, maturity and decline. according to this pattern, new products are introduced in developed and later diffused to less-developed nations over time. hence, these fundamental stages substantially influence the location decision of multinational corporations to,inter alia, set-up production facilities abroad and to benefit from lower production cost and address concerns of growing demand in less developed countries. consistent with apkan et al. (2014), the electric paradigm conceived by dunning (1988, 1993, 2000) provide a general perspective for rationalizing fdi location decisions by multinational companies. according to this model, factors like the geography, scope and industrial elements of fdi by multinationals corporations are substantially affected by interactions in the following three sets of interdependent indicators: location specificity, strategic ownership advantages and internationalisation. this is broadly consistent with the recent survey of theoretical underpinnings on determinants by faeth (2009). 2.2. empirical literature in the second section, we devote some space to discussing the findings of the empirical literature on the determinants of fdi/fla. consistent with recent literature (akpan et al., 2014), it depends on a number of factors, among others: estimation techniques, context of papers, data span and proxies used for indicators (moosa, 2002; asiedu, 2006; hajzler, 2014; moosa & cardak, 2006; asiedu, 2002; ranjan & agrawal, 2011; buchanan et al., 2012; sekkat & veganzones-varoudakis, 2007). we follow asongu & nguena (2014) in discussing them in six main strands, namely: quality of business climate (infrastructure, trade, returns & institutions), tenure security, weak governance, resource-grab motivations, regional factors and global economic shocks. the first is linked to fdi while the others broadly apply to fdi & fla. on the first strand that is focused on business climate, amendolagine et al. (2013) have investigated factors motivating fdi and concluded that features such as local partners, market factors and time are significant. other documented characteristics are infrastructure and return to capital (asiedu, 2002), market size, trade openness and availability of infrastructure s. asongu / european journal of government and economics 8(2), december 2019, 161-188 166  (vijayakumar et al., 2010; bartels et al., 2009; kinda, 2010; darley, 2012; anyanwu, 2012; akpan et al., 2014; büthe & milner, 2008; bartels et al., 2014; jadhav, 2012); the abundance of cheap labour and incentive packages (vijayakumar et al., 2010; tuomi, 2011; asongu, 2014b). institutional factors entail corruption-control (de maria, 2010; wei, 2010), democracy (asiedu & lien, 2011), political stability (busse & hefeker, 2007), economic governance (jadhav & katti, 2012) and good institutional quality (gastanaga et al., 1998; neumayer & spess, 2005; kinda, 2010; tuomi, 2011; asongu, 2012; cleeve, 2012; abdioglu et al., 2013; hayakawa et al., 2013; bartels et al., 2014). in the second strand, areski et al. (2013) who document the attractive force of weak governance have also confirmed business climate quality is an attractive feature for fdi/fla. here, bad governance motivates foreign investments. while kolstad & wiig (2011) have confirmed poor institutional quality as the primary factor motivating fdi from china to africa, asongu & aminkeng (2013) have balanced the narrative by concluding that the motivations of western companies are not much different from those of chinese corporations. the third strand focuses on land tenure security issues which have been documented as an important factor in fla (un, 2010; arezki et al., 2015). systems of land tenure affect food security (economic commission for africa, 2014) and have been identified as one of the fundamental factors influencing fdi/fla (ingwe et al. 2010; okoth-ogendo, 2008). the narrative which is in line with wouterse et al. (2011) broadly characterises the issues as “taking away the land of peasants which are possessed on communal tenure systems that starkly contrast with official land titles related to ‘indigenous colonialist’ controlled neoliberal capitalist systems who have used various forms of manipulation in the past to alienate africans from their land” (asongu & nguena, 2014, p.4). german et al. (2011) argue that in spite of their recognition, customary rights are not fundamentally protected by fla agreements. along the same lines, thaler (2013) concludes that foreign investment targets countries that are characterised by authoritarian and corrupt governments associated with weak land tenure security; in countries where the rights of the local population are not clearly articulated and governance is poor, flas are linked to substantial risks for the population (liu, 2013), and local institutions do not substantially affect decisions in fla because of overwhelming state power (osabuohien, 2014). resource-seeking motivations constitute the fourth strand (aleksynska & havrylchyk, 2013; lay & nolte, 2014). whereas a negative nexus has been established between natural resource-wealth and fdi in the presence of protectionist policies (jadhav, 2012; rogmans & ebbers, 2013), the mainstream narrative suggests a reverse relationship. for instance, lay & nolte (2014) have extended arezki et al. (2015) to confirm the positive connection between natural resource endowment and fdi. the kostad & wiig (2011) conclusion on a resourcethirsty china has also been debunked by asongu & aminkeng (2013), who conclude that the resource motivations of western nations are very much identical to those of china. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 167  table 1: stylized facts on brics and mint gdp (constant 2005 us$, billions) gdp per capita (constant 2005 us$) gdp growth (annual %) gdp per capita growth (annual %) fdi net inflows (bop current us$ billions) population growth (annual %) population, total, millions natural resources, share of gdp* human development index (hdi) brazil 1136.56 5721.23 0.87 0.00 71.54* 0.87 198.66 5.72 0.73 china 4522.14 3348.01 7.80 7.28 280.07* 0.49 1350.70 9.09 0.70 india 1368.76 1106.80 3.24 1.94 32.19* 1.26 1236.69 7.36 0.55 indonesia 427.47 1731.59 6.23 4.91 19.24* 1.25 246.86 10.00 0.63 mexico 997.10 8250.87 3.92 2.65 21.50* 1.24 120.85 9.02 0.78 nigeria 177.67 1052.34 6.55 3.62 8.84* 2.79 168.83 35.77 0.47 russia 980.91 6834.01 3.44 3.03 55.08* 0.40 143.53 22.03 0.79 south africa 307.31 6003.46 2.55 1.34 5.89* 1.18 51.19 10.64 0.63 turkey 628.43 8492.61 2.24 0.94 16.05* 1.28 74.00 0.84 0.72 *2011 data. source of data: undp (2013), world bank (2013) and akpan et al. (2014) in the fifth strand, we find literature on global shocks like food and financial crises as the principal drivers of fdi/fla for agriculture purposes (wouterse et al., 2011). after the 2008 global food crises, countries that substantially relied on food imports began acquiring land abroad for food security agricultural purposes (un, 2010). according to clapp (2013), fairbairn (2013) and isakson (2013), financial investors and private sectors seized the opportunity of speculative investments when in 2008 about 25 countries imposed food export restrictions. investment banks that engaged in such speculation with agricultural investment funds are knight frank in the uk, goldman sachs & black rock in the usa and deutsche bank in germany. in summary, consistent with german et al. (2011), the increasing interest in biofuels and rapid growth of emerging economies are some factors that have influenced the speed and scale of fdi/fla. factors in the sixth strand are regional. before the 2007/2008 food and financial crises, asiedu (2002) had established that sub-saharan africa (ssa) received relatively less fdi by virtue of its geographic location. anyanwu (2012), who does not subscribe to asiedu’s position, concludes that the eastern and southern sub-regions in africa are predisposed to obtain more fdi. a new stream of research is consistent with the view that ssa is a good candidate for fla location decisions because of, among other things: the relatively low use of water supply which currently stands in the neighbourhood of 2% of underground reserves (un, 2010), well-nurtured north-south fdi relations (aleksynska & havrylchyk, 2013), the existence of local partners based on strong colonial networks (amendolagine et al., 2013), and china’s strategy that is oriented towards non-interference and partnership (yin & vaschetto, 2011). s. asongu / european journal of government and economics 8(2), december 2019, 161-188 168  2.3. stylized facts table 1 below presents some stylized facts of the brics and mint countries. consistent with apkan et al. (2014), the former accounted for 15% of world gdp and attracted 26% of global fdi. some interesting common features among brics and mint are: membership in the great 20 (g20), excluding nigeria, burgeoning youth population, and fdi-friendly policies. other stylized facts presented in the table clearly articulate the evolving importance of these nations. for instance, between 2001 and 2012, fdi to the nine countries rose to 510.4 billion from 113.6 billion (in current usd). within the same horizon, these countries accounted for 51% of the population in the world, attracted about 30% of global fdi and 19% of world gdp (world bank, 2013). 3. data and methodology 3.1. data the study assesses a panel of the nine brics (brazil, russia, india, china & south africa) and mint (mexico, indonesia, nigeria & turkey) fast-growing emerging countries with data from akpan et al. (2014) for the period 2001-2011. the geographical and temporal scopes of the study are based on data availability constraints at the time of the study. data from the underlying study, which is consistent with unctad's classification of fdi determinants (see table 2), is obtained from the world development indicators and the world governance indicator databases of the world bank. two dependent variables are used in the analysis, notably gross fdi and net fdi. the choice of these dependent variables is in accordance with the underpinning literature which is based on four principal types of fdi, namely: net fdi inflow as a percentage of gdp (lehnert et al., 2013), net fdi flows as a percentage of gdp (asiedu, 2002), unidirectional fdi inflow into recipient countries (rogmans & ebbers, 2013) and net fdi inflow (jadhav, 2002). following apkan et al. (2014), we use net fdi and gross fdi. the motivation for using both measures is to control for capital consumption (or depreciation). accordingly, gross fdi is total investments on new inputs of capital while net fdi is the gross fdi that is adjusted for depreciation (or capital consumption). the adopted determinants or independent variables have been discussed in the literature above. they are in accordance with the unctad’s classification in table 2. the exogenous variables are: natural resources, inflation, infrastructure, bank credit and ten governance variables. while the first-four are control variables, the governance dynamicsare the key variables of interest. theyare (i) voice & accountability, (ii) poltical stability, (iii) regulation quality, (iv) government effectiveness, (v) the rule of law and(vi) corruption-control, (vii) political governance, (viii) economic governance, (ix) institutional governance and (x) general governance. the latter four of the governance dynamic are principal components (pcs) generated by bundling the former six individual governance variables using the principal component analysis (pca) method explained in section 3.2.1 below. on the expected signs of s. asongu / european journal of government and economics 8(2), december 2019, 161-188 169  the governance variables, a key point is noteworthy here. as discussed in the preceding section, there is as yet no consensus on the effects of governance dynamics. this is thepartial motivation here for bundling and unbundling the impact of governance elements (pelizzo & nwokora, 2016, 2018; pelizzo, araral, pak &xun, 2016; asongu & nnanna, 2019; asongu & odhiambo, 2019a, 2019b, 2019c). with respect to the set of control variables, we expect a positive relationship with fdi, except for inflation. high inflation is potentially detrimental to fdi.inflation which is measured with the consumer price index is consistent with barro (2003). the anticipated sign could also be positive because low and stable inflation has been documented to be conducive for a promising economic outlook (asongu 2013a). this draws from the intuition that chaotic inflation is linked to uncertainty and investors prefer to engage with less ambiguous economic strategies (le roux & kelsey, 2017, 2018). the choice of bank credit as a control indicator is in line with asongu (2015) and the anticipated positive nexus with investment is consistent with the theoretical and empirical evidence from the financial development literature (see levine, 2005). in essence, credit availability offers investment opportunities to economic operators. the choice of natural resources which is in accordance with fosu (2013) is essentially motivated by the documented evidence that the exploitation and exportation of natural resources is directly linked with fdi in developing countries (see amavilah, 2015). the importance of infrastructure as a determinant has been recently documented by sahoo et al. (2010). the positive role of infrastructure in determining fdi location decisions is consistent with asiedu (2002) and sekkat and veganzones-varoudakis (2007), who have also used mobile phones’ (per 100 people) as a proxy. in essence, mobile telephony has been substantially documented to be driving investment and growth in developing nations (afutu-kotey et al., 2017; asongu & boateng, 2018; bongomin et al., 2018 ; gosavi, 2018; hubani & wiese, 2018; isszhaku et al., 2018; minkoua nzie et al., 2018; muthinja& chipeta, 2018; abor et al., 2018). the summary statistics of the variables are presented in table 3 below. it could be inferred from it that the variables are comparable. moreover, the degree of variation is quite substantial, and hence, reasonable estimated relationships are expected to emerge. table 2: unctad's classification of fdi determinants determining variables examples policy variables tax policy, trade policy, privatization policy, macroeconomic policy business variables investment incentives market-related economic determinants market size, market growth, market structure resource-related economic determinants raw materials, labour costs, technology efficiency-related economic determinants transport and communication costs, labour productivity source: unctad (2002) and akpan et al. (2014) s. asongu / european journal of government and economics 8(2), december 2019, 161-188 170  table 3: summary statistics of variables mean s.d min max obs net foreign direct investment (nfdi) 28.979 46.359 -2.977 280.07 99 foreign direct investment (fdi) 2.402 1.348 -1.855 6.136 99 infrastructure (number of mobile phones per 100 people) 52.433 39.220 0.210 179.31 99 bank credit (on gdp) 85.019 63.492 4.909 201.58 99 natural resources (on gdp) 9.003 8.110 0.294 38.410 99 inflation (consumer price index) 8.580 7.519 -0.765 54.400 99 voice & accountability -0.192 0.680 -1.681 0.727 99 political stability -0.826 0.613 -2.193 0.286 99 regulation quality -0.104 0.437 -1.322 0.778 99 government effectiveness -0.100 0.454 -1.200 0.691 99 rule of law -0.428 0.458 -1.522 0.279 99 corruption control -0.431 0.462 -1.333 0.612 99 political governance 0.000 1.153 -2.210 1.976 99 economic governance -0.000 1.372 -3.291 2.639 99 institutional governance 0.000 1.348 -3.048 2.412 99 general governance 0.000 2.124 -4.650 3.765 99 3.2. methodology 3.2.1. principal component analysis consistent with asongu & nwachukwu (2015), the substantial degree of substitution among governance indicators in table 5 implies some overlapping information. we employ principal component analysis (pca) to address this concern. the use of the pca technique also enables us to bundle governance variables. this statistical method facilitates the reduction of a high set of correlated variables into a smaller combination of uncorrelated indicators known as principal components (pcs). in the process, four more governance indicators are blended from the six individual governance variables identified in section 2.1. the pc governance dynamics comprise: political governance, which measures the election and replacement of political leaders is approximated by: voice & accountability and political stability/non-violence; economic governance, which is the formulation and implementation of policies that deliver public commodities, is denoted by regulation quality and government effectiveness ;institutional governance, which is defined as the respect of the state and citizens of institutions that govern interactions between them is measured by the rule of law and corruption-control (andrés et al., 2015). the policy interest of bundling and unbundling governance variables is to avoid conceptual conflation in the governance literature. for instance, it is inappropriate to use the term “political governance” unless it captures “political governance” and “voice & accountability”7. moreover, 7 it is important to note that there is an evolving stream of literature on the need to bundle and unbundle governance variables in order to limit conceptual conflation (asongu, 2016; asongu & ssozi, 2016; ajide & raheem, 2016a, 2016b; asongu et al., 2018, 2019). s. asongu / european journal of government and economics 8(2), december 2019, 161-188 171  kangoye (2013) has used governance to qualify a study when corruption is the main governance dynamic used in the study. according to this study, general governance can only be employed to qualify a context, if it entails political stability/no violence, voice & accountability, regulation quality, government effectiveness, corruption control and the rule of law. the general governance indicator can be derived by means of pca. consistent with the underlying literature (tchamyou, 2017, 2019), we use the kaiser (1974) and jolliffe (2002) criterion for the retention of common factors. hence, we retain factors or pcs with an eigenvalue higher than the mean (or one). in table 4 below, it can be observed that: general governance (g.gov) which is a first pc has an eigenvalue of 4.514 and represents about 75% of the variation in the original six individual governance variables. in spite of the bundling of variables in order to improve policy implications and avoid conceptual conflations, the study addresses the issues of high degrees of substitution among governance variables by employing them in distinct specifications to limit concerns of multicollinearity. borrowing from asongu & nwachukwu (2016), we devote some space to discussing potential concerns that may arise when regressors originate from previous regressions. three issues have been documented by pagan (1984, p. 242) on the quality of resulting estimators. they are: (i) efficiency, (ii) consistency and, (iii) validity of inferences at the second stage of the estimation. according to the conclusions of the author, whereas estimators from a two-step procedure are consistent and efficient, inferences provided by a few are valid. this narrative is broadly in accordance with recent literature on the use methods such as pca, which relies on a two-step regression modelling (oxley & mcaleer, 1993; mckenzie & mcaleer, 1997; ba & ng, 2006; westerlund & urbain, 2013a). table 4: principal component analysis (pca) for governance (gov) principal components component matrix(loadings) proportion cumulative proportion eigen value va ps rq ge rl cc first pc (g.gov) 0.305 0.385 0.440 0.441 0.409 0.452 0.752 0.752 4.514 second pc 0.848 -0.461 -0.207 -0.115 0.096 0.048 0.121 0.874 0.731 third pc 0.337 0.532 -0.240 0.192 -0.714 0.012 0.064 0.938 0.385 first pc (polgov) 0.707 0.707 --------0.664 0.664 1.329 second pc -0.707 0.707 --------0.335 1.000 0.670 first pc (ecogov) ----0.707 0.707 ----0.941 0.941 1.883 second pc -----0.707 0.707 ----0.058 1.000 0.116 first pc (instgov) --------0.707 0.707 0.909 0.909 1.818 second pc ---------0.707 0.707 0.090 1.000 0.181 “p.c: principal component. va: voice & accountability. rl: rule of law. r.q: regulation quality. ge: government effectiveness. ps: political stability. cc: control of corruption. g.gov (general governance): first pc of va, ps, rq, ge, rl & cc. polgov (political governance): first pc of va & ps. ecogov (economic governance): first pc of rq & ge. instgov (institutional governance): first pc of rl & cc”. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 172  table 5: correlation analysis va ps rq ge rl cc polgov ecogov instgov g.gov 1.000 0.329 0.542 0.457 0.538 0.623 0.815 0.515 0.614 0.648 va 1.000 0.774 0.759 0.579 0.752 0.815 0.790 0.698 0.817 ps 1.000 0.883 0.716 0.886 0.807 0.970 0.840 0.934 rq 1.000 0.827 0.861 0.746 0.970 0.885 0.936 ge 1.000 0.818 0.685 0.795 0.953 0.868 rl 1.000 0.849 0.900 0.953 0.959 cc 1.000 0.800 0.804 0.899 polgov 1.000 0.889 0.963 ecogov 1.000 0.958 instgov 1.000 g.gov “p.c: principal component. va: voice & accountability. rl: rule of law. r.q: regulation quality. ge: government effectiveness. ps: political stability. cc: control of corruption. g.gov (general governance): first pc of va, ps, rq, ge, rl & cc. polgov (political governance): first pc of va & ps. ecogov (economic governance): first pc of rq & ge. instgov (institutional governance): first pc of rl & cc”. the use of pcs within the framework of this analysis has been documented by westerlund & urbain (2012, 2013b) who have built on previous papers (pesaran, 2006; stock & watson, 2002; bai, 2003; bai, 2009; greenaway-mcgrevy et al., 2012). as to what error are inherent in pc regressors, they have remarked on the possibility of normal inferences with pc-factors augmenting regressions, if the coefficients that are estimated converge toward their real values at the rate: nt , (where t is the number of time series and n, the number of cross sections). we argue that any potential issues of small sample bias are not very feasible here because we are constrained by the sample size. in essence, only nine countries constitute the mint and brics among fast-growing developing countries. 3.2.2. estimation technique we assess contemporary and non-contemporary determinants using panel regressions. the choice between panel fixed effects (fe) or random effects (re) is decided by the outcome of the hausman test for endogeneity. assuming the hausman test for endogeneity is rejected, eq.(1) and eq. (2) below denote the corresponding contemporary and non-contemporary specifications respectively of fe regressions. tititih n h jti wfdi ,,, 1 ,     [1] tititih n h jti wfdi ,,, 1 ,     [2] tititih n h jti wfdi ,1,, 1 ,      [3] s. asongu / european journal of government and economics 8(2), december 2019, 161-188 173  where: tifdi , is the foreign direct investment for country i at period t ; is a constant,w is the vector of determinants i is the country-specific effect, and ti, the error term. the regressions are specified with heteroscedasticity and autocorrelation consistent (hac) standard errors. the pairwise correlation matrix in table 6 below helps us in mitigating potential problems arising from multicollinearity and overparameterization. interestingly, the linear association between gross fdi and our governance variables, with the exception of political stability/nonviolence is negative. the inference is that reforms in these other governance variables by themselves could potentially reduce the attractiveness of brics and mint economies as destinations for fdi. such provides support for the weak governance effect suggested by areski et al. (2013). this finding is consistent with our indicators of political governance, regulation quality, corruption-control, institutional governance and general governance regardless of whether the gross fdi or net fdi were considered in the correlation analysis. by contrast, the correlation coefficients for the indicators of economic governance, government effectiveness and the rule of law reverted to a positive sign when net fdi was used in the pairwise correlation. we may surmise that policy actions which enhance the quality of institutions in terms of these last three dynamics may help curtail the problem of reverse investment or disinvestment in our brics and mint states, even if they might not necessarily lead to a significant increase in inward direct investment. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 174  table 6: correlation matrix (n for panel a =90) control variables governance dynamics foreign investment infra infla credit nres va ps pgov rq ge egov rl cc ingov ggov fdi nfdi 1.000 -0 .102 0.210 0.277 0.032 0.291 0.198 0.291 0.190 0.248 0.132 0.141 0.143 0.212 0.136 0.183 infra 1.000 -0.0004 0.077 -0.061 -0.274 -0.205 -0.124 -0.254 -0.193 -0.150 -0.253 -0.211 -0.219 -0.157 -0.251 infla -0.488 0.114 0.548 0.406 0.585 0.682 0.658 0.716 0.703 0.744 0.668 -0.031 0.174 credit 1.000 -0.269 -0.228 -0.305 -0.261 -0.345 -0.312 -0.490 -0.455 -0.495 -0.397 0.057 0.049 nres 1 .000 0.329 0.815 0.542 0.457 0.515 0.538 0.632 0.614 0.648 -0.392 -0.056 va 1.000 0.815 0.774 0.759 0.790 0.579 0.752 0.698 0.817 0.137 0.221 ps 1.000 0.807 0.746 0.800 0.685 0.849 0.804 0.899 -0.156 -0.209 pgov 1.000 0.883 0.970 0.716 0.886 0.840 0.934 -0.113 -0.028 rq 1.000 0.970 0.827 0.861 0.885 0.936 -0.143 0.128 ge 1.000 0.795 0.900 0.889 0.963 -0.143 0.051 egov 1.000 0.818 0.953 0.868 -0.247 0.028 rl 1.000 0.953 0.959 -0.087 -0.067 cc 1.000 0.958 -0.175 -0.020 ingov 1.000 -0.151 -0.028 ggov 1.000 0.448 fdi 1.000 nfdi infra: infrastructure. infla: inflation. credit : domestic credit. nres: natural resources. va: voice & accountability. ps: political stability. polgov: political governance. rq: regulation quality. ge: government effectiveness. egov: economic governance. rl: rule of law. cc: corruption-control. ingov: institutional governance. ggov: general governance. fdi: gross fdi. nfdi: net fdi. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 175  4. empirical results 4.1. presentation of results table 7 and table 8 below present contemporary and non-contemporary determinants of fdi, respectively. panel a of either table has gross fdi as the dependent variable, while the dependent variable for panel b is net fdi. the decision as to whether a fe or re model is a better fit is contingent on the outcome of the hausman test. a rejection of the test implies the fe model is a better fit. the following broad finding can be established. while the determinants of gross fdi are significant in panel a, they are overwhelmingly insignificant for panel b on net fdi. we may therefore suppose that governance reforms in countries with similar long-term attributes such as language, culture, religion, climate, demography and ethnicity, would have a comparable effect on inward and outward direct investment decisions. this inference is consistent with both contemporary and noncontemporary specifications8 in tables 7 and 8, respectively. the results from the contemporary specifications may be summarised as follows. first, the significances of governance dynamics are as follows in increasing order of magnitude9: general governance (0.561), political governance (0.595), economic governance (0.832), political stability (1.006), regulation quality (1.669) and government effectiveness (2.035). second, while institutional governance and its corresponding components (rule of law and corruption-control) have insignificant effects, the impact of voice & accountability is persistently negative. a possible explanation for this surprising result is that freedom of speech, accountability and press reporting on matters such as minimum wages, health and safety, environmental controls, tax evasion and human rights abuse may not favour returns to direct investment. third, the motivation to bundle governance variables is articulated by the effect of political governance which is significantly positive while one of its components (voice & accountability) is consistently negative. fourth, the signs for the coefficients for most of our governance dynamics reverted from negative in the pairwise correlation analysis to positive in the panel fixed and random effect models. this may be construed as an indication that fdi flows are not simply motivated by governance reforms per se, but by the interrelatedness between these structural adjustments and the above-mentioned persistent country attributes. fifth, the significant control variables have the expected signs. accordingly, infrastructure and natural resources positively influence gross fdi flows. 8‘both specifications’are used subsequently to refer to ‘contemporary and non-contemporary’ specifications. 9 the magnitude imply that one governance dynamic is more important than others to increase fdi flows and by extension should be given policy priority in decisions to attract fdi. s. asongu / european journal of government and economics 8(2), december 2019, 161-188 176  table 7. contemporary determinants (panel fixedand random-effects). panel a : gross fdi constant 1.754*** 2.82*** 2.059* 1.504 1.199 0.981 1.341 2.483** 1.580 1.536 (0.003) (0.003) (0.066) (0.144) (0.226) (0.356) (0.115) (0.013) (0.160) (0.157) voice & accountability -0.761** ------------------ (0.039) political stability --1.01*** ---------------- (0.007) political governance ----0.595** -------------- (0.029) regulation quality ------1.669** ------------ (0.044) government effectiveness --------2.035** ---------- (0.024) economic governance ----------0.832*** -------- (0.001) rule of law -------------0.525 ------ (0.443) corruptioncontrol ---------------0.004 ---- (0.714) institutional governance ----------------0.483 -- (0.100) general governance ------------------0.561*** (0.006) nresources 0.033 0.072** 0.052*** 0.064** 0.073** 0.079** 0.046 0.046 0.047** 0.065** (0.220) (0.015) (0.000) (0.044) (0.029) (0.041) (0.105) (0.176) (0.028) (0.016) infrastructure 0.007** 0.008*** 0.017*** 0.025*** 0.023*** 0.027*** 0.009*** 0.024*** 0.020*** 0.023*** (0.013) (0.006) (0.000) (0.000) (0.000) (0.000) (0.005) (0.000) (0.000) (0.000) inflation -0.020 -0.021 -0.016 -0.011 -0.010 -0.009 -0.019 0.014 0.0009 -0.002 (0.305) (0.261) (0.367) (0.567) (0.442) (0.543) (0.352) (0.424) (0.963) (0.880) domestic credit -0.001 -0.006 -0.007 0.0005 0.003 0.003 0.0003 -0.004 -0.001 -0.003 (0.797) (0.318) (0.568) (0.967) (0.800) (0.819) (0.951) (0.714) (0.894) (0.819) hausman test 8.547 6.011 18.404*** 11.258** 12.836** 14.800** 7.262 15.652*** 12.562** 23.843*** time effects no no yes yes yes yes no yes yes yes log-likelihood -132.1729 -159.038 ---------142.920 ------ within variance 0.733 0.646 --------0.733 ------ between variance 0.587 1.874 --------0.939 ------ within r² ----0.437 0.450 0.434 0.462 --0.452 0.423 0.462 fisher ----7.273*** 7.524*** 7.222*** 7.749*** --7.553*** 7.019*** 7.741*** observations 90 90 90 90 90 90 90 90 90 90 s. asongu / european journal of government and economics 8(2), december 2019, 161-188 177  panel b : net fdi constant 39.079 -19.468 40.571 42.172 32.557 36.040 -0.599 45.951 42.000 41.944 (0.193) (0.531) (0.160) (0.152) (0.187) (0.176) (0.984) (0.187) (0.154) (0.152) voice & accountability -7.631 ------------------ (0.845) political stability ---5.848 ---------------- (0.618) -2.515 political governance ----(0.811) -------------- regulation quality ------3.684 ------------ (0.889) government effectivness --------47.677 ---------- (0.220) economic governance ----------8.324 -------- (0.457) rule of law ------------18.723 ------ (0.415) corruptioncontrol --------------7.570 ---- (0.788) institutional governance ----------------5.163 -- (0.670) general governance ------------------3.233 (0.705) nresources -0.424 1.382 -0.449 -0.367 0.275 -0.059 1.481 -0.404 -0.370 -0.287 (0.747) (0.142 (0.736) (0.755) (0.769) (0.950) (0.111) (0.735) (0.725) (0.809) infrastructure -0.044 0.436*** -0.020 -0.008 0.100 0.065 0.414*** 0.001 -0.0005 0.003 (0.911) (0.000) (0.952) (0.980) (0.710) (0.827) (0.000) (0.997) (0.998) (0.991) inflation 0.773 0.658 0.805 0.803 0.873 0.840 0.674 0.912 0.956 0.862 (0.158) (0.288) (0.174) (0.162) (0.147) (0.146) (0.279) (0.230) (0.187) (0.162) domestic credit -0.448 0.032 -0.453 -0.476 -0..379 -0.439 -0.035 -0.495 -0.488 -0.491 (0.245) (0.888) (0.265) (0.247) (0.252) (0.247) (0.875) (0.243) (0.246) (0.236) hausman 21.169*** 7.146 20.575*** 17.58*** 10.931* 13.75** 8.536 24.613*** 11.964** 17.77*** time effects yes no yes yes yes yes no yes yes yes log-likelihood ---482.063 ---------481.896 ------ within variance --635.748 --------634.301 ------ between variance --2335.06 --------1964.07 ------ adjusted r² 0.352 --0.352 0.352 0.371 0.358 --0.352 0.354 0.354 fisher 11.292*** --11.297*** 11.28*** 11.726*** 11.425*** --11.302*** 11.342*** 11.329*** observations 90 90 90 90 90 90 90 90 90 90 *, **, ***: significance levels of 10%, 5% and 1% respectively. the random effects specifications are not modelled with timeeffects due to issues in degrees of freedom. accordingly, the matrices become ‘positive definite’ when the model is specified with ‘time-effects’. the following outcomes are established for non-contemporary specifications in table 8. first, the significances of the governance dynamics are as follows in increasing order of magnitude: economic governance (0.427), institutional governance (0.485), general governance (0.489), corruption-control (0.578), political governance (0.802) and political stability (0.908). second, while regulation quality and government effectiveness have insignificant effects on gross fdi, their combined impact as captured s. asongu / european journal of government and economics 8(2), december 2019, 161-188 178  by the economic governance variable is significantly positive at the ten percent level. third, the decision to bundle governance dynamics is justified by the effects of political governance, economic governance and institutional governance, which varied markedly from those of their individual elements either in terms of sign, size and level of statistical significance. for instance, (i) political governance is significantly positive while one of its components (voice & accountability) is negative (ii) economic governance is significantly positive while its components (regulation quality and government effectiveness) are not and (iii) institutional governance is significant while one of its components (rule of law) is not. fourth, the significant control variables have the expected signs. accordingly, infrastructure, domestic credit and natural resources positively influence gross fdi while inflation has a negative effect. overall, irrespective of tables, it is worthwhile to articulate that the large constant terms, especially for net fdi, suggest that much of the variations in fdi are not explained by governance, natural resources, infrastructure, inflation, and domestic rate of interest. 4.1. further discussion of results and policy implications we discuss the results in four main strands: differences in tendencies of effect on gross fdi versus net fdi10; comparing and contrasting contemporary and non-contemporary specifications in terms of significance & magnitude and interest of bundling & unbundling governance dynamics on contemporary &non-contemporary specifications. first, the fact that the governance dynamic effects on gross fdi are significant while they are insignificant for net fdi logically implies that the effects of governance may be more apparent in fdi outflows or disinvestment. the results are broadly consistent with akpan et al. (2014) that used net fdi and found no significant effect between governance and the dependent variable. the rule of law estimate, which is consistently insignificant across contemporary and non-contemporary specifications, is contrary to jadhav (2012), who concluded that it plays a significant positive role in attracting fdi into the brics. given that we have enlarged the dataset, the insignificance could be traceable to the mint countries, the methodology of estimation and conditioning informationset or control variables. it should be noted that the present line of inquiry and jadhav (2012) have sample periodicities that are almost similar (2001-2011 versus 2000-2009, respectively). the favourable effects of regulation quality and government effectiveness from jadhav & katti (2012), who have used thesame periodicity as jadhav(2012), is confirmed only in contemporary specifications of the present study. only the negative effect of voice & accountabilityis confirmed in both contemporary and noncontemporary specifications. similarly, the positive effects ofpolitical stability,political governanceand general governanceare persistently significant in both contemporary and noncontemporary models. the reasons for these differences is the same as those presented for deviations from the findings of jadhav (2012), notably the addition of mint to the sample, estimation technique and data employed. 10 it is important to note that gross fdi is total investments on new inputs of capital while net fdi is the gross fdi that is adjusted for depreciation (or capital consumption). s. asongu / european journal of government and economics 8(2), december 2019, 161-188 179  table 8: non-contemporary determinants (panel fixed-and random-effects) panel a : gross fdi constant 2.103*** 2.78*** 0.410 -0.795 1.719** -1.196 1.848** -0.310 -0.857 -1.001 (0.000) (0.004) (0.511) (0.377) (0.010) (0.225) (0.033) (0.639) (0.288) (0.222) voice & accountability (-1) -0.777** ------------------ (0.021) political stability (-1) --0.908** ---------------- (0.017) political governance (1) ----0.802** -------------- (0.026) regulation quality (-1) ------0.748 ------------ (0.201) government effectiveness(-1) ---------0.890 ---------- (0.179) economic governance (-1) ----------0.427* -------- (0.069) rule of law (-1) -------------0.121 ------ (0.862) corruption-control (-1) --------------0.578*** ---- (0.007) institutional governance (-1) ----------------0.485* -- (0.089) general governance (-1) ------------------0.489** (0.016) natural resources (-1) 0.036 0.074** 0.052 0.045 0.049* 0.053 0.055* 0.036 0.039 0.055 (0.173) (0.014) (0.106) (0.217) (0.073) (0.132) (0.061) (0.292) (0.246) (0.105) infrastructure (-1) 0.003 0.003 0.007 0.011 0.003 0.012 0.004 0.010 0.011 0.012* (0.323) (0.339) (0.329) (0.176) (0.308) (0.110) (0.236) (0.233) (0.145) (0.090) inflation (-1) -0.051** -0.05*** -0.012 -0.006 -0.05** -0.005 -0.05** 0.0009 0.007 0.002 (0.011) (0.007) (0.190) (0.533) (0.013) (0.572) (0.017) (0.955) (0.610) (0.850) domestic credit (-1) 0.001 -0.000 0.011 0.021*** 0.004 0.023*** 0.002 0.019*** 0.020** 0.020** (0.778) (0.994) (0.170) (0.007) (0.366) (0.005) (0.678) (0.007) (0.012) (0.012) hauman test 7.767 5.196 17.40*** 10.983* 9.124 11.055* 8.670 10.194** 9.944* 17.83*** time effects no no yes yes no yes no yes yes yes log-likelihood -113.00 -139.056 -----119.311 ---126.621 ------ within variance 0.7136 0.632 ----0.706 --0.693 ------ between variance 0.472 1.823 ----0.618 --0.919 ------ within r² ----0.497 0.435 --0.442 --0.429 0.446 0.474 fisher ----8.011*** 6.827*** --6.949*** --6.729*** 7.025*** 7.541*** observations 81 81 81 81 81 81 81 81 81 81 s. asongu / european journal of government and economics 8(2), december 2019, 161-188 180  *, **, ***: significance levels of 10%, 5% and 1% respectively. the random effects specifications are not modelled with time-effects due to issues of degree of freedom. second, it is worthwhile comparing and contrasting contemporary and non-contemporary specifications in terms of significance & magnitude. on similarities: (i)voice & accountability, political stability political governance and general governance are consistently significant in both types of specifications (contemporary and non-contemporary); (ii) economic governance (regulation quality and government effectiveness) is only significant in contemporary specifications, while; (iii) institutional governance and corruption-control are exclusively significant in non-contemporary specifications. these comparisons are relevant for the timing of fdi location decisions or its targeting. for instance, while factors in (i) can be considered in the same year that the fdi flows are being targeted, as well as the preceding year, those indicators in (ii) and (iii) are exclusively relevant only in the planning of present and future fdis respectively. panel b : net fdi constant 117.108* 3.216 125.10* 121.89* 13.855 108.01* 17.688 121.013 1.721 119.448 (0.098) (0.923) (0.098) (0.098) (0.624) (0.018) (0.594) (0.112) (0.950) (0.113) voice & accountability (-1) -28.834 ------------------ (0.530) political stability (-1) ---0.535 ---------------- (0.965) political governance (1) ----2.438 -------------- (0.833) regulation quality (-1) ------1.474 ------------ (0.953) government effectiveness(-1) --------37.063 ---------- (0.151) economic governance (-1) ----------7.593 -------- (0.471) 21.148 rule of law (-1) ------------(0.394) ------ corruption-control (-1) ---------------17.232 ---- (0.505) institutional governance (-1) -----------------0.879 -- (0.919) general governance (-1) ------------------2.546 (0.749) natural resources (-1) -0.1902 0.859 -0.050 -0.080 0.902 0.194 0.906 -0.085 0.836 -0.002 (0.848) (0.394) (0.959) (0.926) (0.357) (0.774) (0.361) (0.941) (0.399) (0.997) infrastructure (-1) -0.1745 0.392*** -0.075 -0.068 0.394*** -0.001 0.375*** -0.135 0.371*** -0.051 (0.699) (0.001) (0.821) (0.832) (0.000) (0.997) (0.001) (0.704) (0.002) (0.878) inflation (-1) -0.010 -0.106 0.084 0.100 -0.127 0.148 -0.056 -0.187 -0.200 0.153 (0.970 (0.870) (0.787) (0.738) (0.848) (0.656) (0.931) (0.671) (0.775) (0.664) domestic credit (-1) -0.519 0.022 -0.655 -0.625 -0.058 -0.572* -0.034 -0.616 0.069 -0.629 (0.108) (0.927) (0.120) (0.111) (0.797) (0.092) (0.885) (0.120) (0.770) (0.126) hausman 13.123** 4.603 15.77*** 16.964*** 8.577 11.736** 6.434 19.33*** 9.085 13.928** time effects yes no yes yes no yes no yes no yes log-likelihood ---432.367 -----434.242 ---433.583 ---430.950 -- within variance --681.532 ----654.042 --672.085 --681.048 -- between variance --2769.32 ----1926.97 --2276.81 --1717.61 -- adjusted r² 0.347 --0.341 0.340 --0.346 --0.345 --0.342 fisher 12.262*** --12.124*** 12.109*** --12.241*** --12.21*** --12.13*** observations 81 81 81 81 81 81 81 81 81 81 s. asongu / european journal of government and economics 8(2), december 2019, 161-188 181  with respect to the magnitude of estimated coefficients in the contemporary model, the dominance of economic governance and its key components (regulation quality and government effectiveness) are consistent with the recent findings of oluwatobi et al. (2015). they have shown that these dimensions are the most effective governance dynamics for attracting innovation into africa. this inference is contingent on the hypothesis that fdi could also be a proxy for innovation (andrés et al., 2015, p.692). with regards to non-contemporary specifications, political stability and political governance are most relevant. two policy implications boldly standout: while economic governance matters most for present fdi location decisions, political governance is the most important factor for one-year future fdi targets. third, the reasons for bundling and unbundling governance dynamics which have partially motivated this line of inquiry have been confirmed in the analysis. they are more apparent in noncontemporary estimations. in contemporary estimations, we have observed that while the effect of political governance is positively significant, that of voice & accountability,which is one of its constituents, is not. this implies, foreign investors may look beyond voice & accountability and consider the ‘elections and replacement of political leaders’ all together in their fdi location decisions.the inference and policy implication applies to the interesting findings of non-contemporary specifications, notably: economic governance is significant while its components (regulation quality and government effectiveness) are not; institutional governance is significant while one of its components (rule of law) is not and general governance is significant while its components (the rule of law, government effectiveness and regulation quality) are not. the findings are consistent with asongu & nwachukwu (2016) in which lifelong learning (which is the consolidation of knowledge acquired during three-levels of education) has a higher effect on political stability than the individual independent effects of various educational channels. as a policy implication, established insights into the significant components of the political, economic and institutional governance reforms (as part of a structural adjustment program) could clarify the attractiveness of our brics and mint economies as a future destination for fdi. 5. conclusions we have assessed the drivers of fdi in a panel of brics (brazil, russia, india, china & south africa) and mint (mexico, indonesia, nigeria & turkey) countries for the period 2001-2011. we have bundled and unbundled governance determinants using a battery of contemporary and noncontemporary estimation techniques based on randomand fixed-effects regressions. we have also used a principal component analysis technique in amalgamating six governance dimensions into four dynamics. they comprise (i) political governance (voice &accountability and political stability), (ii) economic governance (regulation quality and government effectiveness), (iii) institutional governance (rule of law and corruption-control), and general governance (political, economic and institutional, governance dynamics). the following four broad general findings are established. first, while the majority of our governance determinants of gross fdi are significant, they are overwhelmingly insignificant for net s. asongu / european journal of government and economics 8(2), december 2019, 161-188 182  fdi. this is consistent with both contemporary and non-contemporary specifications. second, with respect to the contemporary specifications, the significance of the governance dynamics in increasing order of magnitude are as follows: general governance (0.561), political governance (0.595), economic governance (0.832), political stability (1.006), regulation quality (1.669) and government effectiveness (2.035). then in addition, while institutional governance and its corresponding components (rule of law and corruption-control) have insignificant effects, the contributions of political governance and its dimensions (voice &accountability and political stability) and economic governance and its elements (regulation quality and government effectiveness) are significantly different from zero. besides, the decision to bundle governance variables is justified by the effect of political governance which is significantly positive, although the effect of one of its components (voice &accountability) is significantly negative. third, in terms of non-contemporary relationships, we note that the significance of the governance dynamics in ascending order of magnitude are: economic governance (0.427), institutional governance (0.485), general governance (0.489), corruption-control (0.578), political governance (0.802) and political stability (0.908). further, while regulation quality and government effectiveness have insignificant separate effects, their combined impact as captured by the economic governance indicator is significantly positive at the ten percent confidence level. moreover, the motivation to blend governance variables is further demonstrated by the effects of political governance, economic governance and institutional governance. for example, political governance is significantly positive, while one of its components (voice &accountability) is significantly negative. economic governance is significantly positive, while its components (regulation quality and government effectiveness) are not. institutional governance is significantly positive while one of its components (rule of law) is not. fourth, the magnitude of the estimated coefficients in the non-contemporary model is all below one, indicating a decreasing impact of past governance reforms on subsequent fdi flows, even if the effect of political stability adjustment is the most persistent. policy implications have been discussed, notably: (i) the importance of governance reforms in both current and future fdi location decisions, (ii) the persistence of the impact of governance determinants on the real-time and one-period gross and net fdi flows and (iii) the extent to which a synchronized implementation of governance reforms could improve positive fdi location decisions. acknowledgement the author is indebted to the editor and reviewers for constructive comments. references abdioglu, n., khurshed, a., & stathopoulos, k., (2013). foreign institutional investment: is governance quality at home important? journal of international money and finance, 32(february), 916–940. doi: https://doi.org/10.1016/j.jimonfin.2012.08.001 abor, j. y., amidu, y., & issahaku, h., (2018). mobile telephony, financial inclusion and inclusive growth. journal of african business, 18(4), 430-453. doi: https://doi.org/10.1080/15228916.2017.1419332 s. asongu / european journal of government and economics 8(2), december 2019, 161-188 183  afutu-kotey, r. l. , gough, k. w., & owusu, g., (2017). young entrepreneurs in the mobile telephony sector in ghana: from necessities to aspirations. journal of african business, 18(4), 476-491. doi: https://doi.org/10.1080/15228916.2017.1339252 ajide, k. b, & raheem, i. d., (2016a). institutions-fdi nexus in ecowas countries. journal of african business, 17(3), 319-341. doi: https://doi.org/10.1080/15228916.2016.1180778 ajide, k. b, & raheem, i. d., (2016b). the institutional quality impact on remittances in the ecowas sub-region. african development review, 28(4), 462–481. doi: https://doi.org/10.1111/14678268.12224 akpan, u. s., isihak, s. r., & asongu, s. a., (2014). determinants of foreign direct investment in fast-growing economies: a study of brics and mint, african governance and development institute working paper, 14/002, yaoundé. doi: https://doi.org/10.2139/ssrn.2493410 aleksynska, m., & havrylchyk, o. (2013). fdi from the south: the role of institutional distance and natural resources. european journal of political economy, 29(march), 38–53. doi: https://doi.org/10.1016/j.ejpoleco.2012.09.001 amavilah, v., (2015). social obstacles to technology, technological change, and the economic growth of african countries: some anecdotal evidence from economic history, mpra paper no. 63273, munich. amendolagine, v., boly, a., coniglio, n. d., prota, f., & seric, a., (2013). fdi and local linkages in developing countries: evidence from sub-saharan africa. world development, 50(october), 4156. doi: https://doi.org/10.1016/j.worlddev.2013.05.001 andrés, a. r., & asongu, s. a. (2013). fighting software piracy: which governance tools matter in africa? journal of business ethics, 118(3), 667-682 doi: https://doi.org/10.1007/s10551-013-16207 andrés, a. r., asongu, s. a., & amavilah, v. h. s., (2015). the impact of formal institutions on knowledge economy. journal of the knowledge economy, 6(4), 682-703. doi: https://doi.org/10.1007/s13132-013-0174-3 anyanwu, j., (2013). determining the correlated of poverty for inclusive growth in africa. african development bank working paper, 181, tunis. anyanwu, j., (2012). why does foreign direct investment go where it goes? new evidence from african countries. annals of economics and finance, 13(2), 425-462. anyanwu, j., & augustine, d., (2013). gender equality in employment in africa: empirical analysis and policy implications. african development review, 25(4), 400-420. doi: https://doi.org/10.1111/1467-8268.12038 arezki, r., deininger, k., & selod, h., (2015). what drives the global land rush? the world bank economic review, 29(2), 207-233. doi: https://doi.org/10.1093/wber/lht034 asiedu, e., (2002). on the determinants of foreign direct investment to developing countries: is africa different? world development, 30(1), 107–119. doi: https://doi.org/10.1016/s0305750x(01)00100-0 asiedu, e., (2006). foreign direct investment in africa: the role of natural resources, market size, government policy, institutions and political instability. the world economy, 29(1), 63–77. doi: https://doi.org/10.1111/j.1467-9701.2006.00758.x asiedu, e., & lien, d. (2011). democracy, foreign direct investment and natural resources. journal of international economics, 84(1), 99–111. https://doi.org/10.1016/j.jinteco.2010.12.001 asongu, s. a., (2012), government quality determinants of stock market performance in african countries. journal of african business, 13(2), 183-199. doi: https://doi.org/10.1080/15228916.2012.727744 asongu, s. a., (2013a). fighting consumer price inflation in africa: what do dynamics in money, credit, efficiency and size tell us. journal of financial economic policy, 5(1), 39-60. doi: https://doi.org/10.1108/17576381311317772 asongu, s. a., (2013b). how would population growth affect investment in the future? asymmetric panel causality evidence for africa. african development review, 25(1), 14-29. doi: https://doi.org/10.1111/j.1467-8268.2013.12010.x asongu, s. a., (2014a). correcting inflation with financial dynamic fundamentals: which adjustments matter in africa? journal of african business, 15(1), 64-73. doi: https://doi.org/10.1080/15228916.2014.881231 asongu, s. a., (2014b). financial development dynamic thresholds of financial globalization: evidence s. asongu / european journal of government and economics 8(2), december 2019, 161-188 184  from africa. journal of economic studies, 41(2), 166 – 195. doi: https://doi.org/10.1108/jes-032012-0039 asongu, s. a., (2015). finance and growth: new evidence from meta-analysis”, managerial finance, 41(6), pp. 615-639. https://doi.org/10.1108/mf-09-2013-0249 asongu, s. a., (2016). determinants of growth in fast developing countries: evidence from bundling and unbundling institutions. politics & policy, 44(1), 97-134. doi: https://doi.org/10.1111/polp.12148 asongu, s. a., & aminkeng, g. a. a., (2013). the economic consequences of china-africa relations: debunking myths in the debate. journal of chinese economic and business studies, 11(4), 261277. https://doi.org/10.1080/14765284.2013.838384 asongu, s. a., & boateng, a., (2018). introduction to special issue: mobile technologies and inclusive development in africa. journal of african business, 19(3), 297-301. doi: https://doi.org/10.1080/15228916.2018.1481307 asongu, s. a., & kodila-tedika, o., (2016). fighting african conflicts and crimes: which governance tools matter? international journal of social economics: 43(5), 466-485. doi: https://doi.org/10.1108/ijse-11-2014-0233 asongu, s. a., le roux, s., nwachukwu, j. c., & pyke, c., (2019a). the mobile phone as an argument for good governance in sub-saharan africa. information technology & people. doi: https://doi.org/10.1108/itp-01-2018-0011 asongu, s.a. & nguena, c.l. (2015). equitable and sustainable development of foreign land acquisitions: lessons, policies, and implications. in e. osabuohien (ed.), handbook of research on in-country determinants and implications with foreign land acquisitions (pp.1-20). new york: igi global. doi: https://doi.org/10.4018/978-1-4666-7405-9.ch001 asongu, s. a., & nnanna, j., (2019). foreign aid, instability and governance in africa. politics & policy, 47(4), 807-848. https://doi.org/10.1111/polp.12320 asongu, s. a., & nwachukwu, j. c., (2016a). revolution empirics: predicting the arab spring. empirical economics, 51(2), 439-482. doi: https://doi.org/10.1007/s00181-015-1013-0 asongu, s. a., & nwachukwu, j. c., (2016b). the role of lifelong learning in political stability and non-violence: evidence from africa. journal of economic studies, 43(1), 141-164. doi: https://doi.org/10.1108/jes-06-2014-0087 asongu, s. a., & odhiambo, n. m., (2019a). financial access, governance and insurance sector development in sub-saharan africa. journal of economic studies, forthcoming. doi: https://doi.org/10.2139/ssrn.3431813 asongu, s. a., & odhiambo, n. m., (2019b). income levels, governance and inclusive human development in sub-saharan africa. applied research in quality of life. doi: https://doi.org/10.1007/s11482-019-09755-8 asongu, s. a., & odhiambo, n. m., (2019c). inequality thresholds, governance and gender economic inclusion in sub-saharan africa. international review of applied economics 34(1), 94114. doi: https://doi.org/10.2139/ssrn.3401332 asongu, s. a., & ssozi, j., (2016). sino-african relations: some solutions and strategies to the policy syndrome. journal of african business, 17(1), 33-51. doi: https://doi.org/10.1080/15228916.2015.1089614 asongu, s. a., tchamyou, v. s., asongu, n., & tchamyou, n., (2019b). fighting terrorism in africa: evidence from bundling and unbundling institutions. empirical economics, 56(3), 883–933. doi: https://doi.org/10.1007/s00181-017-1378-3 ba, j., & ng s., (2006). confidence intervals for diffusion index forecasts and inference for factoraugmented regression. econometrica, 74(4), 1133-1150. doi: https://doi.org/10.1111/j.14680262.2006.00696.x bai, j., (2003). “inferential theory for factor models of large dimensions”. econometrica, 71(1), 135173. doi: https://doi.org/10.1111/1468-0262.00392 bai, j., (2009). panel data models with interactive fixed effects. econometrica, 77(4), 1229-1279. doi: https://doi.org/10.3982/ecta6135 barro, r., (2003). determinants of economic growth in a panel of countries, annals of economics and finance, 4(2), 231-274. bartels, f. l., alladina, s. n., & lederer, s., (2009). foreign direct investment in sub-saharan africa: motivating factors and policy issues. journal of african business, 10(2), 141-162. doi: s. asongu / european journal of government and economics 8(2), december 2019, 161-188 185  https://doi.org/10.1080/15228910903187791 bartels, f. l. napolitano, f., & tissi, n. e., (2014). fdi in sub-saharan africa: a longitudinal perspective on location-specific factors (2003–2010). international business review, 23(3), 516529. doi: https://doi.org/10.1016/j.ibusrev.2013.08.013 bongomin, g. o. c., ntayi, j. m., munene j. c., & malinga, c. a., (2018). mobile money and financial inclusion in sub-saharan africa: the moderating role of social networks. journal of african business.18(4), 361-384. doi: https://doi.org/10.1080/15228916.2017.1416214 buchanan, b. g., le, q. v., & rishi, m., (2012). foreign direct investment and institutional quality: some empirical evidence. international review of financial analysis, 21(january), 81–89. doi: https://doi.org/10.1016/j.irfa.2011.10.001 busse, m., & hefeker, c. (2007). political risk, institutions and foreign direct investment. the european journal of political economy, 23(2), 397-415. doi: https://doi.org/10.1016/j.ejpoleco.2006.02.003 büthe, t., & milner, h. v., (2008). the politics of foreign direct investment into developing countries: increasing fdi through international trade agreements? american journal of political science, 52(4), 741–762. doi: https://doi.org/10.1111/j.1540-5907.2008.00340.x clapp, j., (2013). financialization, distance and global food politics. international conference, yale university, september 14-15. retrieved from: http://www.tni.org/sites/www.tni.org/files/download/5_clapp_2013.pdf cleeve, e., (2012). political and institutional impediments to foreign direct investment inflows to subsaharan africa. thunderbird international business review, 54(4), 469-477. doi: https://doi.org/10.1002/tie.21477 collier, p., (2008). the politics of hunger. how illusion and greed fan the food crisis. foreign affairs, 87 (6), 67-79. darley, w. k., (2012). increasing sub-saharan africa's share of foreign direct investment: public policy challenges, strategies, and implications. journal of african business, 13(1), 62-69. de janvry, a., & sadoulet, e., (2010). agricultural growth and poverty reduction: additional evidence. world bank research observer, 25 (1), 1-20. doi: https://doi.org/10.1093/wbro/lkp015 dunning, j. h. (1988). explaining international production. london: unwin hyman. dunning, j. h. (1993). multinational enterprises and the global economy. wokingham, berkshire: addison wesley. dunning, j. h. (2000). the eclectic paradigm as an envelope for economic and business theories of mne activity. international business review, 9 (2), 163-190. doi: https://doi.org/10.1016/s09695931(99)00035-9 economic commission for africa (2004). land tenure systems and their impacts on food security and sustainable development in africa, eca/sdd/05/09. faeth, i. (2009). determinants of foreign direct investment: a tale of nine theoretical models. journal of economic surveys, 23 (1), 165–196. doi: https://doi.org/10.1111/j.14676419.2008.00560.x fairbairn, m., (2013). ‘like gold with yield’: evolving intersections between farmland and finance. international conference, yale university, september 14-15. retrieved from: http://www.tni.org/sites/www.tni.org/files/download/6_fairbairn_2013.pdf fosu, a., (2013). growth of african economies: productivity, policy syndromes and the importance of institutions. journal of african economies, 22(4), 523-551. doi: https://doi.org/10.1093/jae/ejs034 gastanaga, v. m., nugent, j. b., & pashamova, b. (1998). host country reforms and fdi inflows: how much difference do they make? world development, 26(7), 1299-1314. doi: https://doi.org/10.1016/s0305-750x(98)00049-7 german, l., schoneveld, g., mwangi, e., (2011). processes of large-scale land acquisition by investors: case studies from sub-saharan africa. paper presented at the international conference on global land grabbing 6-8 april 2011. gosavi, a., (2018). can mobile money help firms mitigate the problem of access to finance in eastern sub-saharan africa. journal of african business. 18(4), 343-360. doi: https://doi.org/10.1080/15228916.2017.1396791 greenaway-mcgrevy, r., han, c., & sul, d., (2012). asymptotic distribution of factor augmented estimators for panel regression. journal of econometrics, 169(1), 48-53. doi: https://doi.org/10.1016/j.jeconom.2012.01.003 hajzler, c., (2014). “resource-based fdi and expropriation in developing economies”. journal of s. asongu / european journal of government and economics 8(2), december 2019, 161-188 186  international economics, 92(1), 24–146. doi: https://doi.org/10.1016/j.jinteco.2013.10.004 hennart, j. (1982). a theory of multinational enterprise. ann arbor: university of michigan. hubani, m., & wiese, m., (2018). a cashless society for all: determining consumers’ readiness to adopt mobile payment services. journal of african business, 18(4), 409-429. https://doi.org/10.1080/15228916.2017.1396792 ingwe, r., okoro, j., & ukwayi, j. k., (2010). the new scramble for africa: how large-scale acquisition of sub-saharan africa’s land by multinational corporations and rich countries threaten sustainable development. journal of sustainable development in africa, 12(3), 28-50. isakson, r., (2013). financialization and the transformation of agro-food supply chains: a political economy. international conference, yale university, september 14-15. retrieved form: http://www.tni.org/sites/www.tni.org/files/download/9_isakson_2013.pdf issahaku, h., abu, b. m., & nkegbe, p. k., (2018). does the use of mobile phones by smallholder maize farmers affect productivity in ghana? journal of african business, 19(3), 302-322. doi: https://doi.org/10.1080/15228916.2017.1416215 jadhav, p., (2012). determinants of foreign direct investments in brics economies: analysis of economic, institutional and political factors. procedia-social and behavioral sciences , 37, 5-14. doi: https://doi.org/10.1016/j.sbspro.2012.03.270 jadhav, p., & katti, v., (2012). institutional and political determinants of foreign direct investment: evidence from brics economies. poverty & public policy, 4(3), 49-57. doi: https://doi.org/10.1002/pop4.5 jolliffe, i. t. (2002) principal component analysis (2nd ed.) new york: springer. kaiser, h. f. (1974) an index of factorial simplicity. psychometrika 39(1), 31–36. doi: https://doi.org/10.1007/bf02291575 kangoye, t., (2013). does aid unpredictability weaken governance? evidence from developing countries. the developing economies, 51(2), 121-144. doi: https://doi.org/10.1111/deve.12008 kelsey, d., & le roux, s., (2017), dragon slaying with ambiguity: theory and experiments. journal of public economic theory, 19(1), 178–197. doi: https://doi.org/10.1111/jpet.12185 kelsey, d., & le roux, s., (2018). strategic ambiguity and decision-making. an experimental study. theory and decision, 84(3), 387–404. doi: https://doi.org/10.1007/s11238-017-9618-8 kinda, t., (2010). investment climate and fdi in developing countries:firm-level evidence. world development, 38(4), 498-513. doi: https://doi.org/10.1016/j.worlddev.2009.12.001 kindlerberger, c. (1969). american business abroad: six lectures on direct investment. new haven: yale university press. kolstad, i., & wiig, a., (2011). better the devil you know? chinese foreign direct investment in africa, journal of african business, 12(2), 31-50. doi: https://doi.org/10.1080/1536710x.2011.555259 lay, j., & nolte, k., (2014). land grabs or simply investments? determinants of foreign large scale land acquisitions in developing countries, giga german institute of global and area studies, hamburg, germany and university of göttingen. lehnert, k., benmamoun, m., & zhao, h. (2013). fdi inflow and human development: analysis of fdi’s impact on host countries’ social welfare and infrastructure. thunderbird international business review , 55 (3), 285-298. doi: https://doi.org/10.1002/tie.21544 levine, r., (2005). finance and growth: theory and evidence. in philippe aghion & steven durlauf (eds.), handbook of economic growth, (1st edition, pp. 865-934). doi: https://doi.org/10.1016/s1574-0684(05)01012-9 lipton, m., (2009). land reform in developing countries: property rights and property wrongs. new york: routledge. doi: https://doi.org/10.4324/9780203876251 liu, p., (2013). impacts of foreign agricultural investment on developing countries: evidence from case studies. food and agricultural organization. retrieved from: file:///c:/users/user/downloads/liu%20(1).pdf. doi: https://doi.org/10.4324/9780203876251 loayza, n. v., & raddatz, c., (2010). the composition of growth matters for poverty alleviation. journal of development economics, 93 (1), 137-51. doi: https://doi.org/10.1016/j.jdeveco.2009.03.008 mckenzie c, r., & mcaleer m (1997). on efficient estimation and correct inference in models with generated regressors: a general approach. japanese economic review, 48(4), 368-389. doi: https://doi.org/10.1111/1468-5876.00060 s. asongu / european journal of government and economics 8(2), december 2019, 161-188 187  minkoua nzie, j. r., bidogeza, j. c., & ngum, n. a., (2018). mobile phone use, transaction costs, and price: evidence from rural vegetable farmers in cameroon. journal of african business, 19(3), 323342. doi: https://doi.org/10.1080/15228916.2017.1405704 muthinja, m. m., & chipeta, c., (2018). what drives financial innovations in kenya’s commercial banks? an empirical study on firm and macro-level drivers of branchless banking. journal of african business, 18(4), 385-408. doi: https://doi.org/10.1080/15228916.2017.1405705 moosa, i. (2002). foreign direct investment: theory, evidence and practice. london: palgrave. doi: https://doi.org/10.1057/9781403907493 moosa, i. a., & cardak, b. a. (2006). the determinants of foreign direct investment: an extreme bounds analysis. journal of multinational financial management, 16(2), 199–211. doi: https://doi.org/10.1016/j.mulfin.2005.07.002 neumayer, e., & spess, l., (2005). do bilateral investment treaties increase foreign direct investment to developing countries? world development, 33(10), 1567–85. doi: https://doi.org/10.1016/j.worlddev.2005.07.001 okoth-ogendo, h. w. (2008). the last colonial question: an essay in the pathology of land administration systems in africa. in helge onsrud and elena busch (eds.) norwegian land tools relevant to africa report, hpnefoss: norwegian mapping and cadastre authority. oluwatobi, s., efobi, u., olurinola, i., & alege, p., (2015). innovation in africa: why institutions matter? south african journal of economics, 83(3), 390-410. doi: https://doi.org/10.1111/saje.12071 osabuohien, e. s., (2014). land-scale agricultural land investments and local institutions in africa. land use policy, 39(july), 155-165. doi: https://doi.org/10.1016/j.landusepol.2014.02.019 osabuohien, e. (ed.) (2015). handbook of research on in-country determinants and implications with foreign land acquisitions. new york: igi global. doi: https://doi.org/10.4018/978-1-4666-7405-9 oxley l, & mcaleer, m., (1993). econometric issues in macroeconomic models with generated regressors. journal of economic surveys, 7(1), 1-40. doi: https://doi.org/10.1111/j.14676419.1993.tb00158.x pagan, a., (1984). econometric issues in the analysis of regressions with generated regressors. international economic review, 25(1), 221-247. doi: https://doi.org/10.2307/2648877 pelizzo, r., araral, e., pak, a., & xun, w., (2016). determinants of bribery: theory and evidence from subsaharan africa. african development review, 28(2), 229-240. doi: https://doi.org/10.1111/1467-8268.12192 pelizzo, r., & nwokora, z., (2016). bridging the divide: measuring party system change and classifying party systems. politics & policy, 44(6), 1017-1052. doi: https://doi.org/10.1111/polp.12188 pelizzo, r., & nwokora, z., (2018). party system change and the quality of democracy in east africa. politics &policy, 46(3), 505-528. doi: https://doi.org/10.1111/polp.12255 pesaran, m. h., (2006). estimation and inference in large heterogeneous panels with a multifactor error structure. econometrica, 74 (4), 967-1012. doi: https://doi.org/10.1111/j.14680262.2006.00692.x ranjan, v., & agrawal, g. (2011). fdi inflow determinants in bric countries: a panel data analysis. international business research, 4(4), 225-263. doi: https://doi.org/10.5539/ibr.v4n4p255 rogmans, t., & ebbers, h. (2013). the determinants of foreign direct investment in the middle east north africa region. international journal of emerging markets, 8(3), 240-257. doi: https://doi.org/10.1108/17468801311330310 rolfe, r. j., & woodward, d. p., (2004). attracting foreign investment through privatization: the zambian experience. journal of african business, 5(1), 5-27. doi: https://doi.org/10.1300/j156v05n01_02 sahoo, p., dash, r., & nataraj, g.,. (2010). infrastructure development and economic growth in china. ide discussion paper no. 261, chiba. sekkat, k., & veganzones-varoudakis, m.-a., (2007). openness, investment climate, and fdi in developing countries. review of development economics, 11(4), 607–620. doi: https://doi.org/10.1111/j.1467-9361.2007.00426.x stock, j. h. & m. w. watson (2002). forecasting using principal components from a large number of predictors. journal of the american statistical association, 97(460), 1167-1179. doi: https://doi.org/10.1198/016214502388618960 s. asongu / european journal of government and economics 8(2), december 2019, 161-188 188  tchamyou, v. s., (2017). the role of knowledge economy in african business. journal of the knowledge economy, 8(4), 1189-1228. doi: https://doi.org/10.1007/s13132-016-0417-1 tchamyou, v. s., (2019). education, lifelong learning, inequality and financial access: evidence from african countries. contemporary social science. doi: https://doi.org/10.1080/21582041.2018.1433314 thaler, k., (2013). large-scale land acquisitions and social conflict in africa. international conference, yale university, september 14-15. retrieved from: http://www.tni.org/sites/www.tni.org/files/download/22_thaler_2013.pdf tuomi, k., (2011). the role of the investment climate and tax incentives in the foreign direct investment decision: evidence from south africa. journal of african business, 12(1), 133-147. doi: https://doi.org/10.1080/15228916.2011.555279 united nations (2010, january). foreign land purchases for agriculture: what impact on sustainable development? sustainable development and innovation briefs issue no. 8, new york. unctad. (2002). world investment report: transnational corporations and export competitiveness. new york: united nations. undp. (2013). human development reports. retrtieved from: : http://www.undp.org/content/dam/undp/library/corporate/hdr/2013globalhdr/english/hdr2013% 20report%20english.pdf(accessed: 15/07/2018). vernon, r., (1966). international investment and international trade in the product cycle. quarterly journal of economics, 80(2), 190-207. doi: https://doi.org/10.2307/1880689 vijayakumar, n., sridharan, p., & rao, k. c., (2010). determinants of fdi in brics countries: a panel analysis. international journal of business science and applied management, 5(3), 1-13. wei, s.-j. (2000). how taxing is corruption on international investors? review of economics and statistics, 82(1), 1–11. doi: https://doi.org/10.1162/003465300558533 world bank. (2007). agriculture for development. (world development report 2008). washington, d.c.: world bank and oxford university press. world bank. (2013). world development indicators. (uk data service). world bank, washington, d.c. wouterse, f., deininger, k., selod, h., badiane, o., swinnen, j., von braun, j., & zilberman, d., (2011). foreign direct investment in land in west africa: the status quo, lessons from other regions, implications for research. west & central africa office thematic research note 1. westerlund, j., & urbain, j-p., (2013a). on the estimation and inference in factor-augmented panel regressions with correlated loadings. economic letters, 119(3), 247-250. doi: https://doi.org/10.1016/j.econlet.2013.03.022 westerlund, j., & urbain, j-p., (2013b). on the implementation and use of factor-augmented regressions in panel data. journal of asian economics, 28(october), 3-11. doi: https://doi.org/10.1016/j.asieco.2013.02.002 westerlund, j., & urbain, j-p., (2012). cross-sectional averages versus principal components. maastricht university, rm/11/053, maastricht. yin, j. z., & vaschetto, s., (2011). china’s business engagement in africa. the chinese economy, 44 (2), 43-57. doi: https://doi.org/10.2753/1097-1475440203 economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt vol.8 • no.1 2019 issn: 2254-7088 european journal of government and economics 8(1), june 2019. european journal of government and economics issn: 2254-7088 number 8, issue 1, june 2019 doi: https://doi.org/10.17979/ejge.2018.8.1 the evaluation index improvement for the relocation of research equipment 6-29 donghun yoon residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) 30-47 josé ramón cardona, daniela álvarez bassi, maría dolores sánchez-fernández social capital and banking system profitability: a survey of european union countries 48-62 arash nayebyazdi governance and domestic investment in africa 63-80 chimere okechukwu iheonu economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt 81-95 marwa shibl biltagy give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia 96-109 jaroslav mihálik, peter horváth, martin švikruha https://doi.org/10.17979/ejge.2018.8.1 european journal of government and economics 8(1), june 2019, 81-95 european journal of government and economics issn: 2254-7088 economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt marwa shibl biltagya* a cairo university, egypt *corresponding author at: bilmarwa@feps.edu.eg article history. received 5 october 2018; first revision required 24 october 2018; accepted 18 march 2018. abstract. education provides good health, empowerment and employment to all people in the society. accordingly, education can make a lifelong difference in individuals’ lives. it is proved that, on average, there is a positive relationship between each additional year of schooling and the income of an individual. the main objective of this paper is to estimate the demand and supply functions for schooling in egypt using the data of the egypt labor market panel survey 2012. it is estimated that the private rate of return to education is 4.5%. moreover, the rate of return to the number of years of experience is 2.6%, i.e. each additional year of experience will increase the earnings of an individual by 2.6% on average. it is concluded that individuals from poor families have an increased likelihood of receiving financial aid, which decreases their discounting rate of interest. keywords: investment in education, marginal rate of return, supply functions for schooling, skill mismatch. jel classification: i21, i26 doi. https://10.17979/ejge.2019.8.1.4574 1. introduction education affects the lives of individuals, their participation in economic activities, and overall economic development in various ways. this paper, however, will focus on the economic returns to education and its opportunity cost. human capital investment in the form of education has an important economic value. the returns to investment in education are generally estimated by micro and macro levels. investment in education has many benefits on both individual and social levels as well (lucas, 1988). this means that there are two types of returns to investment in education; namely, social and private rate of return to education. at the individual level, each additional year of schooling will increase the individual's earnings (mincer, 1974). there is a kind of spillover from the individual investment in human capital (i.e. education) into the social level. externalities of investment in schooling include, for example, low mailto:bilmarwa@feps.edu.eg https://10.0.70.59/ejge.2019.8.1.4574 http://en.wikipedia.org/wiki/economic_value marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 82 levels of crime, high levels of democracy and high levels of political participation. these kinds of externalities are too difficult to be measured; accordingly, the returns to investment in education are underestimated. furthermore, becker (1994) assumes that there are two kinds of schooling costs; namely, direct and indirect costs. the direct costs consist of costs of teachers, administration staff, books, and other fees. the foregone earnings while in school represent the indirect costs of schooling. he assumes that the direct costs of schooling can be estimated easily. however, foregone earnings are too difficult to be estimated. the main objective of this paper is to estimate the demand and supply functions for schooling in egypt, using the data of the egypt labor market panel survey 2012 (elmps 2012). the methodology is based on the mincerian estimation of the earnings-schooling model. it can be said that individuals invest in schooling until the marginal rate of return to schooling equals the discounting rate of interest. this paper is organized as follows; the second part surveys the related literature review. the third section presents the data description and the theoretical framework of the marginal rate of return and supply functions for schooling. part four proposes the empirical model, estimation and the results and fifth part concludes. 2. literature review the issue of the rate of return to education has dominated research in the area of economics of education for the last several decades based on the work of schultz (1961) and becker (1962). more precisely, the concept of human capital was first introduced by mincer (1958) and then elaborated by two nobel prize winners, schultz (1961) and becker (1962). it emphasizes that people get information and skills by education and on-the-job training to increase their earnings in the future. caudill and mixon (2012) stated that people increase their investment in human capital when the returns on other types of capital decrease. the private rate of return to education can be estimated by two methods; either by the internal rate of return to education or by the mincerian earnings function, introduced initially by mincer (1974). the internal rate of return is called the full or elaborate method of estimating the demand functions of schooling. it can be defined as the discount rate that equates the net present value of schooling for an individual at a given point of time. psacharopoulos (1994) presented a survey, which included the results related to estimating the rate of return to education for about 70 countries. the rate of return to education is particularly high when the supply of educated labor is rather scarce. psacharopoulos (1994) and psacharopoulos and patrinos (2002) concluded that primary education yields a higher rate of return than other advanced stages of schooling since the expansion of primary education, in particular, reduces the marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 83 level of income inequality. also, education may create more job opportunities and hence decreases the rate of unemployment. moreover, in general, private rates of return are higher than social rates of return because education has large impacts on earnings, employment in addition to non-market private effects. menzies et al. (2000) stated that there is a kind of spillover from the individual investment in human capital (i.e. education) into the social level. these kinds of externalities are too difficult to be measured; accordingly, the social return to investment in education is usually underestimated. ashenfelter and rouse (1998) estimated a model of schooling investment using data on 700 identical twins. the empirical results indicated that higher ability individuals reach more schooling levels because of the advantage of lower marginal costs. the estimated model implies that genetically identical individuals have the same levels of schooling investment. the results suggested that the rate of return to schooling is about 9%. moreover, it is concluded that, the marginal benefits of schooling decrease with the attainment of higher levels of education and there is a negative relationship between the level of ability and the marginal costs of schooling. regan et al. (2006) derived the demand and supply functions of schooling using data from the national longitudinal survey of youth 1979 (nlsy79). the authors estimated the demand and supply functions for schooling depending on the earnings-schooling relationship. according to the analysis of regan et al. (2006), individuals invest in schooling until the marginal rate of return to schooling equals the discounting rate of interest. the results of estimating these equations specified that the discounting rates of interest are lower for individuals from wealthier families. moreover, the results showed that individuals with higher ability pass through school faster. the estimation of the model indicated that the marginal rate of return to schooling is 9.6 % per year. gasparini et al. (2011) analyzed the wage differentials and trends in the demand and supply of employees by educational level for some countries in latin america over the decades of the 1990s and the 2000s. they explained the decline in income inequality in the specified countries by the labor force educational improvement. the authors found that the demand factors are more powerful than the supply factors in explaining the decrease in wage premia for tertiary educated individuals. furthermore, they clarified that the change in labor demand can be attributed to some factors, such as, preferring the low skilled workforce in some industries, technological transmission or skill mismatch that may reduce the labor productivity of highly-educated people. psacharopoulos and patrinos (2018) highlighted the newest developments of returns to investment in education based on human capital theory depending on a record of 139 countries. they emphasized that the yearly private average global rate of return to one extra year of schooling is about 9%. they argued that private returns to higher education have increased over time and females continue to have higher average rates of return to schooling. moreover, the returns to education are higher in developing countries. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 84 3. skill mismatch: some important facts educational institutions become a source of human resource development. however, the educational sector in egypt faces many difficulties and challenges. among the challenges facing educational institutions is the issue of incompatibility between outputs of higher education on the one hand and the requirements of the labor market on the other, which resulted in the existence of large numbers of graduates without work (high rates of unemployment). the relationship between the outputs of the educational process and the requirements of the labor market is one of the strategic issues that affect the development in egypt in terms of its negative repercussions on the unemployment rates, labor productivity and structural imbalances experienced by labor markets in egypt. the problem of unemployment in the egyptian economy can be explained not only by the quantitative gap between the size of educational outputs and available job opportunities but also by the failure of educational institutions to prepare graduates capable of responding to the requirements of the labor markets and competition at local and international levels. 1. the reasons for the gap between the outputs of higher education and the requirements of the labor market (biltagy, 2013). education policies and plans are adopted without taking into account the suitability of these plans and policies for the requirements of the labor market, which contributed to the existence of a real gap. for example, when the labor market is saturated with a number of disciplines such as legal and commercial studies, higher education institutions still accept large numbers of students in these disciplines, which causes unemployment amongst university graduates. moreover, the current curricula and courses are among the reasons that led to the incompatibility between the requirements of the labor market and the outputs of the educational process. the difficulty of addressing the problem of rising unemployment rates among young graduates and low levels of labor productivity is a multidimensional issue. the determination of the size of employment opportunities and their sectoral and vocational distribution is mainly based on the package of economic and social policies contained in development plans and reform programs. thus, employment opportunities are based on economic growth rates, income distribution issues, investment trends, social welfare levels, and export development policy. the outputs of the educational process are based on the objectives of providing basic education for the largest number of population, increasing the enrollment rates in education, quality assurance and accreditation policies, stimulating the demand for technical education and diversifying the disciplines, in order to achieve the desired structural balance in the supply of educational services. accordingly, a comprehensive study of the link between supply forces and demand trends in the labor market is necessary. in addition to the above, the mismatch between supply and demand in the labor market reflects marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 85 a number of factors, including the inability of many university graduates to obtain a job opportunity in the fields they studied, particularly graduates of law, commercial and agricultural studies. furthermore, employers require graduates with technical knowledge in the field of work, and who also have initial communication skills, teamwork, problem solving and adaptability to the work environment. it is important to enhance the curriculum, provide advanced practical training and reconsider the student admission policy in higher educational institutions, to keep pace with the dynamics of labor markets and to formulate a distinct personality for the student. it can be said that technical factors are among the most important aspects in the inadequacy of university graduates for the requirements of the labor market, such as poor english language, lack of computer knowledge, lack of expertise, and weak analytical abilities. as for higher education patterns, some theoretical disciplines in university education, whose graduates suffer from the absence of suitable employment opportunities in both the public and private sectors, should be reduced. figure 1 illustrates some of the difficulties facing higher education institutions. figure 1: some difficulties facing higher education institutions in achieving the needed alignment with the labor market. source: own elaboration based on the survey of idsc, 2012. reasons of mismatch abundance in graduates of theoretical faculties the limited capacity of applied sciences faculties no available accurate data for the labor market needs labor market demands are variable and unstable marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 86 the graduates of social and human sciences reached 78.1% of total graduates in 2010/2011 i.e. the graduates of faculties of commerce, arts and law represented 28.9%, 21.1% and 16.7%, respectively of the total graduates (the survey of the incompatibility of university graduates' skills and labor market requirements for commercial, legal and agricultural studies, 2012). the following figure illustrates the main reasons for the discrepancy between the outputs of education and the requirements of the labor market, according to the views of graduates. figure 2: the percentage distribution of the causes of imbalances in the educational system. source: own elaboration based on the survey of idsc, 2012. in summary, the main causes of imbalances and skill mismatch in the egyptian educational system are, the low internal efficiency of educational institutions, due to the curricula and the type of tests that depend mainly on memorization, in addition to low quantitative and qualitative external efficiency, represented in the graduation of huge numbers of graduates in disciplines not needed by the labor market, with a deficit in other specialties. 2. suggested strategies to achieve the desired alignment between educational institutions and labor market requirements strategy 1: to develop the main elements of the educational process in terms of objectives, curriculum, content, practical experiences, advanced assessment methods and the educational environment through: 0 5 10 15 20 25 30 35 % marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 87 • reviewing the policy of students’ admission in universities. • conducting a comprehensive study of the needs of the labor market and then arranging specialties according to its importance and applicability. • appraising existing educational frameworks and developing new policies to provide students with a range of social, personal and technical skills that help them join the productive labor market. • expanding the use of modern and advanced teaching techniques, focusing on computer skills and proficiency in foreign languages. • providing appropriate educational buildings and achieving the principle of equal opportunities in education for all students in different geographical regions. • linking educational institutions with advanced educational research centers in order to develop curricula according to the needs of the labor market. • increasing the orientation of students towards the required scientific disciplines in the labor market. strategy 2: prepare a teacher adapted to the needs of the labor market by: • updating the systems of preparing the teacher, in order to improve his professional and social level and enable him to acquire knowledge and skills in various scientific fields. • it is not sufficient to focus only on the mental skills of the students, but it is important to take into consideration the emotional side, including values and principles. this depends on the good model of the lecturer. strategy 3: provide the necessary funding resources for the development of the educational institutions through (biltagy, 2015): • adopting certain policies to increase the government financial resources directed to the educational sector. • increasing the participation of civil society and businesspeople in support of government efforts. • rationalizing the principle of free education for all, rich and poor. • related to higher education, it is useful to transfer egyptian universities to the model of productive universities, by transforming their academic units into productive research units in various fields of work and services. • funding technical education and vocational training, partly by governments, but most of the funding must come from the beneficiaries, which would give them an important role in preparing study programs and training to meet the requirements of the labor market. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 88 4. marginal rate of return and supply functions for schooling: theoretical framework and data description as mentioned above, there are two kinds of economic rate of return to schooling, i.e. social and private. it is more difficult to estimate the social rate of return to education because of the complexity of capturing and calculating the externalities of education1. however, the relation between education and increased earnings is explained in the literature, depending on the mincerian earnings function. figure (3) provides a general overview of the economic returns to education. figure 3: economic rate of return to education. source: own elaboration based on becker (1962). the social rate of return to schooling includes public expenditure on education, in addition to the private cost considered for the calculation of the private rate of return. accordingly, the social rate of return takes into consideration the total cost of education. in general, the social rate of return is higher in primary education, if compared to secondary and higher education because public expenditure per student tends to increase significantly with the level of education. the analysis is conducted using the egypt labor market panel survey 2012 (elmps 2012). the elmps 2012 is the third round of the longitudinal survey, which was also done in 1998 and 2006. the elmps was presented by the economic research forum (erf) in cooperation with the central agency for public mobilization and statistics (capmas) in egypt. the elmps 2012 intends 1 foster and rosenzweig (1995) ascertain that there are positive externalities of education on other people in the neighborhood in addition to the externalities on the educated individual’s own family. economic returns to education social private externalities of education higher productivity for example: better health, lower mortality of children, lower population growth and lower crime rates increased earnings marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 89 to analyze the characteristics of the egyptian labor market. it is considered as a follow-up survey to the same households that were interviewed in 2006, in addition to a new sample, which was selected from a random sample to participate in the survey, to be able to analyze the evolution of the labor market in egypt over time. for the purpose of this study, the size of the total sample is 7573 observations. the sample contains waged workers whose ages range from 15 to 64 years. those individuals answer all the questions needed for the estimation of basic earnings function and the supply functions for schooling. this section includes the descriptive statistics of all variables used in the model together with a brief description of each variable. table 1 presents the variables, which are used in estimating the demand and supply functions for schooling as follows, • age: the age ranges from 15 to 64 years old; the mean value of age is 36.7 years. • w: this variable refers to the total wage of the individual. the minimum and maximum values of the total wage in this sample are 240 and 54000 egyptian pounds per month, respectively and the mean value is l.e. 3545 per month. • s: the variable s represents the number of years of schooling of an individual. the mean value of this variable is 11.8 years. the variable s takes values between 0 and 16, where the value 0 refers to illiterates and the value 16 refers to the university education2. • experience (t): this variable stands for the number of years of experience. the mean value of this variable is 16.24 years. • fsl: this variable symbolizes the father’s schooling level. the mean value of this variable is roughly 6 years; similarly, msl corresponds to the mother’s schooling level. the mean value of this variable is around 3 years. • n: this variable represents the family size. the variable n is used in estimating the supply function for schooling. the mean value of this variable is approximately 5 persons. it can be noticed from table 1 that, the percentage of males in the sample is 77.2% while the percentage of females is 22.8%. in addition, the table ascertains that 78% of the individuals in the sample used are married and 56% of those individuals live in urban areas. furthermore, 57% of the individuals in the sample used are employed in government and public enterprises. moreover, the average number of working days is roughly 6 days per week and the average number of working hours is 8.35 hours per day. 2 this paper follows the study of bratsberg and terrell (2002) in defining the variable of the number of years of schooling. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 90 table 1: descriptive statistics of the variables variable no. of observations mean std. dev. min max age 7573 36.7 10.87 15 64 gender 7573 0.772 0.419 0 1 w 7573 3545.3 3675.5 240 54000 s 7573 11.79 4.47 0 16 ss 7573 159.2 81.44 0 256 experience 7573 16.24 11.17 0 58 urban/rural 7573 0.56 0.49 0 1 marital status 7573 0.78 0.42 0 1 sector of emp. 7573 0.57 0.49 0 1 n. of days/w 7573 5.70 0.84 1 7 n. of hours/day 7573 8.35 2.28 1 24 n 7573 4.45 1.82 1 21 fsl 7573 5.58 5.63 0 20 msl 7573 2.91 4.84 0 20 source: own elaboration based on elmps 2012. 5. empirical model, estimation and results this paper derives the demand and supply functions for schooling depending on the earningsschooling relationship3. it can be said that individuals invest in schooling until the marginal rate of return to schooling equals the discounting rate of interest. the study considers individuals who earned approximately 300 egyptian pounds per month. the dependent variable in the log earnings function is the log of an individual's total wage. the marginal rate of return to schooling is, s w ∂ ∂ = lnδ , [1] where jδ = f (sj). [2] and s is the number of years of schooling for an individual. this implies that the first derivative of the log earnings function with respect to schooling yields an individual’s demand function for schooling, i.e. the rate of return to education (the demand function for schooling) is a function in the number of years of schooling. the previous studies, such as, schultz (1989), ashenfelter and krueger (1994), glewwe (1996), menzies et al. (2000) and boero et al. (2003) ascertained that the earnings of an individual depend 3 the empirical framework follows mincer’s estimation of the schooling model. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 91 on many other factors, for example, whether an individual lives in urban or rural areas, whether he or she works in the government sector or private sector. in addition, the earnings differ depending on gender differences. moreover, the previous studies showed that earnings differ from one person to another, depending on the number of working days per week and the number of working hours per day. the earnings also depend on the marital status of an individual; that is, marriage makes individuals more productive. the following earnings function considers all these factors. jjjjj uxttsw 14 2 3210ln +′++++= θθθθθ , [3] where t signifies the number of years of experience, x represents a vector of variables that affects the earnings of an individual and u1 is ∼ iid n (0, 2 1σ ). the results of the estimation of this function are illustrated in table 2. any individual seeks to maximize the present value of his/her lifetime earnings over time, and it can be represented by the log form as follows, 𝑃𝑃 = 𝑙𝑙𝑙𝑙𝑙𝑙 − 𝑖𝑖𝑖𝑖 − 𝑙𝑙𝑙𝑙𝑖𝑖 [4] where p is the present value of lifetime earnings and i is the discounting rate of interest. the following first-order condition can be obtained by taking derivatives with respect to s, δ= i. [5] an individual’s supply function for schooling investment can be derived by using the present value function as defined in [4]. the discounting rate of interest can be defined as a function of an individual’s family characteristics (i.e. family income levels and family size). the individual's supply function for schooling can be written as, jjmjfjj unssi 23210 ++++= ββββ , [6] where fs and ms are the levels of father's and mother's schooling, n denotes the family size and u2 is ∼ iid n (0, 2 2σ ). in equation [6], the family income levels can be represented by the schooling levels of an individual's parents. as shown in table 2, it is estimated that the private rate of return to education is 4.5% per year, noting that, private rate of return to education = (exp coefficient – 1) * 100. moreover, the rate of return to the number of years of experience is 2.6% i.e. each additional year of experience will increase the earnings of an individual by 2.6% on average. since θ3 is negative, the earnings will increase over time by decreasing rate. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 92 table 2: the results of estimation of earnings function lnw coefficient std. err. t p> ׀t׀ s 0.0442086 0.0018834 23.47 0.000 experience 0.0262104 0.0023612 11.10 0.000 expexp -0.0002842 0.0000549 -5.18 0.000 gender 0.2554301 0.0189975 13.45 0.000 marital status 0.1105411 0.0196645 5.62 0.000 urban/rural 0.1927933 0.0149652 12.88 0.000 sector of emp. 0.0099666 0.0175634 0.57 0.570 n. of days/week -0.0104486 0.0088547 -1.18 0.238 n. of hours/day 0.0270247 0.0034664 7.80 0.000 cons 6.513099 0.0708414 91.94 0.000 source: author’s calculations based on elmps 2012. gender has a significant effect on the earnings of an individual. this means that males get more monthly earnings than females by 29.1% on average. furthermore, keeping other factors constant, the individual who lives in urban areas gains more earnings than the one who lives in rural areas by 21.3%. the marital status of an individual plays a significant role in determining his/ her earnings. the married individual gets more earnings than a single person by 11.7%. in addition, the results show that, at 5% significance level, the variable of the sector of employment is insignificant, while working for long hours per day is associated with higher levels of earnings for an individual. as mentioned above, the supply function for schooling (the discounting rate of interest for individual j, ij) is a function in an individual's family characteristics. the results of the estimation of the supply function for schooling are demonstrated in table 3. table 3: the results of estimation of supply functions for schooling i coefficient std. err. t p> ׀t׀ fsl 0.0112895 0.0017748 6.36 0.000 msl 0.0106348 0.0020655 5.15 0.000 n -0.0213314 0.0042941 -4.97 0.000 source: author’s calculations based on elmps 2012. table 3 shows that the variables that represent the family income (fsl and msl) and the family size, n are significant at a 5% significance level. the supply function for schooling, i, can vary among individuals, for example, the discounting rate of interest would be higher for individuals from poor families. the same could be said for individuals from larger families as compared to individuals from smaller families. there is an indirect effect of family income on the discounting rate of interest of an individual via financial aid, then; individuals from poor families have an increased likelihood of receiving financial aid, which decreases their discounting rate of interest. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 93 6. conclusions and policy recommendations there is a positive relationship between the number of years of schooling and the private rate of return to schooling. it is estimated that the private rate of return to education is 4.5%. the results ascertain that working in urban areas is associated with more earnings for an individual. moreover, married individuals get more earnings because they are more productive. the model proved that, the main independent variable that is contained in the individual's demand function for schooling is the number of years of schooling, s. on the other hand, the main independent variables that are included in the individual's supply function for schooling in egypt are: the father's schooling level (fsl), the mother's schooling level (msl) and the family size (n). the best possible schooling level is determined when the demand and supply functions for schooling are equal. the rate of return to education is considered one of the most important determinants in the decision-making process of investment in education. egypt should reallocate its public resources in favor of education, especially the primary stage. government expenditure on education has grown remarkably during the past few years. the amount of the government's budget directed to education increased from l.e. 40 billion in 2009/10 to l.e. 64.5 billion in 2012/13. moreover, the budget in 2014/15 assigned l.e. 94 billion and 355 million to the education sector with an increase of l.e. 11 billion, compared to the amount devoted to education in 2013/14 (mof, 2014). in 2016/2017, the state public expenditure on education was l.e. 103 billion and 962 million, which represents 10.7% of the state total public expenditure (egypt in figures, 2018). the participation of the private sector and civil society is also important in developing the education sector in egypt. moreover, it is essential to put controls on the huge population increase in egypt because of the direct relationship between the family size and the discounting rate of interest. this paper demonstrates that family background factors are fundamental in determining the education decisions. this coincides with coleman (1966). he showed that family background factors are very crucial in explaining different levels of academic achievement among individuals. accordingly, it is important to increase the level of schooling of an individual's parents and to increase the real income of the family because there is a positive relationship between the family income levels and the level of schooling. acknowledgments this work has benefited from a financial grant from euro-mediterranean network for economic studies (emnes), financially supported by the european commission. marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 94 references ashenfelter, o. & krueger., a. (1994). estimates of the economic return of schooling from a new sample of twins, american economic review, 84(5). ashenfelter, o. & rouse. c. (1998). income, schooling and ability: evidence from a new sample of identical twins, quarterly journal of economics, 113(1). doi: https://doi.org/10.3386/w6106 becker, g. (1962). investment in human capital: a theoretical analysis, journal of political economy, 70. doi: https://doi.org/10.1086/258724 becker, g. (1994). human capital: a theoretical and empirical analysis, with special reference to education, (3rd ed.), chicago, university of chicago press. doi: https://doi.org/10.7208/chicago/9780226041223.001.0001 biltagy, m. (2013). the skills required for higher education graduates to join the labor market, policy paper presented to the egyptian cabinet, information and decision support center (idsc). biltagy, m. (2015). financing higher education in egypt: problems and suggested alternatives, feps journal, journal of faculty of economics and political science, cairo university, 16(3). boero, r., m. castellani, & squazzoni, f. (2003). labor market, entrepreneurship and human capital in industrial districts. an agent-based prototype. proceedings of the wild@ace workshop, turin, italy. doi: https://doi.org/10.1142/9789812702258_0016 bratsberg, b. and d. terrell. (2002). school quality and returns to education of u.s. immigrants, economic inquiry, 40(2). doi: https://doi.org/10.1093/ei/40.2.177 capmas. central agency for public mobilization and statistics. (2018). egypt in figures, arab republic of egypt. caudill, s. & mixon. f. (2012). human capital investment and the internment of japanese americans during wwii: a public choice approach, international journal of applied economics, 9(1). coleman, j. (1966). equality of educational opportunity. washington, d.c.: u.s. government printing office. elmps. (2012). egypt labor market panel survey. capmas and economic research forum. foster, a. & rosenzweig. m. 1995. learning by doing and learning from others: human capital and technical change in agriculture, journal of political economy, 103(6). doi: https://doi.org/10.1086/601447 gasparini, l., s. galiani, g. cruces acosta, p. (2011). educational upgrading and returns to skills in latin america: evidence from a supply-demand framework, 1990-2010, iza dp no. 6244. doi: https://doi.org/10.1596/1813-9450-5921 glewwe, p. (1996). the relevance of standard estimates of rates of return to schooling for education policy: a critical assessment, journal of development economics, 51(2). doi: https://doi.org/10.3386/w6106 https://doi.org/10.1086/258724 https://doi.org/10.7208/chicago/9780226041223.001.0001 https://doi.org/10.1142/9789812702258_0016 https://doi.org/10.1093/ei/40.2.177 https://doi.org/10.1086/601447 https://doi.org/10.1596/1813-9450-5921 marwa shibl biltagy / european journal of government and economics 8(1), june 2019, 81-95 95 https://doi.org/10.1016/s0304-3878(96)00415-4 idsc, information and decision support center, egyptian council of ministers. (2012). the incompatibility of university graduates' skills and labor market requirements for commercial, legal and agricultural studies, cairo. lucas, r. (1988). on the mechanics of economic development, journal of monetary economics, 22, 3-42. doi: https://doi.org/10.1016/0304-3932(88)90168-7 menzies, m., h. barwick & link, v. (2000). scholarships, fellowships and human capital formation in national innovation systems. ministry of research science and technology, new zealand. mincer, j. (1958). investment in human capital and personal income distribution, journal of political economy, 66. doi: https://doi.org/10.1086/258055 mincer, j. (1974). schooling, experience, and earnings. national bureau of economic research, new york. mof. (2014). ministry of finance, statistical report, the state’s general budget, 2014/2015. psacharopoulos, g. (1994). returns to investment in education: a global update, world development, 22(9). doi: https://doi.org/10.1016/0305-750x(94)90007-8 psacharopoulos, g. & patrinos, h,. a. (2002). returns to investment in education: a further update. world bank policy research working paper, no. 2881. psacharopoulos, g. and patrinos, h,. a. (2018). returns to investment in education: a decennial review of the global literature. world bank, policy research working paper, no. 8402. doi: https://doi.org/10.1596/1813-9450-8402 regan, l., g. burghardt & oaxaca, r. (2006). a human capital model of the effects of abilities and family background on optimal schooling levels. institute for the study of labor, iza. schultz, t. (1961). investment in human capital, american economic review, 51(1) schultz, t. (1989). returns to women’s education, phrwd background paper 89/001, world bank: population, health, and nutrition department, washington dc. https://doi.org/10.1016/s0304-3878(96)00415-4 https://doi.org/10.1016/0304-3932(88)90168-7 https://doi.org/10.1086/258055 https://doi.org/10.1016/0305-750x(94)90007-8 https://doi.org/10.1596/1813-9450-8402 front81 ejge oficial 8-1 contents81 number 8, issue 1, june 2019 4574 deciding on financial renegotiation in public-private partnership projects vol.7 • no.2 2018 issn: 2254-7088 european journal of government and economics 7(2), december 2018. european journal of government and economics issn: 2254-7088 number 7, issue 2, december 2018 doi: https://doi.org/10.17979/ejge.2017.7.2 transparency and local government corruption: what does lack of transparency hide? 106-122 juan luis jiménez, daniel albalate 106-122 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real. 123-137 carlos fernández_herraiz, antonio javier prado domínguez, carlos pateiro-rodriguez, jesus m. garcia-iglesias path – dependence and european fisheries management 138-153 federico martín palmero, fernando gonzález laxe why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem 154-170 filipa figueira deciding on financial renegotiation in public-private partnership projects 171-199 wiston risso https://doi.org/10.17979/ejge.2017.7.2 european journal of government and economics 7(2), december 2018, 171-199. european journal of government and economics issn: 2254-7088 deciding on financial renegotiation in public-private partnership projects winston rissoa, * a institute of economics (iecon), university of the republic, uruguay * corresponding author at: universidad de la república, instituto de economía, joaquín requena 1375 / 12000 / montevideo , montevideo , uruguay. email: arisso@iecon.ccee.edu.uy article history. received 24 january 2018; first revision required 23 april 2018; accepted 22 may 2018. abstract. this paper analyzes the renegotiation problem in the context of public-private partnership projects. utilizing a game-theoretic approach, an equilibrium is found in which the government finds that accepting renegotiation can be efficient. a first indicator is proposed based the public sector comparator (psc) that can be estimated by policymakers as an additional tool when deciding about renegotiation. a second more theoretical indicator is derived to analyze the economic and financial variables affecting renegotiation. this indicator is applied to four case studies in different countries (england, taiwan, portugal and china) and the results suggest that the model performs well. keywords. public-private partnership, renegotiation, value for money, game theory doi. https://doi.org/10.17979/ejge.2018.7.2.4508 1. introduction in the last two decades, public-private partnership (ppp) projects are a growing research topic as asserted by cui et al. (2018). the authors indicate some research gaps and research directions, one of which would be to explore and improve the efficiency of contract negotiation and renegotiation in the context of achieving flexible contracts. they remark that although the concept of ‘flexible contracts’ and ‘renegotiation’ introduce various alternative methodologies or guidelines, they are still at a nascent stage and offer enormous scope for refinement. in fact, cruz et al. (2014) highlight that renegotiations are clearly one of the most critical issues affecting the performance of concessions. the present work focuses on financial renegotiations involving ppp projects, in particular, as it uses a game-theoretic framework to develop tools that assist in renegotiation decisions by relying on certain conditions to generate a practical method that can be applied to cases in the real world. the article is organized as follows. in the next section, the literature about renegotiation in ppp is reviewed. in section three, the main causes of renegotiation are identified. in section four, a renegotiation game framework is explained. in section five, two methods are introduced mailto:arisso@iecon.ccee.edu.uy https://doi.org/10.17979/ejge.2018.7.2.4508 w. risso / european journal of government and economics 7(2), 171-199. 172 that identify whether to reject or accept a renegotiation proposal. section six analyzes four real case studies of ppp renegotiation. finally, section seven draws some conclusions. 2. literature review a recent review of the ppp literature by cui et al. (2018) identified six potential research topics: financial packages and ppp application (g1), economic viability and value for money (vfm) (g2), risk management and success factors (g3), procurement and contract management (g4), performance management (g5), and governance and regulation (g6). renegotiation appears as an important topic in the procurement and contract management (g4) group. ho (2009) argues that financial renegotiation plays a critical role in the success of a ppp project. financial renegotiation refers to revising financial subsidy measures after a ppp contract has been executed, when conditions have changed unfavorably and significantly. nikolaidis and roumboutsos (2013) describe the process of renegotiation in ppp projects as a bargaining process in which the parties seek to reach an agreement on one particular option from a set of available alternatives. ho (2006) contends that the fact that the government may bail out a distressed project and renegotiate with the developer in ppp projects can lead to serious opportunism problems in project administration. engel et al. (2014) posit that governments should limit contract renegotiations because they can erase the efficiency gains associated with ppps. hasselgren et al. (2014) point out that sometimes objective reasons arise that make renegotiation necessary, even in contracts that initially seemed to consider all possible contingencies. this is related to the paradigm of incomplete contracts established by grossman and hart (1986), hart and moore (1990), and hart (1995). bitran et al. (2013) assert that contract renegotiations typically mean high fiscal costs and more onerous contractual terms. guasch et al. (2014) reported that a significant number of ppp projects have been renegotiated shortly after the contracts have been signed. these renegotiations frequently occur a short period of time after – and sometimes even before – the financial terms have closed. the authors claim that 68% of ppp projects are renegotiated and that the average time before renegotiation is 1 year after execution. in sectors such as transport or water, this problem is even more evident, with percentages of 78% and 87% and an average time of 0.9 and 0.8 years before renegotiation, respectively. the empirical evidence on ppp renegotiations at an international level is diverse. gifford et al. (2014) argue that united states (us) ppp renegotiations have not experienced the same level of analysis as those in international markets, which might be explained by the relatively few us ppps and scarce data availability. infrastructure-oriented ppps in the us constitute a relatively new market that is still growing. the authors posit that there is not enough evidence regarding the causes of renegotiations in us ppps. baeza and vasallo (2010) note that spain has a long history of partnering with the private sector to help build and operate public infrastructure and that its ppp contracts suffer from significant income overestimation and are w. risso / european journal of government and economics 7(2), 171-199. 173 subject to frequent renegotiations. cruz and marques (2013) examine portugal (the european country with the most ppp projects) and identify specific variables that explain why ppps are so likely to be renegotiated in portugal – the average renegotiation rates are 67% and 100% in the water and transportation sectors, respectively – such as the magnitude of concessions and the dearth of regulations governing the executed contracts. guasch et al. (2014) analyze the renegotiation experience in latin america and find a renegotiation rate of 61%. chang (2013) asserts that compared with the developed countries, china is a latecomer to the practice of engaging in ppps. historically, the private sector was strictly forbidden to take part in the construction and operation of public infrastructure. hasselgren et al. (2014) point out that after several years and hundreds of ppp projects, there have been no renegotiations in india. to avoid moral hazards in the bidding process, it was established that renegotiation is only allowed for exceptional cases. renegotiation can be initiated by the government, the developer or by both parties. the government tends to renegotiate when there are changes to the ruling party, changes in priorities or because it cannot fulfill its contractual obligations. conversely, the private sector can initiate renegotiation for opportunistic reasons and in seeking to maximize the net present value (npv) of the ppp contract. guasch et al. (2014) found that 61% of renegotiations are initiated by the developer. hart and moore (1988) and dewatripont (1988) discuss the essence of renegotiation and its impacts. there are certain models of renegotiations, such as maskin and moore (1999), segal (1999), anderlini and felli (2001), ishiguro and itoh (2001) and tiong and alum (1997). the present paper relies on ho (2006) and develops an indicator when a government encounters the decision to rescue (or not rescue) a distressed project and what the impact is in terms of vfm. developing useful tools for policy makers facing the dilemma of whether to renegotiate a ppp is important for several reasons: 1) if a request for renegotiation is always granted, the developers would then be incentivized to bid optimistically to win projects that will likely be renegotiated and 2) if renegotiation is expected, the agent may choose inefficient actions that reduce overall or social efficiency but increase the agent’s payoff. 3. causes of renegotiation before analyzing the causes of renegotiation, it is important to distinguish the different types of ppp depending on the bundling and the compromises of the private sector. the build-operatetransfer (bot) model implies that the private sector builds, operates, and transfers the asset to the public sector at the end of the period; dbot means that the private sector designs (d), builds (b), operates (o) and transfers (t) the asset; the dbfot model includes assistance in financing (f) a project. there are variations of these models depending on the bundling. for instance, the dbfom includes maintaining (m) the infrastructure but there is no transfer; this model is also called private finance initiative (pfi) arrangements in the united kingdom (uk). w. risso / european journal of government and economics 7(2), 171-199. 174 actually, as asserted by makovšek and veryard (2016), the typical ppp contract is a dbfom type, where the private partner designs, builds, finances and operates the asset. the present work will focus on this model including private financing. iossa and martimort (2012, 2015) indicate that the financier can benefit the project by providing information and experience. in this sense, external finances would alleviate the moral risk of the contract both in the construction and operation stages, thus improving the risk allocation. note that dbfom is the model that includes more services, increasing the probability of renegotiation at some stage. tirole (1999) analyzes renegotiations in terms of transaction costs, explaining that renegotiation occurs because the cost of writing complete contracts can be high. in addition to contractual incompleteness, marques and berg (2010) consider that imperfect allocation of risk is an important factor in renegotiations. fatokun et al. (2015) identify the following main factors involved in renegotiations of ppp contracts for road construction: the lack of an adequate contract design; frequent opportunistic behavior by both the public and private partners during implementation of ppps; material changes in those conditions affecting revenue and costs that are beyond the reasonable assumptions in the original contract; corruption; and political and economic instability, which typically acts to reduce the chances of the public partner achieving its objective of vfm. fatokun et al. (2015) posit that the driving factors behind ppp contract renegotiations in spain include faulty contract design, defective regulations, over-estimation of traffic, inflexible contracts, changing construction risks and inadequate strategic network planning, among others. hasselgren et al. (2014) classify the different reasons why renegotiations might arise into the following four categories that are presented in table 1: the first category is technocratic and endogenous factors, which is related to the business and management aspects of ppp contracts. the second category is technocratic and exogenous factors, which can result in ppps failing to meet their performance goals, generally due to the macroeconomic environment. the third category is subjective and endogenous factors, which is related to imperfect and asymmetrical information. in this regard, the private sector may bid aggressively with the expectation to renegotiate after winning. the public sector can be affected by selection bias, which tends to choose bids that promise more than can realistically be delivered. it can result in the so-called “winner’s curse,” where the bidder tends to bid below the actual cost, since he believes that this is the only way to win. in fact, ho (2006) contends that opportunistic bidding behavior in ppps indicates that bidders’ proposals intentionally understate the possible risks involved or overstate project profitability to win contracts from their competitors. athias and nuñez (2008) found evidence of a strong winner’s curse effect when the public agency has limited experience with ppps. cruz and marques (2013) also conclude that the correlation in portuguese ppps between the type of award and ex post renegotiation supports both the winner’s curse and the strategic underbidding theories. finally, the fourth category is subjective and exogenous reasons, which involve the behavior of the government and public sector. in democratic countries, the governors have a term of four or five years. this can generate a tendency to think in the short term and lose the long-term w. risso / european journal of government and economics 7(2), 171-199. 175 perspective. when this happens, politicians can commit to ppp projects that they know will be renegotiated in the future when they are no longer in government. there is a severe risk of using ppps to elude budgetary spending restrictions, transferring the fiscal burden to the next government. principal-agent characteristics might also be particularly difficult to incorporate into public sector settings. the actors involved (for example, the legislature, individual politicians, the executive and its agencies) might relate to one another in many different ways in aiming to achieve a wide spectrum of objectives, sometimes consistent – and sometimes inconsistent – with one another. bi and wang (2011) recognize that current scholarship focused on moral hazard in ppp renegotiations is limited. table 1. causes of renegotiation four different categories endogenous factors 1) objective (technocratic) 3) subjective (political) management performance and coordination, construction risk, supply contract risk. winner’s curse, opportunism, strategic misrepresentation. organization theory, engineering contract/game theory, transaction cost theory exogenous factors 2) objective (technocratic) 4) subjective (political) economic downturn, changes in demand patterns, changes in institutional environment. public sector opportunism, principal-agent problems. institutional theory, neoclassical theory public choice theory, political science source: own elaboration based on hasselgren et al. (2014) according to hasselgren et al. (2014), if the risks were properly allocated, the endogenous factors do not justify renegotiation of the contract, since they must be assumed by the corresponding parties (public or private). only exogenous reasons and in extreme cases is the renegotiation mechanism enabled. an example of this could be the imminent bankruptcy of the country. cruz and marques (2013) assert that having a strong regulatory body may diminish the rate of renegotiations. 4. renegotiation game framework the present work is based on the game-theoretic framework developed by ho (2006), who investigates when and how the government rescues a distressed project. in this sense, the model focuses on financial renegotiation involved in revisiting financial subsidy negotiations after a contract was executed, particularly when conditions have changed unfavorably and w. risso / european journal of government and economics 7(2), 171-199. 176 significantly. the present paper introduces benefit functions for the two players – the developer and the government – allowing an analysis of the conditions that favor the renegotiation of a contract for purposes of obtaining an indicator of the efficiency of renegotiating a contract. figure 1 presents the dynamic game in extended form. it is a two-stage game with two players (developer and government). the game begins from adverse situations in which the developer has two strategies: 1) they can declare the project bankrupt and the game is over; or 2) the developer can attempt to ask the government to rescue and subsidize at a percentage of the initial investment and the cost flow (g). this second strategy implies that the developer must negotiate with the government. if the developer chooses to negotiate, the government has two strategies: 1) to reject renegotiation (letting the developer go bankrupt and retendering the project); or 2) to rescue the project and subsidize it. if the developer chooses project bankruptcy at the first stage or the government rejects to negotiate a subsidy (in a second stage), then the developer must pay a cost. here, it is assumed that the payoff is 0, since the value of the equity shares held by the developer should approach zero prior to bankruptcy. in both cases, the government must retender the project with a cost associated with project retendering. then, the payoff for the government is a new vfm given by vfm0 /(1+ ρt), where vfm0 is the initial or original positive value for money, ρ is the social discount rate and t is the waiting time until a new developer is contracted. the term 1/(1+ρt) may be thought of as a discount factor. in this sense, the new vfm implies that the larger the ρt, the smaller the new vfm is. figure 1. renegotiation game’s equilibrium path. source: own elaboration. alternatively, as shown in figure 1, the developer can negotiate subsidy g as a percentage developer government request a subsidy, (g) project bankruptcy negotiate a subsidy, (g) reject [ p(g)-p(0), vfm0(g) ] [ 0, )1( 0 t vfm ρ+ ] [ 0, )1( 0 t vfm ρ+ ] w. risso / european journal of government and economics 7(2), 171-199. 177 of the initial investment and the cost flow, thus obtaining an annual increment of the payoff p(g)p(0). after the developer’s request for a subsidy, the government can reject the subsidy as mentioned before and receive a payoff (0, vfm0/(1+ρt)). however, if the government decides to negotiate a subsidy, rescuing the developer will reduce the vfm from the original amount to vfm(g). therefore, the payoff to the developer and the government will be (p(g)-p(0), vfm0(g)). ho (2006) argues that there is a political cost associated with rescuing the project and subsidizing a private party. cruz and marques (2013) assert that renegotiation costs are difficult to calculate not only because there are different forms but also because the increased costs are frequently offset (at least in part) by certain benefits. note that the best developer strategy is to request a subsidy because failing to ask results in a payoff of 0 (and if the request is rejected, the payoff is 0), but it is also possible that the request will lead to a positive payoff of p(g)-p(0). the government has two options: “negotiate the subsidy” or “reject” the proposal. guasch et al. (2014) posit that governments with weak institutions face a crucial tradeoff between canceling and renegotiating a ppp contract to avoid a potential bankruptcy, among other outcomes. thus, the number of cancelled contracts in latin america is low (but has been increasing over the last 30 years) at approximately 5% of total ppp projects granted. this rate is slightly higher than the average for developing countries, which is 4.3%. the renegotiation will be the best strategy under the following condition: 0 )1( )( 0 0 ≥ + ≥ t vfm gvfm ρ (1) the condition means that the impact of the reduction in the vfm resulting from the subsidy is less than the effect of retendering the project and waitgin for a new developer to take on the project. note that vfm is considered as the objective function of government. following kennedy (2013), the criterion for a ppp to be rescued is that it continues to provide greater vfm than other possible procurement methods, including retendering the ppp to another private party. to understand this condition, the payoff model for the developer and the government must be explained. we consider a ppp project whose substance is to design-build-finance-operatemaintain (dbfom) a determined infrastructure. the developer will receive a constant disposal payment (p) from the government and let us assume for the sake of simplicity that the project is infinite in time and that the annual cost to operate and maintain the infrastructure is constant. the developer will receive a payment corresponding to a npv equal to zero for a determined internal rate of return (irr). the following is the developer npv: dteiicipinpv kt−+∞ ∫ −−−+−= 0 000 ])1([ ee (2) where ε is the equity percentage, i0 is the initial investment amount, c is the annual cost as a w. risso / european journal of government and economics 7(2), 171-199. 178 percentage of the investment, i is the private interest rate and k is the irr required for the project to be sustainable. these two financial variables will be expressed as follows: i=r+p and k=r+σ. thus, the private financial rate (i) is equal to the sovereign financial rate plus a spread (p), and the irr is equal to the sovereign rate (r) plus the required risk (σ). solving the equation when npv=0, the annual developer payment is obtained as shown in equation (3). ))1((0 cprip +−++= eeσ (3) the government must decide between traditional public procurement and pursuing the project through a ppp. sarmento (2010) remarked that vfm is one of the leading tools available to public managers to assess the value of pursuing a project through a ppp versus traditional procurement because it provides the public sector with a simple methodology to estimate the benefits, costs and risks involved in the project. fatokun et al. (2015) assert that even if the importance of evaluating renegotiation in terms of vfm achievement is previously established, few empirically assess the relationship between these two concepts to address the challenge of underor non-achievement of vfm in ppp road project transactions. equation (4) presents the vfm of an infrastructure project. the equation has two main terms. the first term shows the government cost of following the traditional option in which r is a public discount rate and δ is a parameter representing expected risk. this parameter is typically estimated in an infrastructure project using a percentage of cost overruns multiplied by the probability of cost overrun. cost overruns are an important parameter determining the infrastructure risk. see for instance, singh (2010), arvan and leite (1990), cantarelli et al. (2010), flyvbjerg et al. (2002), flyvbjerg et al. (2003), flyvbjerg et al. (2004). see lee (2008) and bansal and kalady (2013) for the importance of cost overruns and their impact in infrastructure projects.      +−−−     ++= −+∞−+∞−+∞ ∫∫∫ dteciidtepdteciivfm rtrtrt 0 0000 00 )1()1( dλd (4) the second part of the equation represents the government cost if the project is pursued using the ppp option. in this sense, we have the flow of payments (p) to the developer and the proportion of the risk retained by the administration (1-λ). when the cost of the traditional public option is larger than the ppp option, vfm is positive and it will be efficient to pursue the ppp. equation (4) can be simplified by (5). dtepdteciivfm rtrt −+∞−+∞ ∫∫ −     ++= 00 00)1( dλ (5) solving the integrals in equation (5) and substituting the required payment presented in equation (3), we obtain the following expression of the vfm. w. risso / european journal of government and economics 7(2), 171-199. 179 [ ]))1(()(0 pcr r i vfm eeσdλ −+−+= (6) note that equation (6) shows that vfm is positive when the percentage of the risk transferred that is associated with the public financing rate and the annual cost is greater than the weighted private risk and the required financial spread. thus, if the developer has a high private risk or the financial spread is high with respect to the transferred public cost, it will be determined that the ppp option is not efficient. arguments related to efficiency should explain why the private sector has a lower risk value than the public sector. some authors assert that efficiency should be the only reason why the public sector is willing to pay a high price to the private sector. in fact, engel et al. (2010) suggest that the higher cost of financing ppps is not an argument in favor of public provision, since it appears to reflect both the combination of deficient contract design and the cost-cutting incentives embedded in ppps. thus, in the case of a correctly designed ppp contract, a higher cost of capital may be the price paid for the efficiency advantages accruing to ppps. note that vfm is increasing with c, as the annual percentage cost of the project rises, it is more convenient to follow the ppp option. this finding suggests that projects with important costs relative to initial investment are more likely to be implemented using a ppp structure. even if equations (3) and (6) are the developer and government payoffs when the contract is signed, we are interested in the conditions contemplated when the project is negotiated. in this case, the developer npv is recalculated using the following equation, in which g is the percentage of the initial investment and the annual cost. dtegiigcipgignpv kt−+∞ ∫ +−−+−++−= 0 000 )]1()1()1([)1()( ee (7) in this case, the developer will request a new payment determined by equation (8). ))1()(1()( 0 cprgigp +−+++= eeσ (8) equation (8) has an effect on the vfm of the government in determining a reduction with regard to the original amount when the developer won the tender, as shown in equation (9). [ ]))1()(1())(()( 0 pgcrg r i gvfm eeσdλ −++−+−= (9) note that the renegotiation variable g has two effects on vfm. increasing g will reduce the benefits of the ppp project, thus reducing the effect of the transferred risk and increasing the w. risso / european journal of government and economics 7(2), 171-199. 180 private cost that is measured by the private risk and the financial spread. equation (9) is the payoff to the government of renegotiating the subsidy to the developer; in the case in which the request is rejected, the payoff is given by equation (10). [ ] )1( ))1(()( )1( 0 t pcr r i t vfm ρ eeσdλ ρ + −+−+ = + (10) the renegotiation will result in a nash equilibrium insofar as equation (9) will be larger than equation (10). expression (11) shows the condition in which the request (g) made by the developer is consistent with the renegotiation equilibrium. [ ] [ ]pcr pcr t tg )1( )1()( )1( eeσ eeσdλ ρ ρ −+++ −−−+ + ≤ (11) considering that vfm is positive, when the condition takes the equality, it makes no difference to the government to “renegotiate” or to “reject” the proposal. notably, the renegotiation level is positively related to the social discount rate, the waiting time until the new developer takes on the project, the cost overrun, the level of transferred risk, the percentage cost, and the public interest rate and is negatively related to the private risk level, the financial spread requested by the financial system and equity level, since typically σ-p>0. 5. methodological proposal and the indicator of renegotiation this section introduces two possible practical tools to analyze the decision of a government facing a renegotiation. the first methodology is based on the public sector comparator (psc), and the second methodology is based on the design of a general indicator. it is well known that the psc is used by a government to make decisions about vfm in ppp projects. in fact, there are many government manuals in different parts of the world that consider this methodology. examples of these manuals are the vdtf (2001) in victoriaaustralia, the pbc (2003) in canada, the ntsa (2004) in south africa, the dper (2007) in ireland, the coulson (2008) in uk, the mappp (2008) in france, the australian government (2008) in australia, the utfp (2009) in italy, the dnp (2009) in colombia, the kdi (2010) in south korea, the shcp (2010) in mexico, the mef-cnd (2012) in uruguay, the mef (2014) in peru and the world bank (2012) as a general guide. based on these methodologies, to compute the vfm, we can establish a form to evaluate the convenience of renegotiation. sarmento (2010) indicates that to compute the psc, we need to estimate the public project cash flow including the full costs, revenues and risks discounted at the public sector rate to determine the npv. comparing this amount with the discounted value of payments (including the risks and costs retained by the public sector) to the private supplier, we obtain the w. risso / european journal of government and economics 7(2), 171-199. 181 estimation of the vfm. the author also considers the convenience of conducting the psc prior to the bid. considering this method, which is also included in the abovementioned manuals, we can compute the vfm as in equation (12), where the present value of ct represents the most efficient form of public procurement, the present value of pt is the ppp provision, r is the public interest rate and n is the end of the concession. ∑∑ = = = = + − + = nt t t t nt t t t r p r c vfm 00 )1()1( (12) renegotiating implies vfm(g), which is given by equation (13). ∑∑ = = = = + + − + = nt t t t nt t t t r pg r c gvfm 00 )1( )1( )1( )( (13) rejecting the subsidy implies vfm(ρt), which is given by equation (14). note that (1+ρt) is a discount factor indicating that the government has to postpone the project a time t. ∑∑ = = = = ++ − ++ = nt t t t nt t t t tr p tr c tvfm 00 )1()1()1()1( )( ρρ ρ (14) there are two conditions for renegotiation: 1) vfm(g)-vfm(ρt) >0, where the vfm with the subsidy is larger than that to postpone the original vfm for a time t; and 2) vfm(g)>0, where the vfm is still positive with the subsidy, otherwise it is convenient to the public provision. rearranging vfm(g)-vfm(ρt), we obtain the equation (15). when rvfm is positive, renegotiation will be convenient instead of retendering the project. ( ) ( ) ∑∑ = = = = + −      + − + = nt t t t nt t t tt vfm r gp r pc t tr 00 )1()1(1 ρ ρ (15) note that computing derivatives of rvfm with respect to g, ρ and t we obtain the expected results. on the one hand, ∂rvfm/∂g < 0 implies that the larger the subsidy, the lower the convenience to renegotiate. on the other hand, ∂rvfm/∂ρ> 0 and ∂rvfm/∂t > 0 implies that the larger the social discount rate (ρ) or the larger the time to postpone the project (t), the larger the convenience to renegotiate. one of the limitations of this indicator is the political cost, which is difficult to measure and to include. it is necessary to consider the evaluation of the political costs of saving the project and postponing the project. a second limitation is related to the ability to avoid the “winner’s curse”. in this case, it is difficult to compare the proposed renegotiation and the postponed project because it is necessary to work with a better estimation than the w. risso / european journal of government and economics 7(2), 171-199. 182 vfm0/(1+ρt), since the vfm0 of the original winner project would be underestimated. we can propose a more general second tool considering equations (9) and (10). this indicator has similar characteristics and limitations to those of the first tool, but it will be useful for examining the economic and financial relationships and analyzing the study cases. considering condition (11), it is possible to define the “renegotiate” variable, r. this variable is positive when renegotiate is the more efficient decision and negative when rejecting renegotiation is the best strategy. { } { } ))1(()1()()1( pgttcrgttr eeσρρρdλρ −+++−++−= (16) where r is a function of a vector x of ten variables (ρ, t, δ, λ, g, r, c, ε, σ, p). when complete information about the project is available for each element, r is calculated deterministically. however, when a variable in x is not known, a probability distribution will be assigned based on the empirical evidence. therefore, it is possible to apply monte carlo simulations to obtain the simulated distribution of r(x). in particular, it will be informative to test the decision to renegotiate by a determined amount expressed as a percentage of the initial investment, which is easily obtained by computing (p(g)/k p(0)/k)/i0) –the difference between the total amount of payments following acceptance of the renegotiation minus the total amount of the original payment flows over the initial investment. the total cost of renegotiating as a percentage of i0 is g(k+c)/k. the next step is to make an assumption for each variable considering that the particular projects in some cases may have some information about some variables. for instance, ρ generally takes a value of 12% internationally. according to esty (2000), ε typically takes a value that is between 10% and 35% in these types of projects compared with a range of 65% and 75% in industrial firms. an important consideration should be noted in the case of the cost overrun variable, . this variable is typically modeled as normal, but some evidence indicates that it is not the most efficient distribution because the probability of obtaining variables larger than the mean is higher than obtaining variables lower than the mean. although flyvbjerg et al. (2002) assume that cost overruns are normal, a number of studies have shown that overruns are not normal. love et al. (2014) find that cost overruns follow a generalized log-logistic distribution function. bertisen and david (2008) argue that the evidence suggests that cost overruns are shifted lognormal. horman (2001) highlights that a lognormal distribution best suit cost overruns because this distribution is best for stochastic estimates of cost variables, as shown by wall (1997). since most of the cost overrun information used in this study is based on flybjerg et al. (2007), which provides only the mean and standard deviation, a generalized log-normal distribution will be approximated as the empirical cost overrun probability distribution function for simulations of r. w. risso / european journal of government and economics 7(2), 171-199. 183 table 2. data inputs for the analysis of the percentages of ppp renegotiations variable code la la la la us france france uk total electricit y transpor t water highway s highway s parking total annual cost (a) c 15.00% 15.00% 15.00% 15.00% 15.00% 15.00% 15.00% 15.00% public interest rate(b) r 4.07% 4.07% 4.07% 4.07% 2.27% 3.13% 3.13% 3.13% equity (a) ε 10.00% 10.00% 10.00% 10.00% 10.00% 10.00% 10.00% 10.00% private interest rate(b) i 5.32% 5.32% 5.32% 5.32% 4.02% 4.38% 4.38% 4.38% cost overruns (mean) (c) δ 27.60% 27.60% 20.40% 33.80% 8.40% 22.40% 22.40% 25.70% transfer risk (a) λ 80% 80% 80% 80% 80% 80% 80% 80% social discount rate(a) ρ 12.00% 12.00% 12.00% 12.00% 12.00% 12.00% 12.00% 12.00% public waiting time (a) t 3 3 3 3 3 3 3 3 irr (b) k 11.77% 13.85% 12.09% 12.96% 10.17% 11.11% 11.11% 9.95% renegotiations (%) (d) 68% 41% 78% 92% 40% 50% 73% 55% prob. renegotiation (e) 22.68% 19.45% 14.31% 24.57% 3.94% 16.84% 16.83% 23.02% source: own calculations. (a) estimated parameters based on nao (2009) and esty (2000), and transfer risk in the boo model suggested by allan (2001) and ho (2009); (b) data from the damodaran website: http://people.stern.nyu.edu/adamodar/; (c) taken from flyvbjerg (2007); (d) taken from guasch et al. (2014); (e) the simulated probability that the estimated r is positive and then renegotiation is convenient. before conducting the analysis of the four case studies, an exercise was carried out to determine whether the level of ppp contract renegotiations might have some connection with the probability of renegotiating. table 2 shows the input data for the exercise based on the levels of renegotiation presented by guasch et al. (2014) for different sectors and regions, including latin america (la), united states (us), england (uk) and france. it is important to clarify that hasselgren et al. (2014) stipulate that one should be cautious with the conclusions obtained by analyzing the percentages of renegotiations. due to practical differences in the nature and intrinsic particularities of each case, it is not advisable to use them to estimate the frequency of ppp renegotiation, considering that no important variables (such as the amount of investment projects requested in each case) are identified and their capacities and institutional strengths are not known. the objective here is to show graphically whether there can be a relationship that presents a greater propensity to renegotiate because the probability of accepting a renegotiation is higher. for this, the information in table 2 was used for each type of project. monte carlo simulations were conducted for a renegotiation request (g) of 10%, considering an order for an initial amount given, and the acceptance percentage of the renegotiation was taken therefrom. http://people.stern.nyu.edu/adamodar/ w. risso / european journal of government and economics 7(2), 171-199. 184 figure 2. percentage of renegotiations and the probability of renegotiation. source: own calculations based on data from guasch et al. (2014) figure 2 shows the cloud of points between renegotiation and the percentages indicated by guasch and the calculated probability. if a straight line is fitted, a positive correlation of 0.48 between the variables can be gleaned, which seems to suggest a possible relationship between the probability that renegotiation related to a project will be accepted and the percentage of renegotiations that were realized. unfortunately, there is not more data to undertake a deeper analysis in this regard, although a more rigorous analysis requires considering each particular project and other variables, including institutional variables. in fact, iossa and saussier (2018) assert that weak institutions reduce the ability to transfer the risks and commitments assumed in the contract credibly. furthermore, iossa and martimort (2016) indicate that weak political and regulatory institutions reduce the performance of ppp contracts by raising the risk of corruption. iossa and saussier (2018) remark that the government’s lack of commitment to renegotiate, as well as weak governance, explains the reasons for the frequent revisions of contracts in la countries. in many of these countries, the regulatory agencies receive little training and instruments to carry out their mandate. in many cases, these agencies are under political control that affects their autonomy. considering la countries as a proxy of weak institutions and developing countries, a larger average renegotiation rate and probability are observed compared to developing countries. in fact, the average renegotiation rate in la countries is 70% and the average probability is 20%, whereas in developed countries, the average renegotiation rate is 55% and the average probability is 15%. note that the estimated probabilities seem to predict higher renegotiation rates in la countries with respect to developed countries. 6. case studies and empirical evidence the following four case studies were selected considering the available information, the number of studies analyzing the cases and, principally, because they are from the same sector (light 0% 5% 10% 15% 20% 25% 30% 35% 45% 55% 65% 75% 85% 95% pr ob ab ili ty o f r en eg oa tia te 1 0% percentage of renegotiations for regions and sectors w. risso / european journal of government and economics 7(2), 171-199. 185 railways) but differing governance environments. the china case was also selected as a control case, since it is an example of a successful project after two years of hard negotiations. the two european countries, uk and portugal, are different. the uk economy (gdp of 2,629.20 billion in 2016) is larger than that of portugal with an estimated gdp of usd 204.80 billion. according to cruz and marquez (2011), the latter country has the largest spending on ppps as a percentage of gdp in europe. china and taiwan are from different governance environments in east asia. china is the second largest economy in the world with a gdp of usd 11,218.30 billion and an economy that is highly regulated by the government. chang (2013) argues that china is a latecomer to ppp practices, at least compared with developed countries. historically, the chinese private sector was strictly forbidden from getting involved in the construction and operation of any public infrastructure. taiwan is one of the four asian tigers that underwent rapid industrialization and maintained exceptionally high growth rates. nowadays, it has a highincome economy with a gdp of usd 528.60 billion in 2016. considering the global competitiveness index 2017-2018 computed by the world economic forum, in terms of infrastructure, the uk ranked 11th, taiwan 15th, portugal 18th and china 46th. note that there are differences with respect to the public sector performance; according to the same index, the uk ranks 15th, china 26th, taiwan 38th and portugal is in a far position at 95th. according to bing et al (2005), the uk prefers to allocate the majority of the risk to the private sector. on the contrary, ke et al. (2010) found that preferences in china indicate that no risk fell solely on the private sector. governance in taiwan is also different from china with respect to the private sector. chan et al. (2010) suggest that one of the major challenges to the successfully implementation of ppp projects is the lack of an effective risk assessment model and an equitable risk-sharing mechanism. it is interesting to note that in the included project from china, 80% of the risks were allocated to the private sector, and the risks beyond the control of the private sector were shared or bore by the government. table 3 summarizes all the available parameters, assumptions and the computed r indicator for the four cases. case 1 corresponds to the london underground, a metro system serving large areas of london, england. the information of this case is reported in the nao (2009) and was analyzed by kennedy (2013), hallikeri (2015) and sheikh et al. (2015). by the end of the 1990s, the public was becoming less confident about the safety and efficiency of the system. several accidents had occurred on the london underground, and the uk government announced a modernization program to be implemented by means of a ppp. nao (2000) asserts that in march 2000, the contracts were expected to deliver £12-13 billion of capital and maintenance work during the first 15 years of the contract. w. risso / european journal of government and economics 7(2), 171-199. 186 table 3. parameters of the four case studies variable description metronet (uk) mts (portugal) thsr (taiwan) metro line 4 (china) c annual cost/investment 14.96% 14.96% 14.96% 14.96% r public interest rate 3.50% 6.00% 4.07% 6.00% ε % equity 12.00% 15.00% 25.00% 30.00% σ % project risk 7.61% 4.45% 8.89% 1.50% i private interest rate 4.38%(a) 7.00% 5.32% 6.30% p spread (i-r) 0.88% 1.00% 1.25% 0.30% δ cost overrun (44.7%, 38.4%)(b) (44.7%, 38.4%)(b) (44.7%, 38.4%)(b) (44.7%, 38.4%)(b) λ transferred risk 80.00% 80.00% 80.00% (70%, 80%, 90%) (c) ρ social discount rate 12.00% 12.00% 12.00% 12.00% t social waiting time (1, 2, 3) (c) 3 3 (2, 3) (d) k irr 11.11%(a) 11.11% 12.96% (7%, 8%) (d) g(c+k)/k renegotiated amount 43.90% 35.56% 12.20% 20.00% r renegotiation indicator -0.03494 -0.02320 -0.00396 0.00174 source: own calculations. (a) the irr and engineering/construction interest rate in europe are taken from the damodaran website: http://people.stern.nyu.edu/adamodar/; (b) taken from flyvbjerg (2007). lognormal distribution is assumed (mean, standard deviation). (c) triangular distribution is assumed (minimum, more probable, maximum). (d) uniform distribution is assumed (minimum, maximum). the contracts were bid on and won by two private consortiums, metronet and tube lines. they were schedule to work over a 30-year period, beginning in 2004. however, by 2007, metronet had already entered into insolvency administration after it could not meet its obligations. one important limitation is the lack of complete information; no one knew the conditions of the less accessible parts of the tube’s infrastructure, making difficult to accurately estimate the cost of future maintenance. considering the cost uncertainty of the project, the government decided to break the 30-year project in four 7.5-year sections, allowing for periodic reviews. by 2010, tube lines had also entered into insolvency administration under similar circumstances. the result was that the ppp had completely failed, resulting in huge losses to the public and several government-led inquires followed that heavily criticized metronet and tube lines. nao (2009, p. 17) presents the cost structure of the metronet project after 7.5 years: considering that investment is ₤4.1 billion and operational expenditures, administration and financing costs are ₤4.6 billion, the c is estimated as an average of 14.96% per year. the same report asserts that the government’s discount rate, r, is 3.5% and the equity level is 12%. project risk was estimated to be 7.61%, based on the damodaran website for information in the engineering and construction sectors in europe, where the private interest rate is 4.38% and the irr is 11.11%. generally, transferred risk is approximately 80%, and the social discount rate is typically set at 12%. cost overrun is modeled by applying a lognormal distribution, based w. risso / european journal of government and economics 7(2), 171-199. 187 on flyvbjerg et al. (2004), who indicate that the raw mean and standard deviation of rail projects are 44.7% and 38.4%, respectively. flyvbjerg (2007) claims that there is no significant difference between urban and other types of rail in terms of cost overruns. finally, social waiting time is modeled using a triangular distribution, with a minimum of 1 year and a maximum of 3 years until a new developer is operational. according to nao (2009), considering the 7.5 years, the capital expenditure was ₤4.1 billion and the developer required an additional ₤1.8 billion; this additional amount (g(c+k)/k) represents 43.90% of the investment. note that applying vfm equation (6) to the project before the required amount generates vfm equivalent to ₤5.76 billion when applying the mean cost overrun of 44.7%. however, the impact of the 43.90% required by the developer would change the decision to maintain the contract. the government notes that for a period of 7.5 years, the expected investment was ₤4.1 billion, but the developer required an additional ₤1.8 billion (see figure 3). thus, evaluation of vfm must be undertaken in this context. figure 3. metronet’s project spend as of july 2007. source: own elaboration based on nao (2009, p. 17). the first period runs from april 2003 to october 2007 (7.5 years). in a deterministic context, considering a cost overrun of 44.70% and a social waiting time of 2 years until a new developer is operational, the r indicator is -0.03494. in fact, the decision of the government was to reject the proposal and reopen the project to public administration. the simulated distribution of r is obtained running 1,000,000 trials. as shown in figure 4, the probability of renegotiation is 2.72%, which means that under this structure, there is a 97.28% likelihood of rejecting the renegotiation. note that even when the convenience of renegotiation is clearly rejected, the government did not decide to tender the project, but decided to return the project to the public administration. according to hallikeri (2015) the decision may be the correct one because there were difficulties in measuring costs and allocating risks, and in this case, undertaking a ppp is the wrong strategy for a public sector that seeks value for money. w. risso / european journal of government and economics 7(2), 171-199. 188 figure 4. simulated distribution of r for the metronet case. source: own calculation applying crystal ball. case 2 is the metro transportes do sul (mts) in portugal. information about the case can be found in ferreira (2014). according to cruz and marques (2011), portugal shows the largest ppp spending in europe as a percentage of gdp. in fact, the use of the ppp model as a financing scheme has led to underestimations regarding future annual burdens, and the fiscal impact due to ppp renegotiations is high. one large ppp project is the concession of the southern tagus surface metro to metro transportes do sul (mts). in 1999, a tender was launched in connection with the project, pursuant to which the private partner would be responsible for the construction, equipment supply, and financing for the operation and maintenance (dboft) of the southern tagus surface metro. in 2002, the concession was awarded to mts for 30 years. irr is estimated at 11.11% after reviewing the damodaran website’s cost of equity for engineering and construction in europe. the public discount rate in the case of portugal is 6%, whereas the social discount rate is 12%. the social waiting time is estimated at 3 years by calculating the period from the tender until the mts was functional. as in the other cases, the cost overrun is modeled by applying the mean and standard deviation of rail projects at 44.7% and 38.4%, respectively. the private interest rate is 7% and the transfer risk is approximately 80%. the concessionaire contributed approximately 15%. the total investment cost was € 284 million, and the vfm was € 283 million. the renegotiation implies a cost of € 101 million (35.56% of the investment) which in our case reduces vfm to € 127.31 million if renegotiation is accepted or € 181.08 million if the project must be retendered. in this case, the indicator r is -0.02320, which suggests not rescuing the concession. figure 6 shows the simulated distribution for the mts case with an 11.39% probability of accepting the renegotiation and an 88.61% chance of rejecting it. although the recommendation is negative, portugal accepted renegotiations with mts. ferreira (2014) asserts that this case shows that the government’s bad preparation when negotiating the concession contract led to its lack of decision power during the construction process. w. risso / european journal of government and economics 7(2), 171-199. 189 figure 6. simulated distribution of r for the mts case. source: own calculation applying crystal ball. all this leads to a misallocation and an inadequate management of risk and renegotiation. cruz and marques (2011) stress that between 2002 and 2004, when renegotiations took place, a total of 8 ministers, 3 secretaries of state and 3 project leaders were involved in the project. from the public sector side, there was thus a clear governance problem, which weakened advocacy for the public interest when negotiating with the private sector (which is stable and builds up knowledge over time). this analysis (at least partially) explains why the government's decision does not match the recommendation of the indicator suggested under both deterministic and probabilistic result analyses. it is also consistent with the low world position of public-sector performance. portugal has the worst rank (position 95) of the four countries in the public-sector performance index compared to the rest of the four countries case 3 corresponds to the high speed rail in taiwan. the case information is reported by dutzik et al. (2011) and also analyzed by ho (2009). in 1998, the taiwan high speed rail corporation (thsrc) was awarded a 35-year concession to build and operate the taiwan high speed rail (thsr), partially based on thsrc’s promise to build the system without government financing. however, the company began to run into difficulties after the asian financial crisis in the late 1990s, when it was forced to take out high-interest loans to pay for the project. the investment cost is $16.4 billion, the equity level is 25%, the social waiting time is 3 years, and the social discount rate is 12%. the annual cost as a percentage of the investment is considered the same percentage as in the metronet case, i.e., 14.96%. project risk was estimated at 8.89%; the damodaran website was again checked to determine that the private interest rate for the engineering and construction sectors in emerging markets was 5.32%; the irr is 12.09%; and as discussed above, the transfer risk is approximately 80%. cost overrun is modeled by applying a lognormal distribution, based on flybjerg et al. (2004), using the raw mean and standard deviation for rail projects. when the contract was executed, the vfm – including a cost overrun of 44.7% – is estimated at $14.69 billion and even after including the $2 billion requested by the developer, the vfm remained at $9.63 billion. considering that the vfm is $9.4 billion if the contract is retendered, w. risso / european journal of government and economics 7(2), 171-199. 190 the r indicator of -0.00396 suggests that it is not efficient to renegotiate the contract. figure 5 shows a simulated distribution of r obtained by running 1,000,000 trials. the probability of renegotiation is 28.32%, which indicates that under this structure there is a 71.68% chance of rejecting the renegotiation. thus, the probability of renegotiation in this case is higher than in the metronet one. figure 5. simulated distribution of r for the high speed rail case. source: own calculation applying crystal ball. ho (2009) contends that the taiwanese case is an important project and that projects that are too important or too expensive for society to allow them to fail (or default) are not good candidates for ppps. the costs requested are similar in both cases, $1.8 billion in the metronet case and $ 2 billion in the thsr case. however, the proportion of the investment is different; in the first case it represents 43.90% of the investment and in the high speed rail case, it represents 12.20%. the taiwanese government might have allowed the thsrc to go bankrupt and cease operations when the company ran into financial trouble. however, doing so would have resulted in the loss of a critical public asset, leaving the government with little choice but to prop up the failed business plan of a private developer with public funds. in any case, the indicator does not consider the mistakes made by the government. as mentioned, a megaproject is not a good candidate for a ppp and the political costs of not recuing such a project are very high. in addition, ho (2009) indicates that the government was very aggressive at adopting ppps for almost all public infrastructure projects, encouraging an opportunistic attitude and many related problems. case 4 is the beijing metro line nº4 in china and the case was analyzed by chang (2013), liu and wilkinson (2015) and li (2017). this is an example of a successful case and renegotiation occurred during the negotiation of the contract. in this sense, the case is included as a control case to compare the results. the latter authors claim that the project was the mainland china’s first ppp project involving the development and operation of an urban rail transit system. according to chang (2013), the beijing infrastructure investment corporation (biic) decided to implement a ppp model for the w. risso / european journal of government and economics 7(2), 171-199. 191 beijing no. 4 line project for a 30-year period. the objective of the project was to accelerate beijing’s metro development for the 2008 olympic games. in 2004, following a public tender process, the beijing mtr corporation was selected to undertake this development. the contract was under negotiation for one more year, and it was not finalized until 2006. liu and wilkinson (2015) posit that irr is estimated between 7-8%, and it is assumed that irr is distributed uniformly between 7-8%. chang (2013) indicates that the discount rate is 6%, the interest rate is 6.3% and the equity level is 30%. the annual cost is estimated at 14.96% (as in the previous cases), and the social discount rate is 12%. considering the periods between the public tender and the operation of the developer, the time is distributed uniformly between 23 years. transfer risk is assumed to be triangularly distributed (70%, 80%, 90%). since the project is an urban railway, cost overruns are modeled by applying the mean and standard deviation of rail projects, which are 44.7% and 38.4%, as in the previous cases. the total investment cost was $2.4 billion, and considering a transfer risk of 80%, an average irr of 7.5% and an average cost overrun of 44.7%, the vfm was $2.73 billion. renegotiation estimated a cost of $480 million, representing 20% of the investment. according to chang (2013), this percentage is usual for beijing projects. renegotiation would reduce the vfm to $2.16 billion if renegotiation was accepted and to $2.10 billion if the project was retendered. in this case, the indicator r is 0.00174, thus suggesting that the renegotiation should be accepted. figure 7 shows the simulated distribution for the beijing metro line 4 case with a 39.38% probability of accepting the renegotiation. this probability is the largest among the four cases presented. figure 7. simulated distribution of r for the metro line 4 beijing case. source: own calculation applying crystal ball. w. risso / european journal of government and economics 7(2), 171-199. 192 we analyzed four cases of renegotiation for similar types of projects under different governances. it seems clear that the uk has the most solid institutions and the choice to not renegotiate according to the indicator was decided by the government. as mentioned, china was a control case because it is an example of a successful case, but note that it behaved as forecasted by the indicator. the outliers are the cases from portugal and taiwan. taiwan decided to renegotiate but the indicator suggested the opposite; however, in this case, there may be some issues not captured by the indicator. for instance, it is highlighted that the project is too important to fail and too expensive to default and there is a hidden cost of abandoning the project. it is also important to remark the economic context in taiwan due to the asian crisis at the end of the 1990s and the increases in the interest rates, which would also affect the cost of retendering the project. in this situation, the vfm of retendering the project could be overestimated, but in any case, the project should not have been executed as a ppp. portugal renegotiated the project but the indicator has the largest probability of no renegotiation. this seems to be a clear case of weak governance, as suggested by cruz and marques (2011) and ferreira (2014). it is also illustrative that portugal has the worst rank in the public-sector performance index among the four countries. marques (2017) considers that a contract alone does not protect the public interests; it is necessary to have regulations and an independent regulatory authority. note that even when these governments did not compute this indicator, we can see the potential to correctly predict the government’s behavior in the cases of uk (rejecting renegotiation) and china (accepting renegotiation). taiwan is a case where the project was not recommended as a ppp. finally, portugal is an example of the effects of a lack of regulations and poor performance of public institutions affecting the renegotiation decision. . 7. conclusions this study aims to contribute to the study and analysis of the subject of financial renegotiations regarding ppp projects. guasch et al. (2014) contend that the observed high incidence of renegotiations is a serious problem and thus question the effectiveness of ppp programs as a general matter, suggesting an abuse of the instrument and its use for opportunistic reasons rather for well-founded contractual reasons. rozas et al. (2012) note that excess renegotiations are one of the main problems mentioned in the infrastructure development literature. as hasselgren et al. (2014) argue, renegotiation should not be treated casually. it should be used only exceptionally, as the direct effect will typically be adverse to the public interest. the reason to develop useful tools for policy makers when deciding to renegotiate a ppp is well established by ho (2006). in fact, ho (2006) emphasizes two reasons related to the problems of asymmetric information, i.e., the principal-agent problem and moral hazard. the first occurs when a future developer knows that a request for renegotiation is generally (or even always) granted; in that case, he would then have substantial incentive to bid optimistically to w. risso / european journal of government and economics 7(2), 171-199. 193 win the project regardless of costs. the reason that an overly optimistic proposal can have a higher chance of winning is because some crucial and developer-specific information regarding the project is difficult for the government to verify and can be untruthfully revealed in the development proposal as a result. the second reason occurs after the contract is signed: if renegotiation is expected, the developer may engage in inefficient actions that will reduce overall social efficiency but increase the developer’s ultimate payoff. in this context, we employed the dynamic game-theoretic framework involving financial renegotiation developed by ho (2006). in this way, we made contributions regarding the incorporation of payment functions that allow conclusions regarding the influence of certain variables in the renegotiation of a ppp project. in addition, we were able to apply the model in the real world. considering the applied computation of the vfm based on the psc, we propose a method for estimating the convenience of renegotiation versus postponement of the project. this method can be practically computed given that the vfm is generally estimated in many countries such as the uk, which explains the method in coulson (2008). of course, the indicator suffers from the same limitations as the psc. large transfer risk does not assure more vfm as this does not represent an optimal risk allocation between the public and private sectors. a context of weak governance and the lack of regulations does not ensure the reliability of the results. the political costs of retendering or rejecting the project are difficult to estimate, even if in some cases they could be compensated. the effects of asymmetric information affect the indicator. for instance, the winner’s curse can overestimate the vfm. for this reason, we consider the important recommendation of sarmento (2010) to make the effort to conduct a psc prior to the bid. we also derived an alternative r indicator to distinguish when a project was more likely to be renegotiated. according to the indicator obtained, whether a ppp project should be renegotiated positively depends on the social discount rate, public waiting time, extra costs, level of risk transferred, percentage of cost and the public financing interest rate and negatively depends on the level of private risk, the spread required by the financial system and the equity level. in the present work and due to limited information, we applied the second indicator to four case studies of ppp projects in which renegotiations were requested. we note that in the uk case, which has more solid institutions, the decision to not renegotiate is in accordance with the results of the indicator. china also behaves as suggested by the indicator by renegotiating the project. the taiwan case is more complicated due to the financial crisis that affected the region. in this case, large interest rates may affect the vfm of retendering the project and may not be considered in the computation of the indicator. however, the main mistake was to adopt a very important and expensive project as a ppp. finally, portugal seems to be a clear case of weak governance, and the indicator suggested not renegotiating with a probability of 11.39%. however, the government renegotiated the conditions with the developer for reasons explained by ferreira (2014) and marquez and cruz (2011). these authors blame this miscalculation on a weak government position when renegotiating due to personnel changes in the public stakeholders (there were 8 different ministers, 3 secretaries of state and 3 project leaders w. risso / european journal of government and economics 7(2), 171-199. 194 involved) and poor government preparation of concession contracts. it is clear that asymmetric information and know-how differences between the companies and the government are large, as suggested by marques (2017). note also that portugal, as mentioned above, has the worst rank (position 95) of the four countries in the public-sector performance index compared to the rest of the four countries. in fact, considering the efficiency of the legal framework for settling disputes, portugal ranks 121st out of 137 countries. marques (2017) asserts that contract management is not enough and does not replace renegotiation, highlighting the important role of an independent regulator. the r indicator is more theoretical and its application in this paper is based on poor information about the different analyzed projects because the indicator is used to study relations between economic and financial variables and the correct decision. however, we consider that the alternative indicator based on the psc could be estimated by the government, since it utilizes many of the available tools applied to compute the vfm. although the indicator can be improved, it can be used as an additional tool when analyzing the convenience of renegotiation. references allan, j. (2001). public-private partnerships: a review of literature and practice. saskatchewan institute of public policy. anderlini, l., & felli, l. (2001). costly bargaining and renegotiation. econometrica 69(2), 377-411. doi: https://doi.org/10.1111/1468-0262.00196 arvan, l., & leite, a. (1990). cost overruns in long term projects. international journal of industrial organization 8(3), 443-467. doi: https://doi.org/10.1016/0167-7187(90)90007-n athias, l., & nuñez, a. (2008). winner’s curse in toll road concessions. economics letters 101(3), 172-174. https://doi.org/10.1016/j.econlet.2008.07.017 australian government (2008). national public private partnership guidelines. volume 4: public sector comparator guidance. baeza, m. d. l. á., & vassallo, j. m. (2010). private concession contracts for toll roads in spain: analysis and recommendations. public money & management 30(5), 299-304. doi: https://doi.org/10.1080/09540962.2010.509179 bansal, n., & kalady, r. (2013). study on project schedule and cost overruns: expedite infrastructure project, klynveld peat marwick goerdeler (kpmg)-project management institute (pmi), india bertisen, j., & davis, g. (2008). bias and error in mine project capital cost estimation. the engineering economist 53(2), 118-139. doi: https://doi.org/10.1080/00137910802058533 bi, x., & wang, h. (2011). the control of moral hazard in ppp project renegotiation. in industrial engineering and engineering management (ie&em), 2011 ieee 18th international conference on institute of electrical and electronics engineers (ieee), (pp. 232-236), hong kong, china. https://doi.org/10.1111/1468-0262.00196 https://doi.org/10.1016/0167-7187(90)90007-n https://doi.org/10.1016/j.econlet.2008.07.017 https://doi.org/10.1080/09540962.2010.509179 https://doi.org/10.1080/00137910802058533 w. risso / european journal of government and economics 7(2), 171-199. 195 doi: https://doi.org/10.1109/icieem.2011.6035147 bing, l., akintoye, a., edwards, p., & hardcastle, c. (2005). the allocation of risk in ppp/pfi construction projects in the uk. international journal of project management 23(1), 25-35. doi: https://doi.org/10.1016/j.ijproman.2004.04.006 bitran, e., nieto-parra, s., & robledo, j. (2013). opening the black box of contract renegotiations. organisation for economic co-operation and development (oecd) development center, working paper no. 317 cantarelli, c., flyvbjerg, b., molin, e., & wee, b. (2010). cost overruns in large-scale transportation infrastructure projects: explanations and their theoretical embeddedness. european journal of transport and infrastructure 10 (1), 5-18. chan, a., yeung, j., yu, c., wang, s., & ke, y. (2010). “empirical study of risk assessment and allocation of public-private partnership projects in china. journal of management in engineering 27(3), 136-148. doi: https://doi.org/10.1061/(asce)me.1943-5479.0000049 chang, z. (2013). public–private partnerships in china: a case of the beijing no. 4 metro line. transport policy, 30, 153-160. doi: https://doi.org/10.1016/j.tranpol.2013.09.011 coulson, a. (2008). value for money in pfi proposals: a commentary on the 2004 uk treasury guidelines for public sector comparators, institute of local government studies, university of birmingham. cruz, c., marques, r., & cardoso, p. (2014). empirical evidence for renegotiation of ppp contracts in the road sector. journal of legal affairs and dispute resolution in engineering and construction, 7(2). doi: https://doi.org/10.1061/(asce)la.1943-4170.0000151 cruz, c., & marques, r. (2013). exogenous determinants for renegotiating public infrastructure concessions: evidence from portugal. journal of construction engineering and management 139(9), 1082-1090. doi: https://doi.org/10.1061/(asce)co.1943-7862.0000710 cruz, c., & marques, r. (2011). revisiting the portuguese experience with public-private partnerships. african journal of business management 5(11), 4023-4032. cui, c., liu, y., hope, a., & wang, j. (2018). review of studies on the public–private partnerships (ppp) for infrastructure projects. international journal of project management 36(5), 773794.doi: https://doi.org/10.1016/j.ijproman.2018.03.004 dewatripont, m. (1988). commitment through renegotiation-proof contracts with third parties. the review of economic studies, 55(3), 377-390. doi: https://doi.org/10.2307/2297390 dnp (2009). comparador público-privado para la selección de proyectos app, departamento nacional de planeación (dnp), ministerio de hacienda y crédito público, colombia. dper (2007). public private partnership. value for money and the public private partnership. procurement process, department of public expenditure and reform (dper) http://ppp.gov.ie dutzik, t., schneider, j., & baxandall, p. (2011). high-speed rail: public, private or both? assessing the prospects, promise and pitfalls of public-private partnerships. engel, e., fischer, r., & galetovic, a. (2010). the economics of infrastructure finance: publicprivate partnerships versus public provision. eib papers 15(1), 40-69. engel, e., fischer, r., & galetovic, a. (2014). renegotiations in public-private partnerships: theory https://doi.org/10.1109/icieem.2011.6035147 https://doi.org/10.1016/j.ijproman.2004.04.006 https://doi.org/10.1061/(asce)me.1943-5479.0000049 https://doi.org/10.1016/j.tranpol.2013.09.011 https://doi.org/10.1061/(asce)la.1943-4170.0000151 https://doi.org/10.1061/(asce)co.1943-7862.0000710 https://doi.org/10.1016/j.ijproman.2018.03.004 https://doi.org/10.2307/2297390 w. risso / european journal of government and economics 7(2), 171-199. 196 and evidence. international transport forum discussion papers, 2014/17, organisation for economic co-operation and development (oecd) publishing, paris. esty, b. (2000). big deals: financing large-scale investments. working knowledge: a report on research at harvard business school, 4(2). fatokun, a., akintoye, a., lasanthi, c., (2015). renegotiation of public private partnership road contracts: issues and outcomes. in: 31st annual arcom conference, 7 9 september 2015, lincoln, pp. 1249-1258. ferreira, b. (2014). case study: the metro transportes do sul concession & risk sharing and public financial compensations, doctoral dissertation, universidade católica portuguesa, portugal. flyvbjerg, b., holm, m., & buhl, s. (2002). underestimating costs in public works projects: error or lie? journal of the american planning association 68 (3), 279-295. doi: https://doi.org/10.1080/01944360208976273 flyvbjerg, b., skamris holm, m., & buhl, s. (2003). how common and how large are cost overruns in transport infrastructure projects? transport reviews, 23 (1), 71-88. doi: https://doi.org/10.1080/01441640309904 flyvbjerg, b., skamris holm, m., & buhl, s. (2004). what causes cost overrun in transport infrastructure projects? transport reviews, 24 (1), 3-18. doi: https://doi.org/10.1080/0144164032000080494a flyvbjerg, b. (2007). cost overrun and demand shortfalls in urban rail and other infrastructure, transportation planning and technology 30 (1), 9-30. doi: https://doi.org/10.1080/03081060701207938 gifford, j., bolaños, l., & daito, n. (2014). renegotiation of transportation public-private partnerships: the us experience. international transport forum discussion papers, 2014/16, organisation for economic co-operation and development (oecd) publishing, paris. grossman, s., & hart, o. (1986). the costs and benefits of ownership: a theory of vertical and lateral integration. journal of political economy 94, 691–719. doi: https://doi.org/10.1086/261404 guasch, j., benitez, d., portabales, i., & flor, l. (2014). the renegotiation of ppp contracts: an overview of its recent evolution in latin america. international transport forum discussion papers, 2014/18, organisation for economic co-operation and development (oecd) publishing, paris. hallikeri, v. (2015). how not to bungle a public-private partnership (ppp): lessons from the london underground ppps, the institute for business in the global context, vol. 12. hart, o. (1995). firms, contracts, and financial structure. oxford university press. doi: https://doi.org/10.1093/0198288816.001.0001 hart, o., & moore, j. (1988). incomplete contracts and renegotiation. econometrica 56(4), 755-785. doi: https://doi.org/10.2307/1912698 hart, o., & moore, j. (1990). property rights and the nature of the firm. journal of political economy 98, 1119–58. doi: https://doi.org/10.1086/261729 hasselgren, b., makovsek, d., & perkins, s. (2014). public private partnerships for transport https://doi.org/10.1080/01944360208976273 https://doi.org/10.1080/01441640309904 https://doi.org/10.1080/0144164032000080494a https://doi.org/10.1080/03081060701207938 https://doi.org/10.1086/261404 https://doi.org/10.1093/0198288816.001.0001 https://doi.org/10.2307/1912698 https://doi.org/10.1086/261729 w. risso / european journal of government and economics 7(2), 171-199. 197 infrastructure: renegotiations, how to approach them and economic outcomes roundtable summary and conclusions, international transport forum discussion papers, 2014/25, organisation for economic co-operation and development (oecd) publishing, paris. ho, s. (2006). model for financial renegotiation in public-private partnership projects and its policy implications: game theoretic view. journal of construction engineering and management 132(7), 678-688. doi: https://doi.org/10.1061/(asce)0733-9364(2006)132:7(678) ho, s. (2009). government policy on ppp financial issues: bid compensation and financial renegotiation. policy, finance & management for public-private partnerships, willey-blackwell, new delhi, 267-300. doi: https://doi.org/10.1002/9781444301427.ch15 horman, m. (2001). modeling the effects of lean capacity strategies on project performance. in proc. ninth annual conference of the international group for lean construction. iossa, e., & martimort, d. (2012). risk allocation and the costs and benefits of public-private partnerships. rand journal of economics 43(3), 442–474. doi: https://doi.org/10.1111/j.17562171.2012.00181.x iossa, e., & martimort, d. (2015). the simple micro-economics of public–private partnerships. journal of public economic theory 17(1), 4-48. doi: https://doi.org/10.1111/jpet.12114 iossa, e., & martimort, d. (2016). corruption in ppps, incentives and contract incompleteness. international journal of industrial organization 44(c), 85-100. doi: https://doi.org/10.1016/j.ijindorg.2015.10.007 iossa, e., & saussier, s. (2018). public private partnerships in europe for building and managing public infrastructures: an economic perspective. annals of public and cooperative economics 89(1), 25-48. doi: https://doi.org/10.1111/apce.12192 ishiguro, s., & itoh, h. (2001). moral hazard and renegotiation with multiple agents. the review of economic studies 68(1), 1-20. https://doi.org/10.1111/1467-937x.00157 kdi (2010). detailed guideline for value-for-money test for buil-transfer-operate (bto) publicprivate partnership (ppp) projects, korea development institute (kdi) public and private infrastructure investment management center, south korea. ke, y., wang, s., chan, a., & lam, p. (2010). preferred risk allocation in china’s public–private partnership (ppp) projects. international journal of project management, 28(5), 482-492. doi: https://doi.org/10.1016/j.ijproman.2009.08.007 kennedy, g. (2013). can game theory be used to address ppp renegotiations? a retrospective study of the of the metronet-london underground ppp, doctoral dissertation, universidade católica portuguesa, portugal. lee, j. (2008). cost overrun and cause in korean social overhead capital projects: roads, rails, airports, and ports. journal of urban planning and development 134 (2), 59-62. https://doi.org/10.1061/(asce)0733-9488(2008)134:2(59) li, r. (2017). infrastructure of ppp mode in chinese-with beijing metro line as an example. scholars journal of engineering and technology, 5(3), 117-121. liu, t., & wilkinson, s. (2015). can the pilot public-private partnerships project be applied in future urban rail development?: a case study of beijing metro line 4 project. built https://doi.org/10.1061/(asce)0733-9364(2006)132:7(678) https://doi.org/10.1002/9781444301427.ch15 https://doi.org/10.1111/j.1756-2171.2012.00181.x https://doi.org/10.1111/j.1756-2171.2012.00181.x https://doi.org/10.1111/jpet.12114 https://doi.org/10.1016/j.ijindorg.2015.10.007 https://doi.org/10.1111/apce.12192 https://doi.org/10.1111/1467-937x.00157 https://doi.org/10.1016/j.ijproman.2009.08.007 https://doi.org/10.1061/(asce)0733-9488(2008)134:2(59) w. risso / european journal of government and economics 7(2), 171-199. 198 environment and property management: a focus on china and hong kong, 205. love, p., sing, c., carey, b., & kim, j. (2014). estimating construction contingency: accommodating the potential for cost overruns in road construction projects. journal of infrastructure systems 21(2), doi: https://doi.org/10.1061/(asce)is.1943-555x.0000221 makovšek, d., & veryard, d. (2016). the regulatory asset base and project finance models. in international transport forum, discussion paper 2016-1. mappp (2008). guide d’utilisation du modèle financier d’evaluation préalable, mission d’appui aux partenariats public-privé (mappp), france. marques, r. (2017). why not regulate ppps? utilities policy 48(c), 141-146. doi: https://doi.org/10.1016/j.jup.2017.04.003 marques, r., & berg, s. (2010). revisiting the strengths and limitations of regulatory contracts in infrastructure industries. journal of infrastructure systems, 16(4), 334-342. doi: https://doi.org/10.1061/(asce)is.1943-555x.0000029 maskin, e., & moore, j. (1999). implementation and renegotiation. the review of economic studies, 66(1), 39-56. doi: https://doi.org/10.1111/1467-937x.00077 mef (2014). metodología para la aplicación del análisis cuantitativo en la elección de la modalidad de ejecución de proyectos de inversión cofinanciados, ministerio de economía y finanzas (mef) de perú, agosto 2014. mef-cnd (2012). determinación del valor por dinero: guía metodológica del comparador públicoprivado para esquemas de participación público-privada en uruguay, ministerio de economía y finanzas (mef) y corporación nacional para el desarrollo (cnd), uruguay. nao (2000). the financial analysis for the london underground public private partnerships, national audit office (nao), london, uk. nao (2009). department of transport-the failure of metronet, national audit office (nao), london, uk. nikolaidis, n., & roumboutsos, a. (2013). a ppp renegotiation framework: a road concession in greece. built environment project and asset management 3(2), 264-278. doi: https://doi.org/10.1108/bepam-05-2012-0031 ntsa (2004). sudafrica public-private partnership manual-module 4: ppp feasibility study, national treasury of south africa (ntsa), pp. 63-66. pbc (2003). the public sector comparator-a canadian best practice guide, partnership british columbia (pbc), canada. rozas, p., bonifaz, j., & guerra-garcía, g. (2012). el financiamiento de la infraestructura: propuestas para el desarrollo sostenible de una política sectorial, comisión económica para américa latina y el caribe (cepal), santiago de chile, chile. sarmento, j. (2010). do public-private partnerships create value for money for the public sector? the portuguese experience. oecd journal on budgeting 10(1), 1-27. doi: https://doi.org/10.1787/budget-10-5km8xx3fgws5 segal, i. (1999). complexity and renegotiation: a foundation for incomplete contracts. the review of economic studies 66(1), 57-82. doi: https://doi.org/10.1111/1467-937x.00078 https://doi.org/10.1061/(asce)is.1943-555x.0000221 https://doi.org/10.1016/j.jup.2017.04.003 https://doi.org/10.1061/(asce)is.1943-555x.0000029 https://doi.org/10.1111/1467-937x.00077 https://doi.org/10.1108/bepam-05-2012-0031 https://doi.org/10.1787/budget-10-5km8xx3fgws5 https://doi.org/10.1111/1467-937x.00078 w. risso / european journal of government and economics 7(2), 171-199. 199 shcp (2010). manual del comparador público privado, secretaría de hacienda y crédito público (shcp), ministerio de hacienda y crédito público, méxico. sheikh, s., asher, m., & ramakrishnan, v. (2015). a case study of metronet ppp in the uk: implications for india. lee kuan yew school of public policy research paper, no 15-26. doi: https://doi.org/10.2139/ssrn.2648788 singh, r., (2010). delays and cost overruns in infrastructure projects: extent, causes and remedies. economic and political weekly 45 (21), 43-54. tiong, r. l. k., & alum, j. (1997). final negotiation in competitive bot tender. journal of construction engineering management 123(1), 6-10. doi: https://doi.org/10.1061/(asce)07339364(1997)123:1(6) tirole, j., (1999). incomplete contracts: where do we stand? econometrica 67(4), 741-781. doi: https://doi.org/10.1111/1468-0262.00052 utfp (2009). la misurazione del value for money nell’esperienza italiana e straniera: analisi dei rischi e psc, unità tecnica finanza di progetto (utfp), dipartimento per la programmazione e il coordinamento della política economica, march 2009. vdtf (2001). partnerships victoria: risk allocation & contractual issues, victorian department of treasury & finance (vdtf), pp. 178-191. wall, d. (1997). distributions and correlations in monte carlo simulation, construction management & economics 15(3), 241-258. doi: https://doi.org/10.1080/014461997372980 world bank (2012). public-private partnerships reference guide version 1.0, washington, us, p. 230. https://doi.org/10.2139/ssrn.2648788 https://doi.org/10.1061/(asce)0733-9364(1997)123:1(6) https://doi.org/10.1061/(asce)0733-9364(1997)123:1(6) https://doi.org/10.1111/1468-0262.00052 https://doi.org/10.1080/014461997372980 ejge_front_7_2 ejge oficial 7-2 content page 7-2-5 abstract. this paper analyzes the renegotiation problem in the context of public-private partnership projects. utilizing a game-theoretic approach, an equilibrium is found in which the government finds that accepting renegotiation can be efficient. a ... keywords. public-private partnership, renegotiation, value for money, game theory doi. https://doi.org/10.17979/ejge.2018.7.2.4508 world bank (2012). public-private partnerships reference guide version 1.0, washington, us, p. 230. european journal of government and economics 10(1), june 2021, 5-29 european journal of government and economics issn: 2254-7088 the effect of euribor on banking profitability: evidence from the spanish banking system david boto-garcía a*, antonio álvarez a, josé francisco baños pinoa a department of economics, university of oviedo, campus del cristo, 33006 oviedo, spain * corresponding author at: davidboto93@gmail.com abstract. this paper studies the relationship between the euribor rate and the return on average assets (roaa) of the spanish banking sector. we use quarterly time series data for the period 1995-2016. our analysis also controls for bank factors, market concentration, the macroeconomic environment and time effects. the main purpose is to examine how the sector's roaa varies with the slope of the yield curve. we find evidence of an inverse relationship between the euribor rate and profitability. we also show that banking profitability is pro-cyclical and is positively related to the stock of performing loans and the national 10-year bond yield. keywords. banking profitability; time series; euribor; interest rate spread; unconventional monetary policy; yield curve jel codes. c01; c22; e44; g21 doi. https://doi.org/10.17979/ejge.2021.10.1.7083 1. introduction a solid banking system has a major impact on the performance of the real economy. as proved in the 2008 financial crisis, when banks run into economic difficulties, their problems spread out to the whole economy. indeed, the potential bankruptcy of the banking sector can lead other sectors to failure, thereby producing systemic crises. this is especially relevant for spain, since its business structure strongly depends upon bank financing (hoffman & sorensen, 2015). in recent years, the banking business has suffered a strong deterioration in its balance sheets. on the one hand, the number of unproductive assets has significantly increased as a consequence of the 2008 economic crisis. apart from not producing revenues, these assets also carry non-negligible costs. this has forced the private sector to conduct a slow deleveraging, which in turn has limited loans' growth. additionally, authorities have increased capital requirements, forcing banks to obtain more profits to attain the same level of profitability. the recovery of profitability to levels prevailing before the great recession is one of the great challenges the sector is facing. the level of profitability of the banking sector is a good indicator of the stability of the financial system. low levels of profitability can induce banks to assume greater risks (bikker & vervliet, 2018), thereby increasing the exposure to systemic crises. in recent years, the european central bank (ecb) has developed an expansive monetary policy that has led to a sharp decrease in the mailto:davidboto93@gmail.com https://doi.org/10.17979/ejge.2021.10.1.7083 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 6 interbank euribor rate, which has even turned negative.1 this policy has important consequences for the stability of the banking system because banks substitute wholesale funding with ecb funding (alvarez et al., 2019). since most of the asset products' rates are linked to euribor and given that a recovery to the pre-crisis levels is not expected in the short run, it is unclear how the downturn in the euribor rate affects bank business and, therefore, the stability of the banking system. using aggregate quarterly data for the whole banking sector covering the period 1995-2016, this paper studies the evolution of the spanish banking sector's average profitability. we specially focus on the effect of the slope of the yield curve (difference between longand short-run interest rates) on the return on average assets. our aim is to explore how the central bank monetary policy in terms of the 12-month euribor rate has affected bank profits. our regression analysis also controls for other relevant banking variables like the stock of performing loans or the ecb assets, the macroeconomic environment measured through the gross domestic product, the market concentration of the sector through the herfindahl-hirschman index and the c5 ratio, and some time effects. banking profitability has been widely studied, both internationally (molyneux & thornton, 1992; garcia-herrero et al., 2009) and in spain (trujillo-ponce, 2013; climent & pavía, 2015). most of these studies use panel datasets with disaggregated information for each bank. we instead use aggregate time series data taking the return on average assets (roaa) of the banking sector as our dependent variable. this is because our objective is not to study the determinants of bank profitability, in which case we would need to exploit the cross-sectional variability across banks, but examine how the sector's average profitability relates to interest rates. our approach has the additional advantage that it avoids the problem of dealing with the great number of mergers between banks that took place during the study period. there is an emerging body of literature concerned about the impact of interest rates and central banks' regulation on bank performance (e.g., borio et al., 2017; chaudron, 2018; lópez et al., 2020). the paper adds to this line of research by exploring how the sector's profitability has been affected by the slope of the yield curve, measured here as the difference between the 10year sovereign bond yield and the 12-month euribor rate. while most studies analyse the role of the spread alone, we conduct two separate analyses, including i) the spread and ii) the two interest rates separately. this allows us to explore the different effects of variations in the slope of the yield curve depending on whether the change is due to variations in the long-term or the short-term interest rate. furthermore, we also study whether the 12-month euribor exerts a different effect on the sector's profitability depending on its level. the paper is structured as follows. in section 2, we survey the related literature. section 3 outlines the empirical model, the database and the variables employed. in section 4, we present the results of the empirical analysis. finally, section 5 discusses the main conclusions. 1 the 12-month euribor rate (nominal terms) was -0.08% by the end of 2016 after suffering a decrease of 100.76 % since 1995. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 7 2. literature review following the pioneering works by flannery (1981) and ho & saunders (1981), many studies have examined banking profitability determinants, either for a multi-country-setup (molyneux & thornton, 1992; demirgüc-kunt & huizinga, 1999; carbó & rodríguez, 2007; albertazzi & gambacorta, 2009; dietrich & wanzenried, 2014; saona, 2016) or focusing on a single banking system (athanasoglou et al., 2008; sufian & habibullah, 2009; dietrich & wanzenried, 2011; garcía-herrero et al., 2009). most of this literature employs panel datasets of banks and relates indicators of profitability to bank-specific characteristics, macroeconomic conditions, and the degree of market concentration in the sector. these studies generally find that bank profits are pro-cyclical and depend on funding costs, the growth of total loans, market concentration and central bank official interest rates. because of its important policy implications, there has been a particular interest in disentangling the relationship between market interest rates and profitability. scholars typically argue that bank net interest margins and, therefore, profitability is higher when the yield curve (the difference between longand short-term interest rates) is steeper. this is because banks' revenues mainly depend on the margin between borrowing funding in the short-run and lending in the long run. importantly, this does not only hold in the short run but also considering long time spams (e.g. busch & memmel, 2017). alessandri & nelson (2015) develop a model of monopolistically competitive banks subject to repricing frictions. based on panel data for uk banks, they show that both the level and the slope of the yield curve positively contribute to profitability. similarly, borio et al. (2017) document a positive relationship between bank returns and the yield curve, which is more significant when rates are low. egly et al. (2018) find that the yield curve together with real gdp growth exert positive and significant effects on the net interest margin of u.s. commercial banks. using a panel dataset of banks from oecd countries during the period 2010-2015, cruz-garcia et al. (2018) show that both the 3-month euribor rate level and the slope of the yield curve exert a positive (although at a decreasing rate) effect on the intermediation margin. this suggests that the effect of the interest rates is greater when they are low. cruz-garcía et al. (2019) report that if the spread between the long-term and the short-term interest rates were increased by 50 basis points, the return of assets would rise by 25bp. notwithstanding this, changes in the slope of the yield curve do not only affect profitability through the intermediation margin but can also have indirect effects through other non-financial revenues (cruz-garcia et al., 2018). accordingly, variations in the spread might have ambiguous effects on profitability. in this sense, the empirical evidence is far from robust. the effect of the slope of the yield curve on profitability seems to also depend on the countries analysed and the time periods covered. based on separate regressions using annual aggregate data for ten countries during the period 1979-2001, english (2002) show important heterogeneity in the relationship between the yield curve slope and net interest margins. whereas the effect is negative for germany, norway, sweden and switzerland, it is positive for the united states and non-significant for the rest. accordingly, the relationship between the interest rates spread and bank profitability is market-specific. this calls for more country-specific studies. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 8 following the 2008 economic crisis, many central banks have introduced a negative interest rate policy (nirp) with the purpose of boosting real spending through increased bank loans' supply and demand. in addition to this, some central banks like the ecb increased their liquidity as a response to the financial turmoil of the great recession. an emerging body of literature has started to study the economic impacts of these unconventional monetary policies (umps) on the profitability of the banking sector. these policies affect the banking sector through different channels. a summary of their effects on the euro area, japan and the united kingdom can be found in dell'ariccia et al. (2018). on the one hand, quantitative easing tends to depress long-term interest rates, which flattens the yield curve and reduces the intermediation margin. since deposit rates are close to the zerolower bound and banks are reluctant to pass negative rates to their customers, this reduces profitability. as a response to this, banks with a heavy reliance on deposit funding take more risks and lower their lending standards, which in turn threatens their financial stability (bikker & vervliet, 2018; heider et al., 2019). lambert & ueda (2014) find that umps increase bank medium-term credit risk in the usa, the euro area and the united kingdom and have ambiguous effects on bank profitability. on the other hand, umps are expected to increase the demand for credit and to reduce the stock of non-performing loans in a low-interest rates environment, which might increase profitability through a quantity effect. nevertheless, borio & gambacorta (2017) report that the low-interest rates environment is not effective in stimulating lending growth. using data for 33 oecd countries, molyneux et al. (2020a) show that bank lending is weaker in nirp-adopter countries. they document that this adverse effect of the policy is stronger for banks that are more dependent on retail deposit funding and operate in more competitive markets. using the same dataset, molyneux et al. (2019) find that the nirp is negatively associated with banking profitability. similarly, mamatzakis & bermpei (2016) report that central bank's assets and excess reserves negatively affects bank performance. considering data from 47 countries, claessens et al. (2018) show that a one per cent drop in interest rates translates into an 8 basis point lower net interest margin. this effect is larger at low rates. as a response to the low-interest rates environment, banks also change the composition of their income in various ways. molyneux et al. (2020b) show that banks boost fees from portfolio management, brokerage, consultancy services and current accounts. brei et al. (2020) report that the low-interest rate environment has caused banks to shift their activities from interestgenerating to fee-related and trading activities. other studies do not find a significant relationship between the nirp and bank profits. altavilla et al. (2018) report that the monetary policy easing has not significantly affected banks' profits and that the negative effect of low rates on profits is counterbalanced by improved macroeconomic conditions. exploiting a large dataset involving 5,200 banks from 27 european and asian countries, lópez et al. (2020) find little overall impact of the current negative nominal rates on bank profitability. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 9 for the spanish case, some scholars have studied the interrelations between efficiency, technical change and deregulations (kumbhakar et al., 2001) whereas others have paid attention to the effect of regulatory changes on the banks' negotiating power (salas & saurina, 2003). another stream of research has explored the market failures behind the great difficulties the banking sector went through after the 2008 crisis (jimeno & santos, 2014). these studies generally agree that the high exposure to the real estate bubble and the excessive risk-taking through over lending during the boom period are the main causes of the financial difficulties that follow. as discussed in fernández-villaverde et al. (2013), political economy factors like governments, developers and the local regulation of cajas further contributed to the huge indebtedness of the spanish banking system. concerning bank profitability, trujillo-ponce (2013) finds that the business cycle, the inflation rate and the degree of concentration in the sector positively contribute to profitability, while a high proportion of low-quality assets tends to lower it. climent & pavía (2015) study the differences in profitability between firms receiving public aid and those which do not. revenues from investments and net financial operation results have a positive impact on profitability, whereas staff and administration costs and provisions for depreciation have negative effects. pérezmontes & ferrer pérez (2018) analyze how bank profits are affected by interest rates. using aggregate time series data and covering the 2000-2016 period, these authors find a non-linear relationship between interest rates and net interest income, which is positive at low levels. more recently, cruz-garcía et al. (2020b) analyze the determinants of banks' net interest margins. they show that market power and risk aversion increase banking profitability while a higher volume of liquid reserves reduces net interest margins. tercero-lucas (2020) examines the effects of non-standard monetary policy measures on the spanish banking sector profitability. his analysis covers the period 2001-2017 and uses panel data for 54 spanish banks. he provides robust evidence that ecb's total assets, excess reserves and the slope of the yield curve do not significantly affect banks' return on assets and interest margins. 3. empirical model and data 3.1. empirical model while most studies about banking profitability adopt a firm-oriented approach, we use time series data to study the behavior of the banking sector as a whole. time series techniques for analyzing the banking industry in spain have been previously used by repullo et al. (2010) and pérezmontes and ferrer-pérez (2018). this avoids the problem of dealing with mergers between entities, which have been very important and numerous in spain over the study period.2 the general model to be estimated has the following linear form: 𝛱𝛱𝑡𝑡 = 𝛼𝛼 + 𝛽𝛽𝛽𝛽𝑡𝑡 + 𝜀𝜀𝑡𝑡 (1) 2 tercero-lucas (2020) overcomes the problem of mergers in a panel data setting by constructing “new virtual entities” through adding their balance sheets. in our view, treating as a single unit banks that (in the past) behave as separate entities because latter they merged is a strong assumption. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 10 where subscript t indicates time (quarters), for 𝑡𝑡 = 1995𝑄𝑄1, … , 2016𝑄𝑄4, 𝛱𝛱𝑡𝑡 is an indicator of the profitability of the aggregate banking sector in period t, 𝛽𝛽𝑡𝑡 includes a set of banking factors together with controls for the macroeconomic environment and an indicator of market concentration, 𝛽𝛽 is a vector of parameters to be estimated, and 𝜀𝜀𝑡𝑡 is the error term that is assumed to be white noise. we next proceed to define the variables considered in equation (1) and the data source. 3.2. variable definition and data source banking profitability (𝛱𝛱𝑡𝑡) return on average assets (roaa) is our main measure of profitability. it is defined as the ratio of net before-tax profits to total assets. it reflects how the banks' assets have been managed to produce revenues and are the most used profitability indicator (molyneux et al., 2019; mamatzakis & bermpei, 2016). some scholars also employ the return on equity (roe) as the dependent variable (dietrich & wanzenried, 2011; trujillo-ponce, 2013). for robustness, we also consider roae. these two ratios are drawn from the income statement of the spanish banking sector provided by the bank of spain statistical bulletin. although there are some differences between roaa and roae (for example, the former collects off-balance sheet activities whereas the latter varies as a function of the entities capital requirements), their time evolution is similar in our study period. as shown in figure 1, both variables show a substantial drop in the period 2012-2013. this was the consequence of the passing of two royal decree-laws in february and june 2012 by the spanish government that forced banks to raise provisions in the balance sheet. one important part of the assets was related to the real state sector which, after the bursting of the housing bubble, were appraised above market prices.3 in this context, it was necessary to deploy resources towards provisioning to cover losses from asset impairment. the above-mentioned decrees, by requiring provisions classified as a normal risk to cover the exposure of the real state sector, led to a substantial negative value of the roaa in 2012 and 2013.4 3 martin et al. (2020) show that during the housing boom there was a slowdown in the growth of non-housing credit. as such, banks' assets at that period were heavily dependent on housing credit. 4 the effort in provisioning (provisions over gross margin) in the period 2000-2008 was on average 25%, and it started to grow with the economic crisis. in 2011, it was 79% but in 2012 it reached the value of 253%. this atypical value has a corresponding effect on profits before taxes. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 11 figure 1. evolution over time of roaa and roae (in percentage). banking factors (𝛽𝛽𝑡𝑡) • performing loans: the main activity of any bank is loan granting. theory suggests that the greater the number of loans granted, the higher the results, ceteris paribus. however, profits do not only depend on the loans' volume but also on their quality. specifically, the relevant ones are performing loans, namely, those which generate earnings. when customers cannot repay a loan –which results in the loan being catalogued as in default – ,5 the asset generates costs related to its management and the required provisioning allowance. due to these reasons, we employ the stock of total loans once discounted those considered as defaulted. we expect this variable to be positively associated with profitability. the performing loans variable has been obtained as the difference between the total loans stock and the defaulted loans per quarter. this information is obtained from the bank of spain statistical bulletin. the original variable has been deflated by the consumer price index (base 2011) and is measured in million euros. • ecb assets: as discussed in section 2, the european central has undertaken several unconventional monetary policies (umps). for instance, by providing liquidity in exchange for low-quality assets, banks have been allowed to cleanse deteriorated balance sheets. mamatzakis & bermpei (2016) document that increases in the fed' assets are negatively associated with bank profitability in the usa, while tercero-lucas does not find a significant effect of ecb's assets growth on spanish bank's roa. 5 the bank of spain defines defaulted loans as those with any expired amounts, either of the principal, the interests or of the expenditures contractually agreed, with more than three months of delay. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 12 to control for this quantitative easing (liquidity injections), we take the value of ecb assets for the euro area as an indicator of the asset purchase programs. this variable is retrieved from the ecb statistical database and is expressed in million euros (also deflated by the consumer price index). since previous literature on its effect is scarce, we do not have a clear expectation about the sign of this variable. • 12-month euribor: banks use euribor as a reference for fixing the prices of many of their financial products. the effect of a rise in the euribor rate on profitability may depend on how they are able to shift this increase onto asset prices. on the one hand, because some of the asset products have an interest rate that is linked to euribor (e.g. mortgages with variable rates), if euribor rises the interest rates banks charge on their variable-rate loans will also increase, although with a lag as rates on most of these loans are revised every six months. on the other hand, if there is an increase in the funding cost in the inter-bank market, banks may compete with each other to fund themselves via deposits, thus pushing liability rates up. therefore, the total effect of euribor on profits will depend on the relative evolution of asset and liability rates. in practice, liability rates tend to reprice faster than asset rates, so it is possible that a decrease in euribor causes an increase in the slope of the interest rate curve. this does not imply that net interest income will necessarily increase, as it also depends on the ""volume"" effect. it thus remains to be empirically answered the effect of the euribor rate on profitability. here we employ the quarterly 12-month euribor rate after discounting the inter-annual inflation rate. the data is drawn from the bank of spain, which supplies the value it would have taken in the period 1995-2000 as a weighted average of the national inter-bank interest rates of the euro zone countries in that period. for robustness, we also use the 3month euribor rate, obtained from the same data source. • 10-year sovereign bond yield: banks usually obtain funding in the short-run, but they lend in the long run (maturity mismatch). consequently, to analyse the effect on profitability of a change in interest rates, it seems better to consider the difference between long-run and short-run interest rates. this is also known as the slope of the yield curve. indeed, the sector has traditionally considered that the steepness of the yield curve positively contributes to bank results. in this vein, saunders & schumacher (2000), albertazzi & gambacorta (2009), alessandri & nelson (2015) and borio et al. (2017), among others, find a positive relationship between the slope of the yield curve and banking profitability. the real 10-year spanish bond yield is the chosen indicator of the long-run interest rate. the data is drawn from the bank of spain and is expressed in real terms. in our analysis, we both consider: i) the rates' difference as a single explanatory variable, and ii) both rates separately. in this way, by including the rate for the long-run together with the one for the short-run, we allow for different effects of changes in the slope of the yield curve depending if they are due to shifts in the short-run or in the long-run interest rates. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 13 in line with the literature, we expect profitability to be positively associated with the rate spread. consequently, we also expect the parameter associated with the 10-year spanish bond yield to be positive and the one associated with the 12-month euribor to be negative. macro environment (𝑍𝑍𝑡𝑡) among the different macroeconomic indicators, gross domestic product (gdp) is the main magnitude due to the close relationship between business cycle fluctuations and banking profitability (demirguc-kunt & huizinga, 1999; albertazzi & gambacorta, 2009; bolt et al., 2012; dietrich & wanzenried, 2014; almeida & divino, 2015).6 • real gdp: a low economic growth environment reduces lending activity because of the drop in aggregate demand. conversely, economic expansions increase the demand for credit and other types of financial products. athanasoglou et al. (2008) find evidence of a positive relationship between bank profits and the business cycle. bolt et al. (2012) indicate that the degree of pro-cyclicality of bank profits is stronger for deep recessions than under normal economic conditions. thus, we expect a positive coefficient for this variable. real gdp (index) for each quarter has been taken from the spanish national accounting (national statistics institute) taking december 2010 as the base year, once corrected for seasonal and calendar effects. sector concentration (𝐶𝐶𝑡𝑡) in line with the structure-conduct-performance paradigm, the market structure is usually considered as a determinant of banking profitability.7 to measure the degree of market concentration, scholars use the herfindahl-hirschman index (djalilov & piesse, 2016) or concentration ratios such as c3 (dietrich & wanzenried, 2014) or c5 (claessens & laeven, 2004). • c5: it is defined as the proportion of the total assets under the control of the five largest banks. therefore, the c5 index can be understood as a proxy of market concentration. although market concentration does not necessarily imply a greater market power, when a reduced number of banks control a large share of total assets it is more likely that collusive behaviours take place (gilbert, 1984). although it is unclear whether structure itself necessarily shapes profitability (smirlock, 1985), it can contribute to the build-up of monopoly profits. from this perspective, we would expect a positive sign for this variable. banks assets have been taken at the end of each quarter from the public balance sheets available in ceca (spanish savings banks confederation) and aeb (spanish banking association). as the information from ceca for the savings banks is only available for the 6 the labour market situation or the population (as an indicator of market size) are other variables that some authors have considered. nevertheless, since demographic changes are mainly due to migratory flows associated with the business cycle and that unemployment directly relates to the aggregate production evolution, we only employ real gdp. 7 the reader is referred to gilbert (1984) for a theoretical characterization and a review of empirical evidence. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 14 period 2002-2014, the calculation of the c5 index was completed using the audit reports of the spanish stock market commission. • herfindahl-hirschman (hh) index: one drawback of the c5 ratio is that it ignores the possible asymmetry in the proportion of assets controlled by the five largest firms and the asset composition of the remaining banks. therefore, the hh index is used as an alternative indicator. it is calculated as the sum of the squares of the asset share of each bank. the greater this index, the greater the degree of market concentration. this information is retrieved from the ecb statistical database. the definitions, notation and expected sign of the variables defined above are shown in table 1. table 1. variables, definitions and expected sign. 3.3. descriptive statistics summary statistics of the variables are presented in table 2. the average return on assets is 0.65%, whereas the average return on equity is notably larger (8.9%). the spanish banking system granted loans with an average value of 1.3 billion euros, whereas the mean value of the ecb assets is 1.74 trillion euros. the average 12-month euribor rate (in real terms) is 0.54% during the study period, while the 3-month rate is almost zero (-0.01%) on average. the mean of the 10-year sovereign bond year is 2.20%, being the average spread 1.66%. about 47% of total assets are under the control of the five largest entities. the high mean value of the hh index and its increase over the study period reveals there has been a rise in the market concentration of the spanish banking industry. as recently shown by cruz-garcia et al. (2020c), the consolidation of the sector has increased the dominant position of incumbent banks and led to a rise in market power. variables definition notation expected sign dependent: profitability (πt) net profits before taxes / average total assets roaa net profits before taxes / banks' average equity roae explanatory: (𝛽𝛽𝑡𝑡) performing loans performing loans stock (million €) loans + ecb assets volume of ecb assets (million €) ecb assets +/ 12-month euribor 12-month inter-bank interest rate in real terms (%) eurib_12 10-year bond yield 10-year sovereign bond yield real interest rate in real terms (%) 10y_sb + interest rate spread difference between 10-year bond yield and 12-month euribor (%) spread + real gdp index. base 2010 gdp + c5 concentration index of the five largest banks. c5 + hh herfindahl hirschman index hh + david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 15 table 2. descriptive statistics. table 3 reports the evolution of the explanatory variables by presenting their values in the last quarter of the years 1995, 2000, 2005, 2010 and 2016. the return on average assets (roaa) remained stable around 1.1% during the decade 1995-2005 but has experienced a notable drop over the last ten years. the 12-month euribor rate (eurib_12) is the variable that has suffered the largest decline, falling from 4.83% in 1995 to -1.64% (in real terms) by the end of 2016. the 10year sovereign bond rate (10y_sb) has also substantially changed over time. the stock of performing loans (loans) has continuously increased during the real estate boom but fell after 2010. on the other hand, the high number of bank mergers is the main reason for the increase in the c5 ratio and the hh index in the period 2010-2016. table 3. explanatory variables evolution. december 1995, 2000, 2005, 2010 and 2016. 3.4. time series stationarity prior to the empirical analysis, we tested the stationarity of the time series. since taking natural logs is a type of box-cox transformation that is used to achieve variance stationarity and it also facilitates the coefficient interpretation, we first log transformed the variables loans, ecb's assets, gdp, c5 and hh.8 because of the quarterly frequency of the data, we then analysed the 8 we do not take the logs of the profitability indicators or the interest rates because these variables take negative values in some periods. variable unit mean sd min max roaa % 0.659 0.998 -3.566 1.570 roae % 8.906 12.240 -41.34 23.71 loans million € 1,345,900 485,440 634,960 2,146,900 ecb assets million € 1.74e09 6.90e-08 1.0e09 3.40e09 eurib_12 % 0.54 1.90 -4.11 5.40 eurib_3 % -0.01 1.44 -4.60 2.36 10y_sb % 2.20 1.80 -1.05 7.16 spread % 1.66 1.28 -0.815 5.37 gdp index (2010=100) 89.47 12.09 65.12 104.4 c5 ratio 0.47 0.06 0.38 0.60 hh index 528.90 188.40 213.00 937.00 variable 1995 2000 2005 2010 2016 roaa 1.05 1.22 1.07 0.40 0.35 roae 13.34 15.32 14.43 5.38 3.105 loans 669,349 943,320 1,598,220 1,984,730 1,254,060 ecb assets 1.15e09 1.18e-09 1.98e-09 3.40e-09 eurib_12 4.83 0.92 -0.95 -1.46 -1.64 eurib_3 1.19 0.98 -1.26 -1.96 -1.88 10y_sb 5.64 1.24 -0.36 2.38 -0.14 spread 0.81 0.32 0.58 3.85 1.50 gdp 66.34 81.64 96.29 100.00 103.11 c5 0.38 0.56 0.44 0.45 0.60 hh 213 561 487 528 937 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 16 existence of seasonal unit roots through the hegy test (hylleberg et al., 1990). finally, the existence of at least two regular unit roots was tested by applying the dickey and fuller adf (dickey & fuller, 1979) and philips and perron (philips & perron, 1988) tests. the reader is referred to appendix i for details. the hegy tests suggest there is a seasonal unit root associated with frequency π (every two quarterly periods) for log loans. consequently, this series has been filtered by δ2 so that the transformed series is given by 𝐿𝐿𝐿𝐿𝐿𝐿 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑡𝑡– 𝐿𝐿𝐿𝐿𝐿𝐿 𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝐿𝑡𝑡−2. the adf and pp tests indicate that the variable 10y_sb is a stationary process in levels. conversely, log ecb's assets, eurib_12, eurib_3, spread, log c5 and log hh are i(1) series that need to be differentiated with respect to their first lag to become stationary. we also document there is a structural break in roaa, roae and gdp. since the adf and pp unit root tests may be biased in the presence of a structural break (perron 1989), we moved to the zívot and andrews (1992) test to determine their integration order. this test shows that roaa and roae are stationary variables, with a structural break in the last quarter of 2011. this can be a consequence of the passing of the previously commented royal decrees-laws in february and june 2012 that forced banks to raise their provisions for impairment losses in the balance sheet. to control for this in the analysis, we add a dummy variable (denoted as d2012) that takes value one from the first quarter of 2012 onwards. finally, the variable log gdp is integrated of order one, presenting a structural break in the last quarter of 2008. apart from the above-mentioned transformations, our model specification incorporates i) a time trend, ii) quarter dummies to control for potential seasonal effects, and ii) a dummy variable for the four quarters of 2010 (d2010). the latter controls for the most critical phase of the greek crisis, which threatened to destabilize the european monetary union and produced a climate of uncertainty in the european banking system.9 4. results we divide our empirical analysis into two parts. first, we consider the spread (δ spread) as a single regressor capturing the slope of the yield curve together with the other banking factors and controls. second, we replace this indicator with both the 12-month euribor rate and the 10-year sovereign bond rate. 4.1. results using the spread table 4 presents the first set of estimation results. in the first column, we regress roaa on the second-lag difference of the log of performing loans (δ2 log loans), the ecb assets growth rate (δ log ecb assets), the first difference of the spread of interest rates (δ spread), the gdp growth rate (δ log gdp), the hh index growth rate (δ log hh), a time trend (trend), the quarter dummies 9 in january 2010, a report from the european commission showed that the greek government had committed severe irregularities in sending tax data to brussels. on april 27, standard & poor’s downgraded the greek bond to the 'junk bond' category. after the bankruptcy of the greek economy, the european commission, the european central bank and the international monetary fund contributed with more than 10,000 million euros to rescue greece during 2010. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 17 (q1, q2 and q3) and the dummy variables that control for the greek crisis (d2010) and the structural break (d2012). the second column adds an interaction term between the hh index growth rate and the dummy for the structural break (δ log hh x d2012). this is because the farreaching restructuring process undergone by the spanish banking industry to correct the imbalances generated during the real estate boom led to an intense wave of corporate mergers between entities.10 as discussed in cruz-garcia et al. (2020a), this balance-sheet clean-up dictated by the two previously commented royal decree-laws affected the sector's market power. therefore, the interaction term aims to explore whether the effect of market concentration on roaa has changed in the 2012-2016 period. column 3 replaces the herfindahl-hirschman index by the c5 ratio. similar to column 2, in column 4 we include an interaction term between the concentration ratio and d2012 (δ log c5 x d2012). finally, column 5 expands the model in column 4 by including δ eurib_3 together with δ spread. all the regressions are estimated by ols. since it is possible that the error term is heteroskedastic and serially correlated, we use robust standard errors, following newey & west (1987). we find that the spread is (marginally) positively associated with roaa when we consider the hh index, but it turns to be non-significant when the market concentration is measured through c5. since the use of the latter indicator provides a better model fit according to aic and r2, from these regressions we would conclude that the spread is not significantly associated with the sector's profitability. interestingly, when we add δ eurib_3 to the specification we find that roaa decreases as the 3-month euribor rate rises (conditional on the spread). nonetheless, this latter regression needs to be interpreted with caution. the reason is that the 3-month and the 12month euribor rates are highly correlated (corr=0.89), so interpretating the δ eurib_3 coefficient, ceteris paribus, is cumbersome. concerning the rest of variables, the stock of performing loans has a (marginally) positive effect on profitability. therefore, the higher the volume of performing loans, the higher the sector's profitability. since we have taken differences with respect to the second lag, a marginal change in its inter-quarterly rate of growth increases roaa by around 7.3 percentage points (hereafter pp). by contrast, the ecb's assets growth is not significantly related with the sector's average profitability. this result is consistent with tercero-lucas (2020). we also find that roaa is positively associated with the gdp growth. a 0.1-point increase in the gdp growth rate from one quarter to the following (mean δ log gdp=0.005) translates into an increase of about 7.6 pp in roaa. this adds more evidence on the pro-cyclicality of banking results, in line with albertazzi & gambacorta (2009) and trujillo-ponce (2013). this finding could support the argument that regulation should focus on forcing banks to increase their capital buffers via reserves (or generic provisions) in expansionary periods. 10 the merger process has been quite dynamic and caused that a substantial share of the sector's assets has ended up under the control of the biggest banks. the c5 ratio (hh index) was 0.48 (596) by the end of 2011 but reached 0.58 (839) by the end of 2014 and 0.60 (937) by the end of 2016. except for banco santander that did not participated in the merging process during the study period, big banks have absorbed the business activity of small banks that were heavily exposed to the real estate boom and were in a delicate economic position. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 18 table 4. estimation results using the spread of the interest rates. (1) (2) (3) (4) (5) coefficient robust se coefficient robust se coefficient robust se coefficient robust se coefficient robust se δ2 log loans 7.353* 3.931 7.183* 3.996 6.507* 3.775 6.154* 3.408 6.328** 3.150 δ log ecb assets 0.144 0.299 0.148 0.294 0.166 0.299 0.150 0.275 0.084 0.264 δ spread 0.343* 0.197 0.357* 0.185 0.312 0.193 0.247 0.196 0.212 0.177 δ eurib_3 -0.237* 0.128 δ log gdp 74.637*** 22.301 76.241*** 22.030 76.717*** 23.297 72.009*** 22.201 69.337*** 23.540 δ log hh 1.506 1.324 1.243 1.186 δ log hh x d2012 1.508 3.945 δ log c5 -2.132 1.755 -0.349 1.023 0.537 0.958 δ log c5 x d2012 22.211*** 6.633 23.817*** 6.046 d2010 -0.505** 0.244 -0.519** 0.241 -0.518** 0.243 -0.495** 0.234 -0.610*** 0.206 d2012 -1.572*** 0.480 -1.622*** 0.463 -1.591*** 0.485 -1.319*** 0.470 -1.331*** 0.426 q1 -0.012 0.151 -0.037 0.160 0.073 0.145 0.016 0.138 0.051 0.127 q2 0.053 0.146 0.050 0.147 0.068 0.141 0.145 0.119 0.176 0.138 q3 0.139 0.170 0.142 0.167 0.145 0.170 0.161 0.170 0.210 0.162 trend 0.023** 0.009 0.023** 0.009 0.023** 0.009 0.021** 0.009 0.021** 0.008 constant -0.714 0.523 -0.728 0.509 -0.718 0.529 -0.589 0.506 -0.615 0.500 𝑅𝑅2 0.649 0.644 0.650 0.681 0.711 aic 149.45 151.29 149.27 143.48 137.25 note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. regarding the sector's competitive structure, neither the hh index nor the c5 ratio are statistically significant. although it is customary to associate a highly concentrated sector with greater profitability, there is no empirical evidence to support this positive relationship. this is consistent with athanasoglou et al. (2008), garcia-herrero et al. (2009) and dietrich & wanzenried (2011), who did not find either a significant effect of the hh index on roa. however, we document a negative and significant effect of the interaction between the c5 ratio and d2012. this implies that concentration negatively affected profitability from 2012 onwards. llorens et al. (2020) show that more productive spanish banks absorbed less productive ones for strategic reasons such as acquiring the branches where they were underrepresented. until some time elapses for the sector to generate the corresponding synergies that dimensional gains offer, the absorption of firms with serious imbalances could hamper the proper functioning of the absorbing bank. as shown by smirlock (1985), mergers do not always lead to efficiency and profitability improvements, at least in the short run. furthermore, the rise in assets' concentration and the drop in the number of banks on operation might have increased competition in the sector among leading banks through reducing information costs for consumers. overall, the intensive merging process appears to have produced a negative effect on the sector's roaa, everything else being equal. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 19 the negative and significant coefficient of the dummy variable that captures the structural break (d2012) can be interpreted as a fall in roaa from 2012 onwards because of the duty for banks to increase the share of provisions imposed by the regulatory decrees. the estimation results also show that the most critical phase of greek crisis (d2010) had a significant negative effect, possibly because of the distrust on the stability of the european monetary union if greece were rescued. the seasonal dummy variables are never significant. therefore, roaa does not significantly differ per quarter. as for the positive and significant coefficient of the time trend, this variable merely controls for stationary deviations of the regressors around its deterministic trend. in any case, it suggests roaa has slightly increased, on average, during the study period. 4.2. results using the two rates table 5 reports the second set of estimation results including both δ eurib_12 and 10y_sb as separate regressors instead of δ spread. as before, the first column considers δ log hh as the indicator of market concentration. the second column adds an interaction term between δ log hh and d2012. the third column replaces δ log hh by δ log c5. the fourth column adds the interaction between δ log c5 and d2012. finally, as a robustness check, the fifth column replaces δ eurib_12 by δ eurib_3 (see below). the estimates are very similar across the different specifications. based on the aic criterion and the r2, the regression in column 4 seems to fit the data best. the 12-month euribor rate (δ eurib_12) exerts a positive and significant effect on roaa, ceteris paribus. using the specification in (4), a marginal drop in the inter-quarterly differential of the 12-month euribor increases roaa by 0.32 pp. this negative effect might be due to two reasons. first, the euribor has fallen during the study period, resulting in banks suffering a reduction in their interest income as many of their asset products are linked to it (e.g., variable interest rate mortgages). second, the fall in euribor has also produced a decrease in financial costs due to the drop in deposit prices. since banks reprice liabilities faster than assets, this allows the difference between the lending and the deposit rate to increase when euribor falls. kumbhakar & lozano-vivas (2004) find that the deposit market is less competitive than the loan one, which therefore makes the effect of a change in interest rates on liabilities to be greater. this is due to the existence of switching costs (klemperer, 1987), by which consumers do not change bank in the case of a drop in deposit rates because they are linked to the firm via asset products, or just because the volume of their deposits does not compensate the transaction costs of opening a new account elsewhere (kim et al., 2003). trujillo-ponce (2013) also finds a negative relationship between the interest rate and the profitability of spanish banks (both roa and roe). this author argues that ""this inverse relation may be caused by a temporary lag when modifying the interest rates applied to customers in such a way that interest rate variations are more rapid for liability products than asset ones. hence, the banking interest rate margin increases with a drop in the reference interest rates and decreases with an increase in the latter"". david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 20 table 5. estimation results using both eurib_12 and 10y_sb. (1) (2) (3) (4) (5) coefficient se coefficient se coefficient se coefficient se coefficient se δ2 log loans 9.576** 3.838 9.559** 3.849 9.191** 3.745 8.333** 3.339 8.693** 3.397 δ eurib_12 -0.324** 0.151 -0.326** 0.153 -0.320** 0.152 -0.323** 0.155 δ eurib_3 -0.303** 0.150 10y_sb 0.182** 0.081 0.182** 0.081 0.186** 0.081 0.161** 0.075 0.165** 0.077 δ log gdp 66.817 23.227 67.095*** 23.629 68.513*** 23.766 65.645*** 23.629 63.424*** 22.808 δ log hh 0.447 0.660 0.400 0.627 δ log hh x d2012 0.646 3.270 δ log c5 -1.633 1.208 0.224 0.916 0.229 0.974 δ log c5 x d2012 23.179*** 5.139 -0.228*** 5.189 d2010 -0.687*** 0.225 -0.689*** 0.226 -0.696*** 0.226 -0.686*** 0.225 -0.635 0.208 d2012 -1.656*** 0.453 -1.673*** 0.455 -1.660*** 0.452 -1.384*** 0.430 -1.302*** 0.4000 q1 0.066 0.115 0.056 0.121 0.122 0.113 0.065 0.103 0.033 0.106 q2 0.110 0.131 0.111 0.131 0.115 0.129 0.179 0.126 0.155 0.119 q3 0.127 0.113 0.128 0.113 0.132 0.113 0.161 0.116 0.148 0.120 trend 0.023** 0.009 0.023** 0.009 0.023*** 0.008 0.021** 0.008 0.019** 0.007 constant -0.990 0.641 -0.992 0.640 -0.997 0.640 -0.840 0.612 -0.752 0.585 𝑅𝑅2 0.712 0.709 0.715 0.748 0.744 aic 149.39 151.34 148.68 138.77 140.07 note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. bank margins also benefit from a fall in interest rates (for new operations) due to a composition effect derived from the transfer of fixed-term deposits to sight deposits. when the remuneration of the deposits is quite small, customers prefer the flexibility of money at sight despite its minimal return, which results in the average deposit rate being lower. during the study period, interest rates fell during two or more consecutive years in 1995-1999, 2001-2005, 2008-2010 and 20122016. as shown in table 6, in all these periods, liability interest rates fell more than asset rates. table 6. drop in the synthetic asset and liability rates. asset interest rates also drop less than deposit rates due to the existence of floor clauses in a great share of mortgage loans. indeed, mortgages represent a large percentage of household periods in which a fall in the interest rates took place 1995-1999 2001-2005 2008-2010 2011-2016 drop in asset rates (%) -49.2 -66.3 -20.8 -43 drop in liability rates (%) -62 -70.9 -32.5 -71 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 21 loans, being the vast majority (around 90% during the real estate boom) at a variable interest rate. a high percentage of the variable-rate mortgages contained floor clauses, which limited the possibility of lowering the monthly payments paid by consumers when the reference rate fell below that fixed by the clause. it is important to highlight that euribor does not only affect pre-tax profits, and hence roaa, in the upper part of the income statement (that is, the net interest income), but it also influences the gross margin via capital gains (or losses) in the trading of financial assets, especially fixedincome portfolios. since the price of some financial assets is inversely related to the interest rate and given that from 2009 onwards, euribor has continuously decreased, the incentive to gain profits through the sale of the large amounts of fixed income assets in the banks' balance sheets experienced a great increase, with the corresponding positive effect on earnings. this further reinforces the negative sign obtained. the sign of the parameter associated with the 10-year sovereign bond yield is positive whilst that of the euribor is negative. both signs make sense. since the price of liabilities is usually linked to short-term interest rates (most of bank funding is short-term), an increase in short-term interest rates (ceteris paribus the remaining rates on the curve) increases the funding costs of banks and damages profitability. moreover, asset prices are usually linked to longer term interest rates because banks invest in long-term assets. therefore, an increase in long-term interest rates encourages banks to formalize operations at higher interest rates, thereby improving their profitability. the sign and significance of the rest of the variables are similar to table 4, so we abstract from commenting on them again. we performed some robustness checks and model extensions. first, some authors like albertazzi & gambacorta (2009) and borio et al. (2017) use the 3-month interbank rate instead of the 12-month one. accordingly, column 5 replaces δ eurib_12 by δ eurib_3. as shown, both the signs and the magnitude of the estimated parameters remain unchanged compared to column 4. second, we repeated the regressions presented in tables 4 and 5 using roae as the dependent variable. the results are consistent with the ones reported and are available from the authors upon request. third and more importantly, during the study period the deflated 12-month euribor rate has been positive and negative, and the inter-month variability has also been positive and negative. studies concerned about the linkages between euribor and banks' profitability have shown that the relationship might be non-linear, so that the effect is greater when the interests are low (borio et al., 2017; cruz-garcia et al., 2018; pérezmontes and ferrer-pérez, 2018). to inspect in more detail the role of euribor on roaa, we take the model specification in table 5 column 4 and add i) an interaction term between δ eurib_12 and a dummy for whether eurib_12>0 (euribor. positive), and ii) an interaction term between δ eurib_12 and a dummy for whether eurib_12 is above the study period mean, which equals 0.549 (euribor. abovemean). the coefficient estimates of these additional regressions are presented in appendix ii table a5. interestingly, we find that the negative effect of the 12-month euribor rate on roaa, conditional on 10y_sb, becomes lower in magnitude i) when the euribor is positive, and ii) when david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 22 the euribor is above the study period mean. indeed, in the latter case the overall effect is even positive. this tentatively suggests that the detected negative effect of euribor on roaa mainly holds at low levels. put another way, drops in the 12-month euribor rate conditional on 10y_sb (i.e. rises in the slope of the yield curve) positively affect the sector's profitability at low levels but are negligible at high levels. this is consistent with evidence presented in borio et al. (2017), cruz-garcia et al. (2018) and pérez-montes and ferrer-pérez (2018). 5. conclusions this study contributes to the literature by uncovering the relationship between euribor interest rate and the banking sector's average returns. a time series regression model with quarterly data is specified where the return on average assets (roaa) is explained by banking factors, gdp, indicators of market concentration and time effects. the main purpose is to study how roaa varies with changes in the slope of the yield curve, which is given by the difference between the 10-year sovereign bond yield and the 12-month euribor rate. focusing on the spanish case and considering a longer time span than related studies (22 years), a novel aspect of this research is that we distinguish the effect of variations in the slope of the yield curve depending on whether they stem from the long or the short interest rate. when we specify the spread as a single regressor, we do not detect a significant relationship with roaa. however, when considering both rates in the model, we find that the sector's profitability increases with the long-term interest rate and decreases with the 12-month euribor rate. because euribor is a reference for fixing asset and liability interest rates, a fall (increase) in this rate increases (reduces) the lending minus deposit spread, ceteris paribus. this negative relationship appears to indicate that banks are able to take advantage of the possibility of lowering the rates on deposits faster than those on loans. interestingly, we also find that the negative effect of the euribor is moderated when this rate is positive and above its mean. this suggests that the negative relationship encountered mainly holds at low levels of euribor. the results also indicate that roaa is positively related to the stock of performing loans and gdp, as expected. we do not find evidence that market concentration exerts a significant effect on profitability, on average, but we document that the intense merging process that took place after 2012 is associated with a lower sector's profitability (at least in the short run). the greater market concentration has potentially increased competition due to the drop in customer search costs. aknowledgements the authors wish to thank helpful comments and suggestions received from alejandro fernández-cerezo, enrique moral-benito and two anonymous referees. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 23 references albertazzi, u., gambacorta, l. (2009). bank profitability and the business cycle. journal of financial stability 5: 393-409. https://doi.org/10.1016/j.jfs.2008.10.002 alessandri, p., nelson, b.d. (2015). simple banking, profitability and the yield curve. journal of money, credit and banking 471: 143-175. https://doi.org/10.1111/jmcb.12172 almeida, f.d., divino, j.a. (2015). determinants of the banking spread in the brazilian economy, the role of micro and macroeconomic factors. international review of economics and finance 40: 29-39. https://doi.org/10.1016/j.iref.2015.02.003 altavilla, c., boucinha, m., peydró, j.l. (2018). monetary policy and bank profitability in a low interest rate environment. economic policy 33(96): 531-586. https://doi.org/10.1093/epolic/eiy013 alvarez, a., fernández, a., garcia-cabo, j., posada, d. (2019). liquidity funding shocks: the role of banks' funding mix. journal of financial services research 55: 167-190. https://doi.org/10.1007/s10693-019-00314-8 athanasoglou, p.p., brissimis, s.n., delis, m.d. (2008). bank-specific, industry-specific and macroeconomic determinants of bank profitability. journal of international financial markets institutions and money 18: 121-136. https://doi.org/10.1016/j.intfin.2006.07.001 bikker, j.a., vervliet, t.m. (2018). bank profitability and risk-taking under low interest rates. international journal of finance & economics 23: 3-18. https://doi.org/10.1002/ijfe.1595 bolt, w., de haan, l., hoeberichts, m., van oordt, m.r.c., swank, j. (2012). bank profitability during recessions. journal of banking and finance 36: 2552-2564. https://doi.org/10.1016/j.jbankfin.2012.05.011 borio, c. & gambacorta, l. (2017). monetary policy and bank lending in a low interest rate environment: diminishing effectiveness? journal of macroeconomics 54(b): 217-231. https://doi.org/10.1016/j.jmacro.2017.02.005 borio, c., gambacorta, l., hofman, b. (2017). the influence of monetary policy on bank profitability. international finance: 201, 48-63. https://doi.org/10.1111/infi.12104 brei, m., borio, c. & gambacorta, l. (2020). bank intermediation activity in a low-interest-rate environment. economic notes, forthcoming. https://doi.org/10.1111/ecno.12164 busch, r., memmel, c. (2017). banks' net interest margin and the level of interest rates. credit and capital markets 50(3): 363-392. https://doi.org/10.3790/ccm.50.3.363 carbó, s., rodríguez, f. (2007). the determinants of bank margins in european banking. journal of banking and finance 31: 2043-2063. https://doi.org/10.1016/j.jbankfin.2006.06.017 claessens, s., coleman, n., donnelly, m. (2018). "low-for-long" interest rates and banks' interest margins and profitability: cross-country evidence. journal of financial intermediation 35: 1-16. https://doi.org/10.1016/j.jfi.2017.05.004 claessens, s., laeven, l. (2004). what drives banking competition? some international evidence. journal of money, credit and banking 363: 563-583. https://doi.org/10.1353/mcb.2004.0044 climent, s., pavía, j.m. (2015). determinants of profitability in spanish financial institutions. comparing aided and non-aided entities. journal of business economics and management 166: 1170-1184. https://doi.org/10.3846/16111699.2013.801881 cruz-garcia, p., fernández de guevara, j., maudos, j. (2018). bajos tipos de interés en la eurozona: impacto sobre los márgenes y la rentabilidad de los bancos. papeles de economía española 132-146. cruz-garcía, p., fernández de guevara, j., maudos, j. (2019). determinants of bank's interest margin in the aftermath of the crisis: the effect of interest rates and the yield curve slope. empirical economics 56: 341-365. https://doi.org/10.1007/s00181-017-1360-0 cruz-garcia, p., fernández de guevara, j., maudos, j. (2020a). banking concentration and competition in spain: the impact of the crisis and restructuring. financial stability review 34: 59-76. cruz-garcía, p., forte, a., peiró-palomino, j. (2020b). on the drivers of profitability in the banking industry in restructuring times: a bayesian perspective. applied economic analysis, forthcoming. https://doi.org/10.1108/aea-01-2020-0003 https://doi.org/10.1016/j.jfs.2008.10.002 https://doi.org/10.1111/jmcb.12172 https://doi.org/10.1016/j.iref.2015.02.003 https://doi.org/10.1093/epolic/eiy013 https://doi.org/10.1007/s10693-019-00314-8 https://doi.org/10.1016/j.intfin.2006.07.001 https://doi.org/10.1002/ijfe.1595 https://doi.org/10.1016/j.jbankfin.2012.05.011 https://doi.org/10.1016/j.jmacro.2017.02.005 https://doi.org/10.1111/infi.12104 https://doi.org/10.1111/ecno.12164 https://doi.org/10.3790/ccm.50.3.363 https://doi.org/10.1016/j.jbankfin.2006.06.017 https://doi.org/10.1016/j.jfi.2017.05.004 https://doi.org/10.1353/mcb.2004.0044 https://doi.org/10.3846/16111699.2013.801881 https://doi.org/10.1007/s00181-017-1360-0 https://doi.org/10.1108/aea-01-2020-0003 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 24 cruz-garcia, p., fernández de guevara, j., maudos, j. (2020c). bank competition and multimarket contact intensity. journal of international money and finance, forthcoming. https://doi.org/10.1016/j.jimonfin.2020.102338 dell’ariccia, g., rabanal, p., sandri, d. (2018). unconventional monetary policies in the euro area, japan, and the united kingdom. journal of economic perspectives 32(4): 147–172. https://doi.org/10.1257/jep.32.4.147 demirguc-kunt, a., huizinga, h. (1999). determinants of commercial bank interest margins and profitability, some international evidence. world bank economic review 13: 379 – 408. https://doi.org/10.1596/1813-9450-1900 dickey, d.a., fuller, w.a. (1979). distribution of the estimators for autoregressive time series with a unit root. journal of the american statistical association 74: 427-431. https://doi.org/10.1080/01621459.1979.10482531 dickey, d.a., pantula, s.g. (1987). determining the order of differencing in autoregressive processes. journal of business and economic statistics 5: 455-461. https://doi.org/10.1080/07350015.1987.10509614 dietrich, a., wanzenried, g. (2011). determinants of bank profitability before and during the crisis, evidence from switzerland. journal of international financial markets, institutions and money 21: 307-327. https://doi.org/10.1016/j.intfin.2010.11.002 dietrich, a., wanzenried, g. (2014). the determinants of commercial banking profitability in low, middle-, and high-income countries. the quarterly review of economics and finance 54: 337354. https://doi.org/10.1016/j.qref.2014.03.001 egly, p.v., johnk, d.w. mollick, a.v. (2018). bank net interest margins, the yield curve, and the 2007-2009 financial crisis. review of financial economics 36: 12-32. https://doi.org/10.1002/rfe.1016 english, w.b. (2002). interest rate risk and bank net interest margin. bank for international settlement (bis) quarterly review: 67-92. fernández-villaverde, j., garicano, l., santos, t. (2013). political credit cycles: the case of the eurozone. journal of economic perspectives 27(3): 145-166. https://doi.org/10.3386/w18899 flannery, m.j. (1981). market interest rate and commercial bank profitability: an empirical investigation. journal of finance 36: 1085-1101. https://doi.org/10.1111/j.15406261.1981.tb01078.x garcia-herrero, a., gavila, s., santabarbara, d. (2009). what explains the low profitability of chinese banks? journal of banking and finance 3311: 2080–2092. https://doi.org/10.1016/j.jbankfin.2009.05.005 gilbert, a. (1984). studies of bank market structure and competition. a review and evaluation. journal of money, credit and banking 164: 617–644. https://doi.org/10.2307/1992096 granger, c.w.j., newbold, p. (1974). spurious regression in econometrics. journal of econometrics 2: 111-120. https://doi.org/10.1016/0304-4076(74)90034-7 heider, f., saidi, f., schepens, g. (2019). life below zero: bank lending under negative policy rates. the review of financial studies 32(10): 3728-3761. https://doi.org/10.1093/rfs/hhz016 ho, t.s.y., saunders, a. (1981). the determinants of bank interest margins: theory and empirical evidence. journal of financial and quantitative analysis 16: 581-600. https://doi.org/10.2307/2330377 hoffman, m., sorensen, b.e. (2015). small firms and domestic bank dependence in europe's great recession. european comission, european economy discussion paper 12: 5 – 35. hylleberg, s., engle, r.f., granger, c.w.j., yoo, b.s. (1990). seasonal integration and cointegration. journal of econometrics 44: 215-238. https://doi.org/10.1016/03044076(90)90080-d jimeno, j.f., santos, t. (2014). the crisis of the spanish economy. series 5: 125-141. https://doi.org/10.1007/s13209-014-0116-8 klemperer, p. (1987). markets with consumer switching costs. the quarterly journal of economics 1022: 375-394. https://doi.org/10.2307/1885068 kumbhakar, s., lozano-vivas, a. (2004). does deregulation make markets more competitive? evidence of mark-ups in spanish savings banks. applied financial economics 14: 507-515. https://doi.org/10.1080/0960310042000216060 https://doi.org/10.1016/j.jimonfin.2020.102338 https://doi.org/10.1257/jep.32.4.147 https://doi.org/10.1596/1813-9450-1900 https://doi.org/10.1080/01621459.1979.10482531 https://doi.org/10.1080/07350015.1987.10509614 https://doi.org/10.1016/j.intfin.2010.11.002 https://doi.org/10.1016/j.qref.2014.03.001 https://doi.org/10.1002/rfe.1016 https://doi.org/10.3386/w18899 https://doi.org/10.1111/j.1540-6261.1981.tb01078.x https://doi.org/10.1111/j.1540-6261.1981.tb01078.x https://doi.org/10.1016/j.jbankfin.2009.05.005 https://doi.org/10.2307/1992096 https://doi.org/10.1016/0304-4076(74)90034-7 https://doi.org/10.1093/rfs/hhz016 https://doi.org/10.2307/2330377 https://doi.org/10.1016/0304-4076(90)90080-d https://doi.org/10.1016/0304-4076(90)90080-d https://doi.org/10.1007/s13209-014-0116-8 https://doi.org/10.2307/1885068 https://doi.org/10.1080/0960310042000216060 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 25 kumbhakar, s., lozano-vivas, a., lovell, c.a.k., & hasan, i. (2001). the effects of deregulation on the performance of financial institutions: the case of spanish savings banks. journal of money, credit, and banking 331: 101-120. https://doi.org/10.2307/2673874 lambert, f., ueda, k. (2014). the effects of unconventional monetary policies on bank soundness. imf working paper. https://doi.org/10.5089/9781498363563.001 lópez, j.a., rose, a.k., spiegel, m.m. (2020). why have negative nominal interest rates had such a small effect on bank performance? cross country evidence. european economic review forthcoming. https://doi.org/10.1016/j.euroecorev.2020.103402 llorens, v., martín-oliver, a., salas-fumas, v. (2020). productivity, competition and bank restructuring process. series 11: 313-340. https://doi.org/10.1007/s13209-020-00214-4 mamatzakis, e., bermpei, t. (2016). what is the effect of unconventional monetary policy on bank performance? journal of international money and finance 67: 239-263. https://doi.org/10.1016/j.jimonfin.2016.05.005 martin, a., moral-benito, e., schmitz, t. (2020). the financial transmission of housing booms: evidence from spain. american economic review, forthcoming. https://doi.org/ 10.1257/aer.20191410 molyneux, p., thornton, j. (1992). determinants of european bank profitability, a note. journal of banking and finance 16: 1173-1178. https://doi.org/10.1016/0378-4266(92)90065-8 molyneux, p., reghezza, a., thornton, j., xie, r. (2020a). did negative interest rates improve bank lending? journal of financial services research 57: 51-68. https://doi.org/ 10.1007/s10693-019-00322-8 molyneux, p., reghezza, a., torriero, c., williams, j. (2020b). banks' noninterest income and securities holding in a low interest rate environment: the case of italy. european financial management, forthcoming. https://doi.org/10.1111/eufm.12268 molyneux, p., reghezza, a., xie, r. (2019). bank margins and profits in a world of negative rates. journal of banking and finance, forthcoming. https://doi.org/10.1016/j.jbankfin .2019.105613 newey, w.k., west, k. (1987). a simple positive semi-definite heteroscedasticity and autocorrelation consistent covariance matrix. econometrica 55: 703-708. https://doi.org/10.2307/1913610 pérez-montes, c. ferrer-pérez (2018). the impact of the interest rate level on bank profitability and balance sheet structure. financial stability review, bank of spain 35: 119-148. perron, p. (1989). the great crash, the oil price shock and the unit root hypothesis. econometrica 57: 1361-1401. https://doi.org/10.2307/1913712 phillips, p., perron, p. (1988). testing for a unit root in time series regression. biometrica 75(2): 335-346. https://doi.org/10.1093/biomet/75.2.335 repullo, r., saurina, j., trucharte, c. (2010). mitigating the pro-cyclicality of basel ii. economic policy 25: 659-702. https://doi.org/10.1111/j.1468-0327.2010.00252.x salas, v., saurina, j. (2003). deregulation, market power and risk behaviour in spanish banks. european economic review 47: 1061-1075. https://doi.org/10.1016/s0014-2921(02)00230-1 saona, p. (2016). intraand extra-bank determinants of latin american banks' profitability. international review of economics and finance 45: 197-214. https://doi.org/10.1016/j.iref .2016.06.004 saunders, a., schumacher, l. (2000). the determinants of bank interest rate margins, an international study. journal of international money and finance, 19: 813-832. https://doi.org/10.1016/s0261-5606(00)00033-4 smirlock, m. (1985). evidence on the (non) relationship between concentration and profitability in banking. journal of money, credit and banking 17: 69-83. https://doi.org/10.2307/1992507 sufian, f., habibullah, m.s. (2009). determinants of bank profitability in a developing economy, empirical evidence from bangladesh. journal of business economics and management 10(3): 207-217. https://doi.org/10.3846/1611-1699.2009.10.207-217 tercero-lucas, d. (2020). non-standard monetary policies and bank profitability: the case of spain. working paper. https://doi.org/10.1002/ijfe.2535 trujillo-ponce, a. (2013). what determines the profitability of banks? evidence from spain. accounting and finance 53: 561-586. https://doi.org/10.1111/j.1467-629x.2011.00466.x https://doi.org/10.2307/2673874 https://doi.org/10.5089/9781498363563.001 https://doi.org/10.1016/j.euroecorev.2020.103402 https://doi.org/10.1007/s13209-020-00214-4 https://doi.org/10.1016/j.jimonfin.2016.05.005 https://doi.org/10.1257/aer.20191410 https://doi.org/10.1257/aer.20191410 https://doi.org/10.1016/0378-4266(92)90065-8 https://doi.org/10.1007/s10693-019-00322-8 https://doi.org/10.1007/s10693-019-00322-8 https://doi.org/10.1111/eufm.12268 https://doi.org/10.1016/j.jbankfin.2019.105613 https://doi.org/10.1016/j.jbankfin.2019.105613 https://doi.org/10.2307/1913610 https://doi.org/10.2307/1913712 https://doi.org/10.1093/biomet/75.2.335 https://doi.org/10.1111/j.1468-0327.2010.00252.x https://doi.org/10.1016/s0014-2921(02)00230-1 https://doi.org/10.1016/j.iref.2016.06.004 https://doi.org/10.1016/j.iref.2016.06.004 https://doi.org/10.1016/s0261-5606(00)00033-4 https://doi.org/10.2307/1992507 https://doi.org/10.3846/1611-1699.2009.10.207-217 https://doi.org/10.1002/ijfe.2535 https://doi.org/10.1111/j.1467-629x.2011.00466.x david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 26 zívot, e., andrews, w.k. (1992). further evidence on the great crash, the oil price shock and the unit root hypothesis. journal of business and economic statistics 10(10): 251-270. https://doi.org/10.1080/07350015.1992.10509904 https://doi.org/10.1080/07350015.1992.10509904 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 27 appendix i. time-series stationarity. to avoid spurious regression, time series must be stationary (granger & newbold 1974). nonstationarity in the mean is caused by the existence of a time trend. a deterministic trend is not problematic since it can be included in the model as a regressor. however, stochastic trends (unit roots) are the main cause of the spurious regression problem. because our time series have quarterly periodicity, as a first step we test the existence of unit roots in the seasonal frequencies. we first conduct the hegy test developed by hylleberg et al. (1990).11 table a1 presents the results. the auxiliary regressions incorporate a constant, trigonometric terms and a time trend. from the results in this table we conclude: i) log loans is i(1) with frequency each two periods (quarters), ii) euribor_3 and euribor_12 could be i(0) or i(1), and iii) the rest of variables seem to be i(1). accordingly, the transformation δ2 is applied on log loans. once having explored seasonal unit roots, unit roots in the zero frequency are analysed using the augmented dickey fuller test (dickey & fuller, 1979) (hereafter adf). this test assumes that the series follows an autoregressive process of order k and hypothesizes that there is a unit root, namely, the time series is ds (difference stationary).12 there are three possible specifications for the test: i) with a constant term, ii) without a constant term and iii) with a constant and a time trend. under the alternative hypothesis, in the two first specifications the series would be stationary around a constant mean. in the third case, it would be stationary around a deterministic trend (linear or polynomial) or process ts (trend stationary). we run the adf test on each time series based on an auxiliary ols regression with a constant term, a time trend and k lags.13 alternatively, the phillips & perron test (pp test) (phillips & perron, 1988) is used.14 table a2 show the results of the adf and pp tests for the variables, both in levels and first differences (to detect potential double unit roots). from this output we conclude: i) 10y_sb is stationary in levels; ii) log ecb assets, eurib_12, eurib_3, spread, log c5, log hh are i(1), so we take their first differences; iii) log gdp could be i(1) or i(2); and iv) roaa and roae could be i(0) or i(1). to properly address whether log gdp is i(1) or i(2), in table a3 we run dickey & pantula test (dickey & pantula, 1987). the double unit root hypothesis is rejected, so log gdp is assumed to be stationary in first differences. as for whether roaa and roae are i(0) or i(1), zívot & andrews (1992) propose a test that analyses the stationarity by endogenously determining a possible point of structural break in the series. since roaa and roae appear to have a structural break in its time evolution (see figure 1), in table a4 we present the zívot & andrews (1992) tests for these variables. the results indicate that roaa and roae can be considered as stationary processes 11 these authors derive the auxiliary regressions and the corresponding critical values for each statistic. a t-test statistic is used for unit roots in the zero and π frequencies. the null hypothesis of a unit root in the complex-conjugate frequencies employs a f-type statistic. 12 non-stationary process with a stochastic trend which requires to be d times differentiated as to support a stationary and invertible arma representation. 13 the election of the number of lags is conducted by the aic criterion so that the optimal k satisfies that aick<aick-1 (dickey and pantula 1987). 14 the main difference with adf is the assumption regarding the distribution of the error term in the auxiliary regression. pp assumes the same basic specification of the adf test but considers that the error term is not a random noise – it can be serially correlated –but a stationary arma process. david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 28 if the structural break from the first quarter 2012 onwards is controlled for in the regression.table a1. hegy tests for seasonal integrability. note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. table a2. adf and phillips-perron tests for unit roots. adf test pp test decision variable levels firstdifferences levels firstdifferences roaa -3.40** -7,14*** -2.67 -7.34*** i(0) or i(1) roae -3.26* -7.00*** -2.59 -7.19*** i(0) or i(1) log ecb assets -0.72 -13.96*** -1.14 -14.45*** i(1) eurib_12 -2.57 -8.47*** -2.83 -8.71*** i(1) eurib_3 -2.44 -8.86*** -3.09 -9.05*** i(1) 10y_sb -2.97** -9.74*** -2.99** -9.78*** i(0) spread -1.99 -7.80*** -2.19 -8.12*** i(1) log gdp -2,10 -1.51 -1.08 -2.52 i(1) or i(2) log c5 -0.74 -8.73*** -1.39 -9.00*** i(1) log hh -1.38 -3.49** -1.76 -10.17*** i(1) note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. table a3. dickey and pantula (1987) test for the existence of a double unit root. variable h0: t-statistic conclusion log gdp i(2) -1.97** i(1) note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. table a4. zívot & andrews' test for unit root with structural break. note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. h0: si (1,1) variable π1=0 (zero frequency) π2=0 (π frequency) π3=π4=0 (π/2 frequency) integrability conclusion roaa -3.23 -4.59*** 60.27*** possibly i(1) roae -2.97 -4.50*** 57.85*** possibly i(1) log loans -0.46 -1.09 12.39*** i(1) with frequency each two quarters log ecb assets -0.37 -3.97*** 20.16*** possibly i(1) eurib_12 -2.92** -5.26*** 39.29*** i(0) or i(1) eurib_3 -4.06*** -2.92*** 18.09*** i(0) or i(1) 10y_sb -2.45 -3.53*** 35.56*** possibly i(1) spread -2.15 -5.08** 43.75*** possibly i(1) log gdp -2.03 -4.64*** 28.89*** possibly i(1) log c5 -1.35 -5.33** 36.50*** possibly i(1) log hh -1.65 -5.00*** 37.22*** possibly i(1) variable t-statistic structural break conclusion roaa -5.84*** 2011:4 i(0) with structural break from 2011:4 roae -6.35*** 2011:4 i(0) with structural break from 2011:4 log gdp -5.21** 2008:2 i(1) with structural break from 2008:2 david boto-garcía et al. / european journal of government and economics 10(1), june 2021, 5-29 29 appendix ii. model extensions. table a5. coefficient estimates for model extensions. (1) (2) coefficient se coefficient se δ2 log loans 7.812*** 2.931 7.264** 2.880 δ eurib_12 -0.507** 0.195 -0.437** 0.172 δ eurib_12 x euribor. positive 0.375** 0.186 δ eurib_12 x euribor. abovemean 0.447** 0.195 δ 10y_sb 0.182** 0.072 0.160** 0.072 δ log gdp 70.624*** 22.676 67.294*** 22.346 δ log c5 0.119 1.023 -0.030 1.053 δ log c5 x d2012 -22.881 4.823 -22.558*** 5.248 d2010 -0.809*** 0.217 -0.694*** 0.209 d2012 -1.501*** 0.399 -1.381*** 0.407 q1 0.075 0.098 0.103 0.099 q2 0.198 0.122 0.211* 0.123 q3 0.097 0.098 0.158 0.106 trend 0.021** 0.007 0.018** 0.007 constant -0.909 0.568 -0.725 0.557 𝑅𝑅2 0.763 0.709 aic 134.20 151.34 note: (*), (**) and (***) indicate significance at 10%, 5% and 1% level respectively. european journal of government and economics 12(1), june 2023, 79-101 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 panel data analysis of internal conflict and income inequality brandon parsons a*, shahdad naghshpour b a pepperdine university, united states b university of southern mississippi, united states * corresponding author at: brandon.parsons@pepperdine.edu abstract. the study determines how worsening internal and external conflict affects income inequality. the paper accounts for contributing variables and analyzes panel data in an unbalanced panel of 106 countries from 1988 to 2018— the panel data model groups by development status. the econometric model uses driscoll and kraay standard errors to account for heteroscedasticity, cross-sectional dependence, and autocorrelation. worsening internal conflict increases income inequality in developing countries but not in developed countries. worsening of internal conflict by one standard deviation increases income inequality by 0.068 in developing counties. external conflict does not noticeably affect income inequality in developed or developing panels. keywords. income inequality, gini coefficient, internal conflict, external conflict doi. https://doi.org/10.17979/ejge.2023.12.1.9283 1. introduction1 for centuries, income inequality has been a topic of interest (sahota, 1978). plato discusses income inequality and its effects on the polis in the republic. income inequality has increased in many countries over the last forty years (united nations, 2021). its growth and persistent is an important subject because of its adverse effects economically, socially, and politically (posner, 1997; barry, 2005; karklins, 2005; dorling, 2011; dabla-norris et al., 2015; ariely and uslaner, 2017). additionally, studying income inequality is vital because current income inequality can lead to future income inequality. specifically, intergenerational transmission of income inequality can lead to future undesirable economic, social, and political effects (ermisch and smeeding, 2012). many studies on the conflict-inequality nexus focus on income inequality as a driver of conflict (bircan, brück, and vothknecht, 2017). some researchers claim income inequality can cause conflict (feirarbend and feierabend, 1966; gurr, 1968; huntington, 1968; dobson and romlogan, 2009; haggard and kaufman, 2012). de tocqueville (1835) claims that most uprisings and 1 all data used in the study are publicly available and accessible. we want to acknowledge funding for the icrg dataset from pepperdine university and helpful suggestions and comments from david smith, jared ashworth, blake withall, and the anonymous referee of the european journal of government and economics. any remaining shortcomings are those of the authors. https://creativecommons.org/licenses/by-nc/4.0/ mailto:brandon.parsons@pepperdine.edu https://doi.org/10.17979/ejge.2023.12.1.9283 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 80 revolutions are instigated because of high income inequality. in addition, theoretically, the greater the income inequality in a country, lower income individuals are worse off and have a greater incentive to rebel (haggard and kaufman, 2012). there is less research on the effects of conflict on income inequality, and findings are mixed (alesina and perotti, 1996; collier, 1999; bircan, brück, and vothknecht, 2017). one claim is civil wars affect short-run economic performance but not fundamental drivers of longer-term economic performance (blattman and miguel, 2010). thus, blattman and miguel (2012) claim income distribution remains persistent regardless of conflict, including civil war. on the other hand, other researchers claim conflict increases income inequality through its effects on economic growth and investment (alesina and perotti, 1996; bircan, brück, and vothknecht, 2017). this study examines the effects of conflict on income inequality measured by the net gini coefficient (income distribution after tax and transfer). the gini coefficient is an index between (0) and (100), with the former representing perfect income equality while the latter shows the utmost inequality. the study proposes that an increase in internal conflict causes higher income inequality. the study's premise is that conflict affects different segments of a society differently and causes changes in income distribution among these segments. firstly, conflict is less likely to disrupt upper income individuals' incomes than middleand lower-income individuals. middleclass and lower income groups are more likely to feel the adverse effects of economic disruptions, e.g., they are more likely to lose their jobs (alesina and perotti, 1996). affluent production owners can safeguard their wealth and income by changing production to military goods or by moving abroad and continuing to earn income (pugh, 2003). secondly, since conflict disrupts economic activities, it reduces tax revenue (personal income, business, and tariffs on imports), which reduces the amount that can be redistributed to lower income groups, thus increasing income inequality (pugh, 2003). thirdly, not only does government revenue decline when there is conflict, but the government also allocates the reduced receipts to the military and police instead of redistributing it, especially away from education (milanovic, 2016). subsequently, low income individuals that have the potential to increase their incomes through education may not have the opportunity. the effects of conflict on income inequality can happen quickly or over time. low income individuals may lose their job as soon as conflict disrupts the economy, impacting income inequality immediately. at the same time, a generation of lower income individuals may have less access to subsidized education if there is either a reduction in gross tax revenue or the allocation of tax revenue is used on the military instead of education. the study explores the persistence of income inequality by lagging the gini coefficient for up to five years. the research extends the literature on income inequality by analyzing more comprehensive conflict measures, extending the analysis length, and using data as recent as 2018. this study also improves the econometric model, controls for important covariates, and analyzes differences between developed and developing countries. finally, the study finds worsening internal conflict causes an increase in income inequality in developing countries but not developed ones. the finding is unique to this study and supports kuznets's theory of the impact of economic parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 81 development, economic growth, and income inequality. the literature review focuses on the relationship between income inequality and conflict, while the data section discusses the sources of variables in the study. the design of the panel data analysis is in the methods section. the findings, discussion, and conclusion sections offer insight into the results and consider the study's contributions and limitations. 2. literature review – conflict as a determinant of income inequality internal and external conflict can lead to instability through coups, constitutional crises, terrorism, war, and others (posner, 1997). research finds economic effects of conflict and instability may lead to changes in income inequality (nagel, 1974; hardy, 1979; weede, 1981; alesina and perotti, 1996; cederman, weidmann, and gleditsch, 2011). for example, alesina and perotti (1996) claim that conflict increases instability, which affects income inequality by hindering economic growth and discouraging foreign investment. alesina and perotti use a panel of 70 countries from 1960-85 and measure political instability through an index based on the number of politically motivated assassinations, coup d’états, and regime type. this study extends the literature by analyzing more comprehensive measures of conflict, including "perceived" risk. the international country risk guide (icrg) measure of internal conflict includes the risk of civil war/ coup d'état threat, terrorism/political violence, and civil disorder. although conflict can increase economic uncertainty and disrupt economic activity, its effect on income inequality is not certain. economic growth's impact on income inequality depends on the level of economic development. kuznets (1955) and kaldor (1957) find that as a country’s income grows, inequality initially worsens and then improves later. they posit that when a country is underdeveloped, the per capita income is low for most people. consequently, measures such as the gini coefficient indicate low income inequality. as the country begins to develop, economic growth begins, which makes the associated economic sector and the region experience higher per capita income and income inequality. it will result in an increase in the gini coefficient for the country. as the country continues to develop and more sectors and regions receive a higher income, it reduces income inequality, hence the gini coefficient. thus, the effects of conflict on income inequality through economic growth may partially depend on a country's place on the kuznets curve. in addition, foreign direct investment and free trade may lead to growth and expansion in industries that lead to higher income inequality. on the other hand, what if conflict lead to economic turmoil, recession, and broad economic despair regardless of whether a country is in the highor low-income category? in this scenario, conflict could lead to greater income equality. war is a subcomponent of both the internal and external icrg conflict measures. evidence suggests wars, especially civil wars, adversely affect gross national income (gni) and income distribution (collier and hoeffler, 2007). also, studies find the emergence of a minority class of war profiteers can increase the gini coefficient (pugh, 2003; bircan, brück, and vothknecht, 2017). if the upper income group is immune to conflict, while the middle and lower classes lose parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 82 income due to the destruction of the economy and job opportunities, then conflicts cause income inequality. collier (1999) researches the consequences of civil wars on gdp and the economy's composition and finds that gross domestic product (gdp) per capita declines by 2.2%, primarily through a reduction in production and loss of capital stock. collier's (1999) research is limited by its small sample size and focus on the extreme cases of full-blown civil wars. this research analyzes more comprehensive measures of internal and external conflict, which allows a larger panel of countries. the long run implications of conflict are there may be fewer public funds for education and social welfare programs because of less gni and subsequent tax revenue (deininger and olinto, 2000). additionally, the need for military funds reduces resources available for education and social wellness programs which worsens income inequality (collier and hoeffler, 2002). bircan, brück, and vothknecht (2017) use panel data from 161 countries from 1960-2014 to analyze the effects of civil wars on income inequality and find an increase in income inequality ranging from 1.9 to 2.5 gini points during a civil war and it continues up to ten years after a civil war. bircan, brück, and vothknecht (2017) fail to control for important covariates, such as the dependency ratio (ratio of elderly and children to labor force), and their model does not control for the presence of cross-sectional dependence or autocorrelation, which impact large scale cross-sectional panels (driscoll and kraay, 1998). this study adds important covariates, improves the model to account for cross-sectional dependence and autocorrelation, and explores more comprehensive measures of conflict. an additional consideration is some conflicts have increased gni. for example, the united states entering world war ii (wwii) spurred the united states economy and is broadly viewed as the catalyst that ended the great depression. the outcome of the engagement of the united states in world war ii seems to be an exception to the norm. the income inequality in the united states declined immediately after wwii until around 1967 (duquette, 2018). another contributing factor is that the united states did not have any damages to its production facilities or infrastructure, apart from the shipyard facilities in hawaii. instead, the united states increased its output in almost every area, partly to provide goods, services, and weapons to europe, which had lost substantial production capacity due to damages ((duquette, 2018). the evidence suggests that when a country is not invaded, it can continue to increase its production capacity to get ahead of war-torn countries. one reason for england's ability to rise to global supremacy, despite its size and lack of substantial resources, was that it was rarely invaded, unlike other european countries (von schulze-gaevernitz, 1915). additionally, although there may be some economic growth cases from war, they come with high costs. war casualties and the need to retool factories for consumer goods have adverse shortand long-term effects. there may be few transferable skills employees can use once the economy shifts away from war-based manufacturing. war-driven economies can persist postconflict and lead to high levels of military spending that continue to reallocate funds from education and social programs to the military (pugh, 2003). parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 83 3. methods and data 3.1 data description and variable selection appendix a provides the name, source, and summary statistics of variables. the dependent variable is the net gini coefficient as measured by the standardized world income inequality database (swiid) (solt, 2015). the net gini measures income inequality after tax and transfer and is on a scale between (0) equal distribution to (100) most unequal distribution. the gini coefficient is derived from the lorenz curve that measures income distribution across households (lorenz, 1905). the more evenly income is distributed across a population, the smaller the gini coefficient. table 1 shows differences in the net gini coefficient between developed and developing countries. the mean net gini is (10.2) points less in developed countries than in developing countries, which signifies lower income inequality. the minimum and maximum for developed countries are (16.8) for slovakia and (51.6) for chile. the range for developing countries is (22.6) for belarus and namibia (65.4). table 1. net gini coefficient by panel. full panel developed countries developing countries countries in panel 106 42 64 observations 2,744 1,132 1,612 mean 38.5 32.0 42.2 standard deviation 8.72 7.02 7.35 min-max 16.8-65.4 16.8-51.6 22.6-65.4 internal conflict measures the sum of the three subcomponents of risk of civil war/coup d'état threat, terrorism/political violence, and civil disorder. a score of (12) signifies the absence of conflict, no opposition to the government, and a government not engaging in arbitrary violence against its people. a score of (0) denotes an active civil war. as discussed in the literature review, the expectation is worsening internal conflict leads to an increase in the net gini (alesina and perotti, 1996). table 2 shows differences in internal conflict by panel. the mean internal conflict is (2.17) points higher in developed countries than in developing countries, which reflects less internal conflict. the minimum scores for developed countries are (3.0) for chile and israel during periods of severe internal conflict. many developed countries have had the highest possible score of (12). for developing countries, the minimum is (0) for lebanon, sri lanka, and el salvador during periods of active civil war. several developing countries have had the highest score of (12). parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 84 table 2. internal conflict by panel. full panel developed countries developing countries countries in panel 106 42 64 observations 2,744 1,132 1,612 mean 9.02 10.4 8.23 standard deviation 2.23 1.51 2.22 min-max 0-12 3-12 0-12 external conflict measures the three subcomponents of the risk of an external war, crossborder conflict, and foreign pressure (e.g., sanctions and withholding of aid). a score of (12) indicates very low risk, while (0) signifies an active cross-border war. the researchers expect worsening external conflict will have less of an effect on income inequality than internal conflict for two reasons. first, external conflict, which includes nonviolent subcomponents as well as lesser conflicts including cross-border tension, is less likely to destroy production facilities or disrupt the production of goods and services. second, the external conflict measure includes nonviolent factors such as sanctions and the withholding of aid which might not affect income inequality as much as violent conflict. table 3 shows differences in external conflict by panel. the mean external conflict (1.11) points higher in developed countries, showing less external conflict. iran and lebanon have the minimum score of (0) in the developing panel of countries, representing an active cross-border war. the minimum for developed countries is (3) for cyprus and israel. many countries for both developed and developing panels have the max score of (12). the econometric model includes covariates accounting for demographics, economic development, trade/globalization, political economy, investment risk, economic growth, and unemployment. the covariates are based on the relationship with conflict or the need to account for confounding variables that otherwise could lead to spurious results. first, conflict by nature, is the crystallization of economic, social, and political problems. when there is a large discrepancy in income, conflict can manifest through economic, social, and political factors. therefore, the model includes the covariates of per capita gdp, economic growth, political economy, unemployment, investment risk, institutional quality, trade, and the human capital index. second, the model includes important covariates to account for the economic structure and population distribution. table 3. external conflict by panel. full panel developed countries developing countries countries in panel 106 42 64 observations 2,744 1,132 1,612 mean 9.82 10.5 9.39 standard deviation 1.84 1.42 1.93 min-max 0-12 3-12 0-12 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 85 the natural log of per capita gdp accounts for differences in income and economic development (heston et al., 2012). the economic growth rate is the annual gdp growth rate. kuznets (1955) and kaldor (1957) find economic growth increases income inequality when a country is in earlier stages of development, while economic growth can decrease income inequality as countries become richer. therefore, the expectation is increases in the natural log of per capita gdp will increase income inequality for developing countries, but decrease income inequality for developed countries. the human capital index scores a country's average schooling years and returns to education (feenstra, inklaar, and timmer, 2015), as conflict can impact both. a country experiencing conflict may have fewer resources for education. additionally, returns to education may be reduced because of economic disruptions, including loss of job opportunities. the expectation is a decrease in the human capital index will increase income inequality in most cases (parsons, 2023). the effects of human capital on income inequality are partly determined by the dispersion of education in labor markets. the composition effect increases the wage of the smaller segment of the population with more human capital, which increases income inequality (knight and sabot, 1983). as the dispersion of human capital spreads more widely across a population, more of the population has higher wages, and wages compress since the supply of educated workers increases (knight and sabot, 1983), decreasing income inequality. polity data evaluates a country's regime. conflict may be less in regimes where the government is responsive and held accountable, and citizens have a part in selecting the regime. the polity indicator ranges from (0) for strongly autocratic to (6) for strongly democratic regimes. the score is based on competitiveness, openness, political participation, and checks on executive authority. the effects of increases in democratization on income inequality are indeterminant (olson, 1965; acemoglu et al., 2013). the study controls for institutional strength and quality of bureaucracy, and it is on a scale of (0) low institutional strength and bureaucracy quality to (4) high institutional strength and bureaucracy quality. countries with better institutions and bureaucracy may have less conflict since government services have fewer disruptions. additionally, better institutions and a quality bureaucracy can help cushion the effects of some forms of conflict. for example, although riots were at the united states capitol, the united states electoral college and the legislative branch prevented a larger conflict. the expectation is that increasing institutional strength and bureaucracy quality will reduce income inequality (huber, 2002). trade and globalization is measured through the gdp-adjusted sum of imports and exports. globalization and trade can lead to conflict. trade, especially in the form of cheaper imports from multinational enterprises (mne), affects domestic markets, which could lead to greater unemployment and subsequently cause conflict (stiglitz, 2013). the effects of trade on income inequality are inconclusive (heckscher, 1919; ohlin, 1933). if economic growth through trade increases the income of low income groups disproportionally higher than higher-income groups, income inequality will decrease (stolper-samuelson, 1941). on the other hand, if trade increases incomes of high income groups more than low income groups, income inequality will increase. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 86 the investment risk profile is on a scale of (0) high risk to (12) low risk and is based on the subcomponent measures of expropriation, profit repatriation, and payment delays. high investment risk can lead to conflict through its impact on the economy and foreign investment. a country with high risk, where foreign investors fear expropriation, will have less foreign investment, adversely affecting the economy. in this case, conflict can arise, and poorer economic performance may lead to higher income inequality (alesina and perotti, 1996). the unemployment rate is the percentage of the labor force not working or seeking employment. unemployment can lead to conflict (stiglitz, 2013). higher levels of unemployment increase income inequality since a larger percentage of the population receive no labor income (furceri and ostry, 2019). the age distribution of a population affects working-class versus non-working-class percentages, leading to income inequality variations among countries (burtless, 2009). the age dependency ratio is the percentage of people younger than 15 and older than 64 compared to the working-age population. it measures the burden of the working-age population caring for both the young and old. the larger the ratio, the greater the burden for the working class to support the young and old. additionally, as the segment of the population outside the workforce increases, the larger the segment of the population without a working income, which increases income inequality. the percentage of the labor force employed in manufacturing accounts for the economic structure of a country. the labor force transition from agriculture to manufacturing can decrease income inequality as low income rural households obtain higher incomes in urban manufacturing. young (2013) finds that urban-rural income inequalities may account for 40 percent of intercountry income inequality. 3.2 empirical framework the study includes 106 countries in the following panels: full panel, developed countries, and developing countries. using different panels isolates differences based on development status, and variables, such as per capita gdp, will have different relationships based on development status (kuznets, 1955). appendix b lists the countries in each specific panel. the data is unbalanced, with 2,744 observations in the full panel with country-level data from 1988 to 2018. although balanced panels are preferred for analysis, the dataset is more extensive and representative, with more countries and observations. also, more extensive country representation increases observations, and the larger sample reduces selection bias. model (1) borrows from barro (2000) and lundberg and squire (2003), who study economic growth and income inequality determinants. the panel model regresses the net gini coefficient on the determinants of internal conflict, external conflict, and the previously explained covariates. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 87 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝐼𝐼𝐺𝐺𝐺𝐺𝐼𝐼𝐼𝐼𝐺𝐺𝐼𝐼𝑖𝑖𝑖𝑖 + 𝐸𝐸𝐸𝐸𝐺𝐺𝐼𝐼𝐼𝐼𝐺𝐺𝐼𝐼𝑖𝑖𝑖𝑖 + 𝑋𝑋𝑖𝑖𝑖𝑖 + 𝜇𝜇𝑖𝑖 + 𝜆𝜆𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖 𝑎𝑎𝐺𝐺𝑎𝑎 (𝐺𝐺 = 1, … 𝐺𝐺; 𝐺𝐺 = 1, …𝑇𝑇) (1) 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 is the measure of income inequality for country (i) and time (t). 𝐼𝐼𝐺𝐺𝐺𝐺𝐼𝐼𝐼𝐼𝐺𝐺𝐼𝐼𝑖𝑖𝑖𝑖 is the measure of internal conflict that varies across time and country. 𝐸𝐸𝐸𝐸𝐺𝐺𝐼𝐼𝐼𝐼𝐺𝐺𝐼𝐼𝑖𝑖𝑖𝑖 is the measure of external conflict that varies across time and country. xit is the vector set of control variables used in the model that vary across time and countries. the parameter 𝛼𝛼 contains a constant and individual-specific variable invariant over time. the 𝜇𝜇𝑖𝑖 captures unobservable individual-specific effects and 𝜆𝜆𝑖𝑖 captures unobservable time-specific effects. 𝜀𝜀𝑖𝑖𝑖𝑖 is the error term. see appendix d for specification testing results. model specification analysis leads to selecting fixed effects models for both country and time. heteroskedasticity, autocorrelation, and crosssectional dependence are also evident in the data. to account for heteroskedasticity, autocorrelation, and cross-sectional dependence, the study uses driscoll and kraay (1998) standard errors which use cross-sectional averages of nonparametric standard errors to adjust for all three. the newey-west (1987) modifications to cross-sectional averages while adjusting the standard error estimates are used to ensure the covariance matrix estimators remain consistent and independent of the cross-sectional dimensions. although the lengths of the periods range from 18 to 31 years in this study, and driscoll and kraay (1998) standard errors rely on large-t asymptotics, the data set (panel) is not too short. lags of one, three, and five years are used to account for correlations and endogeneity. using longer lags, such as five years, analyzes the extent conflict leads to longer-lasting effects on income inequality. pugh (2013) claims internal conflict can have an enduring impact on income inequality by reducing tax revenue, reducing funds available to redistribute, and reallocating resources from education to the military. a potential analytical concern is the assumption of the exogeneity of conflict variables. omitted variable bias is possible if an unobserved variable jointly determines conflict and income inequality. the issue is mitigated through the fixed effects estimations, which control for unobservable factors (baltagi, 2001). the structure of panel data analysis accounts for omitted variables. a second potential issue is reverse causality due to the possibility of higher income inequality leading to internal and external conflicts. although high income inequality can be synonymous with deprivation, political protest, and violence (feirarbend and feirarbend , 1966; gurr, 1968; huntington, 1968), many cross-national studies consistently find they do not increase risks of conflict (fearon and laitin, 2003; collier and hoeffler, 2004; buhaug, et al., 2014). to eliminate this possibility, the dependent variable is lagged up to five years which mitigates the endogeneity and reverse causality issues. selection bias is a potential issue if certain countries are underrepresented because of sparse data. the large panel sizes for developing and developed countries help reduce the problem. a tradeoff is the larger grouping reduces generalizability. for example, since the developing country panel include low income through upper middle income countries, some caution should be used when generalizing the finding across an entire panel group. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 88 4. results 4.1 panel regressions table 4 provides three different models utilizing different groups of countries when the dependent variable is lagged 3 years. appendix e and f does the same while changing the lag length to 1 and 5 years. the results are consistent across the models with different lags, suggesting the persistence and stickiness of income inequality (ermisch and smeeding, 2012; pugh, 2013; bircan, brück, and vothknecht, 2017). the results show worsening internal conflict causes an increase in income inequality in developing countries. the decrease of one unit of internal conflict, for example worsening internal conflict from (9) to (8), will increase the net gini by (0.133) for developing countries. a one standard deviation worsening of income inequality will increase the net gini by about (0.068). for a specific example, cameroon has a 2018 internal conflict score of around 6, while the united states has a score near 10. if cameroon had the united states' lower conflict level while controlling for all other variables, cameroon's net gini coefficient would have been lower by (0.798). the results also show worsening internal conflict causes a statistically significant increase in the net gini in the full and developing countries panels but not for the developed country panel. an important question is why is there a difference in statistical significance between developing and developed countries. as discussed earlier, kuznets (1955) indicates that as developed economies grow, their income inequality declines. consequently, developed countries have lower income inequality than developing countries (see table 1). furthermore, their income inequalities are close once adjusting for outliers. finally, they have less internal conflicts than developing countries (see table 2). lack of internal conflict means that it cannot change income inequality. to demonstrate causality, it is necessary for the "cause" to change the "affected" or the anticipated outcome. it is clearly evident in table 4, where none of the variables depicting different indicators of conflict are statistically significant for developed countries, even though the model, as a whole, is significant. only "trade and globalization" and "age dependency ratio" are statistically significant, which could be due to random outcome. in addition, developed countries tend to have better institutions and bureaucracy, which may help absorb minor internal conflict shocks. other possibilities include that an increase in trade causes an increase in exporting and importing sectors more than other sectors, increasing income inequality. table 4 also provides the results of external conflict. external conflict is not statistically significant for any of the development stages. we offer two explanations for this outcome. first, although theoretically, an external conflict should adversely affect many social and economic measures of a country, and hence, income inequality, however, there are not enough of them to allow a meaningful impact on income inequality. to test this hypothesis, it is necessary to only consider cases where there are significant changes in external conflict and assess its impact. second, much literature focuses specifically on war, while the icrg external conflict measure considers the subfactors of diplomatic pressures, including sanctions, withholding of aid, and trade restrictions. therefore, the distinctive differences in the measures, especially between a parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 89 cross-border war and diplomatic pressures, may also explain the statistical insignificance compared to the literature (bircan, brück, and vothknecht, 2017). table 4. internal and external conflict – 3 year lag. full panel developed countries developing countries no. in group 106 42 64 obs. 2,744 1,132 1,612 f 1.35e+09 5233819 1.14e+07 r2 .234 .196 .308 internal conflict -.129*** (.055) -.120 (.130) -.133** (.054) external conflict .071* (.041) .056 (.053) .017 (.071) per capita gdp (log) 2.23*** (.683) -1.18 (.730) 3.47*** (.692) investment profile .005 (.035) .074 (.050) .015 (.047) political economy .088 (.071) .123 (.102) .105 (.086) institutional strength and quality of bureaucracy -.617*** (.147) -.323 (.292) -.733*** (.164) industry % of economy -.049** (.020) .009 (.041) -.045* (.023) trade and globalization .005 (.003) .011*** (.003) -.003 (.005) human capital index -1.76*** (.492) .570 (.378) -3.82*** (.848) gdp growth rate .016 (.018) .051 (.038) .001 (.013) unemployed rate .041** (.020) .024 (.022) .025 (.027) age dependency ratio .110*** (.018) .081*** (.017) .086*** (.015) note: *** p <0.01, ** 0.01<p<0.05, * 0.05<p<0.10. dependent variable is the net gini coefficient. standard errors in parenthesis. 4.2 results – panel regressions with low, medium, and high internal conflict as indicator variables the study also examines the effects of different internal conflict risk levels on income inequality. we create three groups of indicator variables based on risk categories. the categories include high risk countries with an internal conflict score between (0) to (4), the medium risk group has internal conflict between (5) to (8), and the low risk group has internal conflict between (9) to (12). table 5 provides three different models utilizing these 3 groups when the dependent variable is lagged 3 years. the base group is the medium risk countries, and the results show the effects of a country moving from medium risk to high risk or medium risk to low risk. the results indicate a decrease in the net gini of (0.431) when a country moves from the medium risk category to the parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 90 low risk category. the results further support the literature's claim that lower internal conflict reduces income inequality. as noted previously, the lack of statistical significance for developed countries relates to a small variance in internal conflict and income inequality after accounting for outliers. an interesting finding is the movement from medium risk to high risk is statistically insignificant. an explanation for this outcome is the distribution of internal conflict is rightward skewed, and most observations are in the medium and low risk categories. the small subset of high internal countries are outliers. see appendix g for internal conflict distribution. therefore, a reason for the lack of statistical significance is we are comparing the medium risk countries to a group of outliers. table 5. internal conflict as high, medium, and low risk – 3 year lag. full panel developed countries developing countries no. in group 106 42 64 obs. 2,744 1,132 1,612 f 1.35e+09 60857 9460780 r2 .234 .196 .309 high internal conflict (scores of 0 to 4) -.347 (.262) .209 (.489) -.412 (.300) medium internal conflict (scores 5 to 8) base comparison group low internal conflict (scores 9 to 12) -.315** (.140) .459 (.325) -.431*** (.147) external conflict .028 (.043) .028 (.050) -.021 (.076) per capita gdp (log) 2.26*** (.693) -1.13 (.710) 3.50*** (.687) investment profile .002 (.033) .079 (.053) .005 (.049) political economy .065 (.071) .101 (.084) .091 (.083) institutional strength and quality of bureaucracy -.632*** (.154) -.509 (.345) -.739*** (.166) industry % of economy -.055** (.021) .004 (.051) -.049* (.024) trade and globalization .005 (.003) .011*** (.004) -.004 (.005) human capital index -1.71*** (.471) .632* (.361) -3.75*** (.811) gdp growth rate .014 (.018) .051 (.040) -.001 (.013) unemployed rate .045** (.019) .030 (.021) .035 (.027) age dependency ratio .113*** (.018) .087*** (.020) .085*** (.015) note: *** p <0.01, ** 0.01<p<0.05, * 0.05<p<0.10. dependent variable is the net gini coefficient. standard errors in parenthesis. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 91 5. conclusion the study investigates the impact of internal and external conflict on the net gini coefficient. the study finds that worsening internal conflict causes a statistically significant increase in income inequality in developing countries, but not developed countries. the findings support kuznets's theory (1955) that explains smaller gini coefficients in low income developing countries, larger gini coefficients at the intermediate levels of development, and then smaller gini coefficients in high income developed countries. there is less internal conflict in developed countries, and developed countries have better institutional buffers, which leads to the lack of statistical significance. external conflict is not a statistically significant determinant of income inequality in developed or developing countries. the authors propose that since the icrg external conflict measure includes nonviolent subcomponents of withholding of aid, trade restrictions, and sanctions, the broadness of the measure may partly explain the lack of statistical significance. additionally, since there is little variation in external conflict, studies must focus on countries where external conflict changes, and then assess the effects on income inequality. this article expands the current literature in several ways. first, as noted earlier, we find the effects of internal conflict are greater in developing countries than developed ones, a point that other studies have missed. second, in developing countries, the study finds improvement in risk from medium to low reduces a country's income inequality. third, it uses more comprehensive measures of conflict. the icrg conflict data include a minimum of three subcomponents. the broadness of the conflict measures has advantages and disadvantages. the benefit is analyzing a more extensive set of factors of conflict. the broader lens of conflict improves the breadth and depth of the internal and external conflict measures. on the other hand, the broader measures limit the generalizability of specific subfactors. thus, it could be the case that civil war matters, but civil disorder matters less. alternatively, cross-border wars matter, but diplomatic pressures do not matter as much. in other words, as the subcomponents of each internal or external conflict measure get compiled into one overall rating, one needs to be thoughtful and careful with generalizations. the fallacy of division would lead to a faulty conclusion that because the broader external conflict measure is statistically insignificant, each of the three subcomponents must also be insignificant. additionally, it is a mistake to conclude that since the internal conflict measure is statistically significant for developed countries, it must mean all three subcomponents are significant, as well. other contributions to the literature include the model. it accounts for cross-sectional dependence, which may cause biased estimators in cross-national income inequality studies. the research also includes important covariates, such as the dependency ratio, which is missing in some models in the literature. the study also includes recent data and covers the last 30 years in 106 countries. the researchers note the limitations of the study. potential omitted-variable bias exists, but the two-way fixed effects and the panel data structure mitigate the problem. missing data and selection bias always cause some concern, but it is mitigated through the panel structure and parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 92 examining groups based on development status. future research should explore other mechanisms which affect income inequality through conflict and stability. also, measures isolating the three subcomponents of internal and external conflict could lead to important insights. acknowledgments all data used in the study are publicly available and accessible. we want to acknowledge funding for the icrg dataset from pepperdine university and helpful suggestions and comments from david smith, jared ashworth, blake withall, and the anonymous referee of the european journal of government and economics. any remaining shortcomings are those of the authors. references acemoglu, d., naidu, s., restrepo, p., and robinson, j. (2013). democracy, redistribution and inequality. national bureau of economic research working paper 19746. https://doi.org/10.3386/w19746 alesina, a., and perotti, r. (1996). income distribution, political instability, and investment. european economic review, 40(6), 1203-1228. https://doi.org/10.1016/0014-2921(95)00030-5 ariely, g., and uslaner, e. (2017). corruption, fairness, and inequality. international political science review, 38(3), 349-362. https://doi.org/10.1177/0192512116641091 baltagi, and badi, h. (2001). econometric analysis of panel data. wiley, john and sons. barro, r. (2000). inequality and growth in panel of countries. journal of economic growth, 5, 5-32. https://doi.org/10.1023/a:1009850119329 barry, b. (2005). why social justice matters. cambridge: polity press. bircan, c., brück, t. and vothknecht, m. (2017). violent conflict and inequality. oxford development studies, 45:2, 125-144. https://doi.org/10.1080/13600818.2016.1213227 buhaug, h., cederman, l., and gleditsch, k. s. (2014). square pegs in round holes: inequalities, grievances, and civil war. international studies quarterly, 58, 418-431. https://doi.org/10.1111/isqu.12068 burtless, g. (2009). demographic transformation and economic inequality, ch. 18 in w. salverda, b. nolan, and t. m. smeeding (eds), the oxford handbook of economic inequality, oxford, oxford university press, 435-454. cederman, l., weidmann, n. b., and gleditsch, k. s. (2011). horizontal inequalities and ethnonationalist civil war: a global comparison, american political science review, 105, 478-495. https://doi.org/10.1017/s0003055411000207 collier, p. (1999). on the economic consequences of civil war. oxford economic papers, 51, 168183. https://doi.org/10.1093/oep/51.1.168 collier, p., and hoeffler, a. (2002). military expenditure: threats, aid and arms races (world bank policy research working paper no. 2927). washington, dc: world bank. https://doi.org/10.1596/1813-9450-2927 collier, p., and hoeffler, a. (2004). greed and grievance in civil war. oxford economic papers, 56, 563-595. https://doi.org/10.1093/oep/gpf064 collier, p., and hoeffler, a. (2007). civil war. in t. sandler and k. hartley (eds.), handbook of defense economics (pp. 711-739). elsevier. https://doi.org/10.1016/s1574-0013(06)02023-0 dabla-norris, e., kochhar, k., suphaphiphat, n., ricka, f. and tsounta, e. (2015). causes and consequences of income inequality: a global perspective. imf staff discussion note sdn/15/13. washington, dc: imf. https://doi.org/10.5089/9781513555188.006 deininger, k., and olinto, p. (2000). asset distribution, inequality, and growth (world bank policy research working paper no. 2375). washington, dc: world bank. https://doi.org/10.1596/18139450-2375 https://doi.org/10.3386/w19746 https://doi.org/10.1016/0014-2921(95)00030-5 https://doi.org/10.1177/0192512116641091 https://doi.org/10.1023/a:1009850119329 https://doi.org/10.1080/13600818.2016.1213227 https://doi.org/10.1111/isqu.12068 https://doi.org/10.1017/s0003055411000207 https://doi.org/10.1093/oep/51.1.168 https://doi.org/10.1596/1813-9450-2927 https://doi.org/10.1093/oep/gpf064 https://doi.org/10.1016/s1574-0013(06)02023-0 https://doi.org/10.5089/9781513555188.006 https://doi.org/10.1596/1813-9450-2375 https://doi.org/10.1596/1813-9450-2375 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 93 deininger, k., and squire, l. (1996). a new data set measuring income inequality. the world bank economic review. 10(3), 565-591. https://doi.org/10.1093/wber/10.3.565 dobson, s., and ramlogan, c. (2009). is there an openness kuznets curve?" kyklos. 62(2), 226238. https://doi.org/10.1111/j.1467-6435.2009.00433.x dorling, d. (2011). injustice and the 1%. london and new york, ny: verso. driscoll, j., and kraay, a. (1998). consistent covariance matrix estimation with spatially dependent data. review of economics and statistics, 80, 549-560. https://doi.org/10.1162/003465398557825 duquette, n. (2018). inequality and philanthropy: high-income giving in the united states 1917-2012. explorations in economic history, 70, 25-41. https://doi.org/10.1016/j.eeh.2018.08.002 ermisch, j, and smeeding, t.m. (2012). from parents to children: the intergenerational transmission of advantage, new york: russell sage foundation. fearon, j. d., and laitin, d. (2003). ethnicity, insurgency, and civil war. american political science review, 97, 75-90. https://doi.org/10.1017/s0003055403000534 feenstra, r., inklaar, r., and timmer, m. (2015). the next generation of the penn world table. american economic review, 105(10), 3150-3182. https://doi.org/10.1257/aer.20130954 feierabend, k., and feierabend, r. (1966). aggressive behavior within polities. journal of conflict resolution, 10, 249-271. https://doi.org/10.1177/002200276601000301 furceri, d., and ostry, j. (2019). robust determinants of income inequality. oxford review of economic policy, 35(3), 490-517. gurr, r. (1968). a causal model of civil strife: a comparative analysis using new indices. american political science review. 62, 1104-1124. https://doi.org/10.2307/1953907 haggard, s., and kaufman, r. (2012). inequality and regime change: democratic transitions and the stability of democratic rule. american political science review, 106(3), 1-22. https://doi.org/10.1017/s0003055412000287 hardy, m. (1979). economic growth, distributional inequality, and political conflict in industrial societies. journal of political and military sociology, 5, 209-227. heckscher, e. (1919). the effect of foreign trade on the distribution of income. a translation is provided by h. flam and m. j. flanders, eds., heckscher-ohlin trade theory (cambridge, ma: mit press, 1991). this article originally appeared in swedish in ekonomisk tidskerift, 21, 497512. heston, a., summers, r., and aten, b. (2012). penn world table version 7.1. center for international comparisons of production, income and prices at the university of pennsylvania. huber, j. (2002). deliberate discretion? japanese journal of political science, 4, 157-159. https://doi.org/10.1017/cbo9780511804915 huntington, s. (1968). political order in changing societies. new haven, ct: yale univ. press. kaldor, n. (1957). a model of economic growth. the economic journal, 67(268), 591:624. https://doi.org/10.2307/2227704 karklins, r. (2005). the system made me do it: corruption in post-communist societies. armonk, n.y: m.e. sharpe. knight, j., and. sabot, r. (1983). educational expansion and the kuznets effect, american economic review. 73 (5), 1132-1136. kuznets, s. (1955). economic growth and income inequality. the american economic review, 45(1), 1-28. lorenz, o. (1905). methods of measuring the concentration of wealth. publications of the american statistical association, 9(70), 209-219. https://doi.org/10.1080/15225437.1905.10503443 lundberg, m., and squire, l. (2003). the simultaneous evolution of growth and inequality. the economic journal, 113(487), 326-344. https://doi.org/10.1111/1468-0297.00127 milanovic, b. (2016). global inequality: a new approach for the age of globalization. cambridge, ma: harvard university press. https://doi.org/10.4159/9780674969797 nagel, j. (1974). inequality and discontent: a nonlinear hypothesis. world politics. 22, 453-472. https://doi.org/10.2307/2010097 newey, w., and west, k. (1987). a simple, positive semi-definite, heteroscedasticity and autocorrelation consistent covariance matrix. econometrica, 55, 703-708. https://doi.org/10.2307/1913610 ohlin, b. (1933). interregional and international trade. cambridge, ma: harvard university press. olson, m. (1965). the logic of collective action: public goods and the theory of groups. cambridge: harvard university press. parsons, b. (2023). panel data analysis of the human capital index and income inequality. applied econometric and international development, 23(1), 5-32. https://doi.org/10.1093/wber/10.3.565 https://doi.org/10.1111/j.1467-6435.2009.00433.x https://doi.org/10.1162/003465398557825 https://doi.org/10.1016/j.eeh.2018.08.002 https://doi.org/10.1017/s0003055403000534 https://doi.org/10.1257/aer.20130954 https://doi.org/10.1177/002200276601000301 https://doi.org/10.2307/1953907 https://doi.org/10.1017/s0003055412000287 https://doi.org/10.1017/cbo9780511804915 https://doi.org/10.2307/2227704 https://doi.org/10.1080/15225437.1905.10503443 https://doi.org/10.1111/1468-0297.00127 https://doi.org/10.4159/9780674969797 https://doi.org/10.2307/2010097 https://doi.org/10.2307/1913610 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 94 plato. republic. books i, ii, iv and "the myth of the cave: from book vii." morality and the good life: an introduction to ethics through the classical sources. 5th ed. eds. robert c. solomon, clancy w. martin, and wayne vaught. trans. g.m.a. grube. boston: mcgraw-hill, 2009. 81-104. posner, r. (1997). equality, wealth, and political stability. journal of law, economics, and organizations, 13(2), 344-364. https://doi.org/10.1093/oxfordjournals.jleo.a023387 pugh, m. (2003). protectorates and spoils of peace: political economy in south-east europe. in d. jung (ed.), shadow globalization, ethnic conflicts and new wars. london: routledge. https://doi.org/10.4324/9780203216965 sahota, g. (1978). theories of personal income distribution: a survey. journal of economic literature, 16(1), 1-55. solt, f. (2015). economic inequality and nonviolent protest. social science quarterly, 96(5), 13141327. https://doi.org/10.1111/ssqu.12198 stiglitz, joseph. (2013). the price of inequality. new perspectives quarterly. 30(1). https://doi.org/10.1111/npqu.11358 stolper, w., and samuelson, p. (1941). protection and real wages. review of economic studies, 9(1), 58-73. https://doi.org/10.2307/2967638 tocqueville, a. (1835). democracy-in america, volume ii. new york: schocken books. united nations development program. (2021). human development report 2020. new york: oxford press. von schulze-gaevernitz, g. (1915). origins of british supremacy of the seas. current history, 3(3), 560-563. https://doi.org/10.1525/curh.1915.3.3.560 weede, e. (1981). income inequality, average income, and domestic violence. journal of conflict resolution, 25, 639-653. https://doi.org/10.2307/2600358 young, a. (2013). inequality, the urban-rural gap, and migration. quarterly journal of economics, 128(4), 1727-1785. https://doi.org/10.1093/qje/qjt025 https://doi.org/10.1093/oxfordjournals.jleo.a023387 https://doi.org/10.4324/9780203216965 https://doi.org/10.1111/ssqu.12198 https://doi.org/10.1111/npqu.11358 https://doi.org/10.2307/2967638 https://doi.org/10.1525/curh.1915.3.3.560 https://doi.org/10.2307/2600358 https://doi.org/10.1093/qje/qjt025 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 95 appendix a. summary statistics variable source observ. mean st. dev. min max net gini coefficient swiid 2,744 38.12 8.81 19.5 66.5 external conflict icrg 2,744 10.2 1.56 2 12 internal conflict icrg 2,744 9.02 2.05 0 12 age dependency ratio world bank 2,744 61.9 18.1 27.0 117 employment in manufacturing world bank 2,744 21.6 7.85 2.54 46.0 gdp growth rate world bank 2,744 2.23 4.09 -40.7 24.0 human capital index penn world tables (10.0) 2,744 2.52 .700 1.05 3.97 imports + exports % gdp world bank 2,744 77.8 50.74 11.1 437.3 investment risk profile icrg 2,744 7.94 2.31 0 12 natural log of gdp per capita world bank 2,744 8.68 1.49 5.21 11.6 quality of bureaucracy and institutions icrg 2,744 2.4 1.10 0 4 unemployment rate world bank 2,744 8.27 6.33 .1 38.8 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 96 appendix b. panel list, by country full panel: albania, algeria, angola, argentina, armenia, australia, austria, bangladesh, belgium, bolivia, botswana, brazil, brunei, bulgaria, burkina faso, cameroon, canada, chile, china, columbia, costa rica, cote d'ivoire, croatia, cyprus, czechia, denmark, dominican republic, ecuador, egypt, el salvador, estonia, finland, france, gambia, germany, ghana, greece, guatemala, guyana, honduras, hong kong, hungary, iceland, india, indonesia, iran, ireland, israel, italy, jamaica, japan, jordan, kazakhstan, kenya, latvia, lithuania, luxembourg, madagascar, malawi, malaysia, mali, mexico, moldova, mongolia, morocco, mozambique, namibia, netherlands, new zealand, nicaragua, niger, nigeria, norway, pakistan, panama, paraguay, peru, philippines, poland, portugal, qatar, romania, russia, senegal, sierra leone, singapore, slovakia, slovenia, south africa, south korea, spain, sri lanka, sudan, sweden, switzerland, tanzania, thailand, tunisia, turkey, uganda, ukraine, united kingdom, united states, uruguay, venezuela, yemen, zambia, zimbabwe. developed countries: australia, austria, belgium, brunei, canada, chile, croatia, cyprus, czechia, denmark, estonia, finland, france, germany, greece, hong kong, hungary, iceland, ireland, israel, italy, japan, latvia, lithuania, luxembourg, netherlands, new zealand, norway, poland, portugal, qatar, romania, singapore, slovakia, slovenia, south korea, spain, sweden, switzerland, united kingdom, uruguay, and united states. developing countries: albania, algeria, angola, argentina, armenia, bangladesh, bolivia, botswana, brazil, bulgaria, burkina faso, cameroon, china, columbia, costa rica, cote d'ivoire, dominican republic, ecuador, egypt, el salvador, gambia, ghana, guatemala, guyana, honduras, india, indonesia, iran, jamaica, jordan, kazakhstan, kenya, madagascar, malawi, malaysia, mali, mexico, moldova, mongolia, morocco, mozambique, namibia, nicaragua, niger, nigeria, pakistan, paraguay, peru, philippines, russia, senegal, sierra leone, south africa, sri lanka, sudan, tanzania, thailand, tunisia, turkey, uganda, ukraine, venezuela, yemen, zambia, and zimbabwe. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 97 appendix c – correlation matrix internal conflict external conflict gdp per capita depend. ratio invest. profile employed industry% imp + exp (% gdp) human capital gdp growth unempl. internal conflict 1.00 external conflict .568 1.00 gdp per capita .500 .260 1.00 depend. ratio -.412 -.197 -.726 1.00 invest. profile .411 .218 .524 -.399 1.00 employed industry% .352 .184 .596 -.679 .213 1.00 imp + exp % gdp .320 .188 .298 -.322 .298 .170 1.00 human capital .461 .202 .817 -.781 .496 .549 .274 1.00 gdp growth .082 .019 -.039 -.141 .096 .048 .100 .038 1.00 unempl. -.087 -.004 -.032 .040 -.117 .012 -.071 -.058 -.136 1.00 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 98 appendix d. specification testing 1. hausman test for fixed versus random effects=fixed effects 2. joint test for time fixed-effect=indicator variable for time. 3. wald test for heteroscedasticity = presence of heteroskedasticity 4. pesaran (cd) test for cross-sectional dependence = presence of crosssectional dependence 5. woolridge test for autocorrelation = presence of autocorrelation 6. variance inflation factor (vif) test for multicollinearity=1.87 parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 99 appendix e. table 2. internal and external conflict – 1 year lag full panel developed countries developing countries no. in group 106 42 64 obs. 2,744 1,132 1,612 f 4.99e+09 239496 2.11e+07 r2 .234 .200 .301 internal conflict -.129*** (.044) -.120 (.108) -.134*** (.043) external conflict .071* (.037) .056 (.047) .017 (.064) per capita gdp (log) 2.23*** (.526) -1.18* (.628) 3.43*** (.562) investment profile .005 (.030) .074 (.045) .015 (.041) political economy .088 (.055) .123 (.098) .105 (.066) institutional strength and quality of bureaucracy -.617*** (.119) -.323 (.266) -.733*** (.133) industry % of economy -.049*** (.015) .009 (.041) -.045*** (.019) trade and globalization .005* (.003) .011*** (.003) -.004 (.004) human capital index -1.76*** (.382) .570* (.317) -3.81*** (.669) gdp growth rate .016 (.017) .051 (.037) .001 (.013) unemployed rate .041** (.016) .024 (.018) .025 (.024) age dependency ratio .110*** (.015) .081*** (.017) .086*** (.012) note: *** p <0.01, ** 0.01<p<0.05, * 0.05<p<0.10. dependent variable is the net gini coefficient. standard errors in parenthesis. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 100 appendix f internal and external conflict – 5 year lag full panel developed countries developing countries no. in group 106 42 64 obs. 2,744 1,132 1,612 f 1.10e+09 913616 1.68e+07 r2 .234 .196 .308 internal conflict -.129*** (.055) -.120 (.138) -.133** (.058) external conflict .071* (.040) .056 (.052) .016 (.071) per capita gdp (log) 2.23*** (.746) -1.18 (.720) 3.47*** (.745) investment profile .005 (.034) .074 (.049) .015 (.042) political economy .088 (.079) .124 (.093) .105 (.094) institutional strength and quality of bureaucracy -.617*** (.166) -.323 (.271) -.733*** (.187) industry % of economy -.049** (.022) .009 (.051) -.045 (.027) trade and globalization .005 (.003) .011*** (.004) -.003 (.005) human capital index -1.76*** (.835) .570 (.378) -3.82*** (.939) gdp growth rate .016 (.017) .051 (.037) .001 (.011) unemployed rate .041** (.020) .024 (.024) .026 (.026) age dependency ratio .110*** (.020) .081*** (.022) .086*** (.017) note: *** p <0.01, ** 0.01<p<0.05, * 0.05<p<0.10. dependent variable is the net gini coefficient. standard errors in parenthesis. parsons and naghshpour / european journal of government and economics 12(1), june 2023, 79-101 101 appendix g. internal conflict distribution 18 25 36 50 96 170 294 511 722 902 883 900 0 20 0 40 0 60 0 80 0 10 00 fr eq ue nc y 0 1 2 3 4 5 6 7 8 9 10 11 12 internal conflict 0=worst 12=best internal conflict distribution 1. introduction0f 2. literature review – conflict as a determinant of income inequality 3. methods and data 3.1 data description and variable selection 3.2 empirical framework 4. results 4.1 panel regressions 4.2 results – panel regressions with low, medium, and high internal conflict as indicator variables 5. conclusion acknowledgments vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 119-142 european journal of government and economics issn: 2254-7088 how the uefa financial fair play regulations affect football clubs’ priorities and leagues’ competitive balance? pedro garcia-del-barrioa*, giambattista rossib a universitat internacional de catalunya, spain b birkbeck university of london, united kingdom * corresponding author at: pgarcia@uic.es article history. received 14 december 2019; first revision required 23 january 2020; accepted 20 february 2020. abstract. institutional and legal reforms applied in the context of sports competitions often provoke structural changes that can be empirically investigated. using a data set of 560 observations (20 teams per season, from 2009/10 to 2015/16, of teams playing in the premier league, la liga, serie a and ligue 1), this paper examines how the uefa financial fair play (ffp) regulations may have altered the football clubs’ decisions concerning their sport and financial priorities. moreover, based on a simple theoretical description, the paper shows that the increasing financial stability promoted by the ffp rules might actually imply – as an undesired side effect – a declined competitive balance affecting the european football leagues. finally, the paper discusses some policy issues and recommendations. keywords. institutional regulations; professional football; financial fair play; competitive balance; wage-to-revenue ratios jel codes. d22; j24; j33 doi. https://doi.org/10.17979/ejge.2020.9.2.5842 1. introduction and purpose this paper focuses on the impact that the introduction of the uefa1 financial fair play (ffp) regulations may have had on the football clubs’ priorities concerning their sport and financial outcomes. this feature is critical since, among other things, the increasing financial stability produced by the ffp rules could eventually entail – as an undesired side effect – a declined competitive balance across the football teams competing in european leagues. institutional and legal reforms affecting sports competitions typically generate structural changes that can be empirically investigated. examples of such reforms in the context of european professional football include: (i) the re-design in season 1992/93 of the competitive structure of the uefa champions league; (ii) the so-called bosman law introduced in december 1995; and (iii) changes in the bargaining system (individual/collective) of broadcasting contracts. there are various objectives that this paper tries to achieve. first, it re-examines the findings of earlier works, that we mention later in the paper, by conducting an empirical analysis over 1 the uefa, union of european football associations, is the governing body of european football. in this paper, “club” and “team” are often used indifferently. mailto:pgarcia@uic.es https://doi.org/10.17979/ejge.2020.9.2.5842 garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 120 four leagues (premier league, la liga, serie a and ligue 1) and 7 seasons (2009/10 to 2015/16), thereby resulting in a data set of 560 observations (20 teams per league and season). specifically, our analysis verifies the existence of a positive relationship between (i) spending in players’ talent – the clubs’ wage bill − and sports achievements; and between (ii) teams’ sporting success and teams’ revenues. second, we study the effects that the approval of new rules, on the part of the football regulatory institutions, may have in altering the clubs’ decisions regarding both sport performances and financial outcomes. specifically, we explore empirically the existence of structural changes resulting from the reform imposed by the uefa ffp regulations, which are intended to create a more sustainable financial situation in european football. to accomplish the papers’ goals, we carry out several empirical analyses: (i) to verify if the evidence supports that the ffp regulations reduce the wage-to-revenue ratios (wrr) and their dispersion; (ii) to identify what domestic leagues were more sensitive to changes in the ffp rules (evidence of structural breaks); and (iii) to study if the ffp regulations will help to limit wage inflation2 while preserving the overall investment in talent. finally, we discuss the implications that the new ffp rules may have, along with its effects in terms of the wrr, on the competitive balance (cb) of european football leagues. the latter aspect is prospectively addressed by means of a simple theoretical elaboration, which shows how the increasing financial stability procured by the ffp rules is likely to deteriorate the current leagues’ cb. accordingly, we wonder if revenue sharing regulations should be imposed along with the ffp rules to prevent the deterioration of cb in the leagues. 2. related literature the broad scope of the paper recommends referring to several fields of related literature. first, following earlier works (cf.: szymanski and smith (1997); forrest and simmons (2002); and barajas and rodriguez (2010), among others), our empirical analysis corroborates the existence of a positive relationship between wage spending in players’ (as a proxy of the level of the teams’ talent) and sport achievements. in line also with the analysis made by szymanski and smith (1997), we find as well the usual positive and statistically significant correlation between teams’ sport success and annual revenues. the other bibliographical references are grouped in two areas: contributions related to the ffp literature and papers on superstar players and overinvestment. 2 contracts to hire football players are undoubtedly influenced by the link between wage bills and transfer fees. this feature is especially relevant in front of the inflation that seems to affect more intensively the transfer fees of superstar players. it may be that transfer fees of the “average” players have not been so much affected by inflation as they are the transfer fees of players at the top of the talent distribution function. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 121 2.1. super-star players and overinvestment in talent given the link between the club’s spending on sport ability and the resulting wage bill, our study also relates to the issue of talent investment to hire players. garcia-del-barrio (2018) and garcia-del-barrio and pujol (2020) show the necessity of the teams to account for popularity and media visibility status, along with sporting skills, when facing hiring (investment) decisions on talent. pawlowski and anders (2012) stress the role of brand investments, which is the result of people showing a preference for strong brands and sport superstars. rohde and breuer (2016) empirically examine a number of european clubs and find that their financial success is driven by brand value as well as by sporting success; i.e. sporting achievements are determined by the amount of investment. then, aguiar-noury and garcia-del-barrio (2019) provide empirical evidence to advocate that the historical brand status of football clubs has to be taken into account for reducing the risk typically associated to investments in the sports industry. besides, football clubs are usually prone to sacrifice economic returns to improve their (short-run) sport achievements. hence, given these perplexing incentives, it is not surprising the advent of mismanagement in the form of over-investment in talent (excessive wage spending). in this context, dietl et al. (2011) study investment in talent to explore how it depends on whether the clubs are more or less win-oriented. overinvestment and revenue dissipation in sports are also treated in the economic literature. (cf.: dietl et al. (2008); and franck (2014)3). although spending in sport talent is investment decision, which is meant to procure future gains, attempts are made to keep excessive rewards (wage bills) under control; for instance, by the institution of salary caps in some american leagues. other regulations on sport institutions were implemented to try fostering competitive balance; examples in the context of the labour market include: amateur drafts and free agency restrictions (caporale and collier, 2015). another issue related to our topic is the escalating spending involving the decisions of firms’ operating in the same sector. the arms race phenomenon arises in many contexts, including advertising campaigns, where companies try to gain market share from the other companies. but then, the competitors engage in the battle, as they face the prospect of losing market share. as a result, there is excessive unnecessary use of economic resources, given that the crucial issue is the relative rather than the absolute level of investment. regarding football, the issue is described in the following quote from andreff (2012): “the promotion/relegation system and win maximization pave the way to a clubs’ arm race (sanderson, 2002) in which each club attempts to recruit the best players on the eve of season in order to improve its relative situation compared with its opponents; in turn, the latter are incited to overbid. thus, each club’s demand for talent is excessive because all clubs aim at winning the same sports contest. however, investing in additional player recruitment is not efficient when a club attempts by all allowed means to have an advantage 3 cf.: franck (2014, p. 6): “football championship races are examples of a certain form of economic competition, which is known as contest in the literature (…). even if we assume rational behaviour and profit maximizing clubs, contests may exhibit an interesting phenomenon of revenue dissipation called overinvestment.” because football leagues are good examples of contests, their analysis can benefit from the contest’s theory: tullock (1980), lazear and rosen (1981). garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 122 over its opponents because, consequently, all clubs are then forced to behave in the same way in order to remain competitive. such investment is socially efficient only if absolute quality (and not relative quality as in a game between two teams) considerations prevail (lazear and rosen, 1981). there are only a few winners (promoted clubs, those qualifying in a european level contest) in the arms race while cost increases – wage and transfer fee inflation – are all the more generalized across all clubs and so they are not slowed down by profit maximization” (p. 96). backman (1968) points out how economists typically advocate that advertising involves a wasteful of resources. the argument is that the very same market share would eventually result had competitor firms agreed on a coordinated reduction of excessive expending in activities like advertising or payrolls. actually, given its inner competitive nature, sport contests are paradigmatic examples where “arms races” can eventually develop into football clubs’ waste of resources. moreover, this phenomenon is especially relevant in the context of the prevailing patterns at hiring talent in sports labour markets. notice that we are not referring here to exactly the same phenomenon than the “college athletics arms race” (cf.: edwards, 1984),4 since even if spending decisions are usually made while taking into account their rivals’ behaviours, universities are bonded by explicit rules that forbid them to compete for hiring new talented players on prices basis. to examine this issue, we explore if deviations from the average wage of domestic competitors works better than absolute values, implying that what matters the most is the comparative position of one team relative to the others. actually, researchers often model behavioural wage equations (the relationship between individual salaries and performances) on the bases of their position with respect to other rivals, rather than on absolute values. (cf.: torgler and schmidt (2007); peeters and szymanski (2014); caporale and collier (2015); and garcia-del-barrio and tena-horrillo (2019)). 2.2. the uefa “financial fair play” in the literature in this paper we explore how football clubs’ priorities and objectives may be affected by institutional and legal reforms. our analysis is actually going to focus on the impact provoked by the ffp regulations introduced by uefa to prevent clubs from spending beyond their economic means. the ffp rules, introduced to promote clubs’ financial discipline and stability, were approved in september 2009, but they were only fully implemented in 2011. since then, for the clubs to be allowed to participate in uefa competitions, they must prove having no unpaid 4 competition among university colleges have developed intense commercial battles (arms races) beyond the sporting field. in recent years, there have been an increasing number of papers addressing the issue of “arms races” in college athletics: cf.: litan et al. (2003); orszag and orszag (2005); orszag and israel (2009); tsitsos and nixon ii (2012); hoffer et al. (2014); caro and elder (2017); etc. the first mentioned paper identifies relevant industries where the arms races have become paradigmatic: the purchase of “weapons arms races” in the defences industry or “medical arms races” in hospitals (health industry). similarly, in the context of sport inter-colleges competitions, investing in sport facilities are described as “athletics arms races”. litan et al. (2003) also refer to the beginning of an “arms race” specific to football and basketball, as increasing revenues and capital investments provide financially support to improved programs to hire and maintain gifted athletes and coaches. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 123 debts affecting other clubs or tax authorities. the uefa first delivering of ffp ruling was made in april based on the clubs’ accounts of seasons 2011/12 and 2012/13. in order to enforce clubs to balance their wage spending with respect to revenues, and to prevent them of accumulating excessive debt, uefa implemented in 2013 a financial ruling: the break-even requirement.5 it was not until may 2014 that uefa imposed the first sanctions to clubs failing to fulfil the break-even requirements and, hence, the first conditions stemming from the non-compliance of those requirements were actually at work for the first time in the 2014/15 season. franck (2014) discusses the effects of the ffp regulations, expressing a concern about the new rules due to several reasons. first, since imposing limits to wage spending (like the salary caps, a common institution in us professional leagues) is usually made to the aim of enhancing the cb (cf.: peeters and szymanski, 2014).6 second, given that, as stressed by vöpel (2011), the uefa regulations may create entry barrier that will further consolidate the current hierarchy of european clubs: the richest clubs become richer and the poorest become poorer. following madden (2015), franck (2014) also alert that ffp rules may be associated to forgone earnings from the “external” resources that could have otherwise expanded the clubs’ payrolls. to our knowledge, there are still few papers focused specifically on studying the impact of the break-even requirement imposed by ffp rules on clubs’ actual wrr. heiskanen (2017) finds that football clubs are now more disposed to be financially responsible and the ffp regulations have led to smaller wrr (mainly in the premier league and spanish la liga), a result that also applies to teams that did not qualify to participate in the uefa’s competitions. ghio et al. (2019) study the effect of ffp rules on the italian calcio empirically. their paper focusses on how these rules affect the clubs’ cost efficiency regarding the trade-off between sporting and economic outcomes, and find that the ffp does not improve the average efficiency of italian teams. gallagher and quinn (2019) empirically study the effect of the breakeven financial constraints on the joint sporting and financial efficiency of 60 english football clubs for the period running from 2003/2004 to 2016/2017. using a stochastic non-parametric efficiency analysis, they find that the breakeven rules bring forth reductions in the clubs’ efficiency (on average), and that uefa financial regulations may force clubs to grant prior attention to the financial achievements, thereby weaken the future competitive intensity in english football.7 in line with heiskanen (2017), one of our paper’s aims is also identifying changes in wrr due to the introduction of the ffp regulations. however, our paper has a broader scope for two reasons. first, it develops the empirical analysis on a richer data set – involving more teams 5 besides, an independent entity, the club financial control body (cfcb), was set up to supervise the fulfilment of the ffp regulations and clubs’ licensing system. to fulfil their role, the cfcb evaluates financial figures of the last three seasons for all clubs in uefa competitions. the punishments imposed by uefa include economic penalizations, but also: transfer bans, warnings, points deductions, ban to register new players and restrictions on the number of players registered to participate in european competitions, etc. 6 however, gains in competitive balance by introducing the new ffp rules seems not to be expected. moreover, this paper argues that the ffp regulations are likely to deteriorate the future competitive balance of european football leagues. 7 notice that our paper does not focus on overall competitive intensity analyzed by these authors, but on the fact that the ffp regulations may presumably diminish the quality of the competitions through undermining of the actual competitive balance of teams competing in both domestic and european leagues. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 124 and years – and, second, it examines how the new rules might have affected the clubs’ priorities concerning sport and financial achievements.8 the final part of the paper focuses on the implications that the new ffp regulations and their effect in reducing the wrr across football clubs might have on the european leagues (un)competitive balance. the issue of competitive balance (cb) and its link to the “outcome uncertainty” (ou) hypothesis was studied right from the birth of sports economics. the contributions by rottenberg (1956) and neale (1964) are considered seminal papers. since then, other papers have studied the issues of cb and the ou hypothesis, and the way they are related to each other (cf.: owen (2014), among others). regarding the empirical studies, there is no agreement on the relevance of the ou hypothesis, given that contrasting empirical evidence is found when studying how cb influences the degree of interest in sports. (cf., for instance: garcia and rodriguez (2002) and coates and humphreys (2012)). as a final remark, we suggest that the introduction of revenue sharing mechanisms could smooth out the otherwise deteriorating degree of cb across teams in european football leagues. indeed, revenue sharing is already at work, especially in the context of broadcasting revenues. this point is precisely congruent with a spanish law, introduced in 2015, enforcing football clubs in la liga to grant their individual rights for joint commercialization and establishing that 50% of the revenues of the first division league must be shared in equal proportions among the 20 teams.9 our study contributes to filling a gap in the literature as it benefits from a very rich database. moreover, our approach to these topics is made keeping in mind the trade-off between win and profit maximising behaviour of football clubs described in the economic literature. 3. methods and data sources the empirical analysis is made upon data at the club level. the estimated models involve two financial variables: annual revenues (rit) and annual wages (wit). our database comprises all the teams competing in the first division leagues of four out of the five main european competitions: the premier league, spanish la liga, italian serie a and french ligue 1. (financial data for german teams was unavailable). we have been actually able to gather a data set of 560 observations: 20 teams per league and season, over the period 2009/10 to 2015/16. table 1 summarises the descriptive statistics of the main variables employed in this paper and, given the relevance of the wrr, the statistics are also shown by seasons in this case.10 8 our empirical analysis is conducted, for the period 2009/10 to 2015/16, using a sample of 20 teams in the first division leagues every year, from four of the main football leagues in europe: premier league, la liga, serie a and ligue 1. (data for the german bundesliga is not available). 9 real decreto-ley 5/2015, de 30 de abril, de medidas urgentes en relación con la comercialización de los derechos de explotación de contenidos audiovisuales de las competiciones de fútbol profesional (exposición de motivos, n.2). 10 annual wages and revenues were obtained from: deloitte “football money league” (dfml) and deloitte “annual report of football finance” (arff) published over the years. other data was collected from the clubs’ accounts or from data bases: sabi, aida, amadeus and hoovers. due to missing values, the regression analysis was made on 558 or 557 (rather than 560) observations for, respectively, the wages and revenue equations. due to strong financial garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 125 table 1. descriptive statistics of the main variables. variable obs. mean std. dev. min. max. % revenue / wage (rwr) 557 160.56 39.69 44.75 305.17 % wage / revenue (wrr) 557 66.24 17.86 32.76 223.41 annual wages (mill.) 558 67.48 65.02 7.64 371.73 annual revenues (mill.) 557 108.50 115.89 13.42 690.10 ucl games 560 0.78 1.79 0.00 8.00 eur games 560 0.65 1.59 0.00 8.00 points in domestic league 560 51.93 16.16 17.00 102.00 league rank 560 10.50 5.77 1.00 20.00 wage-to-rev. ratio (by season) season obs. mean std. dev. min. max. 2009-10 79 18.35 68.72 34.95 144.34 2010-11 79 21.72 70.28 34.56 177.86 2011-12 80 23.65 68.95 35.87 223.41 2012-13 80 15.38 64.88 34.86 128.68 2013-14 80 15.26 63.17 37.61 148.81 2014-15 79 13.19 63.17 32.76 94.51 2015-16 80 13.90 64.54 39.52 106.81 sources: deloitte arff, deloitte dfml and authors’ own collection from clubs’ accounts. the variable “ucl games” accounts for the number of rounds that teams played in the uefa champions league; whereas “eur games” are the corresponding rounds in the uefa europa league. the meaning of the other variables in the table is straightforward. the first task to be accomplished is examining the patterns behind wrr of clubs across leagues and over time. to this aim, the following section presents a number of plots where, instead of grouping relative frequencies, we deliver the smoother graphical representations that result from estimating kernel probability density functions (rather than histograms). 4. disparity in wrr patterns across football leagues a mere inspection of the available data leads to achieving important conclusions. first, figure 1 suggests that a turning point altering the wrr might have occurred in season 2013/14, possibly due to the introduction of the ffp regulations by uefa. (the first release of the ffp ruling was based on the clubs’ financial accounts of 2011/12 and 2012/13. yet, the compliance conditions of the break-even requirements were first adopted in 2014/15). comparing the behaviour of wrr in the years prior to introducing the uefa rules (kernel density functions in red) and the new prevailing pattern that emerges since then (represented in blue), once the conditions for the compliance of the break-even requirements were adopted for the first time, it seems that a structural break occurred at that time. (the annex shows a similar kernel density analysis where each season is treated separately). difficulties, three clubs failed to publish data on annual revenues in one of the seasons: portsmouth (premier league, in 2009/10); hércules cf (la liga, in 2010/11); and parma fc (serie a, 2014/15); and two of them did not publish either the annual wages: portsmouth (premier league, in 2009/10 ); and hércules cf (la liga, in 2010/11). garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 126 figure 1. kernel density functions – wage-to-revenue ratios (changes over the seasons). the analysis of wrr in the four mentioned football leagues is also enlightening. figure 2 shows substantial discrepancies across domestic competitions, where the french ligue 1 stands out with the highest wrr figures, suggesting that wage inflation is more intense than in the other european leagues. the italian serie a appears to be the league with the lowest wrr figures. notice though that we must be cautious here, as some differences may exist in the way how clubs present their financial accounts across countries. (see for this issue “note 11”, which describes the sources of data, including the clubs’ accounts and other databases). figure 2. kernel density functions – wage-to-revenue ratios (by domestic leagues). garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 127 figure 3. standard deviations of wage-to-revenue ratios (by football leagues). a complementary description of the discrepancies across the domestic leagues is shown in figure 3, which illustrates the evolution of the standard deviations (of each league) over the years, based on the figures previously reported in table 1. a simple inspection of the figure reveals that the standard deviations converge to similar values. this feature seems to have especially benefited the spanish la liga, as it has experienced a diminishing wrr disparity across teams. besides, the graphical analysis also points towards a possible structural change occurring soon after the introduction of ffp regulations. for the purpose of our paper, in figure 4, we also analyse the evolution of wrr standard deviations (figure 3) to the evolution of the clubs’ annual wages. apart from the italian serie a, all the leagues display a growing trend, which suggests general wage inflation. as it is obvious from the wrr evolution, annual revenues increased in line to the wages in the lasts years. figure 4. standard deviations of annual wages (by football leagues). 0 5 10 15 20 25 30 35 40 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 premier league la liga serie a ligue 1 total 0 10 20 30 40 50 60 70 80 90 100 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 premier league la liga serie a ligue 1 total garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 128 figure 4 reveals that wage disparity across teams is growing bigger, evidence of increasing financial unbalances, which is often blamed for creating greater sport competitive unbalance too.11 moreover, the financial imbalance becomes even stronger precisely at the time when the ffp rules were implemented. nevertheless, as far as the empirical evidence is concerned, the question about the effects of ffp regulations should be examined upon longer data series (on wages and performances) before a more conclusive judgment on this matter is delivered. anyway, a further and more refined empirical treatment of this issue is carried out in the following section. finally, we consider the implications of estimating models with variables in levels or variables expressed in deviations from their mean. on one side, given that what really matters are the relative strengths between adversaries (cf.: sanderson 2002), in model estimation the latter option (variables in deviations from the league average) is expected to work better than using variables in absolute values. on the other side, from a theoretical perspective, we also expect – given the nature of sports competitions – that models in deviations from the mean will perform better to explain clubs’ behavioural equations than the others. this point is crucial for judging whether the football labor market makes efficient or inefficient use of its resources. if teams’ performances are basically explained by the relative wages (in deviations from the corresponding league’s average), then it is a nonsense for clubs wanting to hire talent to engage in “arms races” with other competitors. that is to say, insofar as the relative wages (rather than their absolute values) are the main driving force to achieve sport success, there is increasing risk of unnecessary waste of resources due to bidding along with other competitors to hire sport talent. the empirical analysis of the next section is also helpful to examine this feature and to see to what extent does it affect the football labor market. 5. empirical analysis and results this section explores if changes in the behavior and objectives of football clubs may be the result of imposing new institutional regulations. specifically, our attention focuses on the changing patterns observed in the clubs’ wrr (as shown in section 3), to examine whether they may be due to the implementation of the ffp rules and break-even requirements by uefa. this empirical analysis would help us to understand to what extent the new financial restrictions might have affected the football clubs’ decisions concerning their sport and economic priorities and ultimately their sport and economic outcomes. to fulfil the aforementioned aims, this empirical section is organised in three main subsections. first, we run dynamic regression “wage-to-revenue ratio” models, including a lag of the dependent variable, to try identifying structural changes in wrr patterns (overall and in each of the domestic leagues). then, we estimate “revenue equations”, where the clubs’ annual revenue is defined as a function of sports performances; and, also “productivity 11 in this regard, see: szymanski (2006), who in page 201 says: “inequality in income has translated more or less directly into inequality in the strength of playing squads”). this point is however a matter of discussion. precisely szymanski (2006) also stresses that inequality in the playing field seems not to have deteriorated according to the increase in the clubs’ financial inequality. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 129 equations”, where sport performances are explained by alternative settings of the football teams’ annual wage bills. initially, we introduce the two fundamental functions on which the empirical analysis is made, although the discussion of further theoretical aspects is reserved for the next section: rit = f (dspit, clpit, eupit) (1) dspit = g (wit) (2) where: rit = total annual revenues club i and season t (millions of €) dspit = domestic sport performance (points) club i and season t clpit = number of qualifying rounds in the champions league club i and season t eupit = number of qualifying rounds in the europe league club i and season t wit = total annual wages club i and season t (millions of €) first, equation (1) postulates a positive empirical relation between annual revenues and sport performances, by taking into account the domestic games but also the sport achievements in european competitions. domestic sport performance is measured by the number of points accumulated at the end of the national league, whereas sport performance in european competitions counts the number of qualifying rounds that the teams went through. notice that as far as the number of league points is concerned, given that all the leagues have the same number of teams (and games) and they obey to very similar competition rules, there is no essential difference between using this variable in levels or in mean-deviations. then, equation (2) is based on the idea that sport achievements are determined by sport talent, as capture by the teams’ annual wage bills. for the sake of simplicity, we constrain here the analysis to domestic sport achievements as dependent variable. previous papers (carmichael et al. (2011), among others) have conducted similar empirical analyses as the ones defined in equation (1) and equation (2). 5.1. estimations of “wage-to-revenue ratio” models section 3 provided evidence on structural changes happening around 2014 that have altered the clubs’ behavior concerning wrr. in this section, this issue is more carefully examined through regression analyses by introducing seasonal dummies in search of significant alterations – across teams of the same league – from the usual wrr figures. we further hypothesize that the fact that wrr changes occurred immediately after the new regulations adopted by uefa hints that the latter fact may be the cause of the former. the regression analysis is initially carried out for the pooled ols model, with robust standard errors. then, to benefit from the additional information contained in panel data analysis, and to account for the potential influence of individual heterogeneity elements, we also estimate fixed effects (fe) and random effects (re) models. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 130 equation (3) describes the specification of these models: 𝑊𝑊𝑊𝑊𝑊𝑊𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 ∙ 𝑊𝑊𝑊𝑊𝑊𝑊𝑖𝑖𝑖𝑖−1 + ∑ ϑ𝑖𝑖 ∙ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖5 𝑖𝑖=1 + ∑ γ𝑗𝑗 ∙ 𝐿𝐿𝑆𝑆𝑆𝑆𝐿𝐿𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑗𝑗 3 𝑗𝑗=1 + 𝜇𝜇𝑖𝑖𝑖𝑖 (3) to elucidate whether the fixed or the random effects model must be preferred, we computed the hausman (1978) test.12 in our case, the “chi-square” (with 6 degrees of freedom) is equal to 88.43, with a p-value of 0.0000, implying that the null hypothesis (ho: difference in coefficients is not systematic) has to be rejected. given the results of the test, which strongly support the fe model, table 2 only reports – along with the pooled model – the fe estimated coefficients. this is aligned with the expected theoretical outcome, as every football club is supposed to have specific characteristics invariant over time. in the regression analysis we include a set of dummy variables to control for distinctive traits of the domestic leagues (with the italian serie a taken as the reference). the results of the pooled model in table 2 suggest that ffp regulations might have provoked a consistent reduction of football clubs’ wrr over time, as the coefficients of the seasonal dummies (with season 2010/11 being the reference value) are negative. this conclusion applies especially to season 2012/13: the size of the coefficient (–4.849) is greater than the ones attached to other seasons; and, more importantly, it is statistically significant. furthermore, examination of the fe estimations reveals that structural changes on wrr are more significant (in size and statistically) when accounting for individual heterogeneity elements, as table 2 also discloses. table 2. wage to revenue ratios – pooled ols and fixed effects models. pooled ols p-value fixed effects p-value lag of wrr 0.5483*** [0.000] 0.1744*** [0.002] premier league 6.3617*** [0.002] spanish la liga 3.6584* [0.088] french ligue 1 9.3253*** [0.000] season 2011/12 –2.2948 [0.438] –1.6074 [0.499] season 2012/13 –4.8494* [0.091] –5.3549** [0.028] season 2013/14 –4.3309 [0.138] –5.0366** [0.041] season 2014/15 –3.7724 [0.178] –5.5113** [0.029] season 2015/16 –3.6351 [0.176] –4.8504* [0.052] constant 28.7302*** [0.000] 59.1155*** [0.000] no. obs. 407 407 no. groups – 102 r-squared 0.4242 0.3269 aic 3,320.34 3,155.58 prob > f [0.000] [0.006] statistical significance: * p < 0.1; ** p < 0.05; *** p < 0.01 | and [p-values] in brackets. 12 the hausman test is based on the following: if the specification of the model is right, the null hypothesis of no correlation is fulfilled and, therefore, the coefficient estimates of the fixed and random effects models should not statistically differ. under fulfilment of the null hypothesis, both fixed and random effects estimators are consistent; but the latter model should be chosen since their estimations are more efficient. however, in our case the null hypothesis is rejected and, as the re estimations are not consistent, the fe model must be chosen. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 131 table 3. wage to revenue ratios – ols regressions by leagues (levels). premierl p-value la liga p-value serie a p-value ligue 1 p-value lag of wrr 0.7742*** [0.000] 0.5122*** [0.008] 0.6426*** [0.000] 0.4010*** [0.000] season 2011/12 –1.8073 [0.507] –7.3778 [0.421] –1.7141 [0.651] 1.2196 [0.801] season 2012/13 2.7377 [0.425] –18.768** [0.029] 3.6121 [0.268] –6.8848* [0.071] season 2013/14 –14.6106** [0.000] –8.4217 [0.370] 5.4714 [0.118] –1.1957 [0.777] season 2014/15 0.5750 [0.836] –12.7850 [0.126] –4.8504 [0.646] 1.9228 [0.426] season 2015/16 –0.0115 [0.996] –16.978** [0.039] 1.7396 [0.639] 1.3489 [0.740] constant 18.6422*** [0.003] 42.4499** [0.011] 18.4191*** [0.000] 47.0782*** [0.000] no. obs. 102 102 101 102 r-squared 0.6583 0.3737 0.4436 0.2002 aic 732.07 916.37 760.13 795.20 prob > f [0.000] [0.000] [0.006] [0.000] statistical significance: * p < 0.1; ** p < 0.05; *** p < 0.01 | and [p-values] in brackets. our results suggest the existence of structural change occurring around the year 2013, a finding that could be arguably related to the uefa ffp restrictions.13 we further refine the analysis by replicating the same estimation models by leagues, in search of distinctive patterns associated to each of them. table 3 gathers the results of the pooled ols model estimation by leagues. (the same specification models were estimated accounting for fe, but they are they yield similar results, with nearly identical significance levels attached to the coefficients of seasonal dummies, and hence are not shown). the analysis by leagues allows us distinguishing the time at which the wrr experienced drops in each of the domestic football leagues, except for the italian serie a, all the other football leagues display sharp reductions of wrr at once: in france and spain it happened in 2012/13, whereas the drop affected the premier league in the following season. for the case of spanish football, it is also worth noting the negative and significant coefficient associated with season 2015/16. this fact may perhaps be related to the new law (real decreto-ley) to regulate football tv commercial rights that was ruled in spain on the 30th april 2015. (cf.: the description offered in the “note 8”). to summarise the findings until now, this section has provided solid evidence that wrr of european clubs tend to converge towards smaller and more similar figures all across the main football leagues. there are little discrepancies in the timing of this process, which is meant to be a direct consequence of uefa rules; namely, ffp regulations and break-even requirements. 5.2. estimation of “revenue equation” models the “revenue equation”, as defined earlier in this section by expression (1), studies the relevance of sport performances to explain football clubs’ capacity to generate revenues. the simple specification of the functional form used in the estimations is described in equation (4): 13 these regulations are imposed on clubs that qualified to the uefa leagues, which must prove to have no overdue debts. although they were initially approved in 2010, the break-even requirements to balance clubs’ spending and revenues were only introduced in 2013, and the first sanctions to clubs were only effective in season 2014/15. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 132 𝑊𝑊𝑖𝑖𝑖𝑖 = 𝛿𝛿0 + 𝛿𝛿1 ∙ 𝑆𝑆𝑆𝑆𝐷𝐷𝑖𝑖𝑖𝑖 + 𝛿𝛿2 ∙ 𝐶𝐶𝐿𝐿𝐷𝐷𝑖𝑖𝑖𝑖 + 𝛿𝛿3 ∙ 𝐸𝐸𝐸𝐸𝐷𝐷𝑖𝑖𝑖𝑖 + ∑ ϑ𝑖𝑖 ∙ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖6 𝑖𝑖=1 + ∑ γ𝑗𝑗 ∙ 𝐿𝐿𝑆𝑆𝑆𝑆𝐿𝐿𝑆𝑆𝑆𝑆𝑆𝑆𝑗𝑗 3 𝑗𝑗=1 + ɛ𝑖𝑖𝑖𝑖 (4) we mentioned earlier that previous research addressed this issue by estimating models like the one in equation (4). nevertheless, the results of table 4 are precious for two reasons: first, the estimations are obtained from a very comprehensive data set, which contains 80 teams per season over 7 seasons; secondly, our analyses are useful to check the findings of previous research on the topic (cf.: szymanski and smith, 1997; and garcia-del-barrio and szymanski, 2009). besides, the results collected in table 4 are pertinent for the discussion on the clubs’ objectives and how their goals may be affected by institutional changes. table 4 reports the pooled ols estimations of three alternative models that differ on the way how the dependent variable was defined. whereas in model (r1) annual revenues are in levels; estimations of model (r2) are made for natural logarithms of the dependent variable; finally, model (r3) takes the revenues also in logs, but expressed in deviations from the league average. our estimations corroborate the main results of previous papers: teams’ sport performance appears as a major factor in generating revenues. furthermore, along with the domestic performances, sport achievements in european competitions are also relevant, especially as far as the uefa champions league is concerned. the case of the uefa europa league is less evident, given the size of the estimated coefficients, and since the coefficient is not statistically significant in model (r1). anyway, each of the models describing the revenue equations’ behavior has advantages and disadvantages. the first one is convenient as it facilitates interpreting the results; while the 2 other models perform better in terms of the r-squared the “akaike information criterion” (aic). table 4. annual revenue explained by sport performance (in both domestic and european leagues). revenues log revenues mean deviat. log revenues model (r1) p-value model (r2) p-value model (r3) p-value domestic points 2.6948*** [0.000] 0.0220*** [0.000] 0.0221*** [0.000] ucl games 32.4765*** [0.000] 0.1801*** [0.000] 0.1781*** [0.000] eur games 1.5394 [0.419] 0.0647*** [0.000] 0.0658*** [0.000] premier league 61.3159*** [0.000] 0.4668*** [0.000] –0.0471 [0.203] spanish la liga –23.9031*** [0.000] –0.5796*** [0.000] –0.1009** [0.015] french ligue 1 –33.1662*** [0.000] –0.5333*** [0.000] 0.0433 [0.283] season 2010/11 1.2078 [0.862] 0.0277 [0.567] 0.0255 [0.579] season 2011/12 12.7741* [0.072] 0.0984* [0.063] 0.0391 [0.445] season 2012/13 17.3383** [0.019] 0.1348*** [0.008] 0.0400 [0.415] season 2013/14 28.9518*** [0.000] 0.1952*** [0.000] 0.0172 [0.734] season 2014/15 39.2270*** [0.000] 0.2636*** [0.000] 0.0132 [0.773] season 2015/16 52.7217*** [0.000] 0.3574*** [0.000] 0.0229 [0.646] constant –81.2029*** [0.000] 2.9704*** [0.000] –1.3353*** [0.000] no. obs. 557 557 557 r-squared 0.7710 0.8483 0.8037 aic 6079.15 365.299 310.343 prob > f [0.000] [0.000] [0.000] statistical significance: * p < 0.1; ** p < 0.05; *** p < 0.01 | and [p-values] in brackets. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 133 5.3. estimation of “sport performance” models in this section, we focus on behavioural equations for “sport performances” of football clubs, as defined in equation (2). we then present estimation results of several alternative models, whose specifications have important implications for the scope of our paper. in the estimations, we rely on different model specifications, which are made explicit in expression (5). in all the models, the dependent variable (sport performance) is the number of points amassed in the domestic league. regarding the principal explanatory variable, annual wages are sometimes measured in levels, but also as deviations from the respective league mean: 𝑆𝑆𝑆𝑆𝐷𝐷𝑖𝑖𝑖𝑖 = 𝛼𝛼0 + 𝛼𝛼1 ∙ 𝑊𝑊𝑖𝑖𝑖𝑖 + 𝛼𝛼2 ∙ (𝑊𝑊 −𝑊𝑊� )𝑖𝑖𝑖𝑖 + ∑ ϑ𝑖𝑖 ∙ 𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖6 𝑖𝑖=1 + ∑ γ𝑗𝑗 ∙ 𝐿𝐿𝑆𝑆𝑆𝑆𝐿𝐿𝑆𝑆𝑆𝑆𝑆𝑆𝑗𝑗 3 𝑗𝑗=1 + 𝜐𝜐𝑖𝑖𝑖𝑖 (5) compared to the two initial models, (w1) and (w2), the third (w3) and fourth (w4) models are estimated after computing logarithmic transformations of the wages. the estimations of similar models are made for the sake of robustness, and the fact that the results are shown all together in table 5 is due to the meaningful conclusions achieved from comparing them. the results shown in table 5 have important implications in the discussion on what the clubs’ priorities are (sport success or economic returns) in decision making at hiring talent in sport labor markets. actually, the annual wage bills of the teams can be interpreted as a proxy of the intensity with which clubs invest in new talent, which is expected to improve the sport achievements while diminishing the clubs’ economic outcomes. table 5. sport performance in domestic competitions explained by sport talent (annual wages). model (w1) p-value model (w2) p-value model (w3) p-value model (w4) p-value annual wages 0.2086*** [0.000] –0.0108 [0.864] deviation wages 0.2237*** [0.000] log of wages 18.2155*** [0.000] –0.5888 [0.901] deviat. log wages 19.0857*** [0.000] premier league –10.1195*** [0.000] 1.0981 [0.727] –12.6099*** [0.000] 0.9730 [0.781] spanish la liga 2.2318* [0.073] 0.8961 [0.484] 7.0640*** [0.000] 0.7644 [0.684] french ligue 1 2.9668** [0.020] –0.4761 [0.767] 5.1370*** [0.000] –0.4831 [0.792] season 2010/11 –0.5973 [0.692] –0.4562 [0.757] –0.7480 [0.598] –0.4539 [0.745] season 2011/12 –1.8169 [0.259] –0.9405 [0.558] –1.9464 [0.228] –0.9524 [0.557] season 2012/13 –1.5105 [0.363] –0.4831 [0.772] –1.3112 [0.409] –0.5084 [0.748] season 2013/14 –1.9940 [0.250] 0.1334 [0.942] –1.8409 [0.251] 0.0894 [0.957] season 2014/15 –4.1499** [0.013] –0.2286 [0.911] –3.3935** [0.020] –0.3249 [0.840] season 2015/16 –6.0630*** [0.001] –0.1693 [0.935] –5.5885*** [0.001] –0.2927 [0.880] constant 41.4768*** [0.000] 52.6828*** [0.000] –15.8716*** [0.000] 54.3312*** [0.002] no. obs. 558 558 558 558 r-squared 0.5835 0.5961 0.6409 0.6511 aic 4217.863 4202.758 4135.132 4121.054 prob > f [0.000] [0.000] [0.000] [0.000] statistical significance: * p < 0.1; ** p < 0.05; *** p < 0.01 | and [p-values] in brackets. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 134 comparing model (w2) against model (w1), as well as (w4) against (w3), one conclusion seems clear: to explain sport success, wage deviations (from the average wage of competitor teams) works much better than absolute values. this result implies that what matters ultimately is the comparative position of a team – regarding investment in sport talent – relative to the other rivals. that is to say, insofar as the presence in the regressions of wages in meandeviations (either “dev. wages” or “dev. log wages”) make statistical insignificant the impact of the wages in levels (“annual wages” or “log wages”), it suggests that it is mainly the relative positioning of the competitors clubs what matters to explain the clubs’ behavior concerning the squad quality and sport achievements. this finding has implications in the area of industrial organization and labor markets. first, it alerts us of presumably an excessive use of resources by over-investing when hiring sports talent (involving not only the transfer fee but also the wage bill agreement). moreover, as empirical studies prove, most of the clubs behave as win rather than profit maximisers and, hence, have incentives to mismanagement in the form of excessive spending in sport talent. inasmuch as other clubs engage in incremental bids to hire players, there is a risk that the whole process becomes a wasteful “arms race” among competitors clubs. this issue is critical as it relates to the side effect of lower competitive balance resulting from the uefa new regulatory framework. despite the fact that ffp rules aim at preventing clubs from being trapped into growing debt and at ensuring a fairer competition, we argue in the next section that they may end up, even if unintendedly, increasing the competitive unbalance of the leagues. 6. prospective theoretical developments in this section, we use a simple theoretical approach to illustrate how the financial stability improvements procured by uefa may end up creating greater competitive imbalance across european teams. in principle, by inducing greater financial responsibility, the ffp regulations are certainly expected to lower the risk of financial failure of football clubs. however, we illustrate here that the break-even restrictions, even if unintendedly, will deteriorate the sport competitive balance by creating greater disparities among the clubs’ wage bills. thus, we next focus on the uefa regulations that take the form of break-even restrictions.14 they are the type of rules that are supposed to have made an impact in explaining the decrease of football clubs’ wrr and their dispersion (as captured by standard deviations). our theoretical analysis develops upon the following idea: given the inner nature of sport contests, it is reasonable to expect that weaker teams may be willing, in the short-run at least, to compensate their poorer sport talent by assuming greater financial instability. this is certainly the case of teams facing relegation risk or those wanting to consolidate their sport brand status. 14 as explained by vöpel (2011), the break-even requirement implies that the clubs’ spending is not allowed to exceed what uefa denotes “relevant revenues”, where income from external patrons or agents is not accounted for. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 135 the literature mentions similar arguments, referring mainly to the fact that restrictions imposed on the teams (like prohibiting the use of external resources) will presumably affect more intensively the functioning of weaker teams. more wide-ranging arguments are invoked by peeters and szymanski (2014)15, who refer to the spill-over effects linked to the drop in the cost of reaching a given level of sport talent and achievements. in summary, along with other papers, we claim here that the break-even restrictions avoid the possibility of deviating from pre-established wrr values, which fall into the interval considered to be free of financial risk. however, these limits hinder the chances that otherwise weaker clubs might have to reduce their wage gap with respect to other competitors; and then, the leagues can no longer compensate the competitive sport inequalities due to financial imbalances. 6.1. a simple theoretical setting the scope of this paper also relates to the discussion on football clubs’ objectives and priorities. the literature recognises a trade-off between two principal objectives of the teams (and clubs’ owners): either they behave as winning or as profit maximising organizations. for the sake of simplicity, we offer just an illustrative framework to describe the main aspects. in section 4, the two main equations (1) and (2) were defined as follows: rit = f (dspit, clpit, eupit) (1) dspit = g (wit) (2) in this section, we describe the simplest conceivable approach by modelling sport performance as a function of wages; and revenues as a function of sport performance. we also assume that there is only one function of sport performance comprising domestic as well as european games. thus, we simplify matters to the point of expressing revenues depending on wages.16 therefore, equation (1) can be re-written as: rit = f ( g (wit) ) = r (wit) (6) given the role of the wrr in this paper, it is also crucial to be aware of the economic 15 the following quote from peeters and szymanski (2014) describes extensively this point: “indeed, some have argued that the break-even rule will reduce competitive balance by limiting the opportunity of smaller teams to erode the dominance of the established teams since they will not be able to use outside resources to fund a challenge. our analysis shows how the impact of the restraint goes beyond those clubs wanting to spend more than their football resources. given that higher spending tends to generate better sporting performance, restraining the spending of some clubs reduces the cost of winning for all clubs. thus, there will be spill-over effects to clubs not directly affected by the limit, by lowering the cost of achieving a given level of success. (…) thus we argue that the vertical restraint introduced by uefa will produce an anti-competitive outcome which is comparable to the horizontal salary cap agreements which exist in the us, without the latter’s pro-competitive balance effects.” (p. 357-8). the issue was implicitly treated by dietl et al. (2009), since their analysis of a salary cap defined as a fixed proportion of the clubs’ income is equivalent to the break-even ffp rules imposed by uefa. 16 assuming that revenues depend on wages, this extremely simple framework actually neglects the possibility to take into account the dynamic behaviour of the equations or a more realistic specification with lagged variables. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 136 meaning attached precisely to the inverse of this ratio. expression (7) offers a simple relationship between relevant rations: the “revenues over wages ratio” (rwr) measures the economic returns and, hence, is equivalent to profits: 1 𝑊𝑊𝑊𝑊𝑊𝑊 = 1 𝑊𝑊 𝑊𝑊� = 𝑊𝑊 𝑊𝑊 = rwr (7) figure 5 illustrates the main variables involved in our description (wages, revenues, and rwr) expressed in all three cases as a function of wages. in the figure, we also represent the values of the interval (0.44; 3.05) containing the rwr values of actual football clubs, according to the information of our database. given that rwr is the inverse of wrr, imposing limits on wrr implies forcing teams to give more priority to profit than to winning maximizing objectives. the graphical characterization made in figure 5 highlights the value 1.5 (=3/2) of the rwr, which is equivalent to 2/3 of the corresponding wrr; a threshold that represents the imposition implied by the break-even requirements. the figure also displays the location of the outcome of the representative profit maximizing team, whose wage spending level (investment in talent) is far behind the usual behavior of football clubs. at the opposite extreme to the right, we find win maximizing teams, as the greater wages, the more talent accumulation and sport achievements. nonetheless, ffp rules in the form of break-even requirements encourage teams, at least those whose wage bill is greater than the pre-established ratio, to converge towards the aforementioned threshold: wrr=2/3 or, equivalently, rwr = 3/2 = 1.5. this implies that the teams’ choices are now distorted against sport performance maximization, while the restrictions on wrr force them to approach the profit maximizing outcome and, hence, to gain greater financial stability. football clubs are typically prone to sacrifice economic returns to improve their (short-run) sport achievements, break-even requirements to participate in uefa competitions do certainly introduce greater economic rationality. figure 5. description of a simple illustrative model. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 137 in this context, the issue of football clubs’ objectives could be discussed based on the elaboration made in equation (8), which may enlighten the discussion on the goals pursued by football clubs. the left-hand side of the equation defines the rwr (similar to clubs’ profits) and is split out into two factors, where the second one accounts for sport performance per unit of wage spending. 𝑊𝑊 𝑊𝑊 = 𝑊𝑊 𝑆𝑆𝑆𝑆 ∙ 𝑆𝑆𝑆𝑆 𝑊𝑊 (8) the analysis of the issue invites exploring the effect of wages on sport performance and the influence of the latter on the revenues, but this new research avenue remains for future research. 6.2. other issues and policy implications in economic analysis, the issue of causality is always controversial. in the previous sections, changes in the evolution (and general patterns) of the clubs’ wrr have been related to the break-even requirements as if the former were necessarily the consequence of the latter. this fact is difficult to prove empirically, but we propose here a simple example that may help to find out how the causality link operates in this regard. consider two rival teams that have initially identical wage bills: w1 = w2. the revenues of the first team are equal to its wages, meaning that it currently uses all its annual revenues to pay wages: w1 = r1. (this situation is not unusual in the short-run, even if it is incompatible with a sustainable financial situation in the long-run). instead, the amount of revenues of the second team are actually twice the value of its wages: w2 = 0.5 ∙ r2, implying that r2 = 2 ∙ r1. in table 6, we present one example where the initial allocations are made according to the previous description. then, the table further describes the effect of introducing break-even requirements implying that w / r < 2/3, a threshold that resembles the actual ffp regulations. table 6. impact of regulating the wrr: an illustrative example. team1 team2 team1/team2 wages w1 w2 w1/w2 revenues r1 r2 r1/r2 w/r w1/r1 w2/r2 initial allocation wages 3 3 1 revenues 3 6 1/2 w/r 1 1/2 scenario 1 wages 2 3 2/3 revenues 3 6 1/2 w/r 2/3 1/2 scenario 2 wages 2 4 1/2 revenues 3 6 1/2 w/r 2/3 2/3 garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 138 in the table, scenario 1 illustrates the case when only team 1 changes the initial allocations to strictly achieve the rule; hence: w1 = 2/3 ∙ r1. the wage spending in the case of the second team does already fulfill the restriction and is thus not forced to change its behavior. however, it may be that all the teams end up converging close to the established ratio, in which case, the figures of scenario 2 would result. regardless of which scenario is more likely to occur, in both cases sport unbalance between these two rival teams has grown bigger in either case: the wage gap between them (initially zero) ranges now between 2/3 and 1/2 of the other team’s wage bill. in summary, by imposing all teams the same wrr, the uefa restrictions actually enforce the salary gap between two teams to end up being similar to their revenue gap. only if teams were indifferent to winning maximizing objectives, would ffp rules make no significant impact on their competitive balance or unbalance. but we take for granted that football teams deviate from the economic objectives of other types of enterprises, as the competitive environment among teams involves a zero-sum outcome: one team wins, one loses, or the two rivals draw (cf.: peeters and szymanski, 2014). our findings merit more research effort, especially as other authors reach different conclusions. actually, the aforementioned and very relevant paper by peeters and szymanski (2014)17, the authors argue that, in addition to limit the clubs’ spending in payrolls, the uefa ffp regulations also have the effect of improving the competitive balance of teams, insofar as they moderate the competitive advantage of teams at the top. 7. conclusions and further research this paper addresses a number of issues in the context of european football, where the clubs must now fulfill certain rules to get the license to participate in the uefa competitions. (cf.: uefa, 2010). using a comprehensive dataset of 560 observations (20 teams per league and season over the period: 2009/10 to 2015/16), our empirical analyses accomplish several objectives. first, we corroborate empirical regularities characterizing the football industry: the existence of a strong positive relationship (i) between annual wage spending (in players’ talent) and sport performances and (ii) between teams’ sporting success and annual revenues. second, we find significant empirical evidence that the ffp rules enacted by uefa are likely to have dropped the football teams’ wrr, both the means and their degree of dispersion. this feature was also examined by leagues, to identify which of them have been more sensitive to changes in the ffp. our empirical analysis also supports that, concerning sport achievements, what matters is the relative – rather than the absolute – quality of the teams (as captured by the clubs’ wage bills) with respect to their competitors. moreover, this feature easily involves excessive spending in hiring talent, which entails a wasteful economic result overall. third, we examined the effects that ffp regulations have on the clubs’ priorities regarding sport and financial achievements. we actually illustrated – through a simple theoretical 17 they explore the consequences of the ffp rules, concluding that even if the uefa regulations do not explicitly impose a limit to the clubs’ wage bills, break-even requirements seem ultimately to produce this very effect. garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 139 framework – that the ffp rules, designed to procure greater financial stability in european football, have favoured to decline the level competitive balance across clubs. for this reason, our final remark suggests that revenue sharing could be a potential measure to counterbalance the expected loss of competitive balance. to conclude, further research could explore whether other types of structural breaks may occur in the football industry whenever structural reforms such as ffp or changes in the broadcasting contracts (from individual bargaining to collective agreements) may affect the professional football industry. references aguiar-noury, a., and garcia-del-barrio, p. (2019). global brands in soccer: identifying low-risk projects and business opportunities, journal of entrepreneurship and public policy, 8(1), 62– 83. https://doi.org/10.1108/jepp-03-2019-103 andreff, w. (2012). sport events, economic impact and regulation. in michel desbordes and andré richelieu (eds.), global sport marketing: contemporary issues and practice. (chapter 5). backman, j. (1968). is advertising wasteful? journal of marketing, 32(1), 2–8. https://doi.org/10.1177/002224296803200102 barajas, a., and rodriguez, p. (2010). spanish football clubs’ finances: crisis and player salaries. international journal of sport finance, 5(1), 52–66. caporale, t., and collier, t.c. (2015). are we getting better or are they getting worse? draft position, strength of schedule, and competitive balance in the national football league. journal of labor research, 36(3), 291–300. https://doi.org/10.1007/s12122-015-9206-z carmichael, f., mchale, i., and thomas, d. (2011). maintaining market position: team performance, revenue and wage expenditure in the english premier league, bulletin of economic research, 63(4), 464–497. https://doi.org/10.1111/j.1467-8586.2009.00340.x caro, c. a., and elder, a. (2017). the relationship between spending and winning in college baseball: is this the new ‘arms race’ in college athletics? international journal of sports science and coaching, 12(3), 381–397. https://doi.org/10.1177/1747954117710504 coates, d., and humphreys, b. r. (2012). game attendance and outcome uncertainty in the national hockey league. journal of sports economics, 13(4), 364–377. https://doi.org/10.1177/1527002512450260 deloitte arff (2000-2019). annual review of football finance. deloitte’s sports business group. https://www2.deloitte.com/uk/en/pages/sports-business-group/articles/annual-reviewof-football-finance.html deloitte dfml (1997-2019). football money league. deloitte’s sports business group. https://www2.deloitte.com/uk/en/pages/sports-business-group/articles/deloitte-footballmoney-league.html dietl, h., franck, e., and lang, m. (2008). overinvestment in team sports leagues: a contest theory model, scottish journal of political economy, 55, 353–368. https://doi.org/10.1111/j.1467-9485.2008.00457.x dietl, h. m., lang, m., and rathke, a. (2009). the effect of salary caps in professional team sports on social welfare’, the b.e. journal of economic analysis and policy, 9(1), 129–151. https://doi.org/10.2202/1935-1682.2034 dietl, h. m., grossmann, m., and lang, m. (2011). competitive balance and revenue sharing in sports leagues with utility-maximizing teams. journal of sports economics, 12(3), 284– 308. https://doi.org/10.1177/1527002511404787 https://doi.org/10.1108/jepp-03-2019-103 https://doi.org/10.1177/002224296803200102 https://doi.org/10.1007/s12122-015-9206-z https://doi.org/10.1111/j.1467-8586.2009.00340.x https://doi.org/10.1177/1747954117710504 https://doi.org/10.1177/1527002512450260 https://doi.org/10.1111/j.1467-9485.2008.00457.x https://doi.org/10.2202/1935-1682.2034 https://doi.org/10.1177/1527002511404787 garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 140 edwards, h. (1984). the collegiate athletic arms race: origins and implications of the “rule 48” controversy. journal of sport and social issues, 8(1), 4–22. https://doi.org/10.1177/019372358400800103 forrest, d., and simmons, r. (2002). team salaries and playing success in sports: a comparative perspective. zeitschrift für betriebswirtschaft, 72(4), 221–238. https://doi.org/10.1007/978-3-663-07711-4_12 franck, e. (2014). financial fair play in european club football what is it all about? university of zurich, department of business administration, uzh business working paper no. 328. https://doi.org/10.2139/ssrn.2284615 gallagher, r., and quinn, b. (2019). regulatory own goals: the unintended consequences of economic regulation in professional football. european sport management quarterly. online: 3 april 2019. https://doi.org/10.1080/16184742.2019.1588344 garcía, j., and rodríguez, p. (2002). the determinants of football match attendance revisited: empirical evidence from the spanish football league. journal of sports economics, 3(1), 18– 38. https://doi.org/10.1177/152700250200300103 garcia-del-barrio, p. (2018). media value in global sports: football versus formula one, international journal of sport management and marketing, 18(3), 241–266. https://doi.org/10.1504/ijsmm.2018.10012851 garcia-del-barrio, p. and pujol, f. (2020). recruiting talent in a global sports market: appraisals of soccer players’ transfer fees. managerial finance, forthcoming. https://doi.org/10.1108/mf-04-2020-0213 garcia-del-barrio, p., and szymanski, s. (2009). goal! profit maximization and win maximization in football leagues, review of industrial organization, 34(1): 45–68. https://doi.org/10.1007/s11151-009-9203-6 garcia-del-barrio, p., and tena-horrillo, j. d. (2019). investment in talent and visibility in the media: a study of professional football in europe. in jaume garcía (ed.), sports (and) economics (pp. 271–294). funcas social and economic studies, 7. madrid, spain. ghio, a.; ruberti, m., and verona, r. (2019). financial constraints on sport organizations' cost efficiency: the impact of financial fair play on italian soccer clubs. applied economics, 51(24), 2623–2638. https://doi.org/10.1080/00036846.2018.1558348 hausman, j. (1978). specification tests in econometrics. econometrica, 46, 1251–1271. https://doi.org/10.2307/1913827 heiskanen, a. (2017). uefa financial fair play’s effect on the salary-to-revenue ratios of european professional football clubs – evidence from top-5 football leagues in europe. https://aaltodoc.aalto.fi/handle/123456789/26796 hoffer, a., humphreys, b. r., lacombe, d. j., and ruseski, j. e. (2015). trends in ncaa athletic spending: arms race or rising tide? journal of sports economics, 16(6), 576–596. https://doi.org/10.1177/1527002515592541 lazear, e.p., and rosen, s. (1981). rank-order tournaments as optimum labor contracts. journal of political economy, 89(5), 841–864. https://doi.org/10.1086/261010 litan r., orszag j. and orszag p. (2003). the empirical effects of collegiate athletics: an interim report. (commissioned by the national collegiate athletic association). retrieved from: https://www.ncaa.org/sites/default/files/empirical_effects_of_collegiate_athletics_interim_rep ort.pdf madden, p. (2015). welfare economics of “financial fair play” in a sports league with benefactor owners. journal of sports economics, 16(2), 159–184. https://doi.org/10.1177/1527002512465759 neale, p. (1964). the peculiar economics of professional sports: a contribution to the theory of the firm in sporting competition and in market competition. quarterly journal of economics, 78(1), 1–14. https://doi.org/10.2307/1880543 orszag, j. m., and orszag, p. r. (2005). the physical capital stock used in collegiate athletics. indianapolis, in: national collegiate athletic association. orszag, j. m., and israel, m. (2009). the empirical effects of collegiate athletics: an update based on 2004-2007 data. indianapolis, in: national collegiate athletic association. https://doi.org/10.1177/019372358400800103 https://doi.org/10.1007/978-3-663-07711-4_12 https://doi.org/10.2139/ssrn.2284615 https://doi.org/10.1080/16184742.2019.1588344 https://doi.org/10.1177/152700250200300103 https://doi.org/10.1504/ijsmm.2018.10012851 https://doi.org/10.1080/00036846.2018.1558348 https://doi.org/10.2307/1913827 https://aaltodoc.aalto.fi/handle/123456789/26796 https://doi.org/10.1177/1527002515592541 https://doi.org/10.1086/261010 https://www.ncaa.org/sites/default/files/empirical_effects_of_collegiate_athletics_interim_report.pdf https://www.ncaa.org/sites/default/files/empirical_effects_of_collegiate_athletics_interim_report.pdf https://doi.org/10.1177/1527002512465759 https://doi.org/10.2307/1880543 garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 141 owen, d. (2014). measurement of competitive balance and uncertainty of outcome. chapter 3 in: handbook on the economics of professional football (ed.: goddard, j.; sloane, p. j.). edward elgar. cheltenham, uk norhampton, ma, usa. pawlowski, t., and anders, c. (2012). stadium attendance in german professional football–the (un)importance of uncertainty of outcome reconsidered. applied economic letters, 19, 1553–56. https://doi.org/10.1080/13504851.2011.639725 peeters, t., and szymanski, s. (2014). financial fair play in european football. economic policy, 29(78), 343–390. https://doi.org/10.1111/1468-0327.12031 rohde, m., and breuer, c. (2016). europe’s elite football: financial growth, sporting success, transfer investment, and private majority investors. international journal of financial studies, 4(2), 12. https://doi.org/10.3390/ijfs4020012 rottenberg, s. (1956). the baseball players labor market. journal of political economy, 64(3), 242–58. https://doi.org/10.1086/257790 sanderson, a. r. (2002). the many dimensions of competitive balance. journal of sports economics, 3(2), 204–228. https://doi.org/10.1177/152700250200300206 szymanski, s., and smith, r. (1997). the english football industry: profit, performance and industrial structure. international review of applied economics, 11(1), 135–153. https://doi.org/10.1080/02692179700000008 szymanski, s. (2006). the future of football in europe. in placido rodriguez, stefan kesenne, and jaume garcia (eds.), sports economics after fifty years: essays in honour of simon rottenberg, oviedo, spain: university of oviedo. torgler, b., and schmidt, s. l. (2007). what shapes player performance in soccer? empirical findings from a panel analysis. applied economics, 39(18), 2355–2369. https://doi.org/10.1080/00036840600660739 tsitsos, w., and nixon ii, h.l. (2012). the star wars arms race in college athletics: coaches’ pay and athletic program status. journal of sport and social issues, 36(1), 68–88. https://doi.org/10.1177/0193723511433867 tullock, g. (1980). efficient rent-seeking; in j.m. buchanan, r.d. tollison and g. tullock (eds.): toward a theory of rent-seeking society. college station: texas a.andm. university press. uefa (2010). uefa club licensing and financial fair play regulations, lyon. vöpel, h. (2011). do we really need financial fair play in european club football? an economic analysis, cesifo dice report. https://doi.org/10.1080/13504851.2011.639725 https://doi.org/10.1111/1468-0327.12031 https://doi.org/10.3390/ijfs4020012 https://doi.org/10.1086/257790 https://doi.org/10.1177/152700250200300206 https://doi.org/10.1080/02692179700000008 https://doi.org/10.1080/00036840600660739 https://doi.org/10.1177/0193723511433867 garcía-del-barrio and rossi / european journal of government and economics 9(2), july 2020, 119-142 142 annex ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5842_v2_g european journal of government and economics 11(2), december 2022, 167-192 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 effect of military spending on private investment in nigeria: does a crowding-out effect exist? isiaka akande raifu a * a department of economics, faculty of economics and management sciences, university of ibadan, ibadan, oyo state, nigeria * corresponding author at: heritagetiamiyu@gmail.com abstract. this study adopts ardl and var estimation methods to examine whether military spending crowd-out or crowd-in private investment in nigeria. we use the data that covers the period from 1970 to 2019. our results, based on the ardl method, show that military spending only crowds-out private investment in the short run. in the long run, military spending crowds in private investment. the results are robust to the use of alternative estimation methods. specifically, irf results show that military spending has a contemporaneous negative effect on private investment. however, the negative effect turns positive after the third period. also, fevd results show that most of the variation in private investment is explained by its shock and few by military spending. our findings have policy implications. while it is advisable to spend more on the military to curtail the activities of insurgents, bandits and kidnappers and to restore confidence in investors, it is important also to take cognisance of the fact that military spending can crowd out private investment. keywords. military spending, private investment, crowding-out effect. jel codes. h56; c23 doi. https://doi.org/10.17979/ejge.2022.11.2.8758 1. introduction the role of government spending in engendering economic growth has received both commendation and condemnation in the economic literature. from a commendation perspective, it is believed that government provides some vital services that aid and sustain the economy. such services include the provision of public goods such as national defense, the rule of law and law enforcement, among others. these goods are not provided by individual citizens or private organisations in most countries because there are no incentives to produce such goods (wagner, 2007). thus, the duty of the government is to ensure that the lives and properties of its citizens and foreigners living within its territory are protected against internal and external invaders. apart from this, when the economy is in crisis, especially during an economic recession, the government is usually called upon to spend more to bail the economy out of recession. government spending during this period, as argued, would raise effective demand, boost production and investment and https://creativecommons.org/licenses/by-nc/4.0/ mailto:heritagetiamiyu@gmail.com https://doi.org/10.17979/ejge.2022.11.2.8758 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 168 spur economic growth through the multiplier effect (keynes, 2018). from a condemnation perspective, however, it is believed that government spending, in most cases, is financed by taxes paid by individual citizens or private organisations. such tax-financing government expenditure increases the tax burden on citizens and private organisations and thereby leads to a reduction in citizens’ spending or private organisations’ spending and investment, which would eventually lead to the “crowding-out” of private investment (stratmann and okolski, 2010). apart from health and education, the government also spends on the military for the procurement of arms and ammunition to provide security against external aggressions. besides the procurement of military arsenals, the government also spends on the military to take care of military personnel welfare, which includes the regular payment of salary and other benefits. while some defense economists believe that military spending can be beneficial to investment, economic growth and employment, others submit that such an increase in military spending has opportunity costs. those who are against the increase in military spending argue that, in the presence of scarce resources, the allocation of more funds to the military implies depriving other areas of the economy. smith (1980) submits that increase in military spending affects the public components of consumer consumption expenditure which is maintained through political pressure. besides, he further argues that military gadgets are produced by highly capitalised industries and thus, an increase in military spending has a direct effect on private investment with an inelastic capacity. hence, military spending and private investment usually compete for a fixed proportion of public resources. in this case, military spending does crowd-out private investment in the long run (scott, 2001; hou and chen, 2014). apart from the arguments for and against a perpetual increase in military spending, there is growing concern among citizens across the world, especially in advanced countries, about the perpetual increase in military spending (sajid, 2021). many of the citizens are openly expressing their dissatisfaction through different media, such as protests, jingles and other means. in the united states, for instance, sajid (2021) noted that a great proportion of the budget is devoted to military spending on an annual basis. in 2020, out of a total of $4.7 trillion budget, about $0.73 trillion went to defense spending, suggesting that other areas of the economy, such as education and health, are likely to be affected. the ultimate consequence of the growing military spending in the country is mounting public debt and budget deficit on an annual basis, which in turn, could have a detrimental effect on the overall economy (caruso and di, domizio, 2017; sajid, 2021). figure 1 shows the evolution of military spending from 1970 to 2019 in nigeria. it is evident from the figure that there was a significant rise in military spending after the civil war of 19671970. the increase in military spending then was to fortify the military, train the newly recruited military officers, build military barracks and training institutions, procure more ammunition and take care of the general welfare of the military officers. thereafter, military spending declined considerably, even during periods of military regimes. however, in recent times, there has been an upsurge in military spending aimed at procuring arms and ammunition to fight against insecurity, emanating majorly from the terrorism activities of boko haram and the islamic state of isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 169 west african province (iswap).1 the terrorist activities of boko haram and iswap have not only led to the loss of lives, loss of properties and displacement of people but also the loss of investment, especially in the northern region. according to the global conflict tracker (2021), more than 350,000 people have lost their lives, more than 3 million people have been displaced from their permanent habitations, and nearly 310,000 people have become refugees.2 besides, farming in the north, especially in the northeast, has been adversely affected, resulting in low agricultural productivity (adelaja and george, 2019). moreover, cross-border trading or business has been drastically reduced between the north and its neighbouring countries such as niger (kimenyi, et al. 2014). in fact, okereocha (2012) submits that the country has lost about n1.33 trillion in foreign direct investment. in light of this, the primary objective of this study is to examine whether or not an increase in military spending crowds out private investment in nigeria. researchers have been keenly interested in studying the relationship between military spending and private investment to detect a crowding-out effect. the empirical findings are diverse depending on a group of countries or a country studied, the models adopted, the estimation techniques used and the measures of military spending used. besides, most of the extant studies focused on developed economies such as oecd member countries and others (smith, 1980-oecd; smith and dunne, 2001; hou and chen, 2014; dunne and smith, 2020; oecd; gold, 1997; atesoglu, 2004; perioni, 2009-usa; scott, 2001-uk, smith and dunne, 2001; malizard, 2015-france; üçler, 2016-turkey; kennedy, 2021-indonesia). studies that examine how military spending affects investment in nigeria are relatively scarce. however, there are a couple of studies that examined the effect of military spending on foreign direct investment but not on domestic private investment (adediran et al., 2018; edith et al., 2019). it is, however, known that foreign investors, in most cases, bring their capital to invest in many developing countries like nigeria. most of them hardly depend on the domestic capital market to finance their investments. hence, the issue of the government competing with them for funds in the financial market does not arise. thus, it cannot be said that military spending crowds out foreign direct investment. the same cannot be said of domestic private investors who, in most cases, seek funds in the domestic financial market. to explore the impact of military spending on private investment, we set out to achieve two objectives. the first objective is to examine whether the crowding out of military spending occurs in the short run or the long run. this is important for policy decision-making. assuming that the crowding-out effect occurs in the short run, it would inform the government that its spending, including military spending, does not have a permanent effect on private investors. however, if it occurs in the long run, this may call for drastic reforms in the way the government gets involved in the financial market. to achieve this objective, we employ a novel autoregressive distributed lag (ardl) estimation method developed by pesaran et al. (2001). the method can be used to distinguish the effect of military spending on private investment in the short run from the long run. 1 the issue of insecurity has become complicated as there are currently rises in the activities of bandits, kidnappers, herders-farmers conflicts and the host of others. it is also observed that military spending is subject occasional fluctuations due to the instability of crude oil prices (raifu and raheem, 2018; aminu and raifu, 2019, raifu and aminu, 2020, raifu, aminu and folawewo, 2020; raifu, 2021) 2 https://www.cfr.org/global-conflict-tracker/conflict/boko-haram-nigeria https://www.cfr.org/global-conflict-tracker/conflict/boko-haram-nigeria isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 170 although error correction model (ecm) can also be used to achieve the same objective, however, the method has some limitations. one of the limitations of using ecm estimation is that they are only applicable when the variables are integrated of order 1, a condition that may not be met in practice. in practice, some variables could be integrated of order 0 or order 1 or a mixture of both. ardl estimation method becomes useful when the variables display these characteristics. it must, however, be stated that ardl can be rendered unusable if any of the variables are integrated of higher order, especially the integrated order of 2. the second objective is to examine the dynamic effect of military spending on private investment over time. in this case, we employ sims (1980)’s vector autoregressive method and compute the necessary impulse response function (irf) as well as forecast error variance decomposition (fevd). with this approach, it is possible to trace the dynamic effect of military spending on private investment over time. the rest of the study is structured as follows. section 2 reviews the existing studies. section 3 presents the theoretical framework, estimation model as well data sources. the results are presented in section 4. section 5 concludes with policy implications. 2. literature review most of the studies on the effect of military spending on the economy are concentrated on how military spending affects economic growth. beginning with benoit (1978), who concluded that military spending is positively correlated with economic growth in developing countries, a huge number of studies had been conducted in different countries and groups of countries to thoroughly examine the effect of military spending on economic growth. however, a concrete consensus as regards the real effect of military spending on economic growth has not been reached. dunne and smith (2013) summarised the rationales for this development as follows: different theoretical considerations and methodological approaches, different countries considered, different data and periods employed (see also alptekin and levine, 2012). consequently, some studies documented the positive effect of military spending on economic growth (saba and ngepah, 2019), while others established a negative effect of military spending on economic growth (shahbaz, afza and shabbir, 2013). there are some strands of studies that did not establish any cogent nexus between military spending and economic growth. (minzt and huang, 1990). as regards the effect of military spending on investment, table 1 summarises the empirical findings from the existing studies. like the studies on the nexus between military spending and economic growth, the relationship between military spending and investment remains inconclusive. from the table, three empirical findings could be deduced irrespective of different estimation techniques deployed, country or countries studied as well as periods considered. the crowding-out effect of private investment by military spending, however, dominates the literature (smith, 1980; dreger, 1986; knight, et al., 1996; scott, 2001; pieroni 2009; hou and chen, 2014; malizard, 2015; lorusso and pieroni, 2017; kennedy, 2021). this notwithstanding, a handful of studies concluded that military spending crowded-in investment (üçler, 2016; kollias and isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 171 paleologou, 2019). the third strand of studies did not find any evidence of a relationship between military spending and investment (smith and dunne, 2001; morales-ramos, 2020; atesoglu, 2004; dunne and smith, 2020). the relationship between military spending and investment could still be complementary. this is established by malizard (2015) who investigated whether military spending crowded-out investment in france using disaggregated data. the evidence of complementarity comes from the relationship between public investment and private investment. according to him, the reason for this phenomenon could be adduced to the fact that the military equipment sector in france is highly capitalistic and, as such, defense in r & d may contribute to the private sector productivity. the way military spending affects private investment could depend on the level of economic development of a country. kollias and paleologou (2019), who investigated the effect of military spending on economic growth and investment, concluded that military spending has a positive and significant effect on investment in high-income countries whereas its effect on low-income and middle-income countries is negative. this outcome for the low-income and middle-income countries is rationalised by the fact that most of the countries belonging to these groups could be facing the challenges of resource constraints. hence, devoting more resources to the military at the expense of other productive projects appears to be disastrous to private investment. the studies reviewed above are done for other countries. most of the studies on military spending in nigeria focused on the effect of military spending on the economy captured by gdp (apanisile and okunola, 2014; ajefu, 2015; temiyope and ajala, 2021; oji and afolabi, 2022). however, a couple of studies have examined the effect of military spending on foreign direct investment in nigeria (aderemi, et al. 2018; edith et al., 2010). some studies also examined the effect of the general government on private investment in nigeria without a specific reference to the influence of military spending on private investment (akinlo and oyeleke, 2018; olaifa and benjamin, 2020). given the paucity of studies on the nexus between military spending and private investment in nigeria, this study is conducted to fill this gap. isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 172 table 1. literature review: summary table. author topic sample/date/method findings smith (1980) military expenditure and investment in oecd countries, 1954-1973 14 oecd 1953–1974 ols negative effect (crowdout) dreger (1986) military expenditure in the third world countries: the economic effect 50 ldcs, 1965–1973 negative effect (crowdout) knight, loayza and villanueva (1996) the peace dividend: military spending cuts and economic growth 79 countries, 1971–1985 fixed effect negative effect (crowdout) gold 1997 evaluating the trade‐off between military spending and investment in the united states usa 1949-1988 ecm short-run negative effect (crowd-out) long-run (no crowd-out) smith and dunne (2001) military expenditure growth and investment 28 oecd countries 1960-1997 ols, fixed effects, random coefficient model and var neither crowd-out not crowd-in scott (2001) does uk defence spending crowd‐out uk private sector investment? uk 1974-96 ols negative effect (crowdout) atesoglu, (2004) defense spending and investment in the united states usa 1947q1–2001q3 johansen cointegration no crowd-out pieroni (2009) does defence expenditure affect private consumption? evidence from the united states usa 1957-2005 vecm negative effect (crowdout) hou and chen (2014) military expenditure and investment in oecd countries: revisited 13 oecd countries 1971–2012 ols, pooled ls, two-way fixed effects and random effect negative effect (crowdout). however, it fizzles out over time, especially after the cold war malizard (2015) does military expenditure crowd out private investment? a disaggregated perspective for the case of france france 1980-2010 dols, fmols and ccr military equipment and private investment are complementary üçler (2016) testing the relationship between military spending and private investments: evidence from turkey turkey 1975-2014 maki’s cointegration, dols and hatemi-j causality test positive effect (crowd-in) lorusso and pieroni (2017) the effects of military and non-military government expenditures on private consumption usa 1960–2013 svar and dsge civilian spending-positive effect (crowd-in) military spending– negative effect (crowd-out) kollias and paleologou (2019) military spending, economic growth and investment: a disaggregated analysis by income group. 65 countries 1971–2014 pvar high-income group positive effect (crowd-in) low-income and middleincome countries-negative (crowd-out) dunne and smith (2020) military expenditure, investment and growth 17 oecd countries pols, fixed effect, pooled mean group neither crowd-out nor crowd-in morales-ramos (2002) defence r&d expenditure: the crowding-out hypothesis the uk1966–1996. ols, 2sls. france, germany, uk, usa and japan, for the period from 1971-1996 no crowding-out-uk no crowding out for others. however, the indirect crowding-out effect occurs through saving, kennedy (2021) the effect of defense spending on private investment in indonesia based on historical data for the period 1981-2010 pakistan 1981-2010 ols negative effect (crowd-out) isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 173 3. theoretical consideration, model specification and data sources 3.1 theoretical consideration and model specification there are three theoretical models (supply model, demand model and supply-demand model) used to explore the crowd-out effect of military spending (see morales-ramos, 2002). in this study, we adopt a demand model. we chose the demand model because we can easily get data for its estimation. the demand model rests on the keynesian theory, which considers military spending as part of aggregate demand. following hou and chen (2014), the demand model begins with the national accounting identity specified as follows: y q w c i m b= − = + + + , [1] where y denotes actual production in the economy, q is the potential output, w refers to the gap between actual output and potential output. , ,c i m and b are aggregate consumption, investment (could be private or public investment), military spending and balance of trade, respectively. equation 1 can be re-specified if we express it as a share of potential output. thus, equation 1 becomes 1i w c m b= − − − − [2] according to smith (1980), the share of consumption is 1 2oc u gα α α= − − [3] here u is the unemployment rate, g is the actual output growth rate. it is believed that when the shares of unemployment and actual output in potential output increase, the share of consumption in it should decline. when this happens, equations 2 and 3 become: 0 1 2(1 ) ( )i u g m w bα α α= − + + − − + [4] assume that ( )w b+ is related unemployment rate as follows: ( )w b uβ+ = . then equation 4 can be rearranged as: 0 1 2(1 ) ( )i u g mα β α α= − − − + − [5] in many empirical studies that employ the demand model to investigate the crowding-out effect of military spending on investment, equation 5 is often adopted. equation 5 can be formalised as an econometric model as follows: isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 174 0 1 2 3t t t ti u g mα α α α ε= + + + + [6] on a priori ground, if the coefficient of military spending is negative, that is, 3α is negative, it means that military spending crowds out private investment. following raifu and afolabi (2022) and raifu, obijole and nnadozie (2022), equation 6 is formalised into the ardl framework as follows: 1 0 1 1 2 1 3 1 4 1 1 1 2 1 1 0 2 3 3 1 4 1 0 0 pl t t t t t t t i i p p t t t i i i i u g m i u g m α α α α α φ φ φ φ ε − − − − − − = = − − = = ∆ = + + + + + ∆ + ∆ + ∆ + ∆ + ∑ ∑ ∑ ∑ [7] in equation 7, ∆ is the first difference operator, 0α is a constant and drift component of the model, 1α to 4α are the long-run coefficient parameters which show the effects of lags of investment, unemployment, actual output growth rate and military spending investment. 1ϕ to 4ϕ are the short-run coefficient parameters showing the effects of the aforementioned independent variables on investment. we test the null hypothesis of the long run, which states that there is no long-run relationship among the variables 0 1 2 3 4( : 0)h α α α α= = = = , against the alternative hypothesis, which stipulates that there is a long-run relationship among them 0 1 2 3 4( : 0)h α α α α≠ ≠ ≠ ≠ . the error correction model, which shows the speed of adjustment from the short-run disequilibrium towards the long-run equilibrium, is specified as follows. 1 2 3 0 1 1 2 1 3 1 4 1 1 1 0 0 0 p p pl t t t t t t t i i i i i i u g m ectα φ φ φ φ λ ε− − − − − = = = = ∆ = + ∆ + ∆ + ∆ + ∆ + +∑ ∑ ∑ ∑ , [8] where ect is the error correction term, it is expected that λ the coefficient of error term must be negative, less than 1 and statistically significant to claim that there is an adjustment from the short-run disequilibrium towards the long-run equilibrium. for the second objective, we use vector autoregressive (var) method. the basic principle of var is that all variables are treated endogenously and expressed as the lags of one another, that is, the lag of dependent variables and independent variables for each variable. our var model consists of four variables which include military spending, private investment, real gdp and unemployment. the general var framework is specified as follows: 0 1 1 k t i t t t x xβ β ε− = = + +∑ , [9] isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 175 where tx = (private investment, military spending, real gdp, unemployment), tε is the error term, 0β denotes the identity matrix and 1 kβ β− are four by four matrices of the coefficients. we compute impulse response which shows the response of private investment to innovative shocks to military spending as well as other variables included in the model. the orthogonalised irs is computed using a cholesky decomposition of the variance-covariance matrix of residuals following shan (2002). thereafter, the fevd, which depicts the proportion of private investment that can be explained by its shock and shock to military spending is also estimated. 3.2 data sources and preliminary findings the data used for analysis are sourced from different databases. military spending, either in dollar value or as a percentage of gdp, is obtained from the stockholm international peace research institute. private investment is gathered from the international monetary fund. real gdp is extracted from the world development indicators and the unemployment rate is sourced from the national bureau of statistics. the data covers the period from 1970 to 2019. figures 1 to 4 show the trend of these variables over the period under consideration. specifically, figure 1 shows the evolution of military spending over the period. it is observed that nigeria spent more on the military around the 1970s as the period saw military spending rose significantly. this could be attributed to the rebuilding of the military in terms of training of newly recruited soldiers, building of military barracks and training schools and procurement of arms and ammunition after the civil wars. however, spending on the military declined throughout the 1980s and 1990s. in recent times, due to internal crises such as the rising rate of insecurity engineered by insurgent activities such as boko haram, iswap, bandits, kidnappers and gunmen, the government has increased military spending. figure 2 depicts the trend of private investment. as shown in the figure, private investment soared around the 1970s to the early 80s. this could be attributed to the indigenisation policy of the government around the period. however, from the middle of the 1980s, private investment declined throughout the 1990s until 2001, when it rose again. this could be attributed to the return of the country to a democratic system of government which restores confidence in investors due to some economic reforms embarked upon by the new civilian government. figures 3 and 4 are the trend of real gdp and unemployment rate, respectively. table 2 presents the results of descriptive statistics. the mean value of private investment stood at $42.83 million. the average military spending in dollar value stood at $1.962 billion. this represents 1.77% of gross domestic product. the mean value of unemployment stood at 11.35% while gdp’s mean value stood at n32, 553.93 billion. in table 3, the results of correlation analysis among the variables are presented. it is obvious from the table that military spending in dollar value is positively and significantly correlated with private investment. however, military sending as a percentage of gdp is negatively correlated with private investment. the negative correlation is insignificant. real gdp and unemployment are positively and significantly correlated with private investment. isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 176 we also conduct a unit root test to determine the stationary properties of our variables of interest. conducting a unit root test is necessary to avoid running spurious regression. we use augmented dickey-fuller (adf), phillips-perron (pp) and (kpss) unit root test methods. while adf and pp assume that variables contain a unit root, kpss assumes that variables are stationary. the results are presented in table 4. evidence from the table shows that all the variables are integrated of order 1, that is, they contain unit root and they become stationary after the first difference. figure 1. military spending in million usd (blue line, left axis) and as a percentage of gdp (orange line, right axis). figure 2. private investment (b’usd). 0,000 1,000 2,000 3,000 4,000 5,000 6,000 7,000 8,000 9,000 0 1000 2000 3000 4000 5000 6000 7000 8000 0 20 40 60 80 100 120 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 177 figure 3. real gdp. figure 4. unemployment rate. table 2. descriptive statistics. variables obs mean s.d. min max p1 p99 skew. kurt. pi 50 42.829 32.861 1.557 108.273 1.557 108.273 0.901 2.479 mil 50 1962.5 1504.737 415.000 7314 415.000 7314.000 1.513 5.139 mil_gdp 50 1.768 2.013 0.348 8.124 0.348 8.124 1.475 3.947 real gdp 50 32553.93 18606.42 14306.1 72094.1 14306.1 72094.1 1.054 2.574 unemp 50 11.354 11.562 1.600 48.820 1.600 48.820 1.524 4.654 note. pi, mil, mil_gdp, real gdp and unemp are private investment, military spending in dollar value, military spending as a percentage of gdp, real gross domestic product and unemployment rate respectively. 0 10000 20000 30000 40000 50000 60000 70000 80000 0,00 10,00 20,00 30,00 40,00 50,00 60,00 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 178 table 3. pairwise correlations. variables pi mil mil_gdp rgdp unemp pi 1 mil 0.357* 1 mil_gdp -0.156 0.656* 1 rgdp 0.646* 0.104 -0.660* 1 unemp 0.433* 0.064 -0.612* 0.871* 1 * shows significance at the 0.05 level note. pi, mil, mil_gdp, real gdp and unemp are private investment, military spending in dollar value, military spending as a percentage of gdp, real gross domestic product and unemployment rate respectively table 4. unit root test result. level first difference decision wc wc&t wdc&t wc wc&t wdc&t augmented dickey-fuller unit root test pi -1.720 -2.032 0.079 -6.610*** -6.563*** -6.649*** i(1) mil -1.470 -1.207 -0.508 -7.436*** -7.511*** -7.505*** i(1) mil_gdp -1.506 -1.750 -1.577 -8.636*** -8.658*** -8.377*** i(1) rgdp 0.653 -1.368 2.428 -2.266 -2.472 -2.654*** i(1) unemp -0.566 -2.298 0.785 -3.200** -3.294* 3.038*** i(1) phillips-perron unit root test pi -1.768 -2.085 0.381 -6.952*** -7.220*** -6.875*** i(1) mil -1.605 -1.344 -0.481 -7.475*** -7.522*** -7.533*** i(1) mil_gdp -1.478 -1.663 -1.563 -8.566*** -8.676*** -8.265*** i(1) rgdp 0.222 -1.214 2.930 -5.387*** -5.511*** -4.610*** i(1) unemp -0.484 -2.345 1.009 -7.466*** -7.580*** -7.377*** i(1) kpss unit root test pi 0.324 0.174** 0.137 0.109 i(1) mil 0.245 0.202** 0.210 0.084 i(1) mil_gdp 0.785*** 0.201** 0.159 0.058 i(1) rgdp 0.831*** 0.211** 0.189 0.092 i(1) unemp 0.768*** 0.162** 0.129 0.041 i(1) note. pi, mil, mil_gdp, real gdp and unemp are private investment, military spending in dollar value, military spending as a percentage of gdp, real gross domestic product and unemployment rate respectively wc, wc&t and wdc&t denote unit root test with constant, with constant and trend and without constant and trend respectively. *, ** and *** denote 10%, 5% and 1% level of significance respectively. 4. empirical results 4.1 ardl results table 6 reports the results of the ardl estimation method, which shows the effect of military spending and other control variables on private investment in the short run and the long run. however, before presenting the main results, which would show whether military spending crowds out or crowds in private investment, it is important to present the results of the ardl bounds testing to discover whether there is cointegration among the variables in the models (military isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 179 spending, investment, economic growth and unemployment). the results of the ardl bounds testing are reported in table 5. the upper part is the result of bounds testing for model 1 – baseline model (model of military spending as a percentage of gdp (military burden) and private investment including control variables (gdp and unemployment)). the lower part is the result of bounds testing for model 2 robustness model (military spending (dollar value) and private investment including control variables). to determine the existence of cointegration, pesaran, et al (2001) provided lower and upper criteria based on f-statistic. if the computed f-statistic value falls below the lower bound f-statistic criteria, there is no cointegration among the variables. on the other hand, if the computed value of the f-statistic falls above the upper bound f-statistic criteria, there is cointegration. however, no decision would be made if the computed value of the f-statistic falls between the lower and upper bounds f-stat criteria. our results, as shown in the table, reveal that the computed values of f-statistic from the two models we estimated fall above the upper bounds of f-statistic criteria, signifying the existence of cointegration among our variables of interest. this implies that there is a long-run relationship among the variables. in the same, the results of the error correction model reported in table 6 show how fast the economy returns to equilibrium when it is temporarily destabilised by either internal or exogenous shocks. to determine whether there would be adjustment in the long run, the coefficient of error correction term must be negatively signed, less than one and statistically significant. our results in the two models (model 1 and model 2) follow a priori expectations in the sense that they are negatively signed, less than one and statistically significant. precisely, the coefficients of ecm in model 1 and model 2 are -0.530 and -0.567, respectively. this implies that there is an adjustment towards the long-run equilibrium from the short-run disequilibrium. thus, we can conclude that about 50% of errors that occurred in the previous year can be corrected in the current year. as regards the crowding-out of military spending, our results show that in the short run, military spending as a percentage of gdp has a negative effect on private investment (model 1). however, the negative effect is statistically significant and occurs with lags, specifically when military spending is lagged for two periods. thus, when military spending as a ratio of gdp increases by 1%, private investment would decline by 0.566% in the short run. this suggests that military spending crowds out private investment in nigeria. this short-run result is similar to most empirical findings in table 1. for instance, smith (1980) submitted that an increase in military spending by 1% would crowd out private investment in oecd countries with a degree of crowding-out very close to 1%. in many other countries, similar findings have also been documented (see dreger, 1986 for ldcs, knight et al., 1996 for 79 countries, scott, 2001 for the uk, pedroni 2009 and lorusso and pedroni, 2019 for the us and kennedy, 2021 for pakistan). however, most of these studies did not distinguish whether the crowding-out effect takes place in the short run or the long run. however, hou and chen (2014) stated that the negative effect of military spending fizzles out over time. after the cold war, they observed that military spending did not have a negative effect on private investment in oecd countries. gold (1997) had previously concluded that the negative effect of military spending on private investment is a shortrun phenomenon as the trade-off between the two variables fizzles out over time. these findings isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 180 are consistent with our long-run results. in the long run, we discover that military spending (% of gdp) has a positive and significant effect on private investment. specifically, an increase in military spending leads to an increase in private investment by 0.548%. this means that military spending crowds-in private investment in the long run. some studies, especially in advanced countries, that documented a positive effect of military spending on private investment explained their findings based on the positive externality effect of the defense industry. specifically, it is argued that the defense industry can develop a new technology from their research and development activities which can quickly diffuse and affect other industries. this may not be a generalised explanation for all countries, especially for developing countries like nigeria, which imports military hardware from developed countries. thus, the crowding-in effect of military spending can be explained from another perspective. for instance, most countries bedevilled with insecurity problems, such as mena countries and nigeria, invested in arms and ammunition to wage war against insurgents to create a conducive environment for investment to thrive. the restoration of a secure environment brings about by the increase in military spending would restore confidence in investors, which in turn could boost investment in the long run. we conduct a series of robustness tests to check the reliability of our results. in table 6 model 2, we use military spending in dollar value to examine the effect of military spending on private investment. we find that our result remains unchanged as it is consistent with the result of model 1 both in the short-run and the long run. apart from this, we examine whether the crowding-in effect of military spending on private investment in the long run, would remain ditto if we apply other long-run estimation methods. consequently, we use other estimation methods, such as dynamic ordinary least squares (dols), fully modified ordinary least squares (fmols) and canonical cointegration regression (ccr). the results of this exercise are reported in table 7. it is evident from the table that military spending crowds-in private investment in the long run irrespective of estimation methods. for other control variables, we find that economic growth has a positive significant effect on private investment. this finding is inconsonant with economic rationality that the period of economic growth leads to a boom in private investment. however, unemployment has a negative insignificant effect on private investment, especially in the long run. similar findings were documented for france by malizard (2015), who found the accelerator effect of economic growth on private investment and the inverse or adverse effect of unemployment on private investment. post-estimation diagnostic test results are also presented in table 6. the diagnostic tests, which show the reliability of the ardl estimation technique adopted in this study include the jarque-bera normality test, breusch-godfrey serial correlation lm test, lm heteroscedasticity test, ramsey reset test and cusum and cusum square tests. most of the statistical tests must not be statistically significant to confirm that the ardl estimation method and the results generated are reliable. apart from the jarque-bera normality test, other tests such as breuschgodfrey serial correlation lm test, lm heteroscedasticity test and ramsey reset test show the ardl results are reliable. isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 181 table 5: bounds testing results. model 1 f-bounds test null hypothesis: no levels relationship value significant level i(0) i(1) asymptotic: n=1000 f-statistic 5.708 10% 2.37 3.2 k 3 5% 2.79 3.67 2.5% 3.15 4.08 1% 3.65 4.66 model 2 f-bounds test null hypothesis: no levels relationship value significant level i(0) i(1) asymptotic: n=1000 f-statistic 5.866 10% 2.37 3.2 k 3 5% 2.79 3.67 2.5% 3.15 4.08 1% 3.65 4.66 note. 1(0) is the lower bounds while i(1) is the upper bounds. f-statistic (at least at 5%) below lower bounds implies no cointegration, above upper bounds means cointegration and in-between the two bounds means indecision. table 6. ardl results: effect of military spending on private investment. variable model 1 model 2 ardl(3, 3, 0, 0) ardl(3, 3, 0, 0) short-run constant -76.661*** -57.833*** d(pi(-1)) 0.180 0.178 d(pi(-2)) 0.226* 0.255* d(mil_gdp) -0.098 d(mil_gdp(-1)) -0.150 d(mil_gdp(-2)) -0.566** d(mil) -0.186 d(mil(-1)) -0.168 d(mil(-2)) -0.621** ect(-1) -0.530*** -0.567*** long-run mil_gdp 0.548*** mil 0.465** rdgp 2.605*** 1.892*** unemp -0.375 -0.395 r-squared 0.8396 0.8394 adjusted r-squared 0.8006 0.8004 f-stat 21.517 (0.0000) 21.490 (0.0000) durbin-watson stat 1.990 1.9435 diagnostic test jaque-bera test 656.869 (0.000) 640.650 (0.0000) b-g serial corr. lm test 1.374 (0.2662) 0.981 (0.3850) arch lm het test 0.0684 (0.7948) 0.067 (0.7964) ramsey reset test 2.278 (0.1174) 2.685 (0.0620) cusum test stable stable cusum of squares test unstable unstable note. pi, mil, mil_gdp, real gdp and unemp are private investment, military spending in dollar value, military spending as a percentage of gdp, real gross domestic product and unemployment rate respectively *, ** and *** denote 10%, 5% and 1% level of significance respectively isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 182 table 7. robustness: long-run effect of military spending on private investment from the dols, fmols and ccr estimation methods. (dols) (fmols) (ccr) (dols) (fmols) (ccr) mil_gdp 0.387* 0.442** 0.431*** mil 0.371* 0.405** 0.420*** (0.216) (0.172) (0.155) (0.215) (0.172) (0.155) real gdp 2.357*** 2.774*** 2.510*** real gdp 1.803** 2.149*** 1.909*** (0.841) (0.531) (0.483) (0.803) (0.499) (0.450) unemp -0.294 -0.472* -0.288 unemp -0.289 -0.487* -0.315 (0.453) (0.263) (0.244) (0.463) (0.256) (0.236) con. -68.781*** -81.313*** -73.454*** con. -54.296** -64.903*** -57.842*** (25.352) (16.040) (14.601) (24.006) (14.899) (13.460) obs. 47 49 49 obs 47 49 49 r-sq 0.713 0.620 0.577 r-sq 0.710 0.606 0.568 standard errors are in parenthesis. *, ** and *** denote 10%, 5% and 1% level of significance respectively note. pi, mil, mil_gdp, real gdp and unemp are private investment, military spending in dollar value, military spending as a percentage of gdp, real gross domestic product and unemployment rate respectively 4.2 var estimation results our second objective is to examine the dynamic effect of military spending on private investment. the optimal lag length selected is 4 using the akaike information criterion. we do not report the result of the var estimation. we only report the results of impulse response function (irf) and forecast error variance decomposition (fevd). the results of the impulse response function are reported in figures 5 and 6 for models 1 and 2. the irf results show that one standard deviation shock to military spending has a contemporaneous negative effect on private investment. this implies that military spending crowds out private investment when there is a sudden shock to military spending. however, we observe that the negative effect of military spending on private investment turns positive after three periods, that is, after three years. this suggests that military spending crowds-in private investment in the long run. as previously mentioned, the increase in military spending in recent times in nigeria is to tackle insecurity and thereby provide enabling environment for investment to thrive in the country. thus, the need for security is a possible explanation for the positive shock of military spending to investment in nigeria (kollias and paleologou 2019). this is contrary to the explanation given by kollias and paleologou (2019) for the high-income or industrialised countries that have military hardware manufacturing companies. the explanation of the persistent positive association between military spending and investment in high-income countries, according to kollias and paleologou (2019) is premised on the supplyside effects which are engineered by technological spillovers from the defense industry to the other sectors of the economy the results of fevd, which shows the proportion of variation in private investment explained by military spending shock, are reported in table 8 for models 1 and 2 (the graphs of fevd are put in the appendix). the results show that most of the variability in private investment is due to its shock. in specific terms, especially from the baseline model (model 1), 100% of the variation in private investment is due to its shock in period 1. however, it can be observed that the variation in private investment to its shock declines over time. in period 10, only about 36.55% variation in private investment is due to its shock. military spending account for a 0% variation in private isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 183 investment in period 1. however, as the period increases, the variation in private investment caused by the military spending shock increases. in period 10, military spending shock explains about a 14.06% variation in private investment, signifying that military spending has a positive effect on private investment over time. -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to private investment -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to real gdp -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to unemployment rate -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to military spennding (% of gdp) figure 5: response of private investment to military spending (% of gdp) and other variables isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 184 -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to private investment -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to real gdp -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to unemployment rate -.2 .0 .2 .4 1 2 3 4 5 6 7 8 9 10 response of private investment to military spennding figure 6: response of private investment to military spending and other variables table 8. forecast error variance decomposition. period standard error private investment real gdp unemployment rate military spending (% of gdp) 1 0.380776 100.0000 0.000000 0.000000 0.000000 2 0.469305 94.29579 1.219679 2.810787 1.673740 3 0.540770 77.19032 6.142631 10.66391 6.003140 4 0.568091 71.77813 8.175018 14.60530 5.441544 5 0.580322 69.93948 10.46599 14.20152 5.393001 6 0.608103 65.14490 11.32612 15.56029 7.968686 7 0.662306 55.26383 12.03657 23.59458 9.105021 8 0.733770 45.42605 12.44033 32.27131 9.862312 9 0.792531 39.93432 13.81394 37.13882 9.112920 10 0.830723 36.55274 15.16569 39.42891 8.852654 cholesky ordering: private investment, real gdp, unemployment rate, military spending (% of gdp) period standard error private investment real gdp unemployment rate military spending 1 0.378043 100.0000 0.000000 0.000000 0.000000 2 0.473183 96.95168 0.407319 0.564018 2.076981 3 0.564764 80.50399 4.396137 8.592811 6.507063 4 0.582110 75.79883 4.903520 13.15615 6.141505 5 0.590456 74.40356 6.228081 13.15127 6.217088 6 0.617982 68.97490 6.104384 14.21623 10.70448 7 0.672146 58.36043 7.284997 20.57896 13.77561 8 0.744521 47.71792 8.502604 28.60753 15.17195 9 0.799590 41.61173 11.38372 32.38208 14.62247 10 0.834403 38.22281 13.97637 33.73727 14.06355 cholesky ordering: private investment, real gdp, unemployment rate and military spending isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 185 4.3 discussion the controversies surrounding the crowding-out effect of military spending depend on many factors such as whether a country is an exporter or importer of military hardware, the stage of development of the country (developed countries or developing countries), how the government raises funds to finance the military spending and allocation of available resources between the military spending and other government spending such as health, education and infrastructure (morales-ramos, 2002; malizard, 2015; atanassovy and nanda 2018). with regard to military hardware exporting countries such as the united states, malizard (2015) argued that an increase in military spending can have a spillover effect on the private sector (investment) through technological progress, which can stimulate growth and investment. however, the same cannot be said about the military hardware importing countries, mainly developing countries. the necessity for ensuring the security of the environment, as submitted by kollias and paleologou (2019), could engineer or justify the increase in military spending in developing countries, especially country like nigeria, which is currently facing severe insecurity that is threatening the foundation of its existence and the economy. the increase in military spending can then be used to procure more arms and ammunition to secure society and provide the enabling environment conducive enough for private investment and economic growth. in light of this, we investigate the crowing-out effect of military spending in nigeria with two objectives in mind. first, we examine whether or not the crowding-out effect of military spending is a short-run or long-run phenomenon or both. the first objective is achieved by using a novel ardl estimation method. second, we examine the response of private investment to an increase in military spending via the impulse response function (ipf) and forecast error variance decomposition (fevd) methods. our findings based on the first objective reveal that the crowding-out effect of military spending is a short-run phenomenon. in the long-run, military spending does crowd-in private investment in nigeria. apart from the fact that our finding is in tandem with submission from existing studies such as gold (1997) and hou and chen (2014), who argued that the negative effect of military spending could fizzle out in the long run, it also justifies the increase in military expenditure in recent time by the government of nigeria. as previously stated, the country is currently facing security challenges which appear to be threatening the foundation of its existence and future economic prosperity. to combat the insecurity coming from not only the terrorists (boko haram and iswap) but also bandits, kidnappers, robbers, hoodlums and many others who have become dangerous species to the lives and properties of law-abiding citizens, the increase in military spending becomes inevitable. many studies have documented the negative effect of insecurity and both private and foreign investment in nigeria (olubunmi, 2018; jelilov, ozden and briggs, 2018; oji and afolabi, 2022; yusuf and mohd, 2022). yusuf and mohd (2022) specifically stated that domestic gross capital formation (domestic investment), unemployment rate, foreign direct investment and government spending on education are negatively affected by the growing insecurity. this happens because insecurity drives away investors, who are always afraid to make investments in an unsecured isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 186 country. the resultant effect of heightened insecurity is a drastic decline in investment activity. such a decline in investment activity is detrimental to the economy. thus, the increase in military spending in nigeria is not a waste in the long run because it signals to the investors that government is serious about providing adequate security. this would, in turn, restore the confidence of the investors and enable them to further make more investments in the country. our findings based on the second objective corroborate the results based on the first objective. based on the ipf results, we discovered that shock to military spending only causes a temporary decline in private investment. the decline effect, as discovered, fizzle out over time, leading to the crowd-in effect of military spending in the long run. however, from fevd, most deviations in private investment are explained by the shock to private investment and such deviations decline over time. only about 14.06% of the deviation in private investment can be explained by the shock to military spending. this discovery solidifies the short-run effect of military spending on private investment in nigeria. while our study has shown the benefit of an increase in military spending on private investment, particularly in the long run, it, however, has its limitation because it does not cover the whole aspect of investment in the country, especially public investment (government investment. in the literature on military spending, it has been argued that an increase in military spending also crowd-out public investment because such an increase in military spending, in the face of resource or budget constraints, implies that other expenditures such as education, health and infrastructure have to decline or remain unchanged. hence, in the case of nigeria, future studies need to investigate the relationship between military spending and public investment. 5. conclusions and policy implications in this study, we investigated the possibility of a crowding-out effect of an increase in military spending on private investment in nigeria using the data that spans from 1970 to 2019. two main objectives were set, and they were achieved through the use of the ardl and var estimation methods. for the first objective, the ardl estimation method enables us to know whether the crowding-out effect of military spending occurs in the short run or the long run. in the case of the second objective, the var estimation method, through irf and fevd, helps to trace the response of private investment to military spending shock and the proportion of variation in private investment explained by military spending shock. we also performed some preliminary analyses, which included correlation analysis and unit root tests. our findings are summarised as follows. ardl results show that military spending crowds out private investment in the short run, whereas it crowds in private investment in the long run. this is supported by the results obtained from irf, which show that shock on military spending has an instantaneous negative effect on private investment. however, the effect of military spending on private investment turns positive after three periods. the fevd results reveal that most variation in private investment is explained by its shock, while military spending explains little variation in isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 187 private investment. given our results, we conclude that while it is important for the government to spend more on the military for the procurement of arms and ammunition to stem the tide of insecurity bedevilling the country from the activities of insurgents, bandits and kidnappers and to restore confidence in investors who are sceptical of investing in the country, it must, however, take cognisance of the fact that military spending can crowd-out private investment. references aderemi, t. a., olayemi, h. o., ebere, c. e., & adeniran, a. a. (2018). security spending and foreign direct investment inflows: evidence from the nigerian economy. iosr journal of humanities and social sciences (iosr-jhss), 23(11), 29-35. ajefu, j. (2015) impact of defence spending on economic growth in africa: the nigerian case. journal of developing areas, 49(4), 227-244. https://doi.org/10.1353/jda.2015.0146 akinlo, t., & oyeleke, o. j. (2018). effects of government expenditure on private investment in nigerian economy (1980-2016). emerging economy studies, 4(2), 147-156. https://doi.org/10.1177/2394901518795067 alptekin, a., & levine, p. (2012). military expenditure and economic growth: a metaanalysis. european journal of political economy 28(4), 636-650. https://doi.org/10.1016/j.ejpoleco.2012.07.002 aminu, a., & raifu, i. a. (2019). dynamic nexus between oil revenues and economic growth in nigeria. econ bull, 39(2), 1556-1570. apanisile, o. t., & okunlola, o. c. (2014). an empirical analysis of effects of military spending on economic growth in nigeria: a bounds testing approach to co-integration 1989-2013. journal of public administration, finance and law, 6(6), 117-130. atanassovy, j., & nanda, v. (2018). crowding-out innovation. in proceedings. annual conference on taxation and minutes of the annual meeting of the national tax association (vol. 111, pp. 1-64). national tax association. atesoglu, h. s. (2004). defense spending and investment in the united states. journal of post keynesian economics, 27(1), 163-170. ayange, a., samuel, u. e., prince, a. i., & ndubuaku, v. (2020). security expenditure on economic growth in nigeria. humanities and social sciences review, 8(3), 553-561. https://doi.org/10.18510/hssr.2020.8360 caruso, r., & di domizio, m. (2017). military spending and budget deficits: the impact of us military spending on public debt in europe (1988-2013). defence and peace economics, 28(5), 534-549. https://doi.org/10.1080/10242694.2016.1228259 deger, s., (1986), military expenditure in the third world countries: the economic effect, routledge and kegan paul, london. dunne, j. p., & smith, r. p. (2020). military expenditure, investment and growth. defence and peace economics, 31(6), 601-614. https://doi.org/10.1080/10242694.2019.1636182 https://doi.org/10.1353/jda.2015.0146 https://doi.org/10.1177/2394901518795067 https://doi.org/10.1016/j.ejpoleco.2012.07.002 https://doi.org/10.18510/hssr.2020.8360 https://doi.org/10.1080/10242694.2016.1228259 https://doi.org/10.1080/10242694.2019.1636182 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 188 edith, e. c., abolore, y. f., oa, a. o., & moses, s. g. (2019). security spending and foreign direct investment inflows in nigeria: an autoregressive distributed lag model approach. gold, d. (1997). evaluating the trade‐off between military spending and investment in the united states. defence and peace economics, 8(3), 251-266. https://doi.org/10.1080/10430719708404880 hou, n., & chen, b. (2014). military expenditure and investment in oecd countries: revisited. peace economics, peace science and public policy, 20(4), 621-630. https://doi.org/10.1515/peps-2014-0031 jelilov, g., ozden, k., & briggs, s. o. (2018). impact of insecurity on investment in nigeria. journal of management, economics, and industrial organization, 2(3), 41-61. https://doi.org/10.31039/jomeino.2018.2.3.3 kennedy, p. s. j. (2021). the effect of defense spending on private investment in indonesia based on historical data for the period 1981-2010. palarch's journal of archaeology of egypt/egyptology, 18(4), 7094-7102. keynes, j. m. (2018). the general theory of employment, interest, and money. springer (original work published 1936). https://doi.org/10.1007/978-3-319-70344-2 kimenyi, m., adibe, j., djiré, m., jirgi, a. j., kergna, a., deressa, t. t., & westbury, a. (2014). the impact of conflict and political instability on agricultural investments in mali and nigeria. africa growth initiative (vol. 17). working paper. knight, m., loayza, n., & villanueva, d. (1996). the peace dividend: military spending cuts and economic growth. staff papers, 43(1), 1-37. https://doi.org/10.2307/3867351 kollias, c., & paleologou, s. m. (2019). military spending, economic growth and investment: a disaggregated analysis by income group. empirical economics, 56(3), 935-958. https://doi.org/10.1007/s00181-017-1379-2 loayza, n., knight, m., & villanueva, d. (1999). the peace dividend: military spending cuts and economic growth. the world bank. https://doi.org/10.1596/1813-9450-1577 lorusso, m., & pieroni, l. (2017). the effects of military and non-military government expenditures on private consumption. journal of peace research, 54(3), 442-456. https://doi.org/10.1177/0022343316687017 malizard, j. (2015). does military expenditure crowd out private investment? a disaggregated perspective for the case of france. economic modelling, 46, 44-52. https://doi.org/10.1016/j.econmod.2014.10.049 mintz, a., & huang, c. (1990). defense expenditures, economic growth, and the "peace dividend". american political science review, 84(4), 1283-1293. https://doi.org/10.2307/1963264 morales-ramos, e. (2002). defence r&d expenditure: the crowding-out hypothesis. defence and peace economics, 13(5), 365-383. https://doi.org/10.1080/10242690213507 oji, c. e., & afolabi, j.a. (2022). economic and peace effects of terrorism in the 21st century. international journal of public law and policy, 8(2): 144-157. https://doi.org/10.1504/ijplap.2022.122127 okereocha, c. (2012), "heartache for the economy", tell, may 14, pp. 46 47. https://doi.org/10.1080/10430719708404880 https://doi.org/10.1515/peps-2014-0031 https://doi.org/10.31039/jomeino.2018.2.3.3 https://doi.org/10.1007/978-3-319-70344-2 https://doi.org/10.2307/3867351 https://doi.org/10.1007/s00181-017-1379-2 https://doi.org/10.1596/1813-9450-1577 https://doi.org/10.1177/0022343316687017 https://doi.org/10.1016/j.econmod.2014.10.049 https://doi.org/10.2307/1963264 https://doi.org/10.1080/10242690213507 https://doi.org/10.1504/ijplap.2022.122127 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 189 olaifa, f. g., & benjamin, o. o. (2020). government capital expenditure and private sector investment in nigeria: co-integration regression and toda-yamamoto causality analysis. advanced journal of social science, 6(1), 71-82. https://doi.org/10.21467/ajss.6.1.71-82 olubunmi, a. a. (2018). insecurity and major determinants of foreign direct investment in nigeria. international journal of humanities and social science research, 4, 54-63. https://doi.org/10.6000/2371-1655.2018.04.06 pesaran, m. h., shin, y., & smith, r. j. (2001). bounds testing approaches to the analysis of level relationships. journal of applied econometrics, 16(3), 289-326. https://doi.org/10.1002/jae.616 pieroni, l. (2009). does defence expenditure affect private consumption? evidence from the united states. economic modelling, 26(6), 1300-1309. https://doi.org/10.1016/j.econmod.2009.06.004 raifu i. a., & afolabi, j. a. (2022). military expenditure and unemployment in south africa: evidence from linear and nonlinear ardl with and without structural break. economic alternative raifu i. a., obijole, e. o. & nnadozie, o. o. (2022). the role of institutional quality in military spending and unemployment nexus in nigeria. peace economics, peace science and public policy, vol. 28 issue 2, pp. 155-194. https://doi.org/10.1515/peps-2022-0006 raifu, i. a. (2021). the institutional role in oil revenue‐economic growth nexus in nigeria. opec energy review, 45(1), 52-84. https://doi.org/10.1111/opec.12197 raifu, i. a., & aminu, a. (2020). a reconsideration of wagner's hypothesis for nigeria using linear and nonlinear ardl methods. international journal of sustainable economy, 12(4), 321-343. https://doi.org/10.1504/ijse.2020.112308 raifu, i. a., & raheem, a. n. (2018). do government revenues matter for economic growth? evidence from nigeria. european journal of government and economics, 7(1), 60-84. https://doi.org/10.17979/ejge.2018.7.1.4333 raifu, i. a., aminu, a., & folawewo, a. o. (2020). investigating the relationship between changes in oil prices and unemployment rate in nigeria: linear and nonlinear autoregressive distributed lag approaches. future business journal, 6(1), 1-18. https://doi.org/10.1186/s43093-02000033-w saba, s. c., & ngepah, n. (2019). military expenditure and economic growth: evidence from a heterogeneous panel of african countries. economic research-ekonomska istraživanja, 32(1), 3586-3606. https://doi.org/10.1080/1331677x.2019.1674179 sajid, m. j. (2021). continued increases in military and police spending can lead to economic collapse. in e3s web of conferences (vol. 253, p. 03036). edp sciences. https://doi.org/10.1051/e3sconf/202125303036 scott, j. p. (2001). does uk defence spending crowd‐out uk private sector investment? defence and peace economics, 12(4), 325-336. https://doi.org/10.1080/10430710108404991 shahbaz, m., afza, t., & shabbir, m. s. (2013). does defence spending impede economic growth? cointegration and causality analysis for pakistan. defence and peace economics, https://doi.org/10.21467/ajss.6.1.71-82 https://doi.org/10.6000/2371-1655.2018.04.06 https://doi.org/10.1002/jae.616 https://doi.org/10.1016/j.econmod.2009.06.004 https://doi.org/10.1515/peps-2022-0006 https://doi.org/10.1111/opec.12197 https://doi.org/10.1504/ijse.2020.112308 https://doi.org/10.17979/ejge.2018.7.1.4333 https://doi.org/10.1186/s43093-020-00033-w https://doi.org/10.1186/s43093-020-00033-w https://doi.org/10.1080/1331677x.2019.1674179 https://doi.org/10.1051/e3sconf/202125303036 https://doi.org/10.1080/10430710108404991 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 190 24(2), 105-120. https://doi.org/10.1080/10242694.2012.723159 shan, j. (2002). a var approach to the economics of fdi in china. applied economics, 34(7), 885-893. https://doi.org/10.1080/00036840110058941 sims, c. a. (1980). macroeconomics and reality. econometrica: journal of the econometric society, 1-48. https://doi.org/10.2307/1912017 smith, r. p. (1980). military expenditure and investment in oecd countries, 19541973. journal of comparative economics, 4(1), 19-32. https://doi.org/10.1016/0147-5967(80)90050-5 smith, r., & dunne, p. (2001). military expenditure growth and investment. birbeck college and middlesex university business school, abril. stratmann, t. & okolski, g.l. (2010). does government spending affect economic growth? monetary policy mercatus on policy series june 10, mercatus center, george mason university. temitope, l., & ajala, o. (2021) military expenditure and economic growth: evidence from nigeria. american journal of economics, 11, 10-18. thia, j. p. (2020). deficits and crowding out through private loan spreads. the quarterly review of economics and finance, 77, 98-107. https://doi.org/10.1016/j.qref.2020.01.009 üçler, g. (2016). testing the relationship between military spending and private investments: evidence from turkey. theoretical & applied economics, 23(3). wagner, r. e. (2007). fiscal sociology and the theory of public finance: an exploratory essay. edward elgar publishing. https://doi.org/10.4337/9781781951354 yusuf, a., & mohd, s. (2022). growth and fiscal effects of insecurity on the nigerian economy. the european journal of development research, 1-27. https://doi.org/10.1057/s41287-02200531-3 https://doi.org/10.1080/10242694.2012.723159 https://doi.org/10.1080/00036840110058941 https://doi.org/10.2307/1912017 https://doi.org/10.1016/0147-5967(80)90050-5 https://doi.org/10.1016/j.qref.2020.01.009 https://doi.org/10.4337/9781781951354 isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 191 appendix 1. 0 20 40 60 80 100 1 2 3 4 5 6 7 8 9 10 private investment real gdp unemployment rate military spending (% of gdp) variance decomposition of private investment (military spending % of gdp) using cholesky (d.f. adjusted) factors isiaka akande raifu / european journal of government and economics 11(2), december 2022, 167-192 192 0 20 40 60 80 100 1 2 3 4 5 6 7 8 9 10 private investment real gdp unemployment rate military spending variance decomposition of private investment (military spending) using cholesky (d.f. adjusted) factors abstract 1. introduction 2. literature review 3. theoretical consideration, model specification and data sources 4. empirical results 5. conclusions and policy implications references appendix the profile of leisure time sports people and their reason for doing sport in spanish sports facilities vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 210-219 european journal of government and economics issn: 2254-7088 the profile of leisure time sports people and their reason for doing sport in spanish sports facilities moisés grimaldi-puyana a, pablo gálvez-ruiz b, manel valcarce-torrente c, ainara bernal-garcía d* a universidad de sevilla, spain b universidad internacional de valencia, spain c valte – us s.l. spin-off at universidad de sevilla, spain d universidad pablo de olavide, spain * corresponding author at: abernal1@us.es article history. received 16 december 2019; first revision required 22 april 2020; accepted 18 june 2020. abstract. in spanish sports habits, there exists a clear increase of the institutionalised sports practice and a distinct trend of leisure and recreational sport. this motivates the current work as it is necessary to go deeply into the knowledge of this practice. the aims proposed are twofold. on the one hand, to perform an analysis of the profile and the characteristics of the behaviour of the users of sports centres and leisure and recreational sportspeople. and, on the other hand, to study the motives of leisure/recreational sports practice. a descriptive quantitative methodology has been followed in this research. the description of the profile of users emerges from the data analysed as well as that they practise sport for exotelic reasons. the results obtained can serve managers to orientate their strategies destined to satisfy the needs of these users. keywords. leisure sport; motives for practice; profile; recreation jel codes. z0 doi. https://doi.org/10.17979/ejge.2020.9.2.5846 1. introduction from a humanist perspective, we understand leisure as an essential part of human development, affecting it as a personal experience and revealed via numerous practices according to the environment and its resources, and the demographic and social context (lazcano and ortuzar, 2016; cuenca, 2000). according to cuenca, aguilar and ortega (2010), this leisure can be expressed in two directions, negative or positive, both having two areas of action. the negative version of leisure is considered to be absent leisure (a personal absence of experiences) or harmful leisure (detrimental experiences, both personal and social). we can in turn differentiate the positive version of leisure between exotelic leisure (practised as a means to achieve a goal) and autotelic leisure (true, which is done for itself, without a specific aim). taking into account the perspective of sport as leisure and adapting it to the theory described of the concept of humanistic leisure (aristegui-fradua and silvestre-cabrera, 2012), and understanding this type of positive leisure from a double functionality as different authors describe (cuenca et al., 2010) means: sport as exotelic leisure (as the sports practice of mailto:abernal1@us.es https://doi.org/10.17979/ejge.2020.9.2.5846 grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 211 recreational leisure to achieve a goal and not as an end in itself) and sport as autotelic leisure (referring to recreational/leisure practice sports which is done satisfactorily without a utilitarian purpose). according to these premises, we note that in the surveys of spanish sports habits there is a clear increase in disinstitutionalised sports practice and a distinct trend to leisure and recreational sport (garcía-ferrando and llopis-goig, 2017). in the andalusian context and for the first time, the law 5/2016 of sport in andalusia, defined “sport of leisure and recreation” as “physical activity which is done in an organisation or outside it, and aimed at achieving noncompetitive aims related with the improvement of health, acquiring sports habits, as well as the active occupation of free time”. this places the sportsperson outside competition and having an evident orientation towards motivations intrinsic to the individual. since 2009, a growing evolution has been noted of people who say that they practice sport in europe, and according to the european commission (2018) the proportion of people doing so increased from 42 % to 46 % in all the continent. in spain, and according to the survey of sports habits (csd, 2015), 46.2 % of people over 15 years old, practice sport at least once a week, and in reference to the sector of gyms and fitness, we find that 40% of the population in our country are enrolled in a centre of this type (europeactive and deloitte, 2019). a reflection of sports practice is the users’ spending in sports-linked services. the 2019 sports statistics yearbook elaborated by the ministry of culture and sport states that in 2017 the annual spending in specific sports-linked goods by spanish households was 5,686.5 million euros, which represents 1.1% of their total spending in goods and services. the average spending per household in sports-linked goods and services was 307.2 euros, and the average spending per person was 123.4 euros. this yearbook highlights that the most significant sportslinked spending corresponds to recreational and sports services, 79.2%. as appears from the results obtained, the average spending per person in this type of goods is higher than the average in the autonomous communities of aragón, the principality of asturias, the balearic islands, cantabria, catalonia, the valencian community, the community of madrid, the region of murcia, navarre and the basque country. analysing the indicators of the evolution of the prices of specific sports-linked products, obtained from the consumer price index (cpi) base 2016 statistics (ine), the results available indicate that the cpi of recreational and sports services has increased by 0.6 points while that of large sports equipment has decreased by 2 points compared with 2017, the increase being 1.7 points in the national cpi. likewise, the 2019 yearbook of sports statistics provides information on the profiles of the users who carry out sports practices (table 1). grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 212 table 1. sports habits according to personal characteristics (ministerio de educación, cultura y deporte, 2018). in percentage of the population analysed in percentage of the population who do sport in the last year at least once a week in the last year at least once a week total 53.5 46.2 100 86.3 gender male female 59.8 47.5 50.4 42.1 100 100 84.3 88.6 age from 15 to 19 years from 20 to 24 years from 25 to 34 years from 35 to 44 years from 45 to 54 years from 55 to 64 years from 65 to 74 years from 74 and over 87.0 78.2 72.6 64.6 53.2 44.5 30.0 10.9 81.5 71.0 62.5 54.0 44.9 37.4 27.0 9.6 100 100 100 100 100 100 100 100 93.7 90.9 86.0 83.5 84.4 83.9 90.1 88.0 level of studies first stage of secondary school or less second stage of secondary school higher education or equivalent 38.8 63.1 73.4 33.4 53.7 64.1 100 100 100 85.9 85.2 87.3 work situation working unemployed retired student household tasks others 65.5 54.4 23.5 85.2 27.8 38.7 55.9 45.2 20.4 78.2 24.7 32.9 100 100 100 100 100 100 85.4 83.1 86.5 91.8 89.1 85.1 on the other hand, and addressing the spanish population’s reasons for doing sport (table 2), the following results are noted. firstly, “to be fit” with 29.9% (32.7% women compared to 27.6% men), second “fun or entertainment” with 23% of the cases (27% men and 18.4% women), third “health reasons” with 14.8%, fourth “a way to relax” with 13.7%, stronger in women than in men; fifth “he/she likes doing sport”, stronger in men than in women (11.9%), and with lower rates “as a form of social relation” (2.6%); “for personal growth” (1.8%), “he/she likes to compete” (1.5%) and “by profession” (0.7%). table 2. distribution of the spanish population for reasons of doing sport (csd, 2015). order reason % population leisure or recreational sport 1 to be fit 29.9 6,053,615.8 2 fun or entertainment 23.1 4,676,873.7 3 health reasons 14.8 2,996,438.6 4 a way to relax 13.7 2,773,730.3 5 he/she likes doing sport 11.9 2,409,298.6 6 as a form of social relation 2.6 526,401.4 7 for personal growth 1.8 364,431.7 19,800,790.1 8 he/she likes to compete 1.5 202,463.6 9 by profession 0.7 141,723.4 344,187.0 total 100 20,144,977,2 note. spanish population over 15 years old. grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 213 likewise, different recent authors and studies confirm the type of sports practice analysed in table 2, oriented towards the view of leisure as a positive factor in people and society, be it exotelic or autotelic. the great majority of the participants of these studies consider doing sport with non-competitive aims, focused on improving health, improving their physical or mental level, or questions related with personal image, self-esteem and relations among people (elasri et al., 2016; garcía-fernández et al., 2017; moreno-murcia et al., 2016; valcarce-torrente, 2016). 2. aim as described above, the present work has a double aim. firstly, to carry out an analysis of the profile and the characteristics of the behaviour of leisure time and recreation sports people users of sports centres. secondly, to study their reasons for doing sport, differentiating those who do sport as a means of recreational leisure from those who do so as an end in itself. 3. methodology 3.1. procedure this cross-sectional study used non-probability convenience sampling. the participants were 1,050 users (426 men and 624 women) users of 52 sports centres belonging to eight large chains in spain. those who agreed to collaborate in the study were informed of the study’s objectives, their voluntary and anonymous participation, in addition to an assurance of the confidentiality of the answers. they then completed the consent form prior to being provided with the link to access the online questionnaire. the participants responded to the questionnaire in about 10–12 min to complete all the items. the data collection was carried out during the months of january 2018 to april 2018. 3.2. measures an online seven-point likert scale questionnaire was used to measure the role of reasons for leisure sportspeople practicing physical activity. the questionnaire initially consisted of 50 items, was used for data collection was composed of a socio-demographic section (gender, age, education level and marital status) and a section of characteristics of the behavior of leisure/recreation sportspeople formed by five variables: type of sports facility, permanence and previous experience, frequency and time use of the sport facility, spending on sports services and products in percentages 3.3. data analysis a descriptive quantitative methodology has been followed in this investigation. for this, the spss v23 program was used for the analysis of the results, and the data were presented in percentages (profile and expense), averages and standard deviations (reasons for practice). grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 214 4. results 4.1. profile of the users firstly, regarding the profile of the users of the sports facilities studied, the analysis of the data reveals that they are mainly either between 21-30 years old (27.0%; n = 284) or are over 60 years old (20.9%; n = 219) and single (47.6%; n = 500) (table 3). table 3. distribution of the sample by age and marital status. frequency percentage age less than 21 years old 95 9.0 21-30 years old 284 27.0 31-40 years old 146 13.9 41-50 years old 176 16.8 51-60 years old 130 12.4 over 60 years old 219 20.9 total 1,050 100.0 marital status single 500 47.6 married 432 41.1 divorced 61 5.8 widow/er 17 1.6 dk/da 40 3.9 total 1,050 100.0 as to the level of studies, 38.5% were graduates and only 5.5% did not have studies or only had primary studies (table 4). the analysis of the level of income variable establishes that the sample is characterised by having mainly low incomes; in fact, 43.9% do not surpass 250 euros (n = 461). table 4. level of studies. frequency percentage level of studies without studies 9 .9 primary studies 49 4.7 secondary studies 167 15.9 undergraduate course 97 9.2 higher technical certificate 156 14.9 degree 404 38.5 doctorate 15 1.4 university master 117 11.1 dk/da 36 3.4 total 1,050 100.0 grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 215 table 5. level of net monthly incomes. frequency percentage income 0-250€ 461 43.9 251-500€ 48 4.6 501-750€ 56 5.3 751-1,000€ 81 7.7 1,001-1,500€ 104 9.9 1,501-2,000€ 90 8.6 2,001-2,500€ 64 6.1 2,501-3,000€ 17 1.6 over 3,000 20 1.9 dk/da 109 10.4 total 1,050 100.0 4.2. characteristics of the behavior of leisure/recreation sportspeople centring our analysis on the characteristics which define the behaviour of the leisure/recreation user, it is established that 52.1% of the customers chose a private gym as sports installation (n = 547) (table 6). table 6. level of net monthly incomes. frequency percentage type of facility privately managed municipal centre 199 19.0 publicly managed municipal centre 101 9.6 private gym 547 52.1 dk/da 203 19.3 total 1,050 100.0 addressing if the user had been enrolled before in another sports installation, 62.1% (n = 652) of them had been customers of another installation. as to the permanence in the sports centre, 34.5% indicated that they had been a user for more than 2 years (n = 362) (table 7). table 7. permanence and previous experience. frequency percentage permanence less than 3 months 202 19.2 3-6 months 178 17.0 6-12 months 121 11.5 1-2 years 187 17.8 more than 2 years 362 34.5 total 1,050 100.0 previous experience no 319 30.4 yes, in the same centre 79 7.5 yes, in another centre 652 62.1 total 1,050 100.0 with relation to the frequency, 39.1% of the users declared that they went three times a week, 36.7% more than four times a week, 16.9% twice a week and 7.9% less than once a grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 216 week. secondly, regarding the time of use, 57.8% declared having stayed in the centre between 60 and 90 minutes, 26% between 30 and 60 minutes and 19% more than 90 minutes (table 8). table 8. frequency and time use of the sport facilities. frequency percentage frequency less than once a week 30 2.9 once a week 47 4.5 twice a week 177 16.9 three times a week 411 39.1 more than four times a week 385 36.6 total 1,050 100 time of use less than 30 minutes 3 .3 30 60 min 241 23.0 60 90 min 607 57.7 more than 90 min 199 19.0 total 1,050 100.0 regarding spending in sports practice, we find that greater percentages are in personal trainers, sportswear, sports shoes and sports nutrition (50€-300€/year), as follows: 50€/year on sports nutrition (59.6%), personal trainers (59.3%), sportswear (31.7%) sports shoes (308%); 50-100€/year on sports shoes (43.1%) and sports nutrition (24.3%); while 101-300€/year on sports shoes (21.5%), sportswear (18.9%), sports nutrition (7.4%) and personal trainers (1.3%) (table 9). table 9. spending on sports services and products in percentages. spending personal trainers sportswear sport shoes sports nutrition less than 50€/year 59.3 31.7 30.8 59.6 50-100€/year 0.6 41.7 43.1 24.3 101-300€/year 1.3 18.9 21.5 7.4 301-500€/year 0.7 4.7 2.2 2.4 501-750€/year 0.6 0.7 0.3 0.9 over 750€/year 0.6 0.4 0.2 0.5 dk/da 37.0 2.0 1.9 5.0 total 100.0 100.0 100.0 100.0 4.3. reasons for leisure sportspeople practicing physical activity as to the reasons for practicing sports of the users of sports facilities installations, we find lower and contrary values to those which define the reasons that characterise leisure and recreational sportspeople, such as: “he/she likes competing” (m = 2.86; sd = 1.96) and “by profession” (m = 2.09; sd = 1.69). this is compared to higher values in leisure and recreational sportspeople, and in the following order: “to be fit” (m = 6.25; sd = 1.14); “health reasons” (m = 5.80; sd = 1.49); “way to relax” (m = 5.26; sd = 1.77); “he/she likes doing sport” (m = 5.21; sd = 1.83); “for personal growth” (m = 4.71; sd = 1.96); and lastly “as a form of social relation” (m = 3.26; sd = 1.90). grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 217 5. discussion few comparisons of the profile of the user obtained in this study can be made with other, similar studies due to works with the same end not being easily found in the literature. nevertheless, though these studies do not base their analysis on the leisure/recreational users of sports practise, we can venture to show coincidences and interpret that the personal and socioeconomic characteristics of these specific users are similar to those of any practitioner, as the works of baena-arroyo et al. (2020), garcía-fernández et al. (2019), garcía-fernández et al. (2017), garcía-fernández et al. (2016), valcarce-torrente et al. (2019) and the 2019 yearbook of sports statistics show. for their part, the results obtained regarding the reasons for practising leisure/recreational sports back those already obtained before by authors such as valcarce-torrente et al. (2019), although in this latter case the results were obtained from the study of the behaviour of users of low-cost fitness centres. then, studies such as those performed by the csd (2015), elasri et al. (2016), garcía-fernández et al. (2017), moreno-murcia et al. (2016) and valcarce-torrente (2016) conclude that the great majority of the participants do sports focusing on health improvement, improving their physical or mental level, or matters related with personal image, self-esteem and the relations between people. so this reinforces the theory that exotelic reasons predominate in the sports practice oriented towards leisure/recreation. 6. conclusions based on the data obtained, it can be concluded that the profile of the leisure/recreational sportsperson is that of a person between 21 and 30 years old, single, with university studies and monthly incomes equal to or less than 250 euros. with respect to the behaviour of the users, they are customers of private gyms who have been enrolled previously in other sports centres and who have permanence in their current installation of fewer than 2 years. they attend three times a week and stay between 60 and 90 minutes at the sports centre. the spending on sports services and products is mainly on personal trainers, sportswear, sports shoes and sports nutrition. finally, the reasons why they do sport can be concluded to be exotelic; that is to say, it is done to attain another goal and not as an end in itself (to be fit, health reasons and a way of relaxing). 7. research limitations the present work presents limitations, which are necessary to collect so that they can serve as a reference for these same researchers or others that may be of interest to this line. fundamentally, we find methodological limitations. on the one hand, as it has been advanced in the discussion section, it is difficult to compare with respect to user profiles because there are no coincidences in the scientific literature in relation to the variables analyzed. on the other grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 218 hand, the sample was not distinguished or segmented based on the sports center model beyond the type of management. we understand that the user profile will depend on the model of the center and may also affect the reasons for practice. 8. future lines of research and implications for management based on the aforementioned limitations, we propose the following lines of work. first of all, increase and segment the sample taking into account the different business models in which sports centers are concerned. on the other hand, perform correlation analysis between the different user profiles or specific characteristics of these with the reasons for practice, as well as being able to study whether sports centers know and adapt to these concerns of their clients. finally, and in relation to the implications for management, it can be stated that the results obtained can serve managers to guide their strategies aimed at satisfying the needs of these users. knowing not only the profile of their clients but the motivations that motivate them to go to the center and practice sports can help them make decisions about the services to offer. references aristegui-fradua, i., and silvestre-cabrera, m. (2012). el ocio como valor en la sociedad actual. arbor, 188(754), 283-291. https://doi.org/10.3989/arbor.2012.754n2002 baena-arroyo, m. j., garcía-fernández, j., gálvez-ruiz, p., and grimaldi-puyana, m. (2020). analyzing consumer loyalty through service experience and service convenience: differences between instructor fitness classes and virtual fitness classes. sustainability, 12(3), 828. https://doi.org/10.3390/su12030828 csd (2015). encuesta de hábitos deportivos en españa 2015. madrid: ministerio de educación, cultura y deporte. lazcano, i., and madariaga, a. (2016). el valor del ocio en la sociedad actual. in m.a. berthet et al. (eds.), la marcha nocturna: ¿un rito exclusivamente español? (pp. 15-33). madrid: centro reina sofía sobre adolescencia y juventud. fundación de ayuda contra la drogadicción (fad). cuenca, m. (2000). ocio humanista. dimensiones y manifestaciones del ocio, documentos de estudios de ocio (16). bilbao: universidad de deusto. cuenca, m. aguilar, e. and ortega, c. (2010). ocio para innovar. documentos de estudios de ocio (42). bilbao: universidad de deusto elasri, a., triadó, x.m., and aparicio, p. (2016). los usuarios de los centros deportivos públicos: una aproximación de segmentación sobre hábitos deportivos y satisfacción. revista de psicología del deporte, 25(3), 15-18. europeactive and deloitte (2019). european health & fitness market report 2018. brussels: europe active. european commission (2018). special eurobarometer 472. sport and physical activity. brussels: european commission. garcía-ferrando, m., and llopis-goig, r. (2017). la popularización del deporte en españa: en cuesta de hábitos deportivos 1980-2015. madrid: cis (centro de investigaciones sociológicas). garcía-fernández, j., gálvez-ruiz, p., and vélez-colón, l. (2017). client profile of spanish fitness centers: segmentation by loyalty and characteristics of the client. in m. peris-ortiz, j. álvarez-garcía y m.c. del río-rama (eds.), sport management as an emerging economic activity (pp. 273-291). cham, switzerland: springer. https://doi.org/10.1007/978-3-31963907-9_17 https://doi.org/10.3989/arbor.2012.754n2002 https://doi.org/10.3390/su12030828 https://doi.org/10.1007/978-3-319-63907-9_17 https://doi.org/10.1007/978-3-319-63907-9_17 grimaldi-puyana et al. / european journal of government and economics 9(2), july 2020, 210-219 219 garcía-fernández, j., sánchez-oliver, a. j., grimaldi-puyana, m., fernández-gavira, j., and gálvez-ruíz, p. (2017). la calidad y la fidelidad del cliente: un análisis de segmentación en centros de fitness low-cost. revista de psicología del deporte, 26(3), 17-22. garcía-fernández, j., grimaldi-puyana, m., gómez, r., and bernal-garcía, a. (2016). qualidade e valor em centros de fitness low-cost: diferenças de acordo com as características dos clientes. revista intercontinental de gestão desportiva-rigd, 6, 63-72 garcía-fernández, j., gálvez-ruiz, p., sánchez-oliver, a. j., fernández-gavira, j., pitts, b. g., and grimaldi-puyana, m. (2020). an analysis of new social fitness activities: loyalty in female and male crossfit users. sport in society, 23(2), 204-221. https://doi.org/10.1080/17430 437.2019.1625332 grimaldi-puyana, m., sánchez-oliver, a.j., macías-plat, r. and álvarez-garcía, j. (2018). perfil, evolución y características de las organizaciones de gestión de campos de golf en españa. revista espacios, 39(33). grimaldi-puyana, m., gálvez-ruiz, p., sánchez-oliver, a.j., and alcaraz-rodríguez, v. (2018). situación de endeudamiento de las organizaciones de gestión de instalaciones deportivas de ocio no competitivo. un estudio empírico. sport tk. revista euroamericana de ciencias del deporte, 7(2), 41-48. https://doi.org/10.6018/sportk.343091 grimaldi-puyana, m., garcía-fernández, j., gómez-chacón, r., and bravo, g. (2016). impacto de la crisis económica en la rentabilidad de las empresas de gestión de instalaciones de ocio deportivo según tamaño, un estudio empírico. revista de psicología del deporte, 25(1), 51-54. grimaldi puyana, m., and sánchez-oliver, a.j. (2017). tamaño y evolución económica de los clubes de baloncesto sad después de la crisis económica de 2008. cuadernos de psicología del deporte, 17(3), 137-142. ministerio de educación, cultura y deporte (2018). anuario de estadísticas deportivas 2019. madrid: secretaría general técnica del mecd. moreno-murcia, j.a., marcos-pardo, p.j., and huéscar, e. (2016). motivos de práctica físicodeportiva en mujeres: diferencias entre practicantes y no practicantes. revista de psicología del deporte, 25(1), 35-41. valcarce-torrente, m. (2016). modelos de negocio deportivos: evolución y futuro. in j. garcía (ed.), emprendiendo en el sector deportivo, de la teoría a la práctica (pp. 113-126). cizur menor: thomson reuters aranzadi. valcarce-torrente, m., garcía-fernández, j., gálvez-ruiz, p., and grimaldi-puyana, m. (2019). el consumo en centros de fitness low-cost: un análisis desde el cliente. in m. valcarcetorrente, v. jabaloyes-sanchís, j. garcía-fernández y m. grimaldi-puyana (eds.), sport business symposium. deporte y empresa, una relación de éxito, 2019 (pp. 39-41). sevilla: wanceulen editorial sl. https://doi.org/10.1080/17430437.2019.1625332 https://doi.org/10.1080/17430437.2019.1625332 https://doi.org/10.6018/sportk.343091 ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5846_g comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 170-180 european journal of government and economics issn: 2254-7088 comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) benito pérez-gonzález a*, luis de la riva b, josé bonal b, álvaro fernández-luna b a universidad isabel i, spain b universidad europea de madrid, spain * corresponding author at: benitoperezgonzalez@gmail.com article history. received 15 january 2020; first revision required 17 march 2020; accepted 25 may 2020. abstract. the purpose was to analyse the income variation of the 13 top clubs of deloitte football money league report (2019) as well as the variation of the perceived value of their squad, obtained from the website www.transfermarkt.es during 9 seasons. friedman's two-way analysis by ranges of selected samples and wilcoxon w tests were used to measure the evolution and relation among the clubs’ income and perceived value of players. the income of clubs was deflected according to inflation (cpi: 2016) and market value was adjusted to a 20 players squad. market value and income have a significant positive variation from one year to the next (p = .000). we found differences between both variables only in the last 2 years on the period analysed (2011-2019). market value has suffered an increase higher than expected. this fact should make football managers reflect on how to control constant inflation in the transfer market. keywords. financial fair play; football; revenue; perceived value; transfer market jel codes. l51; l83; m10; z20 doi. https://doi.org/10.17979/ejge.2020.9.2.5953 1. introduction in recent years there has been an increase in the income of the main european football clubs (deloitte, 2019; solberg, 2016), and also an increase in the market value of footballers, confirmed by the extensive bibliography that links both variables (kuper and szymanski, 2014; rodríguez, 2012; sánchez, barajas, and sánchez-fernández, 2019). in the literature there are also multiple confirmations of the relationship between income and results (hall, szymanski, and zimbalist, 2002; pawlowski, breuer, and hovemann, 2010; plumley, ramchandani and wilson, 2019). however, there are other factors, aside from the economic ones, that play a role when it comes to obtaining a highly valued team, among which is the capacity of its manager or its coaches (pérez-gonzález, 2017). mailto:benitoperezgonzalez@gmail.com https://doi.org/10.17979/ejge.2020.9.2.5953 pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 171 2. literature review transfer market sport is, in many ways, a perfect laboratory to try to obtain intuitions about the way in which humans make decisions (palacios-huerta, 2019). one of the biggest concerns of managers and coaches is to get the best possible staff for their club; but these are subject to numerous biases (varela-quintana and del corral, 2019) that can influence the decisions they make when putting together their teams. bias such as projection bias or over inference can make you decide to buy a player for a price higher than its value. another fundamental factor when making signings is the situation of “dominance or distress” (szymanski, 2019) in which the club is at that time. clubs that have achieved their goals or that have been close to doing so have less pressure than those who did not. losers are subjected to the double urgency of having to make changes and the knowledge that exists in the market of that situation, which usually leads inexorably to the assumption of cost overruns. some types of player overvaluation have been described in the literature: for example, the overvaluation of players of some nationalities, the possible overvaluation of mature players, the overvaluation of players when they respond to some important need in any of the lines of a club or should replace a notorious exit (kuper and szymanski, 2014). in the sense of the latter bias, palacios-huerta (2019) highlights that the best players, those with some distinctive factor, are usually overrated. another important aspect in the transfer market movement is the taxation and salaries of the players (palacios-huerta, 2019). some countries and certain clubs offer economic incentives that are very attractive to football players. being the best offer from a sporting point of view is no longer a guarantee to attract talent. in the literature and in practice, different measures have been studied and applied to improve the competitive balance of competitions (késenne, 2015, 2019): the salary gap not used in europe or the reservation clause are some of these options to restrict the mobility of players. simmons and berri (2019) mention that szymanski (2015) argued that the transfer market should not exist, since it restricts workers’ free mobility, enabling owners to withhold most of the capital gain. according to him, if big clubs exchange players with each other using large transfer fees, an uneven and uncompetitive market will emerge. other authors advocate the existence of the transfer market, since they believe that the payment of the transfer compensates for the loss of a valuable asset (terviö, 2010). the last factor that is affecting the football market is the club’s academies management. successful football clubs are increasingly placing value on their youth academies. in many cases, clubs own facilities that accommodate several youth teams, including those with children who are under 10 years old. clubs in which many youth academy players make it to the first team save large amounts of money that they would otherwise invest in signing players. an illustrative example of this situation is club atlético de madrid, which has managed to be in the top 5 of the most valuable squads despite earning less than one third of the revenue of its direct competitors (pérez-gonzález, 2017). finally, sometimes it is very difficult to know the real market value of a player or his real pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 172 transfer fee. therefore we must highlight the role of the transfermarkt website (www.transfermarkt.de). transfermarkt is a german-based website and the leading website on the football transfer market. the site offers, within other general football-related data, the estimations of market value at the individual and team levels for most professional football leagues, based on the judgement of fans (müller, simons and weinmann, 2017). this source has also been used in recent scientific studies (peeters, 2018) as providing reliable game performance indicators and has been described as a good predictor of real market values (herm, callsen-bracker, and kreis, 2014). however, there are some controversial aspects that could affect directly to the perceived (and real) market value of players. on the one hand, the main issue is the lack of knowledge and experience, as well as the impulsive behaviour of some registered users (lorenz et al., 2011), and on the other hand, possible attempts at manipulation by agents in lower categories such second division. football clubs’ income sources the income of football clubs can be divided into 3 main channels, following the classification proposed by deloitte: commercial, broadcasting and match day. below we show the evolution that the different types of income have had over the last 10 seasons for the first 13 teams in total revenues of the 2019 deloitte football money league report. figure 1. 10-year evolution of the percentage of income of the 13 clubs with the highest profits in the deloitte football money league 2019. source: own elaboration, deloitte data (2010-2019). only a couple of decades ago, more than 80% of revenues came from match day, mainly ticketing revenues, since the exploitation of stadiums for other uses was very residual. for this reason, santiago bernabéu decided to build a stadium with 100,000 spectators (sánchez et al., 2019). today, ticketing has ceased to be the main entrance of funds in the main clubs, but the german stadiums for the 2006 world cup or the inauguration of the emirates stadium of arsenal in that same year, served as an example for many clubs to make significant investments in both construction and remodeling. these clubs have also agreed to obtain 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 2008/09 2009/10 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 match day broadcasting commercial http://www.transfermarkt.de/ pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 173 income through the naming rights or the exploitation of their facilities both in match day and in other uses: catering, congresses, shopping centers, hotels, etc. the wanda metropolitano of the atlético de madrid, released in 2017, or the expensive remodeling projects of the santiago bernabéu (real madrid cf) and nou camp (fc barcelona) stadiums are good examples of this. but the match day, as important as it is, represents an increasingly lower percentage of club revenues, which are increasingly coming from broadcasting and commercial. television broadcasting rights have grown dramatically in the last decade in all major football leagues. for their part, the big clubs are constantly looking for new markets and followers to improve their income through sponsors, merchandising, tours, etc. tv rights are centralized in the major european leagues, including the spanish football league. this league chose the centralized option in 2015, in spite of the reluctance of real madrid and barcelona, the two biggest spanish clubs. centralization has promoted revenue growth in all clubs, and it has also reduced the gap between the small and big football clubs, since small clubs can now access the transfer market with a larger budget. to this end, real madrid has been working side by side with the american technology giant microsoft, whose mission was from the beginning to undertake a "digital transformation", to place fans at the epicenter of all the club's activity, enabling them to participate in it. the objective is to generate income in this way. real madrid is not just a sports club. it is a gigantic means of communication, but not only sports, also entertainment, and a huge social platform. the rise of annual revenue in the main european football leagues seems to indicate that this is a sector benefited by globalization, which has emerged as a result of new communications technology (garcía del barrio and tena horrillo, 2019). the importance of media visibility and fan pressure in the purchase of players represent factors that promote price increases in the transfer market. in spite of the weak competitiveness of european football leagues (causing strong imbalances), revenues have not stopped growing over the last years (fort, 2019). different studies show that european football clubs aim to maximize victories, not profits. lago, simmons and szymanski (2006) and also garcía del barrio and szymanski (2009) showed that there was a clear quest for the maximization of victories in leagues such as the spanish and the british league. more recently, fort (2019) argues that the pressure to achieve victories is much higher in europe than in the usa, in line with other cited studies. in addition, other researchers claim that the pressure to achieve wins is contrary to a management policy focusing on long-term performance (buraimo et al., 2015). these facts explain the strong proclivity of european clubs to reinvest their profit in purchasing talented players, which affects prices. the law of diminishing returns shows that there is a threshold in which the marginal performance of each new unit of talent will not provide a new unit of utility. media visibility is strongly correlated with the sporting talent of the club. but not everything has been revenue growth. the expenses of the clubs have grown linearly and many of them have been in economic difficulties for various reasons such as pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 174 mismanagement, the dizzying increase in the wage bill, the cost of staff renewal or the purchase of certain football clubs for tycoons or states. for all these reasons, some international organizations, such as fifa, have proposed account control mechanisms, seeking sustainable income and that comes exclusively from the football business (mareque, barajas, and lópez-corrales, 2018; sánchez et al., 2019). finally, the pyramidal system of promotions and relegations used in european leagues (szymanski, 2019) is another strong impediment for the maximization of profit. on the other hand, the american league system, which does not rely on promotions and relegations, facilitate agreements between clubs to maximize their profits. thus, based on the growing and changing economic situation of european football clubs and the importance of the testing of crowds in the players’ market value, this research aimed to analyse the income variation of the 13 top football clubs in the ranking of the last deloitte football money league report (2019) during the last 9 seasons, as well as the variation of the perceived value of their squad, during the same period, and if there is a relation between both factors. 3. methodology design transfermarkt (www.transfermarkt.com) is the leading website of the football transfer market. the site offers, within other general football-related data, the estimations of market value at the individual and team levels for most professional football leagues (müller, simons, and weinmann, 2017). the web, which establishes the market value of players based on registered users, has been used in other recent research (peeters, 2018) as a reliable indicator related to game performance (herm, callsen-bracker, and kreis, 2014) and as a good predictor of the real value of players (torgler and schmidt, 2007). on the other hand, deloitte football money league report, published just eight months after the end of the 2017/18 season, is the most contemporary and reliable independent analysis of the clubs’ relative financial performance (deloitte, 2019) and its objective is to profile the highest revenue generating clubs in the world of football. for our study, which analyses the income and the value of players of the last 9 seasons, we have consulted the deloitte football money league reports from 2011 to 2019. otherwise, transfermarkt holds information about the value of football players for the last 9 years. table 1. total data of income and perceived market value used in the study. 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19 total data obtained total data of value players obtained from transfermarkt 307 316 314 298 305 295 296 296 313 3.812 total data of income obtained from deloitte 13 13 13 13 13 13 13 13 13 117 http://www.transfermarkt.com/ pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 175 statistical analysis the evolution of the income obtained from the deloitte football money league reports and the value of the squad (consulted on february 2019) of the 13 selected clubs was analysed, using friedman's two-way analysis by ranges of selected samples. the income of clubs was deflected according to inflation (cpi: 2016). to study the relationship between the two types of variables, in the section of the 9 years of study, wicolxon w was applied for related samples. for both tests, p <0.05 was considered statistically significant. 4. results in table 2 we show the average value of the players, of each of the 13 teams analysed, throughout the 9 seasons object of our study. as indicated previously, these data have been obtained in transfermarkt and are deflated to inflation (cpi: 2016). table 2. average value of player by club and season (m€)a. 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19 real madrid cf 22,3 24,6 25,2 23,8 29,9 36,8 37,7 49,1 41,3 fc barcelona 26,8 29,8 24,6 25,6 27,0 28,4 34,7 48,9 49,5 manchester united 14,7 15,1 17,9 19,8 17,1 17,1 20,4 30,8 30,3 bayern münchen 13,5 16,4 17,8 22,6 25,0 26,0 29,5 36,4 30,4 manchester city 15,5 18,9 19,9 20,9 20,7 21,9 23,0 41,6 45,9 paris saint-germain 5,3 10,1 13,0 17,9 19,5 19,1 17,9 34,7 34,5 liverpool fc 16,4 10,9 11,5 12,7 12,1 13,5 15,6 22,5 39,3 chelsea fc 21,2 19,3 15,3 20,6 22,6 23,3 20,3 34,6 35,1 arsenal fc 13,8 11,0 11,2 13,8 15,8 15,9 18,4 23,2 25,1 tottenham hotspur 11,5 11,3 12,8 12,2 11,8 11,4 16,2 30,8 34,0 juventus 11,5 10,9 11,6 15,3 15,8 16,2 16,7 26,4 33,8 borussia dortmund 5,6 7,4 10,8 13,7 13,8 14,5 12,2 15,4 22,7 atlético de madrid 9,4 8,1 8,7 10,7 13,2 14,9 19,6 33,8 40,2 a: deflected according to inflation (cpi: 2016). on the other hand, in table 3 we show the income of each of the last 9 seasons of the 13 clubs. as indicated, the data have been obtained in the deloitte football money league reports of each season and are deflated to inflation (cpi: 2016). table 3. income by club and season (m€)a. 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19 real madrid cf 482,5 516,9 538,2 543,8 583,6 613,7 641,2 693,5 764,4 fc barcelona 437,9 485,9 507,2 505,8 514,6 597,3 641,3 666,5 702,8 manchester united 384,8 395,6 415,7 444,1 550,1 553,3 712,4 695,2 678,0 bayern münchen 355,3 346,5 386,8 451,9 517,7 504,8 612,1 604,3 640,5 manchester city 168,1 182,8 299,9 331,4 440,1 493,6 542,7 542,5 578,6 paris saint-germain 90,2 107,8 231,5 417,9 503,6 512,1 538,6 499,8 551,5 liverpool fc 247,8 219,2 244,9 252,1 324,9 417,3 417,5 436,1 522,9 chelsea fc 281,5 269,3 338,7 318,0 411,9 447,3 462,6 440,0 514,8 arsenal fc 301,5 270,7 304,8 297,9 381,6 463,8 484,4 501,3 447,1 tottenham hotspur 160,9 195,1 187,1 180,3 229,2 274,2 289,2 365,6 436,0 juventus 225,5 165,9 205,2 285,5 296,7 345,0 352,7 417,1 402,0 borussia dortmund 115,7 149,3 198,6 268,5 277,7 298,8 293,6 341,9 322,9 atlético de madrid 137,0 107,8 113,3 125,8 180,4 199,3 236,4 280,1 309,9 a: deflected according to inflation (cpi: 2016). pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 176 first, applying friedman's two-way analysis by ranges of related samples confirms, as we assumed, that both market value and revenue have a significant positive variation from one year to the next (p = .000). applying the w of wicolxon for related samples (table 4), we verify that there are no differences in the average income and market value adjusted to 20 players in the first 7 years. however, in the last two years, there are significant differences, with a negative sign. that is, the workforce values are not related to income. table 4. average value and wilcoxon test of clubs’ revenue and squad (20 players) perceived market value. 2010/11 2011/12 2012/13 2013/14 2014/15 2015/16 2016/17 2017/18 2018/19 average revenue s/deloitte (m€)a 259,2 261,1 303,8 338,3 398,7 437,5 476,1 496,3 519,2 average perceived market value of the squad (m€)a 14,3 14,8 15,3 17,6 18,9 19,7 21,5 32,4 34,7 z -1,153b -1,433b -,035b -1,083b -,454c -1,642c -1,363c -2,551b -2,970b p value ,249 ,152 ,972 ,279 ,650 ,101 ,173 ,011* ,003* a: deflected according to inflation (cpi: 2016). b: based on positive ranges. c: based on negative ranges. in figures 2 and 3, we can observe the evolution of clubs’ revenue and perceived market value respectively, and how there is a clear growing trend in all clubs, highlighting the scenario of psg and manchester city in terms of revenue, and atlético de madrid on perceived value of players. figure 2. evolution of club’s revenue during 9 seasons (2011-2019) (m€). pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 177 figure 3. evolution of perceived market value during 9 seasons (2011-2019) (m€). 5. conclusions there has been a statistically significant increase, both in club revenues they have grown 100.3% in constant € in 9 years and in the market value of players the average value of players has grown 97, 1% in this period. for a few years, revenue growth and staff valuation had a statistical relationship, as expected. however, in the last two seasons we did not find a statistical relationship between income and staff value, something that could be explained by the large price escalation in the market, with cases as significant as the psg signings, for € 222 and 188 million of neymar and mbappé, in the summer of 2017, which has caused an anchoring effect on the price of the costs of the players from that moment. it has also contributed to this lack of relationship between income and value of players the significant revaluation of the atlético de madrid squad, well above the growth of their income, being in the last 5 years the club with greater efficiency (pérez-gonzález, 2017) of european football. however, sánchez, barajas and sánchez fernández (2019) showed a clear increase in the percentage of expenditure in salaries over the club’s turnover, especially at france ligue 1, up to 2017 season. this authors also argued about the real and efficient application of the financial fair play, consisted on the regulation that controls the economic wealth, salaries and transfers in fifa and its governing bodies. nevertheless, there is a clear association between players’ salaries and transfers fees (kuper and szymanski, 2014). thus, to our knowledge, this is the first study that makes a relation between perceived market value and top clubs’ income for a long timeframe. our results suggest that it is becoming necessary that the leaders and decision-makers in the clubs take into account the assessment biases to avoid incurring them, being especially to find the optimal moment to make buying and selling decisions in the market; and think about the pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 178 improvement of their academies. however, more analysis about transfer market values and club’s income are needed in the coming years to check in the future that the differences identified in this study period remain in time. this paper shows a very incipient analysis, and only provides initial evidence of a possible pattern in future football. references buraimo, b., frick, b., hickfang, m., and simmons, r. (2015). the economics of long-term contracts in the footballers' labour market. scottish journal of political economy, 62(1). https://doi.org/10.1111/sjpe.12064 castillo, d., pérez-gonzález, b., raya-gonzález, j., fernández-luna, á., burillo, p., and lagorodríguez, á. (2019). selection and promotion processes are not associated by the relative age effect in an elite spanish soccer academy. plos one, 14(7), e0219945. https://doi.org/10.1371/journal.pone.0219945 deloitte (2011). football money league 2011: the untouchables. retrieved from https://www2. deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloittefootball-money-league-2011.pdf deloitte (2012). football money league 2012: fan power. retrieved from https://www2.deloitte .com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-footballmoney-league-2012.pdf deloitte (2013). football money league 2013: captains of industry. retrieved from https://www 2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-ukdeloitte-football-money-league-2013.pdf deloitte (2014). football money league 2014: all to play for. retrieved from https://www2. deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloittefootball-money-league-2014.pdf deloitte (2015). football money league 2015: commercial breaks, from https://www2.deloitte .com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-football-moneyleague-2015.pdf deloitte (2016). football money league 2016: top of the table. retrieved from https://www2. deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/uk-deloitte-sportfootball-money-league-2016.pdf deloitte (2017). football money league 2017: planet football. retrieved from https://www2. deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-sportfootball-money-league-2017.pdf deloitte (2018). football money league 2018: rising stars. retrieved from https://www2. deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-sbgdfml2018.pdf deloitte (2019). football money league 2019: bullseye. retrieved from https://www2. deloitte.com/uk/en/pages/sports-business-group/articles/deloitte-football-money-league.html fort, r. (2019). diferencias entre el deporte europeo y el norteamericano: una mirada retrospectiva. papeles de economía española, deporte y economía 159(abril), 61-71. garcía del barrio, p., and szymanski, s. (2009). goal! profit maximization and win maximization in football leagues. review of industrial organization, 34(1), 45-68. https://doi.org/10.1007/s11151-009-9203-6 garcía del barrio, p., and tena horrillo, j. d. d. (2019). inversión en talento y visibilidad mediática: un estudio del fútbol profesional en europa. papeles de economía española, deporte y economía 159(abril), 165-179. hall, s., szymanski, s., and zimbalist, a. s. (2002). testing causality between team performance and payroll: the cases of major league baseball and english soccer. https://doi.org/10.1177/152700250200300204 https://doi.org/10.1111/sjpe.12064 https://doi.org/10.1371/journal.pone.0219945 https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2011.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2011.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2011.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2012.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2012.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2012.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2013.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2013.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2013.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2014.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2014.pdf https://www2.deloitte.com/content/dam/deloitte/uk/documents/sports-business-group/deloitte-uk-deloitte-football-money-league-2014.pdf https://doi.org/10.1007/s11151-009-9203-6 https://doi.org/10.1177/152700250200300204 pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 179 herm, s., callsen-bracker, h. m., and kreis, h. (2014). when the crowd evaluates soccer players’ market values: accuracy and evaluation attributes of an online community. sport management review, 17(4), 484-492. https://doi.org/10.1016/j.smr.2013.12.006 késenne, s. (2015). the optimal competitive balance in a sports league? in p. rodríguez, s. késenne and r. koning (eds.), the economics of competitive sports (pp. 85-92). glos/massachusetts: edward elgar publishing. késenne, s. (2019). ¿cómo puede mejorarse el balance competitivo? papeles de economía española, deporte y economía 159(abril), 32-42. https://doi.org/10.4337/9781783474769.00013 kuper, s., and szymanski, s. (2014). soccernomics. london: harpercollins. lago, u., simons, r., and szymanski, s. (2006). the financial crisis in european football: an introduction. journal of sports economics, 7(1), 3-12. https://doi.org/10.1177/1527002505282871 lorenz, j., rauhut, h., schweitzer, f., and helbing, d. (2011). how social influence can undermine the wisdom of crowd effect. proceedings of the national academy of sciences, 108, 9020-9025. https://doi.org/10.1073/pnas.1008636108 mareque, m., barajas, a., and lópez-corrales, f. (2018). the impact of union of european football associations (uefa) financial fair play regulation on audit fees: evidence from spanish football. international journal of financial studies, 6(4), 92. https://doi.org/10.3390/ijfs6040092 müller, o., simons, a., and weinmann, m. (2017). beyond crowd judgments: data-driven estimation of market value in association football. european journal of operational research, 263(2), 611-624. https://doi.org/10.1016/j.ejor.2017.05.005 palacios-huerta, i. (2019). tras la senda de galileo: deporte como economía. papeles de economía española, deporte y economía 159(abril), 12-31. pawlowski, t., breuer, c., and hovemann a. top clubs’ performance and the competitive situation in european domestic football competitions. journal of sports economics, 11, 186202. https://doi.org/10.1177/1527002510363100 peeters, t. (2018). testing the wisdom of crowds in the field: transfermarkt valuations and international soccer results. international journal of forecasting, 34(1), 17-29. https://doi.org/10.1016/j.ijforecast.2017.08.002 pérez-gonzález, b. (2017). la resistencia a la regresión a la media. el caso del atlético de madrid, campeón en eficiencia económica desde 2014 / the resistence to regression toward the mean. the case of atlético de madrid, leader in economic efficiency since 2014. journal of sport and health research, 9(3), 357-368. plumley, d., ramchandani, g. m., and wilson, r. (2019). the unintended consequence of financial fair play: an examination of competitive balance across five european football leagues. sport, business and management, 9, 118-133. https://doi.org/10.1108/sbm-032018-0025 rodríguez, p. (2012). la economía del deporte. estudios de economía aplicada, 30(2), 387418. https://doi.org/10.25115/eea.v30i2.3533 sánchez, l. c., barajas, á., and sánchez-fernández, p. (2019). finanzas del deporte: fuentes de ingreso y regulación financiera en el fútbol europeo. papeles de economía española, deporte y economía 159(abril), 180-199. simmons, r., and berri, d. (2019). los mercados de trabajo en el deporte de equipo profesional. papeles de economía española, deporte y economía 159(abril), 148-164. solberg, h. a. (2016). the battle for media rights. in u. wagner, r. s. storm and k. nielsen (eds.), european club football. when sport meets business: capabilities, challenges, critiques (pp. 92-107): sage. https://doi.org/10.4135/9781473984226.n7 szymanski, s. (2015). money and football: a soccernomics guide (intl ed): why chievo verona, unterhaching, and scunthorpe united will never win the champions league, why manchester city, roma, and paris st. germain can, and why real madrid, bayern munich, and manchester united cannot be stopped. bold type books. szymanski, s. (2019). dominio y apuros. papeles de economía española, deporte y economía 159(abril), 92-108. https://doi.org/10.1016/j.smr.2013.12.006 https://doi.org/10.4337/9781783474769.00013 https://doi.org/10.1177/1527002505282871 https://doi.org/10.1073/pnas.1008636108 https://doi.org/10.3390/ijfs6040092 https://doi.org/10.1016/j.ejor.2017.05.005 https://doi.org/10.1177/1527002510363100 https://doi.org/10.1016/j.ijforecast.2017.08.002 https://doi.org/10.1108/sbm-03-2018-0025 https://doi.org/10.1108/sbm-03-2018-0025 https://doi.org/10.25115/eea.v30i2.3533 https://doi.org/10.4135/9781473984226.n7 pérez-gonzález et al. / european journal of government and economics 9(2), july 2020, 170-180 180 terviö, m. (2010). transfer fee regulations and player development. journal of the european economic association, 4(5), 957-987. https://doi.org/10.1162/jeea.2006.4.5.957 torgler, b., and schmidt, s. l. (2007). what shapes player performance in soccer? empirical findings from a panel analysis. applied economics, 39(18), 2355-2369. https://doi.org/10.1080/00036840600660739 transfermarkt. (2019) retrieved marzo, 2019, from https://www.transfermarkt.com/ varela-quintana, c., and del corral, j. (2019). economía del comportamiento en el deporte. papeles de economía española, deporte y economía 159(abril), 72-89. https://doi.org/10.1162/jeea.2006.4.5.957 https://doi.org/10.1080/00036840600660739 http://www.transfermarkt.com/ ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5953_g contemporary decision-making of mid-size city in the czech republic using multiple criteria decision-making european journal of government and economics 7(1), june 2018, 44-59. european journal of government and economics issn: 2254-7088 contemporary decision-making of mid-size city in the czech republic using multiple criteria decision-making meaza birhane hailea, *, martin mastalkaa a pardubice university, czech republic * corresponding author at: pardubice university, czech republic. email: meazabi@gmail.com article history. received 4 september 2017; first revision required 20 february 2018; accepted 28 may 2018. abstract. the strategic planning process has been implemented into the local governance environment in the czech republic during last two decades. but the strategic goals that are expressed in the strategic development documents on the local level are only the part of the issues that are to be fulfilled or solved by the local government. in the everyday practical governance there is always a list of projects that appear from the actual technological, technical, public or political demand. the contemporary decision-making process is based on the personal or collective political decision or on the actual technical demand. but there are also municipalities that are already implementing strategic planning and want to avoid unsystematic interventions and decision-making processes. they try to adopt some attitudes from the corporate sphere to make the decision-making process more open and clear. this paper deals with the methodology of the mid-size city and tries to discuss it and offer some improvements. the paper’s other goals are to offer comparison of typical projects that could be found on the local level and to make a model of results provided by the different methodologies used for the decision-making process, specifically weighted sum average and analytic hierarchy process. keywords. analytic hierarchy process; consistency index; public sector; saaty matrix; spearman’s correlation; strategic planning; weighted sum method. jel classification. c00; h00 1. introduction city governance is a very complicated issue combining many different needs of entities across many fields. local government is the closest to the public. in the central european environment inhabitants know their mayor and municipality employees usually in person. therefore the responsibility for the outputs and outcomes of management and city development is much higher than on other levels of governance. this close interaction of the local stakeholders is a positive contribution to the local development but include some threats as well. the group of decision-makers faces new challenges; and the threat of the conceptual lock-in and not absorbing new ideas and impulses from society (rydin, 2010), (barry, 2001). to prevent this situation the main goals of the local society are expressed in the strategic development documents. these documents are created in the cooperation of four main groups of actors in doi: https://doi.org/10.17979/ejge.2018.7.1.4332 mailto:meazabi@gmail.com m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 45 the city/region local government plus public authorities, private companies, universities and the public (pozoukidou et al. 2017) (robert et al. 2010). they set up the direction of the local development, define the goals and aims. the everyday work on the action level is up to the local government. as mentioned above the local government has to balance not only the strategic goals but also national and eu regulations, the goals of their regional policies and of course the duty of ordinary city maintenance. there are many attitudes to city management. the way implemented in central europe tends to the more open and clear way of governance (wolf and steven, 2013). the first step of the process was the implementation of principles of strategic planning and of project management (spee and paula, 2011). after more than 15 years of experience, some cities recapitulated the processes (shrader et al. 1984) and decided to move forward by adopting output-focused attitudes (šilhánková, 2011) and a clear and fair decision-making processes. the methodology introduced in this paper is the practical example of the czech city that is developing its internal processes more clearly and transparently. the city is collecting projects and project ideas, which are periodically evaluated and ranked by different departments of the city according to their responsibilities. the final list is an on-going document that shows the actual position of each project in the investment hierarchy. the purpose of this paper is to evaluate the impact of different methods of ranking to the position of the projects on the list. the hypothesis of the paper is the following assertion: the position of the projects chosen for practical realization or implementation in local government is affected by the methodology used for the evaluation. figure 1 shows the process flow of project selection at the local government of the city studied in this paper. figure 1. process flow of project selection source: own elaboration. m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 46 2. methodology in the past few decades the project selection task has been performed using multiple criteria decision-making (mcdm). in this paper, the project selection process of a mid-sized city administration using weighted sum average and analytic hierarchy process is discussed. the change in ranking of projects as the priorities of criteria, interval of points and decision-making methods vary, are also shown in the paper. • multiple criteria decision-making mcdm is one branch of decision-making that studies decision problems where the decision space is continues; there are many mcdm methods and each has its own characteristics. there are many ways to classify these methods based on the data used such as deterministic, stochastic and fuzzy or on the number of decision-maker involved (panos, 2000). there are well-known multiple criteria decision-making methods, of which: weighted sum method (wsm), analytic hierarchy process (ahp), analytic network process (anp), technique for order preference by similarity to ideal solution (topsis). wsm is the most straight forward and commonly applied method, the total value of an alternative is the total sum of each criteria weight multiplied by the value of the alternative (panos, 2000). analytic hierarchy process (ahp) is a pairwise comparison method that uses ratio scales. ahp has been and is still applied in solving different problems, as a methodology for facilities layout design problem (yang and kuo, 2003), utilizing the analytic hierarchy (ahp) in swot analysis (kurttila et al. 2000) is the most used method in publication (vaidya and kumar, 2006), (tramarico et al. 2015). there is also commonly used extension of ahp for fuzzy analysis, fuzzy ahp (yu-cheng and thomas, 2011). another method based on pairwise comparison ratio is anp, a generalization of ahp. this method is also used in different fields especially when a problem cannot be explained hierarchically, for instance, for selecting software products for a company (bayazit, 2006; diederik, 2007; haile, 2014; yuksel and dagdeviren, 2007). topsis is a method that measures the distance of alternatives from ideal and negative ideal solution and it is applied in solving different problems such as supplier selection (shahroudi and tonekaboni, 2012). many scholars have also researched and applied hybrid of these methods, machine selection by using ahp and topsis (karim and karmaker, 2016), for development of strategy on china’s rural drinking water supply (jia et al. 2016), application of the multi criteria decision-making methods for project selection (prapawan, 2015). in this paper, ahp and wsm are discussed. wsm was chosen for its simplicity; the ahp method was used in the paper to compare the results of the method with wsm to show consistency of results from the two methods and to provide an alternate method for decision-makers in situations where point evaluation for projects is not possible. m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 47 • analytic hierarchy process (ahp) ahp was originally developed by thomas l. saaty; it is a method that derives ratio scales from paired comparisons. input for the method can be obtained from actual measurement such as price, weight, cost, or from subjective opinions such as preference; it allows some inconsistency in judgment. the ratio scales are derived from the principal eigen vectors and the consistency index is derived from the principal eigen value. the ahp method consists of three levels of hierarchy. the first hierarchy level is the goal of the decision-making, the second level of hierarchy is how each of the existing criteria contributes to the goal achievement, and the last level of hierarchy is to find out how each of the alternatives contributes to each of the criteria. but the hierarchy could increase based on the type problem. ahp uses the 9-point rating of saaty (2001a; 2001b; 2005) for pairwise comparison. where the 'reciprocal matrix’ has all positive elements and has the reciprocal: 𝑠𝑠𝑖𝑖𝑖𝑖 = 1 𝑖𝑖 = 1,2, … , 𝑘𝑘 [1] 𝑠𝑠𝑗𝑗𝑖𝑖 = 1 𝑠𝑠𝑖𝑖𝑗𝑗 , 𝑖𝑖, 𝑗𝑗 = 1,2, … , 𝑘𝑘 ⁄ [2] table 1. the saaty's 9-points rating scale source: saaty (1996) intensity of importance definition explanation 1 equal importance two activities contribute equally to the objective 3 moderate importance experience and judgment slightly favour one activity over another 5 strong importance experience and judgment strongly favour one activity over another 7 very strong importance an activity is strongly favored and its dominance demonstrated in practice 9 absolute importance the evidence favoring one activity over another is of the highest possible order of affirmation. 2,4,6,8 intermediate values when compromise is needed reciprocal of above non-zero numbers if activity i has one of the above nonzero numbers assigned to it when compared with activity j, then j has the reciprocal value when compared with i. the criteria weights or variant values are then normalized by geometrical mean of lines of matrix s (saaty, 2008): m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 48 1/ 1 1/ 1 1 1, 2,..., k k ij j i k kk ij i j s v i k s = = =      = =       ∏ ∑ ∏ [3] experts provide weights of criteria or alternatives in comparison to other criteria or alternative using saaty’s 9-points rating scale of preference (saaty, 1996) shown in table 1. in performing pairwise comparison inconsistency could happen. for instance if there are three criteria and an expert evaluates that the first criterion is slightly more important than the second criterion and second criterion is slightly more important than the third criterion. inconsistency in pairwise comparison arises if the expert evaluates by mistake that the third criterion is equally or more important than the first criterion. a consistent evaluation would be that the first criterion is more important than the third criterion. ahp incorporates an effective technique for checking consistency of the evaluations made by experts. 𝐶𝐶𝐶𝐶 = 𝑥𝑥−𝑚𝑚 𝑚𝑚−1 , [4] where x is a scalar average of the elements of the vector whose jth element is the ratio of the jth element of the vector a·w to the corresponding element of the vector w. a perfectly consistent decision-maker obtains ci=0, but small values of inconsistency are tolerated (panos, 2000), (cliff, 1996): 𝐶𝐶𝐶𝐶 𝑅𝑅𝐶𝐶 < 0.1 [5] table 2. consistency index. source: cliff (1996) n 1 2 3 4 5 6 7 8 9 10 ri 0 0 0.58 0.9 1.12 1.24 1.32 1.41 1.45 1.49 • spearman’s coorelation coefficient spearman’s rank correlation is a nonparametric statistic that allows an investigator to describe the strength of an association between two variables x and y without making the more restrictive assumptions of the pearson product-moment correlation (r) (ruscio, 2008). 𝑟𝑟𝑠𝑠 = 1 − 6∑ 𝑑𝑑𝑖𝑖 2𝑛𝑛 1 𝑁𝑁(𝑁𝑁2−1) , [6] where rs is spearman’s correlation coefficient, di are the differences in the ranked scores on x and y for each pair of cases and n is the sample size. spearman’s correlation could be used in situations where: both variables are measured in ordinal scale, if one of the variables is m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 49 measured in ordinal scale but the other in interval or ratio scale, both variables are measured in interval or ration scale but the requirements for pearson’s correlation coefficient test are not met (liwen, 2001). in this paper, spearman’s rank correlation was used to test the correlation of orders of alternative projects in applying different methods of mcdm and different intervals of evaluation points. • weighted sum method (wsm) wsm is probably the most commonly used approach if there are m alternatives and n criteria then the best alternative is the one that satisfies most. the assumption that governs this model is the additive utility assumption. that is the total value of the alternative is equal to the sum of the products (panos, 2000). 𝐴𝐴𝑖𝑖𝑊𝑊𝑊𝑊𝑊𝑊 = ∑ 𝑤𝑤𝑗𝑗𝑎𝑎𝑖𝑖𝑗𝑗 , 𝑓𝑓𝑓𝑓𝑟𝑟 𝑖𝑖 = 1,2,3 … ,𝑚𝑚.𝑛𝑛 𝑗𝑗=1 [7] where wj denotes the relative weight of importance of the criterion cj and aij is the performance value of alternative ai when it is evaluated in terms of criterion cj. in this paper, selection of project and the change in ranking of projects with respect to weights and the interval of evaluation points are discussed using the wsm and ahp. these methods were selected instead of topsis and anp because of the way the data were represented. in order to apply topsis the evaluation points need to have positive values. the anp was not applied in this paper since there is no dependency among the crriteria set selected for the selection process. 3. strategic planning on the local level strategic planning was introduced in corporations in the 1950s and 1960s. influential scholars in strategic planning at the time included alfred chandler, philip selznick, igor ansoff and peter drucker. but it was used in the public sector a few decades later, one of the scholars who agues strongly about the importance of strategic planning in the public sector and nonprofit organizations is john m. bryson. the environment of public and nonprofit organizations has become not only increasingly uncertain in recent years but also more tightly interconnected; thus changes anywhere in the system reverberate unpredictably and often chaotically and dangerously throughout society. strategic planning can help leaders and managers of public and nonprofit organizations think, learn and act strategically (bryson, 2011). the social-political changes in the late 1980s and the early 1990s in eastern europe brought not only new freedom but also new challenges for citizens, companies and governance system. this paper is focused to the situation in the czech republic where the whole public sector was democratized; it was given the local power (and responsibility) to local governments and at the beginning of the new m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 50 millennium also to newly-established regional bodies. the enthusiasm for freedom and defiance to the strict socialistic planning led to abandonment of planning overall. in some cities years passed when the only valid plans were the budget (annual) and master plan (focused strictly to spatial division of functions in the city and restricting the building activities). such a situation was unsustainable. therefore cities started to use at least triennial budgets and economic outlooks. in the late 1990s the three-year outlook became insufficient for planning city development and “new” tools economic development plans – were adopted. but these documents were still focused on economic issues. the late 1990s in the czech republic were the years of rising economic and social problems when there was the demand for new attitudes of governance (haile et al. 2016). analysis and results projects selected and implemented at the local level have to be complemetary to the strategic plan. therefore it is important to use a systematic approach in selecting these projects. the criteria that are designed by the local government and used by the sample municipality for selecting projects are: • urgency, • availability of external sources, • economy, • readiness, • compliance with city strategic development plan, and • synergy with other projects. based on the professional experience and the discussion with other researchers we would enlarge the list with: • the compliance with other sectoral plans, • compliance with politiclal proclamations and • “public demand”. the description of the criteria is as follow: urgency – there are projects that are not a part of the strategic development documents but they would be fulfilled. typically it would be the reaction to disasters, unexpected problems with the technical infrastructure or other unexpected problems like terrorist attacks. availability of external sources – the investment reality in central and eastern europe has been linked to the huge amount of subsidies for the last decade. many local governments have invested only in projects that were supported by the eu or national subsidies. however, this is not an ideal attitude, the subsidies have to be linked to the eu or national policy. therefore the m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 51 local governments could have helped to fulfill the goals on these levels. anyway the availability of the external support is a very important issue during the decision-making project on the local level. economy – this point expresses the level of expected future mandatory expenditures of the municipality budget. this point would help to keep the long-time sustainable budgeting on the local level. readiness – especially “hard” investment projects in the czech republic are run for the longdistance. very complicated laws together with many stakeholders who are supposed and authorized to affect the process would extend the needed planning and permitting process to years. compliance with a strategic plan – the strategic plan would be the base for the development of each settlement and each project of the local government. although the quality of the documents is very variable, the main idea of the strategic plan as the main planning tool (together with master plan and budget) remains. compliance with other sectoral plans – there are the issues on the local level that have to be worked out more deeply (energy, social care, nature protection, climate change resilience/adaptation, security). therefore the sectoral plans are created. although they would be in compliance with the strategic plan, we recommend to keep them as a special point of the evaluation. compliance with political proclamations – as mentioned above, the strategic plan would be the main and crucial document for the development of the municipality. the goals expresseed in the strategic plans would be the goals of all the society, all stakeholders. the political representatives elected in the czech republic for four years usually declare their pririties in the declaration of their political goals and aims. “public demand” – this criterion is designed to be used in the exact city this paper is about. the municipality of the city has a long experience with strategic planning and evaluation of its impacts on the life of the city. the system of indicators has more than 100 items which are evaluated periodically. based on these indicators there could be expressed the level of “public demand”. however we would suggest to design the obligatory methodology for this criterion to avoid mistakes of interpretation or manipulation. synergy with other projects – several projects are linked to each other. as a typical example would serve reconversion of public spaces linked to the reconstruction of all the infrastructure below the surface. this is also the criterion that would be taken into acount. m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 52 the practical part of the paper is based on the modelling of the ranking of the typical projects appearing in the everyday life of local governance. the selection and design of typical projects is based on discussion with the representatives of the city and brainstorming of the experinced professionals. they have designed five typical projects that usually appear in the city´s everyday life. • the first project is the project of the infrastructure that is at the end of its lifetime period and urgently needs inventions. the project is administratevely prepared but not in compliance with the strategic development documents. • the second project is a typical project that is not urgent but is invovlved in the strategic documents, ready to implement and the government is waiting just for the financial sources. • the third project is the project that is not a part of any strategic document and is not needed operationally. the only reason for implementing this project is available resources (subsidy) and political will. such an attitude of governance is not appropriate but it can be seen in practise and therefore we have included it. • the fourth project is not urgent but is involved in all the long-term development documentation as well as supported by the public. • the fifth project is urgent and ready to implement, there are available external sources and it is neutral to the strategic development document. t the municipal representatives demanded the level of difference of the final ranking when using different ranking scales and weights. the weights used for this paper were split in a relatively balanced way. in the time when the paper was proposed, there were no final weights of the criteria established by the working group yet. outputs of the project were supposed to serve as support for this decision. it will be done regarding the impacts that the different methodology can cause. table 3. weighted sum method comparison with the same weight of criteria. criteria wght prj 1 prj 2 prj 3 prj 4 prj 5 urgency 0.11 0.22 -0.10 -0.10 0.00 0.22 availability of external sources 0.11 0.00 0.20 0.22 0.00 0.22 economy 0.11 -0.10 0.00 -0.10 0.00 0.00 readiness 0.11 0.22 0.10 0.11 0.22 0.22 compliance with strategic plan 0.11 -0.10 0.20 0.00 0.22 0.00 compliance with other sectorial plans 0.11 -0.10 0.10 0.00 0.22 0.00 compliance with political proclamation 0.11 0.22 0.10 0.22 0.22 0.00 “public demand” 0.11 0.00 0.10 0.11 0.22 0.00 synergy with other projects 0.11 0.00 0.10 0.00 0.00 0.00 wsm 0.33 0.90 0.44 1.11 0.67 rank 5 2 4 1 3 m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 53 project values discussed below in the tables were assigned values (-2.2) or (-5.5), the intervals show the compatibility of projects with respect to each criterion where -2 or -5 show that projects refered to as (prj) in the tables have least compatibility whereas 2 or 5 show most compatibility with the criteria. the two different evaluation intervals (-2.2) and (-5.5) were used to test if the intervals could result in a different rank of projects. the change of project ranking with respect to the change in weight is demonstrated using wsm and analytic hierarchy process methods. table 3 shows the rank of projects where all criteria have the same weight. the ranks in table 3 were calculated using wsm; that is to multiply the weight based on the results obtaind in the table project 4 (prj 4). the project that is not urgent but is involved in all the long-term development documentation is the first choice to add more values to the city and satisfy more criteria. project 1 (infrastructure) is the least favourable project to implement. if the weight is changed the rank of project will also change accordingly. table 4 shows the change in rank as the weight changes. table 4. weighted sum method comparison with different weights of criteria. criteria wght prj 1 prj 2 prj 3 prj 4 prj 5 urgency 0.20 0.40 -0.20 -0.20 0.00 0.40 availability of external sources 0.10 0.00 0.20 0.20 0.00 0.20 economy 0.20 -0.20 0.00 -0.20 0.00 0.00 readiness 0.20 0.40 0.20 0.20 0.40 0.40 compliance with strategic plan 0.05 -0.05 0.10 0.00 0.10 0.00 compliance with other sectorial plans 0.05 -0.05 0.05 0.00 0.10 0.00 compliance with political proclamation 0.10 0.20 0.10 0.20 0.20 0.00 “public demand” 0.05 0.00 0.05 0.05 0.10 0.00 synergy with other projects 0.05 0.00 0.05 0.00 0.00 0.00 wsm 0.70 0.55 0.25 0.90 1 rank 3 4 5 2 1 based on the new weight, project 5 urgent operational project which is ready to be implemented, is the first choice to add more values to the city and satisfy more criteria. project 3 a project that is not a part of any strategic document and is not needed operationaly is the least favorable project to implement. • analytic hierarchy process ahp was applied for the project selection process since the project selection problem can be represented hierarchically. figure 2 shows the hierarchy for project selection discussed in this paper. m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 54 figure 2. ahp for project selection. based on the points provided by experts, a pairwise comparison matrix was prepared. table 5 shows the pairwise comparison of projects for the first criterion, urgency. the total value of the projects was then calculated and ranked using 0.11 (weight 1) for all the criteria. table 5. comparison of projects using ahp. project 1 project 2 project 3 project 4 project 5 geomean weight project 1 1 7 7 5 1 3.01 0.08 project 2 0.14 1 1 0.33 0.14 0.37 0.01 project 3 0.14 1 1 0.33 0.14 0.37 0.01 project 4 0.20 3 3 1 0.14 0.76 0.02 project 5 1 7 7 5 1 3.01 0.08 based on the results in table 5, project 4 is not urgent but is involved in all the long-term development documentation is the first choice that will add more values to the city and satisfy more criteria. project 3 is not a part of any strategic document and is not needed operationally. is the least favorable project to implement. table 6. results of ahp comparison. criteria wght prj 1 prj 2 prj 3 prj 4 prj 5 urgency 0.11 0.04 0.01 0.01 0.01 0.04 availability of external sources 0.11 0.01 0.03 0.03 0.01 0.03 economy 0.11 0.01 0.03 0.01 0.03 0.03 readiness 0.11 0.03 0.01 0.01 0.03 0.03 compliance with strategic plan 0.11 0 0.04 0.01 0.04 0.01 compliance with other sector plans 0.11 0 0.03 0.01 0.06 0.01 compliance with political proclamation 0.11 0.03 0.01 0.03 0.03 0.01 “public demand” 0.11 0.01 0.02 0.02 0.05 0.01 synergy with other projects 0.11 0.02 0.05 0.02 0.02 0.02 sum 0.99 0.15 0.23 0.15 0.27 0.19 rank 4 2 5 1 3 ... urgency public demand project project 1 project 1 project 3 project 4 economy ... project 5 m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 55 the resulting ranks of projects obtained above by using different methods is then compared to see if there is a correlation. using spearman’s correlation, (equation 6). figure 3 shows the correlation between the ranks of projects obtained using wsm and ahp where the weight of all criteria is 0.11. figure 3. spearman’s correlation of results from wsm and ahp. based on the graph and the result calculated using spearman‘s correlation coefficient, 0.9, there is a strong similarity of results obtained using the two different methods. the same methods were used to compare the same projects with different intervals of project points. table 7 shows the evaluation of projects where weight is 0.11 for all criteria and the point interval is -5 to +5. according to table 7 the most effective project is project 4 while projects 1 and 2 are equally least desired. the results obtained from the interval -5 to +5 shows the mean of the results obtained from the two methods wsm and ahp with the interval -2 to +2. spearman‘s correlation coefficient was also used to compare the results obtained from the two intervals (-2.2 and -5.5). table 7. results of projects comparison with interval points [-5.5]. criteria wght prj 1 prj 2 prj 3 prj 4 prj 5 urgency 0.11 0.56 0.56 0.00 0.11 0.56 availability of external sources 0.11 0.11 0.11 0.56 0.11 0.56 economy 0.11 -0.11 -0.11 0.00 0.11 0.11 readiness 0.11 0.56 0.56 0.33 0.56 0.56 compliance with strategic plan 0.11 0.00 0 0.11 0.56 0.11 compliance with other sectorial plans 0.11 0.00 0.00 0.11 0.56 0.11 compliance with political proclamation 0.11 0.56 0.56 0.56 0.56 0.11 public demand 0.11 0.11 0.11 0.33 0.56 0.11 synergy with other projects 0.11 0.11 0.11 0.11 0.11 0.11 wsm 1.89 1.89 2.11 3.22 2.33 rank 4 4 3 1 2 0 1 2 3 4 5 6 0 2 4 6 correlation ahp w sm m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 56 conclusions the two methods discussed above, wsm and ahp, were used to analyze the same data set and the results were tested using spearman’s correlation which showed that there is a strong correlation between the results obtained from the two methods. these methods have their advantages and disadvantages. the obvious advantage of wsm is its simplicity to apply. and to incorporate posistive and negative values. the ahp is useful when only ordinal values are provided for evaluating projects and to compare identifying criteria. in this case a pairwise comparison could be used to compare alternatives and criteria. in this paper. the change in ranking of projects as the evaluation interval changes was also tested for correlation. figure 4. spearman’s correlation of results from interval [-5.5] and [-2.2] figure 4 shows the correlation among the results obtained from the interval -2.2 and -5.5. the calculated correlation coefficient also shows that there is 0.65 correlation among the two results. the evaluation and comparison of projects using ahp and wsm have shown that the results from the two methods are consistent therefore either method could be used for project selection based on the type of evaluation data obtained from experts. the outputs of the paper will serve local government as one of the sources used for the final internal methodology for the project-ranking in the city decision-making process. the presented objective methods helped to simulate different situations in the planning process. the results of the modelling have shown that the strictly mathematic attitude of weighted decision process brought to the non-model-fitting reality of the governance could affect the results of the process. the implementation of the model-selected projects could prevent the political prioritization of any projects but on the other it is very sensitive to its tuning. the weights and used methodology are clear and could be expressed publicly by the local government but could also cause the lack of responsibility of local politicians. their decisions could be perceived as a result only of a mathematical model. as any other tool used for the city/regional development. the presented methodologies could be very helpful but during their implementation we have to consider what is the level of mathematization and automatization we want to accept in our lives. 0 1 2 3 4 5 0 1 2 3 4 5 6 correlation interval -5,5 in te rv al -2 ,2 m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 57 references barry. a. (2001). political machines: governing a technological society. the athlone press. london. bayazit. o. (2006). use of analytic network process in vendor selection decisions. benchmarking: an international journal. 13(5). 566-579. bryson. j. (2011). strategic planning for public and nonprofit organizations: a guide to strengthening and sustaining organizational achievement. 4th edition. wiley. cliff. n. (1996). ordinal methods for behavioral data analysis. mahwah. nj: lawrence erlbaum associates. diederik. j. d. (2007). analysis of benefits. opportunities. costs and risks(bocr) with ahpanp: a critical validation. mathematical and computer modeling. 46.892-905. gang. k., daji. e., yi. p., and yong. s. (2013). data processing for the ahp/anp. springer heidelberg new york dordrecht london. haile. m. (2014). application of analytic network process for the assessment and evaluation of software product. 15th international conference of phd. students young scientists and pedagogues. nitra. 427-432. haile. m., krupka. j., and mastalka m. (2016). evaluation of strategic planning process using analysis of fuzzy integral. in 11th international scientific conference on distance learning in applied informatics. wolters kluwer. the netherlands. . 503-512. jia. x., ping. f., and peng. y. (2016). research of development strategy on china’s rural drinking water supply based on swot–topsis method combined with ahp-entropy: a case in hebei province. environ earth sci. 75 (58). karim. r., and karmaker. c. (2016). machine selection by ahp and topsis methods. american journal of industrial engineering. 4. 7-13. kurttila. m., pesonen. m., kangas. j., and kajanus. m. (2000). utilizing the analytic hierarchy (ahp) in swot analysis – a hybrid method and its application to a forest-certification case. forest policy and economics. 1. 41-52. https://doi.org/10.1016/s1389-9341(99)00004-0 liwen. v. (2001). statistical methods for the information professional: a practical. painless approach to understanding. using and interpreting statistics (children of the promise). information today. 140. panos. m.p. (2000). multi-criteria decision-making methods: a comparative study. kluwer academic publishers. the netherlands. pozoukidou. g., gavanas. n., and verani. e. (2017). sustainable management and strategic planning. basingstoke. taylor and francis ltd. retrieved from: https://www.witpress.com/elibrary/sdp/12/5/1522) prapawan. p. (2015). application of the multi criteria decision-making methods for project selection. universal journal of management. 3(1). 15-20. http://www.sciepub.com/ajie/content/4/1 https://doi.org/10.1016/s1389-9341(99)00004-0 m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 58 robert. a., anu. j., pasi. k., and tatu. p. (2010). exploring quadruple helix. university of tampere institute for social research work research centre. ruscio. j . (2008). constructing confidence intervals for spearman ’ s rank correlation with ordinal data: a simulation study comparing analytic and bootstrap methods. journal of modern applied statistical methods. 7 (2). 416-434. rydin. y. (2010). planning and the technological society: discussing the london plan. international journal of urban and regional research. 34(2). 243-259. https://doi.org/10.1111/j.1468-2427.2009.00901.x saaty. t.l. (1996). decision-making with dependence and feedback: the analytic network process. rws publications. pittsburgh. saaty. t.l. (1999). fundamentals of the analytic network process. in: isahp. kobe. japan. saaty. t.l. (2001a). decision-making with the analytic network process and its “super decisions’ software: the national missile defense (nmd) example. in: isahp 2001 proceedings. bern. switzerland. saaty. t.l.( 2001b). deriving the ahp 1-9 scale from first principles. in: isahp 2001 proceedings. bern switzerland. saaty. t.l.( 2005). theory and application of the analytic network process: decision-making with benefits. opportunities. costs and risks. rws publications. pittsburgh. saaty. t.l.( 2008). the analytic network process. iran j oper res. 1-27. shahroudi. k., and tonekaboni. s. (2012). application of topsis method to supplier selection in iran auto supply chain. journal of global strategic management. 6 (2). 123-131. https://doi.org/10.20460/jgsm.2012615779 shrader. c.b., lew. t., and dan. r.d. (1984). “strategic planning and organizational performance: a critical appraisal”. journal of management. 10(2). 149-171. https://doi.org/10.1177/014920638401000202 šilhánková. v. (2011). indicators of sustainable development for municipalities. hradec králové: civitas per populi. spee. a.p., and paula. j. (2011). strategic planning as communicative process. organization studies. 32(9). 1217-1245. tramarico. c.. et al. (2015). analytic hierarchy process and supply chain management: a bibliometric study. information technology and quantitative management. procedia computer science. 55. 441-450. vaidya. o., and kumar. s. (2006). analytic hierarchy process: an overview of applications. european journal of operational research. 169 (1). 1-29. https://doi.org/10.1016/j.ejor.2004.04.028 wolf. c., and steven. w.f. (2013). strategic planning research: toward a theory-driven agenda. journal of management. 43(6). 1754-1788. yang. t., and kuo. c. (2003). a hierarchical ahp/dea methodology for facilities layout design problem. european journal of operational research. 147 (1). 128-136. https://doi.org/10.1111/j.1468-2427.2009.00901.x https://doi.org/10.20460/jgsm.2012615779 https://doi.org/10.1177/014920638401000202 http://www.sciencedirect.com/science/journal/03772217/169/1 https://doi.org/10.1016/j.ejor.2004.04.028 m.b. haile and m. mastalka / european journal of government and economics 7(1), 44-59. 59 yu-cheng. t., and thomas. w.l. (2011). application of the fuzzy analytic hierarchy process to the lead-free equipment selection decision international. journal of business and systems research. 5 (1). 35-56. yuksel. i., and dagdeviren. m. (2007). using the analytic network process (anp) in a swot analysis – a case study for a textile firm. information sciences. 177. 3364-3382. https://doi.org/10.1016/j.ins.2007.01.001 https://doi.org/10.1016/j.ins.2007.01.001 abstract. the strategic planning process has been implemented into the local governance environment in the czech republic during last two decades. but the strategic goals that are expressed in the strategic development documents on the local level are... keywords. analytic hierarchy process; consistency index; public sector; saaty matrix; spearman’s correlation; strategic planning; weighted sum method. jel classification. c00; h00 the challenge of measuring poverty and inequality: a comparative analysis of the main indicators european journal of government and economics 7(1), june 2018, 24-43. european journal of government and economics issn: 2254-7088 the challenge of measuring poverty and inequality: a comparative analysis of the main indicators juan ignacio martín-legendrea, * a universidade da coruña, department of economics, elviña, 15071 a coruña, spain * corresponding author at: department of economics, universidade da coruña, elviña, 15071 a coruña, spain. email: ecojml04@udc.es article history. received 29 september 2017; first revision required 23 april 2018; accepted 8 may 2018. abstract. this paper presents a review of the main available indicators to measure poverty and income inequality, examining their properties and suitability for different types of economic analyses, and providing real-world data to illustrate how they work. although some of these metrics, such as the gini coefficient, are most frequently used for this purpose, it is crucially important for researchers and policy-makers to take into account alternative methods that can offer complementary information in order to better understand these issues at all levels. keywords. income inequality, indicators, inequality metrics, poverty metrics jel classification. d31, d63. introduction poverty and inequality have long been topics of interest in the economic literature, because of the concerns about an equitable distribution of the fruits of economic growth. however, before tackling the analysis of the causes and potential consequences of such phenomena, we have to face the issue of which is the best way to measure them. although they are inseparably connected, we should first distinguish between poverty and income inequality. while inequality is a much broader concept, since it focuses on the way income, or wealth or consumption, is distributed in an entire population, poverty focuses on the living conditions of the individuals placed in the lowest end of income distribution, below a threshold called ‘poverty line’. but for the measure of these two variables to be useful, it is desirable that they fulfil certain conditions. in this respect, on the one hand, regarding poverty measures, morduch (2006) mentions the following properties: scale invariance (also known as population size independence), which means that, if the number of individuals in the population is multiplied by a constant for all income levels, the results of the measurement should not change; focus, which implies that the indicator should only be focused on individuals living below a certain level of doi: https://doi.org/10.17979/ejge.2018.7.1.4331 mailto:ecojml04@udc.es j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 25 income, called the ‘poverty line’, so that an improvement or deterioration in the living conditions of those above this level of income should not change the results of the measurement; monotonicity, which means that if an individual living below the poverty line loses income, the results of the measurement should worsen, or at least not improve; transfer sensitivity, also known as pigou-dalton condition, proposed by pigou (1912) and dalton (1920), which implies that if there is a transfer of income from a richer household to a poorer one without changing their relative positions within the income distribution or their average income, the poverty measure must fall, and vice versa; and finally, an additional desirable property is that poverty measures can be decomposed according to different criteria, so that we can analyse the poverty level of different subgroups, being the sum of the subgroup indicators equal to the poverty level of the entire population. a key feature for a measure to be decomposable is that the sub-groups should not overlap and that together they should encompass the entire population. on the other hand, as for inequality measures, haughton and khandker (2009) point out that it is desirable they have as many of the following properties as possible: scale invariance and transfer sensitivity (both explained above); mean independence, which means that multiplying the income of all individuals in the population by a constant should not change the results of the measurement; symmetry or anonymity, which implies that if two individuals of the population exchange places within the distribution, the results of the measurement should not be altered; and, as additional properties, we could also mention decomposability, equivalent to the aforementioned property; fixed range, so that the measurement of inequality is performed on a scale varying between two fixed values, ideally 0 and 1; and statistical testability, which means that the researcher should be able to test for the significance of changes in the indicator over time. thus, splitting the set of indicators in poverty and inequality measures for reasons explained above, this paper is organized as follows: in section 1, we review the main poverty measures; in section 2, we study the most important inequality measures, and finally, in section 3, we present our concluding remarks. 1. poverty indicators 1.1 foster-greer-thorbecke class of measures foster, greer, and thorbecke (1984) developed a group of indicators in order to assess the living standards of individuals that are below the so-called “poverty line”. this threshold can be set by the researchers as a share of the mean or median income of a population, which would be a measure of “relative poverty”, e.g. eurostat sets it at 60% of the median equivalized disposable income after social transfers, or as an arbitrarily selected value, that is, a measure of “absolute poverty”, e.g. the world bank currently sets the “international poverty line” in 1.90 american dollars a day valued at 2011 purchasing power parity. the general expression of the fgt measures can be written as: j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 26 𝑃𝑃𝑎𝑎 = 1 𝑁𝑁 ∑ �𝐺𝐺𝑖𝑖 𝑧𝑧 � 𝑎𝑎 𝑁𝑁 𝑖𝑖=1 , (𝑎𝑎 ≥ 0), where 𝑁𝑁 is the number of individuals in the population, 𝐺𝐺𝑖𝑖 is the difference between the poverty line and the actual income of an individual (being 𝐺𝐺𝑖𝑖 = 0 for those above the poverty line), 𝑧𝑧 is the poverty line, and 𝑎𝑎 is a constant that represents the indicator sensitivity to poverty, i.e., it can take values from 0 to infinity, and by giving it higher and higher values to 𝑎𝑎, we can gradually increase the sensitivity of the indicator to poverty. there are three cases of the fgt measures that are so widely used that researchers designate them with specific names: 𝑃𝑃0 is called ‘headcount ratio’ or ‘at-risk-of-poverty rate’; 𝑃𝑃1 is known as ‘poverty gap index’; and 𝑃𝑃2 is referred to as ‘squared poverty gap index’ or ‘severity index’. the simplest and most popular way to assess the poverty level of a population is the headcount ratio, also known as at-risk-of-poverty rate, which measures the share of people with an equivalized disposable income below the poverty line. it can be calculated using the following formula: 𝑃𝑃0 = 𝑁𝑁𝑃𝑃 𝑁𝑁 , where 𝑁𝑁 is the number of individuals in the population, and 𝑁𝑁𝑝𝑝 is the number of them below the poverty line. using the headcount ratio, the highest levels of poverty in the eu-28 can be seen in romania, latvia, lithuania, spain and bulgaria, where poor households represent more than 20% of the population. at the other end of the scale, we have the czech republic, the netherlands, denmark, slovakia and finland, where this indicator lies somewhere around 12% (the iso codes corresponding to each eu country are in table 1 of the annex). it is also worth mentioning that only nine member states managed to reduce their poverty levels in the last decade (figure 1). although the headcount ratio offers an easy-to-interpret first glance to poverty measurement, it is a simple ratio that does not allow us to quantify the extent to which individuals fall below the poverty line. thus, it does not change if the living conditions of the poor improve or deteriorate as long as they remain below the poverty line. in order to address these flaws, the poverty gap index allows us to measure how far poor individuals fall below the poverty line, and it can be calculated as: 𝑃𝑃1 = 1 𝑁𝑁 ∑ �𝐺𝐺𝑖𝑖 𝑧𝑧 �𝑁𝑁 𝑖𝑖=1 , where 𝑁𝑁 is the number of individuals in the population, 𝐺𝐺𝑖𝑖 is the difference between the poverty line and the actual income of an individual (being 𝐺𝐺𝑖𝑖 = 0 for those above the poverty line), and 𝑧𝑧 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 27 is the poverty line. figure 1. headcount ratio in the eu-28 countries (2006, 2015 and period peak). source: own elaboration based on statistics from eurostat. notes: (i) countries appear ranked according to the highest at-riskpoverty rate in 2015. (ii) croatia's data correspond to 2010 instead of 2006; romania's data correspond to 2007 instead of 2006. also, if we want to increase the “sensitivity to poverty” of our indicator, we can use the squared pgi, also known as severity index, which is calculated by averaging the square of the distance to the poverty line of every individual below this line: 𝑃𝑃2 = 1 𝑁𝑁 ∑ �𝐺𝐺𝑖𝑖 𝑧𝑧 � 2 𝑁𝑁 𝑖𝑖=1 , where all the elements are just the same as in the case of 𝑃𝑃1. by squaring the 𝐺𝐺𝑖𝑖/𝑧𝑧 component of the poverty gap formula, we make the indicator more sensitive to changes in income of individuals far below the poverty line, i.e. distributionallysensitive. 1.2 sen and sen-shorrocks-thon indices in order to assess more dimensions of poverty with the same indicator, sen (1976) proposed a new metric that combines the relative number of poor people, their income level, and the income distribution within the group, which may arguably be considered his main contribution in the measurement of poverty. it can be calculated as: 0 5 10 15 20 25 30 ro lv lt es bg ee el hr it pt pl de uk ie mt cy lu be hu se si at fr fi sk dk nl cz 2006 peak 2015 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 28 𝑃𝑃𝑆𝑆𝑆𝑆𝑁𝑁 = 𝑃𝑃0 �1 − (1 − 𝐺𝐺𝐺𝐺𝑁𝑁𝐺𝐺𝑃𝑃) �̅�𝑥𝑝𝑝 𝑧𝑧 �, where 𝑃𝑃0 is the headcount ratio of the population, 𝐺𝐺𝐺𝐺𝑁𝑁𝐺𝐺𝑝𝑝 is the gini coefficient (see section 2.4) among the poor, �̅�𝑥𝑝𝑝 is the average income of the poor, and 𝑧𝑧 is the poverty line. shorrocks (1995) presented a modified version of the sen index, currently known as sen-shorrocks-thon index, which introduces in the calculation the poverty gap index and the gini coefficient of the poverty gap ratios for the entire population to better gauge poverty intensity. it can be expressed as: 𝑃𝑃𝑆𝑆𝑆𝑆𝑆𝑆 = 𝑃𝑃0𝑃𝑃1 𝑝𝑝(1 − ĝ𝑝𝑝) where 𝑃𝑃0 is the headcount ratio of the population, 𝑃𝑃1 𝑝𝑝 is the poverty gap applied only to those below the poverty line, and ĝ𝑝𝑝 is the gini coefficient of the poverty gaps of the poor. these variables allow researchers to track the source of the changes in poverty levels measured by the sst index in three basic dimensions: number of poor, the depth of their poverty, and income distribution among the poor. 1.3 watts index this last indicator, proposed by watts (1968), was the first distribution-sensitive poverty indicator. it can be calculated using the following formula: 𝑊𝑊 = 1 𝑁𝑁 ∑ [ln(𝑧𝑧) − ln (𝑞𝑞 𝑖𝑖=1 𝑥𝑥𝑖𝑖)], where 𝑁𝑁 is the number of individuals in the population, 𝑞𝑞 is the number of individuals below the poverty line, 𝑧𝑧 is the poverty line, and 𝑥𝑥𝑖𝑖 is the income level of a certain individual below the poverty line. by introducing logarithms, watts makes the indicator more sensitive to changes in the lowest end of the income distribution. this way, the indicator will improve the most when poorer individuals improve their living conditions. 2. inequality indicators the measurement of inequality can be made using the whole income distribution, which includes the income of every individual in the population or sample under analysis, or using quantiles (deciles, percentiles, etc.). the latter is not as informative as the former since they only consider specific segments of the distribution, while ignoring the rest. for this reason, it is important to clearly identify which type of measure is best suited to the needs of the researcher and to maintain consistency throughout the study. j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 29 2.1 income shares the simplest way to assess how income is distributed in a given population is dividing the observations of our sample in quantiles, e.g., quartiles, quintiles, deciles, percentiles, etc., and analyzing the evolution of the income share corresponding to each quantile over time. if the main focus of our analysis is the lowest end of the income distribution, we may choose to evaluate how the income share of the first decile or quintile has evolved over a certain period of time. for the eu-28 in 2015, the countries where households in the lowest end of the distribution had a greater share of equivalized disposable income were the czech republic, finland, slovenia, the netherlands and slovakia, where the income of the poorest 20% almost reached the 10% of the total income. while in the opposite situation, we can find romania, spain, latvia, greece and belgium, where the bottom 20% only amounted to about 6% of the total income. it also should be noted that only in nine out of twenty-eight countries, the income share of the poorest households has increased over the last decade. figure 2. income share by the first two deciles in the eu-28 countries (2006, 2015, and period peak). source: own elaboration based on statistics from eurostat. notes: (i) countries appear ranked according to the highest income share of the poorest 20% in 2015. (ii) croatia's data correspond to 2010 instead of 2006; romania's data correspond to 2007 instead of 2006. furthermore, in order to analyse the accumulation of income by the households at the highest end of the distribution, we may also use the income shares to quantify the share of total income in the hands of these households. the most widely used segments are the tenth decile and the hundredth decile, but to assess 0 2 4 6 8 10 12 cz fi nl si sk be se fr at dkmt lu hu ie cy de pl uk hr pt ee it lv el bg lt es ro 2006 peak 2015 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 30 the increasing importance of the “super rich”, we may want to focus on increasingly small segments of the top of the distribution, e.g., top 0.5%, top 0.1%, top 0.01%. as we can see in figure 3, the eu-28 countries where the top 1% has the larger share of the total income are romania, cyprus, lithuania, bulgaria and the united kingdom, while slovakia, slovenia, sweden, malta and finland are in the other end of the scale. there is no general trend in this regard among the member states, since sharp falls of the income share suffered by the top 1% in slovakia, ireland and hungary coincided in time with strong increases in countries such as romania, cyprus and latvia. figure 3. income share by the hundredth percentile in the eu-28 countries (2006, 2015, and period peak). source: own elaboration based on statistics from eurostat. notes: (i) countries appear ranked according to the highest income share by the richest 1% in 2015. (ii) croatia's data correspond to 2010 instead of 2006; romania's data correspond to 2007 instead of 2006. 2.2 quantile ratios with the purpose of assessing together the aforementioned segments of the income distribution, it may be useful quantifying the gap between the poorest and the richest households. for this purpose, we have at our disposal several ratios that are easy to construct and interpret. nevertheless, we should note that, even though these ratios are widely used, they do not measure inequality properly since they are calculated without taking into account the central segment of the distribution, so they can be considered income polarization indicators. the two most well-known ratios are: (i) the s80/s20 ratio, defined as the ratio of the richest 20% of the population’s share in gross total income, divided by the poorest 20% of the 0 1 2 3 4 5 6 7 8 9 ro cy lt bg uk dk el fr de pt lu lv nl it at ie es pl ee hu cz be hr mt fi se si sk 2006 peak 2015 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 31 population’s share, and used by the united nations development programme human development indicators; and (ii) the palma ratio, developed by the chilean economist gabriel palma, and defined as the ratio of the richest 10% of the population’s share in gross total income, divided by the poorest 40% of the population’s share. palma (2011) proposed using these two particular segments since there is evidence that in most countries the central segment of the income distribution amounts to about 50% of total income while the other 50% is distributed between the top 10% and the bottom 40%. considering that the way this half of the total income is distributed between these two segments varies greatly among countries and over time, this ratio can be extremely useful to track changes in income polarization over time, or to compare income distribution among countries or regions. as we can see in figures 4 and 5, both ratios show similar outcomes since their correlation coefficient for this period is 96.40%. the member states with the highest income polarization are lithuania, romania and bulgaria for both indicators, while slovakia, slovenia, czech republic, finland and sweden are the least polarized countries in terms of income. these results could be complemented with the data presented in figures 2 and 3 to try to establish whether the source of the changes experienced by these ratios is in the highest or lowest end of the income distribution, or both. figure 4. palma ratio in the eu-28 countries (2006, 2015, and period peak). source: own elaboration based on statistics from eurostat. notes: (i) countries appear ranked according to the highest palma ratio in 2015. (ii) croatia's data correspond to 2010 instead of 2006; romania's data correspond to 2007 instead of 2006. 0 0,2 0,4 0,6 0,8 1 1,2 1,4 1,6 1,8 2 lt ro bg lv ee es cy el pt uk it pl hr de fr ie lu mthu dk at nl be cz fi se si sk 2006 peak 2015 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 32 figure 5. s80/s20 ratio in the eu-28 countries (2006, 2015, and period peak). source: own elaboration based on statistics from eurostat. notes: (i) countries appear ranked according to the highest s80/s20 ratio in 2015. (ii) croatia's data correspond to 2010 instead of 2006; romania's data correspond to 2007 instead of 2006. however, there are many other ratios, such as p90/p10, p90/p50 and p50/p10, than can be used to assess the gap between certain segments of a given population. the p ratios are calculated by dividing the incomes at the respective percentiles, rather than the share of income of all those higher or lower than that percentile as in the s ratios. the p ratios therefore have the advantage of being easier to calculate and are quite insensitive to the data-missingness that is more common in the tails of the income distribution. for instance, the p90/p10 that, similarly to the ratios commented before, measures the gap between the highest and lowest ends of the distribution, and, needless to say, it will give results highly correlated with those ratios. moreover, the p90/p50 ratio is used to appraise the gap between the highest income individuals and the median income of the population, whereas the p50/p10 ratio is employed to gauge the divergence of the poorest households from the median income of their population. 2.3 measures of statistical dispersion: squared coefficient of variation (scv) and relative mean deviation (rmd) the following measures are not designed to analyse the level of inequality in a distribution of income; they are indicators used to assess the variability of any set of observations with regard to their average. that is the reason why these general statistics, despite not having been designed 0 1 2 3 4 5 6 7 8 9 ro bg el ee it cy pl ie fr mt at nl si sk 2006 peak 2015 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 33 specifically to analyse the degree of inequality in income distribution, can be used to quantify the dispersion of an income distribution, so that a higher level of dispersion would also mean higher inequality. first, we have the squared coefficient of variation (scv), which is a variant of the coefficient of variation, a measure of dispersion that can be used for any data set. it fulfills all the requirements explained in the introduction, except the additive decomposability and the fixed range, it can take values from 0 to infinity. as will be explained in section 2.6, this indicator is more sensitive to changes at the highest end of the income distribution. for its calculation, the formula is the following: 𝑆𝑆𝑆𝑆𝑆𝑆 = �𝜎𝜎 𝜇𝜇 � 2 , where σ is the standard deviation and µ is the arithmetic mean. as a second measure of dispersion, we can use the relative mean deviation (rmd), which was developed by schutz (1951), and represents the percentage of income that should be transferred from those with higher-than-average income to those with lower-than-average income, so that both groups have exactly the same average income (kakwani, 1980). it can be calculated according to the following formula: 𝑅𝑅𝑅𝑅𝑅𝑅 = 1 𝑁𝑁∑ |𝑥𝑥𝑖𝑖−�̅�𝑥|𝑁𝑁 𝑖𝑖=1 |�̅�𝑥| , where 𝑁𝑁 is the number of individuals in the population, 𝑥𝑥𝑖𝑖 is the income level of a given individual, and �̅�𝑥 is the average income of the population. the main problem with this measure is its insensitivity to transfers between individuals in the same side of the average income. 2.4 the lorenz curve and the gini coefficient the gini coefficient is based on the lorenz curve, which is a graphical representation of a cumulative distribution function, and is mathematically defined as the cumulative share of total income assumed by cumulative shares of the population. the lorenz curve is always represented paired with the line of egalitarian income distribution, that is the 45-degree line, and represents an ideal situation where every individual in the population has the same income level. this way, we can easily compare how far the lorenz curve is from this line of absolute equality. so, graphically, the gini coefficient is defined as the ratio of the area between the line of complete equality and a given lorenz curve, and the total area under the line of perfect equality. in figure 6, it can be calculated as a/(a+b). due to its comparability between regions and through time, regardless of population sizes, exchange rates, price levels, etc., and its easy-to-interpret results, which always range between 0 (“perfect equality”) and 1 (“perfect inequality”), the gini coefficient is the most widely used indicator to measure inequality. j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 34 assuming that the lorenz curve is a finite discrete function, the area between the perfect equality line and the lorenz curve can be approximated as a frequency polygon using the following formula (abounoori and mccloughan, 2003): 𝐺𝐺 = 1 − ∑ (𝑥𝑥𝑖𝑖 − 𝑥𝑥𝑖𝑖−1)(𝑦𝑦𝑖𝑖 − 𝑦𝑦𝑖𝑖−1)𝑁𝑁 𝑖𝑖=1 , where n is the number of intervals into which the population is divided, 𝑥𝑥𝑖𝑖 is the cumulative share of income, and 𝑦𝑦𝑖𝑖 is the cumulative share of population. besides, given that the lorenz curve is a twice-differentiable, monotonic increasing and convex function 𝐿𝐿(𝑥𝑥) where 𝑥𝑥 is the cumulative share of income, we can calculate the gini coefficient this way: 𝐺𝐺 = 1 − 2� 𝐿𝐿(𝑥𝑥)𝑑𝑑𝑥𝑥 1 0 figure 6. lorenz curve for the eu-28 (2015). source: own elaboration based on statistics from eurostat. according to the data presented in figure 7, the most unequal eu-28 countries are, once again, lithuania, romania, bulgaria and latvia, whereas the most egalitarian are slovakia, slovenia, the czech republic and sweden. these results are extremely correlated to the palma and s80/s20 ratios, with coefficients of correlation of 99.19% and 96.78%, respectively. nonetheless, the relative increases and decreases (i.e., in terms of percentages) are far less pronounced in the gini coefficient than in the ratios. this behaviour may be related to the fact that the gini coefficient is more sensitive to changes at the center of the distribution, while the ratios focus exclusively on what happens at its ends. this situation may lead to the researchers more interested in income polarization to use the aforementioned ratios rather than the gini coefficient as inequality indicators, 0 10 20 30 40 50 60 70 80 90 100 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10 % o f c um ul at iv e po pu la tio n % of cumulative income actual income distribution egalitarian income distribution a b j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 35 considering their almost perfect correlation. additionally, the gini coefficient is unable to differentiate between two populations where the area under the lorenz curve is the same, but the shape of the curve is different, i.e. they have different inequality patterns, and it is completely unresponsive to structural demographic changes. finally, the gini coefficient is not easily decomposable as the sum of the gini indices of different subgroups. nonetheless, many techniques for its decomposition have been proposed over the years (pyatt, 1976; lerman and yitzaki, 1985; silber, 1989). figure 7. gini coefficient of equivalized disposable income in the eu-28 countries (2006, 2015, and period peak). source: own elaboration based on statistics from eurostat. notes: (i) countries appear ranked according to the highest gini coefficient in 2015. (ii) croatia's data correspond to 2010 instead of 2006; romania's data correspond to 2007 instead of 2006. 2.5 the hoover index this indicator is closely associated to the gini coefficient because the lorenz curve is also used for its calculation. originally created as a measure of industrial localization (hoover, 1936), it represents the share of income that should be redistributed to attain a hypothetical situation of complete equality, that is why it is also commonly referred as the “robin hood index”. it can be graphically represented as the maximum vertical distance between a given lorenz curve and the 45-degree line of perfect equality (figure 8), and for its calculation we have to use the following formula after dividing the income distribution into quantiles: 𝐻𝐻 = 1 2 ∑ � 𝑆𝑆𝑖𝑖 𝑆𝑆𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 − 𝐴𝐴𝑖𝑖 𝐴𝐴𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡𝑡 �𝑁𝑁 𝑖𝑖=1 , 0 5 10 15 20 25 30 35 40 45 lt ro bg lv ee es el pt cy it uk hr pl de ie fr lu humt dk at nl be fi se cz si sk 2006 peak 2015 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 36 where 𝑁𝑁 is the number of quantiles, 𝐴𝐴 is the width of said quantiles, 𝐸𝐸𝑖𝑖 is the income level of a given quantile, 𝐴𝐴𝑖𝑖 is the number of individuals in the quantiles. although it provides little information about how income is distributed in a population, it can be used to illustrate how far a population is from the egalitarian distribution. figure 8. lorenz curve and hoover index for the eu-28 (2015). source: own elaboration based on statistics from eurostat. 2.6 generalized entropy measures: theil index and mean log deviation (mld) the theil index and the mean log deviation (mld) are special cases for the generalized entropy index (ge), an indicator originated in information theory, and developed by henry theil (1967). the ge can be calculated using the following formula (haughton and khandker, 2009): 𝐺𝐺𝐸𝐸(𝑎𝑎) = 1 𝑁𝑁𝑁𝑁(𝑁𝑁−1) ∑ ��𝑥𝑥𝑖𝑖 �̅�𝑥 � 𝑁𝑁 − 1�𝑁𝑁 𝑖𝑖=1 , (𝑎𝑎 ≥ 0), where 𝑁𝑁 is the number of individuals in the population, 𝑥𝑥𝑖𝑖 is the income level of a given individual, �̅�𝑥 is the average income of the population, and 𝛼𝛼 is the weight for distances between incomes in different parts of the income distribution. this last parameter allows the researcher to adjust the index sensitivity to their preferences, since for values below (above) 1 ge is more sensitive to changes in the lowest (highest) end of the distribution. for 𝑎𝑎 = 1, it applies equal weights across the income distribution. as the poverty indices presented in section 1, there are three cases of the ge that are so 0 10 20 30 40 50 60 70 80 90 100 d1 d2 d3 d4 d5 d6 d7 d8 d9 d10 % o f c um ul at iv e po pu la tio n % of cumulative income actual income distribution egalitarian income distribution 21.6% j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 37 widely used that researchers designate them with specific names: when 𝛼𝛼 = 0, the generalized entropy index is the mean log deviation; when 𝑎𝑎 = 1, it is the theil index; and when 𝑎𝑎 = 2, it is half the squared coefficient of variation (see section 2.3). although these indicators can take values from zero to infinity, and therefore, they do not fulfill the fixed range requirement, they are easily decomposable, allowing both the segmentation of the income distribution according to different criteria, and the disaggregation of total inequality in between and within group components. the decomposability of these indicators allows us to analyse the evolution of inequality patterns over the reference period using many segmentation criteria provided by the household finances surveys. the theil index, also known as theil’s t, can be calculated using the following formula: 𝐺𝐺𝐸𝐸(1) = 𝑇𝑇 = 1 𝑁𝑁 ∑ �𝑥𝑥𝑖𝑖 𝑥𝑥 𝑙𝑙𝑙𝑙 𝑥𝑥𝑖𝑖 𝑥𝑥 �𝑁𝑁 𝑖𝑖=1 , where 𝑁𝑁 is the number of individuals in the population, 𝑥𝑥𝑖𝑖 is the income level of a given individual, and 𝑥𝑥 is the average income of the population. it can be decomposed as the following sum (haughton and khandker, 2009): 𝑇𝑇 = ∑ 𝑠𝑠𝑖𝑖𝑇𝑇𝑖𝑖 + ∑ 𝑠𝑠𝑖𝑖𝑙𝑙𝑙𝑙 𝑥𝑥𝑖𝑖 𝑥𝑥 𝑚𝑚 𝑖𝑖=1 𝑚𝑚 𝑖𝑖=1 , where 𝑚𝑚 is the number of subgroups, 𝑠𝑠𝑖𝑖 is the share of total income of each subgroup, 𝑇𝑇𝑖𝑖 is the theil index of each subgroup, 𝑥𝑥𝑖𝑖 is the average income of each subgroup, and 𝑥𝑥 is the average income of the population. the first term of the expression above is the weighted sum of the theil indices calculated for the different subgroups, where the weights are given by each subgroup’s share on total income. this term represents the component of inequality attributed to income differences within the same group. the second term is the theil index corresponding to a distribution in which each individual receives the average income of their subgroup. this component then represents the income inequality between subgroups of the population. on the other hand, the mean logarithmic deviation (mld), also known as theil’s l, is the percentage of income difference between a randomly selected individual or household of a certain population and the average income of the population. we can calculate it in this way: 𝐺𝐺𝐸𝐸(0) = 𝑅𝑅𝐿𝐿𝑅𝑅 = 1 𝑁𝑁 ∑ 𝑙𝑙𝑙𝑙 𝑥𝑥 𝑥𝑥𝑖𝑖 𝑁𝑁 𝑖𝑖=1 , where 𝑁𝑁 is the number of individuals in the population, 𝑥𝑥𝑖𝑖 is the income level of a given individual, and 𝑥𝑥 is the average income of the population. j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 38 it can be decomposed as the following sum: 𝐿𝐿 = ∑ 𝑛𝑛𝑖𝑖 𝑁𝑁 𝐿𝐿𝑖𝑖𝑚𝑚 𝑖𝑖=1 + 𝑛𝑛𝑖𝑖 𝑁𝑁 ∑ 𝑙𝑙𝑙𝑙 �𝑥𝑥𝚤𝚤��� �̅�𝑥 �𝑚𝑚 𝑖𝑖=1 , where 𝑁𝑁 is the number of individuals in the population, 𝑚𝑚 is the number of subgroups, 𝑙𝑙𝑖𝑖 is the number of individuals in each subgroup, 𝐿𝐿𝑖𝑖 is the mld of each subgroup, 𝑥𝑥𝑖𝑖 is the average income of each subgroup, and 𝑥𝑥 is the average income of the population. similarly to the theil index, the first term represents the inequality within subgroups, while the second represents the inequality between subgroups. using the theil index, i.e. ge(1), the most unequal eu-28 countries were belgium, latvia, greece, lithuania and portugal. the results are almost identical if we focus on the changes in the lowest end of the income distribution, using the mld, i.e. ge(0). nonetheless, using the scv, i.e. twice the ge(2), that focus on the highest end, we can perceive bigger changes: countries like france or cyprus appear for the first time among the most unequal member states. figure 9. generalized entropy measures for the eu-28 countries, 2012. source: own elaboration based on statistics from the european commission. note: countries appear ranked according to the highest theil index. all three measures in figure 9 show high levels of correlation. but, as one might expect, mld and scv are the ones less correlated (76.27%), whilst the correlation coefficients of the other two pairs are above 90%, theil and mld, 94.92%; theil and scv, 91.99% (see table 2 in annex). 0,0 0,1 0,1 0,2 0,2 0,3 0,3 0,4 0,4 bg el pt ro ee it lu hr uk mt dk fi cz si theil (α=1) scv (α=2) mld (α=0) j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 39 2.7 atkinson class of measures similar to the generalized entropy measures, this group of statistics developed by anthony atkinson (1970) allows the researcher to calibrate the indicator’s sensitivity to inequality by giving values to a theoretical constant ε that measures the "inequality aversion level". it can be calculated using the following formula (litchfield, 1999): 𝐴𝐴(𝜀𝜀) = 1 − �1 𝑁𝑁 ∑ �𝑥𝑥𝑖𝑖 𝑥𝑥 � 1−ε 𝑁𝑁 𝑖𝑖=1 � 1 1−ε (𝜀𝜀 ≥ 0), where 𝑁𝑁 is the number of individuals in the population, 𝑥𝑥𝑖𝑖 is the income level of a given individual, 𝑥𝑥 is the average income of the population, and 𝜀𝜀 is the inequality aversion level, also known as atkinson constant. for increasingly higher values of the constant, the atkinson index becomes more and more sensitive to changes in the lowest end of the distribution. by doing so, if we calculate the atkinson index for different levels of inequality aversion, we can determine if the changes in income inequality in a certain population are being driven more by changes at the top or at the bottom of the distribution. to understand how it works, we can compare the different results of the index for some values of the constant. as we can see in figure 10, for a low level inequality aversion, 𝜀𝜀 = 0.5, the most unequal eu countries are bulgaria, latvia, greece, portugal and lithuania, while the most egalitarian are slovenia, slovakia, the czech republic, the netherlands and sweden. if we move towards a neutral level of inequality aversion, 𝜀𝜀 = 1, lithuania is replaced by romania as one of the countries with higher inequality, while finland does the same with sweden in the other group. finally, for a high level of inequality aversion, 𝜀𝜀 = 2, we can see major changes in the first group: austria, spain and italy enter the "most unequal group", which is also made up of greece and romania. at the other end, nevertheless, the czech republic, slovenia, the netherlands and finland remain as the most egalitarian member states, the entry of denmark is the only significant change in this group. these divergences become even more evident if we calculate the correlation coefficient of the atkinson index for the three 𝜀𝜀 values chosen. a(0.5) and a(1) have a 98.64% correlation, whilst a(1) and a(2) show a correlation of 73.32%, and a(0.5) is only 64.94% correlated with a(2) (see table 2 in annex). j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 40 figure 10. atkinson indices for the eu-28 countries (2012). source: own elaboration based on statistics from the european commission. note: countries appear ranked according to the highest atkinson index (ε=1). finally, it should be noted that this indicator is the only measure analysed in this study that fulfills all the conditions presented in the introduction. 3. concluding remarks the purpose of this study is summarizing and reviewing the most widely used poverty and inequality indicators, weighing up their advantages and disadvantages. as we explain in the introduction, it is a critical issue for researchers and policy-makers to know and use these indicators in order to target, analyse and correct both poverty and inequality. since every indicator or group of indicators, presented in this paper provides complementary information, they should be used in conjunction with others for the purpose of having the best possible overall picture of the circumstances in a certain population. researchers should choose the indicators they will use considering their needs and the information that each one can provide. for instance, if we want to focus on the living conditions of the poor, we should choose the fgt class of measures to quantify the number of households below the poverty line and their distance to such threshold. but if we want to have a better understanding of the income distribution among the poor, only one fgt measure would be useful: the severity index, which measures the distribution of income among individuals below the poverty line using the square of the coefficient of variation. a second measure that could be useful for this purpose is the sst index, which takes into account this dimension of the problem by including the gini index as one of its components. moreover, if we want to measure the gap between the rich and the poor, or the distance between these groups and the median household, we should use the quantile ratios since they 0,0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 bg ro pt lt ee cy ie at lu hu dk se fi si ε=1 ε=2 ε=0.5 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 41 are the most suitable indicators for measuring income polarization. additionally, considering their greater variability and their high correlation with the gini coefficient, we could choose them as proxies of income inequality. conversely, if our focus is on income inequality for an entire population, we should use many of the aforementioned measures, but always taking into account the problems they have: although the gini coefficient is the most widely used indicator to this end due to its simple interpretation and its comparability over time and across countries, we must also bear in mind that it is relatively insensitive to changes in the ends of the distribution, it cannot distinguish inequality patterns and it cannot be decomposed; the hoover index provides little information about the way income distributes in a population and should be only used as a first glance for this issue; the generalized entropy measures allow to adjust its sensibility to poverty and are decomposable, which makes them the ideal choice for unraveling the patterns of inequality according to several criteria, but they are not easily comparable since they can theoretically take values from zero to infinity; finally, the atkinson measures suffer none of the drawback listed above, but nevertheless they are relatively little used, so there are scarce data available of them. regarding data availability on these issues, it should be noted that there are several databases where we can easily download normalized macro data for research purposes, eurostat, the world bank, the united nations university world institute for development economics research, the oecd database, the luxembourg income survey, the world wealth and income database, or the standardized world income inequality database (solt, 2009). however, we should also note that these sources only offer time series for selected variables: headcount index, income shares and quantile ratios, and gini coefficient, leaving aside the rest of indicators commented in this paper. lastly, we must bear in mind that the data available on these subjects have many limitations owing to their sources. almost every data source on income distribution comes from household surveys that involve issues such as an ever-growing unit and item non-response rate, and an increasingly large measurement error due to less accurate responses provided by the respondents (meyer et al., 2015). references abounoori, e., & mccloughan, p. (2003). a simple way to calculate the gini coefficient for grouped as well as ungrouped data. applied economics letters, 10(8), 505-509. https://doi.org/10.1080/1350485032000100279 allison, p. d. (1978). measures of inequality. american sociological review, 43(6), 865-880. https://doi.org/10.2307/2094626 atkinson, a. b. (1970). on the measurement of inequality. journal of economic theory, 2(3), 244-263. https://doi.org/10.1016/0022-0531(70)90039-6 https://doi.org/10.1080/1350485032000100279 https://doi.org/10.2307/2094626 https://doi.org/10.1016/0022-0531(70)90039-6 j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 42 dalton, h. (1920). the measurement of the inequality of incomes. economic journal, 30(119), 348-361. https://doi.org/10.2307/2223525 eurostat. http://ec.europa.eu/eurostat foster, j., greer, j. & thorbecke, e. (1984). a class of decomposable poverty measures. econometrica, 52(3), 761-766. https://doi.org/10.2307/1913475 haughton, j. h. & khandker, s. r. (2009). handbook on poverty and inequality. washington, d.c.: world bank publications. hoover, e. m. (1936). the measurement of industrial localization. the review of economic statistics, 18(3), 162-171. https://doi.org/10.2307/1927875 kakwani, n. c. (1980). income inequality and poverty. new york: world bank. lerman, r. i. & yitzhaki, s. (1985). income inequality effects by income source: a new approach and applications to the united states. review of economics and statistics, 67(1), 151-156. https://doi.org/10.2307/1928447 litchfield, j. a. (1999). inequality: methods and tools. world bank. meyer, b. d., mok, w. k. & sullivan, j. x. (2015). household surveys in crisis. the journal of economic perspectives, 29(4), 199-226. https://doi.org/10.1257/jep.29.4.199 morduch, j. (2006). concepts of poverty. in united nations statistics division (ed.), handbook on poverty statistics: concepts, methods and policy use (pp. 23-50). new york: united nations. palma, j. g. (2011). homogeneous middles vs. heterogeneous tails, and the end of the 'inverted‐u': it's all about the share of the rich. development and change, 42(1), 87-153. https://doi.org/10.1111/j.1467-7660.2011.01694.x pigou, a. c. (1912). wealth and welfare. london: macmillan. pyatt, g. (1976). on the interpretation and disaggregation of gini coefficients. economic journal, 86(342), 243-255. https://doi.org/10.2307/2230745 schutz, r. r. (1951). on the measurement of income inequality. american economic review, 41(1), 107-122 sen, a. (1976). poverty: an ordinal approach to measurement. econometrica, 44(2), 219-231. https://doi.org/10.2307/1912718 shorrocks, a. f. (1995). revisiting the sen poverty index. econometrica, 63(5), 1225-1230. https://doi.org/10.2307/2171728 silber, j. (1989). factor components, population subgroups and the computation of the gini index of inequality. review of economics and statistics, 71, 107-115. https://doi.org/10.2307/1928057 solt, f. (2009). standardizing the world income inequality database. social science quarterly, 90(2), 231-242. https://doi.org/10.1111/j.1540-6237.2009.00614.x theil, h. (1967). economics and information theory. amsterdam: north-holland watts, h. (1968). an economic definition of poverty. in d. p. moynihan (ed.), on understanding poverty (pp. 316-329). new york: basic books. https://doi.org/10.2307/2223525 http://ec.europa.eu/eurostat https://doi.org/10.2307/1913475 https://doi.org/10.2307/1927875 https://doi.org/10.2307/1928447 https://doi.org/10.1257/jep.29.4.199 https://doi.org/10.1111/j.1467-7660.2011.01694.x https://doi.org/10.2307/2230745 https://doi.org/10.2307/1912718 https://doi.org/10.2307/2171728 https://doi.org/10.2307/1928057 https://doi.org/10.1111/j.1540-6237.2009.00614.x j. i. martín-legendre / european journal of government and economics 7(1), 24-43. 43 appendix 1 table a.1. iso codes for each eu country. code country at austria be belgium bg bulgaria cy cyprus cz czech republic de germany dk denmark ee estonia el greece es spain fi finland fr france hr croatia hu hungary ie ireland it italy lt lithuania lu luxembourg lv latvia mt malta nl netherlands pl poland pt portugal ro romania se sweden si slovenia sk slovakia uk united kingdom table 2. correlation coefficients of several inequality measures for the eu-28 countries in 2012. source: own elaboration based on statistics from the european commission. gini mld scv theil at0.5 at1 at2 gini 10.000 0.9761 0.8364 0.9821 0.9952 0.9772 0.6213 mld 0.9761 10.000 0.7627 0.9493 0.9860 0.9999 0.7422 scv 0.8364 0.7627 10.000 0.9199 0.8477 0.7644 0.3576 theil 0.9821 0.9492 0.9199 10.000 0.9870 0.9501 0.5755 a(0.5 0.9952 0.9860 0.8477 0.9870 10.000 0.9864 0.6494 a 0.9772 0.9999 0.7644 0.9500 0.9864 10.000 0.7432 at2 0.6213 0.7422 0.3576 0.5755 0.6494 0.7431 10.000 abstract. this paper presents a review of the main available indicators to measure poverty and income inequality, examining their properties and suitability for different types of economic analyses, and providing real-world data to illustrate how they... keywords. income inequality, indicators, inequality metrics, poverty metrics jel classification. d31, d63. 1. poverty indicators 1.1 foster-greer-thorbecke class of measures 1.2 sen and sen-shorrocks-thon indices 1.3 watts index a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 vol.9 • no.1 2020 issn: 2254-7088 european journal of government and economics 9(1), june 2020. european journal of government and economics issn: 2254-7088 number 9, issue 1, june 2020 doi: https://doi.org/10.17979/ejge.2020.9.1 the politics of renewable power in spain 5-25 doi: https://doi.org/10.17979/ejge.2020.9.1.5231 john s. duffield voting turnout in greece: expressive or instrumental? 26-45 doi: https://doi.org/10.17979/ejge.2020.9.1.5426 irene daskalopoulou the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach 46-73 doi: https://doi.org/10.17979/ejge.2020.9.1.5714 luzius stricker and moreno baruffini do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey 74-94 doi: https://doi.org/10.17979/ejge.2020.9.1.5948 esra n. kilci a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 95-114 doi: https://doi.org/10.17979/ejge.2020.9.1.5847 barbara pawełek https://doi.org/10.17979/ejge.2020.9.1 https://doi.org/10.17979/ejge.2020.9.1.5231 https://doi.org/10.17979/ejge.2020.9.1.5426 https://doi.org/10.17979/ejge.2020.9.1.5714 https://doi.org/10.17979/ejge.2020.9.1.5948 https://doi.org/10.17979/ejge.2020.9.1.5847 european journal of government and economics 9(1), june 2020, 95-114 95 european journal of government and economics issn: 2254-7088 a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 barbara pawełeka* a cracow university of economics, poland * corresponding author at: barbara.pawelek@uek.krakow.pl article history. received 16 december 2019; first revision required 13 march 2020; accepted 15 may 2020. abstract. the main purpose of the paper is to present the results of the comparative analysis of the member states of the european union in terms of expenditure on environmental protection made by the public sector. an additional purpose of the paper is to verify whether there is convergence in public spending on environmental protection of the member states of the european union. in the study, the convergence models and cluster analysis were used. the research results indicate, among others, that there was convergence in total public spending on environmental protection in the member states of the european union in 2004-2017, and that the structure of the member states in terms of amounts of public spending on various aspects of environmental protection in 2004-2010 differed from the structure of the member states determined for the years 2011-2017. keywords. convergence; environmental protection; eu member states; public spending jel codes. c38; h50; q01 doi. https://doi.org/10.17979/ejge.2020.9.1.5847 1. introduction the origins of the european union’s environmental policy can be traced back to the meeting of the european council in paris in 1972. at this meeting, the heads of state or government pointed to the need to develop the community environmental policy. since 1973, the european commission has been announcing multi-annual environmental action programmes (eap). they define, among other things, future objectives of the eu environmental policy (european parliament, environment policy: general principles and basic framework). in 2013, the eu council and parliament adopted the seventh eap for the period up to 2020 entitled living well, within the limits of our planet. the seventh eap lists nine priority objectives and what the eu needs to do to achieve them by 2020. they are (european commission, living well, within the limits of our planet. 7th eap – the new general union environment action programme to 2020): ‘1) to protect, conserve and enhance the union’s natural capital; 2) to turn the union into a resource-efficient, green, and competitive low-carbon economy; 3) to safeguard the union’s citizens from environment-related pressures and risks to health and wellbeing; 4) to maximise the benefits of the union’s environment legislation by improving implementation; 5) to increase knowledge about the environment and widen the evidence base for policy; 6) to secure investment for environment and climate policy and account for the environmental costs of any mailto:barbara.pawelek@uek.krakow.pl https://doi.org/10.17979/ejge.2020.9.1.5847 barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 96 societal activities; 7) to better integrate environmental concerns into other policy areas and ensure coherence when creating new policy; 8) to make the union’s cities more sustainable; 9) to help the union address international environmental and climate challenges more effectively’. the implementation of the priority objectives of the eap is supported, among others, by a public expenditure of the european union member states. the information on public spending on environmental protection in the european union member states is available, for example, on eurostat, which publishes data according to the classification of the public sector spending as per the cofog (the classification of the functions of the government) functions, in which public sector means the general government. the cofog encompasses 10 types of functions performed by public authorities (i.e. 10 divisions broken down into groups and classes): gf01 – general public services, gf02 – defence, gf03 – public order and safety, gf04 – economic affairs, gf05 – environmental protection, gf06 – housing and community amenities, gf07 – health, gf08 – recreation, culture and religion, gf09 – education, gf10 – social protection. financial data grouped according to the cofog may be the basis for comparative analyses of the general government expenditure in various countries. unfortunately, problems arise in the case of attempts to classify certain types of expenditure at the levels of the classification lower than the division. as a result, data concerning more detailed levels of the cofog are not always available to the full extent for all member states of the european union. in the literature, there are publications presenting studies on the impact of public spending on the level of the economic development of countries (pitlik and schratzenstaller, 2011; ryu, 2015). they demonstrate the existence of a positive correlation between the gdp and productive expenditure, and a negative correlation between the gdp and unproductive expenditure. pitlik and schratzenstaller (2011), on the basis of the findings of macroeconomic studies on the impact of fiscal policies on the level of economic development, divided public spending according to the cofog into productive and unproductive expenditure. public expenditure on environmental protection was classified as productive expenditure, but with a comment that its impact on the level of economic development was being challenged. countries have also been compared in terms of public expenditure by 10 divisions according to the cofog using cluster analysis methods for this purpose (ferreiro et al., 2010, 2013), and the occurrence of convergence in total public spending has been studied (apergis et al., 2013; ferreiro et al., 2013). the concepts of convergence as a phenomenon occurring in economic systems are quite varied. among the processes that favour the convergence of economic systems, giertz and mehta (1996) list a state policy, understood as the use by the state of various solutions in the institutional, tax, infrastructure and social spheres. ferreiro et al. (2013) presented the findings of the study on the occurrence of convergence in total public spending (as a percentage of gdp) and in public spending by divisions of the economic classification (as a percentage of total public expenditure) and the cofog (as a barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 97 percentage of total public expenditure) in the years 1990-1998 and 1999-2007. the study was carried out using the box plot. first of all, the variability of the distributions of total public expenditure and public expenditure by divisions in the two analysed periods was compared. in addition, the cluster analysis based on public expenditure by divisions of the economic classification and the cofog separately was carried out. the pca (principal component analysis) was performed in the analysis. in the cofog, the data matrix is formed by 20 countries and 11 variables: the shares of the 10 items (as a percentage of total spending) and the size of public expenditure (as a percentage of gdp). the eu member states were clustered based on the five factors indicated by the pca with the application of ward’s criterion. seven clusters were created for the 1990-1998 data. for the data from the period 1999-2007, there was a smaller number of clusters than in the previous period, i.e. five clusters. based on the study findings, the authors concluded that convergence had occurred in public spending in the years 1990-2007. on the other hand, apergis et al. (2013) studied the occurrence of convergence in public spending by 10 divisions of the cofog based on data from 17 eu member states from the period 1990-2012. the methodology proposed by phillips and sul (2007) was applied. on the basis of the estimated regression model, the absence of convergence in total public expenditure and in public spending by divisions of the cofog was confirmed. next, the dip-test (hartigan and hartigan, 1985) was applied to verify the null hypothesis that the distribution was unimodal. at the significance level of 0.10, the test outcomes resulted in the rejection of the null hypothesis for the majority of public spending as per the cofog functions; spending on environment protection was the only exception. hence, when applying two approaches to the convergence study, conflicting results were obtained for public expenditure on environmental protection. among publications on government spending on environmental protection, there are deliberations on selected aspects of the environmental protection. public social expenditure is a popular area of study (e.g. (halaskova, 2018; leitner and stehrer, 2016; szarowska, 2011)). halaskova (2018) clustered countries due to their public social expenditure. convergence in the case of public social expenditure was also studied (de simone et al., 2010). a much smaller number of publications refer to public expenditure on environmental protection only (russu, 2017), also in break-down by groups (ercolano and romano, 2018). russu (2017) analyzed public expenditure on environmental protection made in romania as compared to other european union member states in 1990, 1995, 2000, 2005, 2010 and 2012. ercolano and romano (2018) studied the occurrence of convergence in public spending on environmental protection in terms of its total amount and for six aspects of environmental protection distinguished in the cofog, expressed as a percentage of the gdp. the analysis was based on the data concerning 21 eu member states from 2002-2010. the occurrence of sigma-convergence was studied based on the box plot and the measures of the variability of distributions of total spending in the years 2002-2010, and of spending by six groups in 2002, 2006 and 2010. on the basis of the results obtained, the conclusion was drawn about the barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 98 absence of convergence in total public spending on environmental protection. in turn, the results obtained for the six aspects of environmental protection varied. in the case of expenditure on ‘protection of biodiversity and landscape’ and ‘r&d environmental protection’, convergence was confirmed; in the case of spending on ‘waste water management’ and ‘environmental protection n.e.c.’, divergence was confirmed; and in the case of ‘waste management’ and ‘pollution abatement’, the findings of the analysis were not sufficiently conclusive to confirm either convergence or divergence. in addition, a cluster analysis was performed using ward’s method and the main factors indicated by the pca for object-periods representing 21 eu member states in 2002, 2006 and 2010. the states were partitioned into five clusters. the analysis of the assignment of three objects representing individual member states in 2002, 2006 and 2010 became the basis for the conclusion about the diversity of eu member states in terms of trends existing in public spending on environmental protection in 2002, 2006 and 2010. this paper is part of the research trend represented by apergis et al. (2013) at the level of total public expenditure on environmental protection, and ercolano and romano (2018) at the level of the six groups of public expenditure on environmental protection as per the cofog. modifications of the research approach applied can also be found in the publications of halaskova (2018) and de simone et al. (2010). the main purpose of the paper is to present the results of the comparative analysis of 27 member states of the european union in terms of expenditure on environmental protection made by the public sector. an additional purpose of the paper is to verify whether there is convergence in public spending on environmental protection of the member states of the european union. the analysis covers the years 2004-2017, using the data from eurostat. the following research questions have been formulated: 1. was there any convergence in total public spending on environmental protection in the member states of the european union in 2004-2017? 2. was the structure of countries in terms of amounts of public spending on various aspects of environmental protection the same in 2004-2010 and 2011-2017? 3. was there any relationship between the structure of countries in terms of amounts of public spending on various aspects of environmental protection in 2004-2010 and 2011-2017 and the length of the membership of these countries in the european union? 4. was there any relationship between the structure of countries in terms of amounts of public spending on various aspects of environmental protection in 2004-2010 and 2011-2017 and the level of economic development of these countries? to the author’s best knowledge, to date, there have been no comparative analyses of the 27 member states of the european union in terms of government expenditure on various aspects of environmental protection according to the cofog, or studies on the occurrence of convergence in public spending on environmental protection in the 27 eu member states in 2004-2017. in this context, the presented results of the cluster analysis conducted on the set of barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 99 27 countries from the european union using the data on public expenditure on various aspects of environmental protection in 2004-2010 and 2011-2017, and a study on the occurrence of betaand sigma-convergence in total public spending on environmental protection in the analyzed countries in 2004-2017, are new in the literature. the paper structure is as follows: section 1 presents the objectives of the paper and the research questions; section 2 presents the discussion of the research procedure; section 3 presents the results of the empirical study with a discussion, and section 4 is a summary of the considerations. 2. data and methods the comparative analysis of the european union member states was carried out in accordance with the following research procedure: 1. downloading data from eurostat 2. verifying the data set in terms of its validity and completeness 3. describing selected empirical distribution characteristics 4. studying the occurrence of convergence 5. creating two data sets for the years 2004-2010 and 2011-2017 6. normalising variables 7. determining the optimum number of clusters 8. countries clustering 9. cluster validation 10. presenting the description of the clusters 11. an attempt at cluster profiling the comparative analysis of the european union member states in terms of government expenditure on environmental protection was based on annual data downloaded from eurostat (https://ec.europa.eu/eurostat). the study subjects comprised the member states of the european union (28 states at the outset). the author downloaded the data from eurostat for ‘general government expenditure by function’ (gf05) broken down into: • gf0501 – waste management (million euros) • gf0502 – waste water management (million euros) • gf0503 – pollution abatement (million euros) • gf0504 – protection of biodiversity and landscape (million euros) • gf0505 – r&d environmental protection (million euros) • gf0506 – environmental protection n.e.c. (million euros) in addition, the following data were also downloaded: ‘population (total number)’ and ‘gdp (current prices, million euros)’. next, per capita gdp, per capita gf05 and gf0501,…, gf0506 as a percentage of gdp were calculated. the study period covered the years 2004-2017. until 2004, the european union comprised https://ec.europa.eu/eurostat barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 100 15 member states (austria, belgium, denmark, finland, france, germany, greece, ireland, italy, luxembourg, netherlands, portugal, spain, sweden, the united kingdom). in 2004, 10 countries joined the eu (cyprus, czechia, estonia, hungary, latvia, lithuania, malta, poland, slovakia, slovenia). bulgaria and romania joined the eu in 2007, and croatia in 2013. due to the absence of many data for bulgaria in the database, this state was excluded from the analysis. in the database downloaded from eurostat, there were negative values of public expenditure. based on the clarifications presented in the publication (eurostat and european commission, 2011, p. 40): ‘(…) the problem might be a recording error, or more likely it might be the result of the compilation process, often in connection with the functional classification of final consumption expenditure p.3 and the distribution between the individual and collective components. even in this latter case, however, the negative values should be eliminated because negative values of public expenditure have no economic significance (…)’, negative values were replaced with zeros. the box plot was applied to present selected distribution characteristics of empirical data concerning government expenditure on environmental protection. this graph is used to visualize some distribution characteristics of a variable. in this study, the box plot based on positional measures was used. in the case of classical convergence, beta-convergence and sigma-convergence are distinguished. beta-convergence occurs when countries with lower expenditure per capita increase it faster than the countries with higher expenditure per capita. in the case of betaconvergence, absolute convergence and conditional convergence are also distinguished. the former type of convergence is based on the assumption that amounts of expenditure per capita in individual countries converge to the same level called ‘the steady state’. conditional convergence is a situation in which each country has its individual level of the steady state. sigma-convergence occurs when the variability of the amounts of expenditure per capita in a particular group of countries measured with, for example, a standard deviation, decreases over time. in the study of the occurrence of convergence, the occurrence of absolute beta-convergence and of sigma-convergence was verified: 1. absolute beta-convergence (barro and sala-i-martin, 1992): 1 𝑇𝑇 ln �𝑦𝑦𝑖𝑖𝑖𝑖 𝑦𝑦𝑖𝑖0 � = 𝛼𝛼0 + 𝛼𝛼1 ln(𝑦𝑦𝑖𝑖0) + 𝜀𝜀𝑖𝑖, (1) where: 𝑦𝑦𝑖𝑖0,𝑦𝑦𝑖𝑖𝑇𝑇 – the amount of public spending on environmental protection in the i-th country per inhabitant in the initial and final periods, t – the length of the analyzed period. barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 101 if the regression coefficient is statistically significant, and the assessment of this coefficient is a negative value, there is absolute beta-convergence in public spending on environmental protection in the eu member states. the rate of convergence can be calculated using a regression coefficient assessment according to the following formula: 𝛽𝛽 = − 1 𝑇𝑇 𝑙𝑙𝑙𝑙(1 + 𝑎𝑎1𝑇𝑇). (2) the β coefficient shows the percent of the distance to the steady state that an economy covers in one period. 2. sigma-convergence (friedman, 1992; quah, 1993): 𝜎𝜎𝑡𝑡 = 𝛾𝛾0 + 𝛾𝛾1𝑡𝑡 + 𝜀𝜀𝑡𝑡, (3) where: 𝜎𝜎𝑡𝑡 – standard deviation of logarithmic amounts of public spending on environmental protection in the member states per inhabitant, 𝑡𝑡 – time. if the slope parameter in the linear trend is statistically significant, and the assessment of this parameter is negative, then there is sigma-convergence in government spending on environmental protection in the eu member states. to eliminate the impact of short-term changes on the results of the comparative analysis of member states in terms of public spending on environmental protection, the annual data from the period of seven years were averaged. as a result, two periods were obtained covering the years 2004-2010 and 2011-2017. each period was represented by average values calculated on the basis of annual data for a particular period. the wilcoxon matched-pairs test (wilcoxon, 1945) was used to check if public spending on environmental protection for the two periods differ. variables were normalised with the application of the unitarisation method with fixed parameters over time: 𝑦𝑦𝑖𝑖𝑖𝑖𝑡𝑡 = 𝑥𝑥𝑖𝑖𝑖𝑖𝑖𝑖−min𝑖𝑖,𝑖𝑖�𝑥𝑥𝑖𝑖𝑖𝑖𝑖𝑖� max𝑖𝑖,𝑖𝑖�𝑥𝑥𝑖𝑖𝑖𝑖𝑖𝑖�−min𝑖𝑖,𝑖𝑖�𝑥𝑥𝑖𝑖𝑖𝑖𝑖𝑖� , (4) where: 𝑥𝑥𝑖𝑖𝑖𝑖𝑡𝑡 – the value of 𝑋𝑋𝑖𝑖 variable for the i-th object in period t, 𝑦𝑦𝑖𝑖𝑖𝑖𝑡𝑡 – the normalised value of 𝑋𝑋𝑖𝑖 variable for the i-th object in period t. the member states were clustered using ward’s method. ward’s method minimises the total barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 102 within-cluster variance. at each step, the pair of clusters with minimum cluster distance are merged. to implement this method, at each step one must find the pair of clusters that leads to a minimum increase in total within-cluster variance after merging. the calculations were made using the ‘nbclust’ function from the ‘nbclust’ package in r environment with ‘ward.d2’ option, which implements ward's (1963) clustering criterion (murtagh and legendre 2014). as a result of the application of hierarchical ward’s method, a set of partitions was obtained for each period, from a partition containing 27 single-element clusters to a partition containing only one 27-element cluster. due to the sample size (27 member states), the assumed maximum number of clusters was 6. for partitions containing from 2 to 6 clusters, the values of selected cluster validity measures were calculated: krzanowski-lai (k-l) index (krzanowski and lai, 1988), calinski-harabasz (c-h) index (calinski and harabasz, 1974), davies-bouldin (d-b) index (davies and bouldin, 1979). on the basis of the values of the indices above, the optimum number of clusters was determined. for each of the two partitions (for the periods: 2004-2010 and 2011-2017), the validity of country assignment to the clusters was assessed using the silhouette (s) index (rousseeuw, 1987). finally, the two partitions were compared with the adjusted rand index (rand, 1971; hubert and arabie, 1985; gates and ahn, 2017). the variables that were used in the clustering of objects are the basis for the description of clusters. therefore, the average values calculated for public expenditure made by member states from a given cluster on particular aspects of environmental protection were used to describe the clusters. in the end, an attempt at cluster profiling was made. profiling is carried out on the basis of the variables that were not used in the clustering of objects. its purpose is to indicate such attributes of objects that can explain the discovered structure of clusters. in this study, two qualitative attributes were taken into consideration, i.e.: 1. the period of a state’s membership in the european union: • ‘old’ – a state that joined the eu before 2004, • ‘new’ – a state that joined the eu in 2004 or later; 2. the level of economic development: • ‘gdp_1’ – a state with per capita gdp below the first quartile calculated for the eu member states in a particular period, • ‘gdp_2’ – a state with per capita gdp between the first quartile and the median, • ‘gdp_3’ – a state with per capita gdp between the median and the third quartile, • ‘gdp_4’ – a state with per capita gdp above the third quartile. the selection of attributes for cluster profiling was related to an attempt at investigating whether the diversity of the european union member states in terms of public spending on various aspects of environmental protection was related to the ‘old-15’ or ‘new-13’ status of the member states or to varying levels of their economic development. barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 103 3. findings and discussion selected empirical distribution characteristics of the analysed variables were presented using the box plot. figure 1 presents the distribution characteristics of per capita public expenditure on environmental protection in the european union member states in 2004-2010 and 20112017. median 25%-75% non-outlier range outliers extremes netherlands netherlands gf05_per_capita_2004-2010 gf05_per_capita_2011-2017 0 200 400 600 800 1000 [e ur o] [uxembourg [uxembourg figure 1. empirical distribution characteristics of per capita public expenditure on environmental protection in the eu member states in 2004-2010 and 2011-2017. source: author’s own study carried out using the ‘statistica’ program. when comparing the distribution characteristics, it can be concluded that the average value of per capita public expenditure on environmental protection measured with the median increased in 2011-2017 as compared to the period 2004-2010; the values of the first and third quartiles also increased. in both periods, per capita expenditure made in luxembourg and the netherlands was substantially higher than the expenditure made in other european union member states. it should be noted that the range containing non-outlier values in the period 2011-2017 was shorter as compared to the period 2004-2010. at the significance level of 0.05, the wilcoxon matched-pairs test showed a statistically significant difference between public spending on environmental protection per capita in the european union member states in the two periods under consideration (p-value equals 0.0095). this may indicate the occurrence of convergence in public spending on environmental protection in the european union member states. it seems that the observed regularity may be associated with the progressing economic development as measured with per capita gdp of the member states of the european union (figure 2). in the case of per capita gdp in 20042010 and 2011-2017, an increase in the value of the first and third quartiles with the constant value of the interquartile range and a slight decrease in the value of the median can be barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 104 observed. at the significance level of 0.05, the wilcoxon matched-pairs test showed a statistically significant difference between the gdp per capita in the european union member states in the two periods under consideration (p-value is less than 0.0001). median 25%-75% non-outlier range outliers [uxembourg [uxembourg g5t_per_capita_2004-2010 g5t_per_capita_2011-2017 0 10000 20000 30000 40000 50000 60000 70000 80000 90000 1e5 [e ur o] figure 2. empirical distribution characteristics of per capita gdp in the eu member states in 2004-2010 and 2011-2017. source: author’s own study carried out using the ‘statistica’ program. next, the author verified whether an increase in the average amount of per capita public expenditure on environmental protection was accompanied by an increase in the amount of this type of expenditure in relation to the gdp. figure 3 presents the empirical distribution characteristics of the amounts of expenditure as a percentage of the gdp. when comparing the graphs obtained for the analyzed periods, it can be observed that the average amount of expenditure measured with the median decreased in 2011-2017 as compared to 2004-2010. in addition, the length of the range containing non-outlier values increased in the period 20112017 as compared to 2004-2010. therefore, a situation opposite to that occurring for per capita expenditure is observed. but at the significance level of 0.05, the wilcoxon matched-pairs test did not show a statistically significant difference between public spending on environmental protection as a percentage of the gdp in the european union member states in the two periods under consideration (p-value equals 0.2029). barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 105 median 25%-75% non-outlier range outliers gf05_%_of_gdp_2004-2010 gf05_%_of_gdp_2011-2017 0.0% 0.2% 0.4% 0.6% 0.8% 1.0% 1.2% 1.4% 1.6% 1.8% aalta netherlands figure 3. empirical distribution characteristics of public expenditure on environmental protection (as a percentage of the gdp) in the eu member states in 2004-2010 and 2011-2017. source: author’s own study carried out using the ‘statistica’ program. when analysing medians of public spending on various aspects of environmental protection (figure 4), it can be observed that the median increased in the period 2011-2017 as compared to the period 2004-2010 in the case of the expenditure on the following aspects: ‘waste management’, ‘pollution abatement’ and ‘protection of biodiversity and landscape’. the reduction of the non-outlier range is observed only for expenditure on the aspects: ‘waste management’, ‘waste water management’, ‘protection of biodiversity and landscape’ and ‘r&d environmental protection’. however, at the significance level of 0.05, the wilcoxon matchedpairs test showed a statistically significant difference between public spending on environmental protection (as a percentage of the gdp) in the european union member states in the two periods under consideration only on ‘waste water management’ and ‘environmental protection n.e.c.’ (p-value equals 0.0031 and 0.0299 respectively). barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 106 median 25%-75% non-outlier range netherlands greece czechia 5enmark gf05 01 _% _o f_g dp_2 00 4-2 01 0 gf05 01 _% _o f_g dp_2 01 1-2 01 7 gf05 02 _% _o f_g dp_2 00 4-2 01 0 gf05 02 _% _o f_g dp_2 01 1-2 01 7 gf05 03 _% _o f_g dp_2 00 4-2 01 0 gf05 03 _% _o f_g dp_2 01 1-2 01 7 gf05 04 _% _o f_g dp_2 00 4-2 01 0 gf05 04 _% _o f_g dp_2 01 1-2 01 7 gf05 05 _% _o f_g dp_2 00 4-2 01 0 gf05 05 _% _o f_g dp_2 01 1-2 01 7 gf05 06 _% _o f_g dp_2 00 4-2 01 0 gf05 06 _% _o f_g dp_2 01 1-2 01 7 0.0% 0.1% 0.2% 0.3% 0.4% 0.5% 0.6% 0.7% 0.8% 0.9% 9stonia 9stonia croatiacroatia figure 4. empirical distribution characteristics of public expenditure on various aspects of environmental protection (as a percentage of the gdp) in the eu member states in 2004-2010 and 2011-2017. source: author’s own study carried out using the ‘statistica’ program. the study of the occurrence of convergence in public spending on environmental protection in the member states of the european union began with the estimation of the model (1): 1 13 ln �𝑦𝑦𝑖𝑖,13 𝑦𝑦𝑖𝑖,0 � = 0.1726 − 0.0305 ln�𝑦𝑦𝑖𝑖,0� (𝑅𝑅2 = 0.5719) (13) at the significance level of 0.05, the parameters of the model (13) are statistically significant (p-value is less than 0.01). since the assessment of the regression coefficient (𝛼𝛼1) is a negative value, it can be concluded that there is absolute beta-convergence in public spending on environmental protection in the eu member states. this conclusion means that the member states with a lower expenditure of this type per capita increase their spending on environmental protection faster than those member states with higher per capita public expenditure on environmental protection. in addition, the coefficient of the rate of convergence (2) was calculated: 𝛽𝛽 = 0.0388. the β coefficient indicates that the eu economy covers in one year ca. 3.9% of the distance to the steady state. next, the occurrence of sigma-convergence was verified. to this end, the linear trend function (3) was estimated: 𝜎𝜎𝑡𝑡 = 0.9144 − 0.0190𝑡𝑡 (𝑅𝑅2 = 0.5576) (14) barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 107 at the significance level of 0.05, the parameters of the model (14) are statistically significant (p-value is less than 0.01). since the assessment of the slope value (𝛾𝛾1) in the linear trend is negative, it can be concluded that there is sigma-convergence in government spending on environmental protection in the eu member states. the measures applied to assess the cluster validity are based on various criteria. thus, in research practice, the indications regarding the optimum number of clusters may vary depending on the adopted measure (sokołowski et al., 2019). in this study, three popular measures were used to assess the cluster validity, i.e. the krzanowski-lai index, calinskiharabasz index and davies-bouldin index. table 1 presents three indications for each of the analyzed measures in the order corresponding to the descending cluster validity level. bold font was used to highlight the number of clusters selected for subsequent analysis. table 1. cluster validity measures. measure 2004-2010 2011-2017 optimum number of clusters index value optimum number of clusters index value k-l 3 5 6 1.7001 1.1307 0.9670 6 3 4 1.2323 1.0053 0.9077 c-h 3 2 6 9.5808 9.3014 8.9929 6 5 3 6.4562 6.0149 5.9522 d-b 6 4 5 1.0265 1.0678 1.1881 6 5 4 1.0419 1.2569 1.3497 source: the author's calculations made using the ‘nbclust’ package in r environment (charrad et al., 2014). on the basis of the values of the cluster validity measures from table 1, the author decided that for the years 2004-2010, the member states of the european union would be partitioned into three clusters, and for the years 2011-2017 into six clusters. the higher number of clusters for 2011-2017 than for 2004-2010 can be interpreted as information about the greater diversity of the european union member states in 2011-2017 than in 2004-2010 in terms of the funds allocated to various aspects of environmental protection expressed as a percentage of a particular member state’s gdp. figures 5 and 6 present dendrograms obtained using ward’s method with squared euclidean distance for the years 2004-2010 and 2011-2017. colours are used to mark clusters of member states similar in a given period in terms of funds allocated to various aspects of environmental protection expressed as a percentage of a particular member state’s gdp. barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 108 figure 5. dendrogram for the years 2004-2010. source: author’s own study carried out using the ‘factoextra’ package in r environment (kassambara and mundt, 2017). figure 6. dendrogram for the years 2011-2017. source: author’s own study carried out using the ‘factoextra’ package in r environment (kassambara and mundt, 2017). table 2 presents the compositions of the clusters obtained as a result of the partitioning of the member states of the european union on the basis of the data from 2004-2010 and 20112017. in addition, the table lists the silhouette index values calculated to assess the validity of the assignment of individual member states to the created clusters. barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 109 table 2. compositions of the clusters obtained as a result of the partitioning of the european union member states on the basis of data from 2004-2010 and 2011-2017. 2004-2010 2011-2017 class country index s(i) ue gdp class country index s(i) ue gdp c_1 romania cyprus greece hungary sweden slovakia belgium austria latvia france poland 0.448 0.426 0.419 0.419 0.407 0.396 0.394 0.372 0.367 0.355 0.347 new new old new old new old old new old new gdp_1 gdp_3 gdp_2 gdp_1 gdp_4 gdp_2 gdp_3 gdp_4 gdp_1 gdp_3 gdp_1 c_1 latvia slovakia united kingdom belgium lithuania cyprus france spain sweden romania portugal 0.386 0.375 0.318 0.310 0.280 0.230 0.229 0.159 0.136 0.014 -0.094 new new old old new new old old old new old gdp_1 gdp_2 gdp_3 gdp_3 gdp_1 gdp_3 gdp_3 gdp_3 gdp_4 gdp_1 gdp_2 lithuania germany finland 0.344 0.343 0.292 new old old gdp_1 gdp_3 gdp_4 c_2 malta italy czechia 0.351 0.185 0.141 new old new gdp_2 gdp_3 gdp_2 united kingdom croatia 0.199 0.155 old new gdp_3 gdp_1 c_3 estonia finland 0.202 0.190 new old gdp_2 gdp_4 c_2 italy spain estonia slovenia denmark 0.309 0.301 0.246 0.233 0.200 old old new new old gdp_3 gdp_3 gdp_2 gdp_2 gdp_4 germany austria poland denmark slovenia 0.174 0.144 0.129 0.127 0.002 old old new old new gdp_3 gdp_4 gdp_1 gdp_4 gdp_2 czechia portugal 0.048 0.032 new old gdp_2 gdp_2 c_4 luxembourg hungary 0.435 0.187 old new gdp_4 gdp_1 c_3 netherlands luxembourg 0.222 0.192 old old gdp_4 gdp_4 netherlands ireland 0.157 0.059 old old gdp_4 gdp_4 ireland 0.179 old gdp_4 c_5 greece 0.000 old gdp_2 malta 0.131 new gdp_2 c_6 croatia 0.000 new gdp_1 note: ‘old’ – a state that joined the eu before 2004, ‘new’ – a state that joined the eu in 2004 or later; ‘gdp_1’ – a state with per capita gdp below the first quartile calculated for the eu member states in a particular period, ‘gdp_2’ – a state with per capita gdp between the first quartile and the median, ‘gdp_3’ – a state with per capita gdp between the median and the third quartile, ‘gdp_4’ – a state with per capita gdp above the third quartile. source: the author's calculations made using the ‘nbclust’ package in r environment (charrad et al., 2014) and the ‘factoextra’ package in r environment (kassambara and mundt, 2017). positive silhouette index values obtained for almost all member states prove that the countries were assigned to the clusters in a correct way. portugal in the years 2011-2017 is an exception. values close to zero mean that the assignment of such member states as czechia and portugal to the c_2 cluster in 2004-2010, romania to the c_1 cluster in 2011-2017, slovenia to the c_3 cluster in 2011-2017, and ireland to the c_4 cluster in 2011-2017 should be taken with caution. to assess the similarity of two partitions, the adjusted rand index was used (ari = 0.179). the adjusted rand index suggests that the two partitions (for the periods: 2004-2010 and 20112017) are not very similar to each other. in the next stage of the analysis, the clusters were described. for this purpose, average amounts of public expenditure on particular aspects of environmental protection as a percentage of the gdp were calculated, and the results are shown in table 3. the highest average amounts of public expenditure on particular aspects of environmental protection are marked in green, while the lowest amounts are marked in red. barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 110 table 3. average amounts of public expenditure on particular aspects of environmental protection as a percentage of the gdp. cluster gf0501 gf0502 gf0503 gf0504 gf0505 gf0506 2004-2010 c_1 c_2 c_3 0.278% 0.251% 0.409% 0.159% 0.166% 0.588% 0.060% 0.046% 0.119% 0.034% 0.190% 0.163% 0.015% 0.040% 0.006% 0.092% 0.108% 0.030% 2011-2017 c_1 c_2 c_3 c_4 c_5 c_6 0.384% 0.533% 0.096% 0.249% 0.006% 0.621% 0.093% 0.242% 0.138% 0.393% 0.146% 0.119% 0.078% 0.046% 0.094% 0.134% 0.003% 0.575% 0.049% 0.219% 0.078% 0.102% 0.113% 0.005% 0.016% 0.013% 0.043% 0.007% 0.000% 0.000% 0.079% 0.037% 0.081% 0.046% 0.300% 0.004% source: the author's calculations. the values in table 3 indicate two clusters of leaders in 2004-2010. in the period 2004-2010, in the c_2 cluster, the highest average public expenditure was recorded for ‘protection of biodiversity and landscape’ (gf0504), ‘r&d environmental protection’ (gf0505) and ‘environmental protection n.e.c.’ (gf0506). on the other hand, the member states from the c_3 cluster made the highest average expenditure on ‘waste management’ (gf0501), ‘waste water management’ (gf0502) and ‘pollution abatement’ (gf0503). in 2011-2017, greece – forming the c_6 single-element cluster – dominated in the area of public spending on ‘waste management’ (gf0501) and ‘pollution abatement’ (gf0503). in turn, croatia, forming the second single-element cluster c_5, made the highest public expenditure on ‘environmental protection n.e.c.’ (gf0506). the lowest average expenditure on ‘waste water management’ (gf0502) was recorded in the c_1 cluster, while the highest – in the c_4 cluster. the c_2 cluster is characterised by the highest average expenditure on ‘protection of biodiversity and landscape’ (gf0504). the highest average spending on ‘r&d environmental protection’ (gf0505) was recorded for the c_3 cluster. in order to profile the clusters, table 2 presents information about the length of the membership of a given country in the european union and the level of its economic development. based on the analysis of cluster compositions it cannot be unequivocally confirmed that, in the analyzed periods, there was a correlation between the division of the member states of the european union due to the amounts of funds allocated to various aspects of environmental protection and the division of countries into ‘old’ and ‘new’ ones due to the dates of their accession to the european union, or the division of the member states due to the level of their economic development measured with per capita gdp. it should be noted, however, that for the years 2004-2010, the c_2 cluster included mainly the countries with an average level of economic development (per capita gdp between the first quartile and the third quartile), while the c_3 cluster was dominated by countries with the highest level of economic development (per capita gdp above the third quartile). in the years 2011-2017, it was noticeable that the c_1 and c_3 clusters were dominated by the member states with an above-average level of economic development (per capita gdp barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 111 above the median). the c_2 cluster comprised countries with an average level of economic development (per capita gdp between the first quartile and the third quartile). in contrast, the c_4 cluster was dominated by countries with the highest level of economic development (per capita gdp above the third quartile). the member states forming the c_5 and c_6 singleelement clusters were characterised by the level of economic development below the average level recorded for the european union member states in 2011-2017 (per capita gdp below the median or the first quartile). 4. conclusions the findings of the empirical study made it possible to answer the research questions: 1. there was convergence in total public spending on environmental protection in the member states of the european union in 2004-2017. 2. the structure of the member states in terms of amounts of public spending on various aspects of environmental protection in 2004-2010 differed from the structure of the member states determined for the years 2011-2017. 3. there was no relationship between the structure of countries in terms of amounts of public spending on various aspects of environmental protection in 2004-2010 and 20112017 and the length of the membership of these countries in the european union. 4. there was no unambiguous relationship between the structure of the countries in terms of amounts of public spending on various aspects of environmental protection in 20042010 and 2011-2017 and the level of the economic development of these countries. studies in this area should be continued. when confronting the conclusions formulated on the basis of the findings of the empirical study with the present knowledge on the analyzed problem, it should be stated that they are: • consistent as regards the occurrence of convergence in total public spending on environmental protection with the results obtained using the dip-test by apergis et al. (2013); • inconsistent as regards the occurrence of convergence in total public spending on environmental protection with the conclusions presented by ercolano and romano (2018). the reasons for the consistency or inconsistency of the conclusions drawn on the basis of this study with the findings of other studies on the occurrence of convergence in total public spending on environmental protection may be seen e.g. in the length of the analyzed period – apergis et al. (2013) studied a period of 23 years, i.e. 1990-2012; ercolano and romano (2018) examined the period of only 9 years, i.e. 2002-2010; and in this analysis it was a period of 14 years, i.e. 2004-2017. in further studies, the author intends to: • apply various research approaches to the same data set to check the stability of the analysis findings; barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 112 • continue the study following the direction set in ercolano and romano (2018), i.e. study the relationship between public spending on environmental protection and some environmental outcomes. the results of the conducted research enrich the knowledge on the occurrence of convergence in public spending on environmental protection in the european union countries. additionally, the results of cluster analysis allow indicating the countries belonging to the european union, which were similar in terms of the structure of public spending on environmental protection in the years 2004-2010 and 2011-2017. extend the research to include the assessment of the relationship between public spending on environmental protection and some environmental outcomes should allow finding the optimal model of public spending on environmental protection in the european union. acknowledgements the research behind this publication was financed from the funds granted to the cracow university of economics. references apergis, n., christou, c., and hassapis, c. (2013). convergence in public expenditures across eu countries: evidence from club convergence. economics and finance research 1: 45– 59. https://doi.org/10.1080/21649480.2013.862735 barro, r. j., and sala-i-martin, x. (1992). convergence. journal of political economy 100: 223– 51. https://doi.org/10.1086/261816 calinski, t., and harabasz, j. (1974). a dendrite method for cluster analysis. communications in statistics 3: 1–27. https://doi.org/10.1080/03610917408548446 charrad, m., ghazzali, n., boiteau, v., et al. (2014). nbclust: an r package for determining the relevant number of clusters in a data set. journal of statistical software 61: 1–36. https://doi.org/10.18637/jss.v061.i06 davies, d. l. & bouldin, d. w. (1979). a cluster separation measure. ieee transactions on pattern analysis and machine intelligence 1: 224–227. https://doi.org/10.1109/ tpami.1979.4766909 de simone, e., gaeta, g. l., and ercolano, s. (2010). exploring convergence in some oecd public social expenditure trends. mpra paper 22496: 2–24. ercolano, s., and romano, r. (2018). spending for the environment: general government expenditure trends in europe. social indicators research 138: 1145–1169. https://doi.org/10.1007/s11205-017-1695-0 european commission, living well, within the limits of our planet. 7th eap – the new general union environment action programme to 2020. available from: http://ec.europa.eu/environment/pubs/pdf/factsheets/7eap/en.pdf european parliament, environment policy: general principles and basic framework. available from: http://www.europarl.europa.eu/ftu/pdf/en/ftu_2.5.1.pdf eurostat, european commission (2011), manual on sources and methods for the compilation of cofog statistics. classification of the functions of government (cofog). in: methodologies and working papers, luxembourg: publications office of the european union. https://doi.org/10.1080/21649480.2013.862735 https://doi.org/10.1086/261816 https://doi.org/10.1080/03610917408548446 https://doi.org/10.18637/jss.v061.i06 https://doi.org/10.1109/tpami.1979.4766909 https://doi.org/10.1109/tpami.1979.4766909 https://doi.org/10.1007/s11205-017-1695-0 http://ec.europa.eu/environment/pubs/pdf/factsheets/7eap/en.pdf http://www.europarl.europa.eu/ftu/pdf/en/ftu_2.5.1.pdf barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 113 ferreiro, j., garcía-del-valle, m. t., and gómez, c. (2010). social preferences and fiscal policies: an analysis of the composition of public expenditures in the european union. journal of post keynesian economics 32: 347–370. https://doi.org/10.2753/pke01603477320302 ferreiro, j., garcía-del-valle, m. t., and gómez, c. (2013). an analysis of the convergence of the composition of public expenditures in european union countries. american journal of economics and sociology 72: 799–825. https://doi.org/10.1111/ajes.12028 friedman, m. (1992). do old fallacies ever die? journal of economic literature 30: 2129– 2132. gates, a. j., and ahn, y.-y. (2017). the impact of random models on clustering similarity. journal of machine learning research 18: 1–28. https://doi.org/10.1101/196840 giertz, j.f., and mehta, s. (1996). regional income trends and convergence. institute of government and public affairs, university of illinois. halaskova, r. (2018). structure of general government expenditure on social protection in the eu member states. montenegrin journal of economics 14: 007–021. https://doi.org/10.14254/1800-5845/2018.14-4.1 hartigan, j. a., and hartigan, p. m. (1985). the dip test of unimodality. the annals of statistics 13: 70–84. https://doi.org/10.1214/aos/1176346577 hubert, l. & arabie, p. (1985). comparing partitions. journal of classification 2: 193–218. https://doi.org/10.1007/bf01908075 kassambara, a., and mundt, f. (2017). factoextra: extract and visualize the results of multivariate data analyses. r package version 1.0.5. available from: https://cran.rproject.org/package=factoextra krzanowski, w. j., and lai, y. t. (1988). a criterion for determining the number of groups in a data set using sum-of-squares clustering. biometrics 44: 23–34. https://doi.org/10.2307/2531893 leitner, s., and stehrer, r. (2016). development of public spending structures in the eu member states: social investment and its impact on social outcomes. working paper 128, the vienna institute for international economic studies, vienna. murtagh, f., and legendre, p. (2014). ward’s hierarchical agglomerative clustering method: which algorithms implement ward’s criterion? journal of classification 31: 274–295. https://doi.org/10.1007/s00357-014-9161-z phillips, p. c. b., and sul, d. (2007). transition modeling and econometric convergence tests. econometrica 75: 1771–1855. doi: https://doi.org/10.1111/j.1468-0262.2007.00811.x pitlik, h. & schratzenstaller, m. (2011). growth implications of structure and size of public sectors. wifo working papers 404. quah, d. (1993). galton's fallacy and tests of the convergence hypothesis. the scandinavian journal of economics 95: 427–443. https://doi.org/10.2307/3440905 rand, w. m. (1971). objective criteria for the evaluation of clustering methods. journal of the american statistical association 66: 846–850. https://doi.org/10.1080/01621459.1971.10482356 rousseeuw, p. (1987). silhouettes: a graphical aid to the interpretation and validation of cluster analysis. journal of computational and applied mathematics 20: 53–65. https://doi.org/10.1016/0377-0427(87)90125-7 russu, c. (2017). progress and dysfunctionalities in environmental protection in romania. economic insights – trends and challenges vi: 37–44. ryu, d. (2015). fiscal spending and economic growth. journal of economic development 40: 91–105. https://doi.org/10.35866/caujed.2015.40.4.004 sokołowski, a., denkowska, s., and fijorek, k. (2019). the assessment of selected measures for clustering quality. in: vision 2025: education excellence and management of innovations through sustainable economic competitive advantage: proceedings, ed. khalid s. soliman king of prussia, pa: international business information management association (ibima), p. 10033–10042. szarowska, i. (2011). relationship between government spending and economic growth in the czech republic. acta universitatis agriculturae et silviculturae mendelianae brunensis lix: 415–422. https://doi.org/10.11118/actaun201159070415 https://doi.org/10.2753/pke0160-3477320302 https://doi.org/10.2753/pke0160-3477320302 https://doi.org/10.1111/ajes.12028 https://doi.org/10.1101/196840 https://doi.org/10.14254/1800-5845/2018.14-4.1 https://doi.org/10.1214/aos/1176346577 https://doi.org/10.1007/bf01908075 https://cran.r-project.org/package=factoextra https://cran.r-project.org/package=factoextra https://doi.org/10.2307/2531893 https://doi.org/10.1007/s00357-014-9161-z https://doi.org/10.1111/j.1468-0262.2007.00811.x https://doi.org/10.2307/3440905 https://doi.org/10.1080/01621459.1971.10482356 https://doi.org/10.1016/0377-0427(87)90125-7 https://doi.org/10.35866/caujed.2015.40.4.004 https://doi.org/10.11118/actaun201159070415 barbara pawełek / european journal of government and economics 9(1), june 2020, 95-114 114 ward, j. h. (1963). hierarchical grouping to optimize an objective function. journal of the american statistical association 58: 236–244. https://doi.org/10.1080/01621459.1963 .10500845 wilcoxon, f. (1945). individual comparisons by ranking methods. biometrics bulletin 1: 80–83. https://doi.org/10.2307/3001968 https://doi.org/10.1080/01621459.1963%0b.10500845 https://doi.org/10.1080/01621459.1963%0b.10500845 https://doi.org/10.2307/3001968 number 9, issue 1, june 2020 a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 1. introduction 2. data and methods 3. findings and discussion 4. conclusions acknowledgements references assessing privatization efficiency in cis countries: financial, non-financial, and gdp-linked indicators © the author(s) 2023. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 12, no. 2 (2023), pages 157-174 https://doi.org/10.17979/ejge.2023.12.2.9850 submitted: jul 13, 2023 accepted: nov 4, 2023 published: dec 5, 2023 article assessing privatization efficiency in cis countries: financial, non-financial, and gdp-linked indicators jun chen,1,* talgat yerzhanov,2 svetlana shcheglova,3 anar mirzayev 4 1 sejong university, south korea 2 academy of public administration under the president of the republic of kazakhstan, kazakhstan 3 transbaikal state university, russian federation 4 azerbaijan state university of economics (unec), azerbaijan *correspondence: junchen5284@gmail.com abstract. this study investigates the impact of privatization on the relationship between gross capital accumulation, employment, labor resources, government spending, and gross domestic product (gdp) in commonwealth of independent states (cis) countries. employing econometric analysis and cointegration evaluation, the study reveals that privatization has had a mixed impact on cis economies. while privatization has contributed to increased capital accumulation and employment in some countries, it has also led to reduced labor resources and government spending in others. the study also finds that government spending is an indirect indicator of privatization efficiency that significantly influences gdp in armenia, azerbaijan, kazakhstan, kyrgyzstan, moldova, russia, uzbekistan, and tajikistan. keywords: capital accumulation; efficiency; gdp; government spending; privatization jel classification: l33; a10; o57 1. introduction the formation of the cis countries' national economies under the market model provided for the creation of the private sector of the economy. during the 1990s transitional period, the privatization of state-owned enterprises (ceos) and state property took place in said countries. the major task of the related processes was to increase the efficiency of resource use, intensify economic growth, and improve the citizens' standard of living. various theoretical models predict a positive relationship between privatization, economic liberalization, and economic development. however, the cis countries experienced an economic crisis during the period of privatization, which provided specific conditions for its implementation. it is interesting in this context to study the financial and nonfinancial effects of privatization in the cis countries under conditions of economic crisis. the financial effects of privatization, as a rule, provide for changes in the economic indicators of both the activities of directly privatized enterprises and the macroeconomic indicators of the https://creativecommons.org/licenses/by-nc/4.0/ 158 jun chen et al. economy as a whole. first, privatization influences government spending, which is reduced due to the termination of budget funding for seos. furthermore, privatized enterprises receive additional financing for their activities' modernization, which leads to the accumulation of gross capital in the country. in addition, positive effects are a decrease in unemployment and growing wages over a rise in the privatized enterprises' performance. non-financial performance indicators are much more difficult to determine because of their indirectness. in general, such an indicator of non-financial efficiency at the level of the national economy is the fair distribution of income. by increasing labor productivity, employees can earn higher wages at private enterprises, which leads to an increase in the equity of income distribution. however, the said effects may not always take place since it is necessary to consider the state of the country's economy and the conditions of privatization processes carried out. the purpose of this study is to identify the impact of privatization on the correlations between gross capital accumulation, employment and the volume of labor resources; government expenditures and the level of gross domestic product in the cis countries. the initial stage of the study considers the theoretical issues of studying the privatization effectiveness in different countries of the world. the identification of the theoretical basis is followed by the methodology presentation. the results of the study include an analysis of the privatization models’ impact on the interrelations between gross capital accumulation, employment and the volume of labor resources, government expenditures and the level of gross domestic product in 9 cis countries. the discussion presents different points of view on the studied issue. the conclusions summarize the results of the study. 2. literature review researchers consider privatization as a case when there is a process of transferring ownership rights for state-owned enterprises to private individuals. the privatization programs were developed and implemented by governments in many countries. the programs were based on the assumption of the modern neoliberal microeconomic theory. according to this theory, the transition from state ownership to private ownership stimulates more efficient management of available resources and property. at the same time, the practical inability to provide the expected results in some cases and a more objective approach to public administration, proposed by recent studies in the field of management, refute the mentioned assumption. consequently, some postulates of this theory require a revision (radić et al., 2021). thus, for example, the existing problem of stable electricity supply in nigeria negatively affected the socio-economic development of the country. this caused the state to make a decision on the privatization of the energy sector. as a result of the research and survey of household and commercial consumers of electricity, it was found that private electric distribution companies have not made significant improvements in electricity supply, the system of settlements and payments, accounting, problems of load management, responding to consumer requests, the formation of healthy competition, and market regulation (idowu et al., 2020). assessing privatization efficiency in cis countries nevertheless, there is also a positive experience of privatization. the study of the privatization impact on economic growth, labor productivity, income inequality and unemployment in 22 european countries in the period from 2004 to 2013 confirms its overall effectiveness. it has been empirically proven that privatization positively affects economic growth. at the same time, a noticeable effect on labor productivity, income distribution and unemployment appears two years after its implementation. in addition, it was also determined that privatization reforms could lead to an increase in labor productivity and a decrease in unemployment, but cause inequality of incomes (cuadrado-ballesteros & peña-miguel, 2018). an analysis of the performance of privatized water supply and sewerage companies in england and wales in 1991-2016 showed that during this period their profits and productivity increased by 4% and 2.5% per year, respectively. a positive contribution to profit growth was made by technical changes, the effects of changes in production volumes and tariff margins, and prices for products and production resources. at the same time, the impact of the change in technical efficiency, although positive, still turned out to be insignificant (molinos-senante & alexandros, 2020). this section contains a review of current research on privatization in different countries assessing its efficiency, which in the private and public sectors has long been a hot topic in academic research. findings of a study by the global centre for public service excellence (2015) show there is no unambiguous answer to the question of which form of ownership is more efficient, that is, private or public. in most cases, it depends on the specific area. key issues in comparing the efficiency of public and private ownership models are their range (including hybrid ones) and differences in the efficiency definition. hall (2014) comes to a similar conclusion in his study related to public and private sector efficiency. there is no empirical evidence that the private sector is essentially more efficient. the same results are constantly observed in sectors and services that use outsourcing, such as the telecommunications sector. it is fundamentally important to organize the work of public services in such a way that they effectively fulfill their social objectives. unlike the private sector, public service providers cannot be unquestioningly guided by financial performance. public services should include structures that ensure public goals are continually reinforced and monitored through democratic mechanisms of accountability and public participation. such mechanisms include formal accountability to public bodies such as municipalities or governments; structures for public participation in decision-making, including full transparency of information; and the active participation of representative organizations, for instance, community associations. this means that public and private companies have different performance evaluation criteria, which must be taken into account when assessing the efficiency of privatization processes. shi and sun (2016), studying the impact of privatization on the efficiency and profitability of china-based companies, conclude that this effect is ambiguous. over the past thirty years, china has undergone a major transformation, resulting in a considerably increased share of private companies. by analyzing data from privatized companies, the authors studied the impact of privatization on employment, wages, profits, and other economic indicators. they found that privatization led to a significant decline in the number of employees and an increase in labor productivity and profitability. soejono and heriyanto (2018) studied the economic performance of indonesia's companies before and after privatization using different methods. the results showed that direct-method privatization 160 jun chen et al. has a positive impact and leads to an increase in asset turnover. privatization through the capital market also showed a positive effect on the turnover of assets, but there was also a significant decrease in productivity, an increase in debt, and a return on equity. privatization based on employee buyouts led to a substantial reduction in measuring debt ratio and return on equity performance. thus, different methods of privatization have different consequences for different economic parameters of enterprises. the research as regards the impact of privatization on the financial performance of enterprises in kenya shows a significant improvement in the financial performance of firms privatized (muthii & ambrose, 2018). in particular, such indicators as return on turnover, return on assets, as well as return on equity, improved considerably. such results testify to the positive effect of privatization. estrin and pelletier (2018) examined privatization in developing countries. privatization involves the transfer of production facilities from the state to private hands. such transfers are, by their very nature, politically sensitive and may be linked with corruption. the authors consider different options for privatization and the state tasks at each stage of the privatization to maximize the positive effect. at the same time, they regard the following: social and economic side effects; competition side effects; global impact; political side effects; effects on the distribution of income; and effects on fiscal balance. the research findings demonstrate that different effects can differ significantly depending on the state's role in the privatization process, the effectiveness of state institutions, and opportunities for corruption. specialists from the asian development bank studied the issue of performance differential between private and state-owned companies (phi et al., 2019). using financial performance indicators and various empirical methods, the authors find substantial evidence that state-owned companies have poorer indicators than private ones, as they are less profitable, which is consistent with other similar studies. however, a simple assessment of profitability and consideration of this aspect only can be misleading since many seos are established to provide public goods and services and do not necessarily seek to maximize their profits. authors of the study on the privatization of state-owned enterprises in india conclude that central government ceos have increased their profitability, investment and growth since the country transferred to market prices and incentives with stricter contract enforcement (kim & panchanatham, 2021). the profitability and efficiency of seos in the manufacturing sector are higher than those of private enterprises. however, the efficiency of seos in the service sector is lower than that of private ones there. in general, the authors believe it is possible to improve the efficiency of public companies only through their full or partial privatization. authors of the study related to privatization and economic growth in malaysia (lai et al., 2018), consider privatization as a response to the high costs and poor performance of state-owned enterprises. in terms of macroeconomics, privatization ensured the formation of conditions for international businesses development in the country. privatized enterprises started to more actively develop international trade and attract foreign investment. private companies have ensured the economic growth of the malaysian economy through autonomy in the formation and implementation of strategies in international markets. the globalization of business has made it possible to increase privatized enterprises' revenue and their corporate tax return to the government budget. however, assessing privatization efficiency in cis countries there are still unresolved problems with the reformation of institutional processes to ensure an appropriate regulatory framework to protect consumers' interests and monitor the performance of privatized enterprises. a report on the role of privatization in sustainable development in ethiopia notes that privatization in developing countries is designed to solve the problem of state-owned companies' low performance (kedir & bedasso, 2020). the most common reason for privatization in developing countries is to restore macroeconomic balance, which is often destabilized by subsidizing unprofitable seos. by reducing inflation via its impact on the budget deficit, privatization helps to amplify the effect of expanding exports through currency devaluation. in turn, improved export performance contributes to debt sustainability by facilitating external debt servicing. thus, privatization in ethiopia is a tool for solving the country's macroeconomic problems. after analyzing current research, it can be concluded that the effects of privatization differ significantly from country to country. the economic conditions in the country amid the privatization period are very important for its efficiency. all analyzed literary sources do not regard the efficiency of privatization in countries that transfer from a planned to a market economy. thus, this study aims to close the gap in research related to the effects of privatization. 3. methods and material research design the cointegrating test engle–granger was used to study the impact of privatization on the interrelations between gross capital accumulation, employment and the volume of attracting labor resources, government expenditures and the level of gross domestic product in the cis countries. this method is used to analyze time series and create a model for rectifying (correcting) errors, in case when short-term changes are corrected depending on the degree of their deviation from longterm dependence. the essence of the engle–granger cointegrating test is that if the remaining parts of this model are non-stationary, that is, they have a single root, then there is no time series cointegration in this case. the null hypothesis is possible when there is no cointegration, that is, the presence of a single root in the model errors (cointegration equation). the statistics of the extended dickey-fuller test were additionally applied to test the hypothesis of a single root. the methods of the test procedure were based on the studies by harris, (1995), maddala and kim (1998). sample study different methods of privatization can differently impact the ownership structure and performance of privatized companies. moreover, privatization can provoke the emergence of externalities, which directly influence the market environment and the country's economic growth. privatization contributes to the formation and development of the capital market in the country as well. therefore, this study will consist of two stages. the research at the first stage will determine in what ways and years privatization took place in the cis countries. to this end, information on privatization in each cis country will be examined and applied financial instruments to transfer state-owned enterprises to private ownership will be 162 jun chen et al. determined. based on the information received, the methods of privatization will be specified. the study will consider the 1995-2020 period during which mass privatization was carried out in the cis countries while forming a market economy. the influence of privatization on economic growth in the cis countries will be determined in the second stage. gdp (dependent variable) will be considered as the resulting indicator. it is assumed that privatization impacts a country's gross capital formation, labor force, employment to population ratio, and government spending. hence, these parameters are introduced into the research model as independent variables or regressors. the model is formed as such: 𝐺𝐺𝐺𝐺𝐺𝐺 = 𝛽𝛽0 + 𝛽𝛽1𝐾𝐾 + 𝛽𝛽2𝐺𝐺 + 𝛽𝛽4𝐿𝐿 + 𝛽𝛽5𝐸𝐸 + 𝑒𝑒𝑡𝑡 , [1] where: 𝐺𝐺𝐺𝐺𝐺𝐺 – gross domestic product (current mln us$); 𝐾𝐾 – gross capital formation (current us$); g – general government final consumption expenditure (current us$); l – labor force, total; e – employment to population ratio, 15+, total (%) (modeled ilo estimate. this model will be built and tested for each cis country, as well as determined the impact of privatization on economic growth, which reflects the financial component of the privatization processes' efficiency parameters. to consider privatization, the models will introduce a dummy variable that will indicate the year of privatization in the cis countries. model variables are taken from the official statistical database of the (world bank, 2021). furthermore, data from the world bank statistical database for the 1995-2020 period will be used. research limitations the methodological limitation of the study is that gdp was chosen as the main indicator of the financial efficiency of the privatization results, whereas gross capital formation, general government final consumption expenditure, labor force, total and employment to population ratio (15+, total %) were selected as non-financial one. in this case, it was not possible to study the effect of the gini coefficient and the unemployment rate, since these data were characterized by incomplete filling of the time series. statistical analysis. the basic statistical calculations and data analysis were performed in the microsoft excel program. cointegrating test and regression models for cis countries were calculated in gretl, a special program for econometric analysis of time series of data. 4. results the description of the study results starts with the methods of privatization that were used in the cis from 1990 to 2001. privatization in armenia was launched in 1991 with small-scale privatization, assessing privatization efficiency in cis countries that is, entities of low value were transferred to private ownership. the majority of the privatization processes took place in 1994-1995 with the help of vouchers, which were also the main instrument of privatization in azerbaijan where mass privatization took place from 1995 to 1997. in belarus, privatization was carried out via buying out enterprises by management and employees in 1994. privatization in kazakhstan was performed through the direct sale of seos to private owners in 1996, which was the main stage of the move, although it was preceded in 1994 by the small-scale privatization using vouchers. in kyrgyzstan, privatization started as early as 1991 with small-scale privatization followed by the corporatization of seos in 1992-1993. the campaign was continued by mass privatization in 1996 through vouchers. moldova's major stage of privatization with vouchers took place in 1995. the relevant campaign in russia through vouchers began in 1992 with its key stage observed in 1993. as for uzbekistan, privatization there was carried out by the management/employee buy-out method with its main stage recorded in 1996. therefore, it can be stated that the main method of privatization in the cis was the campaign through vouchers, which took place mainly in the mid-1990s. this is because in the early 1990s, it was necessary to adopt appropriate laws and create a regulatory framework for the privatization process. in addition, it took time to assess the value of state property subject to privatization. the following abbreviations were used in the calculations: gdp (current mln us$) – gdp; gross capital formation (current mln us$) – gkf; general government final consumption expenditure (current mln us$) – gkf; labor force (mln total) – lf; employment to population ratio, 15+ (total %) (modeled ilo estimate) – epr. the impact of privatization on economic growth in each cis country is to be considered in detail. the model of such impact for armenia is presented in table 1. table 1. model of privatization impact on economic growth for armenia (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −11084.3 2702.25 −4.102 0.0006 *** gkf 0.876482 0.0894350 9.800 4.44e-09 *** ggfce 5.17675 0.632573 8.184 8.19e-08 *** lf −5692.24 2680.44 −2.124 0.0464 ** epr 370.907 89.6572 4.137 0.0005 *** time 152.808 43.0569 3.549 0.0020 *** mean dependent var 7158.992 s.d. dependent var 4438.313 sum squared resid 3296096 s.e. of regression 405.9616 r-squared 0.993307 adjusted r-squared 0.991634 log-likelihood −189.6444 akaike criterion 391.2888 schwarz criterion 398.8374 hannan-quinn 393.4625 rho 0.124387 durbin-watson 1.639735 model 1 data show that privatization had a statistically significant but negative impact on armenia's gdp. until 2000, the country experienced a decline in gdp (appendix 1). the other regressors are also statistically significant. at the same time, privatization had an impact on the 164 jun chen et al. employment level, which has been steadily declining over the past 15 years. it can be concluded that the first wave of privatization in the country (1991) took place during the economic recession (1990-1993). the next waves of privatization in 1994 and 1995 were in the period of gdp growth. it suggests that the positive dynamics of gdp growth are due to the development of the private sector of the economy. at the same time, a significant reduction in public spending is seen after the first wave of privatization, which is a positive effect on the state budget. meanwhile, the second and third waves of the campaign failed to influence the reduction in government spending. with reference to the above, the conclusion can be made that the second wave of privatization in armenia is characterized by lower efficiency and, overall, can be deemed to fail. the second example under consideration was azerbaijan. model 2 data suggest that privatization is not a statistically significant regressor of the country's economic growth over the period under review. but at the same time, government spending is a significant regressor of such growth in terms of statistics. in this case, the growth of government spending affected gdp growth (table 2). table 2. model of privatization impact on economic growth for azerbaijan (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −1195.76 41467.2 −0.02884 0.9773 gkf 1.27025 0.583403 2.177 0.0416 ** ggfce 7.24175 1.45505 4.977 7.25e-05 *** lf 18805.5 20552.1 0.9150 0.3711 epr −1020.69 913.347 −1.118 0.2770 time −1486.67 1182.81 −1.257 0.2233 mean dependent var 31535.44 s.d. dependent var 25259.12 sum squared resid 4.98e+08 s.e. of regression 4989.598 r-squared 0.968783 adjusted r-squared 0.960979 log-likelihood −254.874 akaike criterion 521.7491 schwarz criterion 529.2977 hannan-quinn 523.9228 rho 0.449639 durbin-watson 1.070623 at the same time, the relationship between gdp growth and capital accumulation is statistically significant. this is due to the fact that in the process of privatization, the main sectors of the country's economy also received considerable investments, including foreign ones. this resulted in a steady increase of the gdp indicator. however, the negative side of privatization was the reduction of employment due to the restructuring and modernization of industrial enterprises. model 2 data suggest that privatization is a statistically significant regressor of the country's economic growth over the period under review. but at the same time, government spending is a significant regressor of such growth in terms of statistics. in this case, the growth of government spending affected gdp growth. the comparison of the gdp dynamics and government spending with the stages of privatization is required for a more detailed analysis. the data indicate that from 1993 there was a steady downward trend in government spending against a rise in gdp. it can be assumed that during this period, the state started to prepare enterprises for privatization and did not allocate money funds for their operation. after the preparation, privatization took over three years. yet, the 1992-1993 period is interesting since an increase in government spending was along with gdp growth. at that time, the state met aggregate assessing privatization efficiency in cis countries demand to support the national economy. the dynamics above allow the inference that privatization objects were prepared as early as in 2016, and it was due to the growth of privatization in 2017 that the record drop in gdp of previous years was stopped. the first and second waves of privatization in azerbaijan became the gdp growth pre-determinants. for the first wave, it was the period 1995-1997, while for the second 2017-2018 (appendix 1–5). table 3 presents the results of the regression of the privatization impact on economic growth for belarus. table 3. model of privatization impact on economic growth for belarus (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const 59968.6 45727.8 1.311 0.2046 gkf 1.74925 0.148867 11.75 1.97e-010 *** ggfce 1.15675 0.601576 1.923 0.0689 * lf −3116.88 10649.7 −0.2927 0.7728 epr −917.597 430.109 −2.133 0.0455 *** time 1271.69 243.773 5.217 4.19e-05 ** mean dependent var 40718.33 s.d. dependent var 22953.38 sum squared resid 79862900 s.e. of regression 1998.286 r-squared 0.993937 adjusted r-squared 0.992421 log-likelihood −231.0828 akaike criterion 474.1657 schwarz criterion 481.7142 hannan-quinn 476.3394 rho 0.156440 durbin-watson 1.470995 analysis of model 3 reveals that gross capital formation and employment to population ratio is a statistically significant regressor of economic growth, having a positive impact. meanwhile, other regressors are statistically insignificant in this model. comparing the dynamics of gdp growth, government spending, and privatization stages, it can be stated that privatization failed to bring any considerable positive changes in economic development. accordingly, in the mid-90s, the state retained control over key economy sectors and the largest enterprises of the country. belarus saw a steady downward trend in gdp from 1990. the government tried to stimulate the economy by increasing its spending from 1992, but this yielded no desired result. even the 19931994 privatization did not have the intended effect. the rate of gdp decline halted for the subsequent period of 1995-1998, and then the downward trend resumed. in general, it is expected the equity of income distribution in society declines with a decrease in gdp. the state, by increasing its spending, tried to redistribute resources in the economy to improve the equity of income distribution. with reference to the above data, one can conclude that none of the variables is significant in the regression model. the outcomes of comparison of gdp growth, government spending, and privatization dynamics point to the fact that in 2017, privatization was the most noteworthy and allowed not only to slow down the rate of gdp decline but also receive a slight growth. as can be seen from the data given, the second privatization wave was quite effective in terms of gdp growth. however, such a state of affairs was only inherent to the year 2017 and partially 2018. then, the first wave trend continues. in general, privatization in belarus can be deemed extremely low. table 4 below gives the findings of the regression of the privatization effect on kazakhstan's 166 jun chen et al. economic growth. table 4. model of privatization impact on economic growth for kazakhstan (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −156772 238609 −0.6570 0.5187 gkf 1.99942 0.721287 2.772 0.0118 ** ggfce 3.92268 1.38921 2.824 0.0105 ** lf 18422.6 17721.7 1.040 0.3110 epr 340.784 3840.14 0.08874 0.9302 time −1070.37 1058.47 −1.011 0.3240 mean dependent var 106179.3 s.d. dependent var 76831.10 sum squared resid 2.45e+09 s.e. of regression 11072.90 r-squared 0.983384 adjusted r-squared 0.979229 log-likelihood −275.6003 akaike criterion 563.2006 schwarz criterion 570.7492 hannan-quinn 565.3743 rho 0.221860 durbin-watson 1.440469 model 4 did not reveal statistically significant regressors of kazakhstan's economic growth, except gross capital formation and general government final consumption expenditure. over the study period, kazakhstan saw a downward trend in gdp. at the same time, there was also a drop in government spending. noteworthy, during the years of privatization, gdp stopped decreasing, while government spending increased. this trend could be due to the compensation of structural imbalances in the economy by rising government spending. in general, there is no significant effect of privatization on the economic upturn in kazakhstan. the second wave of privatization also did not lead to gdp growth, which is clearly illustrated by the data for 2014-2016, in which the drop in gdp was much more significant compared to 20172020. hence, one can confidently state that the second privatization wave had no consequences for gdp, precisely like the first one. table 5 provides an assessment of the privatization impact on the economic progress of kyrgyzstan. table 5. model of privatization impact on economic growth for kyrgyzstan (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −1144.62 1922.08 −0.5955 0.5582 gkf 0.774563 0.170451 4.544 0.0002 *** ggfce 3.07567 0.428049 7.185 5.88e-07 *** lf −904.219 902.643 −1.002 0.3284 epr 49.8195 50.2811 0.9908 0.3336 time 93.5590 49.9372 1.874 0.0757 * mean dependent var 4393.349 s.d. dependent var 2657.590 sum squared resid 721239.0 s.e. of regression 189.8998 r-squared 0.995915 adjusted r-squared 0.994894 log-likelihood −169.8906 akaike criterion 351.7812 schwarz criterion 359.3297 hannan-quinn 353.9549 rho −0.105485 durbin-watson 2.200404 an analysis of the model 5 results indicates that gross capital formation (p-value < 0.01) and the general government final consumption expenditure (p-value < 0.01) constitute statistically assessing privatization efficiency in cis countries significant gdp regressors in kyrgyzstan. if the increase in the unemployment rate led to a drop in gdp, privatization caused a rise in gdp. at the same time, the country's economy was related to a decrease in gdp over the period under review. the privatization campaign carried out from 1991 to 1993 could not prevent the reduction in gdp. however, the positive effect in this regard was a decrease in government spending on financing state-owned enterprises. this freed up a resource to support the economy, which is evidenced by growing government spending in 1998-2000 in response to the decline in gdp during this period. proceeding from the outcomes obtained for model 5, an inference can be drawn about the critical importance of government spending for gdp indicators. as illustrated above, the period from 2017 to 2019 is characterized by growth of gdp and public expenditures, and their further relative stabilization. nevertheless, the second wave of privatization also does not have any impact on the gdp. an analysis of the privatization impact on economic progress in moldova is given in table 6. table 6. model of privatization impact on economic growth for moldova (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −7849.40 3522.12 −2.229 0.0375 ** gkf 1.09992 0.468599 2.347 0.0293 ** ggfce 3.51802 0.845027 4.163 0.0005 *** lf 2602.31 5104.56 0.5098 0.6158 epr 61.7894 88.8680 0.6953 0.4949 time 209.314 43.9672 4.761 0.0001 *** mean dependent var 5516.545 s.d. dependent var 3773.360 sum squared resid 5983981 s.e. of regression 546.9909 r-squared 0.983189 adjusted r-squared 0.978986 log-likelihood −197.3969 akaike criterion 406.7938 schwarz criterion 414.3424 hannan-quinn 408.9675 rho 0.608817 durbin-watson 0.751981 the only statistically significant gdp regressor in model 6 is gross capital formation and general government final consumption expenditure, which ensures gdp growth. in this case, it can be assumed that privatization had an indirect effect on gdp through government spending. examining the dynamics of gdp, public spending, and privatization in moldova provides a more detailed analysis. privatization in moldova took place in 1995, and in the subsequent period, there was a decrease in both gdp and public spending from 1997. immediately after privatization in 1996-1997, the country enjoyed a short-term increase in gdp, which was not due to the campaign, as the results of model 6 show. the second wave of privatization in moldova began in 2014 and was rather uneven. though, from 2018 to 2020, this indicator rose. furthermore, it is from 2018, a stable gdp growth with a slight advancement in public spending occurred, which enables the forecast that in 2021-2022 the state will boost the pace of privatization as well. analysis of the privatization effect on the economic development of russia is presented in table 7. 168 jun chen et al. table 7. model of privatization impact on economic growth for russia (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −600769 436122 −1.378 0.1836 gkf 1.85731 0.326408 5.690 1.44e-05 *** ggfce 2.81819 0.446157 6.317 3.64e-06 *** lf 14039.7 7933.77 1.770 0.0920 * epr −7053.66 9912.80 −0.7116 0.4849 time 2798.26 2436.23 1.149 0.2643 mean dependent var 1104403 s.d. dependent var 689272.8 sum squared resid 2.79e+10 s.e. of regression 37320.93 r-squared 0.997655 adjusted r-squared 0.997068 log-likelihood −307.1917 akaike criterion 626.3834 schwarz criterion 633.9320 hannan-quinn 628.5571 rho 0.159398 durbin-watson 1.478000 the data show that statistically significant regressors of economic growth are gross capital formation and general government final consumption expenditure, which increases gdp. although privatization itself is not directly statistically significant, public spending is a function of privatization in terms of reducing the amount of state support for enterprises. analyzing the dynamics of russia's gdp and government spending, one can conclude that privatization at the first stage failed to have the desired effect and gdp continued to fall. against this background, government spending grew, which stabilized only a year after the initial stage of the campaign. the privatization of 1994-1995 also had no positive effect on economic development, and the country experienced a decline in gdp from 1997 due to the onset of the crisis. the collected data show that the statistically significant regressors of economic growth are unemployment, which explains the fall in gdp and government spending. although the privatization process in russia took place in 2016, the most considerable gdp drop was recorded in 2015–2016, which suggests that the second wave of privatization did not actually occur. as for tajikistan and uzbekistan, the models of the privatization impact on economic growth there showed the absence of statistically significant regressors (table 8 and 9, respectively). table 8. model of privatization impact on economic growth for tajikistan (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const 8611.16 3513.46 2.451 0.0236 ** gkf 0.511033 0.163823 3.119 0.0054 *** ggfce 4.95175 0.379964 13.03 3.12e-011 *** lf −3735.40 5720.11 −0.6530 0.5212 epr −90.5712 198.724 −0.4558 0.6535 time 243.725 255.885 0.9525 0.3522 mean dependent var 4452.267 s.d. dependent var 3074.223 sum squared resid 1532333 s.e. of regression 276.7971 r-squared 0.993515 adjusted r-squared 0.991893 log-likelihood −179.6871 akaike criterion 371.3741 schwarz criterion 378.9227 hannan-quinn 373.5479 rho −0.144394 durbin-watson 2.169878 assessing privatization efficiency in cis countries table 9. model of privatization impact on economic growth for uzbekistan (cointegrating regression ols, using observations 1995-2020, t = 26, dependent variable: gdp) coefficient std. error t-ratio p-value const −260396 32326.5 −8.055 1.05e-07 *** gkf −0.407366 0.428596 −0.9505 0.3532 ggfce 6.87642 0.695594 9.886 3.84e-09 *** lf 37988.4 4008.45 9.477 7.75e-09 *** epr −1692.66 536.726 −3.154 0.0050 *** time −6733.35 833.194 −8.081 9.97e-08 *** mean dependent var 37948.17 s.d. dependent var 27228.46 sum squared resid 1.24e+08 s.e. of regression 2494.506 r-squared 0.993286 adjusted r-squared 0.991607 log-likelihood −236.8497 akaike criterion 485.6993 schwarz criterion 493.2479 hannan-quinn 487.8730 rho −0.418214 durbin-watson 2.703464 the data show statistically significant regressors of economic growth are gross capital formation and general government final consumption expenditure. at the same time, in case of uzbekistan, there is a statistically significant relationship with almost all regressors, with the exception of the gross capital formation. privatization in 1990 in tajikistan had an insignificant effect on economic growth in the country, and there was a rapid decline in gdp from 1992 and government spending from 1993. the 1996 privatization did not influence the economic development of uzbekistan. from 1995, there was an upward trend in gdp, which continued until 1999 inclusive. the privatization of 1996 did not affect this trend in any way. the cut in public spending had been trending since 1992, and it saw a sharp reduction in 1996. this may be due to privatization in the same year. close study of the models developed for the second privatization wave in tajikistan and uzbekistan enables the deduction that in the case of tajikistan, the most significant variable favorably affecting gdp was gross capital formation and general government final consumption expenditure. in uzbekistan, however, the second wave of privatization had no impact on economic growth, as did the first. since 2017, the country's gdp has fallen seriously, while government spending has increased. the obtained outcomes provide enough data for all the analyzed countries to be conditionally divided according to the criteria of efficiency and success of the first and second privatization waves (table 10). table 10. countries' grouping by privatization efficiency group countries poor efficiency of the first and second privatization waves belarus, russian federation, kyrgyz republic, uzbekistan first privatization wave was more efficient than the second wave armenia the first and second waves of privatization were equally efficient azerbaijan, moldova, tajikistan, kazakhstan thus, in addition to the fact that belarus and russia have failed both waves of privatization, they were marked by highly negative dynamics in income distribution. theoretically, privatization is aimed at improving the efficiency of previously state-owned companies through their transition to private corporate governance. however, in practice there are a number of side effects that can influence the 170 jun chen et al. effectiveness of privatization. subsequently, they lead to a decrease in the employment level and in wages, a restriction of internal competition, monopoly on political power and economic wealth, an increase in income inequality and other factors. 5. discussion relevant literary sources discuss various effects of privatization on the economy, which is not only a direct effect on economic growth or other financial indicators but also an indirect effect on the economy through institutional transformations. examining the effect of privatization on corruption, the authors (reinsberg et al., 2020) note that it significantly reduces the level of control over corruption. this conclusion can be interpreted as evidence that privatization creates highly concentrated rents that increase corruption risks while creating incentives among rent-seeking elites to weaken state capacity. this leads to the weakening of institutions and corruption growth. in this regard, such an effect of privatization can be interpreted as non-financial consequences, which may not be entirely desirable in the economies of developing countries. this conclusion is also relevant for the cis countries. in the 1990s, the shadow sector of the economy was actively formed during the privatization there over the emergence of corruption schemes in the privatization process. in turn, kant (2018) studied the relationship between privatization and economic growth by the example of ex-socialist countries with economies in transition. the author believes that the economic growth in such countries is associated not so much with the privatization of state property, but with the implementation of institutional reforms. the transition from a planned to a market model for national economic development was carried out through the transformation of state institutions. the privatization of state property and the reduction of the latter's share in the economy was one of the elements of said transformation. but it was the institutional changes and certain exsocialist countries joining the eu in 2004 that had a key effect on economic growth. the findings of this research paper also prove the low impact of privatization on the cis countries' economic growth. a study on the impact of privatization on employment and income indicates that there is no evidence of its significant negative effect on them (earle & shpak, 2019). such conclusions are made based on the analysis of the privatization process in hungary, romania, russia, and ukraine. it has been found that there are three channels through which privatization can have an impact on workers: productivity-improvement, cost-reduction, and scale-expansion effects. external privatization has a greater effect that is when state-owned enterprises become the property of foreign companies, which results in an increase in productivity and wages of workers compared to enterprises that were privatized by domestic firms. it is also interesting to compare the findings of studies as regards the privatization impact on economic growth with the example of other countries. so, a study on the impact of privatization on economic development in ethiopia notes that there is a stable long-run relationship between real gdp growth, privatization, inflation, government consumption, government budget balance, gross private domestic investment due to privatization and foreign direct investment due to privatization (krishna & teshome, 2018). the empirical results show that both privatization and foreign direct investment resulting from privatization have a positive effect on economic growth and are assessing privatization efficiency in cis countries statistically significant in the long and short run. this research paper on the example of the cis countries shows the opposite result. perhaps the reason for this is that privatization in the cis coincided with the economic downturn following the ussr collapse and the economic crises in these countries. authors of a study related to the privatization impact on economic growth in nigeria conclude that there is a considerable relationship between gdp and private sector capital spending, as well as a strong and positive relationship between gdp and public sector capital spending (udoka & anyingang, 2012). the combination of private and public sector capital significantly influenced the country's gdp. the study also found that privatization is not a comprehensive solution to the problems of poorly functioning state-owned enterprises. another study related to the effect of privatization on the nigerian economy indicates that privatization increased confidence in the country's economy (nwali et al., 2019). privatization not only increased the status and number of investors and shareholders but also improved the performance of the country's stock exchange and gdp growth. although privatization endangers and diminishes the country's sovereignty status, especially if the economy is dominated by foreign organizations such as the imf and the world bank, the benefits can lead to a reorganization of its economy. as a result of privatization, the efficiency of unprofitable state-owned enterprises is increased, while the state budget receives tax revenue. the research concerning the privatization impact on the economic and social development in bulgaria found that it benefited the country's economy in the long run (toromanov, 2016). it was economically necessary and justified. although gdp is a very aggregated and general indicator giving no details of structural changes, what is important is that it, nevertheless, clearly shows improvement and continued growth. the trend in the unemployment rate in bulgaria in the post-privatization period is positive, especially after the third wave in 1997-2001. the peaks in unemployment are observed in the first stage, 1990-1995, during the "shadow" privatization through seos decapitalization and after the third stage of privatization in 2001. after this period, there is a positive trend. however, considering the impact of privatization on unemployment in the context of the gini coefficient and the poverty line, it has not made a sufficient contribution to overcoming social inequality, mainly over the low minimum and average wages. the same was observed in the cis countries during the 1990s privatization period when the gini coefficient decreased even in the long run. in general, according to recent studies, it has been found that the corruption detection rate decreases as the number of transactions in the privatization process increases. however, the opposite effect is observed when privatization is more important in terms of annual income. in addition, it has been confirmed that the privatization programs carried out since the early 1980s have not been effective in reducing corruption in certain european countries (miguel & cuadradoballesteros, 2019). 6. conclusions privatization in the cis countries was carried out in the 1990s. its main methods were voucher privatization, direct privatization and privatization with the purchase of shares by 172 jun chen et al. management/employees. the results of the study showed that the privatization campaign in the cis had low indicators of financial and non-financial efficiency. the study found that, according to the results of privatization, all regressors of economic growth were statistically significant only in armenia, as well as in uzbekistan, with the exception of gross capital accumulation. considering the reduction of public spending as an indirect positive effect of privatization, it was found that this indicator is statistically significant for armenia, azerbaijan, kazakhstan, kyrgyzstan, moldova, russia, tajikistan and uzbekistan. the study showed that the factor of capital formation in the process of privatization is a statistically significant regressor in 8 cis countries out of 9 studied. that is, the effectiveness of privatization depends on the ability of new owners to renew and increase the fixed capital of the business. the analysis made it possible to divide the countries into three groups according to the degree of privatization impact on their economic development. thus, belarus, russia, kyrgyzstan and uzbekistan were included in the group of states with low efficiency of the first and second privatization waves. the group in which the first wave of privatization was more effective than the second was represented only by armenia. finally, the group representing countries with equal efficiency of both privatization waves included azerbaijan, moldova, tajikistan and kazakhstan. the results obtained differ from the results of studies in other developing countries. this is due to the different initial conditions of privatization. in the cis countries, privatization was carried out against the background of the economic downturn in the 1990s. therefore, the positive effects were leveled. in addition, poorly developed state institutions could not prevent the emergence of corruption schemes, which decreased the effectiveness of privatization. non-financial indicators of the privatization effectiveness tend to decrease in the cis countries in the long term. these conclusions are useful for further investigation of the reasons for the low efficiency of privatization processes in the cis countries. they can also be taken into account when developing modern privatization programs for other countries. references alibekov, e., & ibadoglu, g. (1999). privatization in azerbaijan. warszawa: center for social and economic research. http://www.case-research.eu/files/?id_plik=2134 cuadrado-ballesteros, b., & peña-miguel, n. (2018). the socioeconomic consequences of privatization: an empirical analysis for europe. social indicators research, 139, 163-183. https://doi.org/10.1007/s11205-017-1713-2 earle, j.s., & shpak, s. (2019). impact of privatization on employment and earnings. iza world of labor. https://doi.org/10.15185/izawol.93.v2 ermakovich, v., & pankuv, y. (1996). privatization in kyrgyzstan in 1992-1996. warszawa: center for social and economic research. http://www.case-research.eu/files/?id_plik=2379 ermakovich, v., kozarzhevsky, p., & pankuv, y. (1996). privatization in the republic of kazakhstan. warszawa: center for social and economic research. http://www.caseresearch.eu/upload/publikacja_plik/4886209_085r.pdf estrin, s., & pelletier, a. (2018). privatization in developing countries: what are the lessons of recent experience? the world bank research observer, 33(1), 65-102. https://doi.org/10.1093/wbro/lkx007 global centre for public service excellence. (2015). is the private sector more efficient? https://www.undp.org/sites/g/files/zskgke326/files/publications/gcpse_efficiency.pdf http://www.case-research.eu/files/?id_plik=2134 https://doi.org/10.1007/s11205-017-1713-2 https://doi.org/10.15185/izawol.93.v2 http://www.case-research.eu/files/?id_plik=2379 http://www.case-research.eu/upload/publikacja_plik/4886209_085r.pdf http://www.case-research.eu/upload/publikacja_plik/4886209_085r.pdf https://doi.org/10.1093/wbro/lkx007 https://www.undp.org/sites/g/files/zskgke326/files/publications/gcpse_efficiency.pdf assessing privatization efficiency in cis countries hall, d. (2014). public and private sector efficiency. public services international research unit. https://www.epsu.org/sites/default/files/article/files/public%20and%20private%20sector% 20efficiency%20en%20fin.pdf harris, r.i.d. (1995). using cointegration analysis is econometric modelling. edinburg gate, harlow. prentice hall. idowu, s.s., ibietan, j., & olukotun, a. (2020). privatization of power sector in nigeria: an evaluation of ibadan and ikeja electricity distribution companies performance (2005-2018). international journal of public administration, 43(16), 1413-1420. https://doi.org/10.1080/01900692.2019.1672183 institute of scientific information on social sciences of the russian academy of sciences. (2003). privatization in russia and other cis countries. http://inion.ru/site/assets/files/2601/2003_privatiz_ros_dr_str.pdf kant, c. (2018). privatization and growth: natural experiment of european economies in transition. munich personal repec archive. https://doi.org/10.2139/ssrn.3277194 kedir, a.m., & bedasso, b. (2020). privatisation for sustainable growth in ethiopia: a review. international growth centre. https://www.theigc.org/wp-content/uploads/2021/05/kedirnovember-2020-final-report.pdf kim, k., & panchanatham, n. (2021). reform and privatization of state-owned enterprises in india. in reforming state-owned enterprises in asia (pp. 157-168). singapore: springer. https://doi.org/10.1007/978-981-15-8574-6_8 krishna, r.c.v., & teshome, a. d. (2018). impact of privatization on economic growth in ethiopia. lambert academic publishing. https://www.researchgate.net/publication/327700937_impact_of_privatization_on_economic_ growth_in_ethiopia lai, f.w., shad, m.k., chow, y.y., & wei, c.y. (2018). revisiting privatization and economic growth in malaysia: an empirical examination. in shs web of conferences (vol. 56, p. 05004). edp sciences. https://doi.org/10.1051/shsconf/20185605004 maddala, g.s. & kim, i.-m. (1998). unit roots, cointegration, and structural change. cambridge university press. https://doi.org/10.1017/cbo9780511751974 molinos-senante, m. & maziotis, a. (2020), drivers of profitability and productivity growth in the english and welsh water industry since privatization. international journal of water resources development, 37(5), 865-881. https://doi.org/10.1080/07900627.2020.1782851 muthii, g.j., & ambrose, j. (2018). the effect of privatization on firm financial performance in kenya. international journal of management and commerce innovations, 5(2), 671-680. https://irlibrary.ku.ac.ke/bitstream/handle/123456789/20037/the%20effect%20of%20privatization... ..pdf?sequence=1&isallowed=y nwali, a.c., johnson, n., & oganezi, b. (2019). privatisation of public enterprises in nigeria: challenges and prospects on economic development. mediterranean journal of social sciences, 10(4), 131131. https://doi.org/10.2478/mjss-2019-0059 peña miguel, n., & cuadrado-ballesteros, b. (2019). is privatization related to corruption? an empirical analysis of european countries. public management review, 21, 69 95. https://doi.org/10.1080/14719037.2018.1444192 phi, n.t.m., taghizadeh-hesary, f., chuc, a. t., yoshino, n., & kim, c. j. (2019). performance differential between private and state-owned enterprises: an analysis of profitability and leverage. adbi working paper. https://doi.org/10.2139/ssrn.3470070 radić, m., ravasi, d., & munir, k. (2021). privatization: implications of a shift from state to private ownership. journal of management, 47(6), 1596-1629. https://doi.org/10.1177/0149206320988356 rakova, e. (2008). privatization as a way to finance the trade deficit: opportunities for belarus. ipm research center. http://www.research.by/webroot/delivery/files/wp2008r09.pdf reinsberg, b., stubbs, t., kentikelenis, a., & king, l. (2020). bad governance: how privatization increases corruption in the developing world. regulation & governance, 14(4), 698-717. https://doi.org/10.1111/rego.12265 shi, w., & sun, j. (2016). the impact of privatization on efficiency and profitability: evidence from chinese listed firms, 2001-2010. economics of transition, 24(3), 393-420. https://doi.org/10.1111/ecot.12094 soejono, f., & heriyanto, f.x. (2018). privatization method and firm performance: a study of indonesia's state-owned enterprises. jurnal dinamika manajemen, 9(2), 149-158. https://doi.org/10.15294/jdm.v9i2.14970 https://www.epsu.org/sites/default/files/article/files/public%20and%20private%20sector%20efficiency%20en%20fin.pdf https://www.epsu.org/sites/default/files/article/files/public%20and%20private%20sector%20efficiency%20en%20fin.pdf https://doi.org/10.1080/01900692.2019.1672183 http://inion.ru/site/assets/files/2601/2003_privatiz_ros_dr_str.pdf https://doi.org/10.2139/ssrn.3277194 https://www.theigc.org/wp-content/uploads/2021/05/kedir-november-2020-final-report.pdf https://www.theigc.org/wp-content/uploads/2021/05/kedir-november-2020-final-report.pdf https://doi.org/10.1007/978-981-15-8574-6_8 https://www.researchgate.net/publication/327700937_impact_of_privatization_on_economic_growth_in_ethiopia https://www.researchgate.net/publication/327700937_impact_of_privatization_on_economic_growth_in_ethiopia https://doi.org/10.1051/shsconf/20185605004 https://doi.org/10.1017/cbo9780511751974 https://doi.org/10.1080/07900627.2020.1782851 https://ir-library.ku.ac.ke/bitstream/handle/123456789/20037/the%20effect%20of%20privatization.....pdf?sequence=1&isallowed=y https://ir-library.ku.ac.ke/bitstream/handle/123456789/20037/the%20effect%20of%20privatization.....pdf?sequence=1&isallowed=y https://ir-library.ku.ac.ke/bitstream/handle/123456789/20037/the%20effect%20of%20privatization.....pdf?sequence=1&isallowed=y https://doi.org/10.2478/mjss-2019-0059 https://doi.org/10.1080/14719037.2018.1444192 https://doi.org/10.2139/ssrn.3470070 https://doi.org/10.1177/0149206320988356 http://www.research.by/webroot/delivery/files/wp2008r09.pdf https://doi.org/10.1111/rego.12265 https://doi.org/10.1111/ecot.12094 https://doi.org/10.15294/jdm.v9i2.14970 174 jun chen et al. toromanov, p.e. (2016). social consequences of the privatization in bulgaria and socio-economic impact of the neo-liberal economic theory on the transition to free market and democracy in the period from 1989 to 2015. de l'université de montréal. https://papyrus.bib.umontreal.ca/xmlui/bitstream/handle/1866/19409/toromanov_evgueni y_2017_memoire.pdf?sequence=4&isallowed=y udoka, c.o., & anyingang, r.a. (2012). the effect of privatization on economic growth of nigeria: 19792007 in retrospect. international journal of economic development research and investment, 3(2), 25-35. https://www.researchgate.net/publication/320622920_the_effect_of_privatization_on_econo mic_growth_of_nigeria_1979-2007_in_retrospect world bank (2021). indicators by country. https://data.worldbank.org/indicator?tab=all https://papyrus.bib.umontreal.ca/xmlui/bitstream/handle/1866/19409/toromanov_evgueniy_2017_memoire.pdf?sequence=4&isallowed=y https://papyrus.bib.umontreal.ca/xmlui/bitstream/handle/1866/19409/toromanov_evgueniy_2017_memoire.pdf?sequence=4&isallowed=y https://www.researchgate.net/publication/320622920_the_effect_of_privatization_on_economic_growth_of_nigeria_1979-2007_in_retrospect https://www.researchgate.net/publication/320622920_the_effect_of_privatization_on_economic_growth_of_nigeria_1979-2007_in_retrospect https://data.worldbank.org/indicator?tab=all 1. introduction 2. literature review 3. methods and material 4. results 5. discussion 6. conclusions references do government revenues matter for economic growth? evidence from nigeria european journal of government and economics 7(1), june 2018, 60-84. european journal of government and economics issn: 2254-7088 do government revenues matter for economic growth? evidence from nigeria isiaka akande raifua, *, abiodun najeem raheemb a department of economics, university of ibadan, ibadan, oyo state, nigeria b department of economics, federal university of agriculture, abeokuta, nigeria * corresponding author at: heritagetiamiyu@gmail.com article history. received 28 december 2016; first revision required 18 april 2017; accepted 13 july 2017. abstract. the bursting of crude oil prices in the international market since mid-2014 has resulted in dwindling oil revenue, which has led to economic recession in nigeria. the recession has further exacerbated existing socioeconomic problems bedeviling the country. in the light of this, we examined the effect of government revenues (oil and non-oil revenues) on economic growth, both in the short-run and the long-run using autoregressive distributed lag method. our findings show that government revenues are indispensable to economic growth in nigeria. in addition, we found that economic growth is more responsive to oil revenue than non-oil revenue. based on our findings, we advocate for effective and efficient use of government revenues. furthermore, since oil revenue fluctuates more than non-oil revenue, we further advocate for creation of an enabling business environment geared towards improving the contribution of the non-oil sector to the government revenue base. keywords. oil revenue; economic growth; ardl jel classification. h20; h27; o40; c21 1. introduction nigeria is one of the oil producing countries in the world with crude oil proven reserve of about 37.2 million barrels as at 2010. since the 1970s when crude oil started booming, crude oil has become the major export commodity that the country heavily relies upon as means of revenue generation and foreign earnings. going by history, oil revenue accounted for 66.96% of total revenues in the 1980s and then rose to 75.21% in the 2000s.1 the importance of oil revenue to the country’s economy is enormous. it forms the benchmark for the annual budget formulation and implementation. however, history shows that the oil revenue is susceptible to fluctuations of the oil price in the international market. this often results in the volatility of the revenue, which has had harmful effects on government expenditure with an ultimate effect on the economy. the recent development in the country readily supports the above assertion. in recent time, the 1 oil revenues include revenues from foreign and domestic crude oil and gas sales, receipts from petroleum profit tax and royalties as well as rents. doi: https://doi.org/10.17979/ejge.2018.7.1.4333 mailto:heritagetiamiyu@gmail.com isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 61 decline in the price of crude oil in mid-2014 has resulted in falling government revenues with socioeconomic consequences. this has become a burning issue in nigeria. to reduce these socioeconomic malaises, the current government is determined to diversify the economy away from the exploration, production and export of crude oil to other sectors of the economy, particularly to the neglected or abandoned agricultural sector and the non-exploited mining subsector.2 specifically, the prices of crude oil fell from about $114 per barrel in 2014 to about $50 per barrel in recent time. the fall in the crude oil price has resulted in fall in government revenues, which has thwarted the efforts of governments both at local, state and national levels to finance their developmental projects and to fulfil other mandated responsibilities. this is because the country as a whole depends almost entirely on the oil revenues to finance its developmental projects. this has degenerated into many socio-economic crises such as nonpayment of workers’ salaries, weak demand for industrial products, closing of factories and retrenchment of workers, a high rate of serious crimes and overall decline in the aggregate economy. moreover, oil revenue is not the only source of government revenue in nigeria. part of government revenues are also obtained from the agricultural sector, manufacturing subsector, service sector and other sectors of the economy. the revenues from these sectors come in different forms such as sales, taxes, fines, levies and tariffs. these revenues are referred to as non-oil revenue. even though the non-oil revenue is small compared to oil revenue, it has been parts of the funds being used by government for the execution of both recurrent and capital projects. historically, efforts at generating more revenues from other sources to complement oil revenue have seen the evolvement of different tax reforms, ranging from tax reform system employed before and during the colonial era to the automated tax system (taxpayer’s identification number). in the course of these tax reforms, different tax methods have been adopted ranging from income tax, company tax, petroleum tax, capital gains tax, as well as value added tax introduced in 1994 (odusola, 2006). however, the porosity in tax collection, which gives room to either tax avoidance or tax evasion has undermined the important contribution of taxes to economic growth, causing overreliance on oil revenue as a major source of financing government’s project apart from borrowing. as noted before, the volatility in oil price often results in revenue volatility, expenditure volatility, output volatility and unstable or unsustainable economic growth. in the face of the economic problems, it is expedient to ask these salient questions, “do revenues, either from oil or non-oil sources, matter for economic growth in oil producing countries, particularly with reference to nigeria?” what is the magnitude of the effects of oil and non-oil revenues on economic growth? what is the direction of relationship between government revenues and economic growth? in order to examine the relationship between government revenues and economic growth, there are ample empirical findings, particularly in oil producing countries, albeit the findings are mixed. while some reported positive effects of tax revenue, oil revenue and non-oil export on 2 the current ruling government is led by the president muhammadu buhari who won the 2015 general elections in nigeria. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 62 economic growth, others posited that its impact on economic growth is negative. there are studies that found no significant relationship between revenues and economic growth. the mixed empirical results may not be unconnected to different approaches employed by researchers; the nature of the economy under consideration, the types of revenues that the study focused on and the kind of controlled variables included in their growth model (see literature review section for details). thus, in the light of the inconclusive empirical evidence obtained from the previous studies and to answer the questions formally raised above, this study examines the short-run dynamic and long-run effects of sources of government revenues on economic growth. the contributions of this study are three fold. first, this study does not only consider the dynamic relationship between oil revenue and economic growth but also the dynamic relationship between non-oil revenue and economic growth. besides, in this study we also consider the effect of total government revenues on economic growth. second, instead of using error correction method as found in the literature, we employ autoregressive distributed lag estimation method (ardl) to examine the dynamic relationship between government revenues and economic growth in nigeria. the ardl has two advantages over ecm. first, ardl is not susceptible to order of integration of the variables of interest. in other words, irrespective of the order of the integration of the variables either i(0) or i(1), ardl is applicable. second, by using ardl, it is methodologically possible to capture both the short-run dynamic and long-run effects of government revenues on economic growth simultaneously. thirdly, for robust analysis, we consider another measure of economic performance, which is industrial production index (ipi) as a dependent variable and then examine the relationship between government revenues and ipi using classical ordinary least squares (ols).3 applying both ardl and ols estimation techniques, the results show that the government has generated revenues (either from oil sector or from non-oil sector) that are crucial to economic growth. however, economic growth is more responsive to oil revenue than the non-oil revenue. the results from robust estimation show that government revenues positively influence industrial production index (ipi). we also found the existence of unidirectional relation between government revenues and economic growth as granger-causality runs from government revenues to economic growth. the rest of the paper is organised as follows: section two focuses on the review of extant theoretical and empirical literature. section three contains methodological framework, data sources and description as well as brief stylised facts about the evolution of oil revenues, non-oil revenues and economic growth in nigeria. section four presents the empirical results and discussion. section five concludes with policy recommendations. 3 the ipi measures the amount of output from the sectors of the economy such as manufacturing, mining, construction and agricultural sectors. it is an indicator to measure the production of these sectors. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 63 2. literature review theoretical literature review the issue of public financing for economic growth and sustainable development, especially in the world of uncertainty, has become a burning issue for both academics and policymakers. little wonder policymakers, business tycoons and academics gather at one time or the other to discuss the best way to finance public projects to achieve socio-economic goals in the best possible way.4 for instance in 2002, the united nations organised an international conference that brought heads of state and government, ministers, business leaders, academics and others together in monterrey, mexico to discuss the best ways to finance the millennium development goals earlier set.5 turing to the literature, there have been a plethora of theories, hypothesis and theoretical models that link government revenues and government expenditure with economic growth through different channels, either at micro or macro levels. this extensive literature dates back to ricardo’s pioneer work on the financing system. according to ricardo (1820), the best way to finance government expenditure is through tax financing. according to wagner (1893), real per capita income of an economy increases as the share of public expenditures in total revenues increases. ramsey (1927) pioneered the theory of optimal taxation and this theory was expanded by mirrlees (1971). specifically, the theory of optimal taxation states that the purpose of collecting taxes is to improve social welfare. thus, social planners should design tax system in such a way to maximise the overall social welfare, taking into consideration the individual economic agent’s preference. such a tax system as posited by the theory, is required to cut down inefficiency and any forms of distortions in the market under a given economic consideration (slemrod, 1990). furthermore, keynes (1936) argued that to stimulate effective aggregate demand, increase in government expenditure serves as an appropriate policy instrument to achieve the desired objective. moreover, on the causal relationship between revenues and government, four stands of hypotheses are found in the literature. the first one is the tax-and-spend hypothesis pioneered by friedman (1978) and buchanan and wagner (1978). friedman states that, there exists positive causal relationship between government revenues and its expenditure. accordingly, an increase in revenues spurs expenditure. conversely, buchanan and wagner posit that the relationship between government revenues and expenditure is negative. the second strand of hypothesis is referred to as spend-and-tax hypothesis propounded by peacock and wiseman 4 the socio-economic goals are multidimensional, ranging from provision of public goods (road construction, provision of pipe-borne water, primary health care or streetlights on major roads) to reduction or elimination of poverty, hunger, disease as in the expired millennium development goals and the current sustainable development goals. 5 this conference dubbed, “the united nations international conference on financing for development” was organized to strategise the best ways to fund the millennium development goals so that the targets set can be achieved. the outcome of the conference is known as the “monterrey consensus”. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 64 (1979). according to the spend-and-tax hypothesis, an increase in expenditure translates to an increase in revenues. in other words, tax policy is designed after determining the total government expenditure. this will ensure that the government raises adequate revenues to meet its planned expenditure. musgrave (1966) and meltzer and richard (1981) pioneered the third hypothesis, known as fiscal synchronisation. according to them, the relationship between government revenues and expenditure is bidirectional. this is because the optimal fiscal policy of government (in terms of its revenue and expenditure) depends on voters’ preferences or decisions concerning their demand for public goods or services and their action towards the redistributive function of government. the fourth hypothesis, known as institutional separation hypothesis or fiscal neutrality hypothesis, was pioneered by baghestani and mcnown (1994). this states that government revenues and expenditure are independent of each other. this is based on the premise that constitutionally, the duties of executive and legislative arms of government are different. this presumes no causal relationship between government revenues and government expenditure. review of the empirical literature as related to the effect of government revenues on economic growth either in advanced or developing economies, the overall literature can be dichotomized as follows: first, there are studies that exclusively focus on how tax revenue affects economic growth. second, other studies focus on the relationship between government revenues (that is revenues from other sources besides taxes) on economic growth. the empirical evidence varies depending on the source of revenue being considered. however, there is unanimous empirical evidence from the literature that tax revenue is positively related to economic growth. beginning from engen and skinner (1996), who considered the impact of tax reform (a 5% point cut in marginal tax rates) on the long term economic growth by employing three approaches. in the first approach, the researchers examined the historical record of the economy of the united states of america (usa) to evaluate whether tax cuts have been associated with economic growth. secondly, they considered the evidence on taxation and growth for a large sample of countries. thirdly, they used evidence from micro level studies of labour supply, investment demand and productivity growth. the major econometric techniques include descriptive and simple regression. their results showed that 0.2 to 0.3 percentage differences in growth rates are due to a major tax reform. put differently a cut in tax spurs economic growth. focusing on african continent, babatunde et al. (2017) analysed the effect of tax revenue on economic growth. after a series of preliminary tests with employment of the panel estimation method, their finding showed that tax revenue promotes economic growth in africa. empirical findings from specific studies in developing countries are akin to the one from the study on the usa. for instance, in pakistan, mashkoor et al. (2010) discovered that tax revenues have both short-run and long-run positive effects on economic growth. takumah (2014) examined the effect of tax revenue on economic growth in ghana using quarterly data covering the period of 1986 to 2010. as in the study of isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 65 mashkoor et al., the results showed that tax revenue exhibits both short-run and long-run relationship with economic growth. similar results were obtained from the sundry studies on tax revenues and economic growth in nigeria (jubril et al., 2012; okafor, 2014; ofoegbu et al., 2016; ojong et al., 2016). specifically, jubril, et al. (2012) examined the effect of petroleum profit tax on economic development in nigeria using ols as a method of estimation. their study which covered a period between 2000 and 2010 showed that petroleum profit tax has a positive significant effect on economic growth. okafor (2014) also studied the impact tax revenues on nigerian economic development with the objective to examine the effect of income tax revenues on economic growth. using a data set that covered the period of 1981 to 2007 and ols estimation technique, his finding showed that income tax revenue exerts a positive and significant effect on economic growth. while the studies above examined the impact of tax revenue on economic growth, ofoegbu et al. (2016), on the other hand, analysed the effect of tax revenue on economic development in nigeria using human development index as a proxy for economic development. their findings are similar to the results above as they found positive and significant effects on the economic development in nigeria apart from the effects of taxes on the economic growth, there are also ample studies that specifically focused on the impact of oil and non-oil revenues on economic growth. briefly, dreger and rahmani (2014) critically examined the impact of oil revenue on iranian and gulf states economies by employing panel cointegration technique. their results showed that while the oil revenues exhibited a long-run relationship with economic growth in iraq, such relationship was not found in the gulf states particularly in the investment equation as investment failed to respond to oil revenue in the long-run. similar to this is the study of hamdi and sbia (2013) which focused on the dynamic relationship between oil revenues, government spending and economic growth in the kingdom of bahraina country where oil revenues are the major driver of government expenditure and importation of goods and services. their study which employed a multivariate cointegration, error correction mechanism as well as impulse response function showed that oil revenue remained the driver of economic growth and the channel through which government financed its expenditure in the kingdom of bahrain. in the case of nigeria, the study on the relationship between oil revenue and economic growth remains inconclusive. while ibeh (2013) found no significant relationship between oil revenue and economic growth, kabir (2016) using the vector autoregression (var) technique showed that oil revenue negatively impacted economic growth. in line with ibeh’s study, ijirshar (2015) employed the vecm estimation technique that showed that the coefficient of the error correction term was insignificant which implies that there is no short-run dynamic movement to the long-run. on the role of non-oil revenue on economic growth in nigeria, ude and agodi (2014) discovered that agricultural revenue, manufacturing revenue and interest rates had significant impact on the economy with the speed of adjustment of about 52%. in short, their results showed that non-oil revenues were crucial to economic growth both in the short-run and in the long-run. on whether the domestic revenues are enough to spur economic growth, tuffour (2013) examined the relationship between foreign aid and domestic revenues on the one hand and isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 66 their impact on economic growth in ghana. employing macroeconomic time series data covering the period from 1970 to 2011 and error correction method together with granger causality, the findings show that domestic revenues and foreign aid complement each other for development financing. in addition, it is found that foreign aid is less important compared with domestic revenue for financing domestic development. the study finally shows that the direction of causality runs from domestic revenue, foreign aid to economic growth. in other words, both domestic revenue and foreign aid had causal relationship with economic growth. although there seems to be a consensus on the impact of tax revenue on economic growth in the literature in both advanced and developing economies, such consensus is rare to be found in the literature on the impact of oil revenue on economic growth. in fact, it can be shown from the reviewed literature that the empirical evidences on the subject matter remain inconclusive or better still are mixed. this may be attributed to a number of factors aforementioned above. this, therefore, calls for the re-examination of the relationship between government revenues and economic growth by employing different method in the light of nigeria’s economic situation. 3. empirical methodology 3.1 autoregressive distributed lag framework the ardl method was developed by pesaran et al (2001) to overcome the restrictive assumption upon which the johansen cointegration test is applicable.6 specifically, the johansen cointegration test was designed on the assumption that the fundamental variables must be integrated by order 1 or i(1). however, ardl is used to determine variables’ cointegration irrespective of order of integration of the variables. besides, the ardl method is used to examine simultaneously both short-run dynamic and long-run economic relations. the adrl cointegration framework (p, q) in accordance to pesaran et al. (2001) are specified as follows: 1 ' *' 0 1 1 0 p q t i t i t t i t i i y t y x x uα α φ β β − − − = = = + + + + ∆ +∑ ∑ [1] 1 1 2 2 3 3...t t t t tx p x p x p x ε− − −∆ = ∆ + ∆ + + ∆ + , [2] where xt is k-dimensional i(1) variables which do not cointegrate among themselves. ut and ε t are uncorrelated disturbances with zero means and constant variance-covariance. pi are k x k coefficient matrices such that the var process in δx t becomes stable. the pesaran et al. 6 we actually used the johansen cointegration method to examine whether there exists contigeration among the variables we considered. this is as result of poor performance of the bound testing approach. however, the ardl approach was used for the joint determination of short-run dynamic and long-run relationship between government revenues and economic growth due to its aforementioned uniqueness over the error correction method. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 67 (2001) ardl framework above is based on the null hypothesis that there is no cointegration between or among our variables of interest against the alternative hypothesis that there exists a cointegration among the variables. formally, this is presented as follows: null hypothesis (h0): ∑ ∅𝑡𝑡 = 0𝑛𝑛 𝑡𝑡=1 [3] alternative hypothesis (h1): ∑ ∅𝑡𝑡 ≠ 0𝑛𝑛 𝑡𝑡=1 [4] the decision to accept the null hypothesis or not is based on the comparison of the calculated value of the f-test obtained from the estimation of the equations (1) and (2) with the lower and upper critical values given in the work of pesaran et al. (2001). suppose the calculated value of the f-test is greater than the upper critical value, then there exists a long-run relationship. in other words, there exists cointegration among the variables under consideration. however, if the calculated value of the f-test is less than the critical value, then there is no cointegration. the decision becomes inconclusive if the f-test value lies in between the upper and the lower critical values. based on the results obtained from the cointegration test exercise, we proceed to the estimation of the error correction term (ect) employing ardl. the purpose of ect is to determine the speed of adjustment to a long-run equilibrium after initial short-run economic disruption. two steps are involved in the determination of the error term through the error correction estimation technique. first is the derivation of error term which could be obtained by regressing independent variables on dependent variables. the second step entails subtraction of the actual value of dependent variables from the estimated value obtained from the first step. the framework for the error correction term estimation is given as follows: 0 1 ( ) n t t t t ect y xα λ = = − + ∑ , [5] where ect = error correction term, yt = dependent variable, ∑ 𝑋𝑋𝑡𝑡𝑛𝑛 𝑡𝑡=1 is the set of independent variables and α and λ are constant. 3.2. johansen cointegration framework johansen cointegration methodological framework is presented in this subsection. johansen built his cointegration method on the concept of maximum likelihood estimation. he derived the maximum likelihood estimation using sequential tests for determining the number of cointegrating vectors. specifically, this method relies on the relationship between the rank of a matrix and its characteristic roots. thus, johansen proposed two different likelihood methods, which include the trace test and maximum eigenvalue test.7 7 the trace test and the maximum eigenvalue tests equations are given as follows: isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 68 following hjalmarsson and osterholm (2007), the cointegration framework follows a var of order p specified as follows: 𝑦𝑦𝑡𝑡 = 𝜙𝜙 + 𝐴𝐴𝑡𝑡𝑦𝑦𝑡𝑡−1 + ⋯+ 𝐴𝐴𝑡𝑡𝑦𝑦𝑡𝑡−1 + 𝜀𝜀𝑡𝑡, [6] where 𝑦𝑦𝑡𝑡 is an nx1 vector of variables that are integrated of order one, i(1) and 𝜀𝜀𝑡𝑡is an nx1 vector of innovations. the var can be rewritten as follows: δ𝑦𝑦𝑡𝑡 = 𝜙𝜙 + π𝑦𝑦𝑡𝑡−1 + ∑ δ𝑦𝑦𝑡𝑡−𝑖𝑖 + 𝜀𝜀𝑡𝑡 𝑝𝑝−1 𝑖𝑖=1 , [7] where 1 1 pp t t j i j i a iand a = = + π = − γ = −∑ ∑ if the coefficient matrix π has reduced rank r<n, then there exists nxr matrices α and β each with rank r such that ' ' tand yαβ βπ = is stationary. r is the number of cointegrating relationship, the elements of α is known as adjustment parameter in the vector error correction model and each column of β is a cointegrating vector. 3.3. granger-causality framework the causality test is credited to the work of granger (1969). his purpose is to determine whether one variable causes another, that is, whether x-variable causes y-variable, given the past lags of x-variable, which can be used to predict y-variable. if x-variable can statistically predict y-variable, then we can say that x-variable granger-causes y-variable. the null hypothesis of granger-causality test is that x-variable does not granger-cause y-variable. the null hypothesis is tested against the alternative hypothesis that x-variable granger-causes yvariable. the granger-causality framework in the context of var framework is presented as follows: 𝑋𝑋𝑡𝑡 = ∑ 𝛼𝛼𝑖𝑖𝑋𝑋𝑡𝑡−1 + ∑ 𝛽𝛽𝑗𝑗𝑌𝑌𝑡𝑡−𝑗𝑗 + 𝜀𝜀𝑖𝑖𝑗𝑗𝑛𝑛 𝑖𝑖=1 𝑛𝑛 𝑖𝑖=1 [8] 𝑌𝑌𝑡𝑡 = ∑ 𝜆𝜆𝑗𝑗𝑌𝑌𝑡𝑡−𝑗𝑗 + ∑ 𝛿𝛿𝑖𝑖𝑋𝑋𝑡𝑡−𝑖𝑖 + 𝜀𝜀2𝑡𝑡𝑛𝑛 𝑖𝑖=1 𝑛𝑛 𝑗𝑗=1 [9] the rejection of the null hypothesis or otherwise is based on the f-test result estimated by the following formula: 𝐹𝐹 = (𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅−𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅)/𝑙𝑙 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅/(𝑛𝑛−𝑘𝑘) [10] ^ ^ max 1 (1 ) ln(1 ) n t ttrace i r j t andj tλ λ = + = − − = − −∑ respectively. in this case t is the sample size and ^ tλ is the ith largest canonical correlation. the trace test is designed to test the null hypothesis of r cointegrating vectors against the alternative hypothesis of n cointegrating vectors. however, the maximum eigenvalue test tests the null hypothesis of r cointegration against the alternative hypothesis of r+1 cointegrating vectors. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 69 where: rssr = restricted sum of squares rssur = unrestricted sum of squares l = number of lagged terms k = number of parameters n = number of observations if the computed f-statistics value is greater than the critical f-statistics, then it can be concluded that x-variable granger-causes y-variable, otherwise no granger-causality takes place between the two variables. however, probability-value is used in most of empirical studies to make a decision based on the output from the econometric software. 3.4. ardl model specification following the pesaran et al. (2001) framework discussed above, the estimated ardl models for economic growth as a dependent variable that captures both short-run and long-run effects is presented as follows: 1 1 1 2 1 3 1 4 1 5 1 6 1 1 1 1 1 1 1 1 1 0 0 0 0 0 t t t t t t t n n n n n n t t t t t t t t t t t t t i i i i i i gdp gdp tr se opt fdi inv gdp tr se opt fdi inv α λ λ λ λ λ λ β β β β β β ε − − − − − − − − − − − − = = = = = = ∆ = + + + + + + + ∆ + ∆ + ∆ + ∆ + ∆ + ∆ +∑ ∑ ∑ ∑ ∑ ∑ [11] 1 1 1 2 1 3 1 4 1 5 1 6 1 1 1 1 1 1 1 2 1 0 0 0 0 0 t t t t t t t n n n n n n t t t t t t t t t t t t t i i i i i i gdp gdp or se opt fdi inv gdp or se opt fdi inv α λ λ λ λ λ λ β β β β β β ε − − − − − − − − − − − − = = = = = = ∆ = + + + + + + + ∆ + ∆ + ∆ + ∆ + ∆ + ∆ +∑ ∑ ∑ ∑ ∑ ∑ [12] 1 1 1 2 1 3 1 4 1 5 1 6 1 1 1 1 1 1 1 3 1 0 0 0 0 0 t t t t t t t n n n n n n t t t t t t t t t t t t t i i i i i i gdp gdp nor se opt fdi inv gdp n or se opt fdi inv α λ λ λ λ λ λ β β β β β β ε − − − − − − − − − − − − = = = = = = ∆ = + + + + + + + ∆ + ∆ + ∆ + ∆ + ∆ + ∆ +∑ ∑ ∑ ∑ ∑ ∑ [13] where gdp, tr, or, nor, se, opt, fdp, inv and their lags represent economic growth, total revenues, oil revenues, non-oil revenues, secondary school enrolment, openness of trade, foreign direct investment and investment and their lags respectively. λ’s and β’s are coefficient parameters. moreover going by the error correction framework specified above, we therefore specify the error correction method model as follows: 0 1 2 3 4( )t t t t tect rgdp rev se opt invα λ λ λ λ= − + + + + [14] where ect = error correlation term, and other variables are as defined above. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 70 4.0. data and discussion of the estimated results 4.1. data sources and description the data used for this study are annual and cover the period of 1981 to 2013 and sourced from world development indicators database (2015 version), the central bank of nigeria (2014 version) and international financial statistics. real gdp used to capture economic growth is measured in local currency unit (the naira in nigeria), secondary school enrolment stands for human capital, foreign direct investment captures foreigners’ investment in nigeria (inflow fdi expressed as percentage of gdp), openness of trade is the summation of export and import divided by gdp multiplied by 100 and investment is measured using gross fixed capital formation scaled by gdp. all these variables are extracted from the world development indicators. government revenues such as total revenue, crude oil revenue and non-oil revenue are sourced from cbn statistical bulletin while the industrial production index is sourced from international financial statistics. all the variables are in logged form except for trade openness, gross fixed capital formation expressed as the ratio of gdp and industrial production index. 4.2. stylised facts about gdp, oil price, oil and non-oil revenues in the 60s, the major driver of the nigerian economy was agriculture, which, according to the central bank of nigeria statistical bulletin (2014), accounted for 65% of gdp in 1960. in fact, agriculture was the major means of foreign exchange earnings. in specific terms, between 1960 and 1970, total earnings from non-oil export (non-oil revenues) averaged n546.84 millions.8 table 1. the values and the growth rates of real gdp, oil prices, oil revenues and non-oil revenues year rgdp (n’tri) lrgdp (%) oil_price ($/barrel) loil_price (%) oil_rev (n’b) loil_rev (%) non_oil_rev (n’b) lnorev (%) 1981-1985 18.20 0.05 30.96 -6.36 8.57 7.54 3.74 -0.88 1986-1990 17.17 1.45 17.95 2.67 31.60 58.83 11.91 48.04 1991-1995 19.93 0.50 17.83 -6.10 178.72 42.74 50.51 58.16 1996-2000 22.29 3.26 19.79 16.07 693.20 49.76 191.87 22.91 2001-2005 31.13 11.15 34.22 15.71 2625.98 30.32 651.21 38.89 2006-2010 48.53 7.22 75.19 11.45 4973.82 11.94 1367.67 23.54 2011-2013 60.93 4.85 110.52 12.54 7904.72 13.26 2605.74 15.67 source: authors’ computation using eviews 9. note 1: rgdp = real gdp, lrgdp = growth rate of gdp, oil_price = oil price, loil_price = growth rate of oil price, oil_rev = oil revenues, loil_rev = growth rate of oil revenues, non_oil_rev = non-oil revenues, lnorev = growth rate of non-oil revenues. n’tri = naira value of gdp in trillion, n’b = naira value of oil revenues and non-oil revenues in billion. all 8 cbn statistical bulletin, 2014 version. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 71 averaged over five years interval except from 2011 to 2013, which is averaged over three years interval. note 2: n’tril = naira in trillion while n’b = naira in billion. however, the discovery of crude oil at olobiri9 in the 1950s and its subsequent boom in 1970s led to the displacement of the agricultural sector as the main driver of the economy and the source of foreign exchange earnings. for instance, between 1981 and 1985, revenues realised from the crude oil and non-oil sources stood at n 7.54 billion and n3.74 billion respectively. the upward disparity between oil and non-oil revenues continues till date as shown in table 1. during the same period under consideration, the value of real gdp has been on the increase, especially on average from n 18.20 trillion between 1981 and 1985 to n60.93 trillion between 2011 and 2013. moreover, with the same period, that is, between 1981 and 2013, average price of crude oil in the international market rose from usd 30.96 per barrel to usd 110.52 per barrel. however, over the years the fluctuation of price of crude in the international market has had a negative effect on oil revenue, which at one time or the other might have constituted a fiscal policy problem. for instance, figure 1 shows the growth rate of the crude oil price and the growth rate of oil revenues accrued to the government from 1981 to 2013. from the figure, it can be deduced that crude oil price and oil revenue move together over time. in other words, when crude oil price goes up, oil revenue also goes up as well and vice versa. this movement indicates that the bulk of government revenues are driven by oil revenue which in turn is driven by what happens to the oil price in the international market. -80 -40 0 40 80 120 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 oil price growth rate (%) oil revenue growth rate (%) figure 1: trend of oil price and oil-revenue growth rates (1982-2013) figure 2 depicts the trend of crude oil price and total government expenditure and it shows the extent to which crude oil price affects government revenue and expenditure. in specific terms, the figure shows cyclical movement in oil price as it affects oil revenue and government expenditure. this shows the channel through which crude oil price can have either positive or negative pass-through effects on the economy. for instance, when oil revenue increases due to 9 olobiri is located in the present bayelsa state. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 72 a rise in crude oil prices the economy responds positively. this implies an increase in government expenditure which can be invested in the critical sectors of the economy that can spur economic growth. on the other hand, a decline in oil price means a fall in government revenues. fall in revenues often put a constraint on government expenditure, which automatically translates into a fall in investment in critical sectors and consequently may lead to economic slowdown or downturn. -60 -40 -20 0 20 40 60 80 100 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 oil price growth rate (%) total government expenditure growth rate (%) figure 2. trend of oil price and government expenditure growth rate (1982-2013) in figure 3, the trend of relationship between economic growth proxied by real gdp and oil revenue is presented. from the figure, it can be observed that real gdp and oil revenue also move together. this implies that when crude oil price fluctuates, government revenues fluctuate and ultimately the economy also fluctuates. thus, we can say fluctuation in the price of crude oil has the potential to cause economic downturn or instability in an oil dependent country like nigeria. -20 -10 0 10 20 30 40 -100 -50 0 50 100 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 real gdp growth rate oil revenue growth rate oi l r eve nu e g row th ra te (% ) real gdp growth rate (%) figure 3. trend of real gdp and oil revenue growth rate (1982-2013) to cap it all, figure 4 shows the relationship between the growth rate of non-oil revenue and the economic growth rate. from the figure, it appears that there are no or little predictable patterns in the movement of the growth rate of oil revenue and economic growth over time except during the early 1980s and the period after the global financial crisis of 2008. in other words, non-oil revenue appears to have little impact or effect on the nigerian economy. for instance, between isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 73 1990 and 2000 when the non-oil revenue appears to be volatile, the economy still recorded moderate growth. however, since 2008, the growth rate of non-oil revenue and economic growth seem to be synchronised, that is, to follow the same pattern. -50 0 50 100 150 200 250 -20 -10 0 10 20 30 82 84 86 88 90 92 94 96 98 00 02 04 06 08 10 12 real gdp growth rate non-oil revenue growth rate no n-o il r eve nu e g row th ra te (% ) real gdp growth rate (% ) figure 4. trend of real gdp and non-oil revenues growth rates (1982-2013) from the foregoing analysis, the stylised facts about nigerian economy in the light of the current study are summarised as follows: before and shortly after independence in the 1960s, agriculture was the major driver of the economy. the oil sector took over as the driver of the economy in the 1970s, since when, oil revenue constitutes the bulk of government revenues. however, fluctuations of crude oil price in the international market put constraint on the government revenue generation and this affects the economy. in short, the nigerian economy is more susceptible to oil price fluctuation in the international market, thereby putting constraint on the capacity of government to generate more revenues, to finance its planned expenditure and this ultimately affects economic progress. 4.3. empirical results and discussion 4.3.1. unit root test results in order to avoid spurious regression analysis, it is imperative to first determine whether the variables under consideration contain unit roots or not. in other words, there is need to determine the order of integration of our variables of interest before we carry out the cointegration analysis of the relationship between government revenues and economic growth (real gdp). to achieve this, we employed ng-perron unit root tests. the test is carried out with intercept/constant in the regression. for a decision to be made either to reject the null hypothesis or to accept it, the computed (statistical) value of each element of the test (mza, mzt, msb and mpt) must be less than the asymptotic critical values. thus, table 2 presents the ng-perron unit root test results. the results show that the null hypothesis of no unit toot test cannot be rejected at the level for all the variables considered in this study. this implies that the variables are not stationary at level. they are, however, become stationary after first difference. thus, the variables are integrated at order 1. hence, we proceed to thr cointegration test based isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 74 on the johnasen cointegration test. table 2. ng-perron unit root test results deterministic component: constant variable lag mza mzt msb mpt lrgdp 2 level 0.864 0.484 0.561 25.872 differenced -8.409 ** -2.040 0.243 2.952 lipi 2 level 0.930 0.559 0.601 29.045 differenced -15.323 * -2.763 0.180 1.619 ltrev 2 level 0.340 0.250 0.734 35.703 differenced -7.975 *** -1.984 0.249 3.118 lorev 2 level 0.386 0.304 0.787 40.362 differenced -10.967 * -2.321 0.212 2.313 lnorev 2 level 0.328 0.213 0.650 29.357 differenced -9.983 ** -2.233 0.224 2.460 lse 2 level 0.409 0.210 0.512 21.115 differenced -6.128 *** -1.721 0.281 4.092 lfdi 2 level -0.164 -0.107 0.652 26.690 differenced -10.870 ** -2.326 0.214 2.276 inv_gdp 1 level -1.990 -0.945 0.475 11.747 differenced -14.177 * -2.639 0.186 1.817 openess 2 level -5.732 -1.597 0.279 4.551 differenced -6.130 *** -1.721 0.281 4.092 asymptotic critical values 1% -13.800 -2.58 0.174 1.780 5% -8.100 -1.98 0.233 3.170 10% -5.700 -1.62 0.275 4.450 source: authors’ computation using eview 9 software. note: *, **, and *** represent 1%, 5% and 10% levels of significance respectively. 4.3.2. cointegration test results in this subsection, the question of whether cointegration exists among the variables we considered was addressed. in this case, the johansen cointegration technique was employed. table 3 presents the johansen cointegration test results (with unrestricted intercept and no trend) for three models which include the total revenues model, oil revenues model and non-oil revenues model. the cointegration test was done to determine whether our variables of interest are cointegrated or not, that is, to test whether the long-run relationship holds. since the cointegration tests are sensitive to lag selection criteria, the maximum lag length selected is two for the three models. from table 3, it can be observed that both the total revenues model and oil revenues model have three cointegrating equations while the non-oil revenues model has only two cointegration equations. the overall results, therefore, show that our variables of isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 75 interest are cointegrated at the 5% level of significance. this implies that the long-run and shortrun models can be estimated using the ardl cointegrating short-run and long-run method. table 3. johansen cointegration test results. trace test k = 2 maximum eigenvalues test k =2 ho ha ( λ trace) critical values (5%) ho ha ( λ max) critical values (5%) total revenues model r ≤ 0 r > 0 182.1519 (0.0000) 9.575.366 r ≤ 0 r > 0 87.25267 (0.0000) 4.007.757 r ≤ 1 r > 1 94.89919 (0.0002) 6.981.889 r ≤ 1 r > 1 40.83225 ( 0.0063) 3.387.687 r ≤ 2 r > 2 54.06694 (0.0117) 785.613 r ≤ 2 r > 2 29.32054 (0.0296) 2.758.434 r ≤ 3 r > 3 24.74641 (0.1707) 2.979.707 r ≤ 3 r > 3 15.22371 (0.2735) 21.13162 r ≤ 4 r > 4 9.522698 (0.3192) 1.549.471 r ≤ 4 r > 4 5.699721 (0.6520) 1.426.460 r ≤ 5 r > 5 3.822978 (0.0505) 3.841.466 r ≤ 5 r > 5 3.822978 (0.0505) 3.841.466 oil revenues model r ≤ 0 r > 0 186.2834 (0.0000) 9.575.366 r ≤ 0 r > 0 83.50589 (0.0000) 4.007.757 r ≤ 1 r > 1 102.7776 (0.0000) 6.981.889 r ≤ 1 r > 1 47.88897 (0.0006) 3.387.687 r ≤ 2 r > 2 54.88858 (0.0095) 4.785.613 r ≤ 2 r > 2 30.32064 (0.0217) 27.58434 r ≤ 3 r > 3 24.56794 (0.1775) 2.979.707 r ≤ 3 r > 3 15.39467 (0.2622) 2.113.162 r ≤ 4 r > 4 9.173268 (0.3496) 1.549.471 r ≤ 4 r > 4 5.476585 (0.6809) 1.426.460 r ≤ 5 r > 5 3.696684 (0.0545) 3.841.466 r ≤ 5 r > 5 3.696684 (0.0545) 3.841.466 non-oil revenues model r ≤ 0 r > 0 147.3871 (0.0000) 9.575.366 r ≤ 0 r > 0 63.15822 (0.0000) 4.007.757 r ≤ 1 r > 1 0.717646(0.0023) 6.981.889 r ≤ 1 r > 1 37.93785 (0.0155) 3.387.687 r ≤ 2 r > 2 46.29103(0.0696) 4.785.613 r ≤ 2 r > 2 25.58088 (0.0883) 27.58434 r ≤ 3 r > 3 20.71015(0.3760) 2.979.707 r ≤ 3 r > 3 14.11761 (0.3555) 2.113.162 r ≤ 4 r > 4 6.592536(0.6254) 1.549.471 r ≤ 4 r > 4 6.580745 (0.53990) 1.426.460 r ≤ 5 r > 5 0.011791(0.9133) 3.841.466 r ≤ 5 r > 5 0.011791 (0.9133) 3.841.466 source: authors’ computation using eviews 9 software. note: probability values that signify the level of significance are put in parenthesis. also, r represents number of cointegrating vectors and k represents the number of lags in the unrestricted var model. 4.3.3 granger-causality test in this subsection, we carried out a granger-causality test to further establish the existence of relationship between government revenues and economic growth in nigeria. in its original form, the granger-causality test is designed to determine whether one variable can be used to forecast another variable and it is predicated on the null hypothesis of no granger-causality between two variables. the null hypothesis will be rejected if the computed f-statistical value is greater than the critical f-statistical value. in this study, we make use of probability values obtained from the eviews output to determine whether our variables of interest granger-cause each other. the results of the granger-causality test are presented in table 4. the results show isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 76 that there exists a unidirectional causality between oil revenues, non-oil revenues and total revenues with the direction of causality running from oil revenue, non-oil revenue and total revenue to economic growth. in other words, oil revenue, non-oil revenue and total revenue granger-cause economic growth in nigeria. table 4. granger-causality test results null hypothesis obs f-statistic prob. loirev does not granger cause lrgdp 32 7.5181 ** 0.0104 lrgdp does not granger cause loirev 0.7244 0.4017 llnoilrev does not granger cause lrgdp 32 6.5171 ** 0.0162 lrgdp does not granger cause llnoilrev 0.0537 0.8185 ltrev does not granger cause lrgdp 32 7.4807 ** 0.0105 lrgdp does not granger cause ltrev 1.1113 0.3005 llnoilrev does not granger cause loirev 32 0.6194 0.4376 loirev does not granger cause llnoilrev 6.8965 ** 0.0136 ltrev does not granger cause loirev 32 1.3775 0.2501 loirev does not granger cause ltrev 0.3326 0.5686 ltrev does not granger cause llnoilrev 32 7.3813 ** 0.0110 llnoilrev does not granger cause ltrev 0.0431 0.8370 source: authors’ computation using eviews 9 software. note: *. ** and *** denote 1%, 5% and 10% level of significance respectively. 4.3.4. ardl coefficients for long-run form results having discovered that the variables are cointegrated, the next agendum is to proceed to the estimation of a short-run dynamic and long-run estimation using the ardl method of estimation. table 5 presents the long-run form results for all three models. beginning from the total revenues model, it can be observed that real gdp and total government revenues are positively and significantly related. specifically, a 1% increase in government total revenues leads to 0.12% increase in economic growth, holding other independent variables constant (henceforth the assumption of other independent variables held constant is applicable to all).10 similarly, oil revenue and non-oil revenue and economic growth have a positive and significant relationship. for example, a 1% increase in oil and non-oil revenues will lead to 0.118% and 0.092% respectively. these results show that government revenues, particularly those realised from sales of crude oil, are crucial to economic growth in nigeria. therefore, an increase in 10 this is done to avoid repetition. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 77 government revenues leads to an increase in economic growth. the results are akin to empirical findings by hamdi and sbia (2013), ahmad and masan (2015), jone et al. (2015) and ude and agodi (2014). specifically, hamdi and sbia (2013) found that oil revenue is the main source for economic growth through the channel of financing government spending. similarly, jone et al. (2015) concluded that there is long-run relationship between the real gdp, government expenditure and the government revenues. on the relationship between non-oil revenue and economic growth in nigeria, ude and agodi (2014) noted that non-oil revenues such as agricultural revenue and manufacturing revenue have both a short-run dynamic and long-run relationship with economic growth. it must also be stated that the results could suggest indirectly that a reduction in government revenues will result in an economic growth downturn. this implies that government needs to take the issue of management of its revenues seriously and channel its realised revenues to productive projects that will not only lead to growth that is level, sustained and inclusive. this becomes important considering the source from which the largest chunk of revenues is coming. any internal or external disturbances to the source of revenues will be detrimental for the economy and by extension increase poverty. in addition, human capital, proxied by secondary school enrolment, is also important to economic growth as well as investment (gross fixed capital formation). the two variables (human capital and investment) have positive and significant effects on economic growth. spefically, if human capital and investment increase by 1% in all three models, economic growth will increase by 0.527%, 0.541% , 0.815 and 0.270%, 0.275% and 0.196% respectively. this result is not surprising as the literature is replete with empirical evidence of impact of human capital development and investment on economic growth both in developed and developing countries (see barro, 1991; barro and lee, 2010; cohen and soto, 2007; hanusheck and woessmann, 2009). moreover, it is found that trade openness, a measure of how a country is opened to the rest of the world in terms of trade both in goods and services as well as capital transactions) exhibits a negative relationship with economic growth. thus, an increase in trade openness by 1% will dampen economic growth in all three models by 0.005%, 0.003% and 0.002% respectively. this finding may not be unconnected to the overdependence of the country on imports of all sorts of goods from foreign countries which, over the year, have had negative impacts on the manufacturing sector. however, foreign direct investment (fdi), though it has a positive relationship with economic growth, is not statistically significant.11 4.3.5. ardl cointegration for short-run model results in this subsection, we estimate error the correction mechanism using the adrl method to examine the short-run relationship among the variables. the results are presented in table 6. we can observe that the coefficients of ect follow a priori expectation. specifically, the coefficients are not only negative but also statistically significant at the 1% level of significance. this shows that there is a short-run dynamic adjustment towards the long-run equilibrium. the 11 the coefficient of each variable can also be explained in terms of elasticity isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 78 magnitudes of these coefficients are quite higher which depict a quicker return to the long-run equilibrium in case there is disequilibrium in the system. to be specific, the error correction term coefficients in all the three models are -0.761, -0.731 and -0.645 respectively. this shows that 76.11%, 73.13% and 64.45% errors are corrected for respectively and that it will take less than one-half years for the economics to converge to the long-run equilibrium. table 5. long-run model results. dependent variable: real gross domestic product (rgdp) variable total revenues oil revenues non-oil revenues constant 20.4488 (0.0000) 20.3351 (0.0000) 21.2272 (0.0000) ltrev 0.1197 (0.0000) lorev 0.1182 (0.0000) lnorev 0.0915 (0.0000) lse 0.5272 (0.0019) 0.5414 (0.0019) 0.8154 (0.0006) open -0.0052 (0.0046) -0.0027 (0.0058) -0.0018 (0.2891) lfdi 0.0201 (0.4075) 0.0184 (0.4621) 0.0471 (0.1964) inv_gdp 0.2702 (0.0000) 0.2752 (0.0000) 0.1957 (0.0047) source. authors’ computation using eviews 9 software. note. probability values that signify the level of significance are put in parentheses. as in the case of the long-run estimated model, total revenues and oil revenue are positively and significantly related to economic growth. however, the positive impact of elasticity coefficients of the long-run model is higher than that of the short-run model. this implies that over time government-realised revenue from the sales of crude oil, per adventure due its investment in the critical sectors of the economy, give rise to economic growth in the long-run. it is, however, observed that non-oil revenue though still having a positive relationship with economic growth is not statistically significant in the short-run. this is understandable considering the meagre amount of money being realised from those sectors of the economy. it is found that in the short-run human capital and investment still maintain positive and significant relationships with economic growth, though, at attenuated rates when compared with their effects on economic growth in the long-run model. trade openness, in total revenues and oil revenue models, still exhibits a negative effect on economic growth at the 10% level of significance, however its lag in one period has a positive effect on economic growth at the 1% level of significance. this shows that the initial opening of the economy to the rest of the world may be profitable though dangerous over time. thus, government and its agencies have to be wary with economic opening. finally, foreign direct investment still does not have statistically significance in the short-run. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 79 table 6. autoregressive distribution lag (adrl) cointegrating model results. dependent variable: real gross domestic product (rgdp). variable total revenues oil revenues non-oil revenues d(lrgdp(-1)) 0.4138 (0.0094) 0.3997 (0.0106) 0.4073 (0.0244) d(ltrev) 0.0911 (0.0001) d(loirev) 0.0864 (0.0000) d(lnorev) 0.0350 (0.2542) d(lnolrev(-1)) -0.0485 (0.1030) d(lse) 0.4012 (0.0025) 0.3959 (0.0026) 0.5340 (0.0021) d(lse(-1)) -0.3033 (0.0159) d(openess) -0.0018 (0.0788) -0.0017 (0.0829) -0.0012 (0.3118) d(openess(1)) 0.0024 (0.0096) 0.0024 (0.0084) d(lfdi) 0.0153 (0.4130) 0.0135 (0.4673) 0.0304 (0.1973) d(in_gdp) 0.2056 (0.0006) 0.2013 (0.0005) 0.1261 (0.0326) cointeq(-1) -0.7611 (0.0000) -0.7313 (0.0000) -0.6445 (0.0003) source: author’s computation using eviews 9 software. note: probability values that signify the level of significance in parentheses. 4.3.6. diagnostic test analysis table 7 presents the results of diagnostic tests for the ardl model estimated above. the tests were carried out because the validity of ardl results rests on the satisfaction of the assumptions of classical ols such as normality, linearity, no serial correlation and homoscedasticity. each of these tests has its null hypothesis against which the alternative hypothesis is tested. for example, the null hypothesis under the linearity assumption states that the model is linear in parameter while the null hypothesis of normality test is that the model is normally distributed with zero mean and constant variance. in the same vein, lm heteroscedasticity and serial correlation tests rest on the null hypotheses of homoscedasticity (equal variance) and no serial correlation respectively. the decision is made based on the nonrejection of the null hypothesis. according to the table 7, the results show that all the models pass the tests conducted because the null hypothesis for each test cannot be rejected. this implies that the models are linear in parameter, normally distributed with zero mean and constant variance, homoscedastic (have equal variance) and suffer no serial correlation. thus, the models are reliable and can be employed for economic policy formulation, forecasting and prediction. isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 80 table 7. sensitivity/diagnostic tests test total revenues model oil revenues model non-oil revenues model jacque-bera 0.8229 (0.6627) 0.4279 (0.8074) 1.9677 (0.3739) serial correlation lm test 0.9637 (0.3994) 1.1147 (0.3486) 0.5051 (0.6122) arch heteroscedasticity test 0.6841 (0.4152) 0.6807 (0.4163) 0.0992 (0.7551) linearity test 0.0424 (0.8390) 1.20e-05 (0.9973) 0.8867 (0.3588) source: authors’ computation using eviews 9 software note: probability values that signify the level of significance in parentheses. table 8. robustness check results. dependent variable: industrial production index variable total revenues oil revenues non-oil revenues constant 0.6737 (0.5991) 0.5873 (0.6428) 0.8111 (0.5428) ltrev 0.0442 (0.0431) lorev 0.0427 (0.0465) lnorev 0.0425 (0.0473) lse 0.1204 (0.5015) 0.1253 (0.4842) 0.1261 (0.4819) open 0.0016 (0.4220) 0.0017 (0.4121) 0.0018 (0.3985) lfdi 0.0921 (0.0319) 0.0922 (0.0326) 0.1005 (0.0161) inv_gdp 0.0374 (0.3951) 0.0404 (0.3600) 0.0282 (0.5233) r2 0.8924 0.8919 0.8917 adj r2 0.8724 0.8718 0.8717 f-stat (prob) 44.776 (0.0000) 44.541 (0.0000) 44.483 (0.0000) source: authors’ computation using eviews 9 software. note: probability values that signify the level of significance in parentheses. 4.3.7. robustness check in addition to assessing the impact of government revenues on economic growth proxied by real gdp and using the ardl estimation approach, we test consistency of our findings above by using other means of economic indicator and estimation techniques. in this case, we used industrial production index (ipi) as a proxy for economic performance and ols as a method of isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 81 estimation.12 the results obtained from this exercise, however, are not different from that obtained using real gdp as a proxy for economic growth and the ardl estimation method. specifically, the results in table 8 show that total revenues, oil revenues and non-oil revenues are positively and significantly linked with the industrial production index. it therefore implies that government revenues are indispensable to the sustenance of the nigerian economy. 5. conclusion and policy recommendations the dynamic relationship between government revenues and economic growth has been examined in this study using the ng-perron unit root test technique, the johansen cointegrated approach and the autoregressive distribution lag method. the results reveal that all revenues considered (total revenue, oil revenue and non-oil revenue) have positive effects on economic growth in both the short-run and long-run. however, it is discovered that economic growth is more responsive to oil revenues than non-oil revenues. this explains in part the rationale for economic problem whenever there is revenue shortage, occasioned most of the time by a declining oil price in the international market. based on the results above, it is recommended that the revenues accrued to government should be frugally channelled to the critical sectors of the economy for rapid and sustainable economic growth. specifically, government should make a concerted effort to ensure that accrued revenues are invested in the infrastructural facilities such as electricity, good roads, health care, pipe-borne water and tourism that will improve the environment and encourage economic activities. during the oil boom particularly when the oil price increases in the international market and revenues are accrued to the government, the latter should set aside money for rainy days so as to avoid the current socioeconomic crisis in the country which is caused by lack of funds. this can be achieved by keeping excess oil revenue in a special account which will be solidly backed by the law that will prevent exuberant spending. examples of this revenue management can be adopted from other oil-producing countries such as norway, south arabia and the united arab emirates, which have been able to manage their oil revenues successfully. the saved money can be used to reflate the economy during economic recession in the future. above all, the other sectors of the economy should be improved upon or made attractive for foreign investors so that more revenues that will serve as shock absorbers against oil price volatility can be generated. references adel, shakeeb m, (2015) 'effects of oil and non-oil exports on the economic growth of syria', academic journal of economic studies, vol. 1, no. 2, pp. 69-78 12 by definition, the industrial production index is an economic indicator used to measure the amount of output produced by different manufacturing industries in the economy while ols is the method used in economics or statistics to estimate unknown parameters in a linear regression which could be simple regression or multiple regression isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 82 aghion, philippe and howitt, peter (1992) 'a model of growth through creative destruction' econometrica, 60:323–351. https://doi.org/10.2307/2951599 ahmad, ahmad h. and masan salem (2015) 'dynamic relationship between oil revenue, government spending and economic growth' international journal of business and economic development, vol 3, no.2 barro, robert (991) 'economic growth in a cross section of countries', quarterly journal of economics, 106(2):407–43 ang, james b. (2008) 'a survey of recent developments in the literature of finance and growth', journal of economic surveys, 22(3):536–576. https://doi.org/10.1111/j.14676419.2007.00542.x ayinde, kayode; kuranga, john and lukman adewale f. (2015) 'modelling nigerian government expenditure, revenues and economic growth: co-integration, error correction mechanism and combined estimators analysis approach', asian economic and financial review, 5(6):858-867. https://doi.org/10.18488/journal.aefr/2015.5.6/102.6.858.867 baghestani, hamid and mcnown robert (1994) 'do revenues or expenditures respond to budgetary disequilibria?' southern economic journal, 61, (2), 311-322. https://doi.org/10.2307/1059979 barro, robert. j. and lee jong-wha (1994) 'sources of economic growth', carnegie-rochester conference series on public policy 40:1-46. https://doi.org/10.1016/0167-2231(94)90002-7 barro, robert and lee jong-wha (2010) 'a new data set of educational attainment in the world', national bureau of economic research working paper no. 15902, massachusetts. boarini, romanian; asa johansson and macro mira d'ercode (2006) 'alternative measures of wellbeing', oecd social, employment and migration working paper 33. buchanan, james m. and wagner richard w. 1978) 'dialogues concerning fiscal religion' journal of monetary economics, 4,627-636. https://doi.org/10.1016/0304-3932(78)90056-9 cass, david (1965) 'optimum growth in an aggregative model of capital accumulation' review of economic studies 32, 233–240. https://doi.org/10.2307/2295827 cohen, daniel and soto marcelo (2007) 'growth and human capital: good data, good results' journal of economic growth, 12:51–76. https://doi.org/10.1007/s10887-007-9011-5 dreger, christian and rahmani termur (2014) 'the impact of oil revenues on the iranian economy and the gulf states' discussion paper. engen, eric and skinner jonathan (1996) 'taxation and economic growth', national tax journal, vol.49, no. 4, pp. 617-642. https://doi.org/10.3386/w5826 friedman, milton (1978) 'the limitation of tax limitation' policy review, 7-14. grossman, gene m. and helpman elhanan (1991) innovation and growth in the global economy. cambridge, ma: mit press hanushek, eric a. and woessmann ludger (2009) do better schools lead to more growth? cognitive skills, economic outcomes, and causation nber working paper no. 14633, national bureau of economic research, massachusetts hamdi, helmi and sbia rashid (2013) 'dynamic relationships between oil revenues, government spending and economic growth in an oil-dependent economy' journal economic https://doi.org/10.2307/2951599 https://doi.org/10.1111/j.1467-6419.2007.00542.x https://doi.org/10.1111/j.1467-6419.2007.00542.x https://doi.org/10.18488/journal.aefr/2015.5.6/102.6.858.867 https://doi.org/10.2307/1059979 https://doi.org/10.1016/0167-2231(94)90002-7 https://doi.org/10.1016/0304-3932(78)90056-9 https://doi.org/10.2307/2295827 https://doi.org/10.1007/s10887-007-9011-5 https://doi.org/10.3386/w5826 isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 83 modelling 35, 118-125, http://dx.doi.org/10.1257/aer.20110236. ibeh, francisca ujunwa (2013) 'the impact of oil revenues on the economic growth in nigeria' caristas university. jubrin, success musa; blessing success ejura and ifurueze m.s.k. (2014) 'impact of petroleum profit tax on economic development of nigeria' british journal of economics, finance and management sciences, vol. 5 (2), pp 60-70. jones, charles i. and klenow peter j. (2016) 'beyond gdp? welfare across countries and time', america economies review, 106(9), 2426-2457. https://doi.org/10.1257/aer.20110236 jones, ebieri; ihundinihu john uzoma. u. and nwaiwu, j. n. (2015) 'total revenue and economic growth in nigeria: empirical evidence' journal of emerging trends in economics and management sciences (jetems), 6(1): 40-46 tuffour, joseph k (2013) 'foreign aid, domestic revenues and economic growth in ghana', journal of economics and sustainable development i (online) vol.4, no.8, www.iiste.org kabir, maryam (2016) 'long-run relationship between oil revenues and economic growth in nigeria' archives of business research, 4(2), 37u47 keynes, john maynard (1936) (1964) the general theory of employment, interest, and money, new york: harcourt-brace & world, inc. koopmans, tjalling c. (1965) 'on the concept of optimal economic growth: in the economic approach to development planning', amsterdam: elsevier. levine, ross and zervos sara (1998) 'stock markets, banks and economic growth', american economic review, 88: 537–558 levine, ross (1997) 'financial development and economic growth: views and agenda', journal of economic literature, xxxv:688–726 lucas, robert e. jr. (1988) 'on the mechanics of economic development', journal of monetary economics 22, 3–42. https://doi.org/10.1016/0304-3932(88)90168-7 abata, matthew a. (2014) 'the impact of tax revenues on nigerian economy: case of federal board of inland revenues', journal of policy and development studies, vol. 9, no. 1 november 2014, website: www.arabianjbmr.com/jpds_index.php meltzer, allan h. and richard scott.f. (1981) 'a rational theory of size of government', journal of political economy, 89, 914-927. https://doi.org/10.1086/261013 mirrlees, james a (1971) 'an exploration in the theory of optimal income taxation' review of economic studies 38, 175-208. https://doi.org/10.2307/2296779 musgrave, richard (1966) 'principles of budget determination', in h. cameron and w. henderson (eds.), public finance selected reading. new york: random house. pg. 15-27 north, douglass c. (1990) institutions, institutional change and economic performance. cambridge university press, cambridge. https://doi.org/10.1017/cbo9780511808678 odularu, gbadebo o. (2008) crude oil and the nigerian economic performance: oil and gas business, http://www.ogbus.ru/eng okafor, regina g. (2012) 'tax revenues generation and nigerian economic development', european journal of business and management issn 2222-1905 (paper) issn 2222-2839 http://dx.doi.org/10.1257/aer.20110236 https://doi.org/10.1257/aer.20110236 https://doi.org/10.1016/0304-3932(88)90168-7 https://doi.org/10.1086/261013 https://doi.org/10.2307/2296779 https://doi.org/10.1017/cbo9780511808678 http://www.ogbus.ru/eng isiaka akande raifu and abiodun najeem raheem / european journal of government and economics 7(1), 60-84. 84 (online) vol. 4, no.19, 2012, www.iiste.org peacock, alan t. and wiseman jack (1979) 'approaches to the analysis of government expenditure growth', public finance review, 7, 3-23. https://doi.org/10.1177/109114217900700101 ramsey, frank (1927) 'a contribution to the theory of taxation', economic journal, 37, 47-61. https://doi.org/10.2307/2222721 ricardo, david (1820), 'funding system' in sraffa, p. (ed.) (1951) the works and correspondence of david ricardo, sraffa, piero. (ed.) (1951), cambridge university press. romer, paul m. (1986) 'increasing returns and long run growth', journal of political economy, 94, 1002–1037. https://doi.org/10.1086/261420 romer, paul. m. (1990) 'endogenous technological change' journal of political economy, 98, s71–s102. https://doi.org/10.1086/261725 singh, ajit (1997) 'financial liberalisation, stock markets and economic development', economic journal, 107: 771–782. https://doi.org/10.1111/j.1468-0297.1997.tb00042.x slemrod, joel (1990) 'optimal taxation and optimal tax systems' journal of economic perspectives, 4(1), p158. https://doi.org/10.1257/jep.4.1.157 solow, robert m. (1956) 'a contribution to the theory of economic growth' quarterly journal of economics, 70, 65–94. https://doi.org/10.2307/1884513 ude, damian kalu and agodi, joy e. (2014) 'investigation of the impact of non-oil revenues on economic growth in nigeria', international journal of science and research (ijsr), issn (online): volume 3 issue 11, pp. 2571-2577 undp, (1990). human development report. new york: oxford university press. victor, ushahemba ijirshar (2015) 'the empirical analysis of oil revenues and industrial growth in nigeria', african journal of business management, vol. 9(16), pp. 599607. https://doi.org/10.5897/ajbm2015.7801 wagner, adolph (1893) grundlegung der politischen okonomie. leipzig: c. f. winter. http://www.iiste.org/ https://doi.org/10.1177/109114217900700101 https://doi.org/10.2307/2222721 https://doi.org/10.1086/261420 https://doi.org/10.1086/261725 https://doi.org/10.1111/j.1468-0297.1997.tb00042.x https://doi.org/10.1257/jep.4.1.157 https://doi.org/10.2307/1884513 https://doi.org/10.5897/ajbm2015.7801 abstract. the bursting of crude oil prices in the international market since mid-2014 has resulted in dwindling oil revenue, which has led to economic recession in nigeria. the recession has further exacerbated existing socioeconomic problems bedevili... keywords. oil revenue; economic growth; ardl jel classification. h20; h27; o40; c21 references adel, shakeeb m, (2015) 'effects of oil and non-oil exports on the economic growth of syria', academic journal of economic studies, vol. 1, no. 2, pp. 69-78 aghion, philippe and howitt, peter (1992) 'a model of growth through creative destruction' econometrica, 60:323–351. https://doi.org/10.2307/2951599 ahmad, ahmad h. and masan salem (2015) 'dynamic relationship between oil revenue, government spending and economic growth' international journal of business and economic development, vol 3, no.2 barro, robert (991) 'economic growth in a cross section ... ang, james b. (2008) 'a survey of recent developments in the literature of finance and growth', journal of economic surveys, 22(3):536–576. https://doi.org/10.1111/j.1467-6419.2007.00542.x ayinde, kayode; kuranga, john and lukman adewale f. (2015) 'modelling nigerian government expenditure, revenues and economic growth: co-integration, error correction mechanism and combined estimators analysis approach', asian economic and financial re... baghestani, hamid and mcnown robert (1994) 'do revenues or expenditures respond to budgetary disequilibria?' southern economic journal, 61, (2), 311-322. https://doi.org/10.2307/1059979 barro, robert. j. and lee jong-wha (1994) 'sources of economic growth', carnegie-rochester conference series on public policy 40:1-46. https://doi.org/10.1016/0167-2231(94)90002-7 barro, robert and lee jong-wha (2010) 'a new data set of educational attainment in the world', national bureau of economic research working paper no. 15902, massachusetts. boarini, romanian; asa johansson and macro mira d'ercode (2006) 'alternative measures of wellbeing', oecd social, employment and migration working paper 33. buchanan, james m. and wagner richard w. 1978) 'dialogues concerning fiscal religion' journal of monetary economics, 4,627-636. https://doi.org/10.1016/0304-3932(78)90056-9 cass, david (1965) 'optimum growth in an aggregative model of capital accumulation' review of economic studies 32, 233–240. https://doi.org/10.2307/2295827 cohen, daniel and soto marcelo (2007) 'growth and human capital: good data, good results' journal of economic growth, 12:51–76. https://doi.org/10.1007/s10887-007-9011-5 dreger, christian and rahmani termur (2014) 'the impact of oil revenues on the iranian economy and the gulf states' discussion paper. engen, eric and skinner jonathan (1996) 'taxation and economic growth', national tax journal, vol.49, no. 4, pp. 617-642. https://doi.org/10.3386/w5826 friedman, milton (1978) 'the limitation of tax limitation' policy review, 7-14. grossman, gene m. and helpman elhanan (1991) innovation and growth in the global economy. cambridge, ma: mit press hanushek, eric a. and woessmann ludger (2009) do better schools lead to more growth? cognitive skills, economic outcomes, and causation nber working paper no. 14633, national bureau of economic research, massachusetts hamdi, helmi and sbia rashid (2013) 'dynamic relationships between oil revenues, government spending and economic growth in an oil-dependent economy' journal economic modelling 35, 118-125, http://dx.doi.org/10.1257/aer.20110236. ibeh, francisca ujunwa (2013) 'the impact of oil revenues on the economic growth in nigeria' caristas university. jubrin, success musa; blessing success ejura and ifurueze m.s.k. (2014) 'impact of petroleum profit tax on economic development of nigeria' british journal of economics, finance and management sciences, vol. 5 (2), pp 60-70. jones, charles i. and klenow peter j. (2016) 'beyond gdp? welfare across countries and time', america economies review, 106(9), 2426-2457. https://doi.org/10.1257/aer.20110236 jones, ebieri; ihundinihu john uzoma. u. and nwaiwu, j. n. (2015) 'total revenue and economic growth in nigeria: empirical evidence' journal of emerging trends in economics and management sciences (jetems), 6(1): 40-46 tuffour, joseph k (2013) 'foreign aid, domestic revenues and economic growth in ghana', journal of economics and sustainable development i (online) vol.4, no.8, www.iiste.org kabir, maryam (2016) 'long-run relationship between oil revenues and economic growth in nigeria' archives of business research, 4(2), 37u47 keynes, john maynard (1936) (1964) the general theory of employment, interest, and money, new york: harcourt-brace & world, inc. koopmans, tjalling c. (1965) 'on the concept of optimal economic growth: in the economic approach to development planning', amsterdam: elsevier. levine, ross and zervos sara (1998) 'stock markets, banks and economic growth', american economic review, 88: 537–558 levine, ross (1997) 'financial development and economic growth: views and agenda', journal of economic literature, xxxv:688–726 lucas, robert e. jr. (1988) 'on the mechanics of economic development', journal of monetary economics 22, 3–42. https://doi.org/10.1016/0304-3932(88)90168-7 abata, matthew a. (2014) 'the impact of tax revenues on nigerian economy: case of federal board of inland revenues', journal of policy and development studies, vol. 9, no. 1 november 2014, website: www.arabianjbmr.com/jpds_index.php meltzer, allan h. and richard scott.f. (1981) 'a rational theory of size of government', journal of political economy, 89, 914-927. https://doi.org/10.1086/261013 mirrlees, james a (1971) 'an exploration in the theory of optimal income taxation' review of economic studies 38, 175-208. https://doi.org/10.2307/2296779 musgrave, richard (1966) 'principles of budget determination', in h. cameron and w. henderson (eds.), public finance selected reading. new york: random house. pg. 15-27 north, douglass c. (1990) institutions, institutional change and economic performance. cambridge university press, cambridge. https://doi.org/10.1017/cbo9780511808678 odularu, gbadebo o. (2008) crude oil and the nigerian economic performance: oil and gas business, http://www.ogbus.ru/eng okafor, regina g. (2012) 'tax revenues generation and nigerian economic development', european journal of business and management issn 2222-1905 (paper) issn 2222-2839 (online) vol. 4, no.19, 2012, www.iiste.org peacock, alan t. and wiseman jack (1979) 'approaches to the analysis of government expenditure growth', public finance review, 7, 3-23. https://doi.org/10.1177/109114217900700101 ramsey, frank (1927) 'a contribution to the theory of taxation', economic journal, 37, 47-61. https://doi.org/10.2307/2222721 ricardo, david (1820), 'funding system' in sraffa, p. (ed.) (1951) the works and correspondence of david ricardo, sraffa, piero. (ed.) (1951), cambridge university press. romer, paul m. (1986) 'increasing returns and long run growth', journal of political economy, 94, 1002–1037. https://doi.org/10.1086/261420 romer, paul. m. (1990) 'endogenous technological change' journal of political economy, 98, s71–s102. https://doi.org/10.1086/261725 singh, ajit (1997) 'financial liberalisation, stock markets and economic development', economic journal, 107: 771–782. https://doi.org/10.1111/j.1468-0297.1997.tb00042.x slemrod, joel (1990) 'optimal taxation and optimal tax systems' journal of economic perspectives, 4(1), p158. https://doi.org/10.1257/jep.4.1.157 solow, robert m. (1956) 'a contribution to the theory of economic growth' quarterly journal of economics, 70, 65–94. https://doi.org/10.2307/1884513 ude, damian kalu and agodi, joy e. (2014) 'investigation of the impact of non-oil revenues on economic growth in nigeria', international journal of science and research (ijsr), issn (online): volume 3 issue 11, pp. 2571-2577 undp, (1990). human development report. new york: oxford university press. victor, ushahemba ijirshar (2015) 'the empirical analysis of oil revenues and industrial growth in nigeria', african journal of business management, vol. 9(16), pp. 599607. https://doi.org/10.5897/ajbm2015.7801 wagner, adolph (1893) grundlegung der politischen okonomie. leipzig: c. f. winter. vol.8 • no.2 2019 issn: 2254-7088 european journal of government and economics 8(2), december 2019. european journal of government and economics    issn: 2254‐7088  number 8, issue 2, december 2019 doi: https://doi.org/10.17979/ejge.2018.8.2 why regions fail (or succeed). the role of government institutions in the long-run 114-144 doi: https://doi.org/10.17979/ejge.2019.8.2.4989 filippo bonanno restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case 145-160 doi: https://10.17979/ejge.2019.8.2.5478 joão ricardo catarino and ricardo de moraes e soares fdi in selected developing countries: evidence from bundling and unbundling governance 161-188 doi: https://doi.org/10.17979/ejge.2019.8.2.4970 simplice asongu the relationship between corporate tax rate and economic growth during the global financial crisis: evidence from a panel var governanceand domestic investment in africa 189-202 doi: https://10.17979/ejge.2019.8.2.5074 gamze oz-yalaman the impact of crd iv on bank lending 203-217 doi: https://doi.org/10.17979/ejge.2019.8.2.4656 matias huhtilainen european journal of government and economics 8(2), december 2019, 114-144 european journal of government and economics issn: 2254-7088 why regions fail (or succeed). the role of government institutions in the long-run filippo bonannoa* a universidad complutense de madrid, spain * corresponding author at: filippobonanno3@gmail.com article history. received 21 january 2019; first revision required 1 august 2019; accepted 29 september 2019. abstract. this paper represents an attempt to reconcile some general intuitions provided by daron acemoglu and james a. robinson in the book “why nations fail” with the case of the deep regional disparities in the economic performances observed within the “western” european union during the period 2001-2015. by adopting an approach to growth analysis based on binary response models, this paper quantifies the extent to which the quality of government institutions has shaped regional economic performances in the european union throughout the period comprising the great recession. empirical results show that: 1) the higher is the quality of institutions, the higher is the probability that a region with high income per capita will grow above the levels of the european union as a whole. 2) the higher is the quality of institutions, the lower is the probability that a low-income region will grow below the levels of european union as a whole. 3) the higher is the quality of institutions, the higher (lower) is the probability that any region, regardless of its income per capita, will outperform (underperform) the european union as a whole. 4) the higher is the quality of institutions, the lower is the probability that a region will “fail” to grow. keywords. quality of institutions; probit; regions; european union, economic performances jel codes. h1; n4; r1; c10 doi. https://doi.org/10.17979/ejge.2019.8.2.4989 1. introduction the period 2001-2015 was characterized by very poor regional economic performances in the old member states european union (eu-15) 1 . however, despite the financial crisis and economic downturn that affected the european countries and in particular in the euro area, some regions performed much better than others while a group of 44 regions recorded negative rates of average economic growth. the worst economic performances were generally observed in the italian and greek regions that are also the region with the poorest quality of government institutions. the objective of this essay is to analyse the role of the quality of government in shaping regional performances and intends to contribute to the existing literature about the relationship between economic growth and institutions by adopting an ex-post approach for the analysis of 1 by eu-15 it is meant the group of countries from western europe that have progressively joined the european community/european union long before the waves of enlargements towards the east during the 2000s f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 115    economic growth. regions will be grouped according to their economic performances compared to the performance of the whole european union over the period 2001-2015 and four binary dependent variables will be constructed. these binary dependent variables in the dataset will take on value 1 if one of the four performances is observed or 0 otherwise. next, a probit model will be estimated for each of the four cases in order to explain to what extent institutional quality has shaped regional economic performances during the period under analysis. the empirical findings are remarkably interesting. both the educational attainment of the working-age population and the industrial sector generally play a positive role in shaping regional performances. however, in the long run, the quality of institutions turns out to be always the ultimate determinant of the increase/decrease in the probability that a region will be recording a positive or negative economic performance. 2. literature review neoclassical growth model and its empirical extensions for growth analysis have traditionally focused on exogenous factors such as saving rate, population growth and technological progress (solow, 1956). within the stream of the neoclassical approach, mankiw et al. (1991) have also remarked the relevant role played by human capital in the transitional dynamics to the steady state output per worker. other streams of theoretical and empirical literature about economic growth have focused on human capital and endogenous technological change arguing that innovation is the engine of economic growth and it is endogenously generated (romer, 1986; lucas, 1988). the consequence of the fact that human capital accumulation is not subjected to decreasing returns as physical capital can explain the discontinuity or the slow speeds of convergence often observed within groups of regions or countries (martin and sunley, 1998). by including variables that can broadly be considered as proxies for institutional quality in a cross-section of 98 countries, barro (1991) finds out that economic growth is positively related to measures of political stability and negative related to the market distortions induced in the economy by the political sphere. also in a panel of around 100 countries with data collected from 1960 to 1990, barro (1996) finds out that for a given level of initial gdp per capita, the growth rate is enhanced among others also by better maintenance of the rule of law. as acemoglu et al. (2005:397) remark, although cultural and geographical forces also matter for economic performances, differences in the quality of institutions are the major source of cross-country differences in economic growth and prosperity. north (1990:3) defines institutions as the rules of the game in a society or, more formally, the humanly devised constraints that shape human interaction. according to alonso (2009:9), institutional structure defines the incentives and penalties that influence the behaviour of agents and shape collective action. therefore, in the uncertain world in which independent agents operate with imperfect information, sound institutions reduce uncertainty and transaction costs and facilitate social coordination. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 116    acemoglu and robinson (2012) argue that rich countries are rich because they have inclusive institutions, while poor countries are poor because they have extractive institutions. by “institutions” also the authors mentioned above mean the rules that govern and organize the economic and political life. inclusive institutions create the fundamental incentives and opportunities that stimulate investment and entrepreneurship, while extractive economic institutions consist of a system where a small group of people is permitted to exploit the rest of the population that is kept out the political and economic process. however, throughout history, most of the societies have been ruled by extractive economic institutions with different levels of intensity 2. acemoglu et al. (2001) individuate the origins of the economic backwardness of former western colonies in asia, africa and latin america in the fact that colonial powers set up extractive state structures in those territories. those institutions did not introduce much protection for private property, nor did they provide checks and balances against the government because the explicit aim of the europeans settled in those overseas territories was the sole extraction of resources. this colonization strategy and the associated institutions contrast with the institutions europeans set up in other colonies where they settled in large numbers, for example, the united states, canada, australia, and new zealand (acemoglou and robinson, 2008:4). at the same time, the quality of institutions also affected the economic development of the colonial powers themselves. achemoglu et al. (2002) argue that the discovery of america in 1492 benefited much more, in terms of economic development, the countries that had already established a systems of checks and balance to the monarchy, as great britain and the netherlands, rather than countries where the monarchy was highly absolutist, as in the case of spain or portugal. the institutionalist explanation of disparities in economic development has been well received in academic circles and in the sphere of international organisations (alonso, 2009:12). as a matter of facts, it has progressively arisen an increasing recognition among practitioners in international organizations that corruption and other aspects of poor governance have substantial and adverse effects on economic development (mauro, 2002). mauro (1995) also finds out a negative relationship between the malfunctioning of institutions and investment rate, therefore between institutional inefficiency (and corruption) and economic growth. somewhat contrary to the capital accumulation model of regional growth, institutional theorists argue that differences in growth and prosperity across countries, regions and cities are strictly related to the quality of political and economic institutions that shape the economic activity (huggins and thompson, 2017). 2 in relation to “extractive” economic institutions in the ancient history, prominent historian macmullen (1988) has argued that also the decline and consequent fall of the roman empire was mainly due to the progressive erosion of the solidity of government institutions pursued by a small group of high-ranking bureaucrats and military leaders. the spread of corruption and informal practices had long term devastating effects on the political and economic integrity of the institutional foundations roman empire. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 117    by estimating the contribution to income levels of different forces like geography and trade in a large sample of countries, also rodrik et al. (2002) find out that the quality of institutions is the main determinant of income levels. roughly said, after controlling for institutional quality, in the output of the regression model measures of geography and trade result at best to exert weak effects on income levels while the institutional quality remains the main determinant of economic development. turning to the more focused topic of regional performances in europe, a wide body of literature has been produced to study the relationship between economic growth and the quality of institutions. however, most of these studies use “quality of institutions” as a predictor of economic growth with a “neoclassical approach” where economic growth is regressed on the natural logarithm of initial levels of gdp per capita and other control variables, including quality of institutions or regional geographic spillover (ascani et al., 2012; harris, 2008; feldkircher, 2006; rodríguez-pose, 1998; pons-novell and viladecans-marsal, 1998). other empirical studies assess the impact of the quality of institutions on innovative capacity and investment in european regions or countries (rodriguez -pose and di cataldo, 2018; canton and solera, 2016). this essay contributes to the existing literature by modelling the relationship between regional economic performances and the quality of institutions with an econometric approach based on binary response models. these models attempt to estimate the marginal probability that a certain economic performance will be attained in a region, conditional on the quality of institutions and other control variables commonly used in the analysis of economic growth. 3. data and empirical strategy the annual data are averaged from 2001 to 2015 for each of the 195 sampling units and the dataset includes 14 nuts-1 territories (belgium: brussels; germany: berlin, brandenburg, bremen, hamburg, mecklenburg-vorpommern, saarland, sachsen-anhalt, schleswig-holstein and turingen; france: ile-de-france and nord-pas-de-calais; greece: attica; spain: community of madrid; united kingdom: northern ireland), 1 nuts-0 territory (luxembourg) and 180 nuts-2 regions3 located in eu-15 countries4. the sample is composed only by the regions of western europe where the heterogeneity in terms of initial conditions is smaller (despite the remarkable cross country differences) than the cross regional heterogeneity observed in the post-enlargement european union. therefore, regions of transition economies from eastern europe are not included in the sample because they outperformed the average rate of growth of the european union despite their very low institutional quality and this is mainly due to the fact 3 nuts 0 territories of cyprus and malta are excluded from the sample together with the 7 nuts-2 territories of norway because of the lack of the availability of data concerning quality of institutions and because norway is not a member state of the union. 4 here, by eu-15 group it is meant all the western member states of the european union excluding malta and cyprus. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 118    that they started with very low levels of income per capita their process of transition and economic integration with the western countries of the european union. data related to gdp per capita and investment rates are sourced from cambridge econometrics european regional database. the data concerning the educational attainment of the population are sourced from the regional database made available online from the european commission. data related to the quality of regional governments consist of the eqi score developed by charron at al (2015) from the quality of government institute of the university of gothenburg and made available in the european quality of government index (eqog) database downloadable from the website of the swedish academic institution. eqog database is relatively new, and the indicators have been developed only for the years 2010, 2013 and 2017. given the fact that the “success” or the “failure” of institutions reflects a wide range of historical, local, sociological and anthropological factors (alesina, 2014; greif, 1994), it is possible to suppose that changes and adjustments in the quality of institutions occur very slowly. also, acemoglu and robinson (2012:435) argue that replacing extractive institutions with inclusive institutions is neither an automatic nor a simple process. indeed, it is often needed a convergence of historical or political factors, in particular, a critical conjuncture combined with a broad coalition of people that support and push for the reforms. as a consequence, for the purpose of this analysis, the choice of “eqi score” of the year 2013 as a predictor of economic performances over the period 2001-2015 represents a very reasonable strategy to remedy the fact that the selected time series data for institutional quality are not available on an annual basis from 2001 to 2015. the estimation methodology applied in this essay partially follows the methodology developed by ainginger et al. (2013). regions are assigned to 4 groups according to their economic performances over the period 2001-2015 and their initial income per capita levels: 1. group a (taking off from above): regions with gdp per capita above eu gdp per capita level in 2001 growing faster than the eu average over the period 2001 – 2015. 2. group b (declining from above): regions with gdp per capita above eu gdp per capita level in 2001 growing below the eu average over the period 2001 – 2015 3. group c (converging from below): regions with gdp per capita below eu gdp per capita level in 2001 growing above the eu average over the period 2001 – 2015 4. group d (diverging from below): regions with gdp per capita below eu gdp per capita level in 2001 growing below the eu average over the period 2001 – 2015 regions assigned to group a and c will be further grouped together (a&c) in another group of “successful regions”. regions assigned to group b and d will be further grouped together (b & d) in another group of “unsuccessful regions”. regions that recorded negative rates of income per capita growth will be grouped in the group of “failed” regions (group e). f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 119    the methodology developed in order to construct the binary dependent variables is explained in table 1. table 1. methodology of construction of binary dependent variables. economic performance has the region been “takingoff”? has the region been “declining”? has the region been “converging”? has the region been “diverging”? has the region been “successful”? has the region been “unsuccessful”? has the region been “failed”? taking-off regions (tanking-off from above) (a) 1 0 0 0 0 0 0 declining regions (declining from above) (b) 0 1 0 0 0 0 0 converging regions (converging from below) (c) 0 0 1 0 0 0 0 diverging regions (diverging from below) (d) 0 0 0 1 0 0 0 successful regions (a & c) 0 0 0 0 1 0 0 unsuccessful regions (b & d) 0 0 0 0 0 1 0 after having constructed the four binary variables, with the probit estimation, it will be possible to estimate at first instance what are the structural features that increase (decrease) the probability that a region will be “taking-off” (diverging) with respect to european union as a whole given its gdp per capita in 2001. then it will be possible to estimate both the conditional probabilities that a region will be “successful” (or unsuccessful) at performing better than the european union as a whole in terms of economic growth and the conditional probabilities that a region will “fail” (negative growth) regardless of its gdp per capita levels in 2001. 3.1. some stylized facts about regional economic performances in the european union as already mentioned in the introduction, the period 2001-2015 was characterized on average by very poor regional economic performances because of the fact that most of the countries of the eu (and in particular in the emu) were hit by severe financial and economic crises. figure 1 reports the average rates of regional economic growth observed within the eu as a whole and within the different subgroups of the eu. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 120    figure 1. average annual regional gdp per capita growth by groups of countries (2001-2015) as it possible to notice from figure 1, regional economic growth in the european union was essentially driven by the “converging” economies of eastern europe. indeed, eu-15 regional average economic growth was nearly the half of the economic growth observed in the eu-28 over the period 2001-2015. figure 2 shows the statistical maps of regional performances a, b, a&c and b&d as formalized through the methodology developed in table 1. as it is possible to note in figure 2, most of regions with gdp per capita lower than the european union gdp per capita in 2001 that underperformed the european union (diverging regions) are located in southern europe while the regions that outperformed the european union “from above” are mainly located in central and northern europe (south-east and north germany, most of austrian regions, scotland, north of england and ile-de-france). basque country is the only southern european region that has been “taking-off from above”. at the same time, most of the regions that have been “diverging from below” are located in spain, portugal, italian “mezzogiorno” while most of the regions that have been “unsuccessful” (either by “diverging from below” or “declining from above”) are located in northern and central italy, france and in the southern area of great britain (south of england and wales) plus northern ireland. however, the situation of spain, great britain and france should not be absolutely confused with the situation of italy and greece. indeed, as shown in figure 3, most of the french (14 out 21), british (wells and south of england) and spanish (community of madrid, catalonia and navarra) “declining from above” regions or spanish and portuguese (all the regions excepted metropolitan area of lisbon) “diverging from below” regions underperformed the european union by recording positive (but lower than the eu) rates of economic growth. all the italian “declining from above” and italian and greek “diverging from below” regions underperformed the european union by recording negative average rates of growth. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 121    figure 2. maps of regional performances. figure 3. european regions by group. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 122    3.2. quality of institutions in the european regions. a graphical overview the european quality of government index (eqi) for the year 2013 has been developed by the quality of government institute of the university of gothenburg, and it is the result of a regional survey answered by a large sample of 85.000 citizen respondents5. the data focus “on both perception and experiences with public sector corruption along with the extent to which citizens believe various public sector services are impartially allocated and of good quality”. figure 4 shows the statistical maps of the regional eqi score as calculated for each region of the european union (when the index it is not available at nuts-2 level, the nuts-2 region will take on the score of its upper-level territorial unit nuts-1). as figure 4 shows, eqi score tends to be generally very high or high in northern and central europe (scandinavian, german, austrian and british regions), in france and the majority of regions of the iberian peninsula. at the same time, the wealthiest regions of northern italy like lombardy have the approximately the same eqi score of the madrid community and of catalonia while all the regions of mezzogiorno have the same eqi score of the most of the greek regions. the only italian region with a “northern european” eqi score is the nuts2 region ith1 (trentino-alto adige/südtirol). the regions with the lowest eqi in the western countries of the european union are campania and calabria that are also two of the poorest regions of italy and among the poorest regions in the eu-15. the extent to which the quality of institutions affects regional economic performances can be easily shown by plotting the marginal effects of the eqi score on the predicted probability that a region will be “taking off from above” or “diverging from below”. 5 a more detailed definition of the eqi score is provided in appendix 1. 1<eqi< 2 0<eqi< 1 -1<eqi< 0 -2<eqi< -1 -3<eqi< -2 figure 4 eqi score f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 123    figure 5. probability of “taking-off” from above and “diverging” from below. as it is possible to note in figure 5a, the higher is the eqi score, the higher is the probability that a region will be “taking-off” from above. at the same time, the higher is the eqi score, the lower is the marginal probability that a region will be “diverging” from below (figure 5b). said more technically, as the quality of institutions tends to the maximum score, the predicted probability for a region to be “taking off” approaches 1, while the predicted probability for a region to be “diverging from below” approaches 0. 4. the specification of the model: institutions in the long-run probit regression is nonlinear regression model specifically utilized for binary dependent variables. a model with binary dependent variable models the probability that y=1 when a change in the predictor x occurs. cumulative probability distribution function (c.d.f.) is used in probit regressions because it produces probabilities between 0 and 1 in the form (y=1|x). therefore, when y is binary, its conditional expectation is the marginal probability that the variable y equals 1, and the expected change arising from a change in x is the change in the probability that y=1 (stock and watson, 2012:431). for the probit model f(x´β) is the cumulative density function (c.d.f) of a standard normal distribution 𝑌 𝐹 𝑋`𝛽 𝛷 𝑧 𝑑𝑧 ` 𝑤𝑖𝑡ℎ 𝛷 𝑧 ∈ 0,1 [1] therefore, coefficient estimate β represents the increase/decrease in the z-score of the probability y=1|x when x changes by 1 unit. the model is specified as follows: pr 𝑌 1|𝑋`𝑠 𝛷 𝛼 𝛽 𝑋 , 𝛽 𝑋 , 𝛽 𝑋 , , [2] where y denotes binary dependent variable that can take on value 1 if a region has recorded a f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 124    certain economic performance (“diverging”, “converging”, “unsuccessful”/“successful” or “failed”) and 0 otherwise. x1 denotes the eqi score calculated in 2013, x2 denotes a vector of control variables related to sectorial investment rates to gdp or the regional economic structure while x 3 denotes the vector of control variables related to the educational attainment of the workingage population. all the variables other than eqi score are expressed as annual averages during the period 2001-2015. the model will be estimated for four cases: regions “taking-off” from above, regions “diverging” from below, unsuccessful regions (gdp per capita growth<eu gdp per capita growth), or symmetrically, successful regions (gdp per capita growth > eu gdp per capita growth)6 failed regions (gdp per capita growth < 0). higher eqi score is expected to increase the probability that a region will be “taking off from above” or be “successful” while it is expected to reduce the probability that a region will be “diverging from below” or be “unsuccessful”. investment rates to gdp are expected to affect the binary variable according to the sector of the economy (industry, market services, nonmarket services, agriculture and constructions) while labour force higher education is expected to confirm the results common to the empirical literature about economic growth. in the tables of the regression outputs, i report directly the marginal effects or the predicted probability that y=1 given the values of x1 , x2 …, xk calculated by computing the z-value. indeed, the coefficient β1 is the change in z-value arising from a unit change in x1, holding constant x2 …, x. the probit model is fitted with maximum likelihood estimator and also provides a pseudo r square7, 8. in order to provide a more in-depth analysis, i also run the same models with variables averaged during the period 2001-2008. with this strategy, it is possible both to exclude the crisis period characterized by a general drop in regional economic activity and to compare the role of institutions in the short run and the long run9. 6 the coefficient estimates and their statistical significance of the models with “successful” or “unsuccessful” regions are exactly the same with opposite signs. therefore, in the paper only the regression output for “unsuccessful” regions (regions “declining from above” and “diverging from below”) is reported 7 in table 13 of appendix 2, i will report the regression diagnostic for the detection of the eventual presence of multicollinearity in the model. as it will be shown, no multicollinearity is detected between the variables in the model. 8 figures 10 and 11 in appendix 3 show the plots of the marginal probabilities estimated with the probit technique and reported in tables 2, 3 4 and 5 f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 125    table 2. marginal effects on probability y=1|x. group a. model 1 1b dy/dx dy/dx eqi score .27*** .22*** [.075] [.08] working age population with tertiary education .003 .007** [.003] [.003] working age population primary education or less -.002 -.0006 [.003] [.003] investment in industry to gdp .061*** [.01] investment in construction to gdp -.07 [.07] investment in non-market services gdp -.046** [.02] investment in market services gdp .01 [.009] investment in agriculture to gdp -.20*** [.06] average gva industry .016*** [.005] average gva construction -.022 [.01] average gva non-market services -.014* [.008] average gva market services -.002 [.01] average gva agriculture -.028* [.01] obs. 195 195 lr chi2 79.86 85.09 prob > chi2 0.000 0.000 pseudo r2 0.337 0.363 *** statistically significant at 1% **. statistically significant at 5% *statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 126    table 3. marginal effects on probability y=1|x. group d. model 2a 2b dy/dx dy/dx eqi score -.12.5*** -.137*** [.03] [.03] working age population with tertiary education .003 .001 [.003] [.003] working age population primary education or less .005** .003* [.002] [.002] investment in industry to gdp -.009 [.01] investment in construction to gdp -.027 [.03] investment in non-market services gdp .026** [.01] investment in market services gdp .010 [.007] investment in agriculture to gdp .021 [.03] average gva industry .008 [.005] average gva construction .012 [.01] average gva non-market services .026*** [.007] average gva market services .013 [.008] average gva agriculture .021 [.03] obs. 195 195 lr chi2 124,04 139,35 prob > chi2 0.000 0.000 pseudo r2 0,58 0,65 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 127    table 4. marginal effects on probability y=1|x. group b + group d. model 3a 3b dy/dx dy/dx eqi score .27*** -.24*** [.08] [.08] working age population with tertiary education .002 -.0047 [.004] [.004] working age population primary education or less .013*** .007* [.004] [.004] investment in industry to gdp -.037** [.01] investment in construction to gdp -.07 [.05] investment in non-market services gdp .005 [.01] investment in market services gdp .002 [.01] investment in agriculture to gdp .07 [.067] average gva industry -.014** [.006] average gva construction -.015 [.01] average gva non-market services .001 [.009] average gva market services .005 [.01] average gva agriculture .01 [.01] obs. 195 195 lr chi2 94,99 100,91 prob > chi2 0.000 0.000 pseudo r2 0,34 0,36 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 128    table 5. marginal effects on probability y=1|x. group e. model 4a 4b dy/dx dy/dx eqi score -.19*** -.18*** [.038] [.03] working age population with tertiary education -.006** -.011*** [.003] [.003] working age population primary education or less -.0001 -.001 [.002] [.002] investment in industry to gdp -.013 [.01] investment in construction to gdp .007 [.03] investment in non-market services gdp -.017 [.012] investment in market services gdp .005 [.008] investment in agriculture to gdp .006 [.03] average gva industry -.016*** [.005] average gva construction -.007 [.01] average gva non-market services -.011** [.005] average gva market services -.012** [.006] average gva agriculture -.006 [.008] obs. 195 195 lr chi2 111,09 120,39 prob > chi2 0.000 0.000 pseudo r2 0,54 0,58 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 129    5. robustness check of the model: institutions in the short-run in order to further investigate the role of the quality of institutions in shaping regional economic performances in the eu-15, i run the same model with annual data averaged during the period 2001-2008. with this strategy, it will be possible to insulate the pre-crisis period and compare the long run and short run results of the econometric analyses in order to assess to what extent the quality of government institutions makes european regions more resilient to economic shocks. as already mentioned the data for european quality of government index (eqi score) developed by the quality of government institute are available only for the years 2010, 2013 and 2017. given the fact that changes in the quality of institutions occur very slowly it is plausible to suppose that the quality of government institutions observed in 2010 in the european regions was not sensibly different from the quality of government institutions that had characterised the period 2001-2008. as a consequence, the eqi score of the year 2010 calculated on a survey of 34.000 respondents is chosen as a predictor of regions economic performance. also, the quality of government institute remarks that “a relative stability in quality of government can be noted across the three editions” of the database and such stability further induces to confidently conjecture that quality of government institutions in 2010 was nearly the same of the quality of regional institutions of the previous years. the relationship between the quality of government institutions and economic performances in the short run is expected at best weak or negligible. indeed, countries or regions can experience short run fast economic growth, convergence or growth accelerations regardless the quality of government institutions and as a consequence of episodes of speculation on real estate, stock market and commodity prices that drive economic growth until the bubble bursts and the subsequent macroeconomic adjustment. furthermore, during the periods of financial euphoria or financial/housing bubbles regions or countries with lower income levels can experience short-run periods of fast economic growth and convergence because of high debt-driven investment rates and decreasing returns to physical capital. anyways, stable economic development is ensured in the long run only by sound government institutions as pointed out by acemoglu and robinson (2005). the results of the regressions are reported in tables 6, 7, 8 and 9. as it is possible to note, the quality of institutions is the main factor behind a “taking-off from above” experienced form a region in the short run, while the quality of institutions plays no effects in shaping other regional economic performances. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 130    table 6. marginal effects on probability y=1|x. group a – robustness check. model 6 6b dy/dx dy/dx eqi score .29*** .34*** [.06] [.06] working age population with tertiary education .010*** .011*** [.003] [.004] working age population with primary education or less .0028 .002 [.003] [.003] investment in industry to gdp .028** [.012] investment in construction to gdp .045 [.04] investment in non-market services gdp -.067*** [.01] investment in market services gdp .03 [.02] investment in agriculture to gdp -.008* [.05] average gva industry .00080 [.005] average gva construction .028** [ .014] average gva non-market services -.014* [.008] average gva market services -.012 [.009] average gva agriculture -.048** [.02] obs. 195 195 lr chi2 79,86 85,09 prob > chi2 0.000 0.000 pseudo r2 0,31 0,26 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 131    table 7. marginal effects on probability y=1|x. group b – robustness check. model 7 7b dy/dx dy/dx eqi score -.042 -.039 [.03] [.03] working age population with tertiary education -.005 -.008* [.004] [.004] working age population primary education or less .003 .0034* [.002] [.001] investment in industry to gdp .001 [.01] investment in construction to gdp -.04 [.03] investment in non-market services gdp .011 [.01] investment in market services gdp .008 [.01] investment in agriculture to gdp .003 . [.03] average gva industry -.001 [.006] average gva construction .003 [.018] average gva non-market services .023*** [.008] average gva market services -.0083 [.009] average gva agriculture -.008 [.009] obs. 195 195 lr chi2 33,38 64,4 prob > chi2 0.000 0.000 pseudo r2 0,19 0,37 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10%. standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 132    table 8. marginal effects on probability y=1|x. group b + group d – robustness check. model 8 8b dy/dx dy/dx eqi score .013 .006 [.06] [.06] working age population with tertiary education -.016*** -.018*** [.005] [.004] working age population primary education or less .001 .0062* [.003] [.003] investment in industry to gdp -.01 [.01] investment in construction to gdp -.08 [.05] investment in non-market services gdp -.01 [.01] investment in market services gdp -.003 [.027] investment in agriculture to gdp -.09 [.06] average gva industry -.015** [.006] average gva construction -.072*** [.015] average gva non-market services -.011 [.009] average gva market services -.021** [.009] average gva agriculture -.039** [.018] obs. 195 195 lr chi2 22,5 100,91 prob > chi2 0.000 0.000 pseudo r2 0,18 0,19 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 133    table 9. marginal effects on probability y=1|x. group e – robustness check. model 9 9b eqi score .0008 .008 [.01] [.01] working age population with tertiary education -.001 -.001 [.001] [.001] working age population primary education or less -.0005 -.00005 [.0007] [.0009] investment in industry to gdp -.002 -.002 [.005] [.002] investment in construction to gdp -.003 [.01] investment in non-market services gdp .001 [.004] investment in market services gdp .010 [.006] investment in agriculture to gdp -.016 [.02] average gva industry -.002 [.002] average gva construction .003 [.004] average gva non-market services -.003 [.003] average gva market services .0001 [.001] average gva agriculture -.005 [.008] obs. 195 195 lr chi2 18,27 12,26 prob > chi2 0.000 0.000 pseudo r2 0,54 0,41 *** statistically significant at 1% ** statistically significant at 5% *statistically significant at 10 / standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 134    6. discussion despite the results of the model perfectly fit the expectations derived from the theory and the previous empirics in the literature, it is surprising to empirically demonstrate how the quality of institution turns out to be the most important determinant of regional performances even compared to investment rate in highly productive sectors as the manufacturing industry. in particular, when controlling for sectorial investment rates to gdp, a percentage point increase in eqi score is expected to determine an increase of 27% in the probability that a region will be “taking off from above” (table 2 model 1a) while it is expected to reduce by 15% the probability that a region will be “diverging” from below (table 3, model 1a). the results of the model reported in table 4 (model 1a) also tells that a one-unit increase in the eqi score will reduce by 27% the probability that a region will be “unsuccessful” either “declining from above” or “diverging from below”. symmetrically, a one-unit increase in the eqi score will increase by 27% the probability that a region will be successful either in “taking off from above” or “converging” from below. beside the quality institutions, the role investment in manufacturing industry to gdp seems to be very relevant for increasing the probability that a region will be “taking off from above”, as reported in table 2, and reducing the probability that a region will be “diverging from below” or being “unsuccessful” (table 3 and table 4). as regards the “failed” regions (table 5), the industrial sector size plays a marginal role while the eqi score is the main determinant of regional performances because a one-unit increase in the eqi score is expected to reduce by 19% the probability that a region will "fail” to grow. human capital proxied as the educational attainment of the working-age population generally plays an effect in shaping regional performances, in particular the labour force with primary education or less is expected to increase the probability that a region will be “diverging from below” or “unsuccessful” while the labour force with tertiary education is expected to reduce the probability that a region will fail to grow. 7. conclusions as in the most of the analyses on economic growth, i provided a model that includes indicators for human capital (proxied as the educational attainment of the working-age population) and physical capital accumulation (investment rate in the different sectors of the economy). however, empirical evidence from western european regions shows that the quality of government plays a preeminent role in shaping regional economic performances. the coefficient estimates for the explanatory variable eqi are always statistically significant within a 99% confidence interval and their magnitude is always remarkably much larger than the magnitude of coefficient estimates for variables like investment rate in manufacturing industry (or the size of the industrial sector) or the educational attainment of the working-age population. according to the estimations, eqi score is the most relevant variables in explaining positive (“taking off from above” or “successful”) and negative (“diverging from below” or “unsuccessful”) f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 135    economic performances. the results of the model clearly confirm the thesis of acemoglu and robinson (2012) about to the tight relation between quality of institution and failures of nations, or regions in our case. as shown in the statistical map (figure 4), most of the regions with very low indices of the quality of government institutions are the regions of southern europe, in particular, the italian and the greek ones. those regions were also the ones that suffered more from the financial crisis and sovereign debt crisis (they have been regions “diverging from below”) while most of the northern italian regions with income higher than eu income levels in 2001 have been “declining” from above or unsuccessful. according to the estimations, a one-unit increase in eqi score is expected to increase by 27% the probability that a region will be “taking off from above” (table 2, model 1a), while it will reduce by 12.5% the probability that a region will be “diverging from below” (table 3, model 2a). finally, a one-unit increase in the eqi score is expected to reduce by 27% the probability that a region will be “unsuccessful” either by “diverging from below” or “declining from above” (table 4, model 3a)10. the empirical finding of this paper are very relevant for governments in diverging or declining regions of the european union, mainly in the peripheral countries of the emu. indeed, the results show how the quality of institutions ends up to be the most relevant determinant of regional economic performances. in particular, the regression output for “diverging from below” regions (table 3) clearly shows that for regions, whose per capita income is lower than the per capita income of the european union as a whole, the quality of institutions is much more important than other variables such as the investment rate in manufacturing sector or it size. indeed, according to the coefficient estimates reported in table 3 (model 2a), a one-unit increase in the eqi score is expected to reduce by 12.5% the probability that a region will be “diverging from below” while the coefficient estimates investment rate in industry or the size of the industrial sector are not statistically significant. quality of government institutions exerts a huge impact on the probability that a region will be “unsuccessful” by either “diverging from below” or declining from above”. according the regression output reported in table 4 (model 3a), while a 1% increase in investment rate in industry is expected to reduce by 3.7% the probability that a region will be “unsuccessful”, a one-unit increase in the eqi score is expected to reduce by 27% (more than one quarter) the probability that a region will be “unsuccessful” either by “declining from above” (as in the case of all the northern and central italian regions) or “diverging from below” as in the case of the greek regions. last but not least, according to the estimations reported in table 5, one unit increase in the eqi index is expected to reduce by 19% (model 4a) or 18% (model 4b) the probability that a region will fail to grow. during the period under analysis 20 italian regions out of 21 and 10 greek regions out of 13 recorded negative average gdp per capita growth, meaning that over the same period, 31 regions out of 44 european regions that recorded negative economic 10 symmetrically a one-unit increase in eqi score is expected to increase by 27% the probability that a region will be “successful” with a 99% confidence interval f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 136    growth were greek or italian. the two countries are also the countries with the lowest quality of regional institutions in terms of eqi score. given these further considerations and the empirical evidence provided by the econometric estimations, it is straightforward to conclude that the quality of institutions is the main determinant of regional economic performances in the long run. as a consequence, it is possible to assert that the poor quality of institutions has hampered both the resilience of northern italian regions to the financial crisis and their ability to “take-off from above” as in the case of the most industrialized regions located in germany. at the same time, the poor quality of government institutions has remarkably contributed to the process of “divergence from below” observed in the greek and southern italian regions. as empirically estimated and reported in table 3, the quality of government institutions is much more relevant than the investment rate in the manufacturing industry for lower-income regions to avoid further divergence from the eu income levels. furthermore, by comparing the results estimated for the same model in the long run and in the short run, it is also possible to assert that the soundness of government institutions makes regions more resilient to financial crises or macroeconomic shocks. as a matter of facts, greek regions, that recorded high rates of economic growth before the financial crisis, have recorded the worst economic performances in europe over the period 2001-2015 together with the italian regions. according to rothstein and uslaner (2005), the quality of government institutions mainly reflects the social trust that characterizes a local community, and many empirical studies show that higher degrees of social trust are generally associated with higher levels of quality of government institutions. lack of social trust within a community means that citizens have faith only in their family, clan or social group and this makes them less eager to contribute to the provision of general public goods, such as paying taxes, respecting and protecting public spaces and, very importantly, engaging in social and political mobilisations asking for improvements in quality of government. generally speaking, free-riding becomes more frequent at all social levels. in turn, public authorities lack both adequate resources and incentives to deliver policies, consolidating a “vicious cycle” (charron et all, 2012:10). in this framework, central governments in peripheral countries (especially in italy and greece) should focus their efforts on promoting a set of institutional reforms aimed at breaking up the nexus between the lack of social trust in local communities and political clientelism, thus reforms aimed at breaking the linkages between decentralised governance and rent extraction by private parties, social groups or local bureaucrats. these sets of reforms would be very effective especially in countries like italy where central government assigns large autonomy to regional governments in the management and provision of costly public services in the framework of a process of progressive devolution of the power from the state to regions. however, given the fact that there is no a “one-size fits all” approach to curbing corruption, any measures must take into consideration the political, economic and social environment of a country and address the root causes of corruption rather than adopting a symptomatic approach f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 137    (lee-jones, 2018). such tailor-made institutional reforms should be accompanied by both the setup of independent national authorities responsible for monitoring and combatting corruption and the introduction of a more stringent code of laws to prosecute episodes of corruption of public officials or policy-makers. references acemoglu, d. & robinson, j. a. (2012). why nations fail. the origins of power, prosperity, and poverty. new york: crown business. doi: https://doi.org/10.1355/ae29-2j acemoglu, d. & robinson, j. (2008). the role of institutions in growth and development. the international bank for reconstruction and development / the world bank on behalf of the commission on growth and development. working paper no.10 acemoglu, d., johnson, s. & robinson, j. a. (2005). institutions as fundamental cause of economic growth. in aghion, p. and durlauf, s. (eds.) handbook of economic growth 1a. amsterdam: elsevier b.v. doi: https://doi.org/10.1016/s1574-0684(05)01006-3 acemoglu d., johnson, s. & robinson, j. a. (2005a). the rise of europe: atlantic trade, institutional change, and economic growth. nber working paper no. 9378. doi: https://doi.org/10.1257/0002828054201305 aiginger, k.; firgo, m. & huber, p. (2013). what can the emu’s peripheral countries learn from regional growth? in lacina, l., rosmahel, p. & rusek, a. (eds.) political economy of the eurozone crisis. reforms and their limits. mendel european centre alonso, j. a. (2009). colonisation, formal and informal institutions, and development. icei working paper no.13 ascani, a; crescenzi, r. & iammarino, s. (2012). regional economic development: a review. search working paper 1/03 barro, r. j. (1996). determinants of economic growth. a cross-country empirical study. national bureau of economic research, working paper n. 5698. doi: https://doi.org/10.3386/w5698 barro, r. j. (1991). economic growth in a cross-section of countries. quarterly journal of economics, 106 (2), 407-443. doi: https://doi.org/10.2307/2937943 canton, e. & solera, i. (2016). greenfield foreign direct investment and structural reforms in europe: what factors determine investments? european commission, european economy discussion paper no. 033 charron, n., dijkstra, l. & lapuente, v. (2015). mapping the regional divide in europe: a measure for assessing quality of government in 206 european regions. social indicators research, 122 (2), 315-346. doi: https://doi.org/10.1007/s11205-014-0702-y charron, n., dijkstra, l. & lapuente, v. (2012). regional governance matters. a study on regional variation in quality of government in the eu. european commission working paper no. 01/2012 feldkircher, m. (2006). new regional economics in central european economies: the future of centrope. oesterreichische nationalbank working paper no.9 greif, a. (1994). cultural beliefs and the organization of society. theoretical reflection on collectivist and individualist societies. the journal of political economy, 102 (5), 912-950. doi: https://doi.org/10.1086/261959 harris, r. (2008). models of regional growth: past, present and future. serc discussion papers, sercdp0002. spatial economics research centre (serc), london school of economics and political science. huggins, r, & thompson, p. (2017). introducing regional competitiveness and development: theories and perspectives. in huggins, r. & thompson, p. (eds.) handbook of regions and competitiveness: contemporary theories and perspectives on economic development. chaltam: edward elgar publishing limited. doi: https://doi.org/10.4337/9781783475018.00005 lee-jones, k. (2018). best practices in addressing police-related corruption. transparency f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 138    international lucas, j. r, (1988). on the mechanics of economic development. journal of monetary economics, 22, 3-42. doi: https://doi.org/10.1016/0304-3932(88)90168-7 macmullen, r. (1988). corruption and the decline of rome. new heaven: yale university press mankiw, g.; romer, d. & weil, d. (1992). a contribution to the empirics of economic growth. the quarterly journal of economics, 107 (5), 407-437. doi: https://doi.org/10.2307/2118477 martin, r. & sunley, p. (1998). slow convergence? the new endogenous growth theory and regional development. economic geography, 74(3), 201-227. doi: https://doi.org/10.2307/144374 mauro, p. (2002). the persistence of corruption and slow economic growth. the international monetary fund. imf working paper no.02/13. doi: https://doi.org/10.5089/9781451874952.001 mauro, p. (1995). corruption and growth. the quarterly journal of economics, 110 (3), 681712. doi: https://doi.org/10.2307/2946696 north, d.c. (1991). institutions. journal of economic perspectives, 5(1), 971-112. doi: https://doi.org/10.1257/jep.5.1.97 pons-novell, j. & viladecans-marsal, e. (1998). kaldor’s law and spatial dependence: evidence for the european regions. regional studies, 33(5), 443-451. doi: https://doi.org/10.1080/00343409950081284 rodríguez-pose, a., & di cataldo, m. (2014). quality of government and innovative performance in the regions of europe. journal of economic geography, 15(4), 673-706. doi: https://doi.org/10.1093/jeg/lbu023 rodrik, d.; subramianan, a. & trebbi f. (2002). institutions rule: the primacy of institutions over geography and integration in economic development. nber working paper no. 9305. doi: https://doi.org/10.3386/w9305 romer, p. m. (1986). increasing returns and long run growth. journal of political economy, 94, 1002–37. doi: https://doi.org/10.1086/261420 rothstein, b. & uslaner, e. m. (2005). all for all: equality and social trust. lse health and social care discussion paper no. 15. doi: https://doi.org/10.2139/ssrn.824506 f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 139    appendix 1. composition and definition of the eqi score. eqi score is developed by the quality of government institute of the university of gothenburg within the framework of a european commission-funded project on measuring the quality of government institutions in the european regions. the index is built on the largest survey ever undertaken to measure the quality of government (eqog). at the sub-national level. in order to capture the most relevant sub-national variation in eqog, surveyors focus on three public services that are often financed, administrated and politically accounted for by subnational authorities: education, healthcare and law enforcement. surveyors asked respondents “to rate these three public services with respect to three related concepts of qog – the quality, the impartiality and the level of corruption of said services”. the regional data combine 16 survey questions about the quality of government in a region. to construct the regional index, surveyors followed carefully the guidelines provided in the “handbook on constructing composite indicators; methodology and user guide” published in 2008 by the oecd. all the qog questions are aggregated from the individual to regional level. next, the 16 regional scores are standardised so as to obtain a common range via standardisation. then the standardised scores are assigned to three different groups named “pillars” containing scores related to question about impartiality, corruption and quality. each variable is given the same weight in each pillar. finally, the three pillars are combined using equal weighting to form the regional index (charron et al., 2012). appendix 2 multicollinearity occurs when a high correlation is detected among the predictors in a regression model. more precisely, multicollinearity arises when one of the regressors is a perfect linear combination of the other regressors. in the case of the model specified in this article, investment rates in the different sectors of the economy (or the share of a sector of the total gva). may be correlated, especially averaged over the long period. for example, an increase in investment in the constructions, a very important steel-using sector, may be positively correlated to an increase in the investment rate in the manufacturing activity involved in the production of steel. a first possible strategy to check for multicollinearity is to examine the correlation matrix of the predictors, where correlation coefficients would indicate the presence of multicollinearity. therefore, the pairwise correlation coefficients between two explanatory variables would be close to 1 if a regressor is a linear function of another regressor. tables 10 and 11 report, respectively, the pairwise correlations between the classes of investment rates and between the sectorial components of the total gross value added. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 140    table 10. pairwise correlations between classes of investment rates. investment in industry to gdp investment in construction to gdp investment in non-market services gdp investment in market services gdp investment in agriculture to gdp investment in industry to gdp 1 investment in construction to gdp 0.0491 1 investment in non-market services gdp -0.0230 0.5567 1 investment in market services gdp -0.1743 -0.2238 -0.1493 1 investment in agriculture to gdp 0.0880 -0.0676 0.1690 0.2020 1 table 11. pairwise correlations between the sectorial components of the total gross value added. gva average gva industry average gva construction average gva non-market services average gva market services average gva agriculture average gva industry 1 average gva construction -0.1947 1 average gva non-market services -0.3552 0.1765 1 average gva market services -0.6467 0.0798 -0.2434 1 average gva agriculture -0.0350 0.3607 0.1267 -0.0578 1 figures 8 and 9 shows the plot of the correlation matrix of investment rates and gva shares, respectively. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 141    according to the correlation coefficients reported in tables 10 and 11 (and their graphical representations), the predictors included in the models do not seem to be collinear or linearly dependent. indeed, in table 10, the larger correlation coefficient in absolute terms is the one detected between the investment rate in construction and investment rate in non-market f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 142    services (-.55). in table 11, the larger correlation coefficient in absolute terms is the one detected between industry gva and gva in market services (-.64). all the correlation coefficients in absolute values lay below the threshold of 0.50. table 12. ols version of the original probit models. dependent variable: average gdp per capita growth 2001-2015. model 9a model 9b log of gdp pc in 2001 -.005*** -.009*** [.001] [.002] eqi score .004*** .004*** [.0008] [.0007] working age population with tertiary education .0001** .0002*** [.00005] [.00005] working age population with primary education or less -.0002*** -.0001*** [.00004] [.00004] investment in industry to gdp .0008*** [.0001] investment in construction to gdp .0010375 [0006] investment in non-market services gdp .00005 [.0002] investment in market services gdp -.00007 [.0001] investment in agriculture to gdp -.001* [.0007] average gva industry .00003 [.00009] average gva construction -.00006 [.0002] average gva non-market services -.0003*** [.0001] average gva market services -.0002** [.0001] average gva agriculture -.0004*** [.0001] obs 202 202 prob > chi2 0.000 0.000 r2 0,6 0,6 ***statistically significant at 1% **statistically significant at 5% * statistically significant at 10% standard error in brackets [ ] f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 143    although the correlation matrix analysis indicates that there exists no multicollinearity problem in the model, i also perform the inflation variance factor (ivf) test. in order to perform this test, it is necessary to perform an ols regression because the ivf is an index that measures to what extent the variance of the estimated regression coefficient is increased as a consequence of multicollinearity. let rj2 indicates the coefficient of determination of a regression equation in which xj is regressed on all the other predictors of the model. let vifj be determined as vifj =1/ (1-rj2). for j=1,2,…p-1. therefore, when rj2 is equal to 0 the vifj would be equal to 1 (=1/ (1-02), meaning that the jth is not linearly related with the other predictors. symmetrically when rj2 is equal to 1 (the highest possible coefficient of determination), the vifj would be equal to ∞ (=1/ (1-12), meaning the jth is linearly related with the other predictors. the rule to establish whether there exists multicollinearity within a model is provided by montgomery (2001). that individuates vif threshold values. if the vif values exceed 5 or 10 it means that the model is poorly estimated because of the presence of multicollinearity. table 12 shows the regression output of the ols version of the probit model whose dependent binary variable replaced with the average gdp per capita growth in the period 2001-2015. also, the natural logarithm of gdp per capita in 2001 is included in the linear regression in order to control for initial conditions. once the ols model has been estimated, it is possible to estimate the vif for each variable. the estimation of the two models is reported in table 13. table 13. vif. ivf model 9a ivf model 9b eqi score 4.05 gva manufacturing industry 4.44 working age population with primary education or less 3.48 log of gdp pc in 2001 4.10 investment in construction to gdp 2.44 market services gva 3.51 log of gdp pc in 2001 2.35 non-market services gva 3.42 working age population with tertiary education 2.35 working age population primary education or less 3.22 investment in non-market services gdp 2.09 eqi score 2.77 investment in agriculture to gdp 1.48 construction gva 2.11 investment in maunufacturing industry 1.21 working age population with tertiary education 2.07 investment in market services gdp 1.18 agriculture gva 1.77 the estimation of the vifj’s confirms the absence of multicollinearity in the model. furthermore, when running regression analysis with the econometric software, variables that are perfectly correlated are always automatically drop from the model in order to estimate the coefficients. f. bonanno / european journal of government and economics 8(2), december 2019, 114-144 144    appendix 3 an optimal quantity tax path in a dynamic setting european journal of government and economics 6(2), december 2017, 191-225. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 an optimal quantity tax path in a dynamic setting nasreen nawaz a,* a department of economics, michigan state university, usa * corresponding author at: department of economics, michigan state university, 486 w circle dr, east lansing, mi 48824, u.s.a. email: nawaznas@msu.edu article history. received 6 may 2017; first revision required 30 november 2017; accepted 17 december 2017. abstract. following ramsey, the existing literature on optimal quantity taxation only compares the pre and the post-tax market equilibriums in order to account for the efficiency losses. however, when the government imposes a quantity tax on the consumer, the buyer’s price jumps to the pre-tax equilibrium price plus the amount of the tax, and the supply and the demand of the taxed commodity then adjust over time to bring the new post-tax market equilibrium. the existing literature does not take into account the efficiency losses during the adjustment process while computing the optimal quantity taxes. this paper derives an optimal quantity tax path in a dynamic setting minimizing the efficiency losses (output and/ or consumption lost) during the dynamic adjustment process as well as the post-tax market equilibrium. keywords. quantity tax; dynamic efficiency; adjustment path; equilibrium jel classification. h20; h21; h22 1. introduction optimal taxation in theory is the design of a tax that minimizes inefficiency and distortion due to deviation from the pre-tax efficient market equilibrium under given economic constraints. ramsey (1927) was the first to make a significant contribution to the theory of optimal taxation from an economic standpoint. he developed a theory for optimal commodity taxes and proposed a theoretical solution that consumption tax on each good should be "proportional to the sum of the reciprocals of its supply and demand elasticities". diamond and mirrlees (1971) consider commodity taxation along with the other kinds of taxes. mirrless (1975) modified the standard problem by considering simultaneously excise taxes and a poll tax. diamond (1975) examines the ramsey rule for a many-person economy with excise taxes and a poll tax. atkinson and stiglitz (1976) show that with an optimal nonlinear income tax, discriminatory commodity taxes are only necessary to the extent that individual commodities are not weakly separable from leisure. in deaton (1981), rules for optimal differential commodity taxes have been derived for the three different cases usually studied in the literature: the one consumer economy, the unidimensional continuum of consumers economy, and the finite number of discrete consumers economy. lucas and stokey (1983) derive a time consistent optimal fiscal policy in an economy without capital http://www.ejge.org/ mailto:nawaznas@msu.edu 192 n. nawaz / european journal of government and economics 6(2), 191-225 maximizing the consumer welfare subject to the condition that a competitive equilibrium holds in each time period. in judd (1985), the government taxes capital income net of depreciation at a proportional rate, which is assumed to be constant. chamley (1986) analyzes the optimal tax on capital income in general equilibrium models of the second best. deaton and stern (1986) show that optimal commodity taxes for an economy with many households should be at a uniform proportional rate under certain conditions. cremer and gahvari (1993) incorporate tax evasion into ramsey’s optimal taxation problem. cremer and gahvari (1995) prove that optimal taxation requires a mix of differential commodity taxes and a uniform lump-sum tax. naito (1999) shows that imposing a non-uniform commodity tax can pareto-improve welfare even under nonlinear income taxation if the production side of an economy is taken into the consideration. saez (2002) shows that a small tax on a given commodity is desirable if high-income earners have a relatively higher taste for this commodity or if consumption of this commodity increases with leisure. the quantity taxes are currently more popular in the environmental economics literature, e.g. nordhaus (1993) proposes an optimal carbon tax (tax per ton of carbon). chari, christiano and kehoe (1994) deal with the labor and capital income taxes instead of a quantity tax as in our model. ekins (1996) takes into account the secondary benefits of carbon dioxide abatement for an optimal carbon tax. coleman (2000) derives the optimal dynamic taxation of consumption, income from labor, and income from capital, and estimates the welfare gain that the us could attain by switching from its current income tax policy to an optimal dynamic tax policy. pizer (2002) explores the possibility of a hybrid permit system and a dynamic optimal policy path in order to accommodate growth and not because of the adjustment over time to equalize the marginal benefit and cost. it is implicitly assumed that the marginal cost equals the marginal benefit in each time-period. following ramsey, the existing literature on optimal quantity taxation only compares the pre and the post-tax market equilibriums in order to account for the efficiency losses. however, when the government imposes a quantity tax on the consumer, the buyer’s price jumps to the pre-tax equilibrium price plus the amount of the tax, and the supply and the demand of the taxed commodity then adjust over time to bring the new post-tax market equilibrium. the existing literature does not take into account the efficiency losses during the adjustment process while computing the optimal quantity taxes. this paper derives an optimal quantity tax path in a dynamic setting minimizing the efficiency losses (output and/ or consumption lost) during the dynamic adjustment process as well as the post-tax market equilibrium. the remainder of this paper is organized as follows: section 2 explains how the individual components of the market system are joined together to form a dynamic market model. section 3 provides the solution of the model with a quantity tax imposed. section 4 derives an optimal commodity tax path minimizing the efficiency losses subject to a tax revenue target in a specific time-period. section 5 summarizes the findings and concludes. the appendix presents mathematical details. 193 n. nawaz / european journal of government and economics 6(2), 191-225 2. the model let’s assume that there is a perfectly competitive market of a single homogeneous commodity in equilibrium (so our starting point is when the market is already in equilibrium). there are four types of infinitely-lived agents: a representative -or a unit mass ofproducer (that produces a good, and demand labor and capital), a middleman (who buys the good from firms to sell to consumers, and possibly accumulating inventories), a representative –or a unit mass of– consumer (who buys the good, accumulates capital by investing and supplies labor inelastically), and a government. the role of middleman is motivated by the real world scenario where the producer and the consumer seldom directly meet for a transaction to take place. the existence of retailers, wholesalers, financial institutions, educational institutions and the hospitals reflect the presence of middlemen between producers and consumers in most of the economic activity going on. the producer produces the goods and supplies those to the middleman, who keeps an inventory of the goods and sells those to the consumer at the market price. in the model, the middleman plays a key role, as she sets the selling price 𝑝𝑝 by maximizing the difference between the revenue for selling goods to consumers and the costs of inventories. the buying price paid to the producer is 𝛼𝛼 𝑝𝑝 with 𝛼𝛼 < 1, and the producer is a price taker. the price adjustment mechanism is based on the fact that when a shock leads the market out of equilibrium, the buyers’ and sellers’ decisions are not coordinated at the current prices. an example can illustrate the working of this market. consider that the market is initially in equilibrium. the middleman has an equilibrium stock of inventory. then, an exogenous demand contraction will increase the stock of inventory, due to firms’ output could not match with the –now lower– units demanded by the consumer at the current price. this excess of supply is accumulated in inventory held by the middleman. the middleman will decrease the price so that the producer will find optimal to produce a lower level of output. a new equilibrium with a lower price and a lower level of output is then reached. the equilibrium is defined as follows: (i) the producer and the middleman maximize their profits and the consumer maximizes her utility subject to the constraints they face (mentioned in their individual dynamic optimization problems in section 2). (ii) the quantity supplied by the producer equals the quantity consumed by the consumer (and hence the inventory does not change when the market is in equilibrium). the conditions for the existence of the equilibrium (routh–hurwitz stability criterion, which provides a necessary and sufficient condition for the stability of a linear dynamical system) have been mentioned in section 3. as the set-up is for a perfectly competitive market, therefore, the middleman who sells the goods to the consumer at the market price is a price taker when the market is in equilibrium. when the market is out of equilibrium, the middleman can change the price along the dynamic adjustment path until the new equilibrium arrives, where again the middleman becomes a price taker. the government announces and imposes a commodity tax at the same time (the expectations of the agents will be taken into account in a future research project when the dates 194 n. nawaz / european journal of government and economics 6(2), 191-225 of announcement and implementation of the tax are different). when a commodity/ quantity tax is imposed, the market does not suddenly jump to the post-tax market equilibrium, rather the price adjusts over time to bring the new equilibrium. this adjustment process involves endogenous decision making (in their own interest) by all the agents in the market, i.e. consumer, producer and the middleman as follows: suppose there is a producer in a market who produces a perishable good and sells it to a middleman who further sells it to a consumer living in a community. the producer and the middleman sell a quantity exactly equal to the quantity the producer produces in each time period, and the market stays in equilibrium. if the government announces and imposes a commodity tax on the buyer, which decreases the demand of this product, some of the production sold to the middleman will remain unsold to the consumer and be wasted by the end of the time period in which the tax was imposed. assuming that the producer and the middleman can change the production and the price respectively, immediately, had they known the exact pattern of new demand, they would immediately pick the quantity (by the producer) and the price (by the middleman) to maximize their profits and clear the market without wasting the production. however, they lack this information, so the middleman decreases the price based on her best guess about the new demand (based on the quantity of the unsold production), driving the market close to the new equilibrium. at the lower price, the producer produces a lower quantity than before. if in the following time period, his production sold to the middleman is fully sold out to the consumer, he will know that the new equilibrium has arrived, however, if a part of his production still remains unsold, the middleman will reduce the price further (and the producer, the production accordingly) to bring the market closer to the new equilibrium. the market will eventually settle at the new equilibrium after some efficiency loss. the resources wasted by the imposition of the tax are those that went into the unsold production in each timeperiod during the adjustment process. a new equilibrium with a deadweight loss due to commodity taxation is finally arrived at. the total efficiency loss because of commodity taxation is the loss during the adjustment process plus the loss in the final equilibrium. for the mathematical treatment, the objective of each of the three market agents is maximized through the first order conditions of their objective functions and to capture the collective result of their individual actions, the equations representing their individual actions are solved simultaneously. for simplification, we assume that after the imposition of the quantity tax, the new equilibrium is not too far from the initial equilibrium. this assumption makes the linearization of supply and demand curves quite reasonable; see figure 1 (the time axis is not shown). linearization seems to be a good approximation when we move from point a to b, whereas it is not a good approximation when we move from point a to c. for modeling the movement of the market from point a to c, we need to model a non-linear dynamical system (which is not covered under the scope of this paper). 195 n. nawaz / european journal of government and economics 6(2), 191-225 figure 1: when is linearity a reasonable assumption? 2.1 middleman the middleman purchases goods from the producer and sells those to the consumer for profit. as happens in the real world, the middleman does not buy and sell exactly the same quantity at all points in time, thus he holds an inventory of the goods purchased to be sold subsequently. inventory is an intermediary stage between supply and demand which reflects the quantum of difference between supply and demand of the goods in the market. if the inventory remains the same, it implies that demand and supply rates are the same. an increase or decrease in inventory implies a change in supply, demand or both at different rates. figure 2 helps to understand the link between inventory, supply, demand and prices. when the supply curve shifts to the right (while demand remains the same), the inventory in the market increases at the initial price, and the new equilibrium brings the price down. similarly, when the demand curve shifts to the right (while supply remains constant), the inventory depletes from the market at the previous price and the new equilibrium brings the price up. this shows that there is an inverse relationship between an inventory change and a price change (all else the same). if both the supply and demand curves shift by the same magnitude such that the inventory does not change, then price will also remain the same. inventory unifies the supply and demand shocks in the sense that they are both affecting the same factor, i.e. inventory and are basically the faces of the same coin. therefore each kind of shock is in fact just an inventory shock. from the above mentioned discussion, we have seen that there is an inverse relationship between an inventory change and a price change. 196 n. nawaz / european journal of government and economics 6(2), 191-225 figure 2: movement of price with inventory. now let’s discuss the mechanism which brings about such a change. consider a market of homogeneous goods where the middlemen, such as whole sellers, retailers, etc. hold inventories, incur some cost for holding those, and sell products to the consumers to make profits. the cost is a positive function of the size of an inventory, i.e. a larger inventory costs more to hold as compared to a smaller inventory. in the absence of an exogenous shock, if the supply and demand rates are equal then the system is in equilibrium and the price does not vary with time. suppose that a technological advancement decreases the marginal cost of production and increases the supply rate, whereas the demand rate remains the same. as the demand and supply rates are no longer equal, therefore the difference will appear somewhere in the economy in the form of piled up inventories. as the production flows from the producer to the consumer through the middleman, therefore it is reasonable to assume that the middleman will be holding the net difference (explanation: the piled up inventories can also be in the form of producers’ inventories of finished goods, which does not change the key point that a difference of supply and demand rates directly affect the inventories in the economy). the economy will not be able to sustain this situation indefinitely, and the middlemen will have to think of some means of getting rid of piled up inventories. the only resort they have is to decrease the price which brings the demand up along the demand curve. 197 n. nawaz / european journal of government and economics 6(2), 191-225 in a perfectly competitive market, the price will eventually come down to equalize the new marginal cost, however the adjustment path depends on how the middlemen react to the change in their inventories. notice that the marginal cost of production has decreased but the marginal cost of holding an extra unit of inventory for the middleman has increased. this is an intuitive explanation which is theoretically consistent with the demand, supply, utility and profit maximization by a consumer and a producer respectively. in the real world, we see examples of this behavior of middlemen, e.g. as consumers, we enjoy the end of year sales, offers such as buy one get one free, gift offers if you buy above a certain quantity threshold, etc. for a mathematical treatment, we need to consider the profit maximization problem of the middleman as follows. 2.1.1 short-run problem let’s first consider the short-run problem of the middleman as follows (the middleman’s objective is myopic rather than doing dynamic optimization. in a discrete analog, this is a one period analysis, which is presented for an intuitive purpose as an anticipation of the -more complicated dynamic problem in section 2.1.2). )),,(()(= epmppq ς−π [1] where =π profit, p = market price, )( pq = quantity sold at price ,p =m inventory (total number of goods held by the middleman), e = other factors which influence inventory other than the market price including the middleman’s purchase price from the producer, )),(( epmς = cost as a function of inventory (increasing in inventory). the first order condition (with respect to price) is as follows: 0,=),()),(()()( 1 ... epmepmpqpqp ς−+ [2] the middleman has an incentive to change the price only during the adjustment process and will incur losses by deviating from the price (equal to the marginal cost) when the market is in equilibrium. during the adjustment process, the demand does not equal the supply and the market drifts toward the new equilibrium (however, the price cannot move automatically and it is reasonable to assume that some economic agent moves the price in her own benefit), therefore a price change by the middleman in the direction of bringing the new equilibrium is not against the market forces, so he does not lose business by changing price on the adjustment path unlike when the market is in equilibrium and where the middleman faces an infinitely elastic demand as 198 n. nawaz / european journal of government and economics 6(2), 191-225 follows: ),,()),((=)()( 1 ... epmepmpqpqp ς+ , )( ),()),((=11 . 1 . . pq epmepm ticitydemandelas p ς      + . )( )(= . pq pqpticitydemandelas the right hand side of the above expression is the marginal cost which equals the price when the middleman faces an infinitely elastic demand. suppose that as a result of a supply shock, the marginal cost of production decreases, and the supply curve shifts downwards. now the competitive market is out of equilibrium as the demand does not equal the supply at the previous equilibrium price. the price must eventually decrease to bring the new equilibrium, however, the price will not jump to equalize the demand and supply, and rather the middleman will continue charging a price higher than the new marginal cost until the market forces make him realize that the supply has increased and he needs to lower the price to satisfy the profit maximizing condition. the similar is the case of a reverse supply shock, where the price must eventually increase to bring the new equilibrium. in this case, the middleman will continue charging a price lower than the marginal cost until the market forces make him increase the price, in which case it is the consumer who is the short term beneficiary. again, the consumer will be paying a price less than the marginal cost only during the adjustment process and only until the middleman increases the price. the equilibrium price is equal to the marginal cost of production plus the marginal cost of storage (i.e. the total marginal cost) in the absence of any kind of a tax, so neither does the middleman earn any economic rent, nor does the consumer benefit by paying a price less than the marginal cost when the competitive market is in equilibrium. for the mathematical treatment, suppose that as a result of a supply shock (while demand remains the same) such as a technological advancement which reduces the marginal cost of production and increases the supply by the producers, if the middleman wants to hold an extra unit of inventory, his marginal cost of holding an extra unit, i.e. )( ),()),(( . 1 . . pq epmepmς , is higher at the previous price, because the term )),(( . epmς is higher at the previous price. this might be on account of higher storage charges because of increased demand of warehouses, godowns, etc. after increased supply in the market. the second term, i.e. )( ),( . 1 . pq epm is a function of price, and is the same as before as the price has not changed yet (we are assuming that the middleman’s purchase price is the same as before as the producer is a price taker during the 199 n. nawaz / european journal of government and economics 6(2), 191-225 adjustment process as well and always charges a fixed fraction of the market price to the middleman). a discrete analog of this scenario is that the middleman maximizes profits in each time period without considering the future time periods, and in each time period he takes the purchase price from the producer as given and only chooses the sale price. this implies that on the previous price, now the middleman faces 0,<),()),(()()(= 1 ... epmepmpqpqp p ς−+ ∂ π∂ [3] which means that the middleman must decrease the price to hold an extra unit of inventory to satisfy the profit maximizing condition after the supply shock. please notice that in this static scenario, the short term gains accrued from the decreased marginal cost of production will be reaped by the producer, as his marginal cost has decreased but he charges the same price to the middleman until the middleman changes the price. if we plot together various profit maximizing combinations of inventories and the respective prices chosen by a middleman, we will get a downward sloping inventory curve with the price on the y -axis and the inventory on the x -axis. this is analogous to the concept of supply and demand curves for the profit maximizing producers and the utility maximizing consumers respectively. 2.1.2 dynamic problem now let’s consider the dynamic problem of the middleman. in a dynamic setting, the middleman maximizes the present discounted value of the future stream of profits, and his present value at time zero is as follows: [ ] ,)),(()(=(0) 0 dteepmppqv rt− ∞ −∫ ς [4] where r denotes the discount rate, )(tp is the control variable and )(tm the state variable. the maximization problem can be written as { } [ ] ,)),(()(=(0) 0 )( dteepmppqvmax rt tp − ∞ −∫ ς subject to the constraints )()),(())),((),(()())),((),((=)( . 1 .. 2 .. 1 . tpztpeztpetpmtpztpetpmtm + (state equation, describing how the state variable changes with time; z are exogenous factors), smm =(0) (initial condition), 0)( ≥tm (non-negativity constraint on state variable), )(∞m free (terminal condition). 200 n. nawaz / european journal of government and economics 6(2), 191-225 the current-value hamiltonian for this case is . )),(( ))),((),(())),((),(()()()))),((),((())(()(=~ 1 . . 2 . 1 .         ∗++− ztpe ztpetpmztpetpmtptztpetpmtpqtph µς [5] the maximizing conditions are as follows: )(i )(tp∗ maximizes h~ for all t : 0,= ~ p h ∂ ∂ )(ii , ~ = . m hr ∂ ∂ −− µµ )(iii µ∂ ∂∗ hm ~ = . (this just gives back the state equation), )(iv 0=)()(lim rt t etmt − ∞→ µ (the transversality condition). the first two conditions are as follows: 0,= ~ p h ∂ ∂ [6] and ))).),((),(((= ~ = .. ztpetpm m hr ςµµ ∂ ∂ −− [7] when the market is in equilibrium, 0,=)( . tp and the expression p h ∂ ∂ ~ boils down to the following (see appendix): , ))(( )),(())),((),(( ))(( ))),((),(()))),((),(((=11)( . 1 .. 2 . 1 . .         +      + tpq ztpeztpetpm tpq ztpetpmztpetpm ticitydemandelas tp ς whiche suggests that the price equals the marginal cost (the right hand side of the above expression is the marginal cost in a dynamic setting, which is different from that in a static problem on account of the fact that in a dynamic setting the middleman also takes into account the impact of price chosen on his purchase price from the producer) when the demand is infinitely elastic. now suppose that as a result of a supply shock, if the middleman wants to hold an extra unit of inventory, then the marginal cost of holding an extra unit is higher because the term )))),((),((( . ztpetpmς is higher at the previous price at that point in time. the term in 201 n. nawaz / european journal of government and economics 6(2), 191-225 parentheses in the expression of the marginal cost, i.e. ))(( )),(())),((),(( ))(( ))),((),(( . 1 .. 2 . 1 . tpq ztpeztpetpm tpq ztpetpm + is a function of price and is the same at the previous price. this implies that on the previous price, now the middleman faces 0.< ~ p h ∂ ∂ therefore, in order to satisfy the condition of dynamic optimization, the middleman must decrease the price for an increase in inventory. this implies a negative relationship between price and inventory. the concept of inventory unifies the market supply and demand. if the supply and demand rates are equal, the market is in a steady state equilibrium. if a difference of finite magnitude is created between the supply and demand rates and the consumer and the producer do not react to a price change induced by a difference in the supply and demand rates, the price will continue changing until the system saturates. this behavior can be depicted by the following formulation: inventory.market in changechange price ∝ change. price=p market, in theinventory in change== smmm − at timeinventory =m t, state.steady in inventory =sm ,=)(==outputinput dt dm dt mmd dt dm s− − ( ) .outputinput=or dtm −∫ ( ) or,rate demandratesupply change price dt−∝ ∫ ( ) ,rate demandratesupply = dtkp m −− ∫ where mk is the proportionality constant; supply and demand rate is the supply and demand per unit time respectively. a negative sign indicates that when ( )rate demandratesupply − is positive, then p is negative (i.e. price decreases). the above equation can be re-arranged as follows: ( ) or,=rate demandratesupply mk pdt −−∫ 202 n. nawaz / european journal of government and economics 6(2), 191-225 ( ) ,=0 m i k pdtww −−∫ [8] ,ratesupply =iw ,rate demand=0w =mk dimensional constant. let at time 0=t , supply rate = demand rate (market is in a steady state equilibrium), then eq. [8] can be written as ( ) 0.=0 dtww sis −∫ [9] the subscript s indicates the steady state equilibrium and 0=p in steady state. subtracting eq. [9] from eq. [8], we obtain ( ) ( ) or,=00 m sisi k pdtwwdtww −−−− ∫∫ ( ) ,=0 m i k pdtww −−∫ [10] rate,supply in change== where iisi www − rate. demandin change== 000 www s− ,p iw and 0w are deviation variables, which indicate deviation from the steady state equilibrium. the initial values of the deviation variables are zero. eq. (10) may also be written as follows: ,== mkwdtkp mm −− ∫ [11] where .= 0www i − if p gets a jump as a result of some factor other than an inventory change, such as imposition of a tax on consumer, that is considered as a separate input and can be added to eq. (11) as follows: .== jmkjwdtkp mm +−+− ∫ [11a] similarly, there can be an exogenous shock in inventory other than the price feedback. 203 n. nawaz / european journal of government and economics 6(2), 191-225 2.2 producer the producer maximizes the present discounted value of the future stream of profits, and his present value at time zero is as follows: ( ) ( )( ) ( )[ ] ,)()()(,)(=(0) 0 dtetittltwtltkftpv rt− ∞ ℜ−−∫ α [12] α is the fraction of the market price the producer charges to the middleman. r denotes the discount rate. )(tl (labor) and )(ti (level of investment) are the control variables and )(tk the state variable. the maximization problem can be written as { } ( ) ( )( ) ( )[ ] ,)()()(,)(=(0) 0 )(),( dtetittltwtltkftpvmax rt titl − ∞ ℜ−−∫ α subject to )()(=)( . tktitk δ− (state equation, describing how the state variable changes with time), 0=(0) kk (initial condition), 0)( ≥tk (non-negativity constraint on state variable), )(∞k free (terminal condition). the current-value hamiltonian for this case is ( ) ( )( ) ( ) [ ].)()()()()()(,)(=~ tktittittltwtltkftph δµα −+ℜ−− [13] now the maximizing conditions are as follows: )(i )(tl∗ and )(ti ∗ maximize h~ for all t : 0= ~ l h ∂ ∂ and 0,= ~ i h ∂ ∂ )(ii , ~ = . k hr ∂ ∂ −− µµ )(iii µ∂ ∂∗ hk ~ = . (this just gives back the state equation), )(iv 0=)()(lim rt t etkt − ∞→ µ (the transversality condition). the first two conditions are as follows: 204 n. nawaz / european journal of government and economics 6(2), 191-225 0,= ~ l h ∂ ∂ [14] 0,= ~ i h ∂ ∂ [15] and . ~ = . k hr ∂ ∂ −− µµ [16] in order to satisfy the condition of dynamic optimization after the price increase, the producer must increase the production level (see appendix). let p = market price, c = a reference price (such as the retail price which includes the production cost, profit of producer and profit of the middleman). c is a parameter which may vary with time or be kept fixed for a limited time period, e.g. the cost of a product may vary over time or can also remain constant for a while. it is the reference point with respect to which the variation in p is considered by the producer for decision making. price,in change todue productionin change=mw )( cp − acts as an incentive for the producer to produce more. therefore, or),( cpwm −∝α ).(= cpkw sm − [17] when the market is in equilibrium, then 0,=mw or ).(=0 sss cpk − [18] sk is the proportionality constant. sp and sc are the steady state equilibrium values. subtracting eq. [18] from eq. [17], ( ) ( )[ ] ( ) ,=== εsssssm kpckccppkw −−−−−− [19] where mw , c and p are deviation variables. 205 n. nawaz / european journal of government and economics 6(2), 191-225 2.3 consumer the consumer maximizes the present discounted value of the future stream of utilities, and his present value at time zero is as follows: ,))((=(0) 0 dtetxuv tρ− ∞ ∫ [20] where ρ denotes the discount rate and )(tx is the control variable. the maximization problem can be written as { } ,))((=(0) 0 )( dtetxuvmax t tx ρ− ∞ ∫ subject to )()()()()(=)( . txtptwtatrta −+ (state equation, describing how the state variable changes with time). )(ta is asset holdings (a state variable) and )(tw and )(tr are exogenous time path of wages and return on assets. saa =(0) (initial condition), 0)( ≥ta (non-negativity constraint on state variable), )(∞a free (terminal condition). the current-value hamiltonian for this case is ( ) ( ) ( ) ( ) ( )[ ].)())((=~ txtptwtatrttxuh −++ µ [21] now the maximizing conditions are as follows: )(i )(tx∗ maximizes h~ for all t : 0,= ~ x h ∂ ∂ )(ii , ~ = . a h ∂ ∂ −− ρµµ )(iii µ∂ ∂∗ ha ~ = . (this just gives back the state equation), )(iv 0=)()(lim t t etat ρµ − ∞→ (the transversality condition). the first two conditions are as follows: 206 n. nawaz / european journal of government and economics 6(2), 191-225 ( ) 0,=)()()(= ~ . tpttxu x h µ− ∂ ∂ [22] and ).()(= ~ = . trt a h µρµµ − ∂ ∂ −− [23] if the price of good x increases, the consumer faces (at the previous level of consumption) ( ) 0.<)()()(= ~ . tpttxu x h µ− ∂ ∂ therefore in order to satisfy the condition of dynamic optimization after the price increase, the consumer must decrease the consumption of good x . let the change in demand be proportional to the change in price, i.e. p . then we can write: or,demandin change p∝ .= pkw dd − [24] dw is the change in demand due to p ; when p is positive dw is negative. 3. solution of the model with a quantity tax the solution of the model can be written as ).(=)()()( tckktpkkk dt tdp smdsm ++ [25] if ttc =)( , i.e. the government imposes a per unit tax on producer at 0=t , then the above differential equation becomes as follows: .=)()()( tkktpkkk dt tdp smdsm ++ [26] the routh–hurwitz stability criterion (which provides a necessary and sufficient condition for stability of a linear dynamical system) for the stability of the above differential equation is 0>)( dsm kkk + , which holds as mk , sk and dk are all defined to be positive. this ensures that, away from a given initial equilibrium, every adjustment mechanism will lead to another equilibrium. now let’s look at the dynamics of the price if the quantity tax is imposed on the buyer 207 n. nawaz / european journal of government and economics 6(2), 191-225 instead. the market price is the buyer’s price as before, however, the producer will be taking into account the price before tax for his/ her production decisions. therefore, ).(=)( tptt −ε [27] this implies that ,=)()()( tkktpkkk dt tdp smdsm ++ which is the same as eq. [26]. the solution of the above differential equation with initial conditions of a buyer’s tax is as follows: the solution has the form [ ] .=)( )( 21 tdkskmkecctp +−+ [28] substituting the values of 1c and 2c in eq. [28] we obtain [ ] .=)( )( tdkskmk ds d ds s e kk tk kk tktp +− + + + [29] when 0,=t tp =(0) (the initial condition), and when ,= ∞t ds s kk tkp + ∞ =)( (the final steady state equilibrium value). in the final equilibrium, the quantity demanded must equal the quantity supplied, which holds (see appendix). 4. an optimal quantity tax path the efficiency loss as a result of a tax, generally mentioned in the economics literature is the dead weight loss as a result of comparisons of the pre and post tax market equilibriums. however, the dynamic picture shows that there is some efficiency loss on the dynamic adjustment path to the new equilibrium as well after the tax. after the imposition of the tax, the price jumps to a price equal to the previous equilibrium price plus the tax. the price then adjusts over time to bring the new equilibrium price which is higher than the previous equilibrium price and less than the price at the time the tax was imposed depending on the elasticity of demand and supply schedules. a pile up of inventory indicates a higher supply than demand, and a depletion of inventory occurs when demand is higher than the supply in a given time period. when the demand and supply are the same, there is no efficiency loss. if the demand and supply are different, the output and/ or consumption is being lost at that point in time. therefore if we sum up the inventory change at all points in time, we get the total efficiency loss, which is as follows: 208 n. nawaz / european journal of government and economics 6(2), 191-225 .)(= 0 dttmel ∫ ∞ [30] eq. [30] can be written as [ ] .)(1= 0 dttkkttp k el dm m −−− ∫ ∞ [30a] in figure 3, the inventory difference jumps to tkd , i.e. the decrease in demand because of tax at 0=t . the demand does not equal the supply any longer, and the market forces come into play. the inventory along with the price adjusts over time and arrives at the new equilibrium, i.e. ).(∞m the shaded area is the efficiency loss (the amount of output and/ or consumption lost) during the adjustment process. the area between the lines 0,=)(tm and )(=)( ∞mtm is the efficiency loss resulting from a difference in pre and post tax market equilibriums. the expression for the tax revenue is as follows: [ ].)((0)= tpkwttr did − [31] if we want to minimize the efficiency loss subject to the constraint that tax revenue generated is greater than or equal to g in a given time period, our problem is as follows: .s.t.min gtrel t ≥ figure 3. dynamic efficiency loss because of a quantity tax 209 n. nawaz / european journal of government and economics 6(2), 191-225 the choice variable is the tax rate, and the constraint is binding. the lagrangian for the above problem is as follows: [ ] [ ][ ])((0))(1= 0 tpkwtgdttkkttp k diddm m −−+−−− ∫ ∞ λl [ ] dttk k te kkk tk kkk tk d m tdkskmk dsm d dsm s       ++ + − + − +− ∞ ∫ )( 0 )()( = [ ] .(0) )(                     + + + −−+ +− tdkskmk ds d ds s did e kk tk kk tkkwtgλ taking the first order condition with respect to t , we get: [ ] [ ] . 2 )()( 1(0) = )( )( 0       + + +       + − + −+− +− +− ∞ ∫ tdkskmk ds d ds s d tdkskmk dsm d dsm s m did e kk k kk kk dte kkk k kkk k k kw t λ λ [32] taking the first order condition with respect to λ , we get: [ ] 0.=(0) )(             + + + −− +− tdkskmk ds d ds s did e kk tk kk tkkwtg [33] eq. [32] can also be written as . 2 (0)= q jwt id λ λ − [34] substituting the value of λ into eq. [34], we obtain . 2 4(0)(0) =)( 2 q qgww tt idid −− [35] a negative optimal tax is an optimal subsidy. the second order condition for minimization has been checked (see appendix). suppose that the government wants to generate a revenue of $1000 by imposing tax on a certain good. the initial equilibrium quantity of that good is 100, and the value of each one of ,mk sk and dk is equal to one. substituting these values in eq. [35] yields 11.27,= 2 400010000100=(0) −−t where ,0.50.5= 2teq −+ and at 0,=t 1=q . the tax revenue generated is 210 n. nawaz / european journal of government and economics 6(2), 191-225 [ ] 1000.=(0)= qtwttr id − now when ,= ∞t 0.5=q . this implies that 10.56.= 1 200010000100=)( −− ∞t the tax revenue is again 1000 as desired. therefore the optimal quantity taxation is that the government should impose a tax rate of $11.27 per unit quantity initially and then gradually decrease the tax rate over time up to a final tax rate of $10.56 per unit quantity of the same good. 5. conclusions when a government imposes a quantity/ commodity tax on the consumer, the price jumps to the pre tax equilibrium price plus the amount of the tax. the demand and supply adjust over time to bring the new post tax equilibrium. as a result of a tax, there are efficiency losses during the adjustment process as well as the new post tax equilibrium as compared to the pre-tax efficient equilibrium. it is important to take into consideration the efficiency losses during the adjustment process as well while deriving an optimal tax schedule. eq. (35) gives an optimal quantity tax path over time which generates the same desired revenue at any given point in time considering the adjustment of demand and supply over time. the expression is a function of the slopes of the demand, supply and the inventory curves, and the initial pre-tax equilibrium quantity. the expression is much more complex as compared to the optimal tax expression of ramsey, where he takes into account the efficiency losses just in the final equilibrium. regarding further research, a complete dynamic welfare analysis against various governmental policies, such as value added tax, income tax, toll tax, corporate tax, environmental tax, etc. can be carried out and the optimal governmental policy instruments can be derived following the methodology developed in this paper. acknowledgements i am grateful to two anonymous referees for their valuable comments and suggestions. references atkinson, a. b. and stiglitz, j. e.: (1976). the design of tax structure: direct versus indirect taxation, journal of public economics 6(1), 55-75. https://doi.org/10.1016/00472727(76)90041-4 chamley, c.: (1986). optimal taxation of capital income in general equilibrium with infinite lives, econometrica, 54(3), 607-622. https://doi.org/10.2307/1911310 211 n. nawaz / european journal of government and economics 6(2), 191-225 chari, v. v., christiano, l. j. and kehoe, p. j.: (1994). optimal fiscal policy in a business cycle model, journal of political economy 102(4), 617-652. https://doi.org/10.1086/261949 coleman, w. j. (2000). welfare and optimum dynamic taxation of consumption and income, journal of public economics, 76(1), 1-39. https://doi.org/10.1016/s0047-2727(99)00043-2 cremer, h. and gahvari, f.: (1993). tax evasion and optimal commodity taxation, journal of public economics, 50(2), 261-275. https://doi.org/10.1016/0047-2727(93)90052-u cremer, h. and gahvari, f. (1995). uncertainty and optimal taxation: in defense of commodity taxes, journal of public economics 56(2), 291-310. https://doi.org/10.1016/00472727(94)01422-k deaton, a. (1981). optimal taxes and the structure of preferences, econometrica, 49(5), 12451260. https://doi.org/10.2307/1912753 deaton, a. and stern, n. (1986). optimally uniform commodity taxes, taste differences and lump sum grants, economics letters, 20(3), 263-266. https://doi.org/10.1016/0165-1765(86)900352 diamond, p. a. (1975). a many-person ramsey tax rule, journal of public economics, 4(4), 335342. https://doi.org/10.1016/0047-2727(75)90009-2 diamond, p. a. and mirrlees, j. a. (1971). optimal taxation and public production ii: tax rules, american economic review, 61(3), 261-278. ekins, p. (1996). how large a carbon tax is justified by the secondary benefits of co2 abatement? resource and energy economics, 18(2), 161-187. https://doi.org/10.1016/09287655(96)00003-6 judd, k. l. (1985). redistributive taxation in a simple perfect foresight model, journal of public economics, 28(1), 59-83. https://doi.org/10.1016/0047-2727(85)90020-9 lucas, r. e. and stokey, n. l. (1983). optimal fiscal and monetary policy in an economy without capital, journal of monetary economics, 12(1), 55-93. https://doi.org/10.1016/03043932(83)90049-1 mirrlees, j. (1975). optimal commodity taxation in a two-class economy, journal of public economics, 4(1), 27-33. https://doi.org/10.1016/0047-2727(75)90021-3 naito, h.: (1999). re-examination of uniform commodity taxes under a non-linear income tax system and its implication for production efficiency, journal of public economics, 71(2), 165188. https://doi.org/10.1016/s0047-2727(98)00052-8 nordhaus, w. d. (1993). rolling the ‘dice’: an optimal transition path for controlling greenhouse gases, resource and energy economics, 15(1), 27-50. https://doi.org/10.1016/09287655(93)90017-o pizer, w. a. (2002). combining price and quantity controls to mitigate global climate change, journal of public economics, 85(3), 409-434. https://doi.org/10.1016/s0047-2727(01)00118-9 ramsey, f. p. (1927). a contribution to the theory of taxation, the economic journal, 37(145), 47-61. https://doi.org/10.2307/2222721 saez, e. (2002). the desirability of commodity taxation under non-linear income taxation and heterogeneous tastes, journal of public economics, 83(2), 217-230. https://doi.org/10.1086/261949 212 n. nawaz / european journal of government and economics 6(2), 191-225 https://doi.org/10.1016/s0047-2727(00)00159-6 appendix dynamic problem of the middleman in a dynamic setting, the middleman maximizes the present discounted value of the future stream of profits, and his present value at time zero is as follows: [ ] ,)),(()(=(0) 0 dteepmppqv rt− ∞ −∫ ς [36] where r denotes the discount rate, )(tp is the control variable and )(tm the state variable. the maximization problem can be written as { } [ ] ,)),(()(=(0) 0 )( dteepmppqvmax rt tp − ∞ −∫ ς subject to the constraints that )()),(())),((),(()())),((),((=)( . 1 .. 2 .. 1 . tpztpeztpetpmtpztpetpmtm + (state equation, describing how the state variable changes with time; z are exogenous factors), smm =(0) (initial condition), 0)( ≥tm (non-negativity constraint on state variable), )(∞m free (terminal condition). the current-value hamiltonian for this case is . )),(( ))),((),(())),((),(()()()))),((),((())(()(=~ 1 . . 2 . 1 .         ∗++− ztpe ztpetpmztpetpmtptztpetpmtpqtph µς [37] the maximizing conditions are as follows: )(i )(tp∗ maximizes h~ for all t : 0,= ~ p h ∂ ∂ )(ii , ~ = . m hr ∂ ∂ −− µµ )(iii µ∂ ∂∗ hm ~ = . (this just gives back the state equation), )(iv 0=)()(lim rt t etmt − ∞→ µ (the transversality condition). 213 n. nawaz / european journal of government and economics 6(2), 191-225 the first two conditions are         ∗+−+ ∂ ∂ )),(( ))),((),(())),((),(()))),((),((())(()())((= ~ 1 . . 21 . .. ztpe ztpetpmztpetpmztpetpmtpqtptpq p h ς         + ++ ∗+ )),(())),((),(()),(())),((),(( )),(())),((),(())),((),(( )()( 2 1 .. 221 .. 21 1 .. 12 .. 11 . ztpeztpetpmztpeztpetpm ztpeztpetpmztpetpm tptµ = 0 [38] and ))).),((),(((= ~ = .. ztpetpm m hr ςµµ ∂ ∂ −− [39] when the market is in equilibrium, 0,=)( . tp and the expression p h ∂ ∂ ~ boils down to the following:         ∗+−+ )),(( ))),((),(())),((),(()))),((),((())(()())(( 1 . . 21 . .. ztpe ztpetpmztpetpmztpetpmtpqtptpq ς 0,= , )),(( ))),((),(())),((),(()))),((),(((=))(())(()( 1 . . 21 . ..         ∗++ ztpe ztpetpmztpetpmztpetpmtpqtpqtp ς , ))(( )),(())),((),(( ))(( ))),((),(()))),((),(((=11)( . 1 .. 2 . 1 . .         +      + tpq ztpeztpetpm tpq ztpetpmztpetpm ticitydemandelas tp ς suggesting that the price equals the marginal cost (the right hand side of the above expression is the marginal cost in a dynamic setting, which is different from that in a static problem on account of the fact that in a dynamic setting the middleman also takes into account the impact of price chosen on his purchase price from the producer) when the demand is infinitely elastic. now suppose that as a result of a supply shock, if the middleman wants to hold an extra unit of inventory, then the marginal cost of holding an extra unit is higher because the term )))),((),((( . ztpetpmς is higher at the previous price at that point in time. the term in parentheses in the expression of the marginal cost, i.e. ))(( )),(())),((),(( ))(( ))),((),(( . 1 .. 2 . 1 . tpq ztpeztpetpm tpq ztpetpm + is a function of price and is the same at the previous price. this implies that on the previous price, now the middleman faces 214 n. nawaz / european journal of government and economics 6(2), 191-225         ∗+−+ ∂ ∂ )),(( ))),((),(())),((),(()))),((),((())(()())((= ~ 1 . . 21 . .. ztpe ztpetpmztpetpmztpetpmtpqtptpq p h ς         + ++ ∗+ )),(())),((),(()),(())),((),(( )),(())),((),(())),((),(( )()( 2 1 .. 221 .. 21 1 .. 12 .. 11 . ztpeztpetpmztpeztpetpm ztpeztpetpmztpetpm tptµ 0.< therefore in order to satisfy the condition of dynamic optimization, the middleman must decrease the price for an increase in inventory. this implies a negative relationship between price and inventory. the concept of inventory unifies the market supply and demand. if the supply and demand rates are equal, the market is in a steady state equilibrium. if a difference of finite magnitude is created between the supply and demand rates and the consumer and the producer do not react to a price change induced by a difference in the supply and demand rates, the price will continue changing until the system saturates. this behavior can be depicted by the following formulation: inventory.market in changechange price ∝ change. price=p market, in theinventory in change== smmm − at timeinventory =m t, state.steady in inventory =sm ,=)(==outputinput dt dm dt mmd dt dm s− − ( ) .outputinput=or dtm −∫ ( ) or,rate demandratesupply change price dt−∝ ∫ ( ) ,rate demandratesupply = dtkp m −− ∫ where mk is the proportionality constant. a negative sign indicates that when ( )rate demandratesupply − is positive, then p is negative (i.e. price decreases). the above equation can be re-arranged as follows: ( ) or,=rate demandratesupply mk pdt −−∫ ( ) ,=0 m i k pdtww −−∫ [40] 215 n. nawaz / european journal of government and economics 6(2), 191-225 ,ratesupply =iw ,rate demand=0w =mk dimensional constant. let at time 0=t , supply rate = demand rate (market is in a steady state equilibrium), then eq. (40) can be written as ( ) 0.=0 dtww sis −∫ [41] the subscript s indicates the steady state equilibrium and 0=p in steady state. subtracting eq. (41) from eq. (40) , we get: ( ) ( ) or,=00 m sisi k pdtwwdtww −−−− ∫∫ ( ) ,=0 m i k pdtww −−∫ [42] rate,supply in change== where iisi www − rate. demandin change== 000 www s− ,p iw and 0w are deviation variables, which indicate deviation from the steady state equilibrium. the initial values of the deviation variables are zero. eq. (42) may also be written as follows: ,== mkwdtkp mm −− ∫ [43] where .= 0www i − if p gets a jump as a result of some factor other than an inventory change, such as imposition of a tax on consumer, that is considered as a separate input and can be added to eq. (43) as follows: .== jmkjwdtkp mm +−+− ∫ [43a] similarly, there can be an exogenous shock in inventory other than the price feedback. producer the producer maximizes the present discounted value of the future stream of profits, and his present value at time zero is as follows: 216 n. nawaz / european journal of government and economics 6(2), 191-225 ( ) ( )( ) ( )[ ] ,)()()(,)(=(0) 0 dtetittltwtltkftpv rt− ∞ ℜ−−∫ α [44] where α is the fraction of the market price the producer charges to the middleman. r denotes the discount rate. )(tl (labor) and )(ti (level of investment) are the control variables and )(tk the state variable. the maximization problem can be written as { } ( ) ( )( ) ( )[ ] ,)()()(,)(=(0) 0 )(),( dtetittltwtltkftpvmax rt titl − ∞ ℜ−−∫ α subject to the constraints )()(=)( . tktitk δ− (state equation, describing how the state variable changes with time), 0=(0) kk (initial condition), 0)( ≥tk (non-negativity constraint on state variable), )(∞k free (terminal condition). the current-value hamiltonian for this case is ( ) ( )( ) ( ) [ ].)()()()()()(,)(=~ tktittittltwtltkftph δµα −+ℜ−− [45] now the maximizing conditions are as follows: )(i )(tl∗ and )(ti ∗ maximize h~ for all t : 0= ~ l h ∂ ∂ and 0,= ~ i h ∂ ∂ )(ii , ~ = . k hr ∂ ∂ −− µµ )(iii µ∂ ∂∗ hk ~ = . (this just gives back the state equation), )(iv 0=)()(lim rt t etkt − ∞→ µ (the transversality condition). the first two conditions are as follows: ( ) ( )( ) 0,=)(,)(= ~ . 2 twtltkftp l h − ∂ ∂ α [46] 0,=)()(= ~ tt i h µ+−ℜ ∂ ∂ [47] 217 n. nawaz / european journal of government and economics 6(2), 191-225 and ( ) ( )( ) .)(,)(= ~ = . 1 .     −− ∂ ∂ −− ttltkftp k hr δµαµµ [48) substituting the value of . µ and µ from eq. (47) into eq. (48) yields ( ) ( )( ) 0.=)()()(,)( .. 1 ttrtltkftp ℜ+ℜ+− δα if the price, i.e. )(tp goes up, (at the previous level of investment and labor) the producer faces ( ) ( )( ) 0,>)(,)( . 2 twtltkftp −α ( ) ( )( ) 0.>)()()(,)( .. 1 ttrtltkftp ℜ+ℜ+− δα therefore, in order to satisfy the condition of dynamic optimization after the price increase, the producer must increase the production level. let p = market price, c = a reference price (such as the retail price which includes the production cost, profit of producer and profit of the middleman). c is a parameter which may vary with time or be kept fixed for a limited time period, e.g. the cost of a product may vary over time or can also remain constant for a while. it is the reference point with respect to which the variation in p is considered by the producer for decision making. price,in change todue productionin change=mw where )( cp − acts as an incentive for the producer to produce more. therefore, or),( cpwm −∝α ).(= cpkw sm − [49] when the market is in equilibrium, then 0,=mw or ).(=0 sss cpk − [50] where sk is the proportionality constant. sp and sc are the steady state equilibrium values. subtracting eq. (50) from eq. (49) , we get: ( ) ( )[ ] ( ) ,=== εsssssm kpckccppkw −−−−−− [51] where cwm , and p are deviation variables. 218 n. nawaz / european journal of government and economics 6(2), 191-225 consumer the consumer maximizes the present discounted value of the future stream of utilities, and his present value at time zero is as follows: ,))((=(0) 0 dtetxuv tρ− ∞ ∫ [52] where ρ denotes the discount rate and )(tx is the control variable. the maximization problem can be written as { } ,))((=(0) 0 )( dtetxuvmax t tx ρ− ∞ ∫ subject to )()()()()(=)( . txtptwtatrta −+ (state equation, describing how the state variable changes with time). )(ta is asset holdings (a state variable) and )(tw and )(tr are exogenous time path of wages and return on assets. saa =(0) (initial condition), 0)( ≥ta (non-negativity constraint on state variable), )(∞a free (terminal condition). the current-value hamiltonian for this case is ( ) ( ) ( ) ( ) ( )[ ].)())((=~ txtptwtatrttxuh −++ µ [53] the maximizing conditions are as follows: )(i )(tx∗ maximizes h~ for all t : 0,= ~ x h ∂ ∂ )(ii , ~ = . a h ∂ ∂ −− ρµµ )(iii µ∂ ∂∗ ha ~ = . (this just gives back the state equation), )(iv 0=)()(lim t t etat ρµ − ∞→ (the transversality condition). the first two conditions are as follows: ( ) 0,=)()()(= ~ . tpttxu x h µ− ∂ ∂ [54] and ).()(= ~ = . trt a h µρµµ − ∂ ∂ −− [55] 219 n. nawaz / european journal of government and economics 6(2), 191-225 if the price of good x goes up, the consumer faces (at the previous level of consumption) ( ) 0.<)()()(= ~ . tpttxu x h µ− ∂ ∂ therefore, in order to satisfy the condition of dynamic optimization after the price increase, the consumer must decrease the consumption of good x . let the change in demand be proportional to the change in price, i.e. p . then we can write: or,demandin change p∝ .= pkw dd − [56] where dw is the change in demand due to p ; when p is positive dw is negative. solution of the model with a quantity tax from eqs. )(11a , (19) and (24) we have the following expressions: ),(=)( twk dt tdp m− ),(=)( tktw sm ε− ),()(=)( tptct −ε ),(=)( tpktw dd − and ),()(=)( twtwtw dm − if there is no exogenous change in supply and demand. from the above equations, we obtain [ ])()(=)( twtwk dt tdp dmm −− [ ])()(= tpktkk dsm +−− ε [ ].)()()(= tpkktckk dssm ++−− the above expression can be rearranged as follows: ).(=)()()( tckktpkkk dt tdp smdsm ++ [57] 220 n. nawaz / european journal of government and economics 6(2), 191-225 if ttc =)( , i.e. the government imposes a per unit tax on producer at 0=t , then the above differential equation becomes as follows: .=)()()( tkktpkkk dt tdp smdsm ++ [58] now let’s look at the dynamics of the price if the quantity tax is imposed on the buyer instead. we start from the same expressions as we did for a producer’s tax, i.e. [ ],)()(=)( twtwk dt tdp dmm −− ),(=)( tktw sm ε− ).(=)( tpktw dd − the market price is the buyer’s price as before, however, the producer will be taking into account the price before tax for his/her production decisions. therefore, ).(=)( tptt −ε [59] this implies that { }[ ],)()(=)( tpkttpkk dt tdp dsm +−− ,=)()()( tkktpkkk dt tdp smdsm ++ which is the same as eq. [58]. in order to solve the above differential equation with initial conditions of a buyer’s tax, we proceed as follows. the characteristic function of the differential equation is 0.=)( dsm kkkx ++ the characteristic function has a single root given by ).(= dsm kkkx +− thus, the complementary solution is [ ] .=)( )( 2 tdkskmk c ectp +− the particular solution has the form .=)( 1ctpp thus, the solution has the form [ ] .=)( )( 21 tdkskmkecctp +−+ [60] the constant 1c is determined by substitution into the differential equation: 221 n. nawaz / european journal of government and economics 6(2), 191-225 [ ] [ ] ,=)()()( )( 21 )( 2 tkkeckkkckkkeckkk sm tdkskmk dsmdsm tdkskmk dsm +−+− +++++− .=1 ds s kk tkc + 2c is determined by the initial condition by ,==(0) 2 tc kk tkp ds s + + ds s kk tktc + −=2 ds sds kk tktktk + −+= .= ds d kk tk + substituting the values of 1c and 2c in eq. (60), we obtain [ ] .=)( )( tdkskmk ds d ds s e kk tk kk tktp +− + + + [61] when 0,=t tp =(0) (the initial condition), and when ,= ∞t ds s kk tkp + ∞ =)( (the final steady state equilibrium value). in the final equilibrium, the quantity demanded must equal the quantity supplied. in order to verify that this holds, we proceed as follows: from eq. [56], the change in demand due to a change in price after the tax is as follows: ),(=)( tpktw dd − ),(=(0))(or tpkwtw didnd −− where (0)idw is the initial demand and )(twnd is the new demand after tax, because )(twd is a deviation variable, i.e. deviation from the initial equilibrium value. similarly from eq. [51], for the supply ),(=)( tktw sm ε− [ ].)(=(0))( tptkwtw simnm −−− in the final equilibrium or),(=)( ∞∞ ndnm ww [ ] ),((0)=)((0) ∞−∞−− pkwptkw didsim 222 n. nawaz / european journal of government and economics 6(2), 191-225 which holds as in the initial equilibrium, the quantity demanded equals the quantity supplied, i.e. (0).=(0) idim ww an optimal quantity tax path the efficiency loss as a result of a tax, generally mentioned in the economics literature is the dead weight loss as a result of comparisons of the pre and post tax market equilibriums. however, the dynamic picture shows that there is some efficiency loss on the dynamic adjustment path to the new equilibrium as well after the tax. after the imposition of the tax, the price jumps to a price equal to the previous equilibrium price plus the tax. the price then adjusts over time to bring the new equilibrium price which is higher than the previous equilibrium price and less than the price at the time the tax was imposed depending on the elasticity of demand and supply schedules. a pile up of inventory indicates a higher supply than demand, and a depletion of inventory occurs when demand is higher than the supply in a given time period. when the demand and supply are the same, there is no efficiency loss. if the demand and supply are different, the output and/ or consumption is being lost at that point in time. therefore if we sum up the inventory change at all points in time, we get the total efficiency loss, which is as follows: .)(= 0 dttmel ∫ ∞ [62] from eq. [43a], we have .)(=)( jtmktp m +− the value of j can be found by imposing the initial conditions, i.e. tp =(0) , and tkm d=(0) (the decrease in demand because of tax at 0=t from eq. (56)), therefore .= tkktj dm+ substituting the value of j in eq. ),(43a we get ,)(=)( tkkttmktp dmm ++− and eq. [62] can be written as [ ] .)(1= 0 dttkkttp k el dm m −−− ∫ ∞ [62.a] in figure 3, the inventory difference jumps to tkd , i.e. the decrease in demand because of tax at 0=t . the demand does not equal the supply any longer, and the market forces come into play. the inventory along with the price adjusts over time and arrives at the new equilibrium, i.e. ).(∞m the shaded area is the efficiency loss (the amount of output and/ or consumption lost) 223 n. nawaz / european journal of government and economics 6(2), 191-225 during the adjustment process. the area between the lines 0,=)(tm and )(=)( ∞mtm is the efficiency loss resulting from a difference in pre and post tax market equilibriums. from eq. [56], the change in demand due to a change in price after the tax is as follows: ),(=)( tpktw dd − ),(=(0))(or tpkwtw didnd −− where (0)idw is the initial demand and )(twnd is the new demand after tax, because )(twd is a deviation variable, i.e. deviation from the initial equilibrium value. therefore the expression for the tax revenue is as follows: [ ].)((0)= tpkwttr did − [63] if we want to minimize the efficiency loss subject to the constraint that tax revenue generated is greater than or equal to g in a given time period, our problem is as follows: .s.t.min gtrel t ≥ the choice variable is the tax rate, and the constraint is binding. the lagrangian for the above problem is as follows: [ ] [ ][ ])((0))(1= 0 tpkwtgdttkkttp k diddm m −−+−−− ∫ ∞ λl [ ] dttk k te kkk tk kkk tk d m tdkskmk dsm d dsm s       ++ + − + − +− ∞ ∫ )( 0 )()( = [ ] .(0) )(                     + + + −−+ +− tdkskmk ds d ds s did e kk tk kk tkkwtgλ taking the first order condition with respect to t , we get: [ ] dtk k e kkk k kkk k d m tdkskmk dsm d dsm s       ++ + − + − +− ∞ ∫ 1 )()( )( 0 [ ]             + + + −− +− tdkskmk ds d ds s did e kk k kk ktkw )((0)λ [ ] 0.=)(       + + + + +− tdkskmk ds d ds s d e kk k kk ktkλ 224 n. nawaz / european journal of government and economics 6(2), 191-225 this implies [ ] dte kkk k kkk k k k tdkskmk dsm d dsm s m d       + − + −+ +− ∞ ∫ )( 0 )()( 1 [ ]       + + + + +− tdkskmk ds d ds s d e kk k kk ktk )(2λ (0),= idwλ or [ ] [ ] . 2 )()( 1(0) = )( )( 0       + + +       + − + −+− +− +− ∞ ∫ tdkskmk ds d ds s d tdkskmk dsm d dsm s m did e kk k kk kk dte kkk k kkk k k kw t λ λ [64] taking the first order condition with respect to λ , we get: [ ] 0.=(0) )(             + + + −− +− tdkskmk ds d ds s did e kk tk kk tkkwtg [65] substituting the value of t from eq. [64] into [65], [ ] [ ]       + + +       + − + −+− +− +− ∞ ∫ tdkskmk ds d ds s d tdkskmk dsm d dsm s m did id e kk k kk kk dte kkk k kkk k k kw wg )( )( 0 2 )()( 1(0) (0).= λ λ [ ]       + + + − +− tdkskmk ds d ds s d e kk k kk kk )( [ ] [ ] , 2 )()( 1(0) 2 )( )( 0                     + + +       + − + −+− ∗ +− +− ∞ ∫ tdkskmk ds d ds s d tdkskmk dsm d dsm s m did e kk k kk kk dte kkk k kkk k k kw λ λ ,(0)2(0)(0)2(0)2=4or 222222 jwjwjwwqg idididid λλλλλ +−−− [ ] ,=where )(       + + + +− tdkskmk ds d ds s d e kk k kk kkq 225 n. nawaz / european journal of government and economics 6(2), 191-225 [ ] . )()( 1= )( 0 dte kkk k kkk k k kj tdkskmk dsm d dsm s m d       + − + −+ +− ∞ ∫ this implies that { } 0.=4(0) 222 jqgwid −− λ . 4(0) = 2 qgw j id − λ eq. [64] can also be written as . 2 (0)= q jwt id λ λ − [66] substituting the value of λ into eq. [66], we get: , 4(0) 2 4(0) (0) =)( 2 2 qgw qj j qgw jw tt id id id − − − . 2 4(0)(0) =)( 2 q qgww tt idid −− [67] a negative optimal tax is an optimal subsidy. in order to check the second order condition for minimization, we proceed as follows: the lagrangian can be written as ( )[ ].(0)= qtwtgjt id −−+ λl the bordered hessian matrix of the lagrange function is as follows: , 4(0) 22(0) 2(0)0 = 2           − − − qgw qjqtw qtw bh id id id the determinant of which is negative as ( ) 0,<2(0) 2qtwid −− which implies that the efficiency loss is minimized. abstract. following ramsey, the existing literature on optimal quantity taxation only compares the pre and the post-tax market equilibriums in order to account for the efficiency losses. however, when the government imposes a quantity tax on the consu... keywords. quantity tax; dynamic efficiency; adjustment path; equilibrium jel classification. h20; h21; h22 2. the model i am grateful to two anonymous referees for their valuable comments and suggestions. references an inclusive taxonomy of behavioral biases european journal of government and economics 6(1), june 2017, 24-58. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 an inclusive taxonomy of behavioral biases david peóna, *, manel antelob, anxo calvo-silvosaa a department of business studies, universidade da coruña, spain. b department of economics, university of santiago de compostela, spain. * corresponding author at: departamento de empresa. universidade da coruña, campus de elviña s/n, 15071 a coruña, spain. email: david.peon@udc.es article history. received 25 may 2016; first revision required 9 december 2016; accepted 27 february 2017. abstract. this paper overviews the theoretical and empirical research on behavioral biases and their influence in the literature. to provide a systematic exposition, we present a unified framework that takes the reader through an original taxonomy, based on the reviews of relevant authors in the field. in particular, we establish three broad categories that may be distinguished: heuristics and biases; choices, values and frames; and social factors. we then describe the main biases within each category, and revise the main theoretical and empirical developments, linking each bias with other biases and anomalies that are related to them, according to the literature. keywords. behavioral biases; decision-making; heuristics; framing; prospect theory; social contagion. jel classification. d03; g02; g11; g14; g30 1. introduction the standard model of rational choice argues that people choose to follow the option that maximizes expected utility. however, this ignores the presence of behavioral biases, i.e. the tendency to reason in certain ways that can lead to systematic deviations from a standard of rationality (shefrin, 2006). both psychology and behavioral economics have shown that people are vulnerable to biases and use shortcuts in thinking, exhibit biases in decision-making and frame their decisions, exhibit preference reversals and struggle to commit with their decisions in the past, and they are influenced by others’ behavior. this leads to anomalies and decision effects, that is, empirical results that are difficult to rationalize within the paradigm (khaneman, knetsch and thaler, 1991). this paper surveys the main biases in the behavioral economics and finance, leaving aside their behavioral consequences – anomalies, when they refer to market outcomes or competition among firms, and decision effects, when they refer to people’s actions which, given the number of them and extensive literature, deserve a separate review. the literature of behavioral http://www.ejge.org/ peón et al. / european journal of government and economics 6(1), 24-58 25 biases is so vast and boundless that trying to cover them all in detail would be unfeasible. thus, and in order to make it particularly helpful for non-initiated readers, we contribute in three instances. first, we provide an original taxonomy that is based on the reviews of relevant authors in the field. we then describe the most significant of those biases, and review the main contributions in regards to the theoretical, empirical and experimental developments. the impact of the contributions was filtered by their number of citations in the scopus database. finally, we provide a critical discussion in terms of the biases and anomalies that are linked to them, the lines of open debate and research, as well as the policy implications, according to the literature. the remainder of the article is laid out as follows: section 2 provides a taxonomy of biases classified in three groups; section 3 reviews the main heuristics and judgmental biases; section 4 is dedicated to choices, values and frames; section 5 surveys the main social factors; finally, section 6 analyzes some policy implications of the biases described. 2. searching for an inclusive taxonomy of behavioral most taxonomies of behavioral biases available use diverse classification rules and different names for similar concepts, what makes it difficult to provide an inclusive list satisfying all criteria. to circumvent these limitations, we start from some of the reviews provided by the founders of the field, including some nobel prize winners, to end up blending their views in a more inclusive taxonomy. they follow in order. kahneman, slovic and tversky (1982) list heuristics and biases in seven categories: representativeness, causality and attribution, covariation and control, overconfidence, conservatism, availability, and judgmental biases in risk perception. tversky and kahneman (1992) see five major phenomena: framing effects, nonlinear preferences, source dependence, risk seeking and loss aversion. plous (1993) separates perception, memory, and context; heuristics and biases; framing; models of decision-making; and social effects. kahneman and riepe (1998) classify heuristics, errors of preference –loss aversion and prospect theory (pt) and framing. rabin (1998) distinguishes mild biases (e.g. loss aversion), severe biases in judgment under uncertainty (e.g. confirmatory bias) and those implying a radical critique of the maximizing utility model (framing effects, preference reversals, and self-control). shiller (2000a) includes pt, regret and cognitive dissonance, mental accounting, representativeness, and overconfidence. mullainathan and thaler (2000) note three deviations from the standard model (bounded rationality, bounded willpower and bounded self-interest). barberis and thaler (2003) label beliefs (e.g. representativeness) and preferences (pt and ambiguity aversion). camerer and loewenstein (2004) list probability judgments (e.g. heuristics) and preferences (framing, anchoring, loss aversion, reference dependence, preference reversals, and hyperbolic discounting). akerlof and shiller (2009) note five aspects of animal spirits, including feedback mechanisms, attitudes about fairness, and social contagion. dellavigna (2009) separates non-standard preferences, non-standard beliefs, and non-standard peón et al. / european journal of government and economics 6(1), 24-58 26 decision-making. finally, recent surveys separate investor beliefs and preferences (sahi, arora and dhameja, 2013), sources of judgment and decision biases (hirshleifer, 2015). following the above, our taxonomy separates three categories: heuristics and judgmental biases; choices, values and frames; and social factors. this choice requires some clarification in regards to the terminology used. first, we use the generic term behavioral biases –or, simply, biasesto refer to any of them, while judgmental biases are a specific type of systematic errors that are induced by heuristics. second, the categories are devised following some authors in particular. we initially followed the spirit of kahneman and tversky’s work, which distinguishes (i) the heuristics that people use and the biases to which they are prone when judging in an uncertain context, (ii) the prospect theory, as a model of choice under risk, and loss aversion in riskless choice, and (iii) the framing effects (kahneman, 2003a,b). then, we merged pt (preferences, broadly speaking) and framing in a single category. we do this following tversky and kahneman (1981), who consider two phases in the choice process –an initial of framing and a subsequent of evaluation-, and barberis and huang (2009), who suggest framing and prospect theory form a natural pair. to name this category, we use the term ‘choices, values and frames’ following the classical article of kahneman and tversky (1984). finally, we include a third category of social factors, which refer to cultural and social influences on individuals’ behavior. plous (1993), shefrin (2000), and hens and bachmann (2008), among many others, advocate for this category. 3. heuristics and judgmental biases heuristics refer to economic shortcuts for information processing, or simple rules that ignore information (marewski, gaissmaier and gigerenzer, 2010). since information is vast, disperse, changes continuously and its gathering is costly, people develop rules of thumb to make decisions, what often leads them to make some errors (shefrin, 2000). griffin et al. (2012) provide a historical overview. in its initial conception, heuristics were restricted to the domain of judgment under uncertainty, a scope later broadened (kahneman and frederick, 2002) to a variety of fields that share a common process of attribute substitution. in other words, “difficult judgments are made by substituting conceptually or semantically related assessments that are simpler and more readily accessible” (kahneman and frederick, 2005: 287). open debate researchers focus on whether and when people rely on heuristics (e.g. cokely and kelley, 2009) or how accurate they are for predicting uncertain events (e.g. ortmann et al., 2008). however, two contrary views prevail. authors like gigerenzer and gaissmaier (2011) argue that heuristics are efficient shortcuts for inference, adaptive strategies that evolved in tandem with fundamental psychological mechanisms (goldstein and gigerenzer, 2002). no rule is assumed to be rational per se; what matters is to understand when a given heuristic performs better –a peón et al. / european journal of government and economics 6(1), 24-58 27 concept named ecological rationality. contrariwise, other authors identify two cognitive systems, reason and intuition, being the latter norm. in these dual-process theories (kahneman and frederick, 2005), heuristics would be the fast, intuitive, affect-driven and effortless cognitive system. through the process of attribution substitution, a target attribute of the judged object is substituted by a heuristic attribute, and since the target and heuristic attributes are different, it induces systematic errors in judgment and decision, known as judgmental biases. currently, the debate stands between those who observe a natural tendency to make errors – e.g. lacetera, pope and sydnor (2014) show heuristics matter even in markets with easily observed information and those who favor the ecological rationality – e.g. norman et al. (2014) see that encouraging increasing attention to analytical thinking does not improve diagnostic accuracy. in table 1 we collect some relevant heuristics and the judgmental biases associated to them. since both concepts specify how agents form expectations, there are authors who merge them in the same category. nonetheless, most researchers ― e.g. the original approach by tversky and kahneman (1974) ― consider first the heuristics people use, and then the biases they lead to. 3.1 availability heuristic availability is an information selection bias where the probability of an event is estimated by the ease with which occurrences can be brought to mind (tversky and kahneman, 1973). due to our limited attention, memory and processing capacities, we make decisions based on subsets of information that are easily available. the heuristic contributes to judgmental biases such as attention anomalies and an overreaction to new information (hens and bachmann, 2008), and the hindsight bias (camerer and loewenstein, 2004). related judgmental biases attention is a scarce resource and our ability to process information limited. an attention bias follows if the attributes that catch our attention are not critical, leading to suboptimal choices. memory has a limited capacity, too, so it works by reconstruction. a hindsight bias may result as a side-effect: in hindsight we exaggerate what we might have anticipated in foresight (fischhoff, 1982). the availability heuristic contributes to the bias, because events that occurred are easier to imagine than counterfactual ones (camerer and loewenstein, 2004). classic articles include odean (1999) on the attention bias and the excessive trading in financial markets, barber and odean (2008) on three indicators of attention for stock investors, and pan and statman (2010), who suggest that the hindsight bias amplifies regret. european journal of government and economics 6(1), june 2017, 24-58. table 1. heuristics and judgmental biases. heuristic judgmental biases related concepts literature availability attention bias overreaction availability and overreaction to new info (hens and bachmann, 2008) earnings announcement drift hirshleifer and teoh (2003): attention and earnings drift hindsight bias camerer and loewenstein (2004): availability contributes to hindsight bias representativeness law of small numbers gambler's fallacy tversky & kahneman (1974): gambler's fallacy and law of small numbers hot hand fallacy momentum and reversals rabin and vayanos (2010) extrapolation bias hens and bachmann (2008): extrapolation bias and representativeness base rate neglect cognitive dissonance tversky and kahneman (1982a) illusion of validity tversky and kahneman (1974) causality and attribution kahneman et al. (1982) conjunction & conjunction fallacy firstly considered a consequence of anchoring, but of representativeness after tversky and kahneman (1983). anchoring-andadjustment disjunction fallacies reference points anchoring falls from the heuristics list (kahneman and frederick, 2002) affect risk-as-feelings finucane et al. (2000) familiarity aversion to ambiguity status quo bias familiarity, aversion to ambiguity and status quo bias (ackert et al., 2005) recognition heuristic endowment effect recognition (gigerenzer et al., 1991), fluency (marewski et al., 2010) fluency heuristic home bias, underdiversif. seiler et al. (2013): familiarity and home bias (excessive) optimism wishful thinking barberis and thaler (2003) overconfidence self attribution bias cognitive dissonance moore and healy (2008) daniel et al. (1998): self-attribution and cognitive dissonance underand overreaction odean (1998): overconfidence and under/overreaction confirmation bias illusion of validity griffin and tversky (1992): illusion of validity and confirmation bias illusion of control shefrin (2000): illusion of control and overconfidence european journal of government and economics 6(1), june 2017, 24-58. open debate the clash between the efficient and the inefficient shortcut views stands on whether the availability heuristic is useful to assess probability because instances of large classes are better recalled, or it leads to decision biases since it is affected by factors other than frequency –e.g. imagination, familiarity and salience. thus, heath, larrick and klayman (1998) argue its effects are ubiquitous because of a lack of experience with unusual events. instead, the efficient approach suggests that results like the hindsight bias, rather than a reconstruction of the prior judgment, is a by-product of the adaptive process of updating of knowledge after feedback (hoffrage, hertwing and gigerenzer, 2000). recent research on the availability heuristic shows its effect on social media (chou and edge, 2012). the attention bias might explain the post-earnings announcement drift (hirshleifer and teoh, 2003) and the accruals anomaly (battalio et al., 2012), though cready et al. (2014) criticize the spurious effects attributable to misclassification of transactions. recent research on the hindsight bias includes theoretical (roese and vohs, 2012) and experimental research – chelley-steeley, kluger and steeley (2015) obtain positive results, calvillo (2014) highlights individual differences. 3.2 representativeness heuristic tversky and kahneman (1983) define representativeness as the degree of correspondence between an outcome and a model. it implies a tendency to rely on stereotypes, particularly when it comes to estimating probabilities (shleifer, 2000). hence, the representativeness heuristic explains several biases of judgment under uncertainty. we see them next. related judgmental biases one intuition people have about random sampling is the law of small numbers, a tendency to exaggerate how closely a small sample will resemble the parent population (tversky and kahneman, 1971). linked to representativeness after tversky and kahneman (1974), it leads to a gambler’s fallacy (rabin, 1998), a belief in the hot hand fallacy (rabin, 2002), and the extrapolation bias (shefrin, 2000). the gambler’s fallacy is a classic misconception of what regression to the mean implies: a belief that random sequences should exhibit systematic reversals (rabin and vayanos, 2010). similarly, a hot hand fallacy implies a failure to appreciate statistical independence, but involves instead the belief in an excessive persistence rather than reversals. related to that, the extrapolation bias suggests that people bet on trends (shefrin, 2000). the lack of expertise in probability assessment is related to two other biases. prior probabilities (base-rate frequencies) play a key role in probability assessment but none on peón et al. / european journal of government and economics 6(1), 24-58 25 representativeness, implying a base rate neglect (tversky and kahneman, 1974). prendergast and stole (1996) relate it to a cognitive dissonance reduction, where individuals overweight their own information. moreover, a conjunction fallacy appears when people believe the probability of a conjunction of two events is greater than that of one of its constituents. bar-hillel (1973) set an antecedent, though the fallacy is original of tversky and kahneman (1982b) and their classic linda experiment. finally, two additional judgmental biases related to the representativeness heuristic are an illusion of validity, when the confidence people have in their predictions depends on the degree of representativeness (einhorn and hogarth 1978), and causality and attribution, when people attempt to infer the causes of the effects observed and incur in errors related to salience, availability and representativeness –after attribution theory by weiner (1985). open debate recent advances in the study of representativeness include a memory-based model of probabilistic inference by gennaioli and shleifer (2010), and empirical evidence of a bayesian updating failure (alós-ferrer and hügelschäfer, 2012). there is also consistent evidence of most judgmental biases in different instances. thus, huber, kirchler and stöckl (2010) obtain experimental evidence of a gambler’s fallacy effect in investment decisions, while rieger (2012) and erceg and galic (2014) perform experimental tests of the effects of conjunction and disjunction fallacies on markets. liberali et al. (2012) explore the mechanisms underlying how individual differences in numeracy lead to these biases. notwithstanding, a controversial judgmental bias today is the base rate neglect (gigerenzer, 1991). first, it seems in contradiction to the widespread belief that judgments are affected by stereotypes (landman and manis, 1983). besides, in regards to the efficient shortcuts debate, cosmides and tooby (1990) rephrased in a frequentist way the questions in the experimental research of tversky and kahneman (1982a), and found the base-rate fallacy disappeared. a recent contribution by pennycook et al. (2014) offers a mixed interpretation: though base rates are indeed neglected, they may be accessible through intuitive reasoning. other minor sources of disagreement include whether men (suetens and tyran, 2012) or women (stöckl et al., 2015) are more prone to display a hot hand fallacy. 3.3 affect heuristic the list of heuristics changed after the concept of attribution substitution was introduced by kahneman and frederick (2002). on one hand, anchoring did not fit as a heuristic anymore, as it does not work through the substitution of one attribute for another. ever since, most authors (e.g. camerer and loewenstein, 2004) label it as an error of preference that derives from the existence of reference points (see section 4). on the other hand, it put the affect heuristic (finucane et al., 2000) on the list. the heuristic is driven by affect, a natural assessment, peón et al. / european journal of government and economics 6(1), 24-58 26 automatically computed and always accessible, so the basic evaluative attribute (e.g. good/bad, like/dislike) is a candidate for substitution in any task that calls for a favorable or unfavorable response. open debate failing to identify the affect heuristic “reflects the narrowly cognitive focus that characterized psychology for some decades. there is now compelling evidence that every stimulus evokes an affective evaluation” (kahneman and frederick, 2002: 55). affect provides a faster intuition than retrieving from memory. recent contributions include theoretical (haack, pfarrer and scherer, 2014) and experimental (pachur and galesic, 2013; jaspersen and aseervatham, 2015). a sideline theory is the model of risk-as-feelings (loewenstein et al., 2001, slovic et al., 2002), an alternative to cognitive theories of choice under risk that emphasizes the role affect plays: beliefs about risk would be expressions of emotion that often diverge from cognitive assessments. lupton (2013) further elaborates the theory, arguing that both emotion and risk judgments are collectively configured via social and cultural processes. 3.4 familiarity familiarity is the most common name in the literature to refer to a set of emotionally and cognitively driven heuristics. on one hand, there is evidence we make decisions based on the degree of closeness we feel about different alternatives. thus, familiarity is related to fear of change and the unknown (cao et al., 2011) and to ambiguity aversion. on the other, the recognition (gigerenzer, hoffrage and kleinbölting, 1991), and fluency heuristics (marewski et al., 2010) show that the reasons for familiarity may be cognitive as well. heuristics and related judgmental biases two processes govern the recognition heuristic, recognition and evaluation. recognition is the capacity to make inferences in cases of limited knowledge (goldstein and gigerenzer, 2002: 75): “if one of two objects is recognized and the other is not, recognition heuristic infers that the recognized object has the higher value with respect to the criterion”. evaluation judges the heuristic as ecologically rational whenever the recognition validity for a given criterion is much higher than chance. it allows people to benefit from ignorance by making inferences from memory and patterns of missing knowledge. in case two alternatives are recognized, the fluency heuristic fills the gap: if one alternative is recognized faster than another, the heuristic infers the one with the higher value (schooler and hertwig, 2005). schwikert and curran (2014) analyze the memory processes that contribute to the recognition and fluency heuristics. related to familiarity is an aversion to ambiguity (ackert et al., 2005). if ambiguity is the uncertainty about uncertainties (einhorn and hogarth, 1986), ambiguity aversion describes a peón et al. / european journal of government and economics 6(1), 24-58 27 preference for known over unknown risks, as shown in the ellsberg paradox (thaler, 1983). early papers include fellner (1961), who introduced decision weights. open debate recent advances to understand how familiarity and ambiguity aversion operate include neurogenetic studies (chew, ebstein and zhong, 2012). they would help explain anomalies such as the status quo bias (ackert et al., 2005), underdiversification (boyle et al., 2012), and their implications on insurance (alary, gollier and treich, 2013) and asset pricing (füllbrunn, rau and weitzel, 2014). however, this is an open field of research, as contradictory results were obtained. roca, hogarth and maule (2006) show that the status quo bias could lead to ambiguity seeking, and einhorn and hogarth (1986) specify some conditions for ambiguity seeking and avoidance. etner, jeleva and tallon (2012) provide a review on advances in the field. regarding recognition, being the most frugal heuristics (goldstein and gigerenzer, 1999), the debate centers around its efficiency: if ignorance is systematically distributed, recognition and criterion are correlated and the heuristic leads to efficient results. schooler and hertwig (2005) suggest a beneficial forgetting, where loss of information aids inference heuristics that exploit mnemonic information, while ortmann et al. (2008) get mixed results when analyzing how the heuristic performs in portfolio management. gigerenzer and goldstein (2011) survey the literature. 3.5 excessive optimism and overconfidence excessive optimism and overconfidence are two of the most relevant heuristic-driven biases. however, they are often confounded in the literature. indeed, overconfidence may refer to different concepts, what added more noise to the debate. optimists overestimate favorable outcomes and underestimate unfavorable ones (shefrin, 2006). overconfidence, instead, may refer to three different concepts (moore and healy, 2008): overestimation in estimating our own performance; overplacement (better-than-average effect) in estimating our own performance relative to others; and overprecision, an excessive precision to estimate future uncertainty, what entails a miscalibration of subjective probabilities. open debate behaviorists suggest it is heuristics and cognitive biases that cause the overconfidence phenomenon. however, two alternative views are the brunswikian or ecological models (gigerenzer et al., 1991), according to which people are good judges of the reliability of their knowledge as long as such knowledge is representatively sampled, and thurstonian or error models (erev, wallsten and budescu, 1994), which interpret overconfidence as merely an peón et al. / european journal of government and economics 6(1), 24-58 28 illusion, created by unrecognized regression. despite its popularity, the behaviorist interpretation does not provide a clear answer on which heuristics or biases drive excessive optimism and overconfidence. some authors suggest they may have evolved under natural selection, while others allege drivers such as the illusion of validity (rabin and schrag, 1999), the hindsight bias (fischhoff, 1982), and a confirmation bias (koriat, lichtenstein and fischhoff, 1980) for overconfidence, and affect (bracha and brown, 2012), self-attribution bias (lovallo and kahneman, 2003), as well as wishful thinking and overconfidence itself (barberis and thaler, 2003), for overoptimism. many models in finance use overconfidence to explain over and underreaction (daniel, hirshleifer and subrahmanyam, 1998), asset bubbles (scheinkman and xiong, 2003) and excessive trading volume (odean, 1998). it also helps explain the forward premium puzzle (burnside et al., 2011) and sensation seeking (grinblatt and keloharju, 2009). research on managerial overconfidence is a classic as well, causing excessive business entry (camerer and lovallo, 1999) and high rates of mandas (malmendier and tate, 2005). related judgmental biases people exhibit a self-attribution bias when they attribute to their ability events that validate their actions, while attribute contrary evidence to external noise or sabotage (bem, 1965). daniel et al. (1998) relates it to cognitive dissonance. a confirmation bias is observed when, once formed a strong hypothesis, people pay attention to news that support their views and ignore those that contradict them. griffin and tversky (1992) link it to the illusion of validity to induce overconfidence. finally, people exhibit an illusion of control when they behave as though chance events were subject to their control (langer, 1975). some anomalies attributed to be consequence of a biased self-attribution are feedback effects that may cause over and underreaction (daniel et al., 1998), and the spread of stories that is essential in the formation of speculative bubbles (shiller, 2003). recent literature includes libby and rennekamp (2012) and troye and supphellen (2012). empirical tests on the confirmation bias include duong, pescetto and santamaria (2014) on investors’ use of financial information. finally, recent research on the illusion of validity includes cowley, briley and farrell (2015). 4. choices, values and frames the second group of behavioral biases follows tversky and kahneman (1981, 1992), who consider two phases in the choice process: an initial of framing and a subsequent of evaluation. regarding framing, behaviorists have shown that people do not choose in a comprehensively inclusive context as the rational-agent model predicts. in particular, invariance –i.e., the fact that preferences are not affected by inconsequential variations in the description of outcomes (kahneman, 2003a)is violated, since alternative descriptions lead to different choices by only peón et al. / european journal of government and economics 6(1), 24-58 29 altering the salience of different features. framing effects include a variety of biases related to two classics in the literature: frame dependence and mental accounting (thaler, 1985). in regards to evaluation, we have prospect theory (pt) on one hand (kahneman and tversky, 1979), a descriptive theory of choice that explains how individuals evaluate the outcomes of risky prospects and choose in consequence. on the other, the empirical evidence that people make inconsistent choices in decisions over time led to the literature on intertemporal preferences, which started with problems of self-control (thaler and shefrin, 1981). framing, pt, intertemporal preferences, and the biases related to them are listed in table 2, and reviewed below. table 2. choices: framing and preferences. framing & preferences related concepts literature fr am in g frame dependence narrow framing equity premium puzzle barberis and huang (2007): narrow framing, equity premium puzzle loss aversion tversky and kahneman (1986) money illusion kahneman et al. (1986a) context dependence tversky and simonson (1993) repeated gambles kahneman and riepe (1998) hedonic editing mental accounting house money effect thaler (1999) self-control thaler and shefrin (1981) choice bracketing choice bracketing (read et al. 1999) pr os pe ct t he or y reference dependence anchoring-and anchoring not heuristic, related to reference points (rabin, 1998) adjustment conservatism conservatism: chan et al. (1996) loss aversion myopic loss aversion benartzi and thaler (1995) diminishing sensitivity risk seeking aversion to a sure loss shefrin (2006) favorite longshot bias tversky and kahneman (1992) in te rt em po ra l pr ef er en ce s preference reversals projection bias projection bias: loewenstein et al. (2003) self control precommitment self-control: loewenstein (1996) hyperbolic discounting present bias frederick et al. (2002) 4.1 frame dependence framing, defined as a decision-maker’s conception of the acts, outcomes and contingencies associated with a particular choice (tversky and kahneman, 1981), may produce predictable shifts of preference when the problem is framed differently ― a result known as frame dependence. a basic principle is the passive acceptance of the formulation given (rabin, 1998). framing influences loss aversion and diminishing sensitivity – see pt below. thus, a frame that highlights losses makes a choice less attractive, while if it makes them small relative to the scales involved it exploits diminishing sensitivity, making the choice attractive (tversky and peón et al. / european journal of government and economics 6(1), 24-58 30 kahneman, 1986). besides, related to frame dependence are the concepts of narrow framing, context effects, repeated gambles and hedonic editing. we see them next. related concepts narrow framing (kahneman and lovallo, 1993) is the tendency to analyze problems in a specific context without reflection of broader considerations (hirshleifer and teoh, 2003), such as evaluating risks in isolation, apart from others they already face (barberis and huang, 2009). context dependence (tversky and simonson, 1993) appears when an individual’s preferences among options depend on which other options are in the set (camerer and loewenstein, 2004), in a way that adding or subtracting options in a menu may affect the choice. the literature review of rooderkerk, van heerde and bijmolt (2011) observes a robust evidence of three types of context effects. kahneman and riepe (1998) show that most people do not distinguish between one-time choices and repeated gambles, setting the same cash-equivalent in both cases despite the fact that statistical aggregation reduce the relative risk of a series of gambles. benartzi and thaler (1999) relate the bias to myopic loss aversion. open debate recent articles include lab experiments (schlüter and vollan, 2015) as well as field research (hossain and list, 2012), both with positive results. however, cason and plott (2014) identify four aspects that contribute to the tension between standard preference theory and the theory of framing. some asset pricing models incorporate narrow framing, such as barberis and huang (2009) and de giorgi and legg (2012). in addition, it help explain market anomalies such as the equity premium puzzle (barberis and huang, 2007). finally, cornelissen and werner (2014) reviews framing in the management literature. evidence of choice effects includes empirical (hu and li, 2011) and experimental research (carlsson and martinsson, 2008). in addition, bordalo, gennaioli and shleifer (2012, 2013) analyze the effects of salience in context-dependent consumer choice and choice under risk. finally, regarding repeated gambles, liu and colman (2009) compare them with ambiguity aversion, and lejarraga and gonzalez (2011) observe that decision makers neglect descriptive information when they can learn from experience. 4.2 mental accounting closely related to framing, mental accounting refers to the implicit methods that individuals use to code and evaluate transactions, keeping track of and evaluating them like financial accounting in firms (thaler, 2008). statman (1999: 19) puts it briefly that people think “some money is retirement money, some is fun money, some is college education money, and some is vacation money”. thaler (1985, 1999) explains people engage in mental accounting activities in peón et al. / european journal of government and economics 6(1), 24-58 31 three instances: how outcomes are perceived and decisions are made, how activities are assigned to specific accounts, and the frequency with which accounts are evaluated. related concepts related to both frame dependence and mental accounting, hedonic editing refers to the evidence that people code combinations of events in a way it makes them happier (thaler, 1999). thaler and johnson (1990) provided a theory. choice bracketing refers to the grouping of individual choices into sets (read, loewenstein and rabin, 1999). narrow bracketing leads to myopic risk seeking (haisley, mostafa and loewenstein, 2008) and myopic loss aversion (hardin and looney, 2012). open debate positive empirical results of mental accounting include consumption, when it is temporally separated from purchase (shafir and thaler, 2006), and experimental evidence about inventory decisions (chen, kök and tong, 2013). models based on the mental accounting principle include the behavioral portfolio theory (shefrin and statman, 2000; das et al., 2010). pan and statman (2010) obtain empirical evidence of risk attitude changing across mental accounts of growth and value investments. finally, recent research includes sul, kim and choi (2013), who compare hedonic editing to subjective well-being, and koch and nafzinger (2016), who develop a model of endogenous bracketing where people set either narrow or broad bracketing to tackle self-control problems. 4.3 prospect theory prospect theory is the best known descriptive theory of decision-making under risk. for a closest insight in such an extensive literature we recommend barberis (2013). in short, according to pt, individuals evaluate the outcomes of risky prospects through a value function, where the carriers of value are changes in wealth compared to a reference point rather than final assets, and a probability weighting function, where probabilities are replaced by decision weights –in accordance with the empirical fact that people tend to put much weight on rare events. tversky and kahneman (1992) developed an extended version, cumulative prospect theory. it accounts for a fourfold pattern of risk attitudes confirmed by experimental evidence: people tend to exhibit risk aversion for gains but risk seeking for losses of high probability, and risk seeking for gains but risk aversion for losses of low probability. in addition, a value function that is steeper for losses than for gains implies loss aversion. thus, three features are essential: reference dependence (the carriers of value are gains and losses defined relative to a reference point), loss aversion (the value function is steeper in the negative than in the positive domain) peón et al. / european journal of government and economics 6(1), 24-58 32 and diminishing sensitivity (the marginal value of both gains and losses decreases with their size). this results in a value function that is kinked at the reference point, concave above and convex below, and represents investor’s loss aversion. moreover, diminishing sensitivity applies to the weighting function as well. these three features are analyzed separately in what follows. 4.3.1 reference dependence in pt, it is not final states what carries utility and matters for choice, but changes relative to a reference point. reference dependence is closely related to diminishing sensitivity and loss aversion, and induces two classic behavioral biases, namely, anchoring and conservatism. related concepts anchoring-and-adjustment is a key judgmental bias in risk perception. tversky and kahneman (1974: 1128) first described it as “people make estimates by starting from an initial value that is adjusted to yield the final answer”, an adjustment that is often insufficient. anchoring and reference dependence help to explain decision effects such as the classic status quo bias (tversky and kahneman, 1991). besides, conservatism, defined as the slow updating of models in face of new evidence (shleifer, 2000), explains why markets often respond gradually to new information, what might explain the profitability of momentum strategies (chan, jegadeesh and lakonishok, 1996). open debate though there is extensive evidence that perception is reference dependent, the debate continues in different instances. first, in terms of how reference points are set. common candidates include the buying price in stock markets (shefrin and statman, 1985) and the subject’s rational expectations given the economic environment (kõszegi and rabin, 2006). however, koop and johnson (2012) provide experimental evidence of multiple reference points in risky decision-making, and schmidt and zank (2012) provide a model of endogenous reference points. second, reference points may change over time, following gains and losses. arkes et al. (2008) observe an asymmetric adaptation that suggests hedonic editing: the magnitude of the adaptation is significantly greater following a gain than after a loss of equivalent size. baucells, weber and welfens (2011) find reference points are not recursive, in the sense that the new one is not a combination of the previous one and the new information. arkes et al. (2010) analyze how cultural differences influence reference point adaptation. the debate on anchoring is even better. a first wave of research, which assumed that the reference point was given in the formulation of the problem, is over (epley and gilovich (2010). epley and gilovich (2001, 2006) found anchoring effects for self-generated anchors, hence a second wave of research searched the psychological mechanisms that produce them. peón et al. / european journal of government and economics 6(1), 24-58 33 frederick, kahneman and mochon (2010) provide a theory. finally, a third wave makes predictions on the consequences of anchoring. furnham and boo (2011) provide a review. regarding conservatism, recent research relates return predictability in stock markets to gaap conservatism principle (ball, kothari and nikolaev, 2013). 4.3.2 loss aversion subjects assign more significance to losses than to gains with respect to the reference point. this asymmetry in the value function implies loss aversion: people suffer a loss more acutely than they enjoy a gain of the same magnitude. however, this represents a contradiction to rational choice, because the basic property of expected utility theory that two indifference curves never intersect no longer holds (knetsch, 1989). the influence of loss aversion in choices is observed in different contexts (see novemsky and kahneman, 2005), and it may explain empirical findings like the disposition effect (shefrin and statman, 1985) and why consumers and managers may take fewer risks (rabin, 2000). related concepts the combination of loss aversion and the investors’ common habit of evaluating their portfolios frequently is known as myopic loss aversion (benartzi and thaler, 1995). thaler et al. (1997) provided empirical evidence. langer and weber (2005) extend the concept to myopic prospect theory: when myopic loss aversion combines with diminishing sensitivity and probability weighting, the effect of myopia might increase the willingness to invest. open debate there is plenty of literature, including kahneman and tversky’s research, exposing the impact of loss aversion. moreover, cesarini et al.(2012) show loss aversion is moderately heritable. however, some limits were identified. three examples follow. first, exchange goods given up as intended, like money paid in purchases, do not exhibit loss aversion (novemsky and kahneman, 2005). second, there is mixed evidence of loss aversion on feelings, because judging feelings does not necessarily require comparison (mcgraw et al., 2010). third, polman (2012) shows loss aversion is lessened when we choose for others. finally, regarding myopic loss aversion, gneezy, kapteyn and potters (2003) provide experimental evidence, and fellner and sutter (2009) discuss debiasing techniques. 4.3.3 diminishing sensitivity marginal effects in perceived well-being are greater for changes close to the reference level than for changes further away (rabin 1998). this third essential feature of prospect theory peón et al. / european journal of government and economics 6(1), 24-58 34 applies to both the value and weighting functions. noting diminishing sensitivity is a pervasive pattern of human perception, kahneman and tversky (1979) conjectured the value function would be concave for gains and convex for losses –the latter implying risk seeking to avoid losses. regarding the weighting function, diminishing sensitivity entails that the impact of a given change in probability diminishes with its distance from two natural boundaries, certainty and impossibility, the endpoints of the scale (tversky and kahneman, 1992). consequently, risk-seeking choices are observed in two instances: the aversion to a sure loss, which stems from the shape of the value function, and the favorite-longshot bias –a miscalibration of probabilities often related to the weighting function. related concepts the aversion to a sure loss is a risk-seeking choice in the negative domain. most people are risk averse, but only when confronted with the expectation of a financial gain. instead, when facing the possibility of losing money, they behave as risk lovers, choosing to accept an actuarially unfair risk in an attempt to avoid a sure loss (shefrin, 2006). the favorite-longshot bias is commonly observed in betting markets. bettors put too much weight on rare events (longshot bets) and underestimate the probability of favorites, making the expected return on longshot bets systematically lower than on favorite bets (ottaviani and sorensen, 2007). open debate the favorite-longshot bias is one of the most studied biases. firstly documented in horse-race betting (griffith, 1949), recent studies include derivatives markets (hodges, tompkins and ziemba, 2008), prediction markets (page and clemen, 2013), and sports (lahvicka, 2014). the debate centers around its rationale, including misestimation of probabilities, informational asymmetries (shin, 1992), and limited arbitrage (ottaviani and sorensen, 2007). regarding the aversion to a sure loss, researchers are more focused on its interpretation. adam and kroll (2012) suggest decision makers perceive lotteries as dynamic processes where emotions may lead to attraction to chance, while schwager and rothermund (2013) provide evidence on the effects of framing and attention bias. 4.4 preference reversals intertemporal preferences are rational if they are time consistent. however, empirical evidence shows people do exhibit reversals, have problems to commit with decisions they took in the past, and exhibit present-biased preferences. we see these concepts together under the epigraph of preference reversals, which include problems of self-control, and a present bias in intertemporal decision-making. peón et al. / european journal of government and economics 6(1), 24-58 35 related concepts standard models compare preferences over time with exponential discounting, implying time consistency and 100% short-term patience. however, there is evidence that people exhibit a present bias or hyperbolic discounting, as preferences typically reverse with changes in delay (kirby and herrnstein, 1995). related to such reversals is a projection bias: people exaggerate the degree to which their future tastes will be similar to their current ones, what makes them save less than originally planned as time passes (loewenstein, o’donoghue and rabin, 2003). self-control (and precommitment) relates to that, as being aware in advance that our preferences may change, we sometimes make certain decisions to restrict our own future flexibility (loewenstein, 1996). open debate a classic review by frederick, loewenstein and o’donoghe (2002) observes cross-study differences in discount rates, against the assumption of a single rate under exponential discounting. however, the debate continues today. andersen et al. (2008) showed that a joint estimation of risk and time preferences is required, so the discounting anomalies previously observed had to be re-tested. andersen et al. (2014) find no evidence favorable to hyperbolic discounting. recent advances include a model of preference reversals (tsetsos, chater and usher 2012), and the work of stevens (2016), who suggests people do not discount, rather they compare within attributes (amounts and delays). recent research includes zeisberger, vrecko and langer (2015) about the projection bias, and on self-control an experimental research by burger, charness and lynham (2011) and an interpretation of the cash-credit co-holding puzzle (gatherwood and weber, 2014). 5. social factors the last category compiles the items that refer to the impact of cultural and social factors on individual’s behavior. this is the least developed and structured body of literature in the behavioral economics and finance, but according to hirshleifer (2015: 133): “the time has come to move beyond behavioral finance to social finance, which studies the structure of social interactions, how financial ideas spread and evolve, and how social processes affect financial outcomes.”. the social factors are shown in table 3 and reviewed below. peón et al. / european journal of government and economics 6(1), 24-58 36 table 3. social factors. social factors related concepts literature global culture cultural differences guiso et al. (2006); statman and weng (2010) social contagion obediency to authority herd behavior social contagion: asch (1952). herding: shiller (2000b) communal reinforcement & groupthink (collective) confirmation bias shiller (1984); janis (1972) shefrin and cervellati (2011) status, social comparison self esteem, pride, prejudice rabin (1998) cooperation, altruism fairness and justice kahneman et al. (1986a,b) greed and fear familiarity fear of the unknown and familiarity bias (cao et al., 2011) status quo bias fear of change and status quo bias (samuelson & zeckhauser, 1988) informational cascades availability cascades asset bubbles shiller (2002b): cascades and bubbles herding bikhchandani et al (1998): cascades and herding 5.1 global culture culture is the values that ethnic, religious, and social groups transmit across generations (statman and weng, 2010). shiller (2000a) notices the emergence of a global culture in a convergence of fashions across countries separated by physical and language barriers, and suggests these cultural factors help explain the dot-com bubble. stulz and williamson (2003) claim culture may affect finance through the country values, institutions, and how resources are allocated. related social factors though a global culture might be emerging, cultural differences are also ubiquitous. the best studied case is perhaps the differences between east asians and americans. thus, east asians exhibit a broader perceptual and conceptual view of the world and live in more complex social networks (nisbett and masuda, 2003), and they exhibit different patterns in terms of overconfidence and the disposition effect (chen et al., 2007). open debate culture has had a significant influence on social psychology (e.g. miller, 1984), but economists were reluctant to use it as an explanatory factor because of the vague and ubiquitous ways it can enter the economic discourse, making it difficult to design testable hypotheses (guiso, sapienza and zingales, 2006). recent techniques and data made it possible to identify peón et al. / european journal of government and economics 6(1), 24-58 37 systematic differences in people’s beliefs, and relate them to their cultural legacy (e.g. levinson and peng, 2007). some authors have analyzed how it affects expectations and preferences. these include henrich et al. (2001) on variations across tribes in the ultimatum and dictator games, and hoff and priyanka (2004) who show the effects of social inequality linger: beliefs that are the legacy of extreme inequality for generations determine individual’s expectations that reproduce the inequality. studies on cultural differences in economic and financial variables include statman and weng (2010), who find different borrowing and investing patterns of immigrants long after they settled in their new countries, and beugelsdijk and frijns (2010), who show that the degree of cultural distance between two countries affects foreign asset allocations. recent literature analyzes the effects on corporate structure (bloom, sadun and van reenen, 2015), innovation rates (taylor and wilson, 2012), and corporate mandas (ahern, daminelli and fracassi, 2015). 5.2 social contagion research on cultural differences focuses on inherited, slow-moving components of societies, while social interaction focuses on peer group effects that can be viewed as the fast-moving component of culture (guiso et al., 2006). the antecedents in the study of social contagion are the experiment of sherif (1937) on the autokinetic effect, and the classic experiments of asch (1952). related social factors a classic in the literature is obedience to authority. the experiments of milgram (1963) showed few people have the initiative to resist authority, to the point of performing acts that violate their deepest moral beliefs. years before, festinger (1957) analyzed the effects of forced compliance, showing that a person forced to do something contrary to her opinion may change her view in order to avoid cognitive dissonance. communal reinforcement is a type of social dynamics related to social learning and the psychology of individual suggestibility (katona, 1901). shiller (1984) gives the example of investors who follow gurus, read magazines, discuss investments with other investors... and through this process, market psychology influences markets. groupthink is the tendency of cohesive groups to reach consensus without offering, seeking or considering alternative hypotheses (lunenburg, 2010). janis (1972) identifies some symptoms, like an excessive risk-taking, and members imposing themselves a self-censorship to avoid appearing as a dissenter. shefrin and cervellati (2011) interpret it as a form of collective confirmation bias. peón et al. / european journal of government and economics 6(1), 24-58 38 open debate likewise other social factors, there is an increasing interest in the recent decades for the study of social contagion. nonetheless, the literature review by manski (2000) suggests that the neoclassical view, where non-market interactions are not of interest, ended by the 1970s with the adoption of non-cooperative dynamic game theory. recent contributions include empirical research by rapp et al. (2013), and experimental studies on viral marketing (aral and walker, 2014). in regards to obedience to authority and social contagion, there are two opposite views. the classic one highlights the negative impact they have in financial markets, like herding and asset bubbles (e.g. shiller, 2000b). contrariwise, ent and baumeister (2014) observe that obedience to legitimate authority may be positive, encouraging individuals to set aside their selfish desires for the good of the group. recent research includes mayo-wilson, zollman and danks (2012) on individual and group rationality, and the model by nofsinger (2012) on asset bubbles fueled by groupthink. 5.3 status, envy and social comparison a field of social psychology relevant to economics is the self-perception compared to others, and the feelings of jealousy, self-esteem, pride and prejudice such comparison provokes. we denote this category status, envy and social comparisons following rabin (1998). not all feelings stemming from social comparisons are negative, as cooperation (argyle, 1991) and reciprocal altruism (trivers, 1971) may be included here. open debate early literature already suggested that social comparison occurs in many forms of human interaction, including social status (ball and eckel, 1998), reciprocity and altruism (gilbert, price and allan, 1995), and consumer dissatisfaction, when they compare themselves with the idealized advertising images (richins, 1991). more recently, researchers have focused on testing, whether in the lab or in the field, motivations and effects of social comparison and cooperative behavior. these include experimental tests of the effects of social status (ball et al., 2001), and the motivations for pro-social behavior (carpenter and myers, 2010). 5.4 fairness and justice fairness and justice were recurrently absent from standard economic theory, a striking contrast when compared to other social sciences (kahneman, knetsch and thaler, 1986b). we first find fairness in the literature of efficiency wages, as well as in the literature of customer markets (okun, 1981). three reasons related to fairness why people are willing to spend money are in order to punish others who have harmed them, to reward those who have helped, or to make peón et al. / european journal of government and economics 6(1), 24-58 39 outcomes fairer (camerer and loewenstein, 2004). fairness and justice are related to behavioral effects like money illusion (kahneman, knetsch and thaler, 1986a) and helps to determine people’s reference prices (thaler, 1985). open debate the classic approach to trace evidence of decisions based on fairness and justice analysis is using dictator and ultimatum games (güth, schmittberger and schwarze, 1982). camerer and thaler (1995) provide a review on ultimatum games and list (2007) on dictator games. researchers focus on topics like moral values (sen, 1995), equity and competition (bolton and ockenfels, 2000), perceptions of fairness (nguyen and klaus, 2013), and inequality and preferences for redistribution (durante, putterman and van der weele, 2014). 5.5 greed and fear being emotional factors, greed and fear might indeed be related to the affect heuristic in section 3. however, we opt to classify them as social factors because these biases tend to appear when individuals interact with each other. two related biases are fear of the unknown, an explanation for the familiarity heuristic (cao et al., 2011), and fear of change, a possible explanation for the status quo bias (samuelson and zeckhauser, 1988). open debate the effects of greed and fear are particularly pervasive in financial markets, where they are alleged to play a key role in concepts like market sentiment, bubbles and crashes, and others. indeed, shefrin (2000) identifies human emotions as determinants of risk tolerance and portfolio choice. pan and statman (2010) show risk tolerance varies with test conditions and the emotions associated to them. lo, repin and steenbarger (2005) offer experimental evidence of a negative correlation between successful trading and emotional reactivity. despite these results, the effects of emotions over market efficiency are far from being widely accepted. for instance, shleifer (2004) asserts that the unethical behavior blamed to stem from greed is often a consequence of market competition. recent research on greed and fear includes lee and andrade (2011), who show social projection explains why fear leads to early sell-off in a stock market simulation, and cohn et al.(2015), who provide experimental evidence that fear may play an important role in countercyclical risk aversion. peón et al. / european journal of government and economics 6(1), 24-58 40 5.6 informational cascades we learn by observing what others do, and then we imitate them. imitation would be an evolutionary adaptation for survival, allowing individuals to take advantage of the hard-won information of others. significant market events only occur if large groups of people think the same, and news media might act as precipitators of attention cascades and the spread of ideas. some phenomena such as herding, fads, asset bubbles and crashes might be consequence of informational cascades (bikhchandani, hirshleifer and welch, 1998). related social factors availability cascades are self-reinforcing processes of collective belief formation that have a combination of informational and reputational motives as driving factors (kuran and sunstein, 1999). by the availability heuristic, people judge the importance of a theme according to their ability to remember examples of it. then, as a chain reaction result, the more people talk about an issue the more relevant it seems due to its rising availability in public discourse, leading to a self-reinforcing cycle (hirshleifer, 2008). open debate a line of research today in process focuses on theoretical modeling of the disruptive or corrective nature of informational cascades. for instance, wu (2015) suggests that the probability of wrong cascades decreases if laymen are among a group of experts, while the model of rubin (2014) suggests that cascades inducing larger shocks are more likely to happen in regimes with centralized coercive power. 6. conclusions and policy implications the impact of behavioral biases in financial and consumer markets has many implications for the way in which these markets work. however, there is no consensus on how to address this issue from a public policy intervention perspective. there are three basic approaches –namely, debiasing techniques, liberal paternalism, and active policy-makingand all of them have supporters and detractors. the logic behind debiasing is, if people make biased decisions (positive economics) from what is standard rationality (normative economics), perhaps we may help them to choose better. croskerry, singhal and mamede (2013) provide a recent discussion on several approaches towards debiasing. the idea itself makes no sense for some authors, either under the interpretation of the ecological rationality of the heuristics (gigerenzer and gaissmaier, 2011), or the contrary: when the limitations of the normative model have become so obvious, it is nonsense to insist upon changing humanity to conform to it (frankfurter, mcgoun and allen, peón et al. / european journal of government and economics 6(1), 24-58 41 2004). others advocate for improving financial literacy (e.g. altman, 2012), while for other authors there is evidence that learning and expertise may do little to eliminate biases (rabin, 1998) or even might exacerbate errors (griffin and tversky, 1992). in any case, debiasing would require intervention, since there are many reasons to doubt individuals can debias themselves (kahneman, 2003b). two approaches are trying to increase motivation to perform well, and setting strategies that are closer to normative standards –known as prescriptive decision making. liberal paternalism (thaler and sunstein, 2003) is a smoother approach for prescriptive debiasing. it criticizes the assumption that people always make choices that are in their best interest, and explores different methods to help consumers and investors improving their decision making and enhance their well-being (see ratner et al., 2008). while being paternalistic in the sense that it seeks to help people make better choices, it is liberal in the sense that it also respects freedom of choice. for such purpose, it exploits the passive acceptance of the formulation given (e.g. the status quo bias) or it uses some behavioral traits by the decision maker to reduce other biases –for instance, mental accounting and framing to mitigate self-control problems (thaler and shefrin, 1981). although behavioral biases may affect consumer decisions or lead to anticompetitive behavior by firms, some authors discredit paternalism and oppose public intervention. cooper and kovacic (2012) provide a model that depicts how greater state intervention, especially if oriented to correct firm biases, is likely to lead regulators to adopt policies closer to the preferences of political overseers, either intentionally, or accidentally (due to bounded rationality). the same interpretation would follow in financial markets. behaviorists such as daniel, hirshleifer and teoh (2002) observe that the same psychological biases that affect investors would affect regulators. rather than correcting market pricing errors, for which they do not have a competitive advantage, they advocate for regulators establishing ex ante rules to improve efficiency, such as default-option-setting regulations. to conclude, some examples of recent literature of behavioral biases and policy implications follow in order. briley, shrum and wyer (2013) analyze representativeness and its effect on public policy. some theoretical models interpret excessive optimism as a key factor behind credit booms (e.g. peón, antelo and calvo, 2015), and observe a similar bias in governments’ official forecasts (frankel and schreger, 2013). givoni et al. (2013) offer a heuristic framework to improve the effectiveness of policy interventions. the empirical analysis of hossain and list (2012) suggests some alternatives to increase productivity in factories through simple framing manipulations, while bao et al. (2015) draw policy lessons from mental accounting: authorities often overestimate the traffic of high tolled roads because travelers with low out-of-pocket travel budget perceive a much higher cost. finally, research on social factors includes the effects of culture on innovation rates (taylor and wilson, 2012) and of social contagion: pacheco (2012) models how public opinion influences policy diffusion. besides, altruism and volunteering may be negatively affected by public policies: ariely, bracha and meier (2009) show that extrinsic incentives can reduce charitable donations and volunteering as they dilute the signaling value of peón et al. / european journal of government and economics 6(1), 24-58 42 pro-social behavior. finally, shleifer (2004) observes ethics and efficiency go together when ethical norms promote cooperative behavior, helping for the successful functioning of social institutions. references ackert, l.f., b.k. church, j. tompkins & p. zhang (2005). what's in a name? an experimental examination of investment behavior. review of finance 9, 281304. https://doi.org/10.1007/s10679-005-7594-2 adam, m.t.p. & e.b. kroll (2012). physiological evidence of attraction to chance. journal of neuroscience, psychology and economics 5(3), 152-165. https://doi.org/10.1037/a0029513 ahern, k.r., d. daminelli & c. fracassi (2015). lost in translation? the effect of cultural values on mergers around the world. journal of financial economics 117, 165189. https://doi.org/10.1016/j.jfineco.2012.08.006 akerlof, g.a. & r.j. shiller (2009). how animal spirits destabilize economies. mckinsey quarterly 3, 127-135. alary, d., c. gollier & n. treich (2013). the effect of ambiguity aversion on insurance and selfprotection. the economic journal 123, 1188-1202. https://doi.org/10.1111/ecoj.12035 alós-ferrer, c. & s. hügelschäfer (2012). faith in intuition and behavioral biases. journal of economic behavior and organization 84, 182192. https://doi.org/10.1016/j.jebo.2012.08.004 altman, m. (2012). implications of behavioural economics for financial literacy and public policy. the journal of socio-economics 41, 677-690. https://doi.org/10.1016/j.socec.2012.06.002 andersen, s., g.w. harrison, m.i. lau & e.i. rutström (2008). eliciting risk and time preferences. econometrica 76(3), 583–618. https://doi.org/10.1111/j.14680262.2008.00848.x andersen, s., g.w. harrison, m.i. lau & e.i. rutström (2014). discounting behavior: a reconsideration. european economic review 71, 1533. https://doi.org/10.1016/j.euroecorev.2014.06.009 aral, s. & d. walker (2014). tie strength, embeddedness, and social influence: a large-scale networked experiment. management science 60(6), 135270. https://doi.org/10.1287/mnsc.2014.1936 argyle, m. (1991). cooperation: the basis of sociability. routledge. ariely, d., a. bracha & s. meier (2009). doing good or doing well? image motivation and monetary incentives in behaving prosocially. the american economic review 99(1), 544555. https://doi.org/10.1257/aer.99.1.544 arkes, h.r., d. hirshleifer, d. jiang & s.s. lim (2008). reference point adaptation: tests in the domain of security trading. organizational behavior and human decision processes 105, 67–81. https://doi.org/10.1016/j.obhdp.2007.04.005 https://doi.org/10.1007/s10679-005-7594-2 https://doi.org/10.1037/a0029513 https://doi.org/10.1016/j.jfineco.2012.08.006 https://doi.org/10.1111/ecoj.12035 https://doi.org/10.1016/j.jebo.2012.08.004 https://doi.org/10.1016/j.socec.2012.06.002 https://doi.org/10.1111/j.1468-0262.2008.00848.x https://doi.org/10.1111/j.1468-0262.2008.00848.x https://doi.org/10.1016/j.euroecorev.2014.06.009 https://doi.org/10.1287/mnsc.2014.1936 https://doi.org/10.1257/aer.99.1.544 https://doi.org/10.1016/j.obhdp.2007.04.005 peón et al. / european journal of government and economics 6(1), 24-58 43 arkes, h.r., d. hirshleifer, d. jiang & s.s. lim (2010). a cross-cultural study of reference point adaptation: evidence from china, korea and the us. organizational behavior and human decision processes 112(2), 99-111. https://doi.org/10.1016/j.obhdp.2010.02.002 asch, s.e. (1952). social psychology. prentice hall. https://doi.org/10.1037/10025-000 ball, r., s.p. kothari & v.v. nikolaev (2013). econometrics of the basu asymmetric timeliness coefficient and accounting conservatism. journal of accounting research 51(5), 107197. https://doi.org/10.1111/1475-679x.12026 ball, s. & c. eckel (1998). the economic value of status. journal of socio-economics 27(4), 495-514. https://doi.org/10.1016/s1053-5357(98)80004-8 ball, s., c. eckel, p.j. grossman & w. zane (2001). status in markets. the quarterly journal of economics 116(1), 161-188. https://doi.org/10.1162/003355301556374 bao, y., z. gao, m. xu, h. sun, & h. yang (2015). travel mental budgeting under road toll: an investigation based on user equilibrium. transportation research part a 73, 117. https://doi.org/10.1016/j.tra.2014.12.010 barber, b.m. & t. odean (2008). all that glitters: the effect of attention and news on the buying behavior of individual and institutional investors. review of financial studies 21(2), 785818. https://doi.org/10.1093/rfs/hhm079 barberis, n.c. (2013). thirty years of prospect theory in economics: a review and assessment. the journal of economic perspectives 27(1), 173-195. https://doi.org/10.1257/jep.27.1.173 barberis, n.c. & m. huang (2007). the loss aversion/narrow framing approach to the equity premium puzzle. in r. mehra (ed.) handbook of investments: equity premium, northholland, amsterdam, pp. 199–229. barberis, n.c. & m. huang (2009). preferences with frames: a new utility specification that allows for the framing of risks. journal of economics dynamics and control 33(8), 155576. https://doi.org/10.1016/j.jedc.2009.01.009 barberis, n.c. & r.h. thaler (2003). a survey of behavioral finance. in g.m. constantinides, m. harris and r.m. stulz (eds.), handbook of the economics of finance 1(1), 1053-1128. bar-hillel, m. (1973). on the subjective probability of compound events. organizational behavior and human performance 9, 396-406. https://doi.org/10.1016/0030-5073(73)90061-5 battalio, r.h., a. lerman, j. livnat & r.r. mendenhall (2012). who, if anyone, reacts to accrual information? journal of accounting and economics 53, 205224. https://doi.org/10.1016/j.jacceco.2011.06.007 baucells, m., m. weber & f. welfens (2011). reference point formation and updating: theory and experiments. journal management science 57(3), 506519. https://doi.org/10.1287/mnsc.1100.1286 bem, d.j. (1965). an experimental analysis of self-persuasion. journal of experimental social psychology 1, 199-218. https://doi.org/10.1016/0022-1031(65)90026-0 benartzi, s. & r.h. thaler (1995). myopic loss aversion and the equity premium puzzle. the quarterly journal of economics 110(1), 73-92. https://doi.org/10.2307/2118511 benartzi, s. & r.h. thaler (1999). risk aversion or myopia? choices in repeated gambles and https://doi.org/10.1016/j.obhdp.2010.02.002 https://doi.org/10.1037/10025-000 https://doi.org/10.1111/1475-679x.12026 https://doi.org/10.1016/s1053-5357(98)80004-8 https://doi.org/10.1162/003355301556374 https://doi.org/10.1016/j.tra.2014.12.010 https://doi.org/10.1093/rfs/hhm079 https://doi.org/10.1257/jep.27.1.173 https://doi.org/10.1016/j.jedc.2009.01.009 https://doi.org/10.1016/0030-5073(73)90061-5 https://doi.org/10.1016/j.jacceco.2011.06.007 https://doi.org/10.1287/mnsc.1100.1286 https://doi.org/10.1016/0022-1031(65)90026-0 https://doi.org/10.2307/2118511 peón et al. / european journal of government and economics 6(1), 24-58 44 retirement investments. management science 45(3), 364381. https://doi.org/10.1287/mnsc.45.3.364 beugelsdijk, s. & b. frijns (2010). a cultural explanation of the foreign bias in international asset allocation. journal of banking and finance 34, 212131. https://doi.org/10.1016/j.jbankfin.2010.01.020 bikhchandani, s., d. hirshleifer & i. welch (1998). learning from the behavior of others: conformity, fads, and informational cascades. the journal of economic perspectives 12(3), 151-170. https://doi.org/10.1257/jep.12.3.151 bloom, n., r. sadun & j. van reenen (2015). the organization of firms across countries. quarterly journal of economics 127(4), 1663-1705. https://doi.org/10.1093/qje/qje029 bolton, g.e. & a. ockenfels (2000). erc: a theory of equity, reciprocity, and competition. american economic review 90(1), 166-193. https://doi.org/10.1257/aer.90.1.166 bordalo, p., n. gennaioli & a. shleifer (2012). salience theory of choice under risk. the quarterly journal of economics 127(3), 1243-85. https://doi.org/10.1093/qje/qjs018 bordalo, p., n. gennaioli & a. shleifer (2013). salience and consumer choice. journal of political economy 121(5), 803-843. https://doi.org/10.1086/673885 boyle, p., l. garlappi, r. uppal & t. wang (2012). keynes meets markowitz: the trade-off between familiarity and diversification. management science 58(2), 253272. https://doi.org/10.1287/mnsc.1110.1349 bracha, a. & d.j. brown (2012). affective decision making: a theory of optimism bias. games and economic behavior 75, 67-80. https://doi.org/10.1016/j.geb.2011.11.004 briley, d.a., l.j. shrum & r.s. wyer (2013). factors affecting judgments of prevalence and representation: implications for public policy and marketing. journal of public policy and marketing 32, 112-118. https://doi.org/10.1509/jppm.12.045 burger, n., g. charness & j. lynham (2011). field and online experiments on self-control. journal of economic behavior and organization 77, 393404. https://doi.org/10.1016/j.jebo.2010.11.010 burnside, c., b. han, d. hirshleifer & t.y. wang (2011). investor overconfidence and the forward premium puzzle. review of economic studies 78, 523558. https://doi.org/10.1093/restud/rdq013 calvillo, d.p. (2014). individual differences in susceptibility to misinformation effects and hindsight bias. journal of general psychology 141(4), 393407. https://doi.org/10.1080/00221309.2014.954917 camerer, c.f. & d. lovallo (1999). overconfidence and excess entry: an experimental approach. the american economic review 89(1), 306318. https://doi.org/10.1257/aer.89.1.306 camerer, c.f. & g. loewenstein (2004). behavioral economics: past, present, future. in c.f. camerer, g. loewenstein and m. rabin (eds.), advances in behavioral economics, princeton university press, 3-51. camerer, c.f. & r.h. thaler (1995). anomalies: ultimatums, dictators and manners. journal of https://doi.org/10.1287/mnsc.45.3.364 https://doi.org/10.1016/j.jbankfin.2010.01.020 https://doi.org/10.1257/jep.12.3.151 https://doi.org/10.1093/qje/qje029 https://doi.org/10.1257/aer.90.1.166 https://doi.org/10.1093/qje/qjs018 https://doi.org/10.1086/673885 https://doi.org/10.1287/mnsc.1110.1349 https://doi.org/10.1016/j.geb.2011.11.004 https://doi.org/10.1509/jppm.12.045 https://doi.org/10.1016/j.jebo.2010.11.010 https://doi.org/10.1093/restud/rdq013 https://doi.org/10.1080/00221309.2014.954917 https://doi.org/10.1257/aer.89.1.306 peón et al. / european journal of government and economics 6(1), 24-58 45 economic perspectives 9(2), 209-219. https://doi.org/10.1257/jep.9.2.209 cao, h.h., b. han, d. hirshleifer & h.h. zhang (2011). fear of the unknown: familiarity and economic decisions. review of finance 15(19), 173-206. https://doi.org/10.1093/rof/rfp023 carlsson, f. & p. martinsson (2008). how much is too much? an investigation of the effect of the number of choice sets, context dependence and the choice of bid vectors in choice experiments. environmental and resource economics 40, 165176. https://doi.org/10.1007/s10640-007-9146-z carpenter, j. & c.k. myers (2010). why volunteer? evidence on the role of altruism, image, and incentives. journal of public economics 94, 911920. https://doi.org/10.1016/j.jpubeco.2010.07.007 cason, t.n. & c.r. plott (2014). misconceptions and game form recognition: challenges to theories of revealed preference and framing. journal of political economy 122(6), 123570. https://doi.org/10.1086/677254 cesarini, d., m. johannesson, p.k.e. magnusson & b. wallace (2012). the behavioral genetics of behavioral anomalies. management science 58(1), 2134. https://doi.org/10.1287/mnsc.1110.1329 chan, l.k.c., n. jegadeesh and j. lakonishok (1996). momentum strategies. the journal of finance 51(5), 1681-1713. https://doi.org/10.1111/j.1540-6261.1996.tb05222.x chelley-steeley, p.l., b.d. kluger & j.m. steeley (2015). earnings and hindsight bias: an experimental study. economic letters 134, 130132. https://doi.org/10.1016/j.econlet.2015.07.005 chen, g., k.a. kim, j.r. nofsinger & o.m. rui (2007). trading performance, disposition effect, overconfidence, representativeness bias, and experience of emerging market investors. journal of behavioral decision making 20(4), 425-451. https://doi.org/10.1002/bdm.561 chen, l., a.g. kök & j.d. tong (2013). the effect of payment schemes on inventory decisions: the role of mental accounting. management science 59(2), 436451. https://doi.org/10.1287/mnsc.1120.1638 chew, s.h., r.p. ebstein & s. zhong (2012). ambiguity aversion and familiarity bias: evidence from behavioral and gene association studies. journal of risk and uncertainty 44, 118. https://doi.org/10.1007/s11166-011-9134-0 chou, h.g. & n. edge (2012). "they are happier and having better lives than i am'': the impact of using facebook on perceptions of others' lives. cyberpsychology, behavior, and social networking 15(2), 117-121. https://doi.org/10.1089/cyber.2011.0324 cohn, a., engelmann, j., fehr, e., & maréchal, m.a. (2015). evidence for countercyclical risk aversion: an experiment with financial professionals. american economic review 105(2), 860-885. https://doi.org/10.1257/aer.20131314 cokely, e.t. & c.m. kelley (2009). cognitive abilities and superior decision making under risk: a protocol analysis and process model evaluation. judgmental decision making 4, 20–33. cooper, j.c. & w.e. kovacic (2012). behavioral economics: implications for regulatory behavior. journal of regulatory economics 41, 41–58. https://doi.org/10.1007/s11149-011https://doi.org/10.1257/jep.9.2.209 https://doi.org/10.1093/rof/rfp023 https://doi.org/10.1007/s10640-007-9146-z https://doi.org/10.1016/j.jpubeco.2010.07.007 https://doi.org/10.1086/677254 https://doi.org/10.1287/mnsc.1110.1329 https://doi.org/10.1111/j.1540-6261.1996.tb05222.x https://doi.org/10.1016/j.econlet.2015.07.005 https://doi.org/10.1002/bdm.561 https://doi.org/10.1287/mnsc.1120.1638 https://doi.org/10.1007/s11166-011-9134-0 https://doi.org/10.1089/cyber.2011.0324 https://doi.org/10.1257/aer.20131314 https://doi.org/10.1007/s11149-011-9180-1 peón et al. / european journal of government and economics 6(1), 24-58 46 9180-1 cornelissen, j.p. & m.d. werner (2014). putting framing in perspective: a review of framing and frame analysis across the management and organizational literature. the academy of management annals 8(1), 181-235. https://doi.org/10.1080/19416520.2014.875669 cosmides, l. & j. tooby (1990). is the mind a frequentist? second annual meeting of the human behavior and evolution society, la, ca. cited in g. gigerenzer (1991), how to make cognitive illusions disappear: beyond 'heuristics and biases', in w. stroebe and m. hewstone (eds.), european review of social psychology, wiley. cowley, e., d.a. briley & c. farrell (2015). how do gamblers maintain an illusion of control? journal of business research 68, 2181-88. https://doi.org/10.1016/j.jbusres.2015.03.018 cready, w., kumas, a., & subasi, m. (2014). are trade size-based inferences about traders reliable? evidence from institutional earnings-related trading. journal of accounting research 52(4), 877-909. https://doi.org/10.1111/1475-679x.12056 croskerry, p., g. singhal & s. mamede (2013). cognitive debiasing 1: origins of bias and theory of debiasing. bmj quality and safety 22(2), 58-64. https://doi.org/10.1136/bmjqs2012-001712 daniel, k., hirshleifer, d., & teoh, s.h. (2002). investor psychology in capital markets: evidence and policy implications. journal of monetary economics 49, 139209. https://doi.org/10.1016/s0304-3932(01)00091-5 daniel, k.d., d. hirshleifer & a. subrahmanyam (1998). investor psychology and security market underand overreactions. the journal of finance 53(6), 183985. https://doi.org/10.1111/0022-1082.00077 das, s., h. markowitz, j. scheid & m. statman (2010). portfolio optimization with mental accounts. journal of financial and quantitative analysis 45(2), 311334. https://doi.org/10.1017/s0022109010000141 de giorgi, e.g. & s. legg (2012). dynamic portfolio choice and asset pricing with narrow framing and probability weighting. journal of economic dynamics and control 36, 951972. https://doi.org/10.1016/j.jedc.2012.01.010 dellavigna, s. (2009). psychology and economics: evidence from the field. journal of economic literature 47(2), 315-372. https://doi.org/10.1257/jel.47.2.315 duong, c., g. pescetto & d. santamaria (2014). how value-glamour investors use financial information: uk evidence of investors' confirmation bias. the european journal of finance 20(6), 524-549. https://doi.org/10.1080/1351847x.2012.722117 durante, r., l. putterman & j. van der weele (2014). preferences for redistribution and perception of fairness: an experimental study. journal of the european economic association 12(4), 1059-1086. https://doi.org/10.1111/jeea.12082 einhorn, h.j. & r.m. hogarth (1978). confidence in judgment: persistence of the illusion of validity. psychological review 85(5), 395-416. https://doi.org/10.1037/0033-295x.85.5.395 einhorn, h.j. & r.m. hogarth (1986). decision making under ambiguity. journal of business 59(4), part 2: the behavioral foundations of economic theory, s225-s250. https://doi.org/10.1007/s11149-011-9180-1 https://doi.org/10.1080/19416520.2014.875669 https://doi.org/10.1016/j.jbusres.2015.03.018 https://doi.org/10.1111/1475-679x.12056 https://doi.org/10.1136/bmjqs-2012-001712 https://doi.org/10.1136/bmjqs-2012-001712 https://doi.org/10.1016/s0304-3932(01)00091-5 https://doi.org/10.1111/0022-1082.00077 https://doi.org/10.1017/s0022109010000141 https://doi.org/10.1016/j.jedc.2012.01.010 https://doi.org/10.1257/jel.47.2.315 https://doi.org/10.1080/1351847x.2012.722117 https://doi.org/10.1111/jeea.12082 https://doi.org/10.1037/0033-295x.85.5.395 peón et al. / european journal of government and economics 6(1), 24-58 47 ent, m.r. & r.f. baumeister (2014). obedience, self-control, and the voice of culture. journal of social issues 70(3), 574-586. https://doi.org/10.1111/josi.12079 epley, n. & t. gilovich (2001). putting adjustment back in the anchoring and adjustment heuristic: differential processing of self-generated and experimenter-provided anchors. psychological science 12, 391-396. https://doi.org/10.1111/1467-9280.00372 epley, n. & t. gilovich (2006). the anchoring-and-adjustment heuristic: why the adjustments are insufficient. psychological science 17(4), 311-318. https://doi.org/10.1111/j.14679280.2006.01704.x epley, n. & t. gilovich (2010). anchoring unbound. journal of consumer psychology 20(1), 2024. https://doi.org/10.1016/j.jcps.2009.12.005 erceg, n. & z. galic (2014). overconfidence bias and conjunction fallacy in predicting outcomes of football matches. journal of economic psychology 42, 5262. https://doi.org/10.1016/j.joep.2013.12.003 erev, i., t.s. wallsten & d.v. budescu (1994). simultaneous overand underconfidence: the role of error in judgment processes. psychological review 10(3), 519527. https://doi.org/10.1037/0033-295x.101.3.519 etner, j., m. jeleva & j.-m. tallon (2012). decision theory under ambiguity. journal of economic surveys 26(2), 234-270. https://doi.org/10.1111/j.1467-6419.2010.00641.x fellner, g. & m. sutter (2009). causes, consequences, and cures of myopic loss aversion: an experimental investigation. the economic journal 119, 900916. https://doi.org/10.1111/j.1468-0297.2009.02251.x fellner, w. (1961). distortion of subjective probabilities as a reaction to uncertainty. quarterly journal of economics 75, 670-690. https://doi.org/10.2307/1884325 festinger, l. (1957). a theory of cognitive dissonance. stanford university press. finucane, m.l., a. alhakami, p. slovic & s.m. johnson (2000). the affect heuristic in judgments of risks and benefits. journal of behavioral decision making 13(1), 117. https://doi.org/10.1002/(sici)1099-0771(200001/03)13:1<1::aid-bdm333>3.0.co;2-s fischhoff, b. (1982). for those condemned to study the past: heuristics and biases in hindsight. in d. kahneman, p. slovic and a. tversky, judgment under uncertainty: heuristics and biases. cambridge university press. https://doi.org/10.1017/cbo9780511809477.024 frankel, j. & j. schreger (2013). over-optimistic official forecasts and fiscal rules in the eurozone. review of world economics 149, 247-272. https://doi.org/10.1007/s10290-0130150-9 frankfurter, g.m., e.g. mcgoun & d.e. allen (2004). the prescriptive turn in behavioral finance. journal of socio-economics 33, 449468. https://doi.org/10.1016/j.socec.2004.04.006 frederick, s., d. kahneman & d. mochon (2010). elaborating a simpler theory of anchoring. journal of consumer psychology 20, 17-19. https://doi.org/10.1016/j.jcps.2009.12.004 frederick, s., g. loewenstein & t. o'donoghe (2002). time discounting and time preference: a critical review. journal of economic literature 40, 351https://doi.org/10.1111/josi.12079 https://doi.org/10.1111/1467-9280.00372 https://doi.org/10.1111/j.1467-9280.2006.01704.x https://doi.org/10.1111/j.1467-9280.2006.01704.x https://doi.org/10.1016/j.jcps.2009.12.005 https://doi.org/10.1016/j.joep.2013.12.003 https://doi.org/10.1037/0033-295x.101.3.519 https://doi.org/10.1111/j.1467-6419.2010.00641.x https://doi.org/10.1111/j.1468-0297.2009.02251.x https://doi.org/10.2307/1884325 https://doi.org/10.1002/(sici)1099-0771(200001/03)13:1%3c1::aid-bdm333%3e3.0.co;2-s https://doi.org/10.1017/cbo9780511809477.024 https://doi.org/10.1007/s10290-013-0150-9 https://doi.org/10.1007/s10290-013-0150-9 https://doi.org/10.1016/j.socec.2004.04.006 https://doi.org/10.1016/j.jcps.2009.12.004 peón et al. / european journal of government and economics 6(1), 24-58 48 401. https://doi.org/10.1257/jel.40.2.351 füllbrunn, s., h.a. rau & u. weitzel (2014). does ambiguity aversion survive in experimental asset markets? journal of economic behavior and organization 107, 810826. https://doi.org/10.1016/j.jebo.2014.03.013 furnham, a. & h.c. boo (2011). a literature review on the anchoring effect. the journal of socio-economics 40, 35-42. https://doi.org/10.1016/j.socec.2010.10.008 gatherwood, j. & j. weber (2014). self-control, financial literacy and the co-holding puzzle. journal of economic behavior and organization 107, 455469. https://doi.org/10.1016/j.jebo.2014.04.018 gennaioli, n. & a. shleifer (2010). what comes to mind. the quarterly journal of economics 125(4), 1399-1433. https://doi.org/10.1162/qjec.2010.125.4.1399 gigerenzer, g. (1991). how to make cognitive illusions disappear: beyond 'heuristics and biases'. in w. stroebe & m. hewstone (eds.), european review of social psychology, wiley. https://doi.org/10.1080/14792779143000033 gigerenzer, g. & d.g. goldstein (2011). the recognition heuristic: a decade of research. judgment and decision making 6(1), 100-121. gigerenzer, g. & w. gaissmaier (2011). heuristic decision making. annual review of psychology 62, 451-482. https://doi.org/10.1146/annurev-psych-120709-145346 gigerenzer, g., u. hoffrage & h. kleinbölting (1991). probabilistic mental models: a brunswikian theory of confidence. psychological review 98, 506– 528. https://doi.org/10.1037/0033-295x.98.4.506 gilbert, p., j. price & s. allan (1995). social comparison, social attractiveness and evolution: how might they be related? new ideas in psychology 13(2), 149165. https://doi.org/10.1016/0732-118x(95)00002-x givoni, m., j. macmillen, d. banister, & e. feitelson (2013), from policy measures to policy packages. transport reviews 33(1), 1-20. https://doi.org/10.1080/01441647.2012.744779 gneezy, u., a. kapteyn & j. potters (2003). evaluation periods and asset prices in a market experiment. journal of finance 58, 821–837. https://doi.org/10.1111/1540-6261.00547 goldstein, d.g. & g. gigerenzer (1999). the recognition heuristic: how ignorance makes us smart. in g. gigerenzer, p.m. todd, and the abc research group, simple heuristics that make us smart, new york: oxford university press, pp. 37–58. goldstein, d.g. & g. gigerenzer (2002). models of ecological rationality: the recognition heuristic. psychological review 109, 75–90. https://doi.org/10.1037/0033-295x.109.1.75 griffin, d. & a. tversky (1992). the weighing of evidence and the determinants of confidence. cognitive psychology 24(3), 411-435. https://doi.org/10.1016/0010-0285(92)90013-r griffin, d.w., r. gonzalez, d.j. koehler & t. gilovich (2012). judgmental heuristics: a historical overview. in k.j. holyoak & r.g. morrison (eds.), the oxford handbook of thinking and reasoning, oxford: oxford university press, pp. 322345. https://doi.org/10.1093/oxfordhb/9780199734689.013.0017 griffith, r.m. (1949). odds adjustments by american horse-race bettors. the american journal https://doi.org/10.1257/jel.40.2.351 https://doi.org/10.1016/j.jebo.2014.03.013 https://doi.org/10.1016/j.socec.2010.10.008 https://doi.org/10.1016/j.jebo.2014.04.018 https://doi.org/10.1162/qjec.2010.125.4.1399 https://doi.org/10.1080/14792779143000033 https://doi.org/10.1146/annurev-psych-120709-145346 https://doi.org/10.1037/0033-295x.98.4.506 https://doi.org/10.1016/0732-118x(95)00002-x https://doi.org/10.1080/01441647.2012.744779 https://doi.org/10.1111/1540-6261.00547 https://doi.org/10.1037/0033-295x.109.1.75 https://doi.org/10.1016/0010-0285(92)90013-r https://doi.org/10.1093/oxfordhb/9780199734689.013.0017 peón et al. / european journal of government and economics 6(1), 24-58 49 of psychology 62(2), 290-294. https://doi.org/10.2307/1418469 grinblatt, m. & m. keloharju (2009). sensation seeking, overconfidence and trading activity. the journal of finance 64(2), 549-578. https://doi.org/10.1111/j.1540-6261.2009.01443.x guiso, l., p. sapienza & l. zingales (2006). does culture affect economic outcomes? journal of economic perspectives 20(2), 23-48. https://doi.org/10.1257/jep.20.2.23 güth, w., r. schmittberger & b. schwarze (1982). an experimental analysis of ultimatum bargaining. journal of economic behavior and organization 3, 367388. https://doi.org/10.1016/0167-2681(82)90011-7 haack, p., m.d. pfarrer & a.g. scherer (2014). legitimacy-as-feeling: how affect leads to vertical legitimacy spillovers in transnational governance. journal of management studies 51(4), 634-666. https://doi.org/10.1111/joms.12071 haisley, e., r. mostafa & g. loewenstein (2008). myopic risk-seeking: the impact of narrow decision bracketing on lottery play. journal of risk and uncertainty 37, 5775. https://doi.org/10.1007/s11166-008-9041-1 hardin, a.m. & c.a. looney (2012). myopic loss aversion: demystifying the key factors influencing decision problem framing. organizational behavior and human decision processes 117, 311-331. https://doi.org/10.1016/j.obhdp.2011.11.005 heath, c., r.p. larrick & j. klayman (1998). cognitive repairs: how organizational practices can compensate for individual shortcomings. research in organizational behavior 20, 1-37. henrich, j., r. boyd, s. bowles, c. camerer, h. gintis, r. mcelreath & e. fehr (2001). in search of homo economicus: behavioral experiments in 15 small-scale societies. american economic review 91(2), 73-79. https://doi.org/10.1257/aer.91.2.73 hens, t. & k. bachmann (2008). behavioral finance for private banking. john wiley and sons. hirshleifer, d. (2008). psychological bias as a driver of financial regulation. european financial management 14(5), 856-874. https://doi.org/10.1111/j.1468-036x.2007.00437.x hirshleifer, d. (2015). behavioral finance. annual review of financial economics 7(7), 133159. https://doi.org/10.1146/annurev-financial-092214-043752 hirshleifer, d. & s.h. teoh (2003). limited attention, information disclosure and financial reporting. journal of accounting and economics 36(1), 337386. https://doi.org/10.1016/j.jacceco.2003.10.002 hodges, s.h., r.g. tompkins & w.t. ziemba (2008). the favorite longshot bias in sandp 500 and ftse 100 index futures options: the return to bets and the cost of insurance. in handbook of sports and lottery markets, elsevier. hoff, k. & p. priyanka (2004). belief systems and durable inequalities: an experimental investigation of indian caste. world bank working paper 3351. https://doi.org/10.1596/18139450-3351 hoffrage, u., r. hertwig & g. gigerenzer (2000). hindsight bias: a by-product of knowledge updating? journal of experimental psychology: learning, memory and cognition 26(3), 566581. https://doi.org/10.1037/0278-7393.26.3.566 hossain, t. & j.a. list (2012). the behavioralist visits the factory: increasing productivity using https://doi.org/10.2307/1418469 https://doi.org/10.1111/j.1540-6261.2009.01443.x https://doi.org/10.1257/jep.20.2.23 https://doi.org/10.1016/0167-2681(82)90011-7 https://doi.org/10.1111/joms.12071 https://doi.org/10.1007/s11166-008-9041-1 https://doi.org/10.1016/j.obhdp.2011.11.005 https://doi.org/10.1257/aer.91.2.73 https://doi.org/10.1111/j.1468-036x.2007.00437.x https://doi.org/10.1146/annurev-financial-092214-043752 https://doi.org/10.1016/j.jacceco.2003.10.002 https://doi.org/10.1596/1813-9450-3351 https://doi.org/10.1596/1813-9450-3351 https://doi.org/10.1037/0278-7393.26.3.566 peón et al. / european journal of government and economics 6(1), 24-58 50 simple framing manipulations. management science 58(12), 21512167. https://doi.org/10.1287/mnsc.1120.1544 hu, y. & x. li (2011). context-dependent product evaluations: an empirical analysis of internet book reviews. journal of interactive marketing 25, 123133. https://doi.org/10.1016/j.intmar.2010.10.001 huber, j., m. kirchler & t. stöckl (2010). the hot hand belief and the gambler's fallacy in investment decisions under risk. theory and decision 68, 445462. https://doi.org/10.1007/s11238-008-9106-2 janis, i.l. (1972). victims of groupthink, houghton mifflin. jaspersen, j.g. & v. aseervatham (2015). the influence of affect on heuristic thinking in insurance demand. the journal of risk and insurance. https://doi.org/10.1111/jori.12088 kahneman, d. (2003a). maps of bounded rationality: psychology for behavioral economics. the american economic review 93(5), 1449-75. https://doi.org/10.1257/000282803322655392 kahneman, d. (2003b). a perspective on judgment and choice: mapping bounded rationality. american psychologist 58(9), 697-720. https://doi.org/10.1037/0003-066x.58.9.697 kahneman, d. & a. tversky (1979). prospect theory: an analysis of decision under risk. econometrica 47(2), 263-292. https://doi.org/10.2307/1914185 kahneman, d. & a. tversky (1984). choices, values and frames. american psychologist 39, 341-350. https://doi.org/10.1037/0003-066x.39.4.341 kahneman, d. & d. lovallo (1993). timid choices and bold forecasts: a cognitive perspective on risk taking. management science 39(1), 17-31. https://doi.org/10.1287/mnsc.39.1.17 kahneman, d. & m. riepe (1998). aspects of investor psychology. journal of portfolio management 24, 52-65. https://doi.org/10.3905/jpm.1998.409643 kahneman, d. & s. frederick (2002). representativeness revisited: attribute substitution in intuitive judgment. in t. gilovich, d. griffin & d. kahneman (eds.), heuristics and biases: the psychology of intuitive thought, cambridge university press, pp. 4981. https://doi.org/10.1017/cbo9780511808098.004 kahneman, d. & s. frederick (2005). a model of heuristic judgment. in k.j. holyoake & r.g. morrison (eds.) the cambridge handbook of thinking and reasoning, cambridge university press. kahneman, d., j.l. knetsch & r.h. thaler (1986a). fairness as a constraint on profit seeking: entitlements in the market. american economic review 76(4), 728-741. kahneman, d., j.l. knetsch & r.h. thaler (1986b). fairness and the assumptions of economics. journal of business 59(4), s285-s300. https://doi.org/10.1086/296367 kahneman, d., j.l. knetsch & r.h. thaler (1991). anomalies: the endowment effect, loss aversion, and status quo bias. the journal of economic perspectives 5(1), 193206. https://doi.org/10.1257/jep.5.1.193 kahneman, d., p. slovic & a. tversky (1982). judgment under uncertainty: heuristics and biases. cambridge university press. https://doi.org/10.1017/cbo9780511809477 katona, g. (1975). psychological economics. elsevier. https://doi.org/10.1287/mnsc.1120.1544 https://doi.org/10.1016/j.intmar.2010.10.001 https://doi.org/10.1007/s11238-008-9106-2 https://doi.org/10.1111/jori.12088 https://doi.org/10.1257/000282803322655392 https://doi.org/10.1037/0003-066x.58.9.697 https://doi.org/10.2307/1914185 https://doi.org/10.1037/0003-066x.39.4.341 https://doi.org/10.1287/mnsc.39.1.17 https://doi.org/10.3905/jpm.1998.409643 https://doi.org/10.1017/cbo9780511808098.004 https://doi.org/10.1086/296367 https://doi.org/10.1257/jep.5.1.193 https://doi.org/10.1017/cbo9780511809477 peón et al. / european journal of government and economics 6(1), 24-58 51 kirby, k.n. & r.j. herrnstein (1995). preference reversals due to myopic discounting of delayed reward. psychology science 6(2), 83-89. https://doi.org/10.1111/j.1467-9280.1995.tb00311.x knetsch, j.l. (1989). the endowment effect and evidence of nonreversible indifference curves. the american economic review 79(5), 1277-84. koch, a.k. & j. nafzinger (2016). goals and bracketing under mental accounting. journal of economic theory 162, 305-361. https://doi.org/10.1016/j.jet.2016.01.001 koop, g.j. & j.g. johnson (2012). the use of multiple reference points in risky decision making. journal of behavioral decision making 25, 49-62. https://doi.org/10.1002/bdm.713 koriat, a., s. lichtenstein & b. fischhoff (1980). reasons for confidence. journal of experimental psychology: human learning and memory 6(2), 107118. https://doi.org/10.1037/0278-7393.6.2.107 kõszegi, b. & m. rabin (2006). a model of reference-dependent preferences. quarterly journal of economics 121(4), 1133-65. https://doi.org/10.1093/qje/121.4.1133 kuran, t. & c. sunstein (1999). availability cascades and risk regulation. stanford law review 51, 683–768. https://doi.org/10.2307/1229439 lacetera, n., d.g. pope & j.r. sydnor (2014). heuristic thinking and limited attention in the car market. american economic review 102(5), 22062236. https://doi.org/10.1257/aer.102.5.2206 lahvicka, j. (2014). what causes the favourite-longshot bias? further evidence from tennis. applied economic letters 21(2), 90-92. https://doi.org/10.1080/13504851.2013.842628 landman, j. & m. manis (1983). social cognition: some historical and theoretical perspectives. advances in experimental social psychology 16, 49-123. https://doi.org/10.1016/s00652601(08)60394-9 langer, e.j. (1975). the illusion of control. the journal of personality and social psychology, 32, 311-328, reprinted short version in d. kahneman, p. slovic & a. tversky (1982), judgment under uncertainty: heuristics and biases, cambridge university press. langer, t. & m. weber (2005). myopic prospect theory vs. myopic loss aversion: how general is the phenomenon? journal of economic behavior and organization 56, 2538. https://doi.org/10.1016/j.jebo.2003.01.005 lee, c.j. & e.b. andrade (2011). fear, social projection, and financial decision making. journal of marketing research 48, s121-s129. https://doi.org/10.1509/jmkr.48.spl.s121 lejarraga, t. & c. gonzalez (2011). effects of feedback and complexity on repeated decisions from description. organizational behavior and human decision processes 116, 286295. https://doi.org/10.1016/j.obhdp.2011.05.001 levinson, j.d. & k. peng (2007). valuing cultural differences in behavioral economics. the icfai journal of behavioral finance 4(1), 32-47. libby, r. & k. rennekamp (2012). self-serving attribution bias, overconfidence, and the issuance of management forecasts. journal of accounting research 50(1), 197231. https://doi.org/10.1111/j.1475-679x.2011.00430.x liberali, j.m., v.f. reyna, s. furlan, l.s. stein & s.t. pardo (2012). individual differences in https://doi.org/10.1111/j.1467-9280.1995.tb00311.x https://doi.org/10.1016/j.jet.2016.01.001 https://doi.org/10.1002/bdm.713 https://doi.org/10.1037/0278-7393.6.2.107 https://doi.org/10.1093/qje/121.4.1133 https://doi.org/10.2307/1229439 https://doi.org/10.1257/aer.102.5.2206 https://doi.org/10.1080/13504851.2013.842628 https://doi.org/10.1016/s0065-2601(08)60394-9 https://doi.org/10.1016/s0065-2601(08)60394-9 https://doi.org/10.1016/j.jebo.2003.01.005 https://doi.org/10.1509/jmkr.48.spl.s121 https://doi.org/10.1016/j.obhdp.2011.05.001 https://doi.org/10.1111/j.1475-679x.2011.00430.x peón et al. / european journal of government and economics 6(1), 24-58 52 numeracy and cognitive reflection, with implications for biases and fallacies in probability judgments. journal of behavioral decision making 25, 361381. https://doi.org/10.1002/bdm.752 list, j.a. (2007). on the interpretation of giving in dictator games. journal of political economy 115(3), 482-493. https://doi.org/10.1086/519249 liu, h. & a. m. colman (2009). ambiguity aversion in the long run: repeated decisions under risk and uncertainty. journal of economic psychology 30, 277284. https://doi.org/10.1016/j.joep.2009.02.001 lo, a.w., d.v. repin & b.n. steenbarger (2005). fear and greed in financial markets: a clinical study of day-traders. the american economic review 95(2), 352359. https://doi.org/10.1257/000282805774670095 loewenstein, g. (1996). out of control: visceral influences on behavior. organizational behavior and human decision processes 65(3), 272292. https://doi.org/10.1006/obhd.1996.0028 loewenstein, g., e.u. weber, c.k. hsee & n. welch (2001). risk as feelings. psychological bulletin 127(2), 267-86. https://doi.org/10.1037/0033-2909.127.2.267 loewenstein, g., t. o'donoghue & m. rabin (2003). projection bias in predicting future utility. quarterly journal of economics 118(4), 120948. https://doi.org/10.1162/003355303322552784 lovallo, d. & d. kahneman (2003). delusions of success: how optimism undermines executives' decisions. harvard business review 81(7), 56-63. lunenburg, f.c. (2010). group decision making: the potential for groupthink. international journal of management, business, and administration 13(1), 1-6. lupton, d. (2013). risk and emotion: towards an alternative theoretical perspective. health, risk and society 15(8), 634-647 https://doi.org/10.1080/13698575.2013.848847 malmendier, u. & g. tate (2005). ceo overconfidence and corporate investment. journal of finance 60(6), 2661-2700. https://doi.org/10.1111/j.1540-6261.2005.00813.x manski, c. (2000). economic analysis of social interactions. journal of economic perspectives 14(3), 115-136. https://doi.org/10.1257/jep.14.3.115 marewski, j.n., w. gaissmaier & g. gigerenzer (2010). good judgments do not require complex cognition. cognitive processing 11(2), 103-121. https://doi.org/10.1007/s10339009-0337-0ç mayo-wilson, c., k. zollman & d. danks (2012). wisdom of crowds versus groupthink: learning in groups and in isolation. international journal of game theory 42, 695723. https://doi.org/10.1007/s00182-012-0329-7 mcgraw, a.p., j.t. larsen, d. kahneman & d. schkade (2010). comparing gains and losses. psychological science 21(10), 1438-45. https://doi.org/10.1177/0956797610381504 milgram, s. (1963). behavioral study of obedience. the journal of abnormal and social psychology 67(4), 371-378. https://doi.org/10.1037/h0040525 miller, j.g. (1984). culture and the development of everyday social explanation. journal of https://doi.org/10.1002/bdm.752 https://doi.org/10.1086/519249 https://doi.org/10.1016/j.joep.2009.02.001 https://doi.org/10.1257/000282805774670095 https://doi.org/10.1006/obhd.1996.0028 https://doi.org/10.1037/0033-2909.127.2.267 https://doi.org/10.1162/003355303322552784 https://doi.org/10.1080/13698575.2013.848847 https://doi.org/10.1111/j.1540-6261.2005.00813.x https://doi.org/10.1257/jep.14.3.115 https://doi.org/10.1007/s10339-009-0337-0%c3%a7 https://doi.org/10.1007/s10339-009-0337-0%c3%a7 https://doi.org/10.1007/s00182-012-0329-7 https://doi.org/10.1177/0956797610381504 https://doi.org/10.1037/h0040525 peón et al. / european journal of government and economics 6(1), 24-58 53 personality and social psychology 46, 961-978. https://doi.org/10.1037/0022-3514.46.5.961 moore, d.a. & p.j. healy (2008). the trouble with overconfidence. psychological review 115(2), 502–517. https://doi.org/10.1037/0033-295x.115.2.502 mullainathan, s. & r.h. thaler (2000). behavioral economics. mit dept. of economics working paper no. 00-27. https://doi.org/10.2139/ssrn.245828 nguyen, b. & p. klaus (2013). retail fairness: exploring consumer perceptions of fairness towards retailers' marketing tactics. journal of retailing and consumer services 20, 311324. https://doi.org/10.1016/j.jretconser.2013.02.001 nisbett, r.e. & t. masuda (2003). culture and point of view. proceedings of the national academy of science 100, 11163-70. https://doi.org/10.1073/pnas.1934527100 nofsinger, j.r. (2012). household behavior and boom/bust cycles. journal of financial stability 8, 161-173. https://doi.org/10.1016/j.jfs.2011.05.004 norman, g., sherbino, j., dore, k., wood, t., young, m., gaissmaier, w., kreuger, s., monteiro, s. (2014). the etiology of diagnostic errors: a controlled trial of system 1 versus system 2 reasoning. academic medicine 89(2), 277284. https://doi.org/10.1097/acm.0000000000000105 novemsky, n. & d. kahneman (2005). the boundaries of loss aversion. journal of marketing research 42(2), 119-128. https://doi.org/10.1509/jmkr.42.2.119.62292 odean, t. (1998). volume, volatility, price and profit when all traders are above average. journal of finance 53(6), 1887-1934. https://doi.org/10.1111/0022-1082.00078 odean, t. (1999). do investors trade too much? the american economic review 89(5), 12791298. https://doi.org/10.1257/aer.89.5.1279 okun, a. (1981). prices and quantities: a macro-economic analysis. the brookings institution. ortmann, a., g. gigerenzer, b. borges & d.g. goldstein (2008). the recognition heuristic: a fast and frugal way to investment choice? in c.r. plott & v.l. smith (eds.) handbook of experimental economics results: vol 1, handbooks in economics 28, 993– 1003. https://doi.org/10.1016/s1574-0722(07)00107-2 ottaviani, m. & p. sorensen (2007). the favorite-longshot bias: an overview of the main explanations. in d. hausch & w. ziemba (eds.) efficiency of sports and betting markets, elsevier: handbook in finance series. pacheco, j. (2012). the social contagion model: exploring the role of public opinion on the diffusion of antismoking legislation across the american states. the journal of politics 74(1), 187-202. https://doi.org/10.1017/s0022381611001241 pachur, t. & m. galesic (2013). strategy selection in risky choice: the impact of numeracy, affect, and cross-cultural differences. journal of behavioral decision making 26, 260271. https://doi.org/10.1002/bdm.1757 page, l. & r.t. clemen (2013). do prediction markets produce well-calibrated probability forecasts? the economic journal 123, 491-513. https://doi.org/10.1111/j.14680297.2012.02561.x pan, c. & m. statman (2010). beyond risk tolerance: regret, overconfidence and other investor https://doi.org/10.1037/0022-3514.46.5.961 https://doi.org/10.1037/0033-295x.115.2.502 https://doi.org/10.2139/ssrn.245828 https://doi.org/10.1016/j.jretconser.2013.02.001 https://doi.org/10.1073/pnas.1934527100 https://doi.org/10.1016/j.jfs.2011.05.004 https://doi.org/10.1097/acm.0000000000000105 https://doi.org/10.1509/jmkr.42.2.119.62292 https://doi.org/10.1111/0022-1082.00078 https://doi.org/10.1257/aer.89.5.1279 https://doi.org/10.1016/s1574-0722(07)00107-2 https://doi.org/10.1017/s0022381611001241 https://doi.org/10.1002/bdm.1757 https://doi.org/10.1111/j.1468-0297.2012.02561.x https://doi.org/10.1111/j.1468-0297.2012.02561.x peón et al. / european journal of government and economics 6(1), 24-58 54 propensities. scu leavey school of business, research paper no. 10-05. pennycook, g., d. trippas, s.j. handley & v.a. thompson (2014). base rates: both neglected and intuitive. journal of experimental psychology: learning, memory and cognition 40(2), 544-554. https://doi.org/10.1037/a0034887 peón, d., m. antelo & a. calvo (2015). a dynamic behavioral model of the credit boom. journal of economic issues 49(4), 1077-99. https://doi.org/10.1080/00213624.2015.1105043 plous, s. (1993). the psychology of judgment and decision making. mcgraw-hill. polman, e. (2012). self-other decision making and loss aversion. organizational behavior and human decision processes 119, 141-150. https://doi.org/10.1016/j.obhdp.2012.06.005 prendergast, c. & l. stole (1996). impetuous youngsters and jaded old-timers: acquiring a reputation for learning. journal of political economy 104, 1105– 34. https://doi.org/10.1086/262055 rabin, m. (1998). psychology and economics. journal of economic literature 36(1), 11-46. rabin, m. (2000). risk aversion and expected-utility theory: a calibration theorem. econometrica 68(5), 1281-92. https://doi.org/10.1111/1468-0262.00158 rabin, m. (2002). inference by believers in the law of small numbers. quarterly journal of economics 117, 775-816. https://doi.org/10.1162/003355302760193896 rabin, m. & d. vayanos (2010). the gambler's and hot-hand fallacies: theory and applications. review of economic studies 77(2), 730-778. https://doi.org/10.1111/j.1467937x.2009.00582.x rabin, m. & j. schrag (1999). first impressions matter: a model of confirmatory bias. quarterly journal of economics 114(1), 37-82. https://doi.org/10.1162/003355399555945 rapp, a., l.s. beitelspacher, d. grewal & d.e. hughes (2013). understanding social media effects across seller, retailer, and consumer interactions. journal of the academy of marketing science 41, 547-566. https://doi.org/10.1007/s11747-013-0326-9 ratner, r.k., soman, d., zauberman, g., ariely, d., et al.(2008). how behavioral decision research can enhance consumer welfare: from freedom of choice to paternalistic intervention. marketing letters 19, 383-397. https://doi.org/10.1007/s11002-008-9044-3 read, d., g. loewenstein and m. rabin (1999). choice bracketing. journal of risk and uncertainty 19, 171-197. https://doi.org/10.1023/a:1007879411489 richins, m.l. (1991). comparison and the idealized images of advertising. journal of consumer research 18(1), 71-83. https://doi.org/10.1086/209242 rieger, m.o. (2012). why do investors buy bad financial products? probability misestimation and preferences in financial investment decision. journal of behavioral finance 13(2), 108118. https://doi.org/10.1080/15427560.2012.680991 roca, m., r.m. hogarth & j. maule (2006). ambiguity seeking as a result of the status quo bias. journal of risk and uncertainty 32(3), 175-194. https://doi.org/10.1007/s11166-006-9518-8 roese, n.j. & k.d. vohs (2012). hindsight bias. perspectives on psychological science 7(5), 411-426. https://doi.org/10.1177/1745691612454303 rooderkerk, r.p., h.j. van heerde & t.h.a. bijmolt (2011). incorporating context effects into a https://doi.org/10.1037/a0034887 https://doi.org/10.1080/00213624.2015.1105043 https://doi.org/10.1016/j.obhdp.2012.06.005 https://doi.org/10.1086/262055 https://doi.org/10.1111/1468-0262.00158 https://doi.org/10.1162/003355302760193896 https://doi.org/10.1111/j.1467-937x.2009.00582.x https://doi.org/10.1111/j.1467-937x.2009.00582.x https://doi.org/10.1162/003355399555945 https://doi.org/10.1007/s11747-013-0326-9 https://doi.org/10.1007/s11002-008-9044-3 https://doi.org/10.1023/a:1007879411489 https://doi.org/10.1086/209242 https://doi.org/10.1080/15427560.2012.680991 https://doi.org/10.1007/s11166-006-9518-8 https://doi.org/10.1177/1745691612454303 peón et al. / european journal of government and economics 6(1), 24-58 55 choice model. journal of marketing research 48, 767780. https://doi.org/10.1509/jmkr.48.4.767 rubin, j. (2014). centralized institutions and cascades. journal of comparative economics 42, 340-357. https://doi.org/10.1016/j.jce.2013.11.002 sahi, s.k., a.p. arora & n. dhameja (2013). an exploratory inquiry into the psychological biases in financial investment behavior. journal of behavioral finance 14(2), 94103. https://doi.org/10.1080/15427560.2013.790387 samuelson, w. & r. zeckhauser (1988). status quo bias in decision making. journal of risk and uncertainty 1(1), 7-59. https://doi.org/10.1007/bf00055564 scheinkman, j.a. & w. xiong (2003). overconfidence and speculative bubbles. journal of political economy 111(6), 1183-1219. https://doi.org/10.1086/378531 schlüter, a. & b. vollan (2015). flowers and an honour box: evidence on framing effects. journal of behavioral and experimental economics 57, 186199. https://doi.org/10.1016/j.socec.2014.10.002 schmidt, u. & h. zank (2012). a genuine foundation for prospect theory. journal of risk and uncertainty 45, 97-113. https://doi.org/10.1007/s11166-012-9150-8 schooler, l.j. & r. hertwig (2005). how forgetting aids heuristic inference. psychological review 112, 610-628. https://doi.org/10.1037/0033-295x.112.3.610 schwager, s. & k. rothermund (2013). motivation and affective processing biases in risky decision making: a counter-regulation account. journal of economic psychology 38, 111126. https://doi.org/10.1016/j.joep.2012.08.005 schwikert, s.r. & t. curran (2014). familiarity and recollection in heuristic decision making. journal of experimental psychology 143(6), 2341-65. https://doi.org/10.1037/xge0000024 sen, a. (1995). moral codes and economic success. in c.s. brittan & a. hamlin (eds.) market capitalism and moral values, brookfield, aldershot. shafir, e. & r.h. thaler (2006). invest now, drink later, spend never: the mental accounting of delayed consumption. journal of economic psychology 27, 694712. https://doi.org/10.1016/j.joep.2006.05.008 shefrin, h. (2000). beyond greed and fear: understanding behavioral finance and the psychology of investing. oxford university press, second edition (2002). shefrin, h. (2006). behavioral corporate finance. mcgraw-hill. shefrin, h. & e.m. cervellati (2011). bp's failure to debias: underscoring the importance of behavioral corporate finance. quarterly journal of finance 1(1), 127168. https://doi.org/10.1142/s2010139211000043 shefrin, h. & m. statman (1985). the disposition to sell winners too early and ride losers too long: theory and evidence. journal of finance 40, 777-790. https://doi.org/10.1111/j.15406261.1985.tb05002.x shefrin, h. & m. statman (2000). behavioral portfolio theory. journal of financial and quantitative analysis 35(2), 127-151. https://doi.org/10.2307/2676187 sherif, m. (1937). an experimental approach to the study of attitudes. sociometry 1, 90https://doi.org/10.1509/jmkr.48.4.767 https://doi.org/10.1016/j.jce.2013.11.002 https://doi.org/10.1080/15427560.2013.790387 https://doi.org/10.1007/bf00055564 https://doi.org/10.1086/378531 https://doi.org/10.1016/j.socec.2014.10.002 https://doi.org/10.1007/s11166-012-9150-8 https://doi.org/10.1037/0033-295x.112.3.610 https://doi.org/10.1016/j.joep.2012.08.005 https://doi.org/10.1037/xge0000024 https://doi.org/10.1016/j.joep.2006.05.008 https://doi.org/10.1142/s2010139211000043 https://doi.org/10.1111/j.1540-6261.1985.tb05002.x https://doi.org/10.1111/j.1540-6261.1985.tb05002.x https://doi.org/10.2307/2676187 peón et al. / european journal of government and economics 6(1), 24-58 56 98. https://doi.org/10.2307/2785261 shiller, r.j. (1984). stock prices and social dynamics. brookings papers on economic activity, 457-498. https://doi.org/10.2307/2534436 shiller, r.j. (2000a). human behavior and the efficiency of the financial system. in b. taylor & m. woodfood (eds.), handbook of macroeconomics, amsterdam: north holland. shiller, r.j. (2000b). irrational exuberance. princeton university press. shiller, r.j. (2003). from efficient market theory to behavioral finance. journal of economic perspectives 17(1), 83-104. https://doi.org/10.1257/089533003321164967 shin, h.s. (1992). prices of state contingent claims with insider traders, and the favoritelongshot bias. the economic journal 102(411), 426-435. https://doi.org/10.2307/2234526 shleifer, a. (2000). inefficient markets: an introduction to behavioral finance. oxford university press. https://doi.org/10.1093/0198292279.001.0001 shleifer, a. (2004). does competition destroy ethical behavior? american economic association papers and proceedings 94(2), 414-418. https://doi.org/10.1257/0002828041301498 slovic, p., m.l. finucane, e. peters & d.g. macgregor (2002). the affect heuristic. in t. gilovich, d. griffin & d. kahneman, eds., heuristics and biases: the psychology of intuitive judgment, cambridge university press, 397420. https://doi.org/10.1017/cbo9780511808098.025 statman, m. (1999). behavioral finance: past battles, future engagements. financial analysts journal 55(6), 18-27. https://doi.org/10.2469/faj.v55.n6.2311 statman, m. & j.a. weng (2010). investment across cultures: financial attitudes of chineseamericans. journal of investment consulting 11(1), 37-44. stevens, j.r. (2016). intertemporal similarity: discounting as a last resort. journal of behavioral decision making 29, 12-24. https://doi.org/10.1002/bdm.1870 stöckl, t., j. huber, m. kirchler & f. lindner (2015). hot hand and gambler's fallacy in teams: evidence from investment experiments. journal of economic behavior and organization 117, 327-339. https://doi.org/10.1016/j.jebo.2015.07.004 stulz, r.m. & r. williamson (2003). culture, openness, and finance. journal of financial economics 70(3), 313-349. https://doi.org/10.1016/s0304-405x(03)00173-9 suetens, s. & j.-r. tyran (2012). the gambler's fallacy and gender. journal of economic behavior and organization 83, 118-124. https://doi.org/10.1016/j.jebo.2011.06.017 sul, s., j. kim & i. choi (2013). subjective well-being and hedonic editing: how happy people maximize joint outcomes of loss and gain. journal of happiness studies 14, 14091430. https://doi.org/10.1007/s10902-012-9379-6 taylor, m.z., & s. wilson (2012). does culture still matter? the effects of individualism on national innovation rates. journal of business venturing 27, 234247. https://doi.org/10.1016/j.jbusvent.2010.10.001 thaler, r.h. (1983). illusions and mirages in public policy. public interest 73, 60-74. reprinted in h. arkes & k. hammond (eds.), judgment and decision making: an interdisciplinary reader, cambridge university press. https://doi.org/10.2307/2785261 https://doi.org/10.2307/2534436 https://doi.org/10.1257/089533003321164967 https://doi.org/10.2307/2234526 https://doi.org/10.1093/0198292279.001.0001 https://doi.org/10.1257/0002828041301498 https://doi.org/10.1017/cbo9780511808098.025 https://doi.org/10.2469/faj.v55.n6.2311 https://doi.org/10.1002/bdm.1870 https://doi.org/10.1016/j.jebo.2015.07.004 https://doi.org/10.1016/s0304-405x(03)00173-9 https://doi.org/10.1016/j.jebo.2011.06.017 https://doi.org/10.1007/s10902-012-9379-6 https://doi.org/10.1016/j.jbusvent.2010.10.001 peón et al. / european journal of government and economics 6(1), 24-58 57 thaler, r.h. (1985). mental accounting and consumer choice. marketing science 4(3), 199-214. https://doi.org/10.1287/mksc.4.3.199 thaler, r.h. (1999). mental accounting matters. journal of behavioral decision making 12(3), 183-206. https://doi.org/10.1002/(sici)1099-0771(199909)12:3<183::aidbdm318>3.0.co;2-f thaler, r.h. (2008). mental accounting and consumer choice: anatomy of a failure. marketing science 27(1), 12-14. https://doi.org/10.1287/mksc.1070.0348 thaler, r.h. & e.j. johnson (1990). gambling with the house money and trying to break even: the effects of prior outcomes on risky choice. management science 36(6), 643660. https://doi.org/10.1287/mnsc.36.6.643 thaler, r.h. & h.m. shefrin (1981). an economic theory of self-control. journal of political economy 89(2), 392-406. https://doi.org/10.1086/260971 thaler, r.h., a. tversky, d. kahneman & a. schwartz (1997). the effect of myopia and loss aversion on risk: taking an experimental test. quarterly journal of economics 112(2), 646661. https://doi.org/10.1162/003355397555226 thaler, r.h., sunstein, c.r. (2003). behavioral economics, public policy, and paternalism: libertarian paternalism. american economic review 93(2), 175– 179. https://doi.org/10.1257/000282803321947001 trivers, r.l. (1971). the evolution of reciprocal altruism. quarterly review of biology 46, 3557. https://doi.org/10.1086/406755 troye, s.v. & m. supphellen (2012). consumer participation in coproduction: "i made it myself" effects on consumers' sensory perceptions and evaluations of outcome and input product. journal of marketing 76, 33-46. https://doi.org/10.1509/jm.10.0205 tsetsos, k., n. chater & m. usher (2012). salience driven value integration explains decision biases and preference reversal. proceedings of the national academy of sciences of the united states of america 109(24), 9659-64. https://doi.org/10.1073/pnas.1119569109 tversky, a. & d. kahneman (1971). the belief in the law of small numbers. psychological bulletin 76(2), 105-110. https://doi.org/10.1037/h0031322 tversky, a. & d. kahneman (1973). availability: a heuristic for judging frequency and probability. cognitive psychology 5(2), 207-232. reprinted in d. kahneman, p. slovic & a. tversky (1982), judgment under uncertainty: heuristics and biases, cambridge university press. tversky, a. & d. kahneman (1974). judgment under uncertainty: heuristics and biases. science 185(4157), 1124-31. https://doi.org/10.1126/science.185.4157.1124 tversky, a. & d. kahneman (1981). the framing of decisions and the psychology of choice. science 211(4481), 453-458. https://doi.org/10.1126/science.7455683 tversky, a. & d. kahneman (1982a). evidential impact of base rates. in d. kahneman, p. slovic and a. tversky, judgment under uncertainty: heuristics and biases, cambridge university press. https://doi.org/10.1017/cbo9780511809477.011 tversky, a. & d. kahneman (1982b). judgments of and by representativeness. in d. https://doi.org/10.1002/(sici)1099-0771(199909)12:3%3c183::aid-bdm318%3e3.0.co;2-f https://doi.org/10.1002/(sici)1099-0771(199909)12:3%3c183::aid-bdm318%3e3.0.co;2-f https://doi.org/10.1287/mksc.1070.0348 https://doi.org/10.1287/mnsc.36.6.643 https://doi.org/10.1086/260971 https://doi.org/10.1162/003355397555226 https://doi.org/10.1257/000282803321947001 https://doi.org/10.1086/406755 https://doi.org/10.1509/jm.10.0205 https://doi.org/10.1073/pnas.1119569109 https://doi.org/10.1037/h0031322 https://doi.org/10.1126/science.185.4157.1124 https://doi.org/10.1126/science.7455683 https://doi.org/10.1017/cbo9780511809477.011 peón et al. / european journal of government and economics 6(1), 24-58 58 kahneman, p. slovic & a. tversky, judgment under uncertainty: heuristics and biases, cambridge university press. https://doi.org/10.1017/cbo9780511809477.007 tversky, a. & d. kahneman (1983). extension versus intuitive reasoning: the conjunction fallacy in probability judgment. psychological review 90(4), 293315. https://doi.org/10.1037/0033-295x.90.4.293 tversky, a. & d. kahneman (1986). rational choice and the framing of decisions. the journal of business 59(s4), s251-s278. https://doi.org/10.1086/296365 tversky, a. & d. kahneman (1991). loss aversion in riskless choice: a reference-dependent model. the quarterly journal of economics 106(4), 103961. https://doi.org/10.2307/2937956 tversky, a. & d. kahneman (1992). advances in prospect theory: cumulative representation of uncertainty. journal of risk and uncertainty 5(4), 297323. https://doi.org/10.1007/bf00122574 tversky, a. & i. simonson (1993). context-dependent preferences. management science 39(10), 1179-89. https://doi.org/10.1287/mnsc.39.10.1179 weiner, b. (1985). an attributional theory of achievement motivation and emotion. psychological review 92, 548–573. https://doi.org/10.1037/0033-295x.92.4.548 wu, j. (2015). helpful laymen in informational cascades. journal of economic behavior and organization 116, 407-415. https://doi.org/10.1016/j.jebo.2015.05.002 zeisberger, s., d. vrecko & t. langer (2015). measuring time stability of prospect theory preferences. theory and decision 72, 359-386. https://doi.org/10.1007/s11238-010-9234-3 https://doi.org/10.1017/cbo9780511809477.007 https://doi.org/10.1037/0033-295x.90.4.293 https://doi.org/10.1086/296365 https://doi.org/10.2307/2937956 https://doi.org/10.1007/bf00122574 https://doi.org/10.1287/mnsc.39.10.1179 https://doi.org/10.1037/0033-295x.92.4.548 https://doi.org/10.1016/j.jebo.2015.05.002 https://doi.org/10.1007/s11238-010-9234-3 abstract. this paper overviews the theoretical and empirical research on behavioral biases and their influence in the literature. to provide a systematic exposition, we present a unified framework that takes the reader through an original taxonomy, ba... keywords. behavioral biases; decision-making; heuristics; framing; prospect theory; social contagion. jel classification. d03; g02; g11; g14; g30 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real vol.7 • no.2 2018 issn: 2254-7088 european journal of government and economics 7(2), december 2018. european journal of government and economics issn: 2254-7088 number 7, issue 2, december 2018 doi: https://doi.org/10.17979/ejge.2017.7.2 transparency and local government corruption: what does lack of transparency hide? 106-122 juan luis jiménez, daniel albalate 106-122 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real. 123-137 carlos fernández_herraiz, antonio javier prado domínguez, carlos pateiro-rodriguez, jesus m. garcia-iglesias path – dependence and european fisheries management 138-153 federico martín palmero, fernando gonzález laxe why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem 154-170 filipa figueira deciding on financial renegotiation in public-private partnership projects 171-199 wiston risso https://doi.org/10.17979/ejge.2017.7.2 european journal of government and economics 7(2), december 2018, 123-137. european journal of government and economics issn: 2254-7088 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real carlos fernández-herraiza, a. javier prado-domínguezb, carlos pateiro-rodriguezb, jesus m. garcia-iglesiasc,* a instituto de estudios bursátiles, spain b universidade da coruña, spain c universidad de extremadura, spain * corresponding author at: jmgarcia@unex.es article history. received 16 july 2018; accepted 6 october 2018. abstract. exchange credibility is a valuable asset for currency policymakers. in this article we intend to analyse exchange credibility from the perspective of carry trade speculators. global speculators' access to shadow banking financing allows them to build dynamic carry trade strategies that are a source of potential financial instability. our view is that the existence of dynamic carry trade opportunities offers a tool for monitoring how market participants asses the credibility of exchange policies. we use the long term performance of different carry trade dynamic specifications to understand how different is the market view of exchange policy credibility in the case of two leading latin america countries, mexico and brazil. our empirical research covers data from may of 2000 to may 2018. in light of the evidence presented, we recognize that mexican peso exchange policy is considered credible but brazilian real exchange policy is not considered credible during the sample period. keywords. currency carry trade, sharpe ratio, exchange policy credibility, shadow banking. doi. https://doi.org/10.17979/ejge.2018.7.2.4520 1. introduction in today's economy, various types of juxtaposition between money markets and financial markets are clearly visible; the confusion between leverage and liquidity, regulated and unregulated players; the impact of technology and the "laissez-faire" promoted by academic orthodoxy and regulators (see shiller, 2012). all this is also combined with the great momentum of shadow banking. hidden or 'shadow' banking covers all those activities in which credit, liquidity or term transformations occur without being possible to establish sufficient supervision over the involved intervening parties. shadow banking represents an uncomfortable reality that weighs just over 48% of the total assets of the global financial system (financial stability board, 2018) and its relative importance continues to grow, especially in foreign exchange markets. in addition, 28% of this activity is considered destabilising shadow banking (financial stability board, 2018), i.e. activities that are not sufficiently supervised and which could jeopardise the integrity of the global financial system. in short, not adequately controlled sources of systemic mailto:jmgarcia@unex.es https://doi.org/10.17979/ejge.2018.7.2.4520 c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 124 risk. carry trade in foreign exchange has been the subject of a long tradition of academic research that goes back to the analysis of the non-interest rate default and the various explanations for the existence of a forward premium on foreign exchange transactions. thus, following the progressive liberalisation of exchange rates in the 1970s, the empirical evidence of this failure and the difficulty of structural models in reproducing the behaviour of currency prices generated an interest from the academic world and from practitioners that has remained active to this day. in our view, the connection between carry trade, shadow banking and foreign exchange policy already implies an in-depth reflection on the functioning of today's financial markets. one might therefore wonder whether it is not the financing of trade in these assets that underpins carry trade strategies, and whether the ocean on which they are traded is not shadow banking. in short, a system in which leverage and liquidity end up being the front and back of the same thing. where leverage (and liquidity) are also pro-cyclical. and where the set of agents that use the same rules: basically carry strategies and portfolio rebalancing based on the use of value at risk (var) models. the foreign exchange market is both the largest financial market and a largely deregulated market. it is also one of the markets where carry trade strategies have been most exploited by professional speculators. our study aims to explore the extent to which the existence of shadow banking rewards speculative strategies based on simple and reproducible rules. these conditions may imply that in the currency markets, destabilizing speculative strategies offer consistently attractive and lasting results over time. if this were so, we would be calling into question the social role of speculators in today's financial markets and thus delegitimising some of the freedom of action enjoyed by financial institutions and institutional investors in the currency markets. the main contribution of the paper to current research in the area of exchange rates and policy intervention in currency markets is the connexion established between exchange policy credibility and ex post carry trade returns from stylized carry trade strategies employed by professional speculators. long lasting attractive performance coming from destabilizing carry trade specifications gives rise to reasonable doubts regarding exchange policy credibility perceived by the markets. the paper is structured as follows. section two covers the foundations of carry trade activity, its connection with shadow banking and our definition of destabilizing currency carry trade. section three defines the set of strategies that we will use in the next section. section four covers our empirical analysis of the mexican peso and the brazilian real within the may-2000 to may-2018 sample and explores the consequences of the evidence presented in terms of exchange rate policy credibility. the last section covers the main conclusions of our analysis and potential next steps in this research. c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 125 2. carry trade in the forex market and a brief reference to carry to risk. the academic interest in the study of the carry trade in the foreign exchange market can be traced from the observation of a double empirical problem. first, the inability of the structural models of the 1970s to provide a better prediction for currencies than a random walk with no trend (meese and rogoff, 1983). secondly, the finding by various authors, including hansen and hodrick (1980) or fama (1984), of a breach of the uncovered interest parity (uirp) on exchange rates, and the "discovery" of the forward premium puzzle. an important part of the literature has been devoted to the discussion of the unfulfilment of the uncovered interest rate parity. in fact, academic interest in the carry trade emerges as a byproduct of empirical evidence of such empirical puzzle. ongoing research has mainly focused on justifying the 'forward premium' in two ways. the first category of explanations, which most authors subscribe to, maintains the assumption of rational expectations, and interprets the systematic component of the forward market prediction error as a risk premium: "time varying risk premium" that fits naturally with models based on the stochastic discount factor. the second category attributes the systematic component of forecasting errors in forward transactions to systematic errors in market participants' expectations, at least within the sample. this phrase is intended to be as general as possible to incorporate issues such as problem weight, learning, adverse selection, and other types of error patterns that seem statistically significant within the sample. the definition does not necessarily have to imply that market participants are irrational in fact they are not so in the learning approach, for example, or in incorporating the assumption of problem weight. however, such an approach also opens the door to behavioural finance approaches. the strategic relevance of exchange rates for financial stability and exchange policies points the way to potential explanations based on the analysis of the credibility of these policies, in view of structural deviations that encourage destabilising carry trade activity. there is a wide recognition among practitioners of the three basic speculation models with a top-down approach to currency portfolios. first, the strategies based on the carry, the foundation of this work. second, momentum strategies, for which a basic rule of thumb is established: invest in those currencies that have appreciated in a given period or invest in those currencies in a basket with the best relative price performance. finally, there are strategies based on the valuation of currencies, based on models that allow us to estimate how reasonable the "price" of a currency is in relation to those other variables or effects that are relevant. these strategies are called "value" strategies. the first two strategies could be considered destabilizing speculation. instead, the third strategy produces stabilizing speculation. in many cases, the literature has worked on the analysis of combinations of these basic strategies. for example, jordá and taylor (2009) or barroso and santa-clara (2015). in our empirical work, for example, carry and momentum strategies will be mixed as a way of improving the performance derived exclusively from the carry. c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 126 the lack of connection with fundamentals that produces the destabilizing characteristics of the first two models was already recognized in a previous article by prado-domínguez and fernández-herraiz (2015). the expression of the carry to risk ratio was formally obtained from the sharpe ratio, showing its neutrality to valuation models, and applicability to carry trade strategies in foreign exchange markets. mounting trading activity based on carry specifications limits the volatility of the currency pair, so it feeds back the carry-to-risk variable, making it progressively more attractive, and therefore, more interesting for the agents involved. in other words, the very activity of carry trade increases the carry to risk (everything else constant) making it more attractive for a speculator who uses this ratio as a trigger for his trading activity. this activity introduces the well-known feedback phenomenon, which can also be found in other forex trading strategies such as momentum strategies (osler, 2012). this type of trading activity corresponds to the way in which this work defines destabilizing speculation. this is a type of speculation that reinforces the progressive detachment between currency valuation and currency price, introducing also a temporary perception of low volatility in the exchange rate. this process permeates the theme of all this work and is clearly reflected in the carry strategies. the destabilizing carry is a process in which a non-trivial set of traders take positions in the same direction, cumulatively over time, reducing volatility and producing a false sense of stability in the strategy. for us, this is the dynamic that usually sets the stage for instability scenarios of the kind that is regularly found in exchange rate crisis models. this type of speculation, much less noisy since it is part of a process of fragile complementarity such as those mentioned by gabor (2014), can be maintained for quite some time and is one of the main causes of the growing deviation between the value and price of the currency. that's why we think this is the real destabilizing speculation, the one that is not seen or, at least, attracts much less attention from policy makers and the media. it is only a matter of time before these developments end up in a currency crisis. the advantage of this explanation is that it supports the same solution for first-generation crisis models and second-generation crisis models in the style of the review discussed in, for example, krugman, rogoff, fisher and mcdonough (1999). this opens a possible reflection for central banks which would be as follows: are there options available to reduce the attractiveness of carry trade strategies, especially for those investors who are more leveraged, or those who notoriously operate at shorter maturities? there is no an easy answer to this question, but the prize is high. for example, reducing the social cost of maintaining international reserves. this true hedging strategy, as such, is subject to significant costs. rodrik (2006) estimates this cost at a loss of income equivalent to 1% of annual gdp. we will see that exchange policy credibility plays a key role in the ability of speculators to use carry trade strategies. for our empirical analysis we will focus on the mexican peso and the brazilian real, and how currency speculators act as potential free riders and not as agents which provide liquidity or ensure valuation stabilization. c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 127 3. the introduction of liquidity in carry trade strategies in foreign exchange. we introduce liquidity in the carry trade strategy including trend following indicators that trigger a trading orders for each of the specifications. the dynamic rules aim to avoid as far as possible the consequences of static carry trade strategies that are sprinkled by specific market events of certain virulence1. such strategies are compatible with the sale of catastrophic insurance on the market activity. the insurer, in our case the speculator in a carry trade strategy, will obtain profits that are not too high but relatively stable, dotted with periods of heavy losses. this is the idea that allows cochrane (1999) to venture the correct hypothesis that carry speculators are implicitly selling puts, that is, selling insurance. in fact, ilmanen (2012) shows that the sale of insurance and lottery tickets is very well paid in the markets, and, in particular, in the carry trade strategies. but going even beyond the ilmanen (2012) appreciation, if it is possible to obtain liquidity in our positions when the trend of positive results breaks down, then, following the simile, it would be possible to obtain a significant part of the "premiums" without fully assuming the cost of the insured "claims". our specifications of dynamic carry trade activities include two potential sources of return. first, the carry, the interest rate differential offered by the currencies. we use the 3-month interest rates2, capitalized over the period in which the position is held. secondly, the evolution in the price of the currency pair in which we are invested. this strategy assumes that the operator is financed at market prices in the financing currency and invests in the target currency, also at market prices. the first specification is a long only strategy. this specification acts as the benchmark and the passive strategy. the trader takes the position at the beginning of the period and makes no adjustments. therefore, this strategy could be a good benchmark for the other specifications. these would be the results of a passive strategy, product of taking exposure to this source of risk on the contrary, the two following specifications share one element in common. liquidity will be used opportunistically to enter and exit the strategy based on parameterized work rules. in our case, the rules will be based exclusively on the use of conventional moving averages. the trigger that will deliver the changes in the positioning of dynamic strategies will be the difference in moving averages between a very short-term average and a longer-term average3. this trigger will allow dynamic strategies to remain liquid temporarily or even take a "short" carry position. 1 such behaviour of carry strategies has often been described by professionals as "picking pennies in front of a rollercoaster" or as "going up by stairs and down the elevator". 2 our strategy is based on the implementation of a carry trade strategy, the most common in academic research, so we use short-term rates. the choice of the three-month rate from among other possible terms (1 month, 6 months, day or week, for example) makes it easier to compare the currencies used since all accounts have this reference in the period considered. because of the close relationship between shortterm rates, we do not believe that repeating the strategy with other rates will significantly alter the results. 3 we need an average with a sufficiently short term to resemble our daily prices. by default, we will use the 5-day average, which represents the trading week. for the long term average, and since we do not intend to make a trading strategy that involves a large amount of trading, we use the market year reference, i.e. 250 days. as this is not an optimised strategy, the results should be robust to changes in the averages used, provided the spirit of the strategy is maintained. c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 128 the first of the dynamic models will be called long-flat strategy, and it is a specification that will remain invested in the “peso” or the “real” until the trigger is activated. the position is then closed. the strategy is reactivated again when the moving averages reverse their order. when the carry strategy is deactivated, we consider that the operator does not obtain remuneration in excess of the monetary assets in dollars. in our view, this is a conservative approach to measure the performance. the third specification, also dynamic, will be called long-short strategy, and is built using the same instruments as the previous strategy, but with a modification of the trading criteria. in this case, when the trigger is activated the trade takes short positions, reversing the strategy towards a long position of usd and short position of the carry currencies. therefore, in this case, the long position closing trigger automatically activates a short position on the carry currency/long position in dollars. the trading motivation behind this third approach is to try to take advantage of the moments of deactivation of the long carry trade strategy, which are normally associated with high volatile market movements in the currencies considered. 4. empirical analysis. the empirical analysis focuses on the last eighteen years, a period from may 2000 until may 2018 and uses daily closing price data for the currencies considered: the mexican peso and the brazilian real. the performance of the carry trade strategies in each of the two individual currencies are presented and commented below. the outcome of the analysis will provide information about the evolution of the carry strategy, for each particular specification. we also present the breakdown of the returns of the carry strategy between those resulting from the appreciation/depreciation of the currencies in question (currency performance) and those resulting from obtaining the interest rate differential (carry component). the performance of the strategies will be measured on the basis of widespread performance indicators, such as the sharpe ratio, sortino ratio and calmar ratio, as well as information on the long term annualised return and volatility. 4.1 carry trade on the mexican peso. the mexican peso is undoubtedly the most liquid currency in the latam region. it is a freely convertible currency, which makes it a candidate for all types of carry strategies. our simplified carry model considers only the returns from the money market remuneration of the exposure to the peso, as well as the impact of changes in the price of the exchange rate pair. the performance of the simplified individual carry strategy in the mexican peso during the period offers little room for comfort to a carry trade speculator. the specification of dynamic strategies does not significantly improve the results of the long only carry. c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 129 table 1: results of the carry strategies on the mxn 5:2000-5:2018, daily data. mxnusd long only long flat long short average annual return 0,65% 0,89% 0,30% standard deviation 11,40% 6,97% 11,40% downside deviation 8,36% 5,06% 7,91% maximum drawdown 39,36% 19,26% 38,23% sharpe ratio 0,0573 0,1274 0,0266 sortino ratio 0,0782 0,1756 0,0384 calmar ratio 0,0166 0,0461 0,0079 source: authors elaboration from financial series obtained from bloomberg. table 1 shows that the annualised excess return on dollar-denominated risk-free assets offered by the strategies is less than 1%. this is a very low yield spread, particularly if we relate it to the volatility of the strategies, between 6.97% and 11.40% for the whole period. one way to look at these outcomes in aggregate is to check sharpe's ratios. with figures between 0.027 and 0.127, the strategies would have offered between 2 and 12 basis points of annual excess yield for every 10 points of volatility during this long period. low returns, high risk and poor performance. even the long flat strategy, characterised by a significant reduction in risk (6.97% volatility compared to 11.40% for the long only and long short strategies), does not provide maximum drawdown levels over the period (max dd) capable of withstanding substantial levels of leverage. the poor results are consistent under any of the performance measures (sharpe, sortino or calmar), in all cases very small. the long only strategy in the mexican peso over the past fifteen years has offered a performance which is apparently compatible with a sound management of the exchange rate policy, or a market perception of exchange credibility. taking the entire sample into account, the yield derived from the interest rate differential in favour of mexico is almost completely offset by the depreciation of the currency, something that is clearly observed in figure 1. : c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 130 figure 1. mxnusd long only carry. source: authors elaboration from financial series extracted from bloomberg. but the focus of our analysis is the dynamic specifications of carry trade speculation. the improvement in annualised returns obtained by the long–flat strategy is not significant, nor would it correspond to the expectations of a trader interested in speculating in the foreign exchange market. the results in terms of performance ratios are better than those of the long only strategy, but as mentioned above, these improvements are too poor to be considered seriously. in addition, the rule used does only partially prevent losses due to the depreciation of the peso, as can be seen in the figure 2. figure 2. mxnusd long flat carry. source: authors elaboration from financial series extracted from bloomberg. a raised key component of the active trading strategy issue is that the simplified trading rule employed does not improve the results of the portfolio. it is therefore logical that the results of the second dynamic strategy, the long-short strategy, which uses the same rule to trade the c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 131 long or short the carry currency, will not achieve significant success either. moreover, in the case of the long-short strategy, there is a negative effect due to the short position built on the strategy in the periods in which the trading rule reverses the positions. in those periods, the strategy is long the us dollar versus the mexican peso, so it yields negative carry returns. therefore, the potential outcome achieved by actively trading the currency is reduced by the negative interest rate differential. in view of the results, it does not seem likely that speculators have generally maintained a continuous destabilizing carry trade strategy throughout the period considered. it looks like mexican arrangements in terms of exchange policy stability are well in place to curb the impact of carry trade speculative strategies similar to those presented in the analysis. 4.2 carry trade on the brazilian real. the brazilian real is an exotic currency, since free convertibility is not allowed. however, brazil has a highly developed domestic financial market according to the regional standards. in this case, liquidity materialises more in the forward markets, specifically in non-derivable products. depending on the capabilities, reputation and scale of the speculator, on shore or off shore carry trade speculative activities could be put in place. our stylized carry model considers only the returns from the money market yield differential of the exposure to the real, as well as the impact of changes in the price of the exchange rate pair. table 2 indicates that the long only strategy in brazilian reals, unlike the case of the mexican peso, does offer attractive returns and performance for potential speculators in the sample period considered. table 2. results of the carry strategies on the brl 5:2000-5:2018, daily data. brlusd long only long flat long short average annual return 6,84% 9,79% 10,96% standard deviation 17,00% 11,05% 17,00% downside deviation 11,91% 7,71% 11,99% maximum drawdown 47,04% 17,04% 26,04% sharpe ratio 0,4024 0,8861 0,6449 sortino ratio 0,5744 1,2704 0,9139 calmar ratio 0,1455 0,5748 0,4210 source: authors elaboration from financial series obtained from bloomberg. the annualised excess return on dollar-denominated risk-free assets offered by the strategies was between 6.84% and 10.96%. it is important to bear in mind that we have the liquidity of the carry strategy and the possibility of using leverage opportunistically. remember c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 132 that the numbers presented in the summary table are considered for a fully funded position, so it would always be possible to increase the profitability and risk of the model. in fact, given that the sharpe ratio is invariant to leverage4, the speculator could decide on the level of market risk chosen, the volatility and, from there, estimate the expected return through the ratio. moreover, the results presented are 'in excess', so that this strategy could be built on a cash position in dollars which would also serve as the collateral for the carry strategy. the carry trade strategy on the brazilian real offers richer options to compare dynamic carry trade specifications, and the different opportunities and risks they offer to the speculator. we observe that active strategies that take advantage of the existing liquidity in the market, considerably improve the performance of the passive strategy. this improvement can be checked in the performance ratios, both in the consideration of symmetric risk (sharpe ratio) and in the consideration of asymmetric risk (sortino ratio) or in the specific analysis of the worst returns obtained (calmar ratio) during the sample period. figure 3 shows that the excess returns and the effect on the exchange rate of the currency pair have the right signs, i.e. the real is depreciated against the dollar over the period as a whole. however, this depreciation does not compensate for the high profitability offered by the carry component. figure 3. brlusd long only carry. source: authors elaboration from financial series extracted from bloomberg. it could be argued that the passive strategy performance is a natural expression of the required risk premium involved in buying brazilian reals. this risk premium for a long term period may signal a situation of permanent low exchange policy credibility. but the focus of our analysis are the dynamic specifications, and how shadow banking tools, liquidity and leverage, 4 the sharpe ratio is invariant to leverage when deposit rates and short-term financing rates are considered equal in the money markets. this is an unrealistic simplification for retail segments, but in the institutional market could be much closer to reality. in addition, if we use derivative instruments for position-taking, the risk-free rate applied in the replication portfolio equations normally considers, for simplicity, the same risk-free rate in the short term. c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 133 potentially creates better options for speculators if not properly handled by currency authorities. the long-flat strategy on brazilian real offers more attractive results than pure exposure to this asset class. the evidence during the sample supports the hypothesis that the opportunistic use of liquidity and the leverage offered by the market increase the potential returns for the strategy. a speculator may consider this strategy for its portfolio of available strategies. an annualised return in excess of 9.79%, with long term volatility around 11.05% over a period of 18 years, represents an attractive outcome. another point of interest of the strategy, which would encourage the use of leverage by an interested speculator, is the low level of maximum drawdown of the strategy, of only 17.04%. the returns are even more compelling when we consider that the strategy keeps cash over long periods of time without considering any excess return. any speculator would look for alternative uses of this liquidity during the “flat” periods. figure 4. brlusd long flat carry. source: authors elaboration from financial series extracted from bloomberg. the yield spread reaches a 7.15% annualised, a lower result than the long only strategy (11.10% annualised), which is systematically invested and therefore obtains a higher carry. this is the product of a lower exposure to the brazilian money market in the period, due to the application of the trading rule. this reduction is offset by a significant improvement in the impact on spot exchange rate losses. in fact, as can be seen in figure 4, the long-flat strategy in the brazilian real allows the speculator to achieve positive returns in the component derived from the evolution of the currency, with annualised returns of 2.76% which are added to the results of the carry. the specification provides an ideal adjustment case. the speculator keeps a comfortable exposure to the interest rate spread offered by the brazilian real, without incurring in high losses due to the depreciation of the exchange rate. a good example of ilmanen (2012) hypothesis about the options for speculators to capture a significant part of the "premiums" without fully assuming the cost of the insured "claims". the combination of both effects has a very positive impact on the performance ratios of the strategy. a sharpe ratio of almost 0.9, sortino ratio higher than 1 and a maximum drawdown of c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 134 only 1.54 times the volatility of the portfolio. since the trading rule significantly improves results compared to the long only strategy, it is reasonable to think that the long-short strategy can also offer attractive results. in the initial summary we saw how the longshort strategy offered results similar in volatility to the long only strategy, but with annualised returns that went from 6.84% to 10.96% in the sample period considered. not only that, but the maximum drawdown suffered by the strategy in the period is much lower (approximately 26%) or around half of the heavy losses of the long only strategy. given the results obtained, it is more than likely that there are traders working with refinements on these strategies within their portfolios. in contrast to the evidence obtained on the mexican peso, in the case of the brazilian real, the use of financing and market liquidity allows for comfortable speculation with an attractive expectation of returns versus risk. the leveraged strategies in this simplified trend following model offer flexibility of accommodation for both domestic-licensed and non-deliverable local-currency derivatives that are settled for differences. the question now is how to reconcile a potential risk premium for the “insurers” of the brazilian real with higher rewards for the “free riders”. the less the credibility of the exchange policy perceived by the market participants, the more the opportunities for dynamic destabilizing carry trade speculation activities. these strategies could turn to market financing (shadow banking) as a source of leverage and liquidity. the dynamic strategies presented do not rely on any model of currency valuation. if these strategies were common in the market, they will generate continuous scenarios of potential destabilizing speculation, in terms of persistent deviation from equilibrium. finally, note that neither the long-short strategy nor the long-flat strategy include any relation to critical variables such as the price of reference commodities, consideration of relevant variables for sovereign risk analysis or reference to a change in appetite for global or specific currency risk. nor do they incorporate a reflection on monetary policy scenarios, considerations on reference rate yield curves, analysis of the country's political strategy, or its capacity to impose certain fiscal or monetary rules. finally, the strategies do not use spot currency flow analysis techniques, or information on non-commercial activity in the currency futures, or information on flows in other related assets such as the local stock exchanges. both brazilian active strategies are agnostic to economic and market developments. this is important because it implies, for example, that the strategy develops its trading rules without reference to a hypothetical anchor of fundamental value of the currency. it could be that the model recommends being long on the currency in a situation of overvaluation, or that it recommends being short on a situation of potential undervaluation. it is in this sense that we affirm that carry strategies in foreign currencies can become destabilizing speculation, delegitimizing at this point the social role of the speculator. evidence presented for the sample period offers two different realities for the mexican peso and the brazilian real. the exchange rate policy of the mexican peso seems to be much more stringent in curving carry trade speculative opportunities. moreover, in view of the market c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 135 participants there seems to be a higher degree of exchange policy credibility attached to the mexican peso in terms of carry trade destabilizing speculation. 5. conclusions. we tested different specifications of carry trade dynamic strategies for the mexican peso and the brazilian real that produced substantially different long term outcomes. the performance of the different specifications signals the exchange policy credibility perceived by the market participants. trend following carry trade strategies can remain active for long periods of time depending on the credibility perceived by the market participants. speculation built on market financing allows participants to buy and sell without relying on currency valuation or other fundamental anchors or predictors. these types of strategies illegitimate the role of speculators, because they propel destabilizing instead of stabilizing speculation. the approach developed suggests that the exchange rate policy of the two currencies is substantially different in the sample period. in addition, our approach suggests that through the lens of different dynamic carry trade specifications it is possible to assess the perception of exchange policy credibility by market actors. the mexican peso offers close to zero returns, and poor performance metrics for the sample of the last 18 years. our analysis suggests that this situation is the outcome of credible exchange policies designed to curb general specifications of dynamic carry trade destabilizing speculation. the evidence provided on the brazilian real, on the contrary, signals potential attractive opportunities for long term dynamic carry trade strategies. the authors understand that the brazilian real is perceived by the market as a currency with a lower level of exchange policy credibility. in that sense, destabilizing speculation as defined in our paper has being possible during the sampling period. the evidence also suggests that the failure of the brazilian real exchange policy to detect and provide the tools to curb this type of speculation during the sample period creates the conditions to reinforce potential carry trade destabilizing activities. in fact, due to the leverage invariance property of the sharpe ratio, and taking the ex post results obtained by this strategy, we found that it could have been implemented in the period with much higher leverage. such leverage may be comparable to the examples by darvas (2008) in his reflection on optimal leverage in dollar-based speculative carry portfolios. it is not impossible that during the sample period a non-trivial number of agents have been pushed to develop such strategies, which could have compromised the integrity of the market and the financial stability of brazilian real denominated assets. this outcome implies strong exchange policy implications for the brazilian real. understanding the link between carry trade strategies and speculators behaviour is important for the correct design of optimal policies of intervention in currency markets. in fact, c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 136 one could also argue that the carry trade strategies, even if destabilizing in the short term, can help exert pressure on governments to conduct credible macroeconomic policies if properly internalized. future research on this topic may extend the scope of the analysis to other emerging market currencies and the way exchange credibility is perceived by the markets. other complementary metrics like the carry to risk ratio and its different specifications could also be explored to develop robustness checks or complementary analysis. references barroso, p., & santa-clara, p. (2015). beyond the carry trade: optimal currency portfolios. journal of financial and quantitative anaysis 50(5), 1037-1056. doi: https://doi.org/10.1017/s0022109015000460 cochrane, j. (1999). new facts in finance. nber working paper num. 7169. doi: https://doi.org/10.3386/w7169 darvas, z. (2008). leveraged carry trade portfolios. discussion papers, institue of economics, hungarian academy of sciences, mt-dp-2008/22.doi: https://doi.org/10.2139/ssrn.1102151 fama, e. (1984). forward and spot exchange rates. journal of monetary economics 14, 319338. doi: https://doi.org/10.1016/0304-3932(84)90046-1 financial stability board (2018). global shadow banking monitoring report. gabor, d. (2014). the imf´s rethink of global banks: critical in theory, orthodox in practice. governance 28(2), 199-218. doi: https://doi.org/10.1111/gove.12107 hansen, l., & hodrick, r. (1980). forward exchange rates as optimal predictors of future spot rates: an econometric analysis. journal of political economy 14(3), 319-338. doi: https://doi.org/10.1086/260910 ilmanen, a. (2012). do financial markets reward buying or selling insurance or lottery tickets? financial analysts journal 68(5), 26-36. doi: https://doi.org/10.2469/faj.v68.n5.7 jordá, o., & taylor, a. (2009). the carry trade and fundamentals: nothing to fear but feer itself. nber working paper num. 15518. doi: https://doi.org/10.3386/w15518 krugman, p., rogoff, k., fisher, s., & mcdonough, w. (1999). currency crises. in m. feldstein, international capital flows. chicago: university of chicago press, 421-470. meese, r., & rogoff, k. (1983). empirical echange models of the seventies. do they fit out of sample? the journal of international economics 14, 3-24. doi: https://doi.org/10.1016/00221996(83)90017-x osler, c. (2012). market microestructure and the profitability of currency trading. annual review of financial economics 4(1), 469-495. doi: https://doi.org/10.1146/annurev-financial110311-101726 https://doi.org/10.1017/s0022109015000460 https://doi.org/10.3386/w7169 https://doi.org/10.2139/ssrn.1102151 https://doi.org/10.1016/0304-3932(84)90046-1 https://doi.org/10.1111/gove.12107 https://doi.org/10.1086/260910 https://doi.org/10.2469/faj.v68.n5.7 https://doi.org/10.3386/w15518 https://doi.org/10.1016/0022-1996(83)90017-x https://doi.org/10.1016/0022-1996(83)90017-x https://doi.org/10.1146/annurev-financial-110311-101726 https://doi.org/10.1146/annurev-financial-110311-101726 c. fernández-herraiz et al. / european journal of government and economics 7(2), 123-137. 137 prado-dominguez, j., & fernandez-herraiz, c. (2015). a sharpe ratio based measure for currencies. european journal of goverment and economics 4(1), 67-75. doi: https://doi.org/10.17979/ejge.2015.4.1.4307 rodrik, d. (2006). the social cost of foreign exchange reserves. nber working paper 11952. doi: https://doi.org/10.3386/w11952 shiller, r. (2012). finance and the good society. new jersey: priceton university press. https://doi.org/10.17979/ejge.2015.4.1.4307 https://doi.org/10.3386/w11952 ejge_front_7_2 ejge oficial 7-2 content page 7-2-2 abstract. exchange credibility is a valuable asset for currency policymakers. in this article we intend to analyse exchange credibility from the perspective of carry trade speculators. global speculators' access to shadow banking financing allows them... keywords. currency carry trade, sharpe ratio, exchange policy credibility, shadow banking. doi. https://doi.org/10.17979/ejge.2018.7.2.4520 2. carry trade in the forex market and a brief reference to carry to risk. 3. the introduction of liquidity in carry trade strategies in foreign exchange. 4. empirical analysis. 4.1 carry trade on the mexican peso. 4.2 carry trade on the brazilian real. 5. conclusions. references female leadership in sports clubs vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 200-209 european journal of government and economics issn: 2254-7088 female leadership in sports clubs alfonso martínez-moreno a*, francisco cavas-garcía a, francisco cano-noguera a, arturo díaz-suárez a a universidad de murcia, spain * corresponding author at: dr.almamo@gmail.com article history. received 12 december 2019; first revision required 21 february 2020; accepted 21 march 2020. abstract. the objective of this study was to determine the type of leadership applied by the directors of amateur sports clubs. this quantitative, descriptive and cross-sectional study was carried out on a sample of 48 directors of sports clubs in total: 23 (47.91%) belonging to collective sports clubs and 25 (52.08%) to individual sports clubs. the study administered the mlq-x5 multifactor leadership questionnaire (45 items). the answers are collected on a likert scale ranging from 0 (never) to 4 (almost always). the results indicated that the directors opt for a transformational leadership style (3.24), preferably followed by the corrective leadership type (3.18) and almost the same level as the developer type (3.17). the study concluded that the collective sports directors opt for the style of transformational leadership by being proactive and promoting the achievement of extraordinary goals. keywords. equality; management; satisfaction; sport; women jel codes. l83; d71; d91 doi. https://doi.org/10.17979/ejge.2020.9.2.5840 1. theoretical foundation equality between men and women is established and recognized at the universal legal level by the united nations charter, the universal declaration of human rights and the united nations convention regarding the elimination of any form of discrimination against women (alfaro, 2012). spain has a gender gap of 74.6%, ranking 29th overall in the european union. the gender inequality index is an indicator of inequality. it unveils the areas in which important normative intervention is needed and promoting proactive thinking and public policies with the aim of overcoming the systematic disadvantages that women experience (undp, 2018). gender stereotypes continue and reinforce female and male professions (arenas, tabernero and briones, 2011). castells (1999) points out that a new society arises when a structural transformation can be observed in the production relations, in the power relations and in the experience relations. the masculinization of the masculine body and the feminization of the feminine body have led to a relationship of power and domination of the former over the latter, with masculinity occupying the central position between the sexes (gao, 2015). puleo (2005) points out the two theoretical approaches that point to the mechanisms of reproduction of the patriarchal system, coercion and consent. sex is something biological, while gender is a broader conception, for gálvez (2001) defines roles and values associated with each on the basis of historical and social elements. this concept is the one that sustains the stereotypes mailto:dr.almamo@gmail.com https://doi.org/10.17979/ejge.2020.9.2.5840 martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 201 and beliefs, culturally shared, that can limit leadership in women (contreras, pedraza and mejías, 2012). the main driving force of the empowerment process must come from the discriminated collective, that is, the political direction of the model cannot come from an external agent (castro, 2015). the situation of women in terms of decision making in sport is not very different from that of other sectors of society, although in the sports sector their representation is much lower than in the political or educational sector, being closer to the labor sector (alfaro, 2010). the studies carried out on sports management positions mark and point out a series of barriers and limits to the promotion of women in decision-making positions (soler, moragas and vilanova, 2018). social stereotypes that classify women in a role that does not correspond to the characteristics of leadership that society considers, so these positions are usually reserved for men (cuestas, 2014). there are endless barriers for women, which make it difficult for them to have access to a free labor market where they can and have access to managerial positions just like men, with the same opportunities within organizations (ibáñez, 2017). the organizational context may be one of the factors that hinder women's access to management positions insofar as, in most cases, it still "favors" or "promotes" certain leadership styles (generally "masculine") (charlo, and torrado, 2012). moragas and puig (2013) establish three key factors to limit the situation of women in sports management i) personal, ii) close environment, iii) sport organization. the different areas of activity in which women predominate may vary by location, but many of them are shared (alario, pascual and baraja, 2009). historically, sport is a masculine space both in its practice and in its management. this is embedded in the organizational structure of sport, which hinders women's access to leadership positions, excluding them from decision-making bodies and therefore making them invisible (alfaro, 2012). the sports practice rate among men and women who are aged between 25 and 34 is higher in men than in women (martín, moscoso, martínez del castillo and ferro, 2009). garcía (2011) defines leadership as the ability to influence others, but it is possible to identify large differences in the way leadership is exercised. pautt (2011) talks about the transdisciplinarity of the concept to express that the study of leadership is carried out from different scientific disciplines with their own perspective, such as anthropology, sociology, psychology, etc. for avolio (2007), leadership is a process of influence, which needs to analyze at least three elements that are essential for this process of influence to occur. i) the role of leader. very often, this individual is able to transmit a vision to his followers about a desired future state. in order for this leader to achieve this vision, he needs to reformulate it into a communicable mission based on concrete actions; ii) the followers of a leader, who will apply (or not) a certain level of effort to achieve that mission; iii) the context and the means by which leaders and followers establish links and relations of social exchange (avolio, walumbwa and weber, 2009; yammarino, dione, schriesheim and dansereau, 2008). the transformational leadership theory (bass, 1985) studies how leadership styles help transform organizations and the human capital that forms them. bass and avolio (1997) differentiate four elements or dimensions that respond to transformational leadership (behavior, attribution, inspirational motivation and intellectual stimulation, two dimensions that make up transactional leadership martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 202 (individual consideration and contingent reward) and a last element that makes up laissez-faire. transactional leadership is identified by the exchange between the leader and the members of his or her group who recognize him or her as leader and accept his or her authority. however, the leader, in return, must provide incentives to the group. perilla and gómez (2017) determine that transformational leadership can especially affect the emotional well-being of employees and the prevention of physical disorders associated with health. the transformational style manages to change attitudes of the followers, infusing the importance and value of the results of the task, as well as the idea of a team obtaining a higher performance than expected and an increase in the satisfaction levels (aguilar-luzón, calvo-salguero and garcía-hita, 2007). according to druskat (1994), the type of traditional organization is not decisive when studying the role of gender in leadership. cuadrado, navas and molero (2004) determine that in general, the “masculine” and “feminine” organizations are not directed using typical masculine and feminine leadership styles. although women at any level of management can "perceive" the glass ceiling, this metaphor is often used to explain the barriers they experience in high-level positions. (charlo, and torrado, 2012). the organizational context may be one of the factors that hinder women's access to management positions cuadrado, navas, and molero (2004). the glass cliff determines that women access managerial positions, largely when there are in a crisis situation (ryan, hersby and bongiorno, 2011). the transformational leadership style "is most desirable in periods characterized by fear, confusion and uncertainty" (monfort, 2015). the characteristics of authentic leadership are closer to the traits that are generally associated with women (ibáñez, 2017). it is necessary to know if one of the possible causes of the under-representation of women in leadership positions is the leadership style they adopt (cuadrado, molero and navas, 2003). growing global competition, the emergence of new markets and rapid technological development have led to the need to generate quick and accurate responses to ensure the survival of blanch, gil, antino, and rodríguez-muñoz organizations (2016). companies need workers, especially in managerial positions, with empathy, understanding, who can communicate and provide help (gartzia, 2011), behaviours inherent to women. women who are competent in their work are often disapproved personally and socially (monzani, hernandez, dick and peiró, 2015). in the current process of change, traditional models of work organization are outdated and not very profitable (gómez and sánchez, 2009). gender and leadership continue to be of great interest, especially given the low representation of women in leadership positions (kapasi, sang, and sitko, 2016). montes and roca (2016) suggest that women are effective, creative, transformational and democratic leaders, characteristics that define the ideal leader of the 21st century. when women have to play the role of leader, they present certain characteristics that make them more appropriate leaders for contemporary organizations (eagly and karau, 2002). the challenge for modern organizations, and for women within them, is to recognize and empower difference as a source of wealth (contreras, pedraza and mejías, 2012). precisely, it is necessary to broaden and continue with the studies from the feminist martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 203 perspective in this field, as well as to revise the current sports legislation (leruite, martos and zabala, 2015). therefore, the objective of this study is to determine the type of leadership applied by the directors of amateur sports clubs in both collective and individual teams. 2. method 2.1. sample the reason for being included in the investigation was to be an active part of the management team of a sports club at the time of the study. the final sample consisted of 48 directors of sports clubs, 23 of them (47.91%) belonging to collective sports clubs and 25 (52.08%) to individual sports clubs. of those, 19 (39.58%) have studies of bup/baccalaureate/higher degree cycles, 3 (6.25%) are graduates in physical education, one (2.08%) has primary studies, 25 (52.08%) have university studies, 9 (18.75%) are in physical education and 16 (33.53%) in other degrees. they play the role of president 11 (22.9%), vice president 4 (8.3%), secretary 10 (20.8%), treasurer 8 (16.7%), member 9 (18.8%) and perform other functions 6 (12.5%). the participants were accidentally selected kerlinger, (2001), being a non-probabilistic sample (hernández, fernández and baptista, 2000). • instrument the multifactor leadership questionnaire (mlq-5x, bass and avolio, 1995) was administered in the factorial version and in spanish by molero, recio and cuadrado (2010). mlq-5x (short form) consists of 45 items, 36 items that respond to the different types of leadership: transformational leadership -behavior, attribution, inspirational motivation and intellectual stimulation-; developmental leadership -individual consideration, contingent reward-; corrective leadership -active exception-; passive leadership -passive exception and laissez-faire-. the remaining nine items evaluate organizational outcome variables 4 items on the leader's effectiveness, 2 on the subject's satisfaction with his/her own methods and work, 3 on the extra effort that influences his/her employees. the answers are collected on a likert scale with a range from 0 (never) to 4 (almost always), which indicates that the higher the scores are, the more they are in line with the type of leadership measured. • procedure after contacting the different sports clubs their participation in the investigation was accepted. they were given an informed consent that specified the objectives of the study, as well as the voluntary, confidential and anonymous nature of their participation. an action protocol was developed to ensure that data collection was similar for all research participants. the questionnaire was self-administered, with an interviewer being present in case the participant wanted to ask any questions that appeared during the process of the questionnaire. martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 204 3. results table 1 shows the descriptions of the type of leadership based on the classification of the sport whether it is collective or individual. the types of leadership that directors determine to apply to a greater extent globally are respectively transformational (3.24), corrective (3.18) and developer (3.17). the passive leadership (1.05) is the one that has obtained the lowest score, being the style that is applied on fewer occasions. as for the type of sport, there are no significant differences as to whether it is collective or individual. the highest average scores, differentiating by type of sport, are reached by the developer (3.27) and corrective (3.26) leadership in individual sports and the transformational (3.24) collective leadership. as for the behaviors that make up each leadership type, both collective (3.37) and individual (3.39) sports behaviors reach the highest values. the passive exception (1.07) in collective sports and the laissez faire (1.00) in individual sports are those that have obtained the lowest scores; therefore, the directors determine that they are the ones that are applied the least. the results also show that they encourage effort, efficiency and satisfaction. table 1. description of the type of leadership depending on the type of sport: collective-individual. total type of sport p-value collective individual transformational_leadership 3.24 (0.4) 3.24 (0.44) 3.24 (0.37) 0.98 behavior 3.38 (0.54) 3.37 (0.52) 3.39 (0.56) 0.948 attribution 3.24 (0.5) 3.22 (0.55) 3.26 (0.47) 0.773 inspirational_motivation 3.28 (0.5) 3.35 (0.48) 3.21 (0.53) 0.337 intellectual_stimulation 3.08 (0.53) 3 (0.57) 3.16 (0.48) 0.315 developer_leadership 3.17 (0.44) 3.07 (0.46) 3.27 (0.39) 0.134 individual_consideration 3.04 (0.63) 2.91 (0.69) 3.18 (0.55) 0.156 contingent_reward 3.28 (0.39) 3.23 (0.37) 3.32 (0.42) 0.417 corrective_leadership 3.18 (0.6) 3.1 (0.68) 3.26 (0.51) 0.37 active_exception 3.18 (0.6) 3.1 (0.68) 3.26 (0.51) 0.37 passive_leadership 1.05 (0.8) 1.08 (0.89) 1.02 (0.73) 0.799 passive_exception 1.05 (0.9) 1.07 (1.07) 1.04 (0.72) 0.929 laissez_faire 1.05 (0.84) 1.1 (0.88) 1.00 (0.82) 0.695 effort 3.23 (0.6) 3.19 (0.52) 3.26 (0.68) 0.672 efficacy 3.12 (0.58) 3.07 (0.51) 3.18 (0.65) 0.516 satisfaction 3.18 (0.56) 3.07 (0.63) 3.29 (0.46) 0.165 4. discussion the present investigation has consisted of analyzing, in a sample of 48 directors of amateur sports clubs, the leadership styles that they apply to their subordinates. in relation to the style of collective sports directives, we agree with martínez-moreno, morales, and angosto, (2018), where cruise crew members presented a higher level of transformational leadership, also with pacsi, estrada, pérez, and cruz (2014) who consider transformational leadership to be the most appropriate; with ruiz (2016) where transformational martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 205 leadership was predominant in their study. as with rozo and abaunza (2010), although the sample was of nurses, they exercise transformational leadership in a stronger way. with gonzález, gómez, and figueroa (2019) where they obtained an acceptable and significant correlation between transformational leadership in relation to trust with the leader, and psychological well-being at work. with martinez, rodriguez, navea, and rojas (2016) where the leader perceives himself mostly transformational, in his research on the leadership style in a brazilian cosmetics company. with martínez (2014) in the direction of an educational center, also with eagly, johannensen-schmidt and van engen (2003), which indicate that women show a more transformational style, with charlo, and torrado (2012) there is an evolution towards flexible and participative innovative management styles, which incorporate the qualities associated with transformational leadership, as well as with part of the sample from the study of cuadrado, molero, and navas (2003). corroborate our data degroot, kiker, and cross (2000), as well as judge and piccolo (2004), who, in their meta-analysis, find positive correlations of transformational leadership with leader effectiveness, execution and subordinate effort, as well as dumdum, lowe and avolio (2002) with effective leadership in relation to transformational leadership, also cáceres, trujillo, hinojo, aznar, and garcía (2012) when talking about a possible female leadership style, with multiple elements of transformational leadership, also cuadrado, and molero (2002), where women tend to score higher on all transformational leadership factors. we also agree with saravia (2014) that the intellectual stimulation dimension, both globally (3.08). as in team sports (3) and in individual sports (3.16), it represents the lowest degree among those that make up the transformational leadership style. regarding the style of the individual sports directors, we agree with the study of ruíz (2016) where the style that achieves the highest score is the developer (2.87) although that is far from those (3.27) that have an average in our study. our results are in line with pedraja-rejas, rodríguez-ponce, delgado-almonte, and rodríguez-ponce (2006), who find that transactional leadership was predominant in their research on small businesses. we also obtain similar scores to the study of turrado (2016) in terms of efficacy (3.12) satisfaction (3.18) and effort (3.23). we also agreed with cuadrado, molero, and navas (2003) in their study, with the women in their sample. 5. conclusions the objective of this study has been to determine the type of leadership applied by the directors of amateur sports clubs in both collective and individual teams. the empirical evidence allows us to conclude that the directors of individual sports clubs in the sample, who constitute the subject of study, encourage exchanges to strengthen their work through rewards. the directors of collective sports, included in the sample, opt for the style of transformational leadership. so, they are proactive, making collective interests prevail, and promoting the achievement of extraordinary goals. finally, some biases such as the innate social desirability of self-evaluation can weaken the martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 206 results. it is necessary to carry out this type of research, all the more so when the glass ceiling, although very slowly, is disappearing in the field of sport and more specifically in sports management. this allows us to determine and know better the leadership styles that apply the directives. however, analysing the subordinates would provide clues to the issue at hand. references aguilar-luzón, m. d. c., calvo-salguero, a., and garcía-hita, m. á. (2007). valores laborales y percepción del estilo de liderazgo en personal de enfermería. salud pública de méxico, 49, 401-407. https://doi.org/10.1590/s0036-36342007000600006 alario, m., pascual, h., and baraja, e. (2009). segregación laboral de las mujeres en la industria rural: el caso de castilla y león. boletín de la agen º, 50, 173-192. alfaro, e. (2010). las mujeres y el deporte: evolución y situación actual. seminario permanente mujer y deporte. fac. cc actividad física y deporte–inef. universidad politécnica de madrid. retrieved from: http://www.munideporte.com/imagenes/documentacion/ficheros /20100805132024elida%20alfaro.pdf alfaro, é. (2012). el liderazgo de las mujeres en la dirección y gestión del deporte. en m. j. mosquera (ed.), i e ii ciclo de conferencias: xénero, actividade física e deporte (pp. 31-49). a coruña: universidad de a coruña. arenas, a., tabernero, c., and briones, e. (2011). ¿qué determina el desempeño en la toma de decisiones de hombres y mujeres? revista de psicología del trabajo y de las organizaciones, 1 (vol. 27), 25-26. https://doi.org/10.5093/tr2011v27n1a6 avolio, b. j. (2007). promoting more integrative strategies for leadership theory building.american psychologist, 62, 25–33. https://doi.org/10.1037/0003-066x.62.1.25 avolio, b. j., walumbwa, f. o., and weber, t. j. (2009). leadership: current theories, research, and future directions. annual review of psychology, 60, 421–49. https://doi.org/10.1146/annurev.psych.60.110707.163621 bass, b. m., and avolio, b. j. (1995). mlq multifactor leadership questionnaire: sampler set. redwwod city. c. a.: mind garden. https://doi.org/10.1037/t03624-000 bass, b. m., and avolio, b. j. (1997). revised manual for the multifactor leadership questionnaire. mind garden, palo alto, ca. bass, b.m. (1985). leadership and performance beyond expectations. nueva york: free press. blanch, j., gil, f., antino, m., and rodríguez-muñoz, a. (2016). modelos de liderazgo positivo: marco teórico y líneas de investigación. papeles del psicólogo, 37(3), 170–176. cáceres, m. p., trujillo, j. m., hinojo, f. j., aznar, i., and garcía, m. (2012). tendencias actuales de género y el liderazgo de la dirección en los diferentes niveles educativos. educar, 48(1), 69–90. https://doi.org/10.5565/rev/educar.36 castells, m. (1999). la sociedad red. la era de la información: economía, sociedad y cultural. madrid: alianza editorial. castro, t. (2015). itinerarios de poder y liderazgo en mujeres extremeñas. una investigación desde la metodología comunicativa crítica. https://doi.org/10.1017/cbo9781107415324 .004 charlo, m. j., and torrado, m. n. (2012). la mujer directiva en la gran empresa española: perfil, competencias y estilos de dirección. estudios gerenciales, 28(124), 87–105. https://doi.org/10.1016/s0123-5923(12)70217-6 contreras, f., pedraza, j. e., and mejía, x. (2012). la mujer y el liderazgo empresarial. diversitas, 8(1), 183. https://doi.org/10.15332/s1794-9998.2012.0001.12 cuadrado, i., and molero, f. (2002). liderazgo transformacional y género: autoevaluaciones de directivos y directivas españoles. revista de psicologia del trabajo y de las organizaciones, 18(1), 39–55. retrieved from http://www.uned.es/474031/rpwo.pdf cuadrado, i., molero, f., and navas, m. (2003). el liderazgo de hombres y mujeres: diferencias en estilos de liderazgo, relaciones entre estilos y predictores de variables de resultado organizacional. acción psicológica, 2 (2), 115–129. https://doi.org/10.5944/ap.2.2.525 https://doi.org/10.1590/s0036-36342007000600006 http://www.munideporte.com/imagenes/documentacion/ficheros/20100805132024elida%20alfaro.pdf http://www.munideporte.com/imagenes/documentacion/ficheros/20100805132024elida%20alfaro.pdf https://doi.org/10.5093/tr2011v27n1a6 https://doi.org/10.1037/0003-066x.62.1.25 https://doi.org/10.1146/annurev.psych.60.110707.163621 https://doi.org/10.1037/t03624-000 https://doi.org/10.5565/rev/educar.36 https://doi.org/10.1017/cbo9781107415324.004 https://doi.org/10.1017/cbo9781107415324.004 https://doi.org/10.1016/s0123-5923(12)70217-6 https://doi.org/10.15332/s1794-9998.2012.0001.12 http://www.uned.es/474031/rpwo.pdf https://doi.org/10.5944/ap.2.2.525 martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 207 cuadrado, m. i., navas, m. s., and molero, f. (2004). el acceso de las mujeres a puestos directivos: género, contexto organizacional y estilos de liderazgo. revista de psicología general y aplicada: revista de la federación española de asociaciones de psicología, 57(2), 181–192. retrieved from: http://dialnet.unirioja.es/servlet/articulo?codigo=997864& info=resumen&idioma=spa cuestas, a.p. (2014). escuela de administración: memorias del seminario de profundización en liderazgo. ”el rol de la mujer en el liderazgo”. universidad del rosario http://repository.urosario.edu.co/handle/10336/11556 degroot, t., kiker, d. s. and cross, t. c. (2000) a meta-analysis to review organizational outcomes related to charismatic leadership. canadian journal of administrative sciences, 17, 356-371. https://doi.org/10.1111/j.1936-4490.2000.tb00234.x druskat, v. u. (1994). gender and leadership style: transformational and transactional leadership in the roman catholic church. leadership quarterly, 5(2), 99-119. https://doi.org/10.1016/1048-9843(94)90023-x dumdum, u. r., lowe, k. b. and avolio, b. j. (2002). a meta-analysis of transformational and transactional leadership correlates of effectiveness and satisfaction: an update and extension. in: b. j. avolio and f. j. yammarino (eds.), transformational and charismatic leadership: the road ahead (pp. 35-66). london: jai press. eagly, a. h., and karau, s. j. (2002). role congruity theory of prejudice toward female leaders. psychological review, 109, 573-598. https://doi.org/10.1037/0033-295x.109.3.573 eagly, a. h., johannesen-schmidt, m. c., and van engen, m. (2003). transformational, transactional, and laissez-faire: a meta-analysis comparing men and women. psychological bulletin, 129, 569-591. https://doi.org/10.1037/0033-2909.129.4.569 gálvez, t. (2001). aspectos económicos de la equidad de género. cepal. gao, q. (2015). bourdieu and body. en l. hunter, w. smith, e. emerald (eds.), pierre bourdieu and physical culture (pp. 143-148). london: routledge. garcía, m. (2010). in paidós (ed.), aprender a liderar. barcelona: paidós garcía, m. (2011). liderazgo transformacional y la facilitación de la aceptación al cambio organizacional. pensamiento psicológico, 6 (9), 41-54. gartzia, l. (2011). empresas del siglo xxi: de pensar en “masculino” a pensar en “andrógino”. diputación foral de gipuzkoa, 7-100. gómez, j., and sánchez, m. (2009). la participación de la mujer en los consejos de administración de empresas del ibex-35. revista de humanidades, 16, 105–140. gonzález, m. b., gómez, j. g. i., and figueroa, e. g. (2019). análisis correlacional del bienestar psicológico en el trabajo, liderazgo transformacional y confianza en el líder: caso de una mype industrial. ricea. revista iberoamericana de contaduría, economía y administración, 8(16), 1–22. https://doi.org/10.23913/ricea.v8i16.139 hernández, r., fernández, c., and baptista, p. (2000). metodología de la investigación. méxico: mc graw hill. ibáñez, m. (2017). mujeres y liderazgo auténtico. ¿podría ser una salida para lograr la igualdad de género? retrieved from: http://repositori.uji.es/xmlui/handle/10234/168744 judge, t. and piccolo, r. (2004). “transformational and transactional leadership: a metaanalytic test of their relative validity”. journal of applied psychology. vol.89, no. 5, 755768. https://doi.org/10.1037/0021-9010.89.5.755 kapasi, i., sang, k. j. c., and sitko, r. (2016). gender, authentic leadership and identity: analysis of women leaders’ autobiographies género, liderazgo autentico e identidad: análisis de las mujeres lideres autobiografías. gender in management: an international journal, 31(5/6), 339–358. https://doi.org/10.1108/gm-06-2015-0058 kerlinger, f., and howard, b. (2001). investigación del comportamiento. métodos de investigación en ciencias sociales. méxico: mc graw hill/interamericana. español de competición desde la perspectiva de protagonistas clave. retos, 28, 3-8. martin, m., moscoso, d., martínez del castillo, j., and ferro, s. (2009). la práctica deportiva en españa: desigualdades entre mujeres y hombres en la población joven, adulta y mayor. fuentes, 30, 43. http://dialnet.unirioja.es/servlet/articulo?codigo=997864&info=resumen&idioma=spa http://dialnet.unirioja.es/servlet/articulo?codigo=997864&info=resumen&idioma=spa http://repository.urosario.edu.co/handle/10336/11556 https://doi.org/10.1111/j.1936-4490.2000.tb00234.x https://doi.org/10.1016/1048-9843(94)90023-x https://doi.org/10.1037/0033-295x.109.3.573 https://doi.org/10.1037/0033-2909.129.4.569 https://doi.org/10.23913/ricea.v8i16.139 http://repositori.uji.es/xmlui/handle/10234/168744 https://doi.org/10.1037/0021-9010.89.5.755 https://doi.org/10.1108/gm-06-2015-0058 martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 208 martínez-moreno, a., morales, v., and angosto, s. (2018). sailing: the crew leadership. journal of human sport and exercise, 13(4), 766–775. https://doi.org/10.14198 /jhse.2018.134.05 martínez, a., rodríguez, h., navea, f., and rojas, c. (2016). identificación del estilo de liderazgo en una empresa brasileña de la industria cosmética. ciencia amazónica (iquitos), 6(1), 40–47. https://doi.org/10.22386/ca.v6i1.107 martínez, y. a. (2014). el liderazgo transformacional en una institución educativa pública. educación, xxiii, 7–28. molero, a., recio, p., and cuadrado, i. (2010). liderazgo transformacional y liderazgo transaccional: un análisis de la estructura factorial del multifactor leadership questionnaire (mlq) en una muestra española. psicothema, 22, 495-501. monfort, r. (2015). una revisión sobre mujeres y liderazgo: ¿los tiempos están cambiando? retrieved from: http://repositori.uji.es/xmlui/bitstream/handle/10234/133445/tfg_monfort+ carrasco_raquel.pdf?sequence=1 montes, c., and roca, d. (2016). el liderazgo femenino en la creatividad publicitaria liderança feminina na criatividade publicitaria. cuadernos. info, 39, 113–131. https://doi.org/10.7764/cdi.39.1039 monzani, l., hernandez, a. s., dick, r. and peiró, j. m. (2015). the synergistic effect of prototypicality ans authenticcity in the relation between leaders’ biological gender and their organizational identification, 132:737–752. https://doi.org/10.1007/s10551-0142335-0 moragas, m., and puig, n. (2013). las mujeres que presiden los clubes deportivos en catalunya. análisis de los factores que inciden en el acceso a os puestos de dirección: resultados preliminares. en j. aldaz, a. dorado, p. j. jiménez, a. vilanova (eds.), libro de actas del xii congreso aeisad. responsabilidad social, ética y deporte (pp. 65-74). madrid: ibersaf. pacsi, a. y., estrada, w., pérez, a., and cruz, p. (2014). liderazgo laissez faire. revista de investigación de administración, 1(1996), 67–72. pautt, g. (2011). liderazgo y dirección: dos conceptos distintos con resultados diferentes. revista facultad de ciencias económicas: investigación y reflexión, 19(1), 213-228. https://doi.org/10.18359/rfce.2269 pedraja-rejas, l., rodríguez-ponce, e., delgado-almonte, m., and rodríguez-ponce, j. (2006). transformational and transactional leadership: a study of their influence in small companies liderazgo transformacional y transaccional: un estudio de su influencia en las pequeñas empresas. ingeniare -revista chilena de ingeniería, 14(2), 159–166. https://doi.org/10.4067/s0718-33052006000100010 perilla, l. e. and gómez, v. (2017). relación del estilo de liderazgo transformacional con la salud y el bienestar del empleado: el rol mediador de la confianza en el líder. revista de psicología del trabajo y de las organizaciones, 33(2), 95-108. https://doi.org/10.1016/j.rpto.2017.02.005 puleo, a. (2005). el patriarcado: ¿una organización social superada? temas para el debate, (133), 39-42. rozo, s. m., and abaunza, m. (2010). liderazgo transaccional y transformacional. avances en enfermería, 28(2), 62–72. retrieved from: http://search.ebscohost.com/ login.aspx?direct=true&db=lth&an=65553984&lang=es&site=ehost-live ruíz, s. (2016). liderazgo transformacional y transaccional: análisis a través de la variable de género. revista iberoamericana de ciencias, 1, 30-45. ryan, k. m., hersby, d., and bongiorno, r. (2011). think crisisthink female: the glass cliff and contextual variation in the think manager-think male stereotype. journal of applied psychology, 3 (96), 470-484. https://doi.org/10.1037/a0022133 saravia, e. (2014). caracterización de estilos de liderazgo mediante el análisis de redes sociales. journal of business, 66(11), 30–42. https://doi.org/10.21678/jb.2014.65 soler, s., moragas, m., and vilanova, a. (2018). the voices of female chairs of sports clubs. en a. elling, j. hovden, a. knoppers (eds.), gender diversity in european sport governance (pp. 102-107). new york: routledge. https://doi.org/10.14198/jhse.2018.134.05 https://doi.org/10.14198/jhse.2018.134.05 https://doi.org/10.22386/ca.v6i1.107 http://repositori.uji.es/xmlui/bitstream/handle/10234/133445/tfg_monfort+carrasco_raquel.pdf?sequence=1 http://repositori.uji.es/xmlui/bitstream/handle/10234/133445/tfg_monfort+carrasco_raquel.pdf?sequence=1 https://doi.org/10.7764/cdi.39.1039 https://doi.org/10.1007/s10551-014-2335-0 https://doi.org/10.1007/s10551-014-2335-0 https://doi.org/10.18359/rfce.2269 https://doi.org/10.4067/s0718-33052006000100010 https://doi.org/10.1016/j.rpto.2017.02.005 http://search.ebscohost.com/login.aspx?direct=true&db=lth&an=65553984&lang=es&site=ehost-live http://search.ebscohost.com/login.aspx?direct=true&db=lth&an=65553984&lang=es&site=ehost-live https://doi.org/10.1037/a0022133 https://doi.org/10.21678/jb.2014.65 martínez-moreno et al. / european journal of government and economics 9(2), july 2020, 200-209 209 turrado, m. (2016). grado universitario en relaciones laborales y recursos humanos facultad de ciencias del trabajo ¿hombres y mujeres utilizan diferentes estilos de liderazgo? retrieved from: https://buleria.unileon.es/bitstream/handle/10612/10182/ turrado%20turrado%2c%20mar%c3%ada.pdf?sequence=1 undp (united nations development program), 2018. retrieved from: http://hdr.undp.org/en/2018-update/download yammarino, f. j., dionne, s. d., schriesheim, c. a., and dansereau, f. (2008). authentic leadership and positive organizational behavior: a meso, multi-level perspective. the leadership quarterly, 19, 693–707. https://doi.org/10.1016/j.leaqua.2008.09.004 https://buleria.unileon.es/bitstream/handle/10612/10182/turrado%20turrado%2c%20mar%c3%ada.pdf?sequence=1 https://buleria.unileon.es/bitstream/handle/10612/10182/turrado%20turrado%2c%20mar%c3%ada.pdf?sequence=1 http://hdr.undp.org/en/2018-update/download https://doi.org/10.1016/j.leaqua.2008.09.004 ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5840_g vol.8 • no.2 2019 issn: 2254-7088 european journal of government and economics 8(2), december 2019. european journal of government and economics    issn: 2254‐7088  number 8, issue 2, december 2019 doi: https://doi.org/10.17979/ejge.2018.8.2 why regions fail (or succeed). the role of government institutions in the long-run 114-144 doi: https://doi.org/10.17979/ejge.2019.8.2.4989 filippo bonanno restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case 145-160 doi: https://10.17979/ejge.2019.8.2.5478 joão ricardo catarino and ricardo de moraes e soares fdi in selected developing countries: evidence from bundling and unbundling governance 161-188 doi: https://doi.org/10.17979/ejge.2019.8.2.4970 simplice asongu the relationship between corporate tax rate and economic growth during the global financial crisis: evidence from a panel var governanceand domestic investment in africa 189-202 doi: https://10.17979/ejge.2019.8.2.5074 gamze oz-yalaman the impact of crd iv on bank lending 203-217 doi: https://doi.org/10.17979/ejge.2019.8.2.4656 matias huhtilainen european journal of government and economics 8(2), december 2019, 189-202 european journal of government and economics issn: 2254-7088 the relationship between corporate tax rate and economic growth during the global financial crisis: evidence from a panel var gamze oz-yalamana* aeskisehir osmangazi university, eskishehir, turkey. * corresponding author at: gamze.ozyalaman@gmail.com article history. received 17 january 2019; first revision required 10 august 2019; accepted 11 september 2019. abstract. this paper compares dynamic relationship between economic growth and corporate tax rate during the recent financial crisis and the non–crisis period using a panel var for 29 oecd countries over the period 1998-2016. the results show that corporate tax rate has a significantly negative effect on economic growth. moreover, the recent financial crisis has had a significant effect on the endogenous interaction between corporate tax rate and economic growth. according to granger causality test, there is only one-way causality from corporate tax rate to economic growth during the non-crisis period. interestingly, there are not any causal relationships between corporate tax rate and economic growth during the crisis period. the results show that the recent crisis has had a significant effect on the endogenous interaction between corporate tax rate and economic growth. key words: corporate tax, oecd countries, panel var. jel code: h25, f23 doi. https://10.17979/ejge.2019.8.2.5074 1. introduction taxation is not only intended to increase the funds required for state expenditure, but it also might contribute to the income distribution, economic stability, resource allocation and economic growth. in fact, with respect to these areas, the impact of taxation on economic growth has received considerable investigation in the literature. in order to better understand the relationship between tax structure and economic growth, it is important to recognize the driving forces of economic growth. in neoclassical framework, in solow’s model (1956), economic growth is determined by physical capital and human capital, and the only factor that can be accumulated is physical capital. both capital and labor income taxes reduce the steady-state level of income, but they have only transitory effects on growth. indeed, taxes have no effect on long-run growth rates. harberger’s (1964) superneutrality conjecture also contends that the tax policy is an ineffective instrument to influence growth. however, according to endogenous growth theory, both physical capital and human capital are reproducible factors and human capital accumulation is seen as a key factor for economic growth. romer (1986) points out that knowledge, which can be seen as an investment on human capital, creates positive externalities. thus, knowledge is assumed to be an input in production that increases marginal g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 190 productivity. in these models, taxation on both physical capital and human capital affect economic growth negatively (lucas, 1990; trostel, 1993). a wide range of further studies support the common findings in endogenous growth literature which suggests that taxation has effects on economic growth. for example, helms (1985) focuses on how state and local taxes effect economic growth and concludes that increase of the state and local tax retard economic growth significantly. koester and kormendi (1989) analyze the effects of average and marginal tax rates on economic growth. their results show that there is a significantly negative relationship between both average and marginal tax rates and economic growth over 63 countries. in addition to these studies, there is a mass of literature that empirically investigates the relationship between economic growth and taxation (king and rebelo, 1990; lucas, 1990; engen and skinner, 1996; gober and burns, 1997; lee and gordon, 2005; dackehag and hansson, 2012; adkisson and mohammed, 2014; stoilova, 2017). on the other hand, the empirical findings of these studies for the presence and direction of this relationship are controversial. for example, some studies support the existence of both negative (lee and gordon, 2005; dackehag and hansson, 2012) and positive (gober and burns, 1997; stoilova, 2017) relationships, while some studies do not support the existence of any relationship between taxation and economic growth at all (stokey and rebelo, 1995; mendoza et al., 1997). these studies also focus on the relationship between different types of taxation and economic growth. some of them focus on personal income tax (pecerino, 1994; gober and burns, 1997; wildmalm, 2001), while some other focus on sales tax or property tax (gober and burns, 1997; ojede and yamarik; 2012; stoilova, 2017). also, there are studies investigating the relationship between corporate tax rates and economic growth (lee and gordon, 2005; dackehag and hansson, 2012; adkisson and mohammed, 2014). considering the tax deduction discussions on corporate tax in policy circles, understanding the connection between corporate tax rate and economic growth is very important. however, there are still few studies comparing the effect of corporate tax rate on economic growth. thus, this paper contributes to the literature by exploring the corporate taxes as one of the robust determinants of economic growth. we use a larger sample and a more comprehensive set of variables. for example, the countries employed in this paper approximately produce 46% of the world gdp. moreover, it is clear that financial crises have a significant effect on economic structure. thus, this paper also contributes to the literature by taking into account the effect of financial crises on the endogenous interaction between corporate tax rate and economic growth. for this reason, we first estimate the model from the full sample period, and then to compare the relationship between non-crisis and crisis period, we re-estimate the same model for the sub-sample periods as the pre-crisis, crisis and post-crisis. the results capture only one-way causality from corporate tax rate to economic growth; moreover, a shock to corporate tax rate decreases the economic growth during non-crisis period. interestingly economic growth does not response a shock in corporate tax rate during the crisis period. it is clear that the recent crisis has significantly affected the endogenous g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 191 interaction between corporate tax rate and economic growth. the following section includes a literature review. section 3 shows description of data and section 4 presents the model specification. section 4 provides the empirical results and discussion, while section 5 concludes. 2. literature review the relationship between taxation and economic growth has been investigated in many papers (lucas, 1990; king and rebelo, 1990; engen and skinner, 1996; lee and gordon, 2005; ojede and yamarik, 2012; atems, 2015). it is important to answer the question of how tax policy affects economic growth. there are several ways. first, corporate tax rates affect investment decision adversely. there are numerous studies in the literature showing that corporate tax rates have a significantly negative effect on investments and foreign direct investments (slemrod, 1990; desai et al. 2004; benassy-quere et al., 2005; de mooij and ederveen, 2005, 2006; djankov et al., 2010). second, taxes may affect labor supply or, in another words, laborleisure choice adversely (engen and skinner, 1996; salanie, 2003; hindricks and myles; 2006). the effect of tax rate on economic growth has been a popular research topic. many papers in the literature emphasize the significantly negative effect of tax rate on economic growth. for example, easterly and rebelo (1993) examine the relationships between fiscal policy and economic growth by using the fiscal variables such as central government surplus, public investment and different measures of tax rates. they conclude that a marginal income tax rate has a negative effect on economic growth. the issue of growth and taxation is also addressed by razin and yuen (1996). they investigate the relationship between capital income taxation and economic growth, considering the international capital mobility and endogenous population. under perfect capital mobility, capital income taxes have larger effects on the long-run growth than under no-capital mobility. engen and skinner (1996) highlight the effect of tax policy on economic growth. in fact, they report that a 5% decrease in a marginal tax rate or a 2.5% decrease in average tax rate would likely increase economic growth by 0.2% or 0.3%. although this effect can be considered to be marginal, it can be clearly seen that there is a big long-term effect by accumulating over the past 36 years. koch, schoeman and van tonder (2005) examine the relationship between taxation and economic growth in south africa from 1960 to 2002. their findings show that there is a significantly negative relationship between tax burden and economic growth. while investigating the relationship between taxation and economic growth, focusing on the different types of taxes, wildmalm (2001) tests the presence of the association between tax structure and economic growth by using data from 23 oecd countries for the period from 1965 to 1990. she finds that personal income taxes have a significantly negative effect on economic growth. lee and gordon (2005) investigate taxation policies and economic growth relationship in a cross-section data set of 70 countries over the period from 1970-1997. their finding shows that there is a significant and negative correlation between statutory corporate tax rate and g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 192 economic growth. pesendorfer (2008) shows that the high level of labor taxes affects the growth potential negatively in austria. ojede and yamarik (2012) test the short and long-run effects of tax policy on state-level growth for 48 contiguous us states over the period from 1967 to 2008. they find a significantly negative effect of property and sales tax rates on long-run income growth, while there is not any significant evidence of income tax rates on growth. dackehag and hansson (2012) investigate the relationship between income taxation and economic growth using statutory tax rates on corporate and personal income for 25 rich oecd countries in the period 1975-2010. they find that both taxation of corporate and personal income have a negative effect on economic growth. in contrast to common view, there are studies suggesting a positive relationship between some taxes and economic growth. gober and burns (1997) state that increases in personal income tax, sales tax and corporate income tax increase gnp, while increases in property tax decrease gnp. stoilova (2017) focuses on the relationship between taxes and economic growth for the 28 eu countries in the period 1996-2013. the findings of that study suggest that the taxes on productions and imports have a strong positive impact on economic growth, but the value added taxes have a negative effect on economic growth. the property taxes and economic growth relationship is insignificant, while the personal income tax has a positive effect. however, the corporate taxes have a positive but very weak effect on economic growth. moreover, the coefficients become insignificant when using instrumental variables. it is also possible to find some studies that cannot capture any significant relationship between taxation and economic growth. for example, wang and yip (1992) state that the negative impact of factor taxes on economic growth is compensated by the positive impact of consumption taxes in taiwan. therefore, aggregate tax rates do not have an effect on long-run economic growth. in addition, their findings suggest that the tax structure is more important than level of taxation. pecorino (1994) also highlights the effect of tax structure. he concludes that the effect of changing the income tax with a consumption tax on the growth rate is estimated to be 1% per capita per year. pecorino (1995) also finds that the negative effect of taxation on growth is mild. stokey and rebelo (1995) investigate the effect of tax reform on economic growth in the u.s. they conclude that tax reform would have little or no effect on the u.s. growth rate. mendoza et al. (1997) examine the effects of tax structure on growth and investment by using panel data for 18 oecd countries. they do not capture any significant evidence supporting the association between tax structure and growth. myles (2000) reviews the association between taxation and economic growth by focusing on both theoretical and empirical frameworks. he concludes that the effect of taxation on growth is negligible. however, when the growth is endogenous, taxation can affect the factors that determine growth. therefore, he states that tax structure is critical. recently, various methods are used in the literature to better model the dynamics between taxation and growth. for example, arnold et al. (2011) use a pooled mean group estimator for 21 oecd countries over the period from 1971 to 2004 and investigate the impact of the tax structure on economic growth. empirical findings of their study emphasize that corporate taxes g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 193 have negative effects on productivity and investments. thus, corporate income taxes can be harmful for growth. atems (2015) uses a different methodology, spatial econometric models as called from previous studies. the main finding of that study is that while 1% increase in state and local taxes is associated with a 0.37% decrease in growth in the short-run, 1% increase in state and local taxes is associated with a 0.33% decrease in growth in the long-run. there are few studies explaining how the financial crisis affects this relationship. for example, adkisson and mohammed (2014) investigate that the relationship between the tax structure and short-run growth through the business cycle, particularly in the 2008 great recession. they use a pool of data on the 50 states of u.s. between 2004 and 2010. their results show that differences in tax structure have little effect on economic growth in the context of great recession in 2008. in addition, short-term decreases in corporate taxes are not recommended to contribute to the recovery. in the light of these considerations, the aim of this paper is to fill the gap in the literature by comparing the dynamic relationship between corporate tax rate and economic growth during the recent financial crisis and non-crisis periods. 3. data the data covers 29 oecd economies for the period between 1998 and 2016, for which the data is available. we use panel data that consists of both cross-sectional and time series information to test for any endogenous interaction between different corporate tax rate (tr) and economic growth rate (gr). the source of data is presented in table 1. we first estimate the model from the full sample period, and then to compare the relationship between the non-crisis and crisis periods, we re-estimate the same model for the sub-sample periods as the pre-crisis, crisis and post-crisis. detection of crisis periods is a controversial topic in the literature. the distributions of data, threshold models and news based framework are some of the common methodologies in the literature (lowell et. al, 1998; favero and giavazzi, 2002; dungey, 2009). following the existing literature, the crisis period is defined from 2007 to 2008 based on a news-based framework (dungey, 2014; nber 2010; yalama, 2012). the precrisis period ranges from 1998 to 2006, the crisis period ranges from 2007 to 2008, and the post-crisis period ranges from 2009 to 2016. table 1. variables and sources variable source tr: effective average corporate tax rate taxation trends in the eu (2016); spengel, christoph et al. (2016), effective tax levels using the devereux/griffith methodology, zew final report 2016 gr: the rate which is calculated by taking the difference between the log of a gdp (constant 2010 us$) world bank database g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 194 table 2 shows that there is a huge variation in corporate tax rates across countries. in other words, there exists a large variation in corporate tax rates across countries. corporate tax rates also vary significantly over the pre-crisis, crisis and post-crisis sub-periods. average corporate tax rates drop from 27.13% in the pre-crisis period, to 24.32% in the crisis period, and then drop again to 23.03% level in the post-crisis period. cross-country variation of corporate tax rates tends to persist over years. table 2. distribution of corporate tax rates in the countries over 1998-2005 corporate tax rates average tax rate country pre-crisis crisis post-crisis below 13% (1 countries) 13.34% ireland 11.91% 14.40% 14.32% 14-17% (2 countries) 14.96% lithuania 16.77% 13.95% 13,66% 15,64% latvia 19.06% 14.05% 13.13% 18-22% (8 countries) 18.04% estonia 20.41% 16.90% 16.30% 18.51% turkey 22.35% 17.90% 16.90% 18.71% switzerland 18.80% 18.75% 18,65% 18.95% slovenia 21.32% 20.45% 16.74% 19.05% hungary 18,64% 19.50% 19.30% 19.94% czech r. 23.83% 19.70% 16.80% 20.16% poland 24.02% 17.40% 17.50% 20,63% slovak r. 24.36% 20.45% 18.26% 23-27% (10 countries) 22.57% sweden 23,66% 24,60% 21.30% 23.85% finland 26.33% 24.50% 21.70% 24.10% denmark 27.11% 22.50% 21.91% 25.36% austria 28.83% 23.00% 22.94% 25.90% greece 29.53% 21.75% 23,67% 25.95% netherlands 31.11% 23.10% 22.25% 26.09% norway 26.40% 26.45% 25.75% 26.89% luxembourg 29.14% 25.90% 25.23% 26.94% uk 29.16% 28,65% 24.81% 27.49% portugal 29.16% 23.70% 26.81% 28-32% (4 countries) 28.40% italy 32.03% 29.55% 25.21% 28.53% belgium 31.84% 25.15% 26.44% 30.23% canada 34.95% 32.90% 26.54% 32.32% germany 37.52% 31.80% 28.15% above 32% (4 countries) 34.20% spain 36.50% 33,65% 32.42% 35,67% france 36.07% 34,60% 35.54% 37.23% usa 38.30% 37.40% 36,61% 40.28% japan 41.70% 41.30% 39.06% average 24.83% 27.13% 24.32% 23.03% g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 195 as can clearly be seen in table 2, corporate tax rates have declined over the last 10 years in most of the countries. it is interesting to test whether the decrease of corporate tax rate brings economic growth in to the oecd countries during the crisis and non-crisis period. there are so many empirical papers emphasizing that higher corporate tax rates should decrease economic growth (see lee and gordon, 2005; arnold et al., 2011; dackehag and hansson, 2012). interestingly, however, there are currently no studies that compares the effect of corporate tax rate on economic growth during the crisis and non-crisis periods. table 3 presents the descriptive statistics for gr and tr with respect to the overall, within and between variations during the periods of pre-crisis, crisis and post-crisis. overall variation represents variation over years and countries. between variation represents variation between countries, and within variation represents variation within each country (over years). table 3: descriptive statistics variable variation mean std. dev. min max pre-crisis gr overall 0.034 0.024679 -0.0614 0.1123 between 0.014724 0.0122 0.0730 within 0.019942 -0.0680 0.8551 tr overall 27.13 7.3289 9.4 41.7 between 7.2523 11.9111 41.7 within 3.4200 16.0470 38.5248 crisis period gr overall 0.028 0.031840 -0.0557 0.1051 between 0.021582 -0.0037 0.0786 within 0.023550 -0.0374 0.0941 tr overall 24.32 8.0461 8.8 41.3 between 8.0593 8.85 41.3 within 0.8535 19.0114 26.311 post-crisis period gr overall 0.009 0.037410 -0.1603 0.2333 between 0.016021 0.0379 0.0505 within 0.033901 -0.1589 0.2001 tr overall 24.03 7.6075 7.9 41.7 between 7.5520 8.35 39.0625 within 1.5075 15.5892 28.5892 full sample period gr overall 0.023 0.034287 -0.1603 0.2333 between 0.011768 0.0022 0.0475 within 0.032274 -0.1749 0.2077 tr overall 24.83 7.8222 7.9 41.7 between 7.2754 9.4166 39.875 within 3.7288 15.0068 40.6121 g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 196 according to table 3, the mean values for gr and tr are 0.034 and 27.13% for the precrisis period; 0.028 and 24.32% for the crisis period; and 0.009 and 24.03% for the post-crisis period, respectively. the minimum values for gr and tr are -0.06 and 9.4% for the pre-crisis period; -0.05 and 8.8% for the crisis period; and -0.16 and 7.9% for the post-crisis period, while the maximum values for gr and tr are 0.11 and 41.7% for the pre-crisis period; 0.10 and 41.3% for the crisis period; and 0.23 and 41.7% for the post-crisis period, respectively. according to standard deviation of gr, we have less between variation from one country to the next than within variation for all the sub-periods, which means the variation of countries over years. according to standard deviation of tr, things change and we have more between variation from one country to the next than within variation for all the sub-periods. 4. methodology this paper uses a panel var approach for several reasons. first, panel data consists of both “time series” and “cross-section” dimensions. thus, we have considerable observation derived from both “time series” and “cross-section” dimensions, which increases degree of freedom and certifies estimation of a more robust model (baltagi, 1995). second, the panel var model allows us to investigate the endogenous interaction between corporate tax rate and economic growth. in other words, the var approach takes into account the fact that tr can have an impact on the gr and, at the same time, tr can be influenced by gr. moreover, the panel var analysis based on the system gmm approach allows us to elaborately analyze a small sample, the omitted variables, and endogeneity problems. third, panel granger causality analysis based on the panel var methodology helps us identify the direction of the relationship between tr and gr. finally, impulse response functions (irfs) allow us to assess the dynamic associations between tr and gr. the model is as follows: 𝐺𝑅 𝛽 𝛽 , 𝐺𝑅 𝛽 , 𝑇𝑅 𝑒 𝑢 [1] 𝑇𝑅 𝛼 𝛼 , 𝑇𝑅 𝛼 , 𝐺𝑅 µ 𝓋 [2] here, i (1,…n) denotes the countries, t (1,…,t) denotes the time period, and z denotes the lag number. 𝑢 and 𝓋 are white noise errors. 𝑒 and µ are individual fixed effects for the panel member. gri,t denotes economic growth rate in different countries, while tri,t denotes corporate tax rates. the basic idea of granger causality is that if past values of tr are significant predictors of the current value of gr even when past values of gr have been included in the model, then tr exerts a causal influence on gr. using the above equation, one might easily test whether tr is said to have a predictive power for gr based on an f-test with the following null hypothesis (𝐻 , : g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 197 𝐻 , 𝛽 , 𝛽 , ⋯ 𝛽 , 0 on the other hand, we test whether the gr is granger cause for tr and the corresponding null hypothesis (𝐻 , is: 𝐻 , 𝛼 , 𝛼 , ⋯ 𝛼 , 0 where tri,t and gri,t are the observations of two stationary variables for the countries in period t. if h0,1 is rejected, it indicates that causality from tr to gr exists. it can be tested for causality in the other direction as well, and it is possible to detect the existence of causality for testing h0,2. if h0,2 is rejected, one can conclude that causality from gr to tr exists. the correct choice of lag length is important to avoid non-trustable results on grangercausality. thus, this paper uses schwartz information criteria (sic) for determining optimal lag. according to sic, we select a lag length of two as optimal. because the priority requirement as stationary is necessary for implementing the grangercausality tests, a panel unit root tests are applied for both gr and tr variables in the all the sub-periods as pre-crisis, crisis and post-crisis. the results are presented in table 4a and 4b. according to table 4a and 4b, we reject the null hypothesis of non-stationarity for the all variables at levels for the full sample and all the related sub-samples. table 4a: panel unit root test for gr * represents a significance level of 0.05. table 4b: panel unit root test for tr pre-crisis crisis post-crisis full sample levin, lin & chu t* -8.7948 (0.0000)* -8.4551 (0.0000)* -11.3154 (0.0000)* -10.5222 (0.0002)* adf fisher chi-square -3.1706 (0.0008)* 107.414 (0.0001)* 224.557 (0.0000)* 211.922 (0.0000)* pp fisher chi-square 103.347 (0.0001)* 119.593 (0.0000)* 197.388 (0.0000)* 209.858 (0.0000)* pre-crisis crisis post-crisis full sample levin, lin & chu t* -2.1973 (0.0140)* -8.1966 (0.0000)* -7.9403 (0.0000)* -3.5988 (0.0002)* adf fisher chi-square 85.9555 (0.0002)* 105.762 (0.0000)* 74.6718 (0.0484)* 136.793 (0.0000)* pp fisher chi-square 125.154 (0.0000)* 101.903 (0.0000)* 95.4330 (0.0008)* 158.391 (0.0000)* * represents a significance level of 0.05. g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 198 table 5: granger causality test results (𝐓𝐑 ↛ 𝐆𝐑 : pre-crisis, crisis, post crisis and full sample period pre-crisis sample tr↛gr gr↛tr lags 2 lags 2 wald stat 22.4924 wald stat 0.2673 p-value 0.0000*** p-value 0.7659 crisis sample tr↛gr gr↛tr lags 2 lags 2 wald stat 0.6966 wald stat 2.9355 p-value 0.7092 p-value 0.2304 post-crisis sample tr↛gr gr↛tr lags 2 lags 2 wald stat 17.0131 wald stat 1.6747 p-value 0.0002*** p-value 0.4328 full sample tr↛gr gr↛tr lags 2 lags 2 wald stat 11.2459 wald stat 0.7988 p-value 0.0036*** p-value 0.6707 * represents a significance level of 0.05. table 5 displays the results from the panel granger causality tests for both the full sample and all the sub-sample periods. the null hypothesis that corporate tax rate does not grangercause economic growth is rejected at the 5 percent level of significance for the full sample period. however, for the sub-sample periods, the results of granger causality tests are very different from those for the full sample period. for instance, for the non-crisis period (both precrisis and post-crisis), the null hypothesis is rejected at the 5 percent level of significance as in the case with the full sample period indicating the existence of granger causality from tr to gr. however, the results dramatically change for the crisis period. interestingly, the null hypothesis is not rejected for the crisis period, which suggests that there is not any significant evidence of causality between tr and gr during the crisis period. the results fail to reject the null hypothesis that gr does not granger-cause tr at the 5 percent level of significance for both the full sample and all the sub-sample periods. the results show that there is no evidence of causality from gr to tr. as a result, we only capture one-way granger causality from tr to gr during the non-crisis period (both for the pre-crisis and post-crisis periods). after detecting granger causality relationship between tr and gr by relying on the panel var methodology, we further investigate the endogenous interaction between tr and gr during the crisis and non-crisis periods by applying structural impulse-response functions (irfs), which assess the dynamic effects for exogenous shocks of tr to gr in the system. figure-1 shows the irfs with their corresponding confidence bands for both the full sample and all the related sub-sample periods. g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 199 figure 1. impulse response functions of gr and tr: pre-crisis, crisis, post crisis and full sample period. according to figure 1, a shock to corporate tax rate decreases the economic growth during non-crisis period. interestingly economic growth does not response a shock in corporate tax rate during the crisis period. it is clear that the impulse responses functions for crisis period are very different from those in the pre-crisis, post-crisis and full sample periods, which suggests that the recent crisis has profoundly affected the endogenous interaction between corporate tax rate and economic growth. 5. conclusions in terms of fiscal policy, it is important that taxes should positively affect economic growth. however, corporate tax has direct impacts on investment decisions, capital structures and earnings of corporations. depending on the extent of taxation of corporate earnings, the entrepreneurs may give up their investment decisions and reduce their existing investments. on the other hand, a low corporate tax rate can generate both a higher rate of capital investment and a more entrepreneurial activity, which will directly affect economic growth. -.010 -.005 .000 .005 .010 .015 .020 1 2 3 4 5 6 7 8 9 10 pre-crisis: response of gr to tr -.06 -.04 -.02 .00 .02 .04 .06 1 2 3 4 5 6 7 8 9 10 crisis: response of gr to tr -.02 -.01 .00 .01 .02 .03 .04 1 2 3 4 5 6 7 8 9 10 post crisis: response of gr to tr -.01 .00 .01 .02 .03 .04 1 2 3 4 5 6 7 8 9 10 total sample data: response of gr to tr g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 200 it is essential that policy-makers detect optimal tax rate so as to meet the expectations of both governments and entrepreneurs. in other words, the tax rate adopted by policy-makers should ensure both economic growth for the government and sustainability of investments for entrepreneurs. the primary aim of this paper is to investigate the relationship between corporate tax rate and economic growth. on the other hand, this paper also tests whether the recent financial crisis has any significant effect on the dynamic relationship between economic growth and corporate tax rate by using panel var for 29 oecd countries over the period 1998-2016. the results show that corporate tax rates have a significantly negative effect on economic growth. moreover, the recent financial crisis has had a significant effect on the endogenous interaction between corporate tax rate and economic growth. according to granger causality test, there is only one-way causality from corporate tax rate to economic growth during the non-crisis period. interestingly, there is not any causal relationship between corporate tax rate and economic growth during the crisis period. the public policy should focus more on corporate tax deduction to increase the economic growth during non-crisis period. interestingly corporate tax rate is not significant policy tool to increase economic growth during the crisis period. future studies may contribute to the literature describing the relationship between tax rates and economic growth by focusing on the impact of different tax rates, on different characteristics of crises or on the impact of econometric methods on these relationships. references adkisson, r. v. & mohammed, m. (2014). tax structure and state economic growth during the great recession. the social science journal, 21, 79-89. doi: https://doi.org/10.1016/j.soscij.2013.10.009 arnold, j.m., brys, a., heady, c., johansson, a., schwellnus, c. & vartia, l. (2011). tax policy for economic recovery and growth. the economic journal, 121, 59-80. doi: https://doi.org/10.1111/j.1468-0297.2010.02415.x atems, b. (2015). another look at tax policy and state economic growth: the long-run and short-run of it. economics letters, 127, 64-67. doi: https://doi.org/10.1016/j.econlet.2014.12.035 baltagi, b.h. (1995). econometric analysis of panel data. wiley: chichester. benassy-quere, a., fontagne, l. & lahreche-revil, a. (2005). how does fdi react to corporate taxation? international tax and public finance, 12 583–603. doi: https://doi.org/10.1007/s10797-005-2652-4 dackehag, m. & hansson, a. (2012). taxation of income and economic growth: an empirical analysis of 25 rich oecd countries. lund university, department of economics, working paper 2012:6. de mooij, r.a. & ederveen, s. (2005). explaining the variation in empirical estimates of tax elasticities of foreign direct investment. tinbergen institute discussion paper, no. 2005108/3. doi: https://doi.org/10.2139/ssrn.869753 de mooij, r.a. & ederveen, s. (2006). what a difference does it make? understanding the empirical literature on taxation and international capital flows. european commission economic paper no. 261. djankov, s., ganser, t., mcliesh, c., ramalho, r. & shleifer, a. (2010). the effect of g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 201 corporate taxes on investment and entrepreneurship. american economic journal macroeconomics, 2(3), 31-64. doi: https://doi.org/10.1257/mac.2.3.31 desai, m.a., foley, c.f. & hines jr., j.r. (2004). foreign direct investment in a world of multiple taxes. journal of public economics, 88, 2727-2744. doi: https://doi.org/10.1016/j.jpubeco.2003.08.004 dungey, m. (2009). the tsunami: measures of contagion in the 2007-08 credit crunch, cesifo forum, 9(4), 33-34. dungey, m. & gajurel, d. (2014). equity market contagion during the global financial crisis: evidence from the world's eight largest economies. economic systems, 38(2), 161-177. doi: https://doi.org/10.1016/j.ecosys.2013.10.003 easterly, w. & rebelo, s. (1993). fiscal policy and economic growth. journal of monetary economics, 32, 417-458. doi: https://doi.org/10.1016/0304-3932(93)90025-b engen, e. & skinner, j. (1996). taxation and economic growth. national tax journal, 49(4), 617-642. doi: https://doi.org/10.3386/w5826 favero, c.a. & giavazzi, f. (2002). is the international propagation of financial shocks nonlinear? evidence from the erm. journal of international economics, 57(1), 231-246. doi: https://doi.org/10.1016/s0022-1996(01)00139-8 gober, j. r. & burn, j. o. (1997). the relationship between tax structures and economic indicators. journal of international accounting, auditing and taxation, 6(1), 1-24. doi: https://doi.org/10.1016/s1061-9518(97)90010-0 harberger, a. c. (1964). taxation, resource allocation, and welfare. in nber and the brookings institution (eds.), the role of direct and indirect taxes in the federal revenue system. new jersey: princeton university press. doi: https://doi.org/10.1515/9781400875931-003 helms, l. j. (1985). the effect of state and local taxes on economic growth: a time seriescross section approach. the review of economics and statistics, 67(4), 574-582. doi: https://doi.org/10.2307/1924801 hindricks, jean & myles, gareth d. (2006). intermediate public economics, the mit press. king, r. g. & rebelo, s. (1990). public policy and economic growth: developing neoclassical implications. journal of political economy, 98(5), 126-150. doi: https://doi.org/10.1086/261727 koch, s. f., schoeman, n. j. & van tonder, j. j. (2005). economic growth and the structure of taxes in south africa: 1960-2002. south african journal of economics, 73(2), 190-210. doi: https://doi.org/10.1111/j.1813-6982.2005.00013.x lee, y. & gordon, r. h. (2005). tax structure and economic growth. journal of public economics, 89, 1027-1043. doi: https://doi.org/10.1016/j.jpubeco.2004.07.002 lowell, j., neu, c.r., & tong, d. (1998). financial crisis and contagion in emerging market countries. monograph rand. mendoza, e. g., milesi-ferretti, g. m & asea, p. (1997). on the ineffectiveness of tax policy in altering long-run growth: harberger’s superneutrality conjecture. journal of public economics, 66, 99-126. doi: https://doi.org/10.1016/s0047-2727(97)00011-x nber. (2010). business cycle dating committee report. technical report, national bureau of economic research (2010). ojede, a. & yamarik, s. (2012). tax policy and state economic growth: the long-run and short-run of it. economics letters, 116, 161-165. doi: https://doi.org/10.1016/j.econlet.2012.02.023 pecorino, p. (1994). the growth rate effects of tax reform. oxford economic papers, 46(3), 492-501. doi: https://doi.org/10.1093/oxfordjournals.oep.a042143 pecorino, p. (1995). tax rates and tax revenues in a model of growth through human capital accumulation. journal of monetary economics, 36, 527-539. doi: https://doi.org/10.1016/0304-3932(95)01224-9 pesendorfer, k. (2008). tax and economic growth in austria. monetary policy and the economy, quarter 3, 21-40. rebelo, s. (1991). long-run policy analysis and long-run growth. the journal of political economy, 99(3), 500-521. doi: https://doi.org/10.1086/261764 romer, p. m. (1986). increasing returns and long-run growth. journal of political economy, g. oz-yalaman / european journal of government and economics 8(2), december 2019, 189-202 202 94(5), 1002-1037. doi: https://doi.org/10.1086/261420 salanie, b. (2003). the economics of taxation. the mit press. slemrod, j. b. (1990). tax effects on foreign direct investment in the united states: evidence from a cross-country comparison. in assaf razin & joel slemrod (eds.) taxation in the global economy (pp. 79-122). university of chicago press. solow, r. m. (1956). a contribution to the theory of economic growth. the quarterly journal of economics, 70(1), 65-94. doi: https://doi.org/10.2307/1884513 stoilova, d. (2017). tax structure and economic growth: evidence from the european union. contaduría y administración, 62(3), 1041-1057. doi: https://doi.org/10.1016/j.cya.2017.04.006 stokey, n. l. & rebelo, s. (1995). growth effects of flat-rate taxes. journal of political economy, 103(3), 519-550. doi: https://doi.org/10.1086/261993 trostel, p. a. (1993). the effect of taxation on human capital. journal of political economy, 101(2), 327-350. doi: https://doi.org/10.1086/261878 wang, p. & yip, c. k. (1992). taxation and economic growth: the case of taiwan. the american journal of economics and sociology, 51(3), 317-331. doi: https://doi.org/10.1111/j.1536-7150.1992.tb03483.x wildmalm, f. (2001). tax structure and growth: are some taxes better than others? public choice, 107(3/4), 199-219. doi: https://doi.org/10.1023/a:1010340017288 yalama, a. (2012). international financial contagion: the role of the uk, bogazici journal, review of social, economic and administrative studies, 26(2), 115-129. doi: https://doi.org/10.21773/boun.26.2.5 the evaluation index improvement for the relocation of research equipment vol.8 • no.1 2019 issn: 2254-7088 european journal of government and economics 8(1), june 2019. european journal of government and economics issn: 2254-7088 number 8, issue 1, june 2019 doi: https://doi.org/10.17979/ejge.2018.8.1 the evaluation index improvement for the relocation of research equipment 6-29 donghun yoon residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) 30-47 josé ramón cardona, daniela álvarez bassi, maría dolores sánchez-fernández social capital and banking system profitability: a survey of european union countries 48-62 arash nayebyazdi governance and domestic investment in africa 63-80 chimere okechukwu iheonu economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt 81-95 marwa shibl biltagy give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia 96-109 jaroslav mihálik, peter horváth, martin švikruha https://doi.org/10.17979/ejge.2018.8.1 european journal of government and economics 8(1), june 2019, 6-29 european journal of government and economics issn: 2254-7088 the evaluation index improvement for the relocation of research equipment donghun yoona, * a korea basic science institute, republic of korea * corresponding author at: njoyoon@gmail.com article history. received 5 october 2018; first revision required 2 january 2019; accepted 14 february 2019. abstract. in this paper, evaluation index improvement for research equipment relocation is presented and discussed to boost the effectiveness of research equipment management. the analytic hierarchy process (ahp) model was designed for the evaluation index improvement for research equipment relocation, and the pairwise comparison scale was set up based on the importance of each evaluation criterion. the consistency rate (cr) was measured, and it was confirmed that the decision-making was reasonable. the improvement of the evaluation index was necessary for the objective and fair relocation of research equipment. therefore, the evaluation index for the relocation of research equipment was designed for an objective and fair evaluation. it is hoped that the study findings will be very useful and will contribute greatly to the professors, researchers, and policymakers involved in science and technology policymaking and r&d. keywords. policy research, evaluation index, research equipment, science and technology, relocation jel codes. h11, h61, m10, m15, m21 doi. https://doi.org/10.17979/ejge.2019.8.1.4572 1. introduction the ministry of science and ict (mist) of south korea introduced a comprehensive review system for research facilities and equipment in 2016 to prevent overlapping investment in research equipment infrastructure and promoted the research institute’s own review system. mist has been conducting a research equipment status survey every year since 2012 to manage research equipment. a roadmap for large research facilities was established in 2013 to strategically introduce large research facilities with high construction costs. mist has also enacted a law for the national research facilities and equipment standard guideline (2016). it has operated research equipment and support services and has trained research equipment manpower through education programs. the allowable rate for the mutual utilization of research equipment was raised to enhance the utilization of research equipment, and the support for promoting the recycling of idle and underutilized research equipment was strengthened. the allowable rate for the mutual utilization of research equipment in south korea was initially 60% but was 60.3% in december 2017. mailto:njoyoon@gmail.com https://doi.org/10.17979/ejge.2019.8.1.4572 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 6 in this paper, evaluation index improvement for research equipment relocation is presented and discussed to boost the effectiveness of research equipment management. mist has transferred 153 idle and underutilized research equipment free of charge to universities, laboratories, and nonprofit organizations, among other institutions. the government allowed researchers to freely apply for research equipment to be transferred to them free of charge. there were so many researchers who had applied for research equipment utilization that it was necessary to develop an evaluation index for the objective and fair relocation of research equipment. in this paper, evaluation index improvement for research equipment relocation is presented and discussed to boost the effectiveness of research equipment management. mist has transferred 153 idle and underutilized research equipment free of charge to universities, laboratories, nonprofit institutions, etc. the government allowed researchers to freely apply for research equipment to be transferred to them free of charge. there were so many researchers who had applied for research equipment utilization that it was necessary to develop an evaluation index for the objective and fair relocation of research equipment. the objective and fair relocation of research equipment is very important for r&d effectiveness and research results. the evaluation index for research equipment relocation needs for the objective and fair relocation of research equipment. therefore, an evaluation index for the relocation of research equipment was designed for an objective and fair evaluation. in this paper, the analytic hierarchy process (ahp) model is utilized as a methodology. the pairwise comparison scale was set up based on the importance of each evaluation criterion and the consistency rate (cr) was measured. this paper consists of introduction, literature review, science and technology environment change, global r&d investment trend, government r&d investment in south korea, relocation of research equipment, methodology, results and discussion, evaluation index improvement research design, and conclusions and policy implication. 2. literature review using the particular case of a large-scale r&d programme concerned with learning technologies within the european community's third framework programme for precompetitive industrial r&d, a participative evaluation method is outlined (stern, 1993). the research evaluation must allow a more systematic evaluation approach to permeate the organisation whereby a common frame of reference regarding the choice of evaluation subjects, choice of strategic focus, and use of fundamental concepts is established (brofoss, 1998). the technometrics approach is a method of evaluation of biotechnology r&d which contributes to ex ante evaluations (meyer-krahmer & reiss, 1992). research evaluation in a context of delegation and as a self-organising system for research actors guaranteed by the state has been strongly developed in the last few years (sanz-menéndez, 1995). in the field of r&d policy at least, reality, theory and therefore the needs of evaluation users seem to have moved well ahead of evaluators' conceptual apparatus (arnold, 2004). because of emerging funding d. yoon / european journal of government and economics 8(1), june 2019, 6-29 7 volumes and increasing expectations in results, concepts for performance measurement and management gain importance (schröder et al., 2014). the evaluation path is better characterised by an experimental attitude of constant change with each new programme, rather than a consolidation of practices learned and tested in previous programmes (silva & henriques, 1995). the research assessment exercise (rae) represents one of the most institutionalised forms of research evaluation in the oecd economies (barker, 2007). widespread and increasing public subsidy for research and development (r&d) has given rise to a large and growing number of evaluation studies (dimos & pugh, 2016). interest in evaluating non-economic social outcomes of science and technology research has risen in policy circles in recent years. the interest in social impacts of research has not yet given rise to a great proliferation of useful, valid techniques for evaluating such impacts (bozeman & youtie, 2017). over recent years there has been a consistent shift in the way in which r&d policy is viewed. at a simplistic level, this has resulted in a convergence of the previously separate domains of s&t and industrial policy into a more coherent innovation policy perspective (cunningham, 2008). grand challenges stress the importance of multi-disciplinary research, a multi-actor approach in examining the current state of affairs and exploring possible solutions, multi-level governance and policy coordination across geographical boundaries and policy areas, and a policy environment for enabling change both in science and technology and in society (amanatidou et al., 2014). innovation underpins competitiveness, is crucial to addressing societal challenges, and its support has become a major public policy goal (edler et al., 2016). government policy objectives can include the development of technology for private users where there is a public policy rationale (rigby et al., 2016). clearly, in its simplest sense, an innovation support measure must be defined as a policy instrument designed to support the process of innovation, at the national, regional or other levels (cunningham, 2008). 2.1. science and technology environment change new industries and markets based on the new-value-creation-type product service are created through industry-ict convergence. the development of the industry infra, such as the smart factory, is accelerated. high-value-added industries will be created by upgrading the mainstream technology related to the industry. digital technology scope expansion is applied to mobile devices and various other products for a society based on the internet (chang et al., 2016; kauffman al., 2015). the development of digital-technology-based products such as 3d printers, autonomous vehicles, and drones will enhance business efficiency. the method change for r&d performance improvement enhances the research trend in a timely manner due to the greater collaboration, higher cost, shortened product technology life cycle, etc. commercialization of technologies on demand should be promoted based on r&d (cerqueti et al., 2016; lynskey, 2016). also, the demand for a technology entrepreneurial ecosystem is increased. the strengthening of scientific and technical responsibilities highlights the role of javascript:; javascript:; javascript:; javascript:; d. yoon / european journal of government and economics 8(1), june 2019, 6-29 8 science and technology in addressing global and social issues such as climate change, aging population, and the emergence and spread of intractable diseases. the responsibility of science and technology in resolving the technological side effects, such as information infringement and security issues, through the development of science and technology is enhanced. labor force replacement is required due to the development of smart technologies and ict, which can extend or replace human intelligence. new industries and services are required to be created to overcome the problems of job polarization and high unemployment rate. 2.2. global r&d investment trend despite the global economic downturn, including the low interest rate and the increasing instability in the international financial and commodity market, the r&d investments of several countries are steadily increasing (lewis & tan, 2016; schatz & bashroush, 2017). developed countries (usa, japan, eu, etc.) are increasing the share of r&d investment in their gdp while maintaining their dominance in r&d investment. in particular, the r&d investment in china is rapidly expanding in both the government and the private sector, amounting to usd294.6 billion in 2013. its proportion of the gdp was 1.39% (usd117,386 million) in 2007 and 2.02% (usd294,621 million) in 2013 for the total r&d investment of china. the proportion of government funding in the total r&d investment (usa, japan, eu, etc.) is slightly fluctuating but is expanding continuously. 2.2.1. usa in the usa, the federal government r&d budget in 2015 was usd136.5 billion, representing a 0.7% increase from the previous year. the non-defense funds (47%) in the 2015 government budget increased by 1.5%. the reduced budget pressure promotes investment in selected sectors under limited financing conditions. the r&d budget growth rate (2014-2015) was 11.7% for energy and 3.5% for the environment. the basic and applied research budgets were reduced by 1.8% for energy and 1.6% for the environment from the previous year. that for energy was usd31.7 billion, and that for the environment was usd34.2 billion. the r&d budget was increased by 3.2%. it was usd67.7 billion in 2013, usd66 billion in 2014, and usd68.2 billion in 2015. the office of science and technology policy (ostp) and the office of management and budget (omb) suggested nine high-priority national priorities (high-tech manufacturing, clean energy, global climate change, policy formulation and management, information technology, national security, biology and neuroscience, stem [science, technology, engineering, & mathematics] education, innovation and change). d. yoon / european journal of government and economics 8(1), june 2019, 6-29 9 2.2.2. japan in japan, the science and technology budget in 2015 was jpy4.03 trillion, representing a 10.4% increase from the previous year. the science and technology budget was stagnated for 10 years. it was the original budget + the revised budget + the science and technology promotion budget + the local public entity budget. the science and technology budget trend was jpy40,490 billion in 2003, jpy42,405 billion in 2008, and jpy44,393 billion in 2013. to promote economic and social development through science and technology innovation and the technological innovation system, the 2014 science and technology innovation strategy was presented. it included the five policy tasks (innovation of a clean and economical energy system, the realization of a healthy and long-living society for leading the international society, construction for pioneering the next-generation infrastructure, new-industry promotion for utilizing the local resources, and early recovery from the great east japan earthquake). to realize the science and technology innovation strategy, the council for science and technology policy (cstp) promoted linkage and cooperation among the ministries through a top-down budget policy. cstp carried out ten major research projects linked to the policy tasks through the strategic innovation creation program (sip; 2015 budget: jpy50 billion). it emphasized the creation of a science and technology innovation environment through r&d enhancement, regional innovation hub establishment, and environment improvement for venture activation. the ministry of education, culture, sports, science, and technology (mext) is forming an innovative hub business centered on the r&d corporations (2015, jpy5 billion). 2.2.3. china in 2015, the science and technology budget was cny277.7 billion, representing a 3.1% increase from the previous year. the rule of science and technology development (2011) suggested r&d investment expansion, the combined improvement of science and technology and the market economy, and the development of human resources for science and technology. among the national r&d investments, the central government funding doubled, from cny108.8 billion in 2008 to cny222.1 billion in 2012. the central government funding was increased by 2.3% (21.6%). the central government and the enterprise investment were cny108.89 and 331.15 billion in 2008 and cny222.14 and 762.5 billion in 2012. in 2013, the science and technology budget of the government was cny614.49 billion, representing a 38.9% increase from the previous year. the science and technology funds in the 2015 budget accounted for 4.43% of the budget. the central and local science and technology budgets were cny164.9 and 157.6 billion in 2009, respectively, and cny227.9 and 345.6 billion in 2013. the ministry of science and technology of the people’s republic of china (most) focused on the eight missions in 2015. also, the government investment had been strengthened for basic science, d. yoon / european journal of government and economics 8(1), june 2019, 6-29 10 frontier science, and economic and social problem-solving. the eight missions are the top-level design promotion of an innovation-driven development strategy, new science and technology plan management model establishment, science and technology system reformation, special project implementation, science and technology innovation capability enhancement, localinnovation-development-level improvement through the regional development strategy, promotion of science and technology achievement and service industry development, and international open cooperation establishment. the basic and frontier sciences are nano, protein, climate change, space, sea, etc. while the economic and social problem-solving is on clean energy, alternative energy vehicles, the information and communication network, remote sensing, and biotechnology. 2.2.4. eu the total budget of horizon 2020 was eur76.9 billion (2013). the science and technology budget was eur7.32 billion in 2014 and 7.06 billion in 2015, representing a 3.7% decrease. the budget was reduced by 6.7% for excellent science, and by 14.4% for social challenges. it was increased, however, by 6.2% for industrial leadership, and by 162.1% for the others. the industry leadership sector includes information & communication technology (ict), nanotechnology (nt), biotechnology (bt), space technology (st), advanced manufacturing, robotics, etc., and the others include the european atomic energy community (euratom). fast track to innovation (fti) was established and supported by eur100 million in 2015. it was based on openness and carried out a cooperation project that went beyond the boundaries (technology, industry-academe-research institutes, etc.). germany, the core of the eu economy, is expanding its investment in the energy and biotechnology sectors, with the federal r&d budget growing at an average annual rate of 2.7% (2011-2014). the r&d investment growth rate of the german government (2011-2014) was 13.1% for energy and 7.6% for biotechnology. 2.3. government r&d investment in south korea the government r&d investment relative to the gross domestic product (gdp) in 2013 was 4.15%. it was no. 1 in the world. the r&d cost was krw54.2 billion, the sixth largest in the world. the total r&d cost increased by 11.8% per annum over the past 5 years (2009-2013). the proportion of investment to the gdp rose by 0.86%, from 3.29% in 2009 to 4.15% in 2013. the government r&d budget growth is slowing down (11.4% in 2009 → 6.4% in 2015), but the expansion of r&d investment is strong and is focused on the key areas. the government r&d budget was krw12,343.7 billion in 2009, krw14,890.2 billion in 2011, krw17,147.1 billion in 2013, and krw18,923.1 billion in 2015. to supplement and lead private r&d, the ministry of science and ict (msit) is exerting efforts to expand the government r&d investment, but the proportion of government funding is lower than in the other advanced countries. msit is d. yoon / european journal of government and economics 8(1), june 2019, 6-29 11 expanding the government r&d investment for the realization of the creative economy and for securing the future growth engine. an attempt is also being made to improve r&d performance and efficiency. with the objective of a creative economy, the government r&d investment is being expanded for the establishment of a new market opening ecosystem and the cultivation of creative talents. the r&d budget for a creative economy was krw5,610.2 billion in 2014 and krw6,218.3 billion in 2015. the government r&d investment in the future growth engines has been expanded to create new industries through the diffusion of convergence researches such as on the internet of things (iot) and customized wellness care. the related r&d budget was krw997.5 billion in 2014 and krw1, 106.3 billion in 2015. the r&d achievement of public research institutes encourages technology transfer. the government r&d project is continuously being improved for r&d efficiency. the r&d project support of the public research institutes was krw15 billion in 2015, the technology commercialization voucher was krw1 billion in 2015, and the technology advancement was krw2 billion in 2015. this study focused on the expected effect of r&d investment on research equipment in south korea. when the government r&d budget is invested, it is essential to monitor and verify the expected effects. also, this research verified the effectiveness of r&d investment through the calculation of the expected effects of r&d investment in research equipment. 2.4. relocation of research equipment when the korea research institute of fraunhofer institute of germany was liquidated in 2013, its research equipment became idle and underutilized. the government could be allowed researchers to freely apply for research equipment to be transferred to them free of charge through the national and local budget for the purchase of research equipment. this was done in accordance with article 25 of the act on the establishment and operation of a public utility corporation in south korea and article 20 of korea research institute of fraunhofer institute. the research equipment has been utilized for r&d on the commercialization of the life sciences and on immune proteins, vaccines, flu, malaria, etc. the national science & technology information service (ntis) was notified of the free-of-charge transfer of 153 research equipment from october 2013 to february 2014. as mentioned above, the government allowed researchers to freely apply for research equipment to be transferred to them free of charge. the re-utilization and relocation of research equipment have a useful advantage for r&d management and the r&d budget reduction. however, since the reutilized and relocated research equipment is not a piece of new research equipment, it is sometimes necessary to repair and replace consumables for normal performance. it can be done through government r&d budget support. the current status of the research equipment is presented in table 1. d. yoon / european journal of government and economics 8(1), june 2019, 6-29 12 table 1. status of research equipment classification (usd) research equipment number purchase cost (usd) 27,713 ~ 9 497,921 18,475-27,713 5 109,930 9,237-18,475 21 285,450 4,618-9,237 11 7,852 ~ 4,618 107 122,863 total 153 1,024,016 3. methodology in this study, the analytic hierarchy process (ahp) model was used. ahp creates a pairwise compression matrix by measuring the weights of the elements in the lower hierarchy based on the weights of the elements in the upper hierarchy. it uses the eigenvalue method to calculate the normalized priority vector for each level of the hierarchy. it also calculates the priority level for the entire hierarchy, which indicates the relative priority levels of the alternatives. the solution to the decision problem for ahp is as follows. first, the decision problems are grouped and arranged into a hierarchy of related decision-making points. the decision hierarchy is then established. decision analysts visualize a number of mutually relevant decisions. the top hierarchy has the most comprehensive purpose of making decisions, and the following hierarchy consists of a variety of factors that affect the purpose of making decisions. the standard hierarchy of ahp is shown in figure 1. second, a matrix is created through pairwise compression in the lower hierarchy, which contributes to the achievement of the goal of the elements in the higher hierarchy. a 9-point likert scale gives importance to the degree to which a binary comparison contributes to a higher component. the upper factors are set on a 9-point scale through pairwise compression. the pairwise compression scale is shown in table 2. the matrix created through comparison according to the evaluation criteria is the pairwise compression matrix. the pairwise compression matrix is shown in table 3. third, weight is estimated. the weight estimate calculates the geometric average of the pairwise compression matrix. the matrix weight estimation for pairwise compression is shown in table 4. fourth, the consistency is judged. the values obtained from the pairwise compression should be considered for the overall consistency of the evaluation index improvement for the relocation of research equipment. d. yoon / european journal of government and economics 8(1), june 2019, 6-29 13 figure 1. standard hierarchy of the analytic hierarchy process (ahp). source: own elaboration decision factor 2 decision factor 3 comprehensive decision objective decision element 2 detail element of decision decision element 3 detail element of decision decision element 1 detail element of decision decision factor 1 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 14 table 2. pairwise compression scale importance definition 1 equal importance 3 moderate importance 5 strong importance 7 very strong importance 9 extreme importance 2, 4, 6, 8 median reciprocal value activity j has a reciprocal value for activity i. 1.1-1.9 for tied activities source: saaty, t. l. (1980) table 3. pairwise compression matrix classification element a element b element c element d element a 1 3 5 7 element b 1/3 1 5 7 element c 1/5 1/5 1 3 element d 1/7 1/7 1/3 1 table 4. matrix weight estimation for pairwise compression classification geometric mean (total=5.93) importance element a (1x3x5x7)1/4=3.2 3.2/5.93=0.54 element b (1/3x1x5x7)1/4=1.85 1.85/5.93=0.31 element c (1/5x1/5x1x3)1/4=0.59 0.59/5.93=0.1 element d (1/7x1/7x1/3x1)1/4=0.29 0.29/5.93=0.05 4. results and discussion the ahp model was designed by the research equipment relocation review committee. the ahp model for evaluation index improvement for the relocation of research equipment is shown in figure 2. pairwise compression was used to evaluate the relative importance in terms of the upper evaluation criteria, through a survey for the evaluation elements. a pairwise compression scale was designed by measuring the importance of each evaluation criterion. the results of the importance evaluation of the evaluation criteria are shown in table 5. the relative importance of each of the four evaluation criteria was studied by the research equipment relocation review committee. the relative importance of each evaluation criterion is shown in table 6. the importance of contribution to r&d activity, utilization, contribution to mutual utilization, and appropriateness of the installation environment of expensive and low-cost research equipment was evaluated. the results of the importance evaluation of contribution to r&d activity are shown in table 7, those of research equipment utilization are shown in table 8, those of contribution to mutual utilization are shown in table 9, and those of appropriateness of the installation environment are shown in table 10. based d. yoon / european journal of government and economics 8(1), june 2019, 6-29 15 on table 4, the importance of each evaluation criterion was calculated. the total relative importance of the evaluation criteria is shown in table 11. the normalized matrix was used through the adjustment of the importance and pairwise compression scale. the weight was calculated by averaging the row. the weight of each evaluation criterion is shown in table 12. it was used to calculate the relative importance (0.24, 0.28, 0.27, and 0.21) of the effects on the evaluation criteria in table 11. the weights of the evaluation elements were calculated in the same way. the elements of the evaluation criteria are shown in table 13. the superior standard was evaluated through the evaluation criteria. the importance was calculated in table 12 and 13. below is the detailed formula that was used for the calculation. the importance of the evaluation criteria for expensive equipment was 0.638, and that for low-cost equipment was 0.458. the ahp application model for evaluation index improvement for research equipment relocation is shown in figure 3. the consistency rate (cr) was measured to determine the level of consistency of the research results. when the evaluation was completed, cr became 0 (zero) and was consistent if the cr was less than 0.1. below is the detailed formula that was used. the random index (ri) value according to n change is shown in table 14. the results of the calculation of the weight matrix are shown in table 15. below are the detailed equations that were used. 0.21 expensive research equipment 0.21 0.79 0.79 0.17 0.36 0.638 low-cost research equipment 0.79 0.21 0.21 0.83 0.24 0.458 0.20        =            ( )( ) ( ) ci consistency indexcr consistency ratio ri random index = ( )( ) 1 max nci consistency index n λ − = − xi wimax n λ ⋅ = ∑ d. yoon / european journal of government and economics 8(1), june 2019, 6-29 16 figure 2. ahp model for evaluation index improvement for research equipment relocation table 5. importance evaluation of the evaluation criteria qualitative evaluation quantitative evaluation hardly important 1/5 less important 1/3 equally important 1 very important 3 absolutely important 5 evaluation index contribution to r&d activity expensive research equipment low-cost research equipment appropriateness of installation environment utilization of research equipment contribution to mutual utilization d. yoon / european journal of government and economics 8(1), june 2019, 6-29 17 table 6. relative importance of the evaluation criteria evaluation criteria contribution to r&d activity utilization of research equipment contribution to mutual utilization appropriateness of installation environment contribution to r&d activity 1/5 3 1 1/5 utilization of research equipment 1 1 3 1/3 contribution to mutual utilization 1/3 3 1 1/3 appropriateness of installation environment 1/3 1/5 1/5 1 table 7. importance evaluation of contribution to r&d activity contribution to r&d activity expensive research equipment low-cost research equipment expensive research equipment 1 1/5 low-cost research equipment 3 1 table 8. importance evaluation of utilization of research equipment utilization of research equipment expensive research equipment low-cost research equipment expensive research equipment 1 5 low-cost research equipment 1/3 1 table 9. importance evaluation of contribution to mutual utilization contribution to mutual utilization expensive research equipment low-cost research equipment expensive research equipment 1 3 low-cost research equipment 1/5 1 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 18 table 10. importance evaluation of appropriateness of installation environment appropriateness of installation environment expensive research equipment low-cost research equipment expensive research equipment 1 1/5 low-cost research equipment 5 1 table 11. relative importance total of the evaluation criteria evaluation criteria contribution to r&d activity utilization of research equipment contribution to mutual utilization appropriateness of installation environment contribution to r&d activity 1/5 3 1 1/5 utilization of research equipment 1 1 3 1/3 contribution to mutual utilization 1/3 3 1 1/3 appropriateness of installation environment 1/3 1/5 1/5 1 total 1.86 7.20 5.20 1.86 table 12. weights of the evaluation criteria evaluation criteria contribution to r&d activity utilization of research equipment contribution to mutual utilization appropriateness of installation environment weight contribution to r&d activity 0.11 0.42 0.19 0.11 0.21 utilization of research equipment 0.54 0.14 0.58 0.18 0.36 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 19 contribution to mutual utilization 0.18 0.42 0.19 0.18 0.24 appropriateness of installation environment 0.18 0.03 0.04 0.54 0.20 table 13. elements of the evaluation criteria classification contribution to r&d activity utilization of research equipment contribution to mutual utilization appropriateness of installation environment expensive research equipment 0.21 0.79 0.79 0.17 low-cost research equipment 0.79 0.21 0.21 0.83 figure 3. ahp application model for evaluation index improvement for research equipment relocation. 0.18 0.26 0.34 0.23 0.457 0.552 evaluation index contribution to mutual utilization contribution to r&d activity utilization of research equipment low-cost research equipment expensive research equipment appropriateness of installation environment d. yoon / european journal of government and economics 8(1), june 2019, 6-29 20 table 14. random index (ri) value according to n change n 1 2 3 4 5 6 7 8 9 10 ri 0.00 0.00 0.58 0.90 1.12 1.24 1.32 1.41 1.45 1.51 table 15. calculation of weight matrix evaluation criteria contribution to r&d activity utilization of research equipment contribution to mutual utilization appropriateness of installation environment total of row 0.21 0.36 0.24 0.20 contribution to r&d activity 0.07 0.84 0.20 0.04 1.14 utilization of research equipment 0.33 0.28 0.60 0.06 1.27 contribution to mutual utilization 0.11 0.84 0.20 0.06 1.21 appropriateness of installation environment 0.11 0.06 0.04 0.18 0.38 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 21 as the research results showed a 0.007 consistency ratio, they have validity in the decision. 4.1. evaluation index improvement research design based on the research results, an evaluation index for research equipment relocation was designed for objective and fair evaluation, for a number of researchers who applied for the transfer of research equipment to them free of charge. the evaluation index was designed to allow research equipment to be relocated to researchers who can manage and operate them and enable researchers to conduct excellent researches. the existing evaluation index was determined through surveys among equipment experts, and the research equipment relocation review committee. the evaluation index for the relocation of existing research equipment is shown in table 16. in the past, one to three researchers applied for one research equipment. the evaluation index was designed for use in the evaluation of the researcher to be selected among the three or more researchers who applied for the transfer of one research equipment to them, as shown in table 16. the basic principle of the research design is that the design should benefit more institutions and that the selection of expensive equipment should not be restricted while low-cost equipment should be limited to a maximum of 5 points (usd18,475) for research institutes or researchers. also selected are the institutions that can utilize and maintain research equipment well after their relocation. the operation control performance rate in the 3rdand 5th-year plans is checked after an agreement is reached for the relocation of research equipment, and the relocation to another research institution or researcher is executed when it is not sufficient for operation and management. if the 3rd-year operation control performance rate is 70% and the 5th-year rate is below 90%, the research equipment will be relocated to other research institutes or researchers. young researchers, however, are given a grace period for the third year. considering that the research equipment that was used in this study was general-purpose, young researchers were selected rather than higher-performing researchers. the target research equipment was divided into expensive and low-cost research equipment. the expensive research equipment was worth more than usd18,475, and the low-cost research equipment was worth less than 1.14 1.27 1.21 0.38 0.21 0.36 0.24 0.20 4.02 4 xi wimax n λ         + + +        ⋅         = = =∑ ( ) 4.02 4 0.02 0.006 1 4 1 3 max nci n λ − − = = = = − − 0.006 0.007 0.90 cicr ri = = = d. yoon / european journal of government and economics 8(1), june 2019, 6-29 22 usd18,475. for the expensive research equipment, the priority levels were determined based on the rate of research equipment utilization, contribution to mutual utilization, contribution to r&d activity, and appropriateness of the installation environment. for the low-cost research equipment, on the other hand, the priority levels were determined based on the appropriateness of the installation environment, the contribution to r&d activity, the rate of research equipment utilization, and the contribution to mutual utilization. the evaluation criteria design for the relocation of research equipment is shown in table 18. the evaluation criteria for the relocation of expensive and low-cost research equipment were classified. the evaluation criteria for the relocation of research equipment are shown in table 18. a summary of the salient points of the discussion on the relocation of research equipment is shown in table 19. d. yoon / european journal of government and economics 8(1), june 2019, 6-29 23 table 16. evaluation index for relocation of existing research equipment classification evaluation index operation and technical review of research equipment are repair and relocation possible? is normal operation possible? is it possible to operate after repair? can parts be secured for equipment repair? research equipment operating life how many years is the research equipment expected to run? relocation period how long will the relocation take? research equipment education method is it necessary to ask the equipment supplier for training on consignment? is it necessary to ask an expert for training on consignment? does the researcher educate himself? is maintenance training necessary? installation environment review is the space suitable for the size of the research equipment? is electrical power available for the research equipment? is water available for the research equipment? is vibration safety secured for the research equipment? is noise safety secured for the research equipment? is dust safety secured for research equipment? are heating and cooling ready? is a fire installation system in place? are there any precautions against dangerous goods? is the locking mechanism in place? is there an internet communication facility? are there any obstacles or risks to the research equipment? operation environment review how many people are operating the equipment? do the operating manpower need to be professionals? how much is the operating cost? d. yoon / european journal of government and economics 8(1), june 2019, 6-29 24 table 17. evaluation criteria design for relocation of research equipment evaluation item detailed evaluation criteria contribution to r&d activity research results (paper, book, patent, etc.) research career (major position, performing research task) whether young scientist or not research planning through research equipment utilization of research equipment research equipment (use number, user, sample) research equipment paper number (sci, non-sci) research equipment patent number (domestic, triad patent families) research equipment technology transfer number number of research equipment education programs (training program, manpower) contribution to mutual utilization mutual utilization system implementation plan and schedule of mutual utilization service performance of mutual utilization service (internal, external) appropriateness of installation environment operation cost of research equipment (operation cost, repair and maintenance cost) manpower of research equipment (regular employee, non-regular employee) others preferential treatment for young scientist (low-cost research equipment) preferential treatment for mutually utilized integrated facilities (expensive research equipment) d. yoon / european journal of government and economics 8(1), june 2019, 6-29 25 table 18. evaluation criteria for relocation of research equipment evaluation item detailed evaluation criteria expensive research equipment low-cost research equipment weight score weight score quantitative evaluation qualitative evaluation quantitative evaluation qualitative evaluation contribution to r&d activity research results (paper, book, patent, etc.) research career (major position, performing research task) whether young scientist or not research planning through research equipment 20 20 30 10 (young scientist) 20 utilization of research equipment research equipment (use number, user, sample) research equipment paper number (sci, non-sci) research equipment patent number (domestic, triad patent families) research equipment technology transfer number number of research equipment education programs (training program, manpower) 35 20 15 20 10 10 contribution to mutual utilization research equipment mutual utilization system research equipment mutual utilization service implementation plan and schedule performance of research equipment mutual utilization service (internal, external) 30 10 20 10 10 appropriateness of installation environment operation cost of research equipment (operation cost, repair and maintenance cost) research equipment manpower (regular employee, non-regular employee) 15 10 5 40 30 10 total 100 40 60 100 60 40 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 26 table 19. discussion for relocation of research equipment evaluation item discussion contribution to r&d activity is the use of research equipment necessary to achieve the research goal? is the research history sufficient, and is the research plan appropriate? should new researchers be selected first? utilization of research equipment is research equipment available according to the plan? is performance management related to research equipment utilization possible? is it continuously available for long periods of time? contribution to mutual utilization is the actual mutual utilization service possible? is it possible to contribute to the promotion of research equipment mutual utilization? appropriateness of installation environment does the research institute have adequate space and environment for research equipment installation? is the research equipment operation cost and manpower secured? d. yoon / european journal of government and economics 8(1), june 2019, 6-29 27 5. conclusions and policy implication in the study, the basic principle and detailed evaluation criteria for the relocation process of research equipment were established. the basic principle of the research design was to benefit more institutions and researchers. moreover, the institutions that could utilize and maintain the research equipment well after their relocation were selected. considering that the target research equipment was the general-purpose one, young researchers were selected rather than higher-performing researchers. duplication was prevented through the advance review of many research equipment. based on the research results, an evaluation index for research equipment relocation was designed for the objective and fair evaluation of the researchers who applied for the transfer of research equipment to them free of charge. as the basic principle of the research design, as mentioned above, was to benefit more institutions, expensive equipment was not restricted from selection, and low-cost equipment was limited to a maximum of 5 points (usd18,475) for research institutes or researchers. moreover, the institutions that could utilize and maintain the research equipment well after their relocation were selected. the target research equipment was divided into expensive and low-cost research equipment, with the former worth more than usd18,475 and the latter worth less than usd18,475. for the expensive research equipment, the priority levels were based on the rate of research equipment utilization, the contribution to mutual utilization, the contribution to r&d activity, and the appropriateness of the installation environment. for the low-cost research equipment, on the other hand, the priority levels were based on the appropriateness of the installation environment, the contribution to r&d activity, the rate of research equipment utilization, and the contribution to mutual utilization. an attempt was made to relocate research equipment to research institutions and researchers by reflecting the characteristics of the research equipment. it was found that the rate of research equipment utilization is the most important element for the expensive research equipment and that the appropriateness of the installation environment is the most important element for the lowcost research equipment. in this paper, evaluation index improvement for research equipment relocation is presented and discussed to boost the effectiveness of research equipment management. we focused on an efficient management method of research equipment through evaluation index improvement for research equipment relocation. the efficient management of research equipment can improve the efficient management of the r&d budget. it can promote government r&d innovation through government r&d investments in various research fields. the ministry of science and ict (mist) of south korea has transferred 153 idle and underutilized research equipment free of charge to universities, laboratories, nonprofit institutions, etc. the government also allowed researchers to freely apply for research equipment to be transferred to them free of charge. there were so many researchers who applied for research equipment to be transferred to them free of charge that it was necessary to develop an evaluation index for the objective and fair relocation of research equipment. therefore, an evaluation index for research equipment relocation was designed for objective and fair evaluation. for d. yoon / european journal of government and economics 8(1), june 2019, 6-29 28 the advancement of south korea’s science and technology infrastructure, the management system through life cycle system installation and operation of research equipment is essential. the relocation of research equipment can increase the efficiency of the national r&d investment. the evaluation method for equipment relocation can replicable and applicable to various research equipment in many countries. it is hoped that the study findings will be very useful and will contribute greatly to the professors, researchers, and policymakers involved in science and technology policymaking and r&d. references amanatidou, e., cunningham, p., gök, a. & garefi, i. (2014). using evaluation research as a means for policy analysis in a 'new' mission-oriented policy context. minerva 52(4), 419-438. doi: https://doi.org/10.1007/s11024-014-9258-x arnold, e. (2004). evaluating research and innovation policy: a systems world needs systems evaluations. research policy 13(1), 3-17. doi: https://doi.org/10.3152/147154404781776509 barker, k. (2007). the uk research assessment exercise: the evolution of a national research evaluation system. research policy 16(1), 3-12. doi: https://doi.org/10.3152/095820207x190674 bozeman, b., & youtie, j. (2017). socio-economic impacts and public value of government-funded research: lessons from four us national science foundation initiatives. research policy 46(8), 1387-1398. doi: https://doi.org/10.1016/j.respol.2017.06.003 brofoss, k. e. (1998). the research council of norway's use of research evaluation: an assessment of research evaluation as a strategic tool. research policy 7(3), 134–140. doi: https://doi.org/10.1093/rev/7.3.134 cerqueti, r., marazzina, d., & ventura, m. (2016). optimal investment in research and development under uncertainty. journal of optimization theory and applications 168(1), 296-309. doi: https://doi.org/10.1007/s10957-015-0751-7 chang, s., li, y., & gao, f. (2016). the impact of delaying an investment decision on r&d projects in real option game. chaos, solitons, and fractals 87, 182-189. doi: https://doi.org/10.1016/j.chaos.2016.03.035 cunningham, p. (2008). monitoring and analysis of policies and public financing instruments conducive to higher levels of r&d investments. the “policy mix” project: thematic report: governance, european commission, 15. cunningham, p. (2008). inno-learning platform mini-studies. european trend chart on innovation policy input report, 6. dimos, c., & pugh, g. (2016). the effectiveness of r&d subsidies: a meta-regression analysis of the evaluation literature. research policy 45(4), 797-815. doi: https://doi.org/10.1016/j.respol.2016.01.002 edler, j., cunningham, p., gök, a., & shapira, p. (2016). chapter 18: conclusions: evidence on the https://doi.org/10.1007/s11024-014-9258-x https://doi.org/10.3152/147154404781776509 javascript:; https://doi.org/10.3152/095820207x190674 https://doi.org/10.1016/j.respol.2017.06.003 https://doi.org/10.1093/rev/7.3.134 https://doi.org/10.1007/s10957-015-0751-7 https://doi.org/10.1016/j.chaos.2016.03.035 https://doi.org/10.1016/j.respol.2016.01.002 d. yoon / european journal of government and economics 8(1), june 2019, 6-29 29 effectiveness of innovation policy intervention in edler, j., cunningham, p., gök, a., & shapira, p. (eds.) handbook of innovation policy impact (pp.543–564). edward elgar publishing. doi: https://doi.org/10.4337/9781784711856.00025 kauffman, r. j., liu, j., & ma, d. (2015). technology investment decision-making under uncertainty. information technology and management 16(2), 153-172. doi: https://doi.org/10.1007/s10799-014-0212-2 lewis, c. m.,& tan, y., 2016, debt-equity choices, r&d investment and market timing. journal of financial economics, 119(3), 599-610. doi: https://doi.org/10.1016/j.jfineco.2016.01.017 lynskey, m. j. (2016). r&d investment in new technology-based firms: strategic and entrepreneurial dynamics and the impact of universities. industry and higher education 30(4), 278-291. doi: https://doi.org/10.1177/0950422216660254 meyer-krahmer, f., & reiss, t. (1992). ex ante evaluation and technology assessment — two emerging elements of technology policy evaluation. research policy 2(1), 47–54. doi: https://doi.org/10.1093/rev/2.1.47 rigby, j., cunningham, p., li, y., yeow, j, edler, j., cox, d., & roper, s. (2016). a review of the small business research initiative. a report commissioned by innovate uk final report august 2015, 25. saaty, t. l. (1980) the analytic hierarchy process. mcgraw-hill, new york. doi: https://doi.org/10.21236/ada214804 sanz-menéndez, l. (1995). research actors and the state: research evaluation and evaluation of science and technology policies in spain. research policy 5(1), 79–88. doi: https://doi.org/10.1093/rev/5.1.79 schatz, d., & bashroush, r. (2017). economic valuation for information security investment: a systematic literature review. information systems frontiers, 19(5), 1205-1228. doi: https://doi.org/10.1007/s10796-016-9648-8 schröder, s., welter, f., leisten, i., richert, a., & jeschke, s. (2014). research performance and evaluation—empirical results from collaborative research centers and clusters of excellence in germany. research policy 23(3), 221–232. doi: https://doi.org/10.1093/reseval/rvu010 silva, c. m. & henriques, l. (1995). r&d evaluation in portugal. research policy 5(1), 89–97. doi: https://doi.org/10.1093/rev/5.1.89 stern, e. (1993). ongoing and participative evaluation: purpose, design and role in the evaluation of a large-scale r&d programme. research policy 3(2), 75–82. doi: https://doi.org/10.1093/rev/3.2.75 https://doi.org/10.4337/9781784711856.00025 https://doi.org/10.1007/s10799-014-0212-2 https://doi.org/10.1016/j.jfineco.2016.01.017 https://doi.org/10.1177/0950422216660254 https://doi.org/10.1093/rev/2.1.47 https://doi.org/10.21236/ada214804 https://doi.org/10.1093/rev/5.1.79 https://doi.org/10.1007/s10796-016-9648-8 javascript:; javascript:; javascript:; javascript:; javascript:; https://doi.org/10.1093/reseval/rvu010 javascript:; javascript:; https://doi.org/10.1093/rev/5.1.89 https://doi.org/10.1093/rev/3.2.75 front81 ejge oficial 8-1 contents81 number 8, issue 1, june 2019 4572 perceptions of gender equality in post-crisis europe european journal of government and economics 6(1), june 2017, 59-77. european journal of government and economics journal homepage: www.ejge.org issn: 2254-7088 perceptions of gender equality in post-crisis europe natalia soboleva laboratory for comparative social research, national research university, higher school of economics, russian federation. email: nsoboleva@hse.ru article history. received 19 july 2016; first revision required 3 january 2017; accepted 24 april 2017 abstract. the paper aims to compare the perceptions of gender equality of individuals more or less affected by economic crisis in europe. crisis touched the economy of most european countries but to a different extent. special focus is given to the perceptions of gender equality of vulnerable groups (female, lesser-educated, one-adult households with children). the data is eurobarometer 2011. the sample is limited to respondents aged 18-65. according to the results of multilevel regression analysis, those who have suffered from crisis assess lower the current level of gender equality whereas perceptions of gender equality do not differ depending on the effect of crisis upon the country. women assess gender equality more positively compared to men. those who live in one-adult households with children have higher perceptions of gender equality compared to those who live in other types of households. the discrepancy between lesser-educated and higher educated is larger in countries that suffered less from crisis. however, when the change in gdp per capita is taken as a measure of crisis the effects for family structure and education are not robust. keywords. gender equality; economic crisis; cross-cultural comparison; europe; vulnerable groups jel classification. j16; j21; j24: j81; c02 1. introduction the article aims at revealing perceptions of gender equality of the individuals affected by economic crisis of 2008 to different extent. i disclose the impact of the changes in the current economic situation upon the assessment of gender equality in the labor market. tolerance toward gender equality is one of the fundamental components of modernization (welzel, 2013). according to the oecd report, gender equality fosters economic growth in education and the labor market (oecd 2012). however, actual gender equality is not possible when traditional gender attitudes prevail. gender attitudes reflect the relation to the position of women in the labor market and society in general (inglehart and welzel, 2005: 111). bergh defines gender attitudes as ‘normative beliefs about what gender relations in society should be like, or the extent to which a person supports the norm of gender equality’ (bergh, 2006: 6). to be more specific, positive attitudes towards traditional gender roles regard men as a breadwinner and women as a homemaker, whereas positive attitudes towards non-traditional gender roles suppose that men and women should share labor market and household duties http://www.ejge.org/ 60 soboleva / european journal of government and economics 6(1), 59-77 (voicu and constantin, 2016). in current paper, i focus more on perceptions of gender equality in european countries, in other words, on individual assessments of gender equality across europe. values and attitudes are largely dependent upon the living conditions. the change in the objective conditions of living in turn leads to the change of attitudes. hence, gender attitudes are formed largely by a current economic situation. economic and political collapses often result in shift to values and attitudes that are more traditional (inglehart and welzel, 2005: 111). economic crisis of 2008-2009 was not as huge but still had an impact upon the conditions of living of people in different countries (world bank indicators). furthermore, it is also likely to affect gender norms (elson 2010; rubery, 2014). the research on gender attitudes is vast. some authors analyzed the structure of gender attitudes and its change it time (alesina and giulino, 2010; bolzendahl and myers, 2004; braun and gloeckner-rist, 2011; cunningham, 2008; guiso, 2003; guveli, need and de graaf, 2007, pampel, 2011; voicu and tufis, 2012; voicu and constantin, 2016). in addition, there is research on the impact of crisis on labor market outcomes in gender perspective (rubery, 2014; bettio and verashchagina, 2014; karamessini and rubery, 2014; seguino, 2009). however, there the issue of perceptions of gender equality for individuals more or less affected by economic crisis of 2008-2009 is not yet covered. in this paper, i analyze a somewhat different aspect, namely perceptions of gender equality in europe after the crisis. crisis is less likely to change the attitudes and perceptions of gender equality completely than to strengthen the current tendencies mostly making them more traditional. therefore, the assessment of the perceptions of gender equality in europe after the economic crisis is a very relevant issue. the economic crisis could be truly called “gendered” because the crisis influenced differently men and women and in some way reinforced gender equality (walby, 2015). economic crisis and recession should change not only objective positions of men and women in the labor market (namely contribution to family income, caring work and access to benefits) but also social norms regarding gender role-attitudes in direction that is more traditional. women are likely to reduce their participation in the labor market (rubery, 2014: 20). furthermore, crisis is likely to have different impact upon various social groups (seguino, 2009; elson, 2010; rubery, 2014). in this respect, the study of the state of gender attitudes after the economic crisis in europe is of great importance. i reveal the effect of crisis both on country level as the change in objective parameters and, on micro-level, i measure the perceived impact of crisis on the individual. hence, the main research objectives are as follows. first, i reveal the state of perceptions of gender equality in post-crisis europe. second, i analyze whether perceptions of gender equality differ for people more or less affected by crisis. finally, i focus upon perceptions of gender equality of vulnerable groups (women, lesser educated, living in one-adult households with children) in post-crisis europe. 61 soboleva / european journal of government and economics 6(1), 59-77 2. the negative impact of crisis the crisis had a negative impact upon the economies of most of the european countries. while in 2008 there was no effect of crisis upon macro-level indicators, most countries experienced tremendous decline in gdp per capita and increase in unemployment rate in 2009. in 2010 the unemployment situation in most countries began to normalize and the level of gdp per capita reached the level of 2008. these two indicators tap slightly different aspects of crisis. change in gdp per capita reflects the general decline of the economic development of the country. change in unemployment rate refers specifically to the situation in the labor market and firstly touches the economically active population. for the paper, the unemployment indicator is of high importance because labor market status is closely connected with gender attitudes and could to a large extent influence them. however, the correlation between the change in gdp and the change in unemployment rate is rather high (-0.503). in figure 1, it could be seen that the most of the countries are located on the same line. however, there are some outliers. the united kingdom experienced a substantial drop in gdp per capita but a low increase in the unemployment rate. estonia and especially lithuania, on the contrary, suffered from high increases in unemployment rates and rather modest change to the gdr.1 figure 1. change of gdp per capita and change in unemployment rate between 2007 and 2010 (in %). source: world bank indicators: http://data.worldbank.org/indicator 1 if we look at gdp per capita change and unemployment rate change between 2008 and 2009 the situation is a bit different. the correlation between two indicators is a bit higher (-0.608). consequently, more countries are located on the same line. however, there is a group of outliers that experienced rather dramatic decline in gdp per capita and rather low increase in unemployment rate, namely sweden, united kingdom, poland, romania and hungary. http://data.worldbank.org/indicator 62 soboleva / european journal of government and economics 6(1), 59-77 the other important point is the diverse effect of crisis upon different groups in society. some authors regard the differences of consequences of crisis for men and women (elson, 2010; walby, 2015). according to walby (2015), the nature of the crisis is gendered. four stages of the crisis could be pointed out namely finance, real economy, labor market and democracy. all four stages could be characterized by the dominance of men. women are absent from decisionmaking process in all the phases of the crisis (elson, 2010; walby, 2015). elson (2010) analyzed the gendered aspects of economic crisis in developing countries. in her analysis, she distinguished gender numbers and gender norms in all economic spheres such as finance, production (formal and informal) and reproduction. in the financial sphere women face more difficulties in commercial bank and crisis may influence female risk-aversion because they are mainly responsible for children’s well-being. in production, the consequences of crisis include the losses of both female and male employment. in general, whether men and women are first to suffer because of the crisis depends on industry. however, if both men and women are employed in the field of work ‘male breadwinner norms’ are likely to prevail. it is considered in the society that it is men’s duty to earn money and consequently women are the first to be fired. governmental policy in its turn is often also more in favor of men. the situation in financial and production spheres are likely to affect the sphere of reproduction. in some cases, sons will have preferences in getting education and women will be firstly responsible for the household. government social policy may play a role in formation of gender norms (elson, 2010). seguino (2009) demonstrated these differences by comparing developing and developed countries. in developing countries, women are more likely to suffer more from crisis because they are more often employed in export manufacturing industries. in developed countries (for instance, us) men are often working in manufacture and construction and hence are likely to be the first to lose jobs. women are likely to lose jobs a bit later when cuts in public service are made. at the same time, women are likely to suffer more from crisis as the gender roles are such that in the case of job scarcity men have more right to have a job. although in developed countries gender attitudes are more egalitarian, still there is evidence that some women withdraw from the labor market (become economically inactive). hence, women are more likely to be targeted by job layoffs and hence the policy of recovery should primarily address women (seguino, 2009). karamessini and rubery (2014) revealed a number of common tendencies of impact of crisis upon female employment in europe and the us. the first one is the growth of this female labor force participation rate in spite of the fall of female employment. the reason of this is that many men lost their jobs. hence, some women became sole breadwinners, other women entered the labor market and switched from being economically inactive to being unemployed. “their growing attachment to careers for reasons of economic independence, personal fulfillment or economic necessity (to secure the family income or to improve living standards, thus seems irreversible in advanced economies even in the context of major economic crisis such as the one we are living through”. second, the crisis led to the narrowing gender inequality in employment and by employment type (part-time and temporary jobs), “because of greater job 63 soboleva / european journal of government and economics 6(1), 59-77 loss and spread of flexible forms of employment among men.” to the third consequence refer austerity and fiscal consolidation policies along with an employment crisis in the public sector in the second phase of the crisis. those employed in the public sector experienced “employment contraction, pay freezes and cuts and deterioration of working conditions”. the public sector is the most popular employer for women (first of all for the more highly educated) (karamessini and rubery, 2014: 346). bettio and verashchagina (2014: 58-59) focused on explanation of the women’s behavior in the labor market during the period of crisis. ‘added’ workers enter the labor market when the heads of the household lose their jobs. according to the segmentation hypothesis, women are more likely to be employed in secondary jobs, and hence are more likely to suffer during the crisis. according to the gender segregation hypothesis, female occupations tend to be more traditional and are likely to be less affected by crisis however, the authors mainly studied the impact of crisis on actual gender equality. rubery (2014: 20)stresses that crisis should touch not only the objective position of men and women in the family and labor market including “contribution to family income, to caring work and access to benefits and public services but also social norms with respect to the appropriate and expected gender roles”. rubery’s assertion fits into the modernization theory of inglehart and welzel. the major economic and political crises, collapses of political regimes and social systems lead to the shift from self-expression to survival values, and to decrease of subjective well-being. such events make life insecure and unpredictable. therefore, in eastern ex-communist countries survival values are more spread than in low-income countries (inglehart and welzel, 2005: 111). however, for instance, it was shown that in russia subjective well-being increased to 2011 but it still was lower than in 1981 (inglehart et al., 2013). survival values are associated with low support of gender equality, whereas self-expression values assume high support of gender equality. gender attitudes are quite polarized among different countries and social groups. (inglehart and welzel, 2005: 54). following the logic above, my first hypothesis is as follows. • individuals who were more affected by crisis or live in countries that were more affected by crisis should have more negative perceptions about current state of gender equality. 3. formation of gender attitudes: theoretical background the causal relationship between gender attitudes and behavior could be regarded in different ways. some researchers view gender attitudes as a cause that influences the gender behavior (campa, 2009; fortin, 2005; stickney and konrad, 2007). others, on the contrary, regard them as an effect (bolzendahl and myers, 2004; braun and gloeckner-rist, 2011; guiso et al., 2003; voicu and tufis, 2012). in myframework, the second approach is more relevant because the economic crisis on individual and country level worsens the objective situation and hence should have an impact upon attitudes and perceptions. 64 soboleva / european journal of government and economics 6(1), 59-77 the impact of current living conditions upon gender attitudes could be explained with the help of several approaches. in accordance with the social role theory, people adopt the attitudes which are consistent with the roles they occupy (eagly and karau, 2002). this could also be explained by the cognitive dissonance theory. when a social role does not match a gender-role attitude, the person is likely to change either the role or the attitude. when the behavioral change is impossible, people modify their attitudes (kroska, 1997). in research of gender attitudes interest-based and exposure-based perspectives of analysis are distinguished. in the framework of interest-based approach, the individuals that benefit from egalitarian gender attitudes usually have more egalitarian work-related gender attitudes. from this perspective, some categories of society (for instance, males, highly educated and highly paid jobs) gain more from being employed. according to exposure-based approach, individuals change their ideas and attitudes when they meet the situation that discords with these ideas. in respect to gender issues it is possible to point out work-force participation, education as well as mother’s work status and mother’s education (bolzendahl and myers, 2004; pampel, 2011). in line with this approach, the gender attitudes of a person are formed in the environment and social groups (family, education, and job). if a person enters a different environment (for example, as a result of crisis), he is likely to change the attitudes. if we follow these conceptions, when the behavior of individuals or their families becomes more egalitarian, their gender attitudes are also likely to change to be more egalitarian in order to correspond to the behavior. the conception by west and zimmerman “doing gender” represents a different framework. the authors emphasize that most differences between genders are not biological but socially constructed. after they are constructed, they sustain and strengthen biological differences. hence, when gender identity is somehow threatened a person is likely to stress it and shift to even more traditional gender identity (west and zimmerman, 1987). voicu and constantin demonstrated that “doing gender” conception could explain gender attitudes in the situation of unemployment by studying the impact of employment status in couples upon gender attitudes. when crisis of identity happens, individuals and couples tend to emphasize their traditional gender roles. female employment does not always foster egalitarian gender attitudes and that partner’s employment status should be taken into account. besides, the income inequality in the country (gini) also affects gender attitudes (voicu and constantin, 2013). i can suggest that the effect of crisis could be the same as the effect of unemployment. in case of crisis, there is usually an involuntary shift in behavior and hence the individuals may tend to emphasize their traditional gender roles. in this respect, it is likely that those who have suffered more after the crisis are likely to become more traditionally oriented. it could concern women, single parents living with children and the lesser educated. as it was shown above, women are more likely to lose jobs as well as shift from being economically inactive to unemployed in order to support the families. this involuntary change according to “doing gender conception” could influence gender attitudes. i can thus derive the following hypothesis. 65 soboleva / european journal of government and economics 6(1), 59-77 • female perceptions towards gender equality are proved to be more egalitarian compared to their male counterparts (alwin et al., 1992; bolzendahl and myers, 2004); however, in countries where the situation worsened more after the crisis female perceptions of gender equality should be more traditional and should differ less from male perceptions. single parents living with children are likely to start with more egalitarian gender attitudes because they are obliged to combine work and family responsibilities (independently from gender). during the period of crisis, this category is likely to be more affected compared to other categories because single parents have to earn money for themselves and for children. this could make them stress the traditional gender identity and lead to the change in attitudes. hence, my next hypothesis is as follows: • perceptions of gender equality of those who live in single-adult households with children are likely to be more positive compared to those of other types of families. however, in countries that were more influenced by crisis they are likely to be less positive. the same logic could be applied to lesser-educated groups that are most likely to suffer as the result or crisis. for them there are fewer possibilities in the labor market so they are more likely to become unemployed. hence, during the crisis they are likely to stress their traditional gender identity (west and zimmerman, 1987). the gender attitudes of lesser-educated groups are likely to become more traditional after the crisis. • perceptions of gender equality of lesser-educated groups are likely to be less positive in the countries where the economy was damaged more by crisis compared to the countries that were less affected by crisis. 4. data, measures, and methods 4.1 data to reveal the perceptions of gender equality depending upon the impact of crisis i use eurobarometer 76.1 dataset of 2011, which covers the following areas: the economic and financial crisis, european parliament energy policy, financial services, corruption, development aid, and gender equality. the eurobarometer is a unique cross-national and cross-temporal survey program conducted on behalf of the european commission (for more information see https://www.icpsr.umich.edu/icpsrweb/icpsr/studies/34552). i have chosen this dataset for a number of reasons. first, it covers a wide range of european countries. the following 27 countries are covered in the chosen dataset and included into analysis: belgium, bulgaria, czech republic, denmark, germany, estonia, ireland, greece, spain, france, italy, cyprus, 66 soboleva / european journal of government and economics 6(1), 59-77 latvia, lithuania, luxembourg, hungary, malta, netherlands, austria, poland, portugal, romania, slovenia, slovakia, finland, sweden and united kingdom. second, it allows us to estimate the impact of crisis not only on country level (by adding macro level) but also on individual level. third, the socio-demographic variables are relevant for testing the hypotheses. in the current research, i am primarily interested in economically active population because they are more likely either to lose a job or begin to work as a necessity during the crisis. i limited the sample by age including those who are 18 – 65 years old. multilevel regression model with maximum likelihood estimation serves as the main research method. for my research, it is a relevant method because i have to distinguish two levels of analysis (albright & marinova, 2010; hox, 2010; snijders & bosker, 1999). individuals are nested in countries. level 1 is individual level and level 2 is country level. i include in the model both individual level and country level indicators. 4.2. measures the main dependent variable is the index of perceptions of gender equality. this index consists of three indicators: 1) women are less interested than men in positions of responsibility; 2) women are less willing than men to fight to make a career for themselves; 3) women do not always have the necessary qualities and skills to fill positions of responsibility. the respondents had to agree or disagree with the statements above. the scaling categories were as follows: totally agree (1), tend to agree (2), tend to disagree (3) and totally disagree (4). the indicators were rescaled from 0 to 10, summed up and then the index was divided by 3. for index construction, first i calculated 2-tailed correlations between items. the correlation between the items 1 and 2 is 0.568**2, the correlation between items 1 and 3 is 0.426** and the correlation between items 2 and 0.387**. then i ran exploratory factor analysis for both pooled data and by country. these three items form in exploratory factor analysis a single factor. higher values indicate more negative perceptions of gender equality. cronbach alpha for three items is 0.719. this index could be named ‘women are not less competent for public duties’. these items tap attitudes regarding female involvement in public space namely work and public duties. this aspect is quite relevant for the focus of the current research because i look at perceptions of women’s position and chances in the labor market in countries more or less suffered by crisis. in the rest of the paper, i will name this index ‘perceptions of gender equality. the agreement with the statements reflects negative perceptions of gender equality. disagreement with the statements could be interpreted as that women are not less suitable for the positions of responsibility than men. the main independent variables on individual level are the following. the indicator of effect of crisis on individual level was measured by the statement: “you have lost your job \ your partner (husband or wife, partner, etc.) has lost his / her job. the baseline includes the 2**p-value<0.01 67 soboleva / european journal of government and economics 6(1), 59-77 respondents to whom this has not happened at all. two dummy variables comprise the following responses: “yes, as a direct consequence of the crisis” and “yes, but not as a direct consequence of the crisis”. the difference between these two dummies is vague to some extent, because it is likely that crisis has contributed to job loss of the people who do not admit it or when it was just not the direct reason. however, i kept the two distinct variables in order to see the difference between these two categories of respondents. furthermore, the following control variables are taken into account. gender is coded as a dichotomous variable (women – 1, men – 0). i distinguish four types of the households: 2-parent household without children (baseline), single-parent household without children, single household with children and 2-parent household with children. single-parent household with children is likely to be the most vulnerable category. by the level of education, respondents were divided into two groups. those who have finished their education after being 20-year old or are still studying refer to highor middleeducated. low-educated are those who finished education before the age of 20 or have not received full-time education. low-educated is a very vulnerable category and are likely to suffer during the crisis. unfortunately, there is no better variable for education in the data and the finer educational categories cannot be distinguished. furthermore, i distinguished three categories by their labor market status: high qualified (baseline), low or middle qualified or not working. it is likely that low qualified and who do not work would be more shaken by crisis than highly qualified. to the high qualified refer the following categories of workers: professionals (lawyer, medical practitioner, accountant, architect, etc.), business proprietors, owner (full or partner) of a company, employed professional (employed doctor, lawyer, accountant, architect), general management, director or top management (managing directors, director general, other director) and middle management, other management (department head, junior manager, teacher, technician). low and middle qualified comprise farmers, fishermen, owners of a shop, craftsmen, other self-employed persons, employed working mainly at a desk, travelling (salesmen, driver, etc.) or in a service job (hospital, restaurant, police, fireman, etc.), supervisors, skilled manual workers, unskilled manual workers and servants. to the category “not working” refer those who never had a job, unemployed or temporarily not working, retired or unable to work through illness as well as students and responsible for ordinary shopping and looking after the home, or without any current occupation, not working. in addition, i use as a control variable a subjective assessment of level in society as a proxy for the material well-being (income). unfortunately, the question about actual income is not available in the dataset. however, the level in society is a more permanent category and is less likely to be influenced by a current situation (for example, the drop of material well-being after the crisis). the question was as follows: “on the following, step ‘1’ corresponds to “the lowest level in the society”; step ‘10’ corresponds to “the highest level in the society”. could you tell me on which step you would place yourself?” finally, respondents were divided into four age groups: 18-24 year old (baseline), 25-39, 4068 soboleva / european journal of government and economics 6(1), 59-77 54 and 55-65. on the contextual level, i use gdp per capita from year 2011 in the logarithmic form to control the level of economic development and modernization. to test the impact of crisis on country level, i use the drop of gdp between 2007 and 2010 (calculated as (gdp2010gdp2007) / gdp2007)) and the drop of gdp between 2008 and 2009 (calculated as (gdp2009-gdp2008) / gdp2008)). i chose the broader period as the main indicator because it is important whether the country managed to overcome the consequences of crisis. in addition, these two measures allow estimating the robustness of the results. the second indicator of impact of crisis is the increase in unemployment rate (ur) between 2007 and 2010 ((ur2010 – ur2007) / ur2007) and between 2008 and 2009 ((ur2009 – ur2008) / ur2008). increase in the unemployment rate is to a lesser extent connected to economic development and hence to some extent is a more accurate measure of the effect of crisis. again, estimation change in unemployment rate for two periods of crisis will show the robustness of the results. using both change in gdp per capita and change in unemployment rate enables me to check for the stability of results. i would like to emphasize that gdp per capita decreases as a result of crisis whereas the unemployment rate increases. therefore, the interaction effects will have different signs. 5. results in the last part of the paper, i will consider the main results of the analysis. first, i ran a zero model where perceptions of gender equality serve a dependent variable. the intercept in this model equals 6.866***, icc is 0.096.3 it means that 9.6 per cent of variance can be explained differences across countries and shows that multilevel analysis is an appropriate method. in the next four models, i included gdp in logarithmic form (lngdp) in 2011 and change in gdp between 2007 and 2010. it is possible to see from all four models that in countries with higher gdp per capita perceptions of gender equality are more positive. however, the effect of gdp per capita is not very strong. it could be due to the fact the discrepancy in gdp per capita in european economies is not very salient. the change in gdp per capita between 2007 and 2010 does not affect perceptions of gender equality (model 1). on the contrary, the effect of crisis on the individual level has a great impact upon perceptions of gender equality. people who or whose partners have lost their jobs perceive gender equality more negatively. the effect is especially strong for those who claim to have lost the job not because of crisis. among the control variables, the main predictors of perceptions of gender equality are gender, age and low level of education. women have more positive perceptions of gender equality, whereas those who finished education before 20 or no full-time and are 55-65 years old perceive gender equality more negatively. surprisingly, those who not work or are low or 3 aic=75196, log likelihood= 44540, n (individuals) = 19729, n (countries) = 27. 69 soboleva / european journal of government and economics 6(1), 59-77 middle qualified do not differ in the perceptions of gender equality with the highly qualified. table 1. multilevel regression models: dependent variable – index of gender attitudes (unstandardized regression coefficients). macro level – gdp per capita and gdp change from 2007 to 2010 model 1 model 2 model 3 model 4 intercept 1.874 1.862 1.902 1.946 individual level yes, as a direct consequence of the crisis -0.120** -0.125** -0.120** -0.120** yes but not as a direct consequence of the crisis -0.541*** -0.546*** -0.542*** -0.539*** finished education before 20 or no full-time education -0.185*** -0.189*** -0.185*** -0.232*** low or middle qualified -0.057 -0.058 -0.056 -0.056 not working -0.086 -0.087 -0.085 -0.086 level in society -0.016 -0.016 -0.015 -0.016 one adult household without children 0.061 0.058 0.061 0.060 one adult household with children 0.148* 0.154** 0.168** 0.145* two+ adult household with children 0.012 0.012 -0.004 0.011 female 0.754*** 0.814*** 0.754*** 0.756*** 25 39 years old -0.081 -0.085 -0.080 -0.081 40 – 54 years old -0.077 -0.079 -0.077 -0.079 55 – 65 years old -0.282*** -0.285*** -0.283*** -0.282*** country level ln gdp 2011 0.484** 0.483** 0.482** 0.481** change in gdp per capita 2007-2010 -1.862 -3.020** -1.611 -0.955 interactions female*gdpchange 2.056*** one adult household without children*gdpchange -0.075 one adult household with children*gdpchange 0.455 two+ adult household with children*gdpchange -0.615 finished education before 20 or no full-time education*gdpchange -1.434*** model fit aic 75199 75175 75,202 75,189 log likelihood -37582 -37568 -37580 -37576 icc 0.075 0.075 0.075 0.075 level 1 (16799 individuals) – r2 0.0583 0.0597 0.0584 0.0590 level 2 ( 27 countries) – r2 0.1292 0.1306 0.1295 0.1299 note: ***p-value<0.01, ** p-value<0.05, * p-value<0.1 (2-tailed tests). data source: eurobarometer 2011. sample of eu27 countries. in models 2, 3 and 4 i estimate the interaction effects between the impact of crisis on country level (gdp change) and micro variables. i also calculated the marginal effects with standard errors of interaction effects for more precise interpretation (figures 2 and 3). from model 2 it can be seen that women in the countries where gdp per capita dropped more there is less discrepancy in women’s and men’s perceptions of gender equality. in countries where gdp change was less, there is more discrepancy between men and women. at the same time in all the countries, women have more favorable perceptions of gender equality than men (figure 2). the same result was found for the drop of gdp between 2008 and 2009. model 3 shows that there is no interaction effect between family structure and the drop of gdp. however, the expected interaction was found for the drop of gdp 2008 and 2009 but it 70 soboleva / european journal of government and economics 6(1), 59-77 cannot be considered robust. furthermore, i found a quite unexpected interaction effect between having finished education before 20 or having no full-time education and unemployment rate growth (model 4, figure 3). in countries where gdp dropped more the discrepancy between perceptions of gender equality of the lesser-educated and middle and higher educated groups is not significant. at the same time, in countries where gdp dropped less, the lowlier educated have less positive perceptions of gender equality. probably it could be explained by the fact that in countries where gdp per capita grew more the qualification in the labor market played a more important role than in countries that experienced higher unemployment rate growth. however, this interaction effect was not significant in the model with gdp change from 2008 to 2009. figure 2. effect of gender upon perceptions of gender equality in countries with different gdp change figure 3. effect of low education upon ga in countries with different gdp change. 71 soboleva / european journal of government and economics 6(1), 59-77 to sum up, only interaction effect between gender and drop of gdp proved to be robust. the effects of family structure and education proved to be not stable. next, i ran models with unemployment rates in 2011 and their changes between 2007 and 2010 as the main country variables in order to confirm the impact of crisis at country level. unemployment rates in 2011 itself as well as changes in unemployment rate between 2007 and 2010 do not influence gender attitudes (model 5). table 2. multilevel regression models: dependent variable – index of gender attitudes (unstandardized regression coefficients). macro level – unemployment rate change 2007-2010 model 5 model 6 model 7 model 8 intercept 6.762*** 6.705*** 6.771*** 6.802*** individual level yes, as a direct consequence of the crisis -0.121** -0.122** -0.120** -0.123** yes but not as a direct consequence of the crisis -0.541*** -0.544*** -0.544*** -0.540*** finished education before 20 or no full-time education -0.187*** -0.194*** -0.187*** -0.248*** low or middle qualified -0.057 -0.057 -0.058 -0.057 not working -0.086 -0.088 -0.089 -0.087 level in society -0.015 -0.015 -0.015 -0.015 one adult household without children 0.062 0.060 0.073 0.061 one adult household with children 0.149** 0.156** 0.288*** 0.149** two+ adult household with children 0.012 0.014 -0.029 0.012 female 0.754*** 0.865*** 0.754*** 0.756*** 25 39 years old -0.080 -0.081 -0.082 -0.082 40 – 54 years old -0.076 -0.073 -0.074 -0.077 55 – 65 years old -0.280*** -0.274*** -0.280*** -0.279*** country level unemployment rate 0.019 0.019 0.019 0.019 ur change 2007-2010 -0.089 0.002 -0.093 -0.134 interactions female* urchange -0.163*** one adult household without children* urchange -0.014 one adult household with children * urchange -0.189** two+ adult household with children * urchange 0.060 finished education before 20 or no full-time education*urchange 0.084** model fit aic 75207 75193 75203 75205 log likelihood -37585 -37577 -37580 -37583 icc 0.099 0.098 0.098 0.098 level 1 (16799 individuals) – r2 0.0332 0.0340 0.0337 0.0334 level 2 (27 countries) – r2 0.1284 0.1292 0.1285 0.1280 note: ***p-value<0.01, ** p-value<0.05, * p-value<0.1 (2-tailed tests). source: eurobarometer 2011. sample of eu27 countries. cross-level interactions with unemployment rates are similar as with the gdp per capita. marginal effects with standard errors are plotted in figures 4 to 6. (figures 4-6). in countries where unemployment rates increased because of crisis women’s perceptions of gender equality differ from men’s to a lesser extent. at the same time in all countries, women evaluate gender equality more positive than men do (model 6, figure 4). hence, it is likely that women suffered from economic crisis more and even if they had to find job or become employed they did not consider it appropriate behavior. the crisis could have stressed the female traditional social role. it is worth emphasizing that change in the unemployment rate is not always less in more economically developed countries. the same tendency is true for those who live in one-parent households with children (model 72 soboleva / european journal of government and economics 6(1), 59-77 7, figure 5). in countries where unemployment rate raised more those who live in one-parent households with children have the same perceptions of gender equality compared to those who live in 2-parent households without children. however, in the countries where unemployment rates did not change, the perceptions of gender equality of those who live in one-adult household with children are more positive compared to the perceptions of those who live in 2parent households. although this group is likely to possess more positive perceptions of gender equality for the countries where unemployment rate dropped more as a result of crisis this is not true. it is possible that in these countries such a situation is not regarded as normal and hence these individuals emphasize more traditional gender roles. when a single breadwinner either loses her or his job or begins to earn less, his or her perceptions of gender equality could be shaken because he or she is unable to cope with the situation alone. finally, again i found a quite unexpected interaction effect between having finished education before the age of 20 or having no full-time education, and unemployment rate growth (model 8, figure 6). in countries where unemployment rates grew more, the discrepancy between perceptions of gender equality of the lesser-educated and middle and higher educated groups is not significant. at the same time in countries where unemployment rates increased to a lesser extent, the lesser-educated have less positive perceptions of gender equality. figure 4. effect of gender upon perceptions of gender equality in countries with different ur change 20072010. figure 6. effect of low education upon perceptions of gender equality in countries with different ur change 2007-2010. 73 soboleva / european journal of government and economics 6(1), 59-77 figure 5. effect of gender upon perceptions of gender equality in countries with different ur change 20072010. all my results were confirmed when i took a shorter time span for change in unemployment rates (from 2008 to 2009). in addition, i checked the models where i controlled for the unemployment rate in 1998 but its effect is insignificant and it does not have impact upon other effects. hence, change in gdp per capita and change in unemployment rates between 2008 and 2009 give almost the same results that show the consistency of measurement. 6. conclusions and discussion in the current study, i examined the perceptions of gender equality of vulnerable groups that were more and less affected by crisis in 27 european countries using eurobarometer data from 2011. i used multilevel regression analysis focusing mainly on cross-level interactions. my hypothesis about the impact of economic crisis on perceptions of gender equality was partly confirmed. on country level change in gdp per capita and in unemployment rate does not influence perceptions of gender equality in general. on the individual level, however, crisis has a strong impact upon perceptions of gender equality. those who or whose partner has lost the job emphasize more negatively perceived gender equality. this tendency is independent from gender, education, having low or high-qualified job or the general impact of crisis in the country. this result means that individuals usually evaluate the situation from the perspective of their family and micro-environment than on the country level. perceptions of gender equality of vulnerable groups vary a lot depending upon the depth of impact of economic crisis in the country. my hypotheses about the impact of economic crisis on women was confirmed. this finding is in line with the “doing gender” conception of west and zimmerman (1987). in the countries where gdp per capita dropped more or unemployment rate increased more after the economic crisis the perceptions of gender equality are worse. although many women entered the labor market during the crisis, their employment is often involuntary 74 soboleva / european journal of government and economics 6(1), 59-77 and this emphasizes their female identity. the hypothesis about the impact of economic crisis on one-parent household with children was partly confirmed. perceptions of gender equality of those who live in one-parent households with children in general are more egalitarian. in countries where unemployment rates increased more after the crisis there is no difference in perceptions of gender equality of different types of households whereas in countries less affected by crisis those who live in one-parent households with children perceive better gender equality. this interaction turned out to be significant for change in gdp per capita from 2008 to 2009 but not from 2007 to 2010. the explanation could be as follows. when the single parent begins to earn less or loses a job it becomes more difficult to cope with the family situation. this change is likely to make the perceptions of gender equality more negative. my hypothesis about the effect of crisis on the lesser-educated was not confirmed. in countries where unemployment rates grew more there is no significant discrepancy in perceptions of gender equality between the lesser-educated and middle and higher educated groups whereas in countries where unemployment rates increased to a lesser extent perceptions of gender equality of the lesser-educated are more negative compared to other educational groups. the same effect was found when i considered gdp per capita change from 2007 to 2010 (but not 2008 to 2009). probably the explanation lies in the fact that in countries that experienced high unemployment rate growth, the level of education and qualification diminish in importance. to sum up, i analyzed the cross-level interactions between individual characteristics (gender, education, family structure) and the changes in economic context. values and attitudes are closely connected with economic development (inglehart and welzel, 2005). changes in economic condition in the previous years are related to perceptions of gender equality. when individuals negatively assess gender equality in the country, it also partly indicates their willingness and capacity to participate in labor market activities. this increasing vulnerability, in its turn, can to some extent hinder economic growth. the findings provide theoretical and empirical challenges for further study of the mechanisms of perceptions of gender equality in post-crisis europe. undoubtedly, the position of women and especially those who live in single-adult households is more vulnerable in countries that suffered more from the crisis. this should be taken into account in elaboration of policy regarding gender equality in the labor market. acknowledgements the study has been funded by the russian academic excellence project '5-100'. 75 soboleva / european journal of government and economics 6(1), 59-77 references albright, j. j., & marinova, d. m. (2010). estimating multilevel models using spss, stata, sas, and r. retrieved from http://www.iub.edu/~statmath/stat/all/hlm/hlm.pdf. alesina, a., giuliano, p., & nunn, n. (2013). the origins of gender roles: women and the plough. the quarterly journal of economics 128 (2), 469-529. https://doi.org/10.1093/qje/qjt005 alwin, d. f., braun, m., & scott, j. (1992). the separation of work and the family: attitudes towards women’s labour-force participation in germany, great britain, and the united states. european sociological review 8(1), 1337. https://doi.org/10.1093/oxfordjournals.esr.a036620 bergh, j. (2006). gender attitudes and modernization processes. international journal of public opinion research 19(1), 5-23. https://doi.org/10.1093/ijpor/edl004 bettio, f., & verashchagina, a. (2014). women and men in the ‘great european recession’, in m. karamessini & j. rubery (eds.), women and austerity. the economic crisis and the future for gender equality (pp. 57-81). london, uk: routledge. bolzendahl, c., & myers, d. j. (2004). feminist attitudes and support for gender equality: opinion change in women and men, 1974-1998. social forces 83 (2), 759790. https://doi.org/10.1353/sof.2005.0005 braun, m., & and gloekner-rist, a. (2011). perceived consequences of female labor-force participation. a multilevel latent-class analysis across 22 countries. obets. revista de ciencias sociales 6(2), 163-184. https://doi.org/10.14198/obets2011.6.2.01 campa, p., casarico, a., & profeta, p. (2009). gender culture and gender gap in employment. econpubblica centre for research on the public sector working paper no. 143. milano: università commerciale luigi bocconi. cunningham, m. (2008). family model: influences of women's employment and education over the life course. social forces 87(1), 299-323. https://doi.org/10.1353/sof.0.0097 eagly, a. h., & karau, s. j. (2002). role congruity theory of prejudice toward female leaders. psychological review 109, 573 – 598. https://doi.org/10.1037/0033-295x.109.3.573 elson, d. (2010). gender and the global economic crisis in developing countries: a framework for analysis. gender & development 18(2), 201212. https://doi.org/10.1080/13552074.2010.491321 european commission, and european parliament (2013). eurobarometer 76.1: financial and economic crisis, financial services, corruption, development aid, and gender equality, september 2011. icpsr34552-v2. cologne, germany: gesis/ann arbor, mi: interuniversity consortium for political and social research [distributors]. https://doi.org/10.3886/icpsr34552.v2 fortin, n. m. (2005). gender role attitudes and the labour-market outcomes of women across oecd countries. oxford review of economic policy 21(3), 416438. https://doi.org/10.1093/oxrep/gri024 http://www.iub.edu/%7estatmath/stat/all/hlm/hlm.pdf https://doi.org/10.1093/qje/qjt005 https://doi.org/10.1093/oxfordjournals.esr.a036620 https://doi.org/10.1093/ijpor/edl004 https://doi.org/10.1353/sof.2005.0005 https://doi.org/10.14198/obets2011.6.2.01 https://doi.org/10.1353/sof.0.0097 https://doi.org/10.1037/0033-295x.109.3.573 https://doi.org/10.1080/13552074.2010.491321 https://doi.org/10.3886/icpsr34552.v2 https://doi.org/10.1093/oxrep/gri024 76 soboleva / european journal of government and economics 6(1), 59-77 guiso, l., sapienza, p., & zingales, l. (2003). people’s opium. the effect of religion on economic attitudes. journal of monetary economics 50, 225– 282. https://doi.org/10.1016/s0304-3932(02)00202-7 guveli, a., need, a., & and de graaf, n. d. (2007). socio-political, cultural and economic preferences and behaviour of the social and cultural specialists and the technocrats. social class or education? social indicators research 81, 597–631. https://doi.org/10.1007/s11205006-9000-7 hox, j. j. (2010). multilevel analysis. techniques and applications. mahwah, nj: lawrence erlbaum associates. inglehart, r. f., foa, r., ponarin, e., & welzel, c. p. (2013). understanding the russian malaise: the collapse and recovery of subjective well-being in post-communist russia. wp brp working paper no. 32, higher school of economics, series: sociology. inglehart, r. f., & welzel, c. p. (2005). modernization, cultural change and democracy, cambridge: cambridge university press. https://doi.org/10.1017/cbo9780511790881 karamessini, m., & rubery, j., (2014). economic crisis and austerity: challenges to gender equality. in m. karamessini, & j. rubery (eds.), women and austerity. the economic crisis and the future for gender equality (pp. 314-351). london and new york: routledge. kroska, a. (1997). the division of labor in the home: a review and conceptualization. social psychology quarterly 60, 304 – 322. https://doi.org/10.2307/2787092 oecd (2012). closing the gender gap: act now. oecd publishing. https://doi.org/10.1787/9789264179370-en pampel, f. (2011). cohort changes in the socio-demographic determinants of gender egalitarianism. social forces 89, 961-982. https://doi.org/10.1353/sof.2011.0011 rubery, j. (2014). from ‘women and recession’ to ‘women and austerity’: a framework of analysis. in m. karamessini, & j. rubery (eds.), women and austerity. the economic crisis and the future for gender equality (pp. 17-36). london and new york: routledge. seguino, s. (2009). the global economic crisis, its gender implications and policy responses. paper prepared for gender perspectives on the financial crisis panel, at the fifty-third session of the commission on the status of women, united nations. retrieved from https://vip2.uvm.edu/~sseguino/pdf/global_crisis.pdf snijders, t. a. b., & bosker, r. j. (1999). multilevel analysis. an introduction to basic and advanced multilevel modeling. thousand oaks, ca: sage. stickney, l. t., & konrad, a. k. (2007). gender-role attitudes and earnings: a multinational study of married women and men. sex roles 57, 801811. https://doi.org/10.1007/s11199007-9311-4 the world bank (2016). world bank open data. available at http://data.worldbank.org/indicator voicu, m., & tufis, p.a. (2012). trends in gender beliefs in romania: 1993–2008. current sociology 60(1), 61-80. https://doi.org/10.1177/0011392111426648 voicu, m., & constantin, a. (2013). unemployment and attitudes towards gender equality. presentation at the 3rd international workshop of the lcsr, saint-petersburg, russia. https://doi.org/10.1016/s0304-3932(02)00202-7 https://doi.org/10.1007/s11205-006-9000-7 https://doi.org/10.1007/s11205-006-9000-7 https://doi.org/10.1017/cbo9780511790881 https://doi.org/10.2307/2787092 https://doi.org/10.1787/9789264179370-en https://doi.org/10.1353/sof.2011.0011 https://vip2.uvm.edu/%7esseguino/pdf/global_crisis.pdf https://doi.org/10.1007/s11199-007-9311-4 https://doi.org/10.1007/s11199-007-9311-4 http://data.worldbank.org/indicator https://doi.org/10.1177/0011392111426648 77 soboleva / european journal of government and economics 6(1), 59-77 voicu, m., & constantin, a. (2016). couple's status on labour market and their attitudes towards gender roles. revista de cercetare si interventie sociala 54, 7-22. walby, s. (2015). gender and the crisis. contested transformations of the gender regime. presentation at the 3rd equal is not enough conference, antwerp, belgium. welzel, c. p. (2013). freedom rising human empowerment and the quest for emancipation, cambridge university press. https://doi.org/10.1017/cbo9781139540919 west, c., & zimmerman, d. h. (1987). doing gender. gender and society 1(2), 125151. https://doi.org/10.1177/0891243287001002002 https://doi.org/10.1017/cbo9781139540919 https://doi.org/10.1177/0891243287001002002 abstract. the paper aims to compare the perceptions of gender equality of individuals more or less affected by economic crisis in europe. crisis touched the economy of most european countries but to a different extent. special focus is given to the pe... keywords. gender equality; economic crisis; cross-cultural comparison; europe; vulnerable groups jel classification. j16; j21; j24: j81; c02 undoubtedly, the position of women and especially those who live in single-adult households is more vulnerable in countries that suffered more from the crisis. this should be taken into account in elaboration of policy regarding gender equality in the... references ©the author 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 189-205 https://doi.org/10.17979/ejge.2024.13.2.9940 submitted: sep 26, 2023 accepted: apr 30, 2024 published: dec 3, 2024 article exploring determinants of vat gaps using structural equation models: a mimic approach iva hasíková 1,* 1 mendel university, czechia *correspondence: xhasikov@mendelu.cz abstract. this study explores the estimation of the vat gap using the multiple indicators and multiple causes (mimic) model, a specific type of structural equation model. the vat gap, a significant indicator of tax revenue loss and inefficiency in vat collection, is treated as a latent variable driven by various causes and reflected through specific indicators. drawing on data from european countries, the model incorporates variables such as economic openness, government spending, corruption perceptions, and the e-government development index, each serving as proxies for underlying vat collection challenges. this study presents a first-time application of the mimic model to the czech republic, estimating the vat gap of 2003 – 2020 and revealing fluctuations between 27 % and 35 % of total tax liability, with stabilization at 31 % since 2016. the findings underscore policy areas for potential improvements in vat compliance and collection efficiency, particularly through enhanced digitalization efforts and governance quality in tax administration by highlighting key factors contributing to the vat gap. keywords: vat gap; determinants of vat gap; efficiency of vat collection; mimic model; structural equation model jel classification: c39 ; c51 ; h26 1. introduction value-added tax (vat) is one of the most significant sources of tax revenue in most european countries. according to eurostat (2024), vat revenues represent around 20 % of total tax revenues, including social contributions. however, vat is frequently associated with the risk of tax evasion. according to estimates by the european union (eu) in 2020, the total vat evasion loss amounted to 93 billion euros, equating to approximately 3,000 euros lost every second throughout the year (baert, 2023). tax evasion poses significant threats to the economic environment, tax system, and society. novysedlák and palkovičová (2012) argue that tax evasion and avoidance undermine the economic environment, as certain entities gain advantages that may encourage others to adopt similar practices in the long term, even if tax evasion was not their initial intent. furthermore, tax evasion is a significant source of inequality, regardless of a country’s redistribution objectives (argentiero et al., 2021). https://creativecommons.org/licenses/by-nc/4.0/ 190 iva hasíková the tax gap provides valuable insight into tax evasion and the efficiency of tax collection. in the context of vat, the vat gap has emerged as an indicator of vat fraud (moravec et al., 2021). the vat gap is defined as the difference between the amount of vat that should be collected under the legislation and the vat that is actually collected (carfora et al., 2020). it encompasses revenues lost due to tax fraud, bankruptcies, taxpayer insolvency, or tax liability miscalculations (gajewski & jonski, 2022). quantifying the vat gap involves econometric methods that allow for the consideration of various factors, such as the quality of public sector institutions or the level of digitalisation of public administration. one such econometric method is the multiple indicators and multiple causes model (mimic). the mimic model, a variant of structural equation models, treats the vat gap as a latent unmeasurable variable that manifests through multiple indicators and is the result of numerous causes (frey & weck-hannemann, 1984; schneider et al., 2010). this paper, while developing the mimic model, considers variables such as the unemployment rate, the openness of the economy, the index of economic freedom, the corruption perception index (cpi), general government expenditure, final consumption expenditure, and the egovernment development index (egdi) as causes of the vat gap. the vat gap is evidenced by the growth of gross domestic product (gdp) per capita and vat revenue relative to gdp. the contribution of this paper is to construct a mimic model with data from selected european countries; this model will be applied to these countries. specifically, the mimic model will be used to estimate the vat gap in the czech republic. this method has not yet been used in the czech republic, providing a new perspective on the vat gap, not only its extent but also its potential sources. the czech republic, like other european countries, is not exempt from the challenges of tax evasion or the need for its elimination. the issue lies not only in the previously described disruptive effects of tax evasion but also in the loss of public funds, which are critically needed given the debt burden of czech public finances. in an international comparison, the czech republic’s debt remains relatively low. however, its debt dynamics have been above average among eu countries in recent years (the czech fiscal council, 2022). 2. literature review the vat gap can be quantified in monetary units and as a proportion of the tax collected relative to the total theoretical vat liability (vttl). ideally, when tax compliance is maximised, and there are no tax credits or deductions, the vat revenue as a proportion of the vttl should be one (cnossen, 2022). to estimate the vat gap and the tax gap in general, the literature identifies three main approaches: the bottom-up approach, the top-down approach, and methods based on econometric modelling (alm, 2012; kasnauskienė & krimisieraitė, 2015; poniatowski et al., 2020). the bottomup approach relies on tax audits and direct interviews with taxpayers (alm, 2012). according to novysedlák and palkovičová (2012), the bottom-up approach includes methods such as the selection of a random sample of taxpayers and targeted controls based on risk analysis. gajewski and jonski exploring determinants of vat gaps using structural equation models (2022) caution that tax authorities using the latter method target taxpayers with the highest expected value of unreported tax liabilities, which could lead to an overestimation of the gap. the top-down approach utilises macroeconomic and international accounts data to cover the entire national economy (poniatowski et al., 2020). the data sources typically include statistical offices’ reports on the production of gdp or supply and use tables that provide information on the output of individual industries and the consumption by these industries and sectors of the national economy. one of the methods based on the top-down approach involves using supply-and-use tables (sut) and input–output tables (iot). these tables are designed so that all production and imports match their uses (intermediate consumption, final consumption, gross capital formation, and exports). valderas-jaramillo et al. (2019) highlight concerns about the delay between the reference year of sut and iot and their publication year; official sut and iot are often published too late to be useful for policy-oriented research. nonetheless, the tables are detailed enough to allow for the direct assignment of a legislated rate to each item. the center for social and economic research (case), a consulting firm, used an effective vat rate in its estimates of the vat gap for the eu (poniatowski et al., 2020). in contrast to the other two approaches, the approach based on econometric modelling not only provides information about the size of tax evasion but also identifies the factors and determinants that influence its magnitude. however, the econometric model can only determine the development (year-on-year changes) of tax evasion over time; its value in the base period must be established by another method (kasnauskienė & krimisieraitė, 2015; schneider, 2005). the choice of the base variable is crucial as it sets the level. the mimic model is another method derived from econometric modelling. this model was pioneered by frey and weck-hanneman (1984), who used it to estimate the extent of the shadow economy in member states of the organisation for economic cooperation and development. schneider et al. (2010) and tedds (2005) also applied it to determine the proportion of the shadow economy in the gdp. frey and weck-hanneman (1984) adopted this method in response to the fact that all approaches used until that time assessed the extent of the shadow economy based on a single indicator, depending on the method employed (the currency in circulation – the demand for currency approach). furthermore, they scarcely considered any other causes of the shadow economy. the mimic model was also utilised to investigate the determinants of the vat gap in lithuania (kasnauskienė & krimisieraitė, 2015). using the mimic model, it is possible to incorporate variables such as egdi or cpi in the calculation of the vat gap; however, these variables have not yet been considered in relation to vat gap calculations in the czech republic. 192 iva hasíková 3. methodology 3.1 variable selection this section introduces the variables for the mimic model. two types of variables are required for this model: causes and indicators. • causes: unemployment rate, the openness of the economy, index of economic freedom, cpi, general government expenditure, final consumption expenditure, egdi. • indicators: growth of gdp per capita, vat revenue on gdp. 3.1.1 unemployment rate the case included the unemployment rate among the potential determinants of the vat gap as an index of taxpayers’ liquidity difficulties (poniatowski et al., 2018). the unemployment rate also reflects income inequality or poverty (reckon, 2009). in this research, the unemployment rate will be considered an index of the economic cycle. 3.1.2 openness of economy the openness of an economy is expressed as the proportion of the sum of imports and exports to a country’s gdp. this variable was examined in relation to the vat gap by aizenman and jinjarak (2008) and, more recently, by carfora et al. (2020). according to these authors, an economy’s openness positively influences the efficiency of vat collection (i.e. it reduces the vat gap). additionally, research by pluskota (2022) suggests that the share of foreign trade (exports and imports) in gdp is significant across the eu. the openness of the economy also presents an opportunity for missing trade, intracommunity fraud, and carousel fraud (frunza, 2019). in this context, the openness of the economy has a negative impact on the efficiency of vat collection. 3.1.3 index of economic freedom the efficiency of the tax system is strongly influenced by the quality of the government, which primarily involves the formulation and implementation of various regulations and the degree of independence of tax administration from political pressure (godin & hindriks, 2015). according to chan and ramly (2018), the redistributive effect of the vat system also depends on the quality of the government structure; with a low-quality structure, the vat system can be highly regressive and exacerbate income inequality. in czechia, slovakia, hungary, and poland, substantial evidence suggests that taxpayers avoid paying taxes if they believe that the public services provided do not justify the taxes collected (hanousek & palda, 2004). the index of economic freedom encompasses 12 principles for sustained progress and prosperity, many of which relate to the quality of government as described above. these principles are organised into four broad categories: exploring determinants of vat gaps using structural equation models • the rule of law (property rights, government integrity, judicial effectiveness); • government size (government spending, tax burden, fiscal health); • regulatory efficiency (business freedom, labour freedom, monetary freedom); • open markets (trade freedom, investment freedom, financial freedom); (the heritage foundation, 2023). 3.1.4 corruption perception index the cpi is another variable incorporated into the model. it measures the perceived level of corruption in each country’s public sector, according to experts and businesspeople. a higher cpi indicates lower perceived corruption. alongside the index of economic freedom, the cpi serves as another indicator of the quality of public institutions. reckon (2009) included cpi in its econometric analysis of the vat gap, finding it to be the variable most strongly associated with the size of the vat gap; lower perceived corruption corresponds to a smaller vat gap. the center for social and economic research also included cpi in its regression analysis of the determinants of vat gaps, but their results showed a positive, albeit insignificant, relationship between the vat gap and cpi; this indicates that an improvement in the perception of corruption within a country is associated with a higher vat gap (barbone et al., 2013). these consulting firms, case and reckon, have reported contrasting results regarding the relationship between cpi and the vat gap. 3.1.5 general government expenditure for the model, government expenditure is expressed as a share of the gdp in each country. reckon (2009) included government expenditure in his research into the causes of the vat gap as it reflects the total tax burden and the size of the public sector, encompassing tax audits and other regulatory types. similarly, zídková and pavel (2016) factored in government spending as a share of gdp in their study on the causes of the vat gap, arguing that a larger public sector size leads to a reduction in the vat gap. 3.1.6 final consumption expenditure final consumption expenditure refers to the portion of expenditure primarily spent by households on goods and services intended to directly satisfy individual needs (eurostat, 2016). according to zídková and pavel (2016), final consumption poses more challenges for vat collection compared to intermediate consumption by manufacturing businesses, which can claim vat deductions on their purchases. these purchases can be made in cash and, as such, may evade the scrutiny of tax authorities. a study by immordino and russo (2018) demonstrates that cashless payments negatively impact the vat gap. 3.1.7 e-government development index the egdi is an index produced by the united nations for its member states to reflect the utilisation of information technology. the index comprises three sub-indices: provision of online services (online service index), human participation (human capital index), and telecommunication 194 iva hasíková connectivity (telecommunication infrastructure index) (united nations, 2023). utilising this index aids in considering the advance of digitalisation, which aims to eliminate tax evasion. case incorporated information technology expenditures related to gdp to gauge the effect of implementing innovative processes into tax administration; their research indicates a statistically significant negative impact of information technology expenditures on the vat gap (poniatowski et al., 2020). digitalisation will enable governments to access and analyse the necessary information, thus increasing tax collection efficiency (alm, 2021). however, alm (2021) also cautions that technological changes will make tax evasion increasingly difficult for entities whose transactions leave an electronic trail or are subject to third-party information reporting. these entities comprise most taxpayers in developing and developed countries. conversely, digitalisation appears to facilitate tax evasion for multinational corporations through profit-shifting, high-income individuals via tax havens and money laundering, and independent contractors who operate without intermediaries, clearing centres, or banks, leaving no electronic trail of their transactions. table 1. candidate causes and indicators of the vat gap. variable explanation author predicted effect on the vat gap source unemployment rate poverty, income inequality, index of the economic cycle poniatowski et al. (2018); reckon (2009) + eurostat gross capital formation investments kasnauskienė and krimisieraitė (2015) − eurostat openness of economy risk of carousel fraud, the openness of the economy zídková and pavel (2016) +/− eurostat index of economic freedom government quality, tax burden, open market godin and hindriks (2015); hanousek a palda (2004) − the heritage foundation cpi government quality, corruption perceived reckon (2009) +/− transparency international general government expenditure size of the public sector reckon (2009); zídková and pavel (2016) +/− eurostat final consumption purchases of final consumers, potential vat base zídková and pavel (2016) + eurostat egdi information technologies in government and tax offices poniatowski et al. (2020) − united nations growth of gdp per capita reflection of tax evasion between taxpayers kasnauskienė and krimisieraitė (2015) + eurostat vat revenue on gdp level of vat revenue in each country kasnauskienė and krimisieraitė (2015) − eurostat exploring determinants of vat gaps using structural equation models 3.1.8 growth of gdp per capita gross domestic product per capita will be used as the primary indicator to reveal vat collection inefficiencies, particularly vat evasion. schneider et al. (2013) utilised this variable as an indicator when examining the shadow economy, arguing that the informal economy must necessarily be reflected in the formal economy captured by statistical offices. kasnauskienė and krimisieraitė (2015) examined the determinants of the vat gap using the mimic model, employing real gdp per capita as one of the indicators to negate the effect of inflation. 3.1.9 vat revenue on gdp tax noncompliance is expected to manifest as a decrease in vat revenue; thus, vat revenue on gdp was selected as a second indicator (kasnauskienė & krimisieraitė, 2015). table 1 summarises each cause and indicator, providing a brief description of its inclusion in the model, the authors who have worked with these variables, the predicted effect on the vat gap, and the source of data. 3.2 methodology – the mimic model the mimic model is founded on the statistical theory of a latent (unobserved) variable, which is ascertained using multiple measurable causes and indicators. multiple causes lead to the existence of a latent variable, while simultaneously, multiple indicators of its presence can be observed (schneider et al., 2010). . figure 1. mimic model – general structure. source: own elaboration from schneider et al. (2010). figure 1 illustrates the general structure of the mimic model. this model is a specific type of structural equations model comprising two components: a structural model and a measurement model (schneider et al., 2010). to estimate the variance of a latent variable, the mimic model utilises unstandardised estimates, meaning the first indicator is permanently fixed at level 1 and is termed the reference indicator. all other estimates vary by a specified coefficient if the reference indicator changes by 1 (acock, 2013). 196 iva hasíková additionally, the two models encompass several measurable (observed) variables and a latent variable, each assuming a distinct role. in the structural model, the latent variable serves as the dependent variable, influenced by the measurable variables entering the model. the equation can be expressed as ηt = γ′xt + ςt, [1] where xt’ is a (1*q) vector of time series xit, i=1,...,q, containing potential causes of the hidden variable ηt, and γ’ is a vector of coefficients describing the relationship between the hidden variable and its causes. ςt represents the error term. in the measurement model, the latent variable is independent, whereas the measurable variables entering the model are dependent on it. yt = ληt + εt [2] where yt’ is a (1*p) vector of a time series of indicators of the hidden variable, λ is a vector of regression coefficients, and ε’ is a vector of white noise. in this scenario, the latent unobserved variable, the vat gap, is initially linked to the observed indicator variables within the measurement model. subsequently, the relationships between the latent unobserved variable and the observed explanatory variables (causes) in the structural model are examined. by employing equation 1 in equation 2, a multiple regression model is derived where the explanatory endogenous variables yjt, j = 1,...,p, are indicators of the latent variable η, and the explanatory exogenous variables xit, i = 1,...,q, are causes. the model can be represented by the following equation: yt = πxt + z, [3] where π = λγ’ is the matrix and z = λς + ε represents the error term, a (p*1) vector of linear combinations of white noise ς and ε from the structural model and the measurement model. the final mimic model will retain only those variables that are significant at a minimum 5% level of statistical significance, using the p-value as an indicator. the mimic index is computed using the structural model equation, equation 1. equation 4 outlines this calculation; it is a modified version of equation 1. 𝜂𝜂�t = γx1t + γx2t. . +γxqt, [4] where x1t to xqt represent the variables of causes at a level of at least 5%. the mimic index merely indicates the relative development of the latent variable, the vat gap in this instance. to translate the relative values into absolute terms, a baseline variable obtained by another method must be employed. exploring determinants of vat gaps using structural equation models for conversion, the following equation is used: �̂�𝜂t = 𝜂𝜂�𝑡𝑡 𝜂𝜂�𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 ∗ �̇�𝜂𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 , [5] where 𝜂𝜂�𝑡𝑡 denotes the value of the mimic index at time t according to equation 4, 𝜂𝜂�𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 is the value of the mimic index in the base period, and �̇�𝜂𝑏𝑏𝑏𝑏𝑏𝑏𝑏𝑏 is an estimate of the latent variable obtained by another method. 3.3 data the mimic model is constructed using panel data from selected european countries (a total of 26 countries) spanning the years 2002 to 2020. due to the availability of data, the maximum time series was selected to provide an adequate database for estimating the mimic model. concerning the availability of the egdi, which represents digitalisation, the time series could not commence earlier than 2002, with the objective of capturing an overview of vat gap development up to the most recent year feasible. regarding other variables, the list of selected countries is presented in appendix 1. the group also includes non-european countries, as the study does not consider eu membership. table 2 contains descriptive statistics. the total dataset comprises 26 panels and 20 time periods (years), with 520 observations collected for each variable. for the mimic model, the data must be stationary. the data were tested for the presence of a unit root using the levin–lin–chu test, designed for panel data. this test is suitable for data where the number of panels does not exceed 100 and the ratio of the number of panels to the number of time periods is close to zero (levin et al., 2002; stata, 2023). the result of the test confirmed the presence of a unit root, indicating that the data are not stationary and require differencing. table 2. descriptive statistics variable number of observations mean standard deviation minimum maximum unemployment rate 520 8.03 4.32 2.00 27.50 openness of economy 520 59.87 30.42 22.80 176.70 economic freedom index 520 69.47 6.25 48.70 82.60 cpi 520 0.65 0.16 0.26 0.95 general government expenditure 520 0.44 0.07 0.24 0.65 final consumption 520 0.53 0.09 0.23 0.70 egdi 520 0.73 0.11 0.47 0.98 growth of gdp per capita 520 1.88 3.95 −14.50 23.20 vat revenue on gdp 520 20.19 4.07 11.20 30.50 198 iva hasíková particularly noteworthy is the maximum unemployment rate of 27.5% recorded in greece in 2013. this is an exceptional case, as the standard deviation is approximately 4.62. the data for gdp per capita growth present another extreme value. the minimum for this variable is -14.5%, recorded in estonia in 2009, while the maximum represents a year-on-year increase of 23.2%, occurring in ireland in 2015. in terms of data variability, the standard deviation provides the most informative measure. if the standard deviation is around zero, the data exhibit low variability despite the presence of outliers. data on the openness of the economy show the highest standard deviation. luxembourg consistently records very high values of economic openness, often exceeding 100%. in contrast, openness is low in spain, greece, and sweden. 4. results 4.1 mimic model table 3 presents the mimic models, detailing both parts: the structural model and the measurement model. in model 1, all variables are included with the goal of retaining only those that are statistically significant in the structural model at a minimum 5% level of significance. a stepwise selection process is employed to systematically eliminate statistically insignificant variables from the structural model; this process is fully detailed in table 3. in the measurement model, both variables – growth of gdp per capita and vat revenue on gdp – are statistically significant at the 1% level, allowing them to remain in all subsequent models. the variable index of economic freedom was the first to be removed from the structural model due to its high p-value. the structural model of model 2 consists of six variables. the unemployment rate also had to be removed due to its high p-value, resulting in model 3, the final model, which consists of only four variables. information criteria are useful when determining which model is optimal. the akaike information criterion (aic) was applied in this research. the most appropriate model is identified by the lowest value of this criterion, which is the final model with the lowest aic. other indicators of model quality are also optimal in the last model. the comparative fit index (cfi) measures how close a given model is to a perfect fit with the data used. it ranges from 0 to 1; a higher value indicates a better model. the final model, with a cfi of 0.86, shows the highest value. the interpretation of coefficients in a structural model of the mimic model closely resembles that in regression analysis. their value indicates the resultant change in the vat gap for a unit change in the causal variable under the ceteris paribus condition. in the following paragraphs, the coefficients from the structural model are interpreted according to model 1, which includes all variables, significant and insignificant. final consumption emerges as the most crucial driver of the vat gap. according to model 1, if final consumption increases by 1 %, then the vat gap also increases by about 5.71 %. final consumption is significant at the 5% level; hence, it is retained in exploring determinants of vat gaps using structural equation models model 2 and the final model, with the value of its coefficient fluctuating around 6 % to 7 %. a higher unemployment rate increases the vat gap; according to model 1, if the unemployment rate increases by 1 %, the vat gap increases by 0.01 %. in model 2, the coefficient of the unemployment rate remains stable. however, this variable was removed from the final model due to an excessively high p-value. greater openness of the economy also raises the vat gap; if the openness of the economy increases by 1 %, then the vat gap increases by 0.02 %. the openness of the economy is significant at the required level, with the value of its coefficient remaining stable in model 2 and the final model. there is no unified conclusion about the effect of the openness of the economy on the vat gap; aizenman and jinjarak (2008) suggest a negative impact, while frunza (2019) argues that an open economy provides opportunities for carousel frauds and other vat frauds, which have a positive effect on the vat gap. this research supports the latter view regarding the positive impact of economic openness on the vat gap. table 3. model mimic for europe-23 (author calculations using stata) – vat revenue on gdp is used as a reference indicator. model 1 model 2 model 3 (final model) structural coefficient (p-value) coefficient (p-value) coefficient (p-value) unemployment rate 0.01 (0.31) 0.01 (0.26) openness of economy 0.02 (0.00) *** 0.02 (0.00) *** 0.02 (0.00) *** index of economic freedom −0.01 (0.36) cpi −0.27 (0.00) *** −0.32 (0.00) *** −0.14 (0.00) *** general government expenditure −4,44 (0.00) *** −5.3 (0.00) *** −4.53 (0.00) *** final consumption 5.71 (0.00) *** 7.08 (0.00) *** 6.05 (0.00) *** egdi −2.01 (0.00) *** −2.44 (0.00) *** −2.20 (0.00) *** measurement vat revenue on gdp 1 1 1 growth of gdp per capita 13.01 (0.00) *** 10.89 (0.00) *** 12.97 (0.00) *** statistics chi-square 50.9 (0.00) *** 60.30 (0.00) *** 42.83 (0.00) *** degrees of freedom 6 5 4 cfi 0.851 0.80 0.86 srmr 0.042 0.056 0.05 rmsea 0.12 0.15 0.140 aic 850.57 −838.84 −2517.93 the index of economic freedom has a negative effect on the vat gap; if the index of economic freedom increases by one percentage point, then the vat gap decreases by 0.01 %. this variable is not statistically significant at the required level. due to an excessively high p-value, it was removed from model 2 and the final model. the cpi also has a negative effect on the vat gap. if the cpi increases by one percentage 200 iva hasíková point – indicating a lower perception of corruption – the vat gap decreases by 0.27 %. the cpi is significant at the required level of significance, so it remains in model 2 and the final model. the value of its coefficient remains stable. the case (barbone et al., 2013) and reckon (2009) examined the influence of cpi on the vat gap and reached opposite conclusions. this research supports the findings of reckon (2009) regarding the negative effect of cpi on the vat gap. general government expenditure also negatively impacts the vat gap, but this variable is more substantial. if the share of general government expenditure in gdp increases by 1 %, then the vat gap decreases by 4.44 %. this variable is significant at the required level of significance, so it remains in model 2 and the final model, with the value of its coefficients fluctuating between 4 and 5.5 %. similar to the case of economic openness, there is no unified conclusion about the effect of general government expenditure on the vat gap; however, this research supports the findings of zídková and pavel (2016) regarding the negative impact of general government expenditure on the vat gap. the egdi also decreases the vat gap, although it is not as strong a variable. if the egdi increases by one percentage point, then the vat gap decreases by 2.01 %. egdi is significant at the required level, so it remains in model 2 and the final model. the value of its coefficient remains stable. table 4. robustness test: mimic model for eu-17. model mimic for the eu-17 structural coefficient (p-value) unemployment rate 0.01 (0.12) openness of economy 0.01 (0.00) *** index of economic freedom −0.01 (0.38) cpi 0.75 (0.00) *** general government expenditure −1.71 (0.00) *** final consumption 6.05 (0.00) *** egdi −0.86 (0.00) *** measurement vat revenue on gdp 1 growth of gdp per capita 15.71 (0.00) *** statistics chi-square 20.65 (0.00) *** degrees of freedom 6 cfi 0.85 srmr 0.04 rmsea 0.09 aic −2517.93 exploring determinants of vat gaps using structural equation models 4.2 test for the robustness of coefficients as outlined in the methodology section, the dataset exhibits high variability, with the openness of the economy introducing the most variation. it is thus prudent to test the robustness of the coefficients of the mimic model. to this end, another mimic model was estimated, including only developed eu countries, while nine transit countries were excluded from the dataset. table 4 presents the results of the mimic model for europe-17. despite the dataset’s limitations, the coefficients of the variables remain consistent, with the same signs and levels of significance. 4.3 application of the mimic model in the czech republic in this subsection, the mimic model is applied to the czech republic. in equation 4, each coefficient of a significant variable is multiplied by the corresponding variable (for example, the coefficient of the openness of the economy is multiplied by the level of openness of the economy in the czech republic), producing the mimic index. the mimic index expresses the relative size of the vat gap. to convert this into absolute values, it is necessary to use a base variable obtained by another method. for this research, an estimate by case (barbone et al., 2013) is used as the baseline variable, setting the vat gap in the czech republic in the baseline year of 2002 at 29% of vttl. values of the vat gap for the period 2003–2020 are determined according to the mimic index, influenced by significant causes of the vat gap. figure 2. the size of the vat gap in the czech republic using the mimic model. source: author’s calculations figure 2 displays estimates of the vat gap in the czech republic from 2003 to 2020, expressed as a percentage of vttl. it peaked in 2005 at almost 35%. the second highest point was in 2015, at 33 %. conversely, the lowest estimated values were around 27 % and occurred in 2003 and 2013, when the vat gap fell to a minimum of 27 %. the vat gap level has been rising steeply since 2013, reaching one of its highest values in 2015 and then decreasing again in 2016. since 2017, the vat gap has stabilised at around 31 % of vttl. 20% 22% 24% 26% 28% 30% 32% 34% 36% 38% 40% 2 0 0 3 2 0 0 4 2 0 0 5 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 2 0 1 0 2 0 1 1 2 0 1 2 2 0 1 3 2 0 1 4 2 0 1 5 2 0 1 6 2 0 1 7 2 0 1 8 2 0 1 9 2 0 2 0 202 iva hasíková 5. conclusions the paper provides a different insight into the vat gap through econometric modelling. the mimic model does not process statistical data from national accounts; instead, it evaluates the relationship between inputted variables and the vat gap. the contribution of the mimic model lies in the evaluation and identification of significant variables that can proxy potential sources of vat collection inefficiencies. such information allows for the development of policy-related recommendations. however, this contribution also conceals a significant limitation of the research conducted: the omission of a vital input variable. this omission could lead to biased results from the mimic model. to avoid such an omission, a literature search was conducted on studies concerning the causes or determinants of the vat gap. the search included studies by both foreign and domestic authors, as well as those under the auspices of the eu, such as case or reckon. significant causes identified include final consumption, general government expenditure, the openness of the economy, the cpi, and the egdi. in contrast, the unemployment rate and the index of economic freedom are not statistically significant variables. in the mimic model, the unemployment rate represents the economic cycle; according to this analysis, the economic cycle does not significantly influence the extent of the vat gap. the index of economic freedom and cpi were included in the model to express the quality of government, among other factors. a key difference between these two variables is that the cpi is based on the perceptions of businesspeople and experts about corruption in their home country, while the index of economic freedom is based on data from the world bank. this underscores the importance of a country not only maintaining high-quality governance but also effectively conveying this to its citizens and taxpayers. recommendations for further development of indirect tax policy should be based on the statistically significant causes of vat evasion. a key recommendation from this research is to focus on the digitalisation of tax offices, which can increase the efficiency of vat collection. digitalisation can also facilitate cooperation between tax offices at home and abroad, potentially mitigating the impact of economic openness. acknowledgements this research was supported by the internal grant agency of fbe, mendel university in brno, registration no.: iga-pef-dp-23-012. exploring determinants of vat gaps using structural equation models references acock, a. c. (2013). discovering structural equation modeling using stata (vol. 1). stata press. aizenman, j., & jinjarak, y. (2008). the collection efficiency of the value added tax: theory and international evidence. journal of international trade and economic development, 17(3), 391-410. https://doi.org/10.1080/09638190802137059 alm, j. (2012). measuring, explaining, and controlling tax evasion: lessons from theory, experiments, and field studies. international tax and public finance, 19(1), 54-77. https://doi.org/10.1007/s10797-011-9171-2 alm, j. (2021). tax evasion, technology, and inequality. economics of governance, 22(4), 321-343. https://doi.org/10.1007/s10101-021-00247-w argentiero, a., casal, s., mittone, l., & morreale, a. (2021). tax evasion and inequality: some theoretical and empirical insights. economics of governance, 22(4), 309-320. https://doi.org/10.1007/s10101-021-00261-y baert, p. (2023). vat in the digital age. european parliamentary research service. retrieved from https://www.europarl.europa.eu/regdata/etudes/brie/2023/739371/eprs_bri(2023)7393 71_en.pdf barbone, l., belkindas, m., bettendorf, l., bird, r. m., bonch-osmolovskiy, m., & smart, m. (2013). study to quantify and analyse the vat gap in the eu-27 member states. case network reports, no. 116, center for social and economic research (case). https://doi.org/10.2139/ssrn.2365189 carfora, a., dongiovanni, s., marabucci, a., & pisani, s. (2020). the impact of domestic factors and spillover effects on eu countries vat gap. argomenti di discussione, 1, 1-35. cnossen, s. (2022). the c-inefficiency of the eu-vat and what can be done about it. international tax and public finance, 29(1), 215-236. https://doi.org/10.1007/s10797-021-09683-0 eurostat. (2016). glossary: final consumption expenditure. retrieved from https://ec.europa.eu/eurostat/statisticsexplained/index.php?title=glossary:final_consumption_expenditure eurostat. (2024). main national tax aggregates. retrieved from https://ec.europa.eu/eurostat/web/products-datasets/-/gov_10a_taxag frey, b. s., & weck-hannemann, h. (1984). the hidden economy as an 'unobserved' variable. european economic review, 26(1), 33-53. https://doi.org/10.1016/0014-2921(84)90020-5 frunza, m. c. (2019). value added tax fraud. routledge. https://doi.org/10.4324/9781315098722 gajewski, d. j., & jonski, k. (2022). 'vat gap' estimation: distinguishing between informality and fraud. ec tax review, 31(3), 124-130. https://doi.org/10.54648/ecta2022012 godin, m., & hindriks, j. (2015). a review of critical issues on tax design and tax administration in a global economy and developing countries. core discussion paper, 2015(28), 2-76. hanousek, j., & palda, f. (2004). quality of government services and the civic duty to pay taxes in the czech and slovak republics, and other transition countries. kyklos, 57(2), 237-252. https://doi.org/10.1111/j.0023-5962.2004.00252.x chan, s. g., & ramly, z. (2018). the role of country governance on value-added tax and inequality. e & m ekonomie a management, 21(4), 79-93. https://doi.org/10.15240/tul/001/2018-4-006 immordino, g., & russo, f. f. (2018). cashless payments and tax evasion. european journal of political economy, 55, 36-43. https://doi.org/10.1016/j.ejpoleco.2017.11.001 kasnauskienė, g., & krimisieraitė, j. (2015). using mimic models to examine determinants of vat gap in lithuania. organisations and markets in emerging economies, 6(1), 107-126. https://doi.org/10.15388/omee.2015.6.1.14230 levin, a., lin, c. f., & chu, c. s. j. (2002). unit root tests in panel data: asymptotic and finite-sample properties. journal of econometrics, 108(1), 1-24. https://doi.org/10.1016/s03044076(01)00098-7 moravec, l., hinke, j., & borsiczká, m. (2021). vat gap estimation and influence of selected methods of financial administration on its reduction. danube, 12(1), 45-60. https://doi.org/10.2478/danb2021-0004 novysedlák, v., & palkovičová, j. (2012). the estimate of the value added tax revenue loss. institute for financial policy, the ministry of finance of the slovak republic. retrieved from https://www.mfsr.sk/en/finance/institute-financial-policy/working-papers/the-estimatevalue-added-tax-revenue-loss/ pluskota, a. (2022). vat gap determinants in the european union and poland. ekonomia i prawoeconomics and law, 21(3), 623-633. https://doi.org/10.12775/eip.2022.033 poniatowski, g., bonch-osmolovskiy, m., duran-cabré, j. m., esteller-moré, a., & śmietanka, a. (2018). study and reports on the vat gap in the eu-28 member states: 2018 final report. case reports, no. 496, center for social and economic research (case). https://doi.org/10.2139/ssrn.3272816 https://doi.org/10.1080/09638190802137059 https://doi.org/10.1007/s10797-011-9171-2 https://doi.org/10.1007/s10101-021-00247-w https://doi.org/10.1007/s10101-021-00261-y https://www.europarl.europa.eu/regdata/etudes/brie/2023/739371/eprs_bri(2023)739371_en.pdf https://www.europarl.europa.eu/regdata/etudes/brie/2023/739371/eprs_bri(2023)739371_en.pdf https://doi.org/10.2139/ssrn.2365189 https://doi.org/10.1007/s10797-021-09683-0 https://ec.europa.eu/eurostat/statistics-explained/index.php?title=glossary:final_consumption_expenditure https://ec.europa.eu/eurostat/statistics-explained/index.php?title=glossary:final_consumption_expenditure https://ec.europa.eu/eurostat/web/products-datasets/-/gov_10a_taxag https://doi.org/10.1016/0014-2921(84)90020-5 https://doi.org/10.4324/9781315098722 https://doi.org/10.54648/ecta2022012 https://doi.org/10.1111/j.0023-5962.2004.00252.x https://doi.org/10.15240/tul/001/2018-4-006 https://doi.org/10.1016/j.ejpoleco.2017.11.001 https://doi.org/10.15388/omee.2015.6.1.14230 https://doi.org/10.1016/s0304-4076(01)00098-7 https://doi.org/10.1016/s0304-4076(01)00098-7 https://doi.org/10.2478/danb-2021-0004 https://doi.org/10.2478/danb-2021-0004 https://www.mfsr.sk/en/finance/institute-financial-policy/working-papers/the-estimate-value-added-tax-revenue-loss/ https://www.mfsr.sk/en/finance/institute-financial-policy/working-papers/the-estimate-value-added-tax-revenue-loss/ https://doi.org/10.12775/eip.2022.033 https://doi.org/10.2139/ssrn.3272816 204 iva hasíková poniatowski, g., śmietanka, a., & bonch-osmolovskiy, m. (2020). study and reports on the vat gap in the eu-28 member states: 2020 final report. case reports, no. 503, center for social and economic research (case). https://doi.org/10.2139/ssrn.3744157 reckon. (2009). study to quantify and analyse the vat gap in the eu-25 member states. report for dg taxation and customs union. schneider, f. (2005). shadow economies around the world: what do we really know? european journal of political economy, 21(3), 598-642. https://doi.org/10.1016/j.ejpoleco.2004.10.002 schneider, f., buehn, a., & montenegro, c. e. (2010). shadow economies all over the world: new estimates for 162 countries from 1999 to 2007. world bank policy research working paper (5356), 3-52. https://doi.org/10.4337/9780857930880.00007 stata. (2023). panel-data unit-root tests. retrieved from https://www.stata.com/features/overview/panel-data-unit-root-tests/ tedds, l. m. (2005). revisiting the latent variable/mimic model approach to measuring the size of the canadian underground economy. in schneider, f., & bajada, c. (eds.), size, causes and consequences of the underground economy (pp. 159-176). ashgate publishing. the czech fiscal council. (2022). development of the national debt of the czech republic. retrieved from https://www.rozpoctovarada.cz/publikace/informacni-studie-vyvoj-statniho-dluhu-ceskerepubliky/ the heritage foundation. (2023). 2023 index of economic freedom: about the index. retrieved from https://www.heritage.org/index/about united nations. (2023). e-government development index (egdi). retrieved from https://publicadministration.un.org/egovkb/en-us/about/overview/-e-governmentdevelopment-index valderas-jaramillo, j. m., rueda-cantuche, j. m., olmedo, e., & beutel, j. (2019). projecting supply and use tables: new variants and fair comparisons. economic systems research, 31(3), 423-444. https://doi.org/10.1080/09535314.2018.1545221 zídková, h., & pavel, j. (2016). what causes the vat gap? ekonomicky casopis, 64(9), 811-826. https://doi.org/10.2139/ssrn.3744157 https://doi.org/10.1016/j.ejpoleco.2004.10.002 https://doi.org/10.4337/9780857930880.00007 https://www.stata.com/features/overview/panel-data-unit-root-tests/ https://www.rozpoctovarada.cz/publikace/informacni-studie-vyvoj-statniho-dluhu-ceske-republiky/ https://www.rozpoctovarada.cz/publikace/informacni-studie-vyvoj-statniho-dluhu-ceske-republiky/ https://www.heritage.org/index/about https://publicadministration.un.org/egovkb/en-us/about/overview/-e-government-development-index https://publicadministration.un.org/egovkb/en-us/about/overview/-e-government-development-index https://doi.org/10.1080/09535314.2018.1545221 exploring determinants of vat gaps using structural equation models appendix 1 selected countries: austria belgium cyprus czech republic denmark estonia france germany greece hungary ireland italy latvia lithuania luxembourg malta netherlands poland portugal romania slovenia spain slovakia sweden non-eu countries: norway switzerland 1. introduction 2. literature review 3. methodology 4. results 5. conclusions references appendix 1 social capital and banking system profitability: a survey of european union countries vol.8 • no.1 2019 issn: 2254-7088 european journal of government and economics 8(1), june 2019. european journal of government and economics issn: 2254-7088 number 8, issue 1, june 2019 doi: https://doi.org/10.17979/ejge.2018.8.1 the evaluation index improvement for the relocation of research equipment 6-29 donghun yoon residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) 30-47 josé ramón cardona, daniela álvarez bassi, maría dolores sánchez-fernández social capital and banking system profitability: a survey of european union countries 48-62 arash nayebyazdi governance and domestic investment in africa 63-80 chimere okechukwu iheonu economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt 81-95 marwa shibl biltagy give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia 96-109 jaroslav mihálik, peter horváth, martin švikruha https://doi.org/10.17979/ejge.2018.8.1 european journal of government and economics 8(1), june 2019, 48-62 european journal of government and economics issn: 2254-7088 social capital and banking system profitability: a survey of european union countries arash nayebyazdia* a mofid university, iran * corresponding author at: a.nayebyazdi@bankmellat.ir article history. received 28 april 2018; first revision required 27 october 2018; accepted 10 january 2019. abstract: over the last years, the concept of social capital as a facilitator of economic activities has been a remarkable issue among economists. in this article, we study the impact of social capital on banking performance focusing on profitability in the european union for period 2008-2016. social capital indicators are applied in the model are "trust in others" and "fair behavior of others". we expect more profitable banks in societies with higher levels of social capital. according to the type of data, we apply gmm estimator to do more efficient estimations. we use auxiliary variables such as bank asset, capital adequacy, real interest rate, the cost to income ratio as micro variables, gdp and inflation are employed as macros. our estimations point at a rejection of the main hypothesis. opportunistic behavior and less social trust result in more profits for european countries. we justify the results in two ways. first, due to the 2008 financial crisis, trust to all institutions has decreased in european countries. the second reason concerns countries with low levels of social capital. the decrease of trust for the banking system is lower than for other institutions. therefore, that sector may benefit is such circumstances. keywords. social capital, banking profitability, european union, generalized method of moments jel codes. a13, g19, g21 doi. https://doi.org/10.17979/ejge.2019.8.1.4575 1. introduction hanifan introduced the notion of social capital in 1916, but it was not until the early 1990s that economists took a serious interest in this concept. there is still no broad consensus on the definition of social capital, (cassar, crowleyand & wydick, 2007), but coleman (1988) has defined it as "all social relations and social structure facilitate some forms of social capital; actors establish relations purposefully and continue them when they continue to provide benefits"(p.105). elsewhere, social capital has been defined as “social networks/relations, and the norms and values that are generated, accumulated and spread through these networks” (westlund & adam, 2010). woolcock (1998) defines social capital as “the information, trust, and norms of reciprocity inhering in one’s social networks”. putnam (1993), who is one of the major contributors to the literature of social capital, has reported a correlation between electoral participation and the quality of government in italy. the economic implications of social capital have been extensively researched, and many studies have confirmed the existence of such implications. (glaeser, libson & sacerdote, 2002). mailto:a.nayebyazdi@bankmellat.ir https://doi.org/10.17979/ejge.2019.8.1.4575 arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 49 according to knack and keefer (1997), social capital has indeed a significant impact on the performance of the economy. in recent years, the economic impacts of social capital have been explored in greater detail. but some believe that social capital-related literature is still underdeveloped and can be compared, in terms of maturity, to the 1960s literature on human capital (grootaert & van bastelaar, 2002). others believe that the research into social capital should undergo a shift from the pre-paradigm to the paradigm phase (fulkerson & thompson, 2008). therefore, social capital has different effects such as economic, behavioral, and institutional effects however, economic consequences are the focus of this paper. some commentators have defined social capital as a network of social relations that affect individual behavior and thereby economic growth (penner & mueller, 1997). the impact of social capital in various economic areas has also been extensively researched. for example, there is a considerable body of literature regarding the role of social capital in the area of economic growth (zak & knack, 2001; lyer, kitson & toh, 2005; akomak & weel, 2009), labor productivity (uphoff & wijayaratna, 2000; hall & jones, 1997), and development of financial systems (ongena & smith, 2000; guiso, sapienza & zingales 2004). by surveying the literature of social capital, the concept can be categorized into three approaches (membiela-pollan & pena-lopez, 2017): 1. micro-social perspective that focuses on the individual as a core and his relations with others to gain benefits (bordieu,1986; lin,2001) 2. meso-social perspective or communitarian. its focus is on community and the benefits from a group rather than individuality (bowels & gintis,2002; fukuyama,1997) 3. macro-social and macroinstitutional perspective: the focal center in this perspective is the entire social and economic network as a whole. some social beliefs and behaviors are so important in this perspective such as social trust, opportunistic behavior and civic-mindedness. (putnam.1993, inglehart,1997; guiso et al.,2004) some social capital indices such as trust to other people and less financial opportunistic behavior can facilitate financial connections and decrease transactional costs and improve social performance (putnam,1993; dasgupta,2005; wolleb,2008). guiso, sapienza and zinalas (2011), consider social capital as a result of social networks and shared common beliefs that assist cooperation in a society. in this article, the macro-social aspect of social capital in terms of common norms is considered. from the above, it can be inferred that social capital has an undeniable impact on the economy. banks, as integral components of the economy, are no exception in this regard. therefore, social capital can affect debt holders’ perception about borrowers that they behave less opportunistically in a high social capital level environment. it can be seen that banks insist less on collateral requirements in the lending process when the firms as borrowers work in an environment with higher levels of social capital and by reducing the number of covenants can decrease transactional costs (hasan, hoi, wu & zhang (2017). arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 50 in this paper, we examine the impact of social capital on the performance of banks within the boundaries of the european union as a relatively homogeneous society. in the next section of this paper, we will review the existing works and theories regarding the subject. in the third section, the employed data and methodology will be described. in the fourth section, the modelling and results will be discussed. and in the fifth section, we will summarize the results and present the conclusions. the hypothesis of this research is the existence of a significant positive relationship between social capital and the profitability of banks. 2. social capital and banks according to jin, kanagaretnam, and lobo (2015), in the countries that enjoy a level degree of social capital, financial reports of banks tend to have enhanced accounting transparency, and their commitment to honor obligations and preserve mutual trust prevent them from self-serving actions such as window dressing. in contrast, in the countries where social capital is low, financial reporting is less transparent and bank failures are more common. hence, it can be concluded that social capital has a significant impact on the performance of the banking system. in a study conducted by xie (2013) on the impact of social capital on the banks’ risk-taking behavior, positive and negative effects of social capital and its relationship with the emergence of non-performing loans were examined. the positive effect of social capital is that ethical principles encourage borrowers to honor the terms of the loan contract and repay the debt. however, the negative effect is that the same premise gives banks a false sense of confidence regarding how borrowers return the money. the abuse of this trust may also undermine the same principles upon which the trust is founded. in other words, opportunists may take advantage of a trusting community to borrow money and then refuse to honor their commitment, ultimately creating a sense of distrust. in conclusion, this study reported that the higher is the social capital of a country, the more stable are its banks. talavera, xiong, and xiong (2012) examined the association of social capital with access to bank financing. after conducting a survey on chinese companies and banks, they concluded that in the countries like china where financial market is dominated by state-owned banks, maintaining a high level of social capital is essential for ensuring that companies and entrepreneurs have fair access to financial resources. they stated that membership in business associations, participation in charitable activities, and membership in the chinese communist party are notable examples of the social capital enhancing factors that may improve the private sector’s access to financial resources. one of the comprehensive studies regarding the relationship of social capital with banks is the research conducted by pastor and tortosa-ausina (2008). according to this research, social capital can affect the bank performance in five ways: (i) by reducing information, transaction, and monitoring costs, (ii) by reducing risk premium, and therefore lowering financial and credit costs, (iii) by reducing loan losses, (iv) by increasing loan supply and reducing loan rationing, and finally (v) by encouraging customers to use bank products. this study argued that “the most direct impact of social capital on banks is through the increase in the confidence and trust of the arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 51 individuals participating in different banking relationships in the institutions and systems that control the social, economic and political welfare of the society”. in a study by cassar, crowley and wydick (2007), it was stated that social capital, when defined as the personal trust between individuals and social cohesion between different groups, has a significant positive effect on the borrowers’ repayment performance; but when defined as a simple familiarity between individuals or an individual’s general trust in society, it has little impact on the borrowers’ performance. from this study, it can be concluded that high social capital has an impact on the repayment of loans, and can, therefore, reduce the default rate and increase banking profitability. a similar conclusion can also be made about the effect of social capital on the banking risk. in a study carried out by guiso et al., (2004) on the role of social capital in financial development in italy, it was concluded that in the areas that enjoy a higher level of social capital, households are more likely to use banking services and invest in stocks, and have better access to formal finance. this research found that social capital was stronger wherever the law was weaker and had a greater impact on less educated people. these results indicated that in the study area, informal relationships had replaced the official law. several studies including elyasiani and goldberg (2004), ferri and messori (2000) have suggested that higher social capital allows banks to lend more easily, thus increasing the loan supply. some studies including dowla (2006), van bastelaer (1999), and quinones and seibel (2000) have stressed the important role of banks in creating social capital. in a case study on bangladesh’s grameen bank, dowla (2006) examined the impact of microfinancing services provided by this bank on the social capital. in this study, it was stated that grameen bank has promoted social capital of bangladeshi community by creating trust, norms, and networks. to be more precise, the founders of grameen bank suggested that it is possible to create trust by forming an institutional incentive structure. in response, customers returned this trust by repaying the loans on time, and this cycle created social capital in the form of trust. furthermore, the emphasis of grameen bank on transparency in financial transactions, credit discipline, and timely repayment turned creditworthiness into a social norm; a contribution that greatly simplified the work of the descendants of this bank. this bank also improved the social capital of bangladeshi society by creating horizontal and vertical networks. in a study conducted by mayoux (2001), the role of microfinancing in creating social capital was investigated in cameroon. according to this study, microfinancing not only increases social capital but also empowers women. in a similar study, anderson, locker, and nugent (2002) stated that microfinancing programs can improve social and human capital by transforming the household’s production and consumption behavior with further focus on women, by encouraging group work, and by enhancing the people’s decision-making and risk-taking capability in regard to lending and borrowing. sanyal (2009) also drew a similar conclusion in regard to the effect of microfinancing on women’s social capital. the above-reviewed body of literature indicates that there is a mutual relationship between arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 52 social capital and the performance of banks. indeed, banks can improve the social capital of a society by means of financing and credit facilities. but the present study is focused on the impact of social capital on the performance of banks, and more specifically their profitability. from the existing literature about the effect of social capital on the performance of banks, the following can be inferred: 1. in a society with a higher level of social capital, information transparency improves banking profitability by reducing the costs of customer identification, collateral registration, and data verification. 2. a higher level of social capital results in banks and customers showing more commitment to the terms of contracts. 3. in a society with a higher level of social capital, people are more likely to use banking and financial services. in general, the existing literature leads us to the hypothesis that higher social capital helps banks earn more profit. in this paper, we test this hypothesis by examining the effect of social capital on the profitability of the banks in the european union countries. 3. methodology and data 3.1. methodology this study was conducted using the generalized method of moments (gmm). since our data were of panel type, they could be processed using the static panel models. but the static method cannot properly deal with serial correlation, heteroscedasticity, and endogeneity of explanatory variables. thus, we chose to use the gmm estimator to avoid these issues. in comparison, the two-stage least squares 2sls estimators are weaker from several perspectives. also, the typical instrumental variables (iv) estimators are ineffective in dealing with heteroscedasticity. the reasons for the use of gmm technique for the estimation of panel data can be summarized as follows: • the dynamic panel data gmm is well-suited for the cases where the number of cross-sectional variables (n) exceeds the number of time periods (t), and has been developed to control the dynamic panel bias in such circumstances (bond & baltagi, 2008). the reason to choose gmm is because of the point that is mentioned: bigger n than t. in the sample: t=9 and n=23. • in dynamic panel data models, the endogeneity problem has been resolved by using the model variables as instrumental variables. in this method of estimation, any regression variable that is not correlated with the error term (including lagged and difference variables) can potentially be an instrumental variable (greene, 2008). • unlike 2sls estimators, this method can properly address the collinearity problem. in the dynamic gmm method, lagged differences and levels of variables are used as instrumental variables, so they are expected to be more orthogonal with respect to other variables. hence, the collinearity problem is partly resolved (hsiao, 2003). arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 53 there are two methods of estimation with dynamic panel gmm. the principles of the dynamic gmm models were first introduced by arellano and bond (1991), who formulated an estimator called first-difference gmm (dgmm) accordingly. later, arellano and bover (1995), blundell and bond (1998) introduced some changes to the dgmm formulations and developed the system gmm (sgmm). the difference between these two methods is in the way individual effects are incorporated into the model. while the arellano-bond method uses the differencing, the arellano-bover method utilizes the orthogonal deviation. in the arellano-bond method, the entire set of lags are used as instrumental variables, but in the sgmm method, only the lagged levels are used for this purpose. while being less renowned than the arellano-bond method, sgmm has a set of merits that make it preferable. the notable advantages of sgmm over dgmm include higher accuracy and lower sample size bias, which lead to relatively more efficient estimations (baltagi, 2008). we used two tests to make sure that the method is wellsuited for model estimation. the first test was the hansen’s j-test, which checks the validity of the overidentifying restrictions (i.e. the validity of instrumental variables). according to baum (2006), hansen’s j-test is the leading tool for assessing the suitability of a gmm model for an estimation. the second test was the test of first-order and second-order serial correlation of residuals (ar1 and ar2), which also checks the validity of instrumental variables. according to arlano and bond (1991), in the gmm estimation, error terms must have a first-order serial correlation (ar1) but not a second-order serial correlation (ar2). here, the rejection of the null hypothesis means that the lags of dependent variables are endogenous, hence the condition stated in the equation is rejected. 3.2. data analyses were performed on the data pertaining to 231 eu countries for the period 2008-2016 the data were collected from the reports of the european central bank (ecb) and other available resources (see table 1). table1. data resources category variable data source banking return on average assets european central bank (ecb) capital adequacy ratio real interest rate asset cost-to-income ratio macroeconomic gross domestic product eurostat european commission inflation macro-social trust in others european social survey(ess) opportunistic behavior the following is a description of the variables used in the analysis. 1 because of some data gaps, croatia, greece, latvia, malta and romania are excluded. arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 54 3.2.1. dependent variables return on average assets (roaa): this rate was obtained by dividing the bank’s net profit by the value of its assets. 3.2.2. independent variables internal bank variables these are the variables that influence the profitability of a bank depending on its capability and performance. these variables were derived from the related literature: a. capital adequacy ratio (car): this ratio was calculated by dividing the bank’s capital by its risk-weighted assets. many researchers including olalekan and adeyinka (2013), mathuva (2009), blum (1998), alper and anbar (2011), and kosmidou (2008) have proven the existence of a relationship between capital adequacy and bank profitability. in our model, this variable is expressed in percentage. b. real interest rate (rir): this is the interest rate after adjustment with respect to the inflation rate. a high rir reduces the demand for loan but instead increases the income earned from each loan. the effect rir on bank profitability has been studied by abreu and mendes (2001), vong and chon (2006), bennaceur and goaied (2008), demiguc-kunt and huizinga (1999), albertazzi and gambacorta (2009). c. cost-to-income ratio (cir): this ratio shows how efficient the bank expenditures are, and how much they result in a profit. naturally, efficient expenditure can be expected to increase the bank’s profitability. many works including olson and zoubi (2011) and ariff and luc (2008) have studied the association of cir with profitability. many other papers such as pasiouras and kosmidou (2007), francis (2013) and hassan and bashir (2005) have confirmed that this ratio strongly influences the bank’s profitability. d. bank size (lnasset): the size of a bank, which is represented by the size of its assets, is one of the variables that affect its profitability. the impact of bank size on its profitability has been confirmed in many studies including athanasoglou, brissimis & delis (2005), gul, irshad and zaman (2011), dietrich and wanzenried (2011), sufian and razali chong (2008), and vong and si chan (2006). in our model, the logarithmic form of this variable was used. macroeconomic variables the performance of a bank as an integral part of a country’s economic system is certainly affected by the macroeconomic conditions of that country. the effects of macroeconomic variables such as employment, gdp growth, and inflation on the performance of banking systems have been explored and emphasized in many studies. a. gross domestic product (lngdp): in our model, the logarithmic form of the gross domestic product (gdp) was used as a measure of the economic growth of countries. the impact of this macroeconomic variable on the profitability of the banks has been investigated by many researchers including albertazzi and gambacorta (2010), bashir (2003), goddard, molyneux, wilson (2004), sufian and chong (2008), tan and floros (2012), aburime (2008). arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 55 b. inflation (inf): inflation is widely regarded among the most important determinants of economic activities. notable works on the effect of inflation on the performance of banks include the studies of boyd, levin and smith (2001), ali, akhtar and ahmed (2011), santoni (1986), tan and folors (2012). main variables: social capital indicators the aim of the study was to investigate the impact of social capital on the profitability of the banks. as it is discussed in the introduction, because of the definition of social capital that is adopted in the article, i used the european social survey (ess)2 results to measure social capital through two variables. the first is social trust and the second is the perception of opportunistic behavior in others. ess has collected the answers through questionnaires by asking two questions as table 2 in a likert scale with an 11-point scale that offers a range of answer options between 0-10. table 2. questions to measure social capital by ess question likert scale indicator of abbreviatio n scale 0 10 most people can be trusted or you can't be too careful you can't be so careful most people can be trusted general trust gt most people try to take advantage of you or try to be fair most people try to take advantage of me most people try to be fair opportunistic /fair behavior fb by studying the answers of questions above, we can see the social capital level in european countries as shown in diagram 1. diagram 1. median of social capital indicators by country source: author's calculations from ess. as can be seen in diagram 1, social capital indicators are at high levels in scandinavian countries such as denmark, finland and sweden. on the contrary, some eastern and southern european countries such as bulgaria, portugal czech republic and cyprus are not at a 2 the surveys are done in 8 rounds from 2002 to 2016. 0 2 4 6 8 10 general trust fair behavior arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 56 desirable level of social capital. in the model, we apply the percentage of optimists as a representative of social capital in two aspects: general trust and fair behavior. the hypothesis is that banking profitability increases with social capital. 4. results this section presents the results obtained by model estimation based on the described data. the results regarding the effect of social capital on the profitability of eu banks are presented in table 3. table 3. gmm model results independent variables model (1)-general trust model (2)fair behavior dependent variable:roaa dependent variable:roaa gt -1.23*** fb -0.31* l1 0.11*** 0.12*** car 0.49*** 0.54*** rir -.051*** -.05*** lnasset -0.25*** lngdp 2.08*** 1.42*** inf 0.03* 0.11*** wald chi2(8) 8987.43*** 6366.59*** sargan test 10.96 12.64 ar(1) -1.66* -1.73* ar(2) 1.32 1.42 n 207 207 source: author's calculations. *, ** and *** represent significance at 10%, 5% and 1% respectively. model (1) suggests that in line with theoretical expectations, capital adequacy (car), which represents the bank’s stability and risk covering capability, has a significant positive effect on the bank’s profitability. the model shows that the real interest rate (rir) has a negative impact on the banks’ profitability. this inverse relationship is expectable since at higher real interest rate, borrowers and businesses become less interested in purchasing financial facilities, and this affects the banks’ main source of income that is the repayment of loans. the cost-to-income ratio (cir), which reflects the banks’ spending inefficiency, was also found to have a significant negative impact on profitability. it means that there are inefficient costs in the banking system. the gdp growth (lngdp) was found to have a significant positive effect on the profitability. it can be argued that an active economy encourages business owners and people to further interact with the banks. inflation also had a significant positive impact in this respect. the main variable, i.e. social capital (general trust), also showed a significant negative impact on profitability. the more general trust, the fewer profits for the banking system. according to the results, the hypothesis is rejected. as we discussed in the literature segment, arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 57 social capital makes banking system agents act with higher levels of trust and with a lower level of transactional costs. lower costs should result in much profitability but it did not occur in this model as in model (2). largely similar results were obtained in model (2), which investigated the effects of variables on return on average asset. but the main difference is the core topic of the debate. it concludes with a rejection of the hypothesis of a negative correlation between opportunistic behavior and economic performance. as in model (2), more opportunistic behavior led to a more profitable performance of banks. it is against our expectations and should be justified. it can be justified by other available data resources. by studying some other surveys such as trust in european institutions (parliament, commission and central bank) it can be concluded that although opportunistic behavior is a negative factor influencing social capital and therefore banking performance but it is shown that in societies such as bulgaria, portugal, and cyprus with high levels of opportunistic behavior, banking is one of the most trustable institutions than others such as parliament and politicians and others. it can be concluded from what happened in 2008-2009. financial crisis occurred and systemic trust in european countries decreased (roth, 2009). however, it can be seen in countries with a low level of social capital that the change in trust to other institutions higher than in the central bank as a supervisory entity that prevent risks and defaults (diagram 2). diagram 2. percent of growth of trust in european institutions (2008-2016) source: author's calculations from eurostat. as it can be seen in diagram 2, decreasing of trust in european institution for central bank is less than others in societies with lower level of social capital. in these condition although trust in banks are decreasing compare with other institution is higher. in a society where people do not trust each other so much, banks are more trustable because transactions in banks are done by supervision of an institution like central bank. the validity of the results was confirmed by the sargan test and also by the ar1 and ar2 tests, which showed the existence of first-order correlation and the inexistence of second-order correlation between residuals. -50% -45% -40% -35% -30% -25% -20% -15% -10% -5% 0% czechia cyprus portugal bulgaria cenral bank european comission european parliament arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 58 5. summary and conclusion this study aimed to determine whether the social capital of a society, as a macro, macroeconomic or macrosocial concept, has an impact on the performance of the banks operating in that society. the definition of social capital in this study was the structure that facilitates the relationship between people and groups, and thereby economic relations as a macro concept. it means that our concentration is general trust and context is the entire social network. using the gmm technique, we analyzed the data pertaining to 23 eu countries for the period 2008-2016 and found that social capital has a negative impact on the profitability of the banks against our hypothesis. it can be argued that less social capital leads to decreased bank's profit. the justification is in the societies with lower levels of social capital trust in banks is more than other institutions. banks contracts are more clear and are under the supervision of a central bank that makes their transactions relatively secure. apart from this, we should consider the 2008-2009 financial crisis that reduces level of social capital in all the european countries but as it was discussed before the native growth rate of trust in the banking system is less than others. in such a condition, banks are more trustable than others. references abreu, m., mendes, v., (2001). commercial bank interest margins and profitability: evidence from some eu countries. in: paper presented at the proceedings of the pan-european conference jointly organised by the iefs-uk & university of macedonia economic & social sciences, thessaloniki, greece, may 17–20, 2001. aburime, t. (2008). determinants of bank profitability: company-level evidence from nigeria. ssrn electronic journal. doi: https://doi.org/10.2139/ssrn.1106825 akçomak, i. s., & ter weel, b. (2009). social capital, innovation and growth: evidence from europe. european economic review, 53(5), 544-567. doi: https://doi.org/10.1016/j.euroecorev.2008.10.001 albertazzi, u., & gambacorta, l. (2010). bank profitability and taxation. journal of banking & finance, 34(11), 2801-2810. doi: https://doi.org/10.1016/j.jbankfin.2010.06.003 ali, k., akhtar, m. f., & ahmed, h. z. (2011). bank-specific and macroeconomic indicators of profitability-empirical evidence from the commercial banks of pakistan. international journal of business and social science, 2(6), 235-242. alper, d & anbar, a. (2011). bank specific and macroeconomic determinants of commercial bank profitability: empirical evidence from turkey. business and economics research journal, 2(2), 139-152. anderson, c. l., locker, l., & nugent, r. (2002). microcredit, social capital, and common pool resources. world development, 30(1), 95-105. doi: https://doi.org/10.1016/s0305750x(01)00096-1 ariff, m., & luc, c. (2008). cost and profit efficiency of chinese banks: a non-parametric https://doi.org/10.2139/ssrn.1106825 https://doi.org/10.1016/j.euroecorev.2008.10.001 https://doi.org/10.1016/j.jbankfin.2010.06.003 https://doi.org/10.1016/s0305-750x(01)00096-1 https://doi.org/10.1016/s0305-750x(01)00096-1 arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 59 analysis. china economic review, 19(2), 260-273. doi: https://doi.org/10.1016/j.chieco.2007.04.001 arellano, m., & bond, s. (1991). some tests of specification in panel data: monte carlo evidence and an application to employment equations. review of economics and statistics 58(2): 277-297. doi: https://doi.org/10.2307/2297968 arellano, m., & bover, o. (1995). another look at the instrumental variables estimation of error components models. journal of econometrics, 68: 29-51. doi: https://doi.org/10.1016/03044076(94)01642-d athanasoglou, p.p., brissimis, s n. & delis, m. d. (2005). bank-specific, industry specific and macroeconomic determinants of bank profitability. mpra paper, no.153. baltagi, b. (2008). econometric analysis of panel data. john wiley & sons. bashir, a. h. m. (2003). determinants of profitability in islamic banks: some evidence from the middle east. islamic economic studies, 11(1), 31-57. baum, c. f. (2006). an introduction to modern econometrics using stata. stata press. bennaceur, s., & goaied, m. (2008). the determinants of commercial bank interest margin and profitability: evidence from tunisia. frontiers in finance and economics, 5 (1), 106-130. doi: https://doi.org/10.12691/ijbrm-1-1-3 berkowitz, j. & white, m.(2004). bankruptcy and small rms access to credit. rand. journal of economics, 35(1), 6984. doi: https://doi.org/10.2307/1593730 blum, j. (1999). do capital adequacy requirements reduce risks in banking? journal of banking & finance, 23(5), 755-771. doi: https://doi.org/10.1016/s0378-4266(98)00113-7 blundell, r., & bond, s. (1998). initial conditions and moment restrictions in dynamic panel data models. journal of econometrics, 87: 11-143. doi: https://doi.org/10.1016/s03044076(98)00009-8 boyd, j. h., levine, r., & smith, b. d. (2001). the impact of inflation on financial sector performance. journal of monetary economics, 47(2), 221-248. doi: https://doi.org/10.1016/s0304-3932(01)00049-6 cassar, a., crowley, l., & wydick, b. (2007). the effect of social capital on group loan repayment: evidence from field experiments. the economic journal, 117(517). doi: https://doi.org/10.1111/j.1468-0297.2007.02016.x chan, h. s., & vong, p. i. (2006). determinants of bank profitability in macau. coleman, j. s. (1988). social capital in the creation of human capital. american journal of sociology, 94, s95-s120. doi: https://doi.org/10.1086/228943 demirgüç-kunt, a., & huizinga, h. (1999). determinants of commercial bank interest margins and profitability: some international evidence. the world bank economic review, 13(2), 379-408. doi: https://doi.org/10.1596/1813-9450-1900 dietrich, a., & wanzenried, g. (2011). determinants of bank profitability before and during the crisis: evidence from switzerland. journal of international financial markets, institutions and money, 21(3), 307-327. doi: https://doi.org/10.1016/j.intfin.2010.11.002 dowla, a. (2006). in credit we trust: building social capital by grameen bank in https://doi.org/10.1016/j.chieco.2007.04.001 https://doi.org/10.2307/2297968 https://doi.org/10.1016/0304-4076(94)01642-d https://doi.org/10.1016/0304-4076(94)01642-d https://doi.org/10.2307/1593730 https://doi.org/10.1016/s0378-4266(98)00113-7 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1016/s0304-3932(01)00049-6 https://doi.org/10.1111/j.1468-0297.2007.02016.x https://doi.org/10.1086/228943 https://doi.org/10.1596/1813-9450-1900 https://doi.org/10.1016/j.intfin.2010.11.002 arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 60 bangladesh. the journal of socio-economics, 35(1), 102-122. doi: https://doi.org/10.1016/j.socec.2005.12.006 elyasiani, e., & goldberg, l. g. (2004). relationship lending: a survey of the literature. journal of economics and business, 56(4), 315-330. doi: https://doi.org/10.1016/s01486195(04)00022-0 ferri, g., & messori, m. (2000). bank–firm relationships and allocative efficiency in northeastern and central italy and in the south. journal of banking & finance, 24(6), 1067-1095. doi: https://doi.org/10.1016/s0378-4266(99)00118-1 francis, m. e. (2013). determinants of commercial bank profitability in sub-saharan africa. international journal of economics and finance, 5(9), 134. doi: https://doi.org/10.5539/ijef.v5n9p134 fulkerson, g. m., & thompson, g. h. (2008). the evolution of a contested concept: a meta‐ analysis of social capital definitions and trends (1988–2006). sociological inquiry, 78(4), 536-557. doi: https://doi.org/10.1111/j.1475-682x.2008.00260.x glaeser, e. l., laibson, d., & sacerdote, b. (2002). an economic approach to social capital. the economic journal, 112(483). doi: https://doi.org/10.1111/1468-0297.00078 goddard, j. a., molyneux, p., & wilson, j. o. (2004). dynamics of growth and profitability in banking. journal of money, credit, and banking, 36(6), 1069-1090. doi: https://doi.org/10.1353/mcb.2005.0015 greene, w. h. (2008). econometric analysis (6th ed.). new jersey, upper saddle river: pearson international. doi: grootaert, c., & van bastelaer, t. (eds.). (2002). the role of social capital in development: an empirical assessment. cambridge university press. doi: https://doi.org/10.1017/cbo9780511492600 guiso, l., sapienza, p., & l. zingales (2011). civic capital as the missing link. in j. benhabib, m. o. jackson and a. bisin, handbook of social economics (vol. 1a., pp. 417-480.) doi: https://doi.org/10.1016/b978-0-444-53187-2.00010-3 guiso, l., sapienza, p., & zingales, l. (2004). the role of social capital in financial development. the american economic review, 94(3), 526-556. doi: https://doi.org/10.1257/0002828041464498 gul, s., irshad, f., & zaman, k. (2011). factors affecting bank profitability in pakistan. romanian economic journal, 14(39). hall, r. e., & jones, c. i. (1997). levels of economic activity across countries. the american ecnonomic review, 87(2), 173-177. doi: https://doi.org/10.2139/ssrn.80211 hanifan, l. j. (1916). the rural school community center. the annals of the american academy of political and social science, 67(1), 130-138 doi: https://doi.org/10.1177/000271621606700118 hasan, i., hoi, c. k., wu, q., & zhang, h. (2017). social capital and debt contracting: evidence from bank loans and public bonds. journal of financial and quantitative analysis, 52(3), 1017-1047. doi: https://doi.org/10.1017/s0022109017000205 https://doi.org/10.1016/j.socec.2005.12.006 https://doi.org/10.1016/s0148-6195(04)00022-0 https://doi.org/10.1016/s0148-6195(04)00022-0 https://doi.org/10.1016/s0378-4266(99)00118-1 https://doi.org/10.5539/ijef.v5n9p134 https://doi.org/10.1111/j.1475-682x.2008.00260.x https://doi.org/10.1111/1468-0297.00078 https://doi.org/10.1353/mcb.2005.0015 https://doi.org/10.1017/cbo9780511492600 https://doi.org/10.1016/b978-0-444-53187-2.00010-3 https://doi.org/10.1257/0002828041464498 https://doi.org/10.2139/ssrn.80211 https://doi.org/10.1177/000271621606700118 https://doi.org/10.1017/s0022109017000205 arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 61 hassan, m. k., & bashir, a. h. m. (2003, december). determinants of islamic banking profitability. in 10th erf annual conference, morocco (pp. 16-18). hsiao, c. (2003). analysis of panel data (2nd ed). cambridge university press. doi: https://doi.org/10.1017/cbo9780511754203 iyer, s., kitson, m., & toh, b. (2005). social capital, economic growth and regional development. regional studies, 39(8), 1015-1040. doi: https://doi.org/10.1080/00343400500327943 jin, j., kanagaretnam, k., lobo, g. j., & lee-chin, m. (2015). social capital and bank accounting transparency. michael lee-chin & family institute for strategic business studies working paper; 2015-09. knack, s., & keefer, p. (1997). does social capital have an economic payoff? a cross-country investigation. the quarterly journal of economics, 112(4), 1251-1288. doi: https://doi.org/10.1162/003355300555475 kosmidou, k. (2008). the determinants of banks' profits in greece during the period of eu financial integration. managerial finance, 34(3), 146-159. doi: https://doi.org/10.1108/03074350810848036 mathuva, d. m. (2009). capital adequacy, cost income ratio and the performance of commercial banks: the kenyan scenario. the international journal of applied economics and finance, 3(2), 35-47. doi: https://doi.org/10.3923/ijaef.2009.35.47 mayoux, l. (2001). tackling the down side: social capital, women’s empowerment and micro‐ finance in cameroon. development and change, 32(3), 435-464. doi: https://doi.org/10.1111/1467-7660.00212 membiela-pollán, m., & pena-lópez, j. a. (2017). clarifying the concept of social capital through its three perspectives: individualistic, communitarian and macrosocial. european journal of government and economics, 6(2), 146-170. doi: https://doi.org/10.17979/ejge.2017.6.2.4327 olalekan, a., & adeyinka, s. (2013). capital adequacy and banks’ profitability: an empirical evidence from nigeria. american international journal of contemporary research, 3(10), 8793. olson, d., & zoubi, t. a. (2011). efficiency and bank profitability in mena countries. emerging markets review, 12(2), 94-110. doi: https://doi.org/10.1016/j.ememar.2011.02.003 ongena, s., & smith, d. c. (2000). what determines the number of bank relationships? crosscountry evidence. journal of financial intermediation, 9(1), 26-56. doi: https://doi.org/10.1006/jfin.1999.0273 pasiouras, f., & kosmidou, k. (2007). factors influencing the profitability of domestic and foreign commercial banks in the european union. research in international business and finance, 21(2), 222-237. doi: https://doi.org/10.1016/j.ribaf.2006.03.007 pastor, j., & torrosa‐ausina, e. (2008). social capital and bank performance: an international comparison for oecd countries. the manchester school, 76(2), 223-265.doi: https://doi.org/10.1111/j.1467-9957.2007.01058.x https://doi.org/10.1017/cbo9780511754203 https://doi.org/10.1080/00343400500327943 https://doi.org/10.1162/003355300555475 https://doi.org/10.1108/03074350810848036 https://doi.org/10.3923/ijaef.2009.35.47 https://doi.org/10.1111/1467-7660.00212 https://doi.org/10.17979/ejge.2017.6.2.4327 https://doi.org/10.1016/j.ememar.2011.02.003 https://doi.org/10.1006/jfin.1999.0273 https://doi.org/10.1016/j.ribaf.2006.03.007 https://doi.org/10.1111/j.1467-9957.2007.01058.x arash nayebyazdi / european journal of government and economics 8(1), june 2019, 48-62 62 pennar, k., & mueller, t. (1997). the ties that lead to prosperity. business week, 15(1), 153154. putnam, r. d. (1993). the prosperous community. the american prospect, 4(13), 35-42 quinones, b. r., & seibel, h. d. (2000). social capital in microfinance: case studies in the philippines. policy sciences, 33(3-4), 421-433. doi: https://doi.org/10.1023/a:1004833123412 santoni, g. j. (1986). a private central bank: some old english lessons. in central bankers, bureaucratic incentives, and monetary policy (pp. 219-239). springer, dordrecht. doi: https://doi.org/10.1007/978-94-009-4432-9_12 sanyal, p. (2009). from credit to collective action: the role of microfinance in promoting women's social capital and normative influence. american sociological review, 74(4), 529550. doi: https://doi.org/10.1177/000312240907400402 sufian, f., & chong, r. r. (2008). determinants of bank profitability in a developing economy: empirical evidence from the philippines. asian academy of management journal of accounting & finance, 4(2) 91-112. talavera, o., xiong, l., & xiong, x. (2012). social capital and access to bank financing: the case of chinese entrepreneurs. emerging markets finance and trade, 48(1), 55-69. doi: https://doi.org/10.2753/ree1540-496x480103 tan, y., & floros, c. (2012). bank profitability and gdp growth in china: a note. journal of chinese economic and business studies, 10(3), 267-273. doi: https://doi.org/10.1080/14765284.2012.703541 tan, y., & floros, c. (2012). bank profitability and inflation: the case of china. journal of economic studies, 39(6), 675-696. doi: https://doi.org/10.1108/01443581211274610 uphoff, n., & wijayaratna, c. m. (2000). demonstrated benefits from social capital: the productivity of farmer organizations in gal oya, sri lanka. world development, 28(11), 18751890. doi: https://doi.org/10.1016/s0305-750x(00)00063-2 van bastelaer, t. (1999). imperfect information, social capital and the poor's access to credit. iris center working paper, 234. vong, p. i., & chan, h. s. (2009). determinants of bank profitability in macao. macau monetary research bulletin, 12(6), 93-113. westlund, h., & adam, f. (2010). social capital and economic performance: a meta-analysis of 65 studies. european planning studies, 18(6), 893-919. doi: https://doi.org/10.1080/09654311003701431 woolcock, m. (1998). social capital and economic development: toward a theoretical synthesis and policy framework. theory and society, 27(2), 151-208. doi: https://doi.org/10.1023/a:1006884930135 xie, w . (2013). the impact of social capital on bank risk-taking (master's thesis, lingnan university, hong kong). doi: http://dx.doi.org/10.14793/econ_etd.24. zak, p. j., & knack, s. (2001). trust and growth. the economic journal, 111(470), 295-321. doi: https://doi.org/10.1111/1468-0297.00609 https://doi.org/10.1023/a:1004833123412 https://doi.org/10.1007/978-94-009-4432-9_12 https://doi.org/10.1177/000312240907400402 https://doi.org/10.2753/ree1540-496x480103 https://doi.org/10.1080/14765284.2012.703541 https://doi.org/10.1108/01443581211274610 https://doi.org/10.1016/s0305-750x(00)00063-2 https://doi.org/10.1080/09654311003701431 http://dx.doi.org/10.14793/econ_etd.24 https://doi.org/10.1111/1468-0297.00609 front81 ejge oficial 8-1 contents81 number 8, issue 1, june 2019 4575 © theauthor(s)2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 1 (2024), pages 67-84 https://doi.org/10.17979/ejge.2024.13.1.9595 submitted: march 26, 2023 accepted: january 24, 2024 published: june 6, 2024 article unveiling the impact of european structural funds for innovation in andalusia, spain diego sande veiga1,* 1 university of santiago de compostela, spain *correspondence: diego.sande.veiga@usc.es abstract. this study evaluates the impact of european union structural funds for innovation on key business indicators related to growth, profitability, and innovation at the regional level. we use the case of andalusia during the period 2007-2020, a spanish region benefiting from these funds, focusing on the erdf-innterconnecta program which supports business collaboration in r&d projects. while some indicators, showed improvement, others did not. by analyzing these mixed results, we aim to inform the planning, design, and implementation of future regional innovation policies. keywords: policy evaluation; business innovation policies; structural funds; business performance; employment; regional development. jel classification: o32; o38; o12; l25 1. introduction while the great recession was spreading across the planet thanks to financial globalization and the limitations of the global financial system, the convergence objective regions of the european union (eu), such as andalusia, saw the exit from austerity of the crisis pushed towards a progressive reduction in the budget allocation of the european structural investment funds (esif) (sande, 2020; 2018). when the crisis broke out, the eu had just approved a program called the technology fund aimed at promoting business research, development and innovation (r&d&i) in the period 2007-2013. this program, which was endowed with more than 2,000 million euros, continued in the period 2014-2020 with the well-known smart growth program (sgp). while it is true that companies play a crucial role within the national (sánchez, martínez, & arellano, 2018) and regional innovation systems (nis/ris) (karlsen, 2013), and that the business fabric tends to be more fragile in the territories of the convergence objective as a consequence of the low interaction with other agents, it is also true that the small size of companies and their productive specialization in low technological intensity sectors (hollanders et al., 2014; 2016; 2019) hinder the ability to https://creativecommons.org/licenses/by-nc/4.0/ 68 sande veiga absorption of resources (sande, 2020; sande & vence, 2021). however, the nature and objective of these funds caused them to be directed mainly at the business fabric and not at the systemic configuration (sande & vence, 2019; cooke, uranga & etxebarria, 1998; león & fernández, 2006; nikitskaya et al., 2014), with the consequences derived from this policy choice. considering the previous starting point, the analysis of the results of the innterconnecta program has been selected as the objective of this original study since this has been the main technological policy aimed at supporting companies in a peripheral region such as andalusia. within this framework, this research aims to evaluate whether the resources and projects financed by the structural funds have had a positive impact on the growth, economic performance, and innovation of andalusian companies. the originality of this work consists of addressing the study of the impact of esif from the microeconomic level, as opposed to the usual macroeconomic approaches. furthermore, it is the first impact study of innovation programs at the andalusian (peripheral region) level that examines the results of this policy according to the set of selected indicators. the importance of the choice of the analyzed indicators is motivated by the fact that they reflect some of the main characteristics of the companies in the territory that could be promoted through the public policies implemented in the region. in this way, the results extracted will allow for the generation of knowledge to improve the planning, design, and application of european policies for regional innovation in the territory. the article is structured as follows: the second section reviews the importance of innovation policies, as well as the promotion of business development in peripheral and technologically backward territories. it also includes a description of the policies analyzed; the third section describes the methodology used and the main data sources; the fourth section analyses the results of the program and assesses the impact on the main business indicators; finally, the fifth section sets out the conclusions drawn from the analysis carried out and makes recommendations for the future planning of innovation policies in the autonomous region. 2. literature review the first part of this section reviews the importance of european policies at the regional level, focusing on the impact of the structural funds for innovation on the business fabric. the programs under study are described below. 2.1 european policies and the regional level: impact of the structural funds for innovation on the business fabric the esif has been one of the eu's main funding instruments, aimed at trying to reduce economic disparities between regions and member states. despite the existence of studies that have questioned the ability of the esif to reduce european regional inequalities over time (rodriguezpose, 2000; rodriguez-pose & fratesi, 2004; ederveen, de groot & nahuis, 2006; van der zwetet al., 2017; neagu et al., 2017; di caro & fratesi, 2021), other research has shown its contribution to economic cohesion in europe (caldas, dollery & marques, 2018; lópez-villuendas & del campo, unveiling the impact of european structural funds for innovation 2022; maynou et al., 2014) and in spain, highlighting the role played in the growth of objective 1 regions (cancelo, faíña & lópez-rodríguez, 2005; sosvilla-rivero, bajo & díaz, 2003; de la fuente, 2003). however, some authors (lembcke & menon, 2017) insist on the criticism that regional inequalities persist over the years and, in some cases, even increase. indeed, for others, the results of cohesion policy do not seem very robust (krieger-boden, 2018). the main feature that defines the results provided by the academic literature on the impact of european policies at the regional level is the disparity of conclusions obtained in the different studies. thus, while some authors (bernini & pelegrini, 2011; vivarelli, 2014) find that subsidized companies improve their production indicators compared to non-subsidized companies, for sande (2022b) and vojtovičj (2016) companies financed by esif do not achieve economic results that contribute to their growth in these indicators. for others (breidenbach, mitze & schmidt, 2019), negative funding effects on the business fabric significantly correlate with lower levels of regional institutional quality. according to sande (2020), part of the inefficiencies in the application of european structural resources earmarked for areas such as innovation in objective 1 regions could be due to factors such as the lack of systemic vertebration at regional level, the lack of alignment with industrial policies and even the existence of leakage of resources towards more advanced and central regions. in the same vein, gancarczyk et al. (2022) argue that the co-evolutionary theoretical framework focuses on the so-called interaction mechanisms (im), meaning the processes underlying industrial policy that allow for a better understanding of policy roles and industrial development paths. in this context, innovation policies would be understood as a proactive complement to the need for structural industrial change. regarding the impact of structural funds for innovation on firms, studies have shown different results. for example, bachtrögler & hammer (2018) have used techniques such as propensity score matching (psm), with which they have found different effects of the structural and cohesion funds for six european countries. in general, according to the authors, firms tend to hire more workers and increase their capital stock. other studies, such as those carried out by sande (2022a), have even shown differences in the impact of the esif in objective 1 regions depending on the size of the recipient firms. furthermore, baláž, jeck & balog (2023) point to the importance of spatial proximity and personal contacts to achieve better innovation results. in summary, some policies have positive impacts on key measures, whereas others do not (oecd, 2023). in a research conducted by dvouletý, srhoj & pantea (2021), the findings show mostly the positive outcomes of the grants on firm survival, employment, tangible/fixed assets, sales/turnover, with mixed findings for labour productivity and total factor productivity (tfp). however, we point out that there are significant differences concerning the time period of analysis (investigating short-term vs longterm outcomes) and, importantly, the heterogeneity of effects concerning firm size and age, region, industry and intensity of support. when talking about the impact of esif on business results, again, the main feature that defines the results provided by the academic literature is the disparity of conclusions obtained in the different studies. while some studies (hartsenko & sauga, 2012; arbidane & tarasova, 2018) find that activities financed by the esif would increase the competitiveness of firms and business activity in general, for others there is generally no impact of structural resources on business 70 sande veiga performance and productivity (sande, 2022b; dumciuviene, stundziene & startiene, 2015; vojtovičj, 2016; bachtrögler & hammer, 2018). other authors (lucaciu, 2018) highlight the positive effect of complementarity of the different funds. this is a fact that cannot be dissociated from the results defended by milio (2007), who asserts that the effects of resources would be linked to the existence of administrative capacity to implement the funds. related exclusively to the impact on a very important indicator such as employment, the literature has traditionally argued that technological development and innovation can help solve socio-economic problems and foster employment (cozzens et al., 2007; alzugaray, medores & sutz, 2012; florio & moretti, 2014). but the truth is that the implementation of policies aimed at technological development and innovation has sometimes implied social inequalities and greater inequalities in the labour market (lee & rodríguez-pose, 2013). so much so that, according to several studies, there is no positive impact of subsidies on business employment (bernini & pelegrini, 2011; bondonio, 2014; bachtrögler & hammer, 2018; sande, 2022b), and sometimes the funds are even used to solve other financial problems of companies (komninos, musyck & iain reid, 2014; sergej, 2016). however, cerqua and pellegrini (2014) found that the impact of subsidies on employment, investment, and turnover is positive and statistically significant, while the effect on productivity is mostly negligible. nevertheless, nemethova, siranova & sipikal (2019) found a positive and significant impact on labour productivity that disappears shortly after 1 year following subsidy allocation. research results have shown that innovation policies must be tailored to the characteristics of territories (tödtling & trippl, 2005; foray & van ark, 2007; mccann & ortegaargilés, 2013; sande, 2020). for this reason, the eu and regional governments have not implemented innovation policies in line with the needs of nis and their enterprises in the design of innovation policies. in this context, regions that need to improve their technological capacities classified as objective 1 regions in the 2000-2006 programming period and convergence regions from 2007-2013-, need to design strategies appropriate to their situation (heijs, 2001; pastor et al., 2010). in the spanish case, andalusia, galicia, extremadura and castilla-la-mancha maintained this situation. support through the esif for the financing of technological innovation has shown mixed results over time. thus, while some studies have found positive results of technological innovation policies for the business fabric (musyck & reid, 2007; croce, martí & murtinu, 2013; bronzini & piselli, 2016; le & jaffe, 2017; segarra-blasco, 2018), other studies reflect moderate results of direct public funding in peripheral contexts (sande & vence, 2021; sande, 2022a; sande & sande, 2023), or even lack of results for certain contexts and indicators (clausen, 2009; blasio, fantino & pellegrini, 2015; lewandowska, stopa & humenny, 2015). in this sense, mieszkowski & barbero (2021) explain less-than-adequate conditions in rural areas and smaller counties, which may limit the potential for attraction and implementation of esif. in order to find out the results of the esif on business innovation -and particularly the results on growth, performance and innovation of companies in the medium term-, this analysis presents the data from an instrument such as the innterconecta programme, firstly belonging to the tf and then to the sgp, which has been applied for almost a decade in a peripheral and moderately innovative autonomous community such as andalusia. unveiling the impact of european structural funds for innovation 2.2 the policies object of study the european council approved the birth of the tf as a program dedicated to the promotion of business r&d&i (ministerio de economía y hacienda, 2007). this tf had a continuity framework for business innovation after the approval of the sgp (ministerio de hacienda y administraciones públicas, 2014). table 1 shows the main descriptive data on this funding, including territorial allocation, objectives, and eligible actions. table 1. descriptive data on the technology fund (tf) and the smart growth program (sgp) technology fund smart growth assignment to spain 2.248,45 m€ + 3.939,18 m€ assignment to andalusia 976,80 m€ 1.612 m€* territorial distribution funds -70% for obj. convergence regions (galicia, andalusia, extremadura and castilla la mancha) -15% for phasing-in regions (growth effect) -10% for competitiveness objective regions -5% for phasing-out regions (statistical effect) -plurirregional objectives -to articulate and integrate the spanish r&d&i system with the regional innovation systems -promote business innovation, especially in smes in convergence objective regions -to support the transfer of research results to companies -promoting r&d and innovation -improving the use, quality and access to information and communication technologies (ict). -improve the communication and competitiveness of smes. -widen the base of the science, technology and enterprises system ( stes) by attracting smes to r&d&i -promote gender equality in r&d&i subsided actions -to vertebrate the innovation system, incorporating smes into innovative activity. -to create and consolidate technology and research centres oriented towards relations with companies. -promote the transfer of research from pris to companies. -attract smes and other agents to innovation and research activity. -capacity building for the development of r&d&i activities supported by competitive scientific infrastructures at european and international level. -stimulating and fostering capacities for the implementation of business r&d&i projects. -promoting the incorporation of researchers and r&d&i personnel and fostering mobility between public sector personnel and the business fabric, as well as the creation of high added value employment. source: own elaboration. *note: total forecast expenditure (boscá et al., 2016). 72 sande veiga table 2. erdf-innterconecta programme: descriptive data. technology fund smart growth assignment to spain 262 m€ 210 m€ territorial distribution -andalusia 150 m€ -galicia: 105 m€ -plurirregional -extremadura: 7 m€ -castilla la mancha: the region does not participate in these call for proposals subsided areas -all, as long as they stimulate employment and increase added value (ministerio de economía y competitividad, 2013) health, demographic change and wellbeing, food safety and quality; safe, efficient and clean energy, smart, sustainable and integrated transport; action on climate change; social change and innovations, digital economy and society; security, safety and defence dimension and amounts subsidized in the projects (andalusia) up to 5 m€ between 1-4 m€ project requirements formation of an economic interest grouping (eig) or consortium project duration two and three-year projects (ministerio de ciencia e innovación, 2012). objectives support for large r&d projects increasing business r&d expenditure use of existing infrastructures mobilisation of smes greater involvement of stakeholders and promotion of innovative culture internationalisation of innovation experimental development and cooperation between companies the european and regional development fund-innterconecta (erdf_interconecta) calls arose in the middle of the 2007-2013 programming period, in view of the low implementation that was being achieved by the tf. the birth of this programme was based on the premise of supporting integrated experimental development projects of a public-private nature, of a strategic nature, large in size and aimed at developing new technologies in technological areas with international economic projection. the aid granted until 2020 under this programme financed projects with no thematic limitation, on the condition that they fostered employment, were of a high technological level, and promoted activities that favoured an increase in the added value of the participating companies (ministerio de economía y competitividad, 2013). the basic information on the innterconecta programme is broken down in table 2. 3. methodology and data sources throughout this section, the information is divided into two sub-sections. the first sub-section explains the methodology used and the limitations of the work, while the second describes the data sources and the main data used in the current research. unveiling the impact of european structural funds for innovation 3.1 methodology this paper proposes a microeconomic analysis with a strong empirical character. the qualitative and quantitative analysis of the data generated for the erdf-innterconecta calls in andalusia has been used, in addition to the data corresponding to the autonomous community in the case of multi-regional calls. we used the propensity score matching (psm) methodology for the statistical analysis, which analyses the covariances between two groups of values: on the one hand, the companies not participating in the policy and, on the other, the participating companies. we performed the statistical test, for which the number of companies in the control sample with data for the indicators was 355, while for those participating in innterconecta it was 337. for each of the indicators, these values may vary due to the occasional lack of data for some entities, which discourages the statistical study from being disaggregated by cnae groups. for both the control sample and the participating companies, we first calculate the number for which matching has taken place. the mean of the values () and the standard deviation () are then studied. in case the value of the standardised mean difference (or smd), measured through the d-index, is greater than 0.1, imbalance would be observed and we should apply the psm, however, in order to provide a broader information of the results, we have chosen to also calculate the psm for those values whose d-index was less than 0.1. the propensity score was then estimated by applying a logit model in which the outcome variable is a binary variable indicating whether the policy was implemented or not, using the r software package matchit. among the different methods to perform the matching (xact matching, nearest neighbour, optimal matching, full matching and caliper matching,...), we selected the nearest neighbour, as we considered it more appropriate to match each individual in the treatment group with the individual in the control group that has the closest propensity score. using one-by-one nearest neighbour ps matching =n(1)ic, one treated unit i ∈ t is matched to one control unit j ∈ c. that is, that individual is selected from the candidates pairing whose propensity score is the most similar to the propensity score of the individual to be paired in the case group. there is a one-to-one matching, in the former an element of the control group is used more than once. the values of the variables have been taken at the end of the period, as a result for these indicators. once the test is completed, we include the p-value, which indicates whether there are significant differences between the group that participates in the policy and the group that does not. nevertheless, the proposed impact measurement study has had to face some problems and limitations. firstly, there is the problem of self-selection, arising from the companies' ability to choose whether or not to participate in the calls for proposals of the program under analysis. secondly, the problem of endogeneity has been addressed, insofar as the decision of public administrations when approving the program has been an external trigger that has allowed firms to participate (garcía-nicolás & cantos, 2015; lago & martínez, 2004). finally, the results could have distorting biases in case there were governmental interests in the selection of funding to projects and companies (martí, 2020). to isolate the effect of these problems, the use of the propensity score matching statistical technique has been proposed, which, by accounting for and analyzing covariances, allows the effect of a policy to be estimated. 74 sande veiga 3.2. data sources and main data this paper proposes a microeconomic analysis with a strong empirical character. the qualitative and quantitative analysis of the data generated for the erdf-innterconecta calls in andalusia has been used, in addition to the data corresponding to the autonomous community in the case of multi-regional calls. the data used have been extracted from various sources, including the spanish ministry of finance, the ministry of economy, finance and european funds of the andalusian regional government, the spanish national statistics institute (ine), and official journals of the administrations, which have enabled an understanding of the current situation in andalusia. however, the centre for technological and industrial development (cdti) has been the main provider of raw data on the projects carried out and the companies participating in the innterconecta programme. finally, it was ardán's1 information that made it possible to construct the data for the indicators analyzed. this section also describes the main data extracted from the projects carried out in the innterconecta program in its calls for proposals in andalusia. to this end, we will first synthesize the information on the projects financed, the samples of companies analyzed, and the technological areas involved. thanks to the projects financed by innterconecta, around 2,000 companies have been able to carry out projects throughout spain. although the tf had mobilised more european resources in andalusia, the slightly smaller size of the multiregional sgp projects has allowed a similar level of business participation to be maintained (table 3). based on previous data, the average number of participating companies per project was 4.17, also taking into account the participation of research organizations in the consortia. the average amount of investment per company participating in the funded projects has been calculated as total amount/numb. of companies. the average budget of each of the 827 participating companies identified amounted to 639,679.85 €, while cdti support covered almost half of this amount on average, with 302,406.91 €. table 3. approved projects and participating companies in innterconecta-andalusia. callf for proposals approved projects (*) numb. companies requested projects (*) numb. companies 1st reg. call 2011 31 195 74 410 2nd reg. call 2013 41 211 59 255 3rd call 2015* 131 511 269 946 4th call 2016* 64 246 231 822 5th call 2018* 67 229 n/a n/a total 334 1392 633 2433 source: own elaboration based on data from cdti and boe. note: *plurirregional 1 the ardan database belongs to the vigo free zone consortium, and provides accounting information from companies' annual accounts. unveiling the impact of european structural funds for innovation of the more than eight hundred andalusian companies identified as participants in the regional and multi-regional innterconecta calls for proposals, data was available for a total of 337 companies that received grants between 2012-2020. in order to gain a deeper understanding of the impact of the innterconecta programme on these companies, a comparison was made between the evolution of their indicators. on the other hand, the sample of companies of andalusian origin participating in the programme analyzed has been compared to another general sample of 355 companies in the autonomous community that have not participated in the policy (represented as cs), and which has been extracted from ardán. the control sample has been selected from a random sample of andalusian companies in the ardán database, but which have not participated in the policy analyzed. in addition, criteria such as the size of the companies, their status as previously innovative or not (in accounting terms) and the sectors of activity to which they belong have been taken into account. with regard to the classification of the innterconecta companies analyzed (337) and the control sample (355), the characteristics of both samples are quite similar (see table 4). taking into account the description of the subsidized projects, we can see that they have been classified in techonological areas. the technological areas to which the 337 companies participating in innterconecta belong are mainly industrial manufacturing activities (34.12%) and professional, scientific and technical activities (27.60%), which often correspond to consultancy and specialised services. the rest of the innterconecta resources went mainly to the following technological areas: information and communication technologies (9.20%), retail and wholesale trade (8.90%) and construction (8.31%). information about the registered offices of the companies participating in the policy analyzed is also provided below. these companies are concentrated primarily in seville, malaga and cordoba, and to a lesser extent in jaen. other andalusian territories have hardly any participation at all (table 5). the map in figure 1 also includes companies participating in the policy from other regions. table 4. descriptive statistics of the projects analyzed at the beginning of the period. number of participating companies // control sample 337 355 small and medium enterprises 247 (73.29%) 345 (97.18%) large enterprises 90 (26.71%) 10 (2.82%) number of companies per project 4.17 role in the projects leaders 63 (18.69%) partners 274 (81.31%) role in innovation of participants // control sample previously innovative (accountancy data) 10 (2.97%) 3 (0.85%) non-innovative (accountancy data) 327 (93.03%) 352 (99.15%) source: own elaboration based on ardán and cdti data 76 sande veiga table 5. regions and provinces to which the companies participating in the policy analyzed belong. region province numb. of firms total (%) region province numb. of firms total (%) andalusia almería 28 8.31% galicia a coruña 3 0.89% cádiz 24 7.12% pontevedra 1 0.30% córdoba 26 7.72% basque contry bizkaia 3 0.89% granada 19 5.64% guipúzcoa 1 0.30% huelva 7 2.08% ávala 1 0.30% jaén 22 6.53% asturias oviedo 1 0.30% málaga 32 9.50% cataluña barcelona 12 3.56% sevilla 90 26.71% cantabria santander 1 0.30% castilla-león león 1 0.30% murcia murcia 1 0.30% navarra pamplona 6 1.78% madrid madrid 57 16.91% valencia valencia 1 0.30% source: own elaboration based on ardán and cdti data. figure 1. spatial location of the companies participating in the innterconecta programme in andalusia, by registered office. source: own elaboration based on ardán data (sande, 2024) 4. data analysis the first part of this section contains a comparative analysis of the evolution of the indicators analyzed. the second part analyses the data using the selected methodology. unveiling the impact of european structural funds for innovation 4.1. comparative evolution of the indicators analyzed the amount of resources allocated to the promotion of business innovation through the erdfinnterconecta programme has been significant for the convergence regions, and especially in andalusia. for this reason, the expected impact should be relevant (although it is true that part of the results can be assessed over a longer period of time). in order to characterise the impact of this programme in andalusia, the behaviour of the main indicators of growth, results and innovation of the companies participating in this programme has been analyzed, without ignoring the fact that the evolution shown by these companies is also influenced by other factors of the socio-economic context, such as the systemic crisis suffered, legislative changes, the multiple corporate business management strategies, and others. this paper deals with the evolution of the following three blocks of business indicators: the first group includes indicators related to business growth [revenue, gross value added (gva) and employment], the second group includes indicators of business performance [profitability and result of the year], while the third group analyses the impact on innovation indicators [investment in research and development]. we take as a starting point the accounting information of the companies participating in innterconecta obtained in raw form from the ardán database. the presentation of the information analyzed will make it possible to visualise the difference in the behaviour of the companies as a result of their participation in the innterconecta programme. when analyzing the aggregate change of the indicators, the andalusian companies participating in the innterconecta programme generally show positive results in all the previously selected indicators: revenue, gva, employment, economic profitability, result for the year and investment in development, with the exception of research investment. table 6 summarizes the information on the relative impact for each sample. utilizing outcome indicators for the difference-in-differences (did) analysis allows assessing the impact of funding on the specific outcomes of interest. this combined approach can help address potential selection bias, control for confounders, and provide a robust estimation of the treatment effect. we estimate the causal effect of funding on these outcomes. did effect has been calculated as follows = (outcome in treatment group, post-intervention outcome in treatment group, pre-intervention) (outcome in control group, post-intervention outcome in control group, pre-intervention). a statistically significant and positive did effect would imply that the intervention (innovation funding) had a positive impact on the outcomes of interest. the results show that, in general, the outcomes for the treatment group (int) did not improved more than those for the control sample (cs) after the intervention (table 7). table 6. aggregate change and relative impact after business participation in innterconecta by indicators sample income (€) gva (€) employment (nº jobs) profitability (%) result for the year (€) research invest. (€) development invest (€) innterconecta companies 4,843,392,963 (+) 5,617,463,938 (+) 48,494 (+) 0.01 (-) -1,103,529,238 (+) 10,320,772.88 (+) 112,441,237.3 (+) control sample 8,722,662,171 (+) 1,879,833,987 (+) 36,414 (+) 0.04 (+) 301,259,773 (+) 11,858,106.54 (+) 75,893,895.83 (+) source: own elaboration from ardán and cdti data. 78 sande veiga table 7. outcome indicators for did of the groups studied (€,%) sample revenue gva employees profitability* result of the year research invest. development invest. int-cs -3,879,269,208 3,737,629,951 12,080 -0.053 -1,404,789,011 -1,537,333.66 36,547,341.43 int-cs (%) -218.17 -216.08 -280.67 -172.60 -347.07 320,981.89 -4,597,553.88 source: own elaboration based on ardán and cdti data. note: *% the graphical analysis shows a positive evolution for the values of all the indicators analyzed, with the sole exception of investment in research. precisely when it comes to a policy aimed at promoting business innovation (see figures 2 to 9). figure 2. comparative evolution of revenue, innterconecta-andalusia companies 2007-2020, (index 2007=100, log10(x)) figure 3. comparative evolution of gva, innterconecta-andalucía companies 2007-2020, (index 2007=100, log10(x)) figure 4. comparative evolution of employment, innterconecta-andalucía companies 2007-2020, (index 2007=100, log10(x)) figure 5. comparative evolution of profitability, innterconecta-andalucía companies 2007-2020, (index 2007=100, log10(x)) figure 6. comparative evolution of result of the year, innterconecta-andalucía companies 2007-2020 (index 2007=100, log10(x)) figure 8. comparative evolution of research investment, innterconecta-andalucía companies 2007-2020, (index 2007=100, log10(x)) unveiling the impact of european structural funds for innovation figure 9. comparative evolution of development investment, innterconecta-andalucía companies 2007-2020, cnae c, f, g, j, m (index 2007=100) 4.2 statistical analysis as it was explained in the methodology section, we used the propensity score matching (psm) methodology for the statistical analysis, which analyses the covariances between two groups of values: on the one hand, the companies not participating in the policy and, on the other, the participating companies. the results of the statistical analysis would indicate that there is a significant difference for both groups. the participation of innovative companies in the innterconecta programme would have a significant impact on four of the indicators analyzed: revenue, gva, employment and investment in development (table 8). table 8. results of the statistical analysis of business indicators using psm. revenue gva employment profitability result of the year research investment development investment companies control sample (mc) 338 334 338 337 336 17 41 companies innterconecta 232 232 231 232 232 16 60  control sample 37,594,951.57 7,748,712.04 145.87 0.07 1,170,650.66 697,571.59 1,851,110.89  innterconecta 240,967,630.71 58,883,060.40 624.60 0.04 14,388,704.16 621,703.48 5,678,831.06  control simple 94,740,323.08 15,045,398.71 305.85 0.13 8,052,397.65 1,859,626.66 6,711,620.05  innterconecta 1,069,090,259.76 251,410,141.20 1,760.90 0.13 185,986,867.98 1,939,440.07 15,409,012.33 d-index (dme) 0.268 0.287 0.379 0.232 0.100 0.040 0.322 p-value 0.007431 0.003692 0.0003532 0.703 0.292 0.7966 0.045 table 9 lists the results observed for each of the main business indicators analyzed. table 9. summary of the results of positive impact (+), or not demonstrated (=) of the analyzed policy, by indicator revenue gva employment profitability result of the year research investment development investment innterconecta companies + + + = = = + 80 sande veiga 5. conclusions the conclusions of this paper can be divided into two sub-sections. the first relates to policy implications, while the second draws the main recommendations derived from the results of this research. 5.1. implications the tf and sgp, endowed with significant amounts to promote technological development in andalusia in the 2007-2013 and 2014-2020 programming periods, raised expectations for the development of business innovation within the andalusian innovation system. however, the positive results observed for several of the indicators (revenue, gva, employment and investment in development) should not distract us from the lack of impact on others (results of the exercise, profitability and research investment), making the achievements more moderate than apparently expected. thus, while previous studies for other autonomous regions showed a positive, albeit moderate, impact of the esif for innovation on the main innovation indicators of firms (sande & vence, 2021), the present study partially confirms this general result (innovation research is an exception). regarding growth and performance indicators, this study confirms the results of previous research in other regions (sande, 2022a; sande, 2024). 5.2. recommendations as a consequence of the above results on the impact of the policy on business innovation in andalusia, and with a view to achieving greater efficiency in the results of r&d&i policies (particularly for investment in research), smaller projects could have been set up, which would have made it possible to finance initiatives that responded to a greater extent to the possible investment needs of the smaller business fabric, particularly smes, which constituted a specific objective of the programme. similarly, more specific objectives could have been included in these innovation programmes, which would facilitate the evaluation of funding for the innovation ecosystem (e.g. indicating expected sectoral impacts in terms of employment, expected benefits, patents, etc.). similarly, it would be worth considering whether these types of programmes aimed at reducing the innovation gap in the peripheral territories should have incorporated measures that would make it possible to more clearly promote the priority thematic areas defined in the regional smart specialisation strategy, which would facilitate greater alignment between policies and strategies while promoting those areas with the greatest technological projection. unveiling the impact of european structural funds for innovation references alzugaray s., medores l., & sutz j. (2012). building bridges: social inclusion problems as research and innovation issues. review of policy research 29, 776–796. https://doi.org/10.1111/j.15411338.2012.00592.x arbidane, i. & tarasova, m. (2018). assessment of the impact of the eu structural funds on business in latvia. society, integration, education. international scientific conference, vol vi, may 25th26th, 15-29 bachtrögler, j & hammer, c. (2018). who are the beneficiaries of the structural funds and the cohesion fund and how does the cohesion policy impact firm-level performance? oecd economics department working paper, nº 1499, 1-47. https://doi.org/10.1787/67947b82-en baláž, v., jeck, t., & balog, m. (2023). knowledge transfers and business performance in creative networks: the case study of the slovak creative voucher scheme. journal of the knowledge economy, 1-27 bernini, c., & pellegrini, g. (2011). how is growth and productivity in private firms affected by public subsidy? evidence from a regional policy. regional science and urban economics, 41(3), 253– 265. https://doi.org/10.1016/j.regsciurbeco.2011.01.005 blasio, g., fantino, d. & pellegrini, g. (2015). evaluating the impact of innovation incentives: evidence from an unexpected shortage of funds. industrial and corporate change, vol. 24(6), 1285-1314. https://doi.org/10.1093/icc/dtu027 bondonio, d. (2014). revitalizing regional economies through enterprise support policies: an impact evaluation of multiple instruments. european urban &regional studies, 21(1), 79103. https://doi.org/10.1177/0969776411432986 boscá, j., escribá, j., ferri, j. & murgui, m.j. (2016). el impacto de los fondos feder (2014‐2020) sobre el crecimiento y el empleo de las regiones españolas. fedea working paper [online]. available at: https://infoitijaen.es/wp-content/uploads/2022/01/evaluacion-ex-ante-de-impacto-macro economico-feder-2014-2020-informe-final.pdf breidenbach, p., mitze, t., & schmidt, c.m. (2019). eu regional policy and the neighbour's curse: analyzing the income convergence effects of esif funding in the presence of spatial spillovers. jcms: journal of common market studies, 57(2), 388-405 bronzini, r. & piselli, p. (2016). the impact of r&d subsidies on firm innovation. research policy, vol. 45(2), 442-457. https://doi.org/10.1016/j.respol.2015.10.008 caldas, p., dollery, b., & marques, r. c. (2018). european cohesion policy impact on development and convergence: a local empirical analysis in portugal between 2000 and 2014. european planning studies, 26(6), 1081-1098 cancelo, j.r., faíña, j.a., lópez-rodríguez, j. (2005). the effect of structural fund spending on the galician region: an assessment of the 1994-1999 and 2000-2006 galician csfs. fundación de las cajas de ahorros, documento de trabajo nº 224/2005. cerqua, a., & pellegrini, g. (2014). do subsidies to private capital boost firms' growth? a multiple regression discontinuity design approach. journal of public economics, 109, 114-126. https://doi.org/10.1016/j.jpubeco.2013.11.005 clausen, t. (2009). do subsidies have positive impacts on r&d and innovation activities at the firm level? structural change and economic dynamics, vol. 20(4), 239-253. https://doi.org/ 10.1016/j.strueco.2009.09.004 cooke, p., uranga, m.g., & etxebarria, g. (1998). regional systems of innovation: an evolutionary perspective. environment and planning a: economy and space, 30(9), 1563–1584. https://doi.org/10.1068/a301563 cozzens s., kallerud e., ackers l., gill b., harper j. & santos-pereira t. (2007) problems of inequality in science, technology, and innovation. policy james martin institute working paper 5, oxford. croce, a., martí, j. & murtinu, s. (2013). the impact of venture capital on the productivity growth of european entrepreneurial firms: ‘screening’ or ‘value added’ effect? journal of business venturing, vol. 28(4), 489-510. https://doi.org/10.1016/j.jbusvent.2012.06.001 de la fuente, a. (2003). el impacto de los fondos estructurales: convergencia real y cohesión interna. hacienda publica española 165, 129-148 di caro, p. & fratesi, u. (2021). one policy, different effects: estimating the region-specific impacts of eu cohesion policy. journal of regional science, vol. 62(1), 307-330. https://doi.org/ 10.1111/jors.12566 dumciuviene, d., stundziene, a., startiene, g. (2015). relationship between structural funds and economic indicators of the european union. inzinerine ekonomika-engineering economics, 2015, 26(5), 507–516. https://doi.org/10.5755/j01.ee.26.5.8831 https://doi.org/10.1111/j.1541-1338.2012.00592.x https://doi.org/10.1111/j.1541-1338.2012.00592.x https://doi.org/10.1787/67947b82-en https://doi.org/10.1016/j.regsciurbeco.2011.01.005 https://doi.org/10.1093/icc/dtu027 https://doi.org/10.1177/0969776411432986 https://infoitijaen.es/wp-content/uploads/2022/01/evaluacion-ex-ante-de-impacto-macroeconomico-feder-2014-2020-informe-final.pdf https://infoitijaen.es/wp-content/uploads/2022/01/evaluacion-ex-ante-de-impacto-macroeconomico-feder-2014-2020-informe-final.pdf https://doi.org/10.1016/j.respol.2015.10.008 https://doi.org/10.1016/j.jpubeco.2013.11.005 https://doi.org/10.1016/j.strueco.2009.09.004 https://doi.org/10.1016/j.strueco.2009.09.004 https://doi.org/10.1016/j.jbusvent.2012.06.001 https://doi.org/10.1111/jors.12566 https://doi.org/10.1111/jors.12566 https://doi.org/10.5755/j01.ee.26.5.8831 82 sande veiga dvouletý, o., srhoj, s. & pantea, s. (2021). public sme grants and firm performance in european union: a systematic review of empirical evidence. small bus econ 57, 243–263. https://doi.org/10.1007/s11187-019-00306-x ederveen, h., de groot, l.f. & nahuis, r. (2006). fertile soil for structural funds?a panel data analysis of the conditional effectiveness of european cohesion policy. kyklos, vol. 59(1), 17-42. https://doi.org/10.1111/j.1467-6435.2006.00318.x florio, m. & moretti, l. (2014). the effect of business support on employment in manufacturing: evidence from the european union structural funds in germany, italy and spain. european planning studies, vol. 22(9), 1802-1823. https://doi.org/10.1080/09654313.2013.805731 foray, d. & van ark, b. (2007). smart specialisation in a truly integrated research area is the key to attracting more r&d to europe. knowledge economists policy brief, no.1 gancarczyk, m, najda-janoszka, m., gancarczyk, j. & hassink, r. (2022). exploring regional innovation policies and regional industrial transformation from a coevolutionary perspective: the case of małopolska, poland. economic geography, november, 1-31. https://doi.org/10.1080/ 00130095.2022.2120465 garcía-nicolás, c. & cantos, j.m. (2015). estudio empírico del impacto de los fondos estructurales sobre la inversión pública: la aplicación del principio de adicionalidad. conference: xxii encuentro de economía pública: reformas y nuevos retos de los estados de bienestar: eficiencia y equidad at: university of cantabria. hartsenko, j., & sauga, a. (2012). does financial support from the eu structural funds has an impact on the firms’ performance: evidence from estonia. in proceedings of 30th international conference mathematical methods in economics (pp. 260-65). karvina: school of business administration, silesian university. heijs, j. (2001). sistemas nacionales y regionales de innovación y política tecnológica: una aproximación teórica. iaif working papers, nº 24, 2001 hollanders, h., es-sadki, n. & kanerva, m. (2016). regional innovation scoreboard 2016. european comission. belgium: brussels hollanders, h., es-sadki, n., buligescu, b., rivera, l., griniece, e., roman, l., et al. (2014). regional innovation scoreboard 2014. maastricht: innometrics-merit, european comission hollanders, h., es-sadki, n., merkelbach, i. (2019). regional innovation scoreboard 2019. maastrich: innometrics-merit, european comission karlsen, j. (2013). the role of anchor companies in thin regional innovation systems lessons from norway. syst pract action res 26, 89–98. https://doi.org/10.1007/s11213-012-9266-4 komninos, n., musyck, b., & iain reid, a. (2014). smart specialisation strategies in south europe during crisis. european journal of innovation management, 17(4), 448-471 krieger-boden, c. (2018). what direction should eu cohesion policy take? in cesifo forum (vol. 19, no. 1, pp. 10-15). münchen: ifo institut-leibniz-institut für wirtschaftsforschung an der universität münchen. lago, s. & martínez, d. (2004). convergencia y política regional: algunas reflexiones sobre el caso español. cuestiones clave de la economía española, perspectivas actuales. centro de estudios andaluces, consejería de la presidencia [online]. available at: https://centrodeestudios andaluces.es/datos/publicaciones/iv_jornadas.pdf#page=66 le, t. & jaffe, a.b. (2017). the impact of r&d subsidy on innovation: evidence from new zealand firms. economics of innovation and new technology, 26 (5), 429-452. https://doi.org/10.1080/10438599.2016.1213504 lee, n. & rodríguez-pose, a. (2013). innovation and spatial inequality in europe and the united states. journal of economic geography, vol. 13, 1-22. https://doi.org/10.1093/jeg/lbs022 lembcke, a., & menon, c. (2017). making policy evaluation work: the case of regional development policy. oecd science, technology and industry policy papers, 38, 2-35. https://doi.org/10.1787/c9bb055f-en león, m.d. & fernández, a.m. (2006). teoría evolucionista y sistema de innovación: implicaciones institucionales y organizaciones de la innovación tecnológica y el desarrollo económico regional. boletín económico de ice, información comercial española, nº 2876, 25-44 lewandowska, a., stopa, m. & humenny, g. (2015). the european union structural funds and regional development. the perspective of small and medium enterprises in eastern poland. european planning studies, 23(4). https://doi.org/10.1080/09654313.2014.970132 lópez-villuendas, a.m., & del campo, c. (2022). analysis of economic convergence in the galicianorthern portugal euroregion in the period 1980-2019. revista galega de economía, 31(2), 119. https://doi.org/10.15304/rge.31.2.8289 lucaciu, l. (2018). a look at the evaluation framework for smart growth programmes. revista românească pentru educaţie multidimensională, 10(3), 60-76 https://doi.org/10.1007/s11187-019-00306-x https://doi.org/10.1111/j.1467-6435.2006.00318.x https://doi.org/10.1080/09654313.2013.805731 https://doi.org/10.1080/00130095.2022.2120465 https://doi.org/10.1080/00130095.2022.2120465 https://doi.org/10.1007/s11213-012-9266-4 https://centrodeestudiosandaluces.es/datos/publicaciones/iv_jornadas.pdf#page=66 https://centrodeestudiosandaluces.es/datos/publicaciones/iv_jornadas.pdf#page=66 https://doi.org/10.1080/10438599.2016.1213504 https://doi.org/10.1093/jeg/lbs022 https://doi.org/10.1787/c9bb055f-en https://doi.org/10.1080/09654313.2014.970132 https://doi.org/10.15304/rge.31.2.8289 unveiling the impact of european structural funds for innovation martí, c. (2020). la colaboración público-privada en la obtención de los medios para la defensa. el caso del fondo europeo de defensa. documento de trabajo opex, nº 105/2020. observatorio de la política exterior española, ministerio de defensa, secretaría general de política de defensa. available at: https://fundacionalternativas.org/publicaciones/la-colaboracion-publico-privadaen-la-obtencion-de-los-medios-para-la-defensa-el-caso-del-fondo-europeo-de-defensa/ maynou, l., saez, m., kyriacou, a. & bacaria, j. (2014). the impact of structural and cohesion funds on eurozone convergence, 1990–2010. regional studies, vol. 50(7), 1127-1139. https://doi.org/10.1080/00343404.2014.965137 mccann, p., & ortega argiles, r. (2013). modern regional innovation policy. cambridge journal of regions, economy and society, volume 6(2),187–216. https://doi.org/10.1093/cjres/rst007 mieszkowski, k., & barbero, j. (2021). territorial patterns of r&d+i grants supporting smart specialisation projects funded from the esif in poland. regional studies, 55(3), 390-401 milio, s. (2007). can administrative capacity explain differences in regional performances? evidence from structural funds implementation in southern italy. regional studies, vol. 41(4), 429-442. https://doi.org/10.1080/00343400601120213 ministerio de ciencia e innovación (2012). orden ecc/1808/2012, de 18 de junio por la que se modifica la orden cin/1729/2011, por la que se establecen las bases reguladoras para la concesión de subvenciones destinadas a fomentar la cooperación estable público-privada en investigación y desarrollo (i+d), en áreas de importancia estratégica para el desarrollo de la economía española (federinnterconecta). madrid, españa: boe núm. 194, de 14-08-2012. available at: https://www.boe.es/buscar/doc.php?id=boe-a-2011-10854 ministerio de economía y competitividad (2013). resolución de 30 de enero de 2013, del centro para el desarrollo tecnológico industrial, por la que se aprueba la convocatoria del año 2013 para la comunidad autónoma de galicia del procedimiento de concesión de subvenciones destinadas a fomentar la cooperación estable público-privada en investigación y desarrollo (i+d), en áreas de importancia estratégica para el desarrollo de la economía española (feder-innterconecta). boe núm. 46, de 22-02-2013. available at: https://www.boe.es/diario_boe/txt.php?id=boe-a2013-2012 ministerio de economía y hacienda. (2007). programa operativo de i+d+i por y para el beneficio de las empresas-fondo tecnológico. madrid: age. available at: http://www.dgfc.sepg.hacienda.gob.es /sitios/dgfc/es-es/ipr/fcp0713/p/pop/documents/pofondotecnologico_07_2011.pdf ministerio de hacienda y administraciones públicas. (2014). programa operativo feder de crecimiento inteligente 2014-2020. available at: https://www.fondoseuropeos.hacienda.gob.es/sitios/dgfc/ es-es/ipr/fcp1420/p/prog_op_plurirregionales/documents/po_feder_crecimiento _inteligente_2014_2020.pdf musyck, b., & reid, a. (2007). innovation and regional development, do european structural funds make a difference? european planning studies, vol. 15(7), 961-983. https://doi.org/10.1080/ 09654310701356696 neagu, o. m., michelsen, k., watson, j., dowdeswell, b., & brand, h. (2017). addressing health inequalities by using structural funds. a question of opportunities. health policy, 121(3), 300-306 nemethova, v., siranova, m., & sipikal, m. (2019). public support for firms in lagging regions—evaluation of innovation subsidy in slovakia. science and public policy, volume 46 (2), 173–183. https://doi.org/10.1093/scipol/scy046 nikitskaya, e. f., safronova, a. a., zhidkova, o. n., anatolyevich, a., ivanov, l. n. i. s., & mamedova, n. a. (2014). evolutionary aspects of innovation development. life science journal, 11(8), 516-519 oecd (2023). framework for the evaluation of sme and entrepreneurship policies and programmes 2023, oecd studies on smes and entrepreneurship, oecd publishing, paris https://doi.org/10.1787/a4c818d1-en pastor, j.m., raymond, j.l., roig, j.l. & serrano, l. (2010). fondos estructurales, capital humano y convergencia en las regiones objetivo 1 en españa. papeles de economía española, nº 123, 3954 rodríguez-pose a. (2000) economic convergence and regional development strategies in spain: the case of galicia and navarre, eib papers 5(1), 89–115 rodríguez-pose a., & fratesi, u. (2004). between development and social policies: the impact of european structural funds in objective 1 regions, regional studies, vol. 38(1), 97–113. https://doi.org/10.1080/00343400310001632226 sánchez, e.c., martínez, r.c., & arellano, y.m. (2018). evolutionism, innovation and national development: the role of national companies in current economic dynamics. competition forum, indiana tomo 16, n.º 1, 116-122 sande, d. (2018). análise dos fondos europeos para innovación rexional: avaliación do programa federinnterconecta do fondo tecnolóxico para galicia en 2007-2015 [phd thesis]. university of santiago de compostela: santiago de compostela https://fundacionalternativas.org/publicaciones/la-colaboracion-publico-privada-en-la-obtencion-de-los-medios-para-la-defensa-el-caso-del-fondo-europeo-de-defensa/ https://fundacionalternativas.org/publicaciones/la-colaboracion-publico-privada-en-la-obtencion-de-los-medios-para-la-defensa-el-caso-del-fondo-europeo-de-defensa/ https://doi.org/10.1080/00343404.2014.965137 https://doi.org/10.1093/cjres/rst007 https://doi.org/10.1080/00343400601120213 https://www.boe.es/buscar/doc.php?id=boe-a-2011-10854 https://www.boe.es/diario_boe/txt.php?id=boe-a-2013-2012 https://www.boe.es/diario_boe/txt.php?id=boe-a-2013-2012 http://www.dgfc.sepg.hacienda.gob.es/sitios/dgfc/es-es/ipr/fcp0713/p/pop/documents/pofondotecnologico_07_2011.pdf http://www.dgfc.sepg.hacienda.gob.es/sitios/dgfc/es-es/ipr/fcp0713/p/pop/documents/pofondotecnologico_07_2011.pdf https://www.fondoseuropeos.hacienda.gob.es/sitios/dgfc/es-es/ipr/fcp1420/p/prog_op_plurirregionales/documents/po_feder_crecimiento_inteligente_2014_2020.pdf https://www.fondoseuropeos.hacienda.gob.es/sitios/dgfc/es-es/ipr/fcp1420/p/prog_op_plurirregionales/documents/po_feder_crecimiento_inteligente_2014_2020.pdf https://www.fondoseuropeos.hacienda.gob.es/sitios/dgfc/es-es/ipr/fcp1420/p/prog_op_plurirregionales/documents/po_feder_crecimiento_inteligente_2014_2020.pdf https://doi.org/10.1080/09654310701356696 https://doi.org/10.1080/09654310701356696 https://doi.org/10.1093/scipol/scy046 https://doi.org/10.1787/a4c818d1-en https://doi.org/10.1080/00343400310001632226 84 sande veiga sande, d. (2020). o estrangulamento tecnolóxico de galiza. análise das políticas europeas para innovación rexional durante a gran recesión. editorial laiovento: santiago de compostela sande, d. (2022a). ¿existe impacto de las políticas europeas de innovación regional en las empresas? análisis del programa feder-innterconecta del fondo tecnológico 2007-2013 en galicia”. cuadernos europeos de deusto, vol. 66, 101-132. https://doi.org/10.18543/ced.2369 sande, d. (2022b). large enterprises vs smes: who shows the greatest impact of the structural funds for business innovation in peripheral regions? analysis of the results of the technology fund 2007-2013 in galicia. revista portuguesa de estudos regionais, vol. 62, 57–76. https://doi.org/10.59072/rper.vi62.560 sande, d. (2024). do the structural funds in innovation influence the growth of companies? analysis through the erdf-innterconecta programme in andalusia differentiating by business size and role in the projects. investigaciones regionales-journal of regional research, vol. 58, 5-29. https://doi.org/10.38191/iirr-jorr.24.001 sande, d., & sande, j.r. (2023). evaluación de las políticas europeas de innovación empresarial en el sector tecnológico medioambiental: análisis de la ejecución del programa feder-innterconecta del fondo tecnológico 2007-2013 en galicia. revista de estudios regionales, vol. 126, 85-121 sande, d., & vence, x. (2019). avaliación do impacto do programa fondo tecnolóxico 2007-2013 en galicia: resultados, concentración das axudas e fugas de recursos. revista galega de economía, vol. 28(3), 92-114. https://doi.org/10.15304/rge.28.3.5926 sande, d., & vence, x. (2021). impacto de los fondos estructurales para innovación sobre la innovación empresarial: un análisis a través de los indicadores de empresas participantes en el programa feder-innterconecta de galicia”. revista galega de economía, vol. 30(2), 1-16. https://doi.org/10.15304/rge.30.2.6865 segarra-blasco, a. (2018). subvenciones, préstamos y desgravaciones a la i+d: ¿cuál es su impacto en las empresas catalanas? investigaciones regionales–journal of regional research, 40 (2018), 109140 sergej, v. (2016). the impact of the structural funds on competitiveness of small and medium-sized enterprises. journal of competitiveness, 8(4), 30 sosvilla-rivero, s., bajo o., & díaz c. (2003). sobre la efectividad de la política regional comunitaria: el caso de castilla-la mancha, working papers 2003-25, fedea tödtling, f., & trippl, m. (2005). one size fits all? towards a differentiated regional innovation policy approach. research policy, 34, (8), 1203-1219 van der zwet, a., bachtler, j., ferry, m., mcmaster, i., & miller, s. (2017). integrated territorial and urban strategies-how are esif adding value in 2014-2020? vivarelli, m. (2014) innovation, employment and skills in advanced and developing countries: a survey of the literature. journal of economic issues 48, 123-154. https://doi.org/10.2753/jei00213624480106 vojtovič, s, (2016). the impact of the structural funds on competitiveness of small and medium-sized enterprises. journal of competitiveness, vol. 8(4), 30-45. https://doi.org/10.7441/ joc.2016.04.02 https://doi.org/10.18543/ced.2369 https://doi.org/10.59072/rper.vi62.560 https://doi.org/10.38191/iirr-jorr.24.001 https://doi.org/10.15304/rge.28.3.5926 https://doi.org/10.15304/rge.30.2.6865 https://doi.org/10.2753/jei0021-3624480106 https://doi.org/10.2753/jei0021-3624480106 https://doi.org/10.7441/joc.2016.04.02 https://doi.org/10.7441/joc.2016.04.02 1. introduction 2. literature review 3. methodology and data sources 4. data analysis 5. conclusions references european journal of government and economics 10(1), june 2021, 46-64 european journal of government and economics issn: 2254-7088 liquidity of commercial banks in portugal and spain marco amarala* a instituto politécnico do cávado e do ave, portugal * corresponding author at: mamaral@ipca.pt abstract. liquidity is very important for the functioning of financial markets, especially for the banking sector, because one of the critical aspects in the banking business is precisely the process of transforming short-term funds and placing them in the medium and long term. this paper aims to comprehensively assess the liquidity positions of portuguese and spanish commercial banks through different liquidity ratios for the period from 2002 to 2015 and understand whether the liquidity management strategy differs by bank size. to this end, unconsolidated balance sheet data were used, which were obtained from the banks annual reports. the sample includes a significant part of the portuguese and spanish banking sector (not only by the number of banks, but also by the representation in banks total assets). the results obtained show that spain's banks' liquidity indicator has decreased over the last four years. in contrast, bank liquidity indicator in portugal varied slightly positively during the period 2002-2006 but decreased sharply between 2010 to 2015. bank liquidity increased slightly during the period of the financial crisis in both countries, namely between from 2007 to 2009. finally, it is concluded that smaller banks have less fluctuating liquidity management, i.e., large and medium-sized banks show greater variation in bank liquidity in the period under analysis, i.e., they are less liquid. keywords. banking liquidity; financial entities; crisis jel codes. g21 doi. https://doi.org/10.17979/ejge.2021.10.1.7165 1. introduction studies on bank liquidity have been increasingly highlighted, particularly after the financial crisis that began in 2007 in the united states, which had the impact of the collapse of several banking entities, leading to strong turmoil in financial markets and greater pressure on the regulation of the financial system. in this sense, the basel committee on banking supervision (bcbs), in relation to bank liquidity, has been implementing a set of new standards and rules on the management and monitoring of bank liquidity. the international financial crisis of 2007-2008 demonstrated the importance of prudent liquidity risk management in the banking system. exposure to liquidity risk was a determining factor for the bcbs to implement common rules for banks on liquidity risk management. one of the most important additions to the basel iii banking regulations was the introduction of two minimum financial liquidity standards. these two new banking liquidity indicators follow two distinct, but complementary objectives. thus, quantitative regulation of liquidity risk was introduced through the liquidity coverage ratio (lcr) and the net stable funding ratio (nsfr). and if, on the one hand, the lcr seeks mailto:mamaral@ipca.pt https://doi.org/10.17979/ejge.2021.10.1.7165 marco amaral / european journal of government and economics 10(1), june 2021, 46-64 47 to ensure that banks are in a permanent position to withstand a very adverse financial shock for at least a very short period of time (one month), they should have a buffer of assets of high liquidity quality (assets that can easily be converted into cash). on the other hand, the nsfr is more structural in nature and aims to define the necessary incentives for banks to resort to rather stable financing sources, in order to ensure their resilience to prolonged financial shocks. thus, the net stable funding ratio (nsfr), complements the liquidity coverage ratio (lcr), because while the lcr, consists in resisting a scenario of financial pressure for 30 days, the nsfr consists in overcoming a broader time horizon (1 year). although the bcbs published the final document of the new regulatory framework for banks' liquidity requirements for the lcr in january 2013 (bcbs, 2013) and for the nsfr in october 2014 (bcbs, 2014), its implementation has been delayed throughout several countries. thus, formal implementation in european countries will take place under the revised capital requirements regulation (crr ii), which was published in june 2019 (european commission, 2020) and is expected to come into force from june 2021. in this context, based on the recent critical liquidity situation faced by the financial system on an international level, we aim to assess the liquidity positions of commercial banks in portugal and spain extensively through different liquidity ratios in the period 2002-2015 and understand whether liquidity management strategy is distinct from bank size. to achieve this objective, the unconsolidated balance sheet data of the sample banks, which were obtained from the annual reports, were used. for the purpose of analysing the liquidity positions of portuguese and spanish commercial banks, two liquidity indicators were used for each bank in the sample, based on bank financial ratios. these ratios make it possible to work with a significant number of variables such as loans, deposits and their modifications, and relate them to their respective assets. specifically, we strive to answer two crucial questions: q1: "the liquidity positions of the portuguese and spanish banking system were similar in the period under review?"; q2: "do banks liquidity positions differ from the size of banks?". the answer to these questions allows us to validate the degree of liquidity of two banking systems and to identify the divergences between bank liquidity and the different sizes of banks. it is expected that this study will provide strong contributions to research in this area: firstly, to contribute to combining the lessons from historical readings on banks' liquidity management with the liquidity-focused approach to their liquidity positions for the extended period from 2002 to 2015; secondly, to contribute to the analysis of the liquidity behaviour of banking institutions in two different financial systems, thus making it possible to portray the reality of portuguese and spanish banking; thirdly, this work allows us to analyse the problem of bank liquidity, since the recent first banking crisis of the 21st century cannot be ignored, especially as it revealed a lack of marco amaral / european journal of government and economics 10(1), june 2021, 46-64 48 liquidity in banks; and finally, fourthly, this study is relevant for the academic world, but it is also interesting for banking institutions, investors and regulators/supervisors of the banking system, as the results obtained in this study help refine and improve standards and rules on banks' liquidity management. the final document of the research work is structured as follows: in addition to the first point of introductory, in point 2, the literature review is presented which offers, in a brief way, a review of the existing literature in this area, where the concepts of bank liquidity and liquidity risk are highlighted. then, in section 3, the methods and data of the sample are presented where the methodology adopted in the study is described, presenting, the liquidity indices used and the specification of the model, and in section 4, the results and discussion of the descriptive statistics data are presented, throughout the period of analysis of the liquidity indices of the banks of each country, as well as the discussion of the results obtained. finally, the main conclusions of this study are presented. 2. literature review bank liquidity has generated much interest in the research community. bank liquidity is the ability of a bank to meet the cash demands of bank customers and honour its commitments to third parties as they fall due, thus comprising the balance between the conversion of assets and the callability of liabilities. the basel committee on banking supervision (bcbs, 2008) defines liquidity as the bank's ability to finance asset growth and meet its payment obligations within the contractual term without incurring significant losses. in the same line of convergence, (goodhart, 2008) states that bank liquidity has two facets: the first, involves maturity transformation, i.e. the relative maturity of a bank's liabilities and assets; and the second, involves the inherent liquidity of a bank's assets, i.e. the degree to which these assets can be sold without significant loss of value under any market conditions. the authors bryant (1980), diamond and dybvig (1983) and deyoung and jang (2016), state that the mismatch of size and maturity between assets and liabilities, both on and off the balance sheet, constitutes a structural risk of banks, thus allowing the creation of liquidity risk. indeed, an important role of banks in the economy is to provide liquidity by funding long-term illiquid assets with short-term liquid liabilities. therefore, banks by transforming these maturities face a liquidity risk if any liabilities invested in illiquid assets are called for in the short term (distinguin et al., 2013). given its nature, for author nikolaou (2009), it is possible to distinguish the term liquidity risk into three types of risk: central bank liquidity risk; funding liquidity risk and market or asset liquidity risk. central bank liquidity risk is non-existent, since the central bank is always able to supply the monetary base and therefore can never be illiquid. marco amaral / european journal of government and economics 10(1), june 2021, 46-64 49 table 1. summary of the main literature review studies. author (year) period of study country definition of dimension indicator conclusion berger and bouwman (2009) 1993-2003 usa commercial banks natural logarithm of the value of gross asset. the demand for higher capital requirements, implies associated reduced liquidity for small banks but not for large banks. cornett et al. (2011) 2006-2009 usa commercial banks natural logarithm of the previous year's asset value. as loans grew faster at the big banks, it eventually had a reflection on the effect of the liquidity crisis on the bigger banks. bonfim and kim (2012) 2002-2009 commercial banks from 43 countries natural logarithm of the asset value. excessive risk-taking in large banks implies associated reduced liquidity, as larger banks are more likely to be bailed out in cases of bankruptcy proceedings. cetorelli and goldberg (2012) 2006-2010 usa banks and foreign subsidiaries sufficient liquidity throughout the organisation, i.e. funds flow regularly between banks and subsidiaries on different foreign markets. horváth et al. (2012) 2000-2010 czech banks natural logarithm of the asset value. strong liquidity expansion until the financial crisis driven by the big banks. vodová (2012) 2001-2010 czech and slovak commercial banks total assets. due to increased lending activity, czech and slovak banks have become less liquid. bord and santos (2014) 2005-2007 usa banks natural logarithm of the asset value. banks with greater liquidity difficulties charge higher rates on loans to companies. acharya and mora (2015) 1994-2009 usa banks the study reveals that unlike countries in europe (uk), bank liquidity management in the us system has declined. khan et al. (2017) 1986-2014 usa banks natural logarithm of the asset value. larger banks take less liquidity risk compared to smaller banks when they have more deposits. lastuvková (2017) 2001-2013 banks of slovenia total assets. with bank size activity decreases; bank creates liquidity or uses it mogro and bravo (2018) 2000-2015 banks of ecuador total assets. small banks do not operate at a disadvantage compared to large banks. bonfim and kim (2019) 2002-2009 commercial banks from 45 countries natural logarithm of the asset value. larger banks show weaker liquidity indicators, for the reason that they hold less liquid assets as they can more easily access markets. al-naimi (2020) 2004-2015 commercial banks of jordan natural logarithm of the asset value. larger banks tend to have more operations than just their core banking business, and as such perform more efficient liquidity management and therefore require lower liquidity levels. hakimi et al. (2020) 2014-2015 commercial banks in the mena region natural logarithm of the asset value. the study shows that liquidity risk increases when the size of the bank also increases. kapoor and peia (2021) 2014-2018 usa banks natural logarithm of the asset value. banks with a higher share of mortgage-backed securities in total assets benefit more in creating bank liquidity. source: own elaboration. since the central bank is the monopoly supplier of liquidity, i.e. the originator of the monetary base, it can dispense liquidity as and when needed, in order to meet the equilibrium demand for liquidity in the banking system and thus avoid cases of excess or deficit liquidity according to its policy. the funding liquidity risk is the risk that the bank will not be able to meet its obligations when they fall due in a given period of time, without affecting daily operations or the financial condition of the institution. this risk translates, in practice, into situations in which cash flows (current and future cash inflows and outflows) that are not offset (brunnermeier and pedersen, 2009 and wójcik-mazur and szajt, 2015). the market or asset liquidity risk results from the lack of opportunities to redeem or sell an asset immediately, i.e. it prevents the bank from recovering, marco amaral / european journal of government and economics 10(1), june 2021, 46-64 50 at a given moment, part or all of the asset applied, and sometimes this recovery is only possible by incurring price penalties or high redemption costs. therefore, exposure to this type of risk translates into the reduced or absent possibility of selling liquid assets at a fair price or at a similar level (tirole, 2011). despite the fact that much research has been conducted in the field of performance and adequacy of risks taken by commercial banks on bank liquidity management, there are only a few relevant studies and their results show that recently bank liquidity management has declined sharply in recent years, essentially in the period of the financial crisis (crockett, 2008; cornett et al., 2011; strahan, 2012; vodová, 2012; acharya and mora, 2015 and lastuvková, 2017). these results, however, are divergent with the study of authors bonfim and kim (2019) who show that banks continued to increase liquidity even during the global financial crisis, as well as the study of authors cetorelli and goldberg (2012) who show that funds flowed regularly between banks' global activities. other authors, such as berger and bouwman (2009), bonfim and kim (2012), horváth et al., (2012), bord and santos (2014), khan et al., (2017), mogro and bravo (2018), have evaluated in their studies the impact of bank size on bank liquidity analysis and concluded that some questions still remain open since different results were obtained (table 1 summary of the main literature review studies). 3. methodology and data this section is composed, firstly, of the analysis and description of the data collected in the statistical publications of the banking sector entities, as well as in the sample considered and, secondly, of the methodology adopted to determine bank liquidity. 3.1. data for data collection in this study, we used the content analysis of the statistical publications of apb portuguese banking association, aeb spanish banking association and ceca spanish savings banks and also of the mandatory publications of bdp bank of portugal and bde bank of spain. to this end, we used the panel data technique (stata 13) that combines cross-section (banks) and time-series (years) data, obtaining an unbalanced panel data. in the 14 years (2002 to 2015) under analysis, there are banks that have ceased their activity, others that have started up during this time span and still others that have merged or completed acquisitions that have transformed them into distinct institutions. the sample considered is composed of 105 commercial banks, of which 22 are portuguese and 83 are spanish (27 from aeb and 56 from ceca), having been analysed in this study the banks that provided the individual annual accounts during the period under analysis, i.e., 2002 to 2015. the total number of observations in the sample amounted to 2.048, for the two liquidity ratios under study, of which 489 observations concern the portuguese banking system and 1.559 marco amaral / european journal of government and economics 10(1), june 2021, 46-64 51 observations concern the spanish banking system. in order to demonstrate the size of the two banking systems under study, a characterisation is made of both systems (table 2) consisting of banks operating in portugal and spain, whose volume of net assets as of 31 december 2015 amounted to: 407.6 thousand million euros, having been represented in the sample 85.4% (347.9 thousand million euros) for all the aggregate net assets of the portuguese banking system; and 2.6 billion euros, having been represented in the sample 85.8% (2.2 billion euros) for all the aggregate net assets of the spanish banking system. table 2. characterisation of the sample in the banking system in portugal and spain year 2015. amounts in millions of euros, except where expressly indicated. source: own elaboration. 3.2. methodology for the methodology adopted in order to calculate bank liquidity, descriptive statistics of two liquidity indices (table 3) were carried out for each commercial bank in the sample in both countries and subsequently, the relationship between bank liquidity and bank size were also assessed. table 3. bank liquidity ratios. ratios form of determination liquidity ratio (1) cldc net loans to customer / customer deposits liquidity ratio (2) clal net loans to customer / total net assets source: own elaboration. • liquidity ratio (1) – loan to deposit ratio value % value % 1. cgd caixa geral de depósitos 90 003 22,1 1. santander banco santander 496 322 18,9 2. mbcp millennium banco comercial português 54 151 13,3 2. bbva banco bilbao viscaya argentaria 397 303 15,2 3. nb novo banco 51 276 12,6 3. la caixa caixabank 317 757 12,1 4. bst banco santander totta 50 232 12,3 4. bankia bankia 208 221 7,9 5. bpi banco português de investimento 32 515 8,0 5. sabadell banco de sabadell 163 630 6,2 6. mg montepio geral 24 335 6,0 6. popular banco popular 148 778 5,7 7. ca crédito agrícola 13 060 3,2 7. bankinter bankinter 58 929 2,2 # oci other credit institutions (7) 32 341 7,9 # oci other credit institutions (31) 459 622 17,5 347 913 85,4 2 250 562 85,8 407 589 100,0 2 622 000 100,0 sample (year 2015) 14 commercial banks sample (year 2015) 38 commercial banks total banking system portugal total banking system spain commercial banks 31.12.2015 commercial banks 31.12.2015 total net assets total net assets (individual accounts) (individual accounts) marco amaral / european journal of government and economics 10(1), june 2021, 46-64 52 in order to carry out the assessment of the liquidity position of portuguese and spanish commercial banks, two liquidity ratios were used, the first one being the following: liquidity ratio (1) cldc = net loans to customer customer deposits a very common indicator, having been adopted in the studies conducted by the authors (vodová, 2012; lópez, 2015; mousa, 2015; trenca, 2015; wójcik-mazur and szajt, 2015; deyoung and jang, 2016 and lastuková, 2017). in the banking sector, it is regularly used by financial system supervisors to measure the liquidity of banks and is called the transformation ratio of customer funds into loans and advances to customers (know as loan to deposit ratio). this indicator measures how much of customers' capital (bank deposits) is tied up in loans and advances to customers of the same banks. in general, this indicator is widely used to assess the liquidity of commercial banks (retail banking), losing much relevance when observed for specialised banking (business area: consumer, automotive, investment, real estate, among others), since its purpose is not to capture bank deposits from customers. for dogan, 2013, this ratio between loans and deposits demonstrates the degree of conversion between the most illiquid assets considered by banks (loans) with the liquid liabilities, namely with the main source of funding of commercial banks (customer funds). the higher this ratio, the more the bank depends on borrowed funds and the less liquid the bank is. it is usually greater than 100%, which shows the bank's lending capacity. conversely, lower values of this ratio mean that loans granted by banks are financed by customer deposits and therefore more liquid is the bank. in recent years the portuguese and spanish banking sector has seen a decrease in the ratio of transformation of deposits into credit, reflecting the application of economic and financial adjustments resulting from the financial crisis period, as shown in figures 1 and 2. figure 1. loan to deposit ratio of the banking sector in portugal. source: apb portuguese banking association overview of the portuguese banking system, may 2016. marco amaral / european journal of government and economics 10(1), june 2021, 46-64 53 figure 2. loan to deposit ratio of the banking sector in spain. source: aeb spanish banking association presentation of the 2016 banking year, april 2017. the effect of the global financial crisis, subsequently aggravated from 2010 by the emergence of the sovereign debt crisis "euro crisis", led to a series of economic and financial assistance programmes (efap) to countries in the eurozone, the so-called "european bailouts" and which also became known as piigs countries (portugal, ireland, italy, greece and spain). in the case of portugal and spain, the redemption requests occurred in may 2011 and june 2012, respectively. in this context, the european stability mechanism (esm) played an important role, requiring a thorough assessment of the banks' credit portfolios and financial statements. in this way, the eurogroup in both financial systems approved an agreement, reflected in the memorandum of understanding (mou), which conditioned the financial assistance to compliance with certain measures reinforcing the financial stability of both countries. the main conditioning factors for the banking sector were the requirements to reinforce bank capital, review provisions, transfer assets and the strong need to control bank liquidity. as such, the regulatory authorities, taking into consideration that the banks were excessively leveraged (figures 1 and 2), demanded a sharp reduction, which put a “brake” on lending and launched a “war” for deposits with the corresponding increase in deposit interest rates. as observed in the portuguese case (figure 1), bank liquidity decreased sharply since the start of the agreement (mou) from 2011 to 2015 went from 140.2% to 102.4% (-37.8 p.p.). in turn, the spanish case (figure 2), also decreased bank liquidity (less markedly than in the portuguese case) since it made the agreement (mou) from 2012 to 2016 went from 123% to 110% (-13 p.p.). the restrictions imposed on portuguese and spanish banks as a result of the financial aid from the eurozone, had the practical consequence of improving the liquidity indicators of the banks, enabling greater financial freedom for the banking institutions in subsequent years. • liquidity ratio (2) – loan participation the second liquidity indicator analysed in this study was as follows: liquidity ratio (2) clal = net loans to customer total net assets marco amaral / european journal of government and economics 10(1), june 2021, 46-64 54 the studies conducted by the authors vodová (2011), munteanu (2012), vodová (2012), roman and sargu (2014), lópez (2015), moussa (2015), roman and sargu (2015) and deyoung and jang (2016) used the indicator that measures the ratio between net loans to customers and total net assets. indicator called loan participation that allows determining the weight of the loan portfolio in the total assets of banks. this indicator is relevant for the assessment of banks' liquidity, since loans are the most representative assets on commercial banks' balance sheets, as well as the most illiquid considering their maturity periods (medium and long term). therefore, the higher the ratio of this indicator, the less liquid is the bank. for a better understanding of the participation of loans and advances to customers in total bank assets, we present table 4 and figure 4 in which we can see that the portuguese and spanish banking systems have similar weights of credit portfolios in relation to their assets of around 60.0%. table 4. loan participation to the banking sector in portugal. 2014 2015 2016 2017 average loans to customers total (million euros) annual rate of change as a % of total assets 220.218 58,9% 211.968 -3,7% 59,5% 208.502 -1,6% 61,9% 203.436 -2,4% 60,6% -2,6% 60,2% source: apb portuguese banking association annual bulletin, 2017. figure 4. loan participation to the banking sector in spain. source: aeb spanish banking association presentation of the 2016 banking year, april 2017. • relationship between liquidity and bank size model specification in order to capture the relationship between bank liquidity and the size of commercial banks in portugal and spain over the period from 2002 to 2015 (14 years), we used the simple linear regression model analysis to assess whether the bank size variable has explanatory power over the two liquidity variables (table 5). in the econometric estimation of the model, the panel data (stata 13.1 statistics data analysis) technique is used, which combines cross-section (banks) loans to customers loans to customers as a % total assets marco amaral / european journal of government and economics 10(1), june 2021, 46-64 55 and time-series (years) data, according to the studies of authors reviewed in the literature, such as bonfim and kim (2012), cetorelli and goldberg (2012), bord and santos (2014), acharya and mora (2015), mogro and bravo (2018), bonfim and kim (2019), al-naime (2020) and hakimi et al. (2020). table 5. linear regression models. liq(1)it = β0 + lnβ1dimit + γ1pmgit + ∑ 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴14 𝑡𝑡=1 + εit liq(2)it = β0 + lnβ1dimit + γ1pmgit + ∑ 𝐴𝐴𝐴𝐴𝐴𝐴𝐴𝐴14 𝑡𝑡=1 + εit where, — liq(1)it dependent variable representing the transformation ratio of deposits into credit of bank i at moment t; — liq(2)it dependent variable representing the loan participation ratio of bank i at time t; — dimit independent variable representing the (natural) logarithm of value of net asset bank i at time t; — pmgit control variable representing the criterion for characterising the size of bank i at time t; — β0 is the constant term; — εit is the statistical error term of bank i at time t. source: own elaboration. 4. results the results of this study include the descriptive statistics of two bank liquidity indices for each bank in the sample in both countries, as previously mentioned in section 3, and also the analysis of the relationship between bank liquidity and the size of commercial banks. 4.1. descriptive statistics of bank liquidity ratios the data of the descriptive statistics of liquidity measured by the liquidity indicator (1), that is, by the so-called loan to deposit ratio, can be seen in table 6. table 6. descriptive statistics for liquidity indicator (1) – loan to deposit ratio. percentage values, except where expressly indicated. 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 portugal average 107.1 108.7 115.0 125.2 146.0 136.8 135.0 132.6 154.0 148.0 123.1 113.2 104.8 95.9 median 105.8 115.7 111.1 128.3 151.7 132.1 132.1 151.4 148.7 128.2 111.7 111.2 105.8 103.8 s.d. 55.6 57.4 70.9 80.3 73.7 82.4 73.6 64.0 78.0 88.8 66.6 56.1 47.1 40.4 minimum 0.1 0.1 0.2 0.2 16.4 7.0 6.5 12.6 8.9 4.9 3.5 3.7 3.2 3.0 maximum 198.9 232.4 310.7 330.5 285.6 295.0 273.6 239.9 289.6 360.6 274.7 250.1 198.8 175.5 nº obs. 18 19 20 19 17 19 18 18 17 17 15 15 15 14 ∑ 241 spain average 91.4 96.5 101.1 108.7 111.5 114.1 110.4 109.1 108.8 114.8 102.9 88.0 84.1 81.3 median 88.0 93.5 95.3 100.1 105.7 108.9 104.9 100.6 100.5 105.0 95.4 78.9 75.4 75.4 s.d. 24.9 28.3 30.3 28.6 26.6 33.1 33.4 36.1 39.4 49.4 39.7 35.7 38.8 36.6 minimum 8.1 10.0 3.5 4.3 3.0 4.0 5.3 2.9 9.8 7.7 12.9 13.1 1.1 0.4 maximum 164.6 203.1 227.2 217.4 185.9 240.2 264.1 239.4 291.4 296.2 205.3 184.7 178.1 158.4 nº obs. 72 71 71 70 70 70 66 65 56 41 35 32 30 29 ∑ 778 source: own elaboration. marco amaral / european journal of government and economics 10(1), june 2021, 46-64 56 as previously mentioned, the higher the ratio between credit and deposits, the less liquid is the bank. thus, observing the evolution of liquidity in the two banking systems confirms that the liquidity of portuguese and spanish commercial banks is improving, i.e. increasing. in the case of spain's commercial banks, between 2002 and 2007 there is a slight upward fluctuation in this indicator (in 2002, it rose on average from 91.4% to 114.1% in 2007). however, its liquidity has improved, mainly in the last four years (it rose on average from 114.8% in 2011 to 81.3% in 2015). in turn, the liquidity of banks in portugal varied more markedly (it rose on average from 107.1% in 2002 to 146.0% in 2006). as in the spanish banking market, bank liquidity in portugal increased between 2010 and 2015, but also much more sharply than in neighbouring spain, from an average liquidity of 154.0% in 2010 to 95.9% in 2015. finally, both countries experienced a very slight increase in liquidity during the financial crisis, namely between 2007 and 2009. in the case of spain, the average rose from 114.1% in 2007 to 109.1% in 2009, and in the case of portugal, the average rose from 136.8% in 2007 to 132.6% in 2009. these results are convergent with the studies of strahan, 2012, vodová, 2012, acharya and mora, 2015 and lastuvková, 2017, which find that bank liquidity has been decreasing, essentially until the period of the financial crisis. in the current study, it is observed that both portuguese and spanish commercial banks their liquidity decreases sharply until the years 2006 and 2007, improving substantially their liquidity after the years 2010 and 2011. as regards the descriptive statistics for the relationship measured by the liquidity indicator (2), i.e. the so-called participation of loan to total bank assets, the respective values can be seen in table 7. table 7. descriptive statistics for the liquidity indicator (2) – loans participation. percentage values, except where expressly indicated. 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 portugal average 56.7 60.4 60.0 58.7 59.1 60.8 60.9 55.5 57.5 56.6 55.0 55.1 53.6 48.7 median 67.9 72.5 71.0 68.8 69.3 69.5 68.7 61.3 61.8 61.5 58.6 55.9 57.2 58.0 s.d. 27.3 26.9 27.7 28.2 27.8 26.5 26.4 23.4 22.5 22.3 19.6 19.0 21.5 21.4 minimium 0.0 0.1 0.1 0.1 3.0 5.2 5.6 8.9 8.0 4.3 3.3 3.5 3.0 2.7 maximum 85.9 83.7 89.4 84.6 85.9 86.4 89.1 83.3 88.6 90.5 89.9 93.0 89.4 70.9 nº obs. 18 19 20 19 19 19 18 18 17 18 16 16 16 15 ∑ 248 spain average 66.5 68.9 71.0 72.3 74.1 74.5 72.8 70.3 70.3 65.1 58.5 56.7 52.7 52.4 median 67.4 70.0 73.6 75.3 76.5 77.2 75.9 72.2 73.5 71.9 61.3 47.3 55.2 55.7 s.d. 15.5 15.5 15.4 15.4 14.6 15.0 14.5 14.1 14.8 21.6 19.7 20.2 21.9 21.3 minimium 2.4 3.1 1.2 1.2 0.7 1.2 1.6 2.4 2.1 0.6 2.5 2.6 0.9 0.3 maximum 90.6 92.2 93.0 94.1 94.0 93.9 87.7 86.2 90.6 90.2 89.3 93.2 88.8 80.8 nº obs. 72 71 71 71 71 70 66 65 56 42 35 32 30 29 ∑ 781 source: own elaboration. like the liquidity indicator (1), a high value of the liquidity indicator ratio (2) means reduced bank liquidity. thus, we can state that the increase in lending activities confirms that portuguese and spanish banks have become less liquid. thus, it can be seen in the case of banks in spain that from 2007 to 2015, lending to customers has reduced substantially, falling on average from 74.5% to 52.4%, respectively. in the case of the banking sector in portugal, the liquidity indicator (2) has been less fluctuating, having only registered a decrease from 55.1% in 2013 to 48.7% in 2015, also revealing an improvement in bank liquidity. in relation to the period of the financial marco amaral / european journal of government and economics 10(1), june 2021, 46-64 57 crisis (2007-2009) in both countries, there was a slight decrease in this indicator, which also showed that bank liquidity increased during this period. in spain, it rose from an average of 74.5% in 2007 to 70.3% in 2009, while in portugal it rose from an average of 60.8% in 2007 to 55.5% in 2009. these results support the findings of authors cornett et al., 2011, strahan, 2012, vodová, 2012, acharya and mora, 2015 and lastuvková, 2017, who observe reduced liquidity until the period of the financial crisis, in contrast to the study of authors bonfim and kim (2019) who show that banks increased their liquidity during the period of the financial crisis. 4.2. liquidity ratios by bank size we now move on to the analysis of the two liquidity ratios associated with the size of commercial banks, i.e. the relationship between bank size and liquidity. firstly, a comparative analysis is carried out between portugal and spain of the two bank liquidity ratios by size (small, medium and large) bank, through the graphical representation of the average of the respective liquidity ratios over the period under analysis. secondly, the functional relationship between the two variables on bank liquidity and bank size, measured by the natural logarithm of net asset value, was analysed. to this end, the statistical technique of simple linear regression was adopted in order to model the relationship between bank liquidity and bank size. • graphical analysis of the evolution of bank liquidity ratios taking into account that the two banking systems under analysis present a very different size in terms of the values of aggregate assets of each banking system, we have also adopted different criteria to characterise the size of banks. thus, the criteria adopted to classify the size of banks in the portuguese banking system was as follows: large banks are those with total assets exceeding eur 25 thousand million; medium-sized banks have total assets between eur 2.5 thousand million and eur 25 thousand million; and small banks are those with total assets below eur 2.5 thousand million. the criteria adopted to classify the size of banks in the spanish banking system was the following: large banks are those with total assets over 100 thousand million euros; medium-sized banks have total assets between 10 thousand million and 100 thousand million euros; and small banks are those with total assets under 10 thousand million euros. the characterisation of the size of banks by the two banking systems under analysis in this study is as follows, according to table 8. table 8. characterization of the size of banks in portugal and spain. dimension n.º banks % criterion n.º banks % criterion large 5 22,7 > 25 thousand million eur 8 9,6 > 100 thousand million eur medium 12 54,6 between 2.5 and 25 thousand million eur 43 51,8 between 10 and 100 thousand million eur small 5 22,7 < 2,5 thousand million eur 32 38,6 < 10 thousand million eur 22 100,0 83 100,0 source: own elaboration. marco amaral / european journal of government and economics 10(1), june 2021, 46-64 58 as we can see in figure 5, the liquidity of portuguese banks measured by the ratio between credit and deposits decreases with bank size, i.e. small banks are the most liquid. in turn, the liquidity of medium and large portuguese banks, are less liquid, being almost always above the banking sector average. the only difference is that the liquidity of medium-sized banks is above the liquidity of large banks, i.e. medium-sized banks are the least liquid in the system. this result is in line with the authors vodová, 2011, bonfim and kim, 2012, bonfim and kim, 2019 and hakimi et al., 2020 who concluded that larger banks have reduced liquidity. in turn, it is divergent from the study of the authors horváth et al., 2012, who concluded in the case of czech banks, that large banks were more liquid. figure 5. ind. liq. (1) – loan to deposit ratio by bank size in portugal. source: own elaboration. figure 6 shows the liquidity of spanish banks also measured by the ratio between credit and deposits and we can see that the liquidity of small and medium-sized banks goes hand in hand with the average for banks, with only small banks being more liquid from 2011 onwards. unlike the portuguese banking system, the large spanish commercial banks are the least liquid of the system. the fact that large and medium-sized banks in both the portuguese and spanish banking system are the least liquid results essentially from the banks' preference for strategies related to balance sheet liabilities, i.e. large and medium-sized banks hold fewer liquid assets and rely more on the interbank market and other sources of funding other than customer resources. this result is similar to the studies of authors vodová, 2011, bonfim and kim, 2012, bonfim and kim, 2019 and hakimi et al., 2020. 0 20 40 60 80 100 120 140 160 180 200 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 average small size bank medium size bank large size bank marco amaral / european journal of government and economics 10(1), june 2021, 46-64 59 figure 6 ind. liq. (1) – loan to deposit ratio by bank size in spain. source: own elaboration. the results of the liquidity indicator relationship relative to the loans participation, as shown in figure 7, reveal that in the case of the portuguese banking system, large and medium-sized commercial banks are more willing to lend and are therefore less liquid. only small banks have a lending activity below the banking sector average, with medium and large banks being slightly above the sector average. this result is in line with the studies of authors cornett et al., 2011 and vodová, 2012. figure 7. ind. liq. (2) – loans participation by bank size in portugal. source: own elaboration. finally, figure 8 shows the spanish banking system's banking liquidity measured by the loans participation. the results obtained are quite surprising, since they reveal that small banks were more willing to lend until 2010 and medium-sized banks are the ones that have a lending activity that is always above the average for the sector, and are therefore less liquid. the situation observed here in small banks is indicative of the strong bank restructuring (know as orderly bank settlement fund) that has taken place in the spanish banking system since 2009, which has 60 80 100 120 140 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 average small size bank medium size bank large size bank 0 10 20 30 40 50 60 70 80 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 average small size bank medium size bank large size bank marco amaral / european journal of government and economics 10(1), june 2021, 46-64 60 resulted in the intervention of a large number of small banks and savings banks. thus, we can conclude that small spanish banks lent heavily until 2010, subsequently experiencing liquidity difficulties. this result corroborates the findings of the studies by authors khan et al., 2017 and al-naimi, 2020. figure 8. ind. liq. (2) loans participation by bank size in spain. source: own elaboration. • econometric results for the estimation of the econometric results the simple linear regression model was used as a way to model the existing relationship between the variables. to this end, the panel data resource was used, through the fixed effects model of the ordinary least squares (ols) and also through the random effects model of the generalised least squares (gls). from the analysis carried out, it was found that the model with the highest quality is that which is used through the random effects model. the results obtained from the regressions indicate that there is a low level of explanation of the size variable for the two bank liquidity indices used in this study, it should be noted that, in both the portuguese and spanish cases, the adjusted r² is often less than 10%. thus, it is understood that the models applied do not have an acceptable quality of adjustment. these results are explained by the fact that the study only considers the size as a determining factor, a situation for which there are still other important factors that were not considered in the model and that influence bank liquidity, i.e., more than 90% are factors that are found in the statistical error term (ε). with regard to the explanatory variable used in the two models used, the results obtained recognise statistical significance in explaining bank liquidity for the two banking systems analysed, according to table 9. 40 45 50 55 60 65 70 75 80 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 average small size bank medium size bank large size bank marco amaral / european journal of government and economics 10(1), june 2021, 46-64 61 table 9. regressions results of bank estimates in portugal and spain. source: own elaboration. in the case of the portuguese banking sector, the results allow for the conclusion that only the medium-sized banks for variable liq1 and the large-sized banks for variable liq2 are statistically significant, for a significance level of 10% (sig = 0.086) and 1% (sig = 0.003), respectively. in turn, in the case of the spanish banking sector, it can be concluded that for small, medium and large banks, the variables (liq1 and liq2) are statistically significant, with the exception of the liq2 variable of large banks. 5. conclusions the objective of this study was, on the one hand, to comprehensively analyse the liquidity positions of portuguese and spanish commercial banks through two very common liquidity indices of the banking sector in the period between 2002 and 2015 and on the other hand, to observe whether liquidity is different from bank size. once the analysis is complete, the questions of the study can be answered: q1: "the liquidity positions of the portuguese and spanish banking system were similar in the period under review?"; q2: "do banks liquidity positions differ from the size of banks?". the results obtained show that after the period of the financial crisis, banking institutions show a marked improvement in liquidity in both countries. spanish banks have sharply increased their variable portugal liquidity ratio (1) – liq1 (loan to deposit) (random effects) liquidity ratio (2) – liq2 (loan participation) (random effects) small medium large small medium large coef sig coef sig coef sig coef sig coef sig coef sig dim 10.79 0.388 -36.67 0.086 (c) -22.51 0,267 -8.10 0.400 6.80 0.158 -13.80 0,003 (a) _cons -5.44 0.944 395.30 0.007 312.12 0.047 82.12 0.162 20.09 0.541 169.0 0.000 n,º of observations: 54 119 68 56 124 68 241 248 ⅀ = 489 n.º of banks: 5 12 5 5 12 5 22 22 r-sq: 0.085 0.000 0.077 0.020 0.024 0.062 variable spain liquidity ratio (1) – liq1 (loan to deposit) (random effects) liquidity ratio (2) – liq2 (loan participation) (rndom effects) small medium large small medium large coef sig coef sig coef sig coef sig coef sig coef sig dim 41,34 0.000 (a) 30.20 0.000 (a) 14.38 0.028 (b) 5.34 0.096 (c) 8.73 0.000 (a) 3.26 0.195 _cons -165.51 0.000 -118.95 0.003 -1.65 0.975 33.84 20.921 6.21 0.722 33.56 0.108 nº of observations: 290 393 95 292 394 95 778 781 ⅀ = 1.559 n.º of banks: 32 43 8 32 43 8 83 83 r-sq: 0.126 0.024 0.022 0.087 0.001 0.257 note: (a), (b), (c) statistically significant results for a significance level of 0.01, 0.05 and 0.10 respectively. marco amaral / european journal of government and economics 10(1), june 2021, 46-64 62 liquidity in the last five years, while portuguese banks have varied their liquidity slightly negatively during the period 2002-2006, increasing liquidity sharply with the worsening of economic conditions from 2010 onwards. thus, we can state that for the first question of the study the management strategy of liquidity positions of portuguese and spanish commercial banks was broadly similar. additionally, it is observed that smaller banks have a less fluctuating liquidity management and therefore small banks are more liquid, however, in the case of the spanish banking system, small banks until 2010 were more willing to grant credit and as such had major bank liquidity problems. thus, we can state that regarding the second question of this study, the liquidity positions of banks can be different depending on the size of the bank, since in the case of the spanish banking system, in a certain period (between 2002 and 2010) small banks were less liquid than medium and large spanish banks due to the exacerbated lending. thus, we conclude that an increasing volume of loans also leads to a decrease in bank liquidity, being in the case of spain, more affected by small banks. on the other hand, we can observe that in general the large banks are the least liquid. this finding is associated with the fact that large banks are "too big to fail", which motivates these banks to have reduced liquidity, because in the event of a liquidity shortage, these banks rely on the liquidity assistance of lender of last resort. finally, we find that medium and large banks are statistically significant in the case of the portuguese banking sector, while in the case of the spanish banking sector, all banks, regardless of their size, influence bank liquidity. the study carried out has, however, some limitations. firstly, to measure the bank liquidity ratio, only two liquidity indices were used (loan to deposit ratio and loan participation), which are regularly applied by financial system supervisors. however, the new banking liquidity indicators, liquidity coverage ratio (lcr) and net stable funding ratio (nsfr) were not applied in this study. secondly, this current study was limited to commercial banks only. third, the period of analysis, although long, from 2002 to 2015 (14 years), could be further extended by integrating more recent periods. thus, the empirical conclusions of this article should be analysed taking into account all these limitations. thus, it is suggested, for future research, to apply other bank liquidity indices, introduce other types of banks and extend the period of analysis because they will certainly improve the results of this article. aknowledgements the author is grateful to the anonymous referees of the journal for their extremely useful suggestions to improve the quality of the article. marco amaral / european journal of government and economics 10(1), june 2021, 46-64 63 references acharya, v., mora, n. (2015). a crisis of banks as liquidity providers. the journal of finance, 70(1), pp. 1-43. https://doi.org/10.1111/jofi.12182 al-naimi, a.a., (2020). liquidity risks models in jordanian commercial banks. paideuma journal, xiii(iv), 58-83. bcbs (basel committee on banking supervision), (2008). principles for sound liquidity risk management and supervision. bis – bank for international settlements, september. bcbs (basel committee on banking supervision), (2013). basel iii: the liquidity coverage ratio and liquidity risk monitoring. bis – bank for international settlements, january. bcbs (basel committee on banking supervision), (2014). basel iii: the net stable funding ratio. bis – bank for international settlements, october. berger, a.n., bouwman, c.h.s. (2009). bank liquidity creation. the review of financial studies, 22(9), 3780-3837. https://doi.org/10.1093/rfs/hhn104 bonfim, d., kim, m. (2012). risco de liquidez sistémico. banco de portugal relatório de estabilidade financeira, 79-99. bonfim, d., kim, m. (2019). liquidity risk and collective moral hazard. international journal of central banking, 15(2), 101-150. bord, v., santos, j. (2014). banks liquidity and the cost of liquidity to corporations. journal of money, credit and banking, 46(1), 13-45. https://doi.org/10.1111/jmcb.12076 brunnermeier, m., pedersen, l. (2009). market liquidity and funding liquidity. the review of financial studies from oxford journals, 22(6), 2201-2238. https://doi.org/10.1093/rfs/hhn098 bryant, j. (1980). a model of reserves, bank runs, and deposit insurance. journal of banking and finance, 4, 335-344. https://doi.org/10.1016/0378-4266(80)90012-6 cetorelli, n., goldberg, l. (2012). liquidity management of u.s. global banks: internal capital markets in the great recession. journal of international economics, 88, 299-311. https://doi.org/10.1016/j.jinteco.2012.05.001 cornett, m.m., mcnutt, j.j., strahan, p.e., tehranian, h. (2011). liquidity risk management and credit supply in the financial crisis. journal of financial economics, 101, 297-312. https://doi.org/10.1016/j.jfineco.2011.03.001 crockett, a. (2008). market liquidity and financial stability. financial stability review from banque de france, 11 special, 13-17. deyoung, r., jang, y. (2016). do banques actively manage their liquidity? journal of banking and finance, 66, 143-161. https://doi.org/10.1016/j.jbankfin.2015.11.013 diamond, d.w., dybvig, p.h. (1983). bank runs, deposit insurance, and liquidity. journal of political economy, 91, 401-419. https://doi.org/10.1086/261155 distinguin, i., roulet c., tarazi, a. (2013). bank regulatory capital and liquidity: evidence from u.s. and european publicly traded banks. journal of banking and finance, 37, 3295-3317. https://doi.org/10.1016/j.jbankfin.2013.04.027 dogan, m., (2013). comparison of financial performance of domestic and foreign banks: the case of turkey. international journal of business and social sciences, 4(2), 233-240. european commission, (2020). no 2019/876 of the european parliament and the council as regards the leverage ratio, the net stable funding ratio, requirements for own funds and eligible liabilities, counterparty credit risk, market risk, exposures to central couterparties, exposures to collective investment undertakings, large exposures, reporting and disclosure requirements, and regulation (eu) no 648/2012. regulation (eu). goodhart, c. (2008). liquidity risk management. financial stability review from banque de france, 11 special, 39-44. hakimi, a., boussaada r., hamdi, h. (2020). the interactional relationships between credit risk, liquidity risk and bank profitability in mena region. global business review, 1-23. horváth, r., seidler j., weill, l. (2012). bank capital and liquidity creation. granger-causality evidence. european central bank working paper series nº 1947. kapoor, s., peia, o. (2021). the impact of quantitative easing on liquidity creation. journal of banking and finance, 122, 1-15. https://doi.org/10.1016/j.jbankfin.2020.105998 khan, m., scheule, h., wu, e. (2017). funding liquidity and bank risk taking. journal of banking and finance, 82, 203-216. https://doi.org/10.1016/j.jbankfin.2016.09.005 https://doi.org/10.1111/jofi.12182 https://doi.org/10.1093/rfs/hhn104 https://doi.org/10.1111/jmcb.12076 https://doi.org/10.1093/rfs/hhn098 https://doi.org/10.1016/0378-4266(80)90012-6 https://doi.org/10.1016/j.jinteco.2012.05.001 https://doi.org/10.1016/j.jfineco.2011.03.001 https://doi.org/10.1016/j.jbankfin.2015.11.013 https://doi.org/10.1086/261155 https://doi.org/10.1016/j.jbankfin.2013.04.027 https://doi.org/10.1016/j.jbankfin.2020.105998 https://doi.org/10.1016/j.jbankfin.2016.09.005 marco amaral / european journal of government and economics 10(1), june 2021, 46-64 64 lastuvková, j. (2017). dimensions of liquidity and their factors in the slovenian bank sector. eam: ekonomie a management, 2(20), 163-174. https://doi.org/10.15240/tul/001/2017-2-012 lópez, a. s. (2015). aplicación de las técnicas multivariantes al sector bancario español: el caso de las entidades afectadas por la restructutación. revista de métodos cuantitativos para la economía y la empresa, 19, 66-100. mogro, s. c., bravo, g. a. (2018). assessing competition in the private banking sector in ecuador: an econometric approach with the panzar-rossr model. cuadernos de economía, 41, 225240. https://doi.org/10.32826/cude.v41i117.75 moussa, m. a. b. (2015). the determinants of bank liquidity: case of tunisia. international journal of economics and financial issues, 5(1), 249-259. munteanu, i. (2012). bank liquidity and its determinants in romania. procedia economics and finance, 3, 993-998. https://doi.org/10.1016/s2212-5671(12)00263-8 nikolaou, k. (2009). liquidity (risks) concepts definitions and interactions. european central bank working paper, 1008, pp. 1-71. roman, a., sargu, a. c. (2014). banks liquidity risk analysis in the new european union member countries: evidence from bulgaria and romania. procedia economics and finance, 15, 569576. https://doi.org/10.1016/s2212-5671(14)00512-7 roman, a., sargu, a. c. (2015). the impact of bank-specific factors on the commercial banks liquidity: empirical evidence from cee countries. procedia economics and finance, 20, 571579. https://doi.org/10.1016/s2212-5671(15)00110-0 strahan, p. e. (2012). liquidity risk and credit in the financial crisis. frbsf economic letter, 15. tirole, j. (2011). “iliquidity and all its friends”, journal of economic literature, 49(2), pp. 287-325. https://doi.org/10.1257/jel.49.2.287 trenca, k. y., petria, n., & corovei, e. a. (2015). impact of macroeconomic variables upon the banking system liquidity. procedia economics and finance, 32, 1170-1177. https://doi.org/10.1016/s2212-5671(15)01583-x vodová, p. (2011). determinants of commercial banks liquidity in the czech republic. recent researches in applied and computational mathematics, 92-97. vodová, p. (2012). liquidity of czech and slovak commercial banks. acta universitatis agriculturae et silviculturae mendelianae brunensis, 50(7), 463-476. https://doi.org/10.11118/actaun201260070463 wójcik-mazur, a., szajt, m. (2015). determinants of liquidity risk in commercial banks in the european union. argumenta oeconomica. 2(35), 25-47. https://doi.org/10.15611/ aoe.2015.2.02 https://doi.org/10.15240/tul/001/2017-2-012 https://doi.org/10.32826/cude.v41i117.75 https://doi.org/10.1016/s2212-5671(12)00263-8 https://doi.org/10.1016/s2212-5671(14)00512-7 https://doi.org/10.1016/s2212-5671(15)00110-0 https://doi.org/10.1257/jel.49.2.287 https://doi.org/10.1016/s2212-5671(15)01583-x https://doi.org/10.11118/actaun201260070463 https://doi.org/10.15611/aoe.2015.2.02 https://doi.org/10.15611/aoe.2015.2.02 european journal of government and economics 11(1), june 2022, 31-50 european journal of government and economics issn: 2254-7088 the explanatory power of expenses in the performance of eurozone mutual funds renato correia-domingues a, *, pablo duran-santomil b, luis otero-gonzalez a a polytechnic institute of tomar, portugal b university of santiago of compostela, spain * corresponding author at: renatodomus@hotmail.com abstract. we studied the relationship between the mutual fund’s expenses and their profitability, as well as their ability to predict future performance. we used regressions in panel data with a sample of equity mutual funds of the eurozone area from 2003 to 2014. we concluded that expenses are an important determinant of performance in the eurozone and we show evidence that there is, in the short-term, a negative relationship between the profitability of the year and the level of expenses of the fund. furthermore, there is also a relationship between the profitability of mutual funds with their level of expenses after one and three years. on the other hand, by applying the quantile regression, it seems clear that the funds with the highest returns have a positive relationship with the level of expenses. keywords. mutual funds, expenses, performance, ratings jel codes. g20; g23; g28 doi. https://doi.org/10.17979/ejge.2022.11.1.8767 1. introduction the financial performance of mutual funds has been widely discussed in the literature, and this kind of product has acquired great importance for investors and savers. there is a debate around the management value added mutual funds and expenses, and the funds with the largest expenditures can bring value to the investors/savers, which has led to a debate between active and passive management in the industry. this debate is related to agency theory, since in many funds the returns of investors are against the returns of the managers. authors such as gruber (1996), livingston and o'neal (1996), capon, fitzsimons and prince (1996), carhart (1997), alexander, jones and nigro (1998), sirri and tufano (1998) or chalmers, edelen and kadlec (1999) document the negative relationship of expenses with the performance of funds, verifying those managers rarely manage to beat the market after discounting costs to gross returns. on the other hand, more recent studies support that management expenses may represent a contribution to management if the fund portfolio differs from the market index. studies such as those carried out by cremers and petajisto (2009) or petajisto (2013) conclude that the most active funds (which therefore differ from the market) have outperformed their benchmarks even © 2022. this work is licensed under a cc by-nc 4.0 license. mailto:renatodomus@hotmail.com https://doi.org/10.17979/ejge.2022.11.1.8767 https://creativecommons.org/licenses/by-nc/4.0/ renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 32 after taking costs into account. the objective of this study is to understand the role of expenses on euro zone funds for predicting future performance in the short and long term or if they affect current performance. this study brings several contributions to the literature. firstly, we use a data sample for the eurozone, knowing that the economic area is a fundamental determinant in the performance of the funds (see, for example, ferreira, keswani, miguel and ramos 2012), and since most of the studies have been carried out for the american or british market. secondly, we use a metric widely used in studies of mutual funds, but little used in studies that analyse the role of expenses (4-factor alpha), in addition to other more traditional ones (sharpe, sortino and returns net). in third place, we carry out an analysis for the main categories of mutual funds in the euro zone. many previous studies do not consider category differences which can influence their results and focus their analysis on the american market. fourth, we use different econometric techniques, including quantile regression for the panel data, to show the complex relationship that may exist between expenses and returns. therefore, this study is not only of interest to the scientific community, but also to professionals, investors, researchers, and financial advisers. the work is structured as follows: section 2 will develop the literature review; section 3 will describe the sample and the performance metric’s, section 4 presents the empirical study, and section 5 concludes. 2. literature review 2.1 negative relationship of expenses with performance much of the mutual fund literature focuses on analysing whether fund managers generate returns to beat their benchmark and justify the associated transaction costs and fees (see for example daniel, grinblatt, titman and wermers, 1997; malkiel, 1995). studies such as gruber (1996), livingston and o'neal (1996), capon, fitzsimons and prince (1996), carhart (1997), alexander, jones and nigro (1998), sirri and tufano (1998), chalmers, edelen, and kadlec (1999) among others, documented the negative relationship between mutual fund´s expenditures and their performance. the reasoning is that the return of the index equals the weighted average return of active and passive investors before investment expenses. therefore, active management would be a zero-sum game. sharpe (1991) calls this fact the arithmetic of active management. the author states that active investors cannot exceed the returns obtained from passive investment strategies. sharpe (1966) found that part of the difference in returns to funds could be explained by the difference in spending levels. grinblatt and titman (1992) found that expenses contributed to the difference between funds classified as "winners" or "losers." carhart (1997) showed that expenses can explain the persistence of adjusted risk return and, therefore, may be used as a good predictor of future performance. o'neal (2004), haslem, baker and smith (2008) gil-bazo and ruiz-verdu (2009) also support the negative relationship between performance and expenses. wermers (2000) related funds that performed worse to those with high asset turnover renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 33 and, therefore, high expenses. hereupon, he argues that funds with more expenses perform worse. also, berk and green (2004) found that the alpha performance metric obtained by investors has zero return after eliminating expenses. on the other hand, investors in funds with higher expenses are considered in the literature as unsophisticated, with low sensitivity to low returns (gil-bazo and ruiz-verdu, 2009). philips and kinniry (2010) show that funds with higher expenses have lower returns than those with lower expenses, suggesting passive investing or indexing as an investment strategy. ferruz and alda (2012) carried out a study of expenses for the spanish market, concluding that there is a negative relationship between the commissions charged and the adjusted return risk provided by the pension plans. fernández and órtiz (2015) also attribute to the high costs paid by members the disappointing performance in pension plans. a very important part of the cost of the funds corresponds to their marketing and the distributor receives, in general, remuneration higher than that of the managers. in addition, part of the commission generates conflicts of interest and the incentive to sell the product for higher commissions. nevertheless, several papers have supported a positive relationship between expenses and performance. diaz-mendoza, lopez-espinosa, and martinez-sedano (2012) suggest that fund´s performance is positively related to the magnitude of the performance fees. flood (2015) found that many of the costs are “hidden costs” that can reach 80% of the total costs in the case of actively managed funds. on the other hand, there is some evidence that the best funds would be related to higher expenses. servaes and sirgusson (2018) studied the funds fees of european union, norway and switzerland, finding that mutual funds with higher expenses achieved superior performance. this relationship is justified by kosowski, timmermann, white and wermers (2006) who find that some managers have superior skills and charge higher expenses but obtain better returns (gruber, 1996; livingston and o'neal, 1996; capon, fitzsimons and prince, 1996 and carhart,1997). also, alexander, jones and nigro (1998), sirri and tufano (1998) or chalmers, edelen and kadlec (1999) support the existence of a positive relationship between expenses and fund performance. in this way we formulate our first hypothesis: h1: mutual fund`s expenses are negatively related to their performance. 2.2 the predictive power of expenses in return/adjusted risk return in addition to studying the relationship between expenses at a given time, it is also interesting to analyse whether expenses can predict the future behaviour of funds in terms of risk-adjusted returns. russel kinnel, (morningstar's managing director of research), indicated that the level of expenses is a good predictor of the future returns of mutual funds, so it should be one of the main renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 34 variables to consider by investors1. financial research corporation (2002) analysed the factors that can predict future performance (including, but not limited to, past performance, morningstar ratings, and the expense ratio), concluding that only the expense ratio is can of predicting future performance. this goes against cremers and petajisto (2009) and petajisto (2013), who found that the expense index is not a good variable to predict future profitability as they are the so-called “active share”, the size or past returns. several studies directly attributed the poor performance of funds to increased expenses resulting from management costs, such as berk and green (2004). the result is repeated in the united kingdom (cuthbertson, nitzsche and o'sullivan, 2006) and spain (matallin-saez, soler-dominguez and tortosa-ausina, 2012), reaching conclusions comparable with studies in the united states. based on the previous literature, our second research hypothesis is: h2: mutual fund expenses have the power to predict future performance. 2.3 morningstar ratings another relevant issue is the relationship between the rating or rating of a fund and its level of expenses. investors frequently use ratings to select investment funds (otero and durán, 2021). howe and pope (1996), blake and morey (2000), morey and gottesman (2006), morey (2005), duret et al. (2008), philips and kinniry (2010) and chotivetthamrong (2015) studied ratings and their predictive power, concluding that they have little power to predict future performance. blake and morey (2000) studied the rating obtained by morningstar, showing that a low rating was a good predictor of future bad behaviour, but they no found evidence that funds with good ratings had a better performance than the average. a later work by morey (2005) focused on the funds with the best ratings, found that these had a high probability of being worse in the next three years: the authors think that this is due to inflows in the funds with better ratings, which makes managers unable to adequately accomplish new flows. on the other hand, otero, durán and domingues (2019) found that morningstar's star rating, which is the main rating analysed by the investor, can predict short and medium-term performance and the authors find evidence that the expenses may partly explain the classification. morey and gottesman (2006), müller and weber (2014), meinhardt (2014), antypas, caporale, kourogenis and pittis (2009) reach similar conclusions. otero and duran (2021) find that star ratings can have predictive power. this power can be improved when it is used to complement another´s variables such as costs and analysts’ ratings. 1 study carried out in 2010 and updated in 2016. see: https://www.morningstar.com/articles/347327/how-expense-ratiosand-star-ratings-predict-success and https://www.morningstar.com/articles/752485/fund-fees-predict-future-success-orfailure the other variable that the study considers to have an impact on future performance, to a lesser extent, are sta ratings. https://www.morningstar.com/articles/347327/how-expense-ratios-and-star-ratings-predict-success https://www.morningstar.com/articles/347327/how-expense-ratios-and-star-ratings-predict-success https://www.morningstar.com/articles/752485/fund-fees-predict-future-success-or-failure https://www.morningstar.com/articles/752485/fund-fees-predict-future-success-or-failure renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 35 h3: combining morningstar star ratings with mutual funds expenses help understand the funds that outperform their benchmark. 3. sample and performance metrics the sample is made up of euro area equity mutual funds rated by morningstar with a quantitative star rating, for the period between 2003 and 2014. we limit our sample to funds included in the following morningstar categories: eurozone flex-cap equity, eurozone large-cap equity, eurozone mid-cap equity and eurozone small-cap equity. the funds are of the "open funds" category, that is, those that have a variable number of shares. fund selection was made for all active and inactive funds to avoid the so-called survival bias. in addition, to avoid multicollinearity problems, a single class was selected for each fund, according to the following preferences (ordered from highest to lowest priority): institutional class, lowest administrative expenses, lowest net expenses, higher volume of assets, older start date and preferably accumulation class (vs. distribution classes). our final sample is an unbalanced panel with data from 2003 to 2014 made up of 1,763 observations. table 1 summarizes the number of funds by category and rating, and the mean and standard deviation of the level of expenditures. thus, the sample is made up of 118 funds from the flex cap category of the eurozone, 53 funds from the mid-cap category of the eurozone, 1,117 funds from the large capitalization category of the zone euro and 56 euro zone small cap category funds. these funds are rated according to morningstar star ratings criteria from one to five stars. it is observed that, for the different categories, funds with fewer stars (worst rating) have a higher level of expenditure than those with more stars (better rating), although this behaviour is not uniform. table 1. number of observations by category, rating and level of expenses. flex cap mid cap stars obs. mean s.d. stars obs. mean s.d. 1 23 3.493 1.439 1 1 2.240 2 14 2.905 1.491 2 7 1.953 0.795 3 31 2.662 1.296 3 16 1.511 0.789 4 26 2.440 1.194 4 20 1.342 0.776 5 24 2.370 0.893 5 9 1.711 1.542 large cap small cap stars obs. mean s.d. stars obs. mean s.d. 1 74 2.105 1.089 1 7 1.856 0.733 2 198 1.640 0.839 2 9 1.764 0.710 3 412 1.342 0.793 3 21 1.829 0.985 4 296 1.377 0.940 4 12 2.505 0.966 5 137 1.448 0.984 5 7 1.986 1.107 renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 36 to measure the performance of mutual funds we used several metrics: carhart's four-factor alpha, sharpe ratio, sortino ratio, and annual return. the sharpe ratio allows you to measure the risk-adjusted return by dividing the excess of a fund's return over the risk-free interest rate by the volatility or standard deviation of that return. the sortino ratio uses in the numerator the standard deviation generated by the negative returns of the portfolio. carhart's four-factor alpha allows you to adjust for other factors that skew the fund's performance analyses (market, value, size, and momentum). net returns allows you to analyse the funds from the view of the retail investor, who does not use risk-adjusted metrics. the four-factor model based on carhart (1997) that includes market return (mkt), size (smb), book-to-market (hml) and momentum (mom) permits to control the effect of different styles of investment in fund performance. in this way, the excess return of the fund over the risk-free interest rate is given by: 𝑅𝑅𝑖𝑖,𝑡𝑡−𝑅𝑅𝑓𝑓,𝑡𝑡 = 𝛼𝛼𝑖𝑖,𝑡𝑡 + 𝛽𝛽𝑖𝑖,1𝑀𝑀𝑀𝑀𝑀𝑀𝑡𝑡 + 𝛽𝛽𝑖𝑖,2𝑆𝑆𝑀𝑀𝑆𝑆𝑡𝑡 + 𝛽𝛽𝑖𝑖,3𝐻𝐻𝑀𝑀𝐻𝐻𝑡𝑡 + 𝛽𝛽𝑖𝑖,4𝑀𝑀𝑀𝑀𝑀𝑀𝑡𝑡 + 𝜀𝜀𝑖𝑖,𝑡𝑡 [1] where: ri,t is the return of fund i at the moment t; rf,t is the return on a risk-free asset at time t; mktt; is the monthly return of the market portfolio minus the risk-free rate smbt captures exposure to the returns of diversified portfolios of smalland largecap stocks; hmlt captures exposure to returns between high and low book-to-market portfolios; momt is the momentum; εi,t is the stochastic error term. the β parameters in equation (1) measure the sensitivity of the excess return of fund i to each risk factor, where α is the excess return of the fund. the sharpe ratio metric that is given by: 𝑆𝑆ℎ𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑖𝑖,𝑡𝑡 = (𝑅𝑅𝑖𝑖,𝑡𝑡−𝑅𝑅𝑓𝑓,𝑡𝑡 ) 𝜎𝜎𝑖𝑖,𝑡𝑡 [2] where: 𝑅𝑅𝑖𝑖,𝑡𝑡 represents the profitability of fund i at time t, 𝑅𝑅𝑓𝑓,𝑡𝑡 represents the risk-free rate (1-month euribor) at time t, 𝜎𝜎𝑖𝑖,𝑡𝑡 is the standard deviation of background i at time t, the sortino ratio is a variant of the sharpe ratio that only considers the deviation of declines (negative returns) or downside deviation in the denominator. 𝑆𝑆𝑀𝑀𝑎𝑎𝑀𝑀𝑆𝑆𝑆𝑆𝑀𝑀𝑖𝑖,𝑡𝑡 = (𝑅𝑅𝑖𝑖,𝑡𝑡−𝑅𝑅𝑓𝑓,𝑡𝑡 ) 𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑑𝑖𝑖𝑑𝑑𝑑𝑑𝜎𝜎𝑖𝑖,𝑡𝑡 [3] renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 37 where 𝑑𝑑𝑀𝑀𝑑𝑑𝑆𝑆𝑑𝑑𝑆𝑆𝑑𝑑𝑎𝑎𝜎𝜎𝑖𝑖 ,𝑡𝑡 is the downside deviation finally, within the variables explained, we use a performance metric without adjusting for risk, the annual net return (annual return). the other variables that we will use later as explanatory ones are the fund's expense ratio (net expenses) and the global star rating of the fund awarded by mornigstar. the expense ratio of an investment fund has been calculated with the difference between the fund's gross and net return. morningstar's rating is based on the fund's position within its category based on a risk-adjusted performance measure (the so-called morningstar risk-adjusted return, mrar). the top 10% of funds receive 5 stars, the next 22.5% receive 4 stars, the next 35% 3 stars, then 22.5% 2 stars, and the 10% worst funds 1 star. morningstar establishes its 3, 5 and 10 year ratings for each fund, with the overall rating being a weighted average of the above. finally, as control variables, we used the dummy variable to age (yield per year), dummies of the morningstar fund categories, the years of experience of the fund managers and size. the source of the data is morningstar direct for all variables, except for carhart's alpha, which are based on own calculations from the website of professor kenneth r. french2. to assess the independence degree between the indicators and the other variables, an analysis of the correlations between variables and the calculation of the vif (variance inflation factor) was performed, ruling out the existence of multicollinearity problems between the variables studied. 4. empirical study 4.1. econometric models the first model used was a regression with panel data with random effects. the methodology based on panel data can control individual effects with advantages such as reduction of collinearity and efficiency, among others (baltagi, 2010). armstrong, genc and verbeek (2018) also used the panel data methodology to evaluate the performance of mutual funds using ratings. in this sense, we estimate the following equations: 𝑌𝑌𝑖𝑖𝑡𝑡 = 𝛼𝛼𝑖𝑖 + 𝛽𝛽1 net expenses + 𝛽𝛽2tenure (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽3 𝑓𝑓𝑓𝑓𝑆𝑆𝑑𝑑 𝑦𝑦𝑎𝑎𝑎𝑎𝑎𝑎𝑑𝑑 (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽4 𝑁𝑁𝑎𝑎𝑀𝑀𝑎𝑎𝑑𝑑𝑑𝑑𝑎𝑎𝑀𝑀𝑑𝑑 + 𝛽𝛽𝑖𝑖 ∑ 𝑌𝑌𝑎𝑎𝑎𝑎𝑎𝑎𝑖𝑖𝑖𝑖 + 𝛽𝛽𝑗𝑗 ∑ 𝐶𝐶𝑎𝑎𝑀𝑀𝑎𝑎𝐶𝐶𝑀𝑀𝑎𝑎𝑦𝑦𝑗𝑗𝑗𝑗 + 𝜀𝜀𝑖𝑖𝑡𝑡 [4] 2 see https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html. https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/data_library.html renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 38 where: y_it is the performance measure for fund i at time t. the measures used are the 4-factor alpha (alpha), the sharpe ratio (sharpe), the sortino ratio (sortino), and the net annual return. net expenses is total expenses for fund i. tenure is the average years of management experience of the fund managers. fund years is the age (years) of the fund. 𝑁𝑁𝑎𝑎𝑀𝑀 𝐴𝐴𝑑𝑑𝑑𝑑𝑎𝑎𝑀𝑀𝑑𝑑 is the logarithm of the fund's net assets. year is a dichotomic variable that assume 1 when is the year in cause. categoryj is a dummy variable of the morningstar fund categories. αi and β are the parameters to estimate in the regressions and ε_ (i, t) are the estimation errors. in the second models, the overall morningstar ratings are also entered. in this way, we estimate the following equations: 𝑌𝑌𝑖𝑖𝑡𝑡 = 𝛼𝛼𝑖𝑖 + 𝛽𝛽1𝑁𝑁𝑎𝑎𝑀𝑀 𝑎𝑎𝑒𝑒𝑎𝑎𝑎𝑎𝑆𝑆𝑑𝑑𝑎𝑎𝑑𝑑 (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽2tenure (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽3 (𝑓𝑓𝑓𝑓𝑆𝑆𝑑𝑑 𝑦𝑦𝑎𝑎𝑎𝑎𝑎𝑎𝑑𝑑) (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽4 𝑁𝑁𝑎𝑎𝑀𝑀 𝐴𝐴𝑑𝑑𝑑𝑑𝑎𝑎𝑀𝑀𝑑𝑑 (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽𝑘𝑘 ∑ 𝑅𝑅𝑎𝑎𝑀𝑀𝑆𝑆𝑆𝑆𝐶𝐶 𝑖𝑖 (𝑀𝑀 − 1)𝑖𝑖𝑡𝑡 + 𝛽𝛽𝑖𝑖 ∑ 𝑌𝑌𝑎𝑎𝑎𝑎𝑎𝑎𝑖𝑖 𝑖𝑖 + 𝛽𝛽𝑗𝑗 ∑ 𝐶𝐶𝑎𝑎𝑀𝑀𝑎𝑎𝐶𝐶𝑀𝑀𝑎𝑎𝑦𝑦𝑗𝑗𝑗𝑗 + 𝜀𝜀𝑖𝑖𝑡𝑡 [5] where rating is the morningstar rating being the dummy variable for funds from 1 to 4 stars. based on chen and huang (2011) we also used quantile regression to extend the regression model to conditional quantiles of different performance metrics, because it is more appropriate for a heterogeneous universe of mutual funds, where strategies and objectives may vary. this model makes it possible to capture information on the coefficients in different quantiles of the dependent variable, given the set of endogenous variables (classification by stars). furthermore, the conditional quantile regression developed by koenker and bas-sett (1978) is well suited to distorted distributions of fund returns. in particular, we adopted the method proposed by efron (1979). as a vector of exogenous variables that represent the classification of funds, the quantile model can be written as: 𝑦𝑦𝑖𝑖 = 𝑋𝑋𝑖𝑖´𝛽𝛽𝜙𝜙 + 𝑓𝑓𝜙𝜙𝑖𝑖 [6] where: 𝑄𝑄𝑓𝑓𝑎𝑎𝑆𝑆𝑀𝑀𝜙𝜙(𝑦𝑦𝑖𝑖|𝑋𝑋𝑖𝑖) = 𝑋𝑋𝑖𝑖´𝛽𝛽𝜙𝜙 [7] 𝑄𝑄𝑓𝑓𝑎𝑎𝑆𝑆𝑀𝑀𝜙𝜙(𝑓𝑓𝜙𝜙𝑖𝑖|𝑋𝑋𝑖𝑖)=0 renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 39 4.2. results in table 2 we estimate the dynamic data panel models using regression with random effects. it can be seen a negative and significant relationship between the risk-adjusted returns (sharpe and sortino) and the level of expenses of the funds themselves in the year. there is also a negative, although not significant, relationship with the 4-factor (alpha) adjusted performance measures and the net annual return. in addition, it appears to be a negative and significant relationship between the performance and age of the fund, and a positive relationship with respect to size. annex 1 also analyses the ability of past expenses to predict future performance. there is a negative and significant relationship for risk-adjusted measures. this means that one year's expenses have some power to predict risk-adjusted or non-risk-adjusted performance one year in advance, given that higher expenses funds will underperform significantly in the next year. table 3 shows that when we use a 3-year lag, that is, the expenses of the t-3 period, there is also a negative and significant relationship for all the performance measures used (alpha, sharpe, sortino, and annual return). in this way we obtain evidence of the predictive power of expenses on the future performance of mutual funds at 1 and 3 years. table 2. regression of expenses and performance, with random effects. variable alfa sharpe sortino annual return net expenses -0.0335 -0.0238*** -0.0478*** -0.2040 tenure -0.0059 0.0005 0.0003 0.0021 net assets 0.0051 0.0181*** 0.0262*** 0.0840 fund years -0.0044 -0.0049*** -0.0078*** -0.0563** flexcap 0.4586** 0.0982 0.1908 2.3688* largecap 0.3339* 0.0057 0.0051 0.3598 midcap 0.5914*** 0.2075** 0.3424** 6.6403*** smallcap 0.5986*** 0.1531* 0.2310 5.6062*** _cons 0.1932 1.0635*** 2.0937*** 1.7214 n 1411 1185 1185 1388 r2_w 0.5796 0.9511 0.9153 0.9432 note: alpha represents carhart factor adjusted return; sharpe and sortino are risk-adjusted profitability measures; annual return is the net annual return; net expenses are the expenses of the funds for year n; flexcap, largecap, midcap and smallcap are control variables that adjust the morningstar categories of equity funds; tenure is a control variable for management experience; net assets is the control variable for the size of the fund, while years of the fund measures the age of the fund. the estimation of the dummy variables for time is not shown. n is the number of funds and r2_w the adjusted coefficient of regression. * significant at 10%; ** significant at 5% and *** significant at 1%. renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 40 table 3. regression of panel data, of the expense ratio (t-3) of funds with respect to their performance, with random effects. variable alfa sharpe sortino annual return net expenses(t-3) -0.0477*** -0.0258*** -0.0394** -0.5287*** tenure (t-3) -0.0001 -0.0012 -0.0029 -0.0200 net assets (t-3) 0.0097 -0.0100** -0.0074 0.0915 funds years(t-3) 0.0002 -0.0029 -0.0068* -0.0439* flexcap 0.1173 -0.0062 0.0157 -0.1748 largecap 0.1412 -0.0422 -0.1100 -0.9987 midcap 0.0296 0.2375** 0.4300** 2.7538* smallcap 0.2878** 0.2125** 0.3142 3.7286*** constant 0.2764 1.5384*** 2.6899*** 3.6384* n 1154 1163 1163 1163 r2_w 0.7487 0.9413 0.9066 0.9232 table 4. regression in the panel data, of the expense ratio (t-3) of the funds with respect to their performance with star rating control variables (t-3), with random effects. variable alpha sharpe sortino annual return net expenses (t-3) -0.0555*** -0.0374*** -0.0651*** -0.5939*** tenure (t-3) 0.0004 -0.0013 -0.0037 -0.0389 net assets (t-3) 0.0156 -0.0115** -0.0056 0.0940 funds years (t-3) 0.0019 -0.0022 -0.0041 -0.0144 1 star (t-3) -0.0480 -0.0851** -0.3035*** -1.5617* 2 stars (t-3) -0.0947 -0.0175 -0.1518*** -1.0800 3 stars (t-3) -0.0243 -0.0210 -0.1263** -1.1791* 4 stars (t-3) -0.0407 -0.0387* -0.1219** -0.2052 flexcap 0.0750 0.0377 0.1047 -0.2923 largecap 0.0790 0.0449 0.0396 -0.8909 midcap -0.0157 0.2455** 0.3878* 2.0371 smallcap 0.3006* 0.2934*** 0.4884** 3.0196* constant 0.2508 1.5205*** 2.6476*** 4.2060 n 887 895 895 895 r2_w 0.7607 0.9465 0.9107 0.9282 note: stars 1, 2, 3, and 4 are morningstar's quantitative ratings of 1, 2, 3, and 4 stars respectively. table 4 shows the relationship between expenses and performance, using morningstar ratings as a control variable. all variables are lagged for 3 years. it turns out that, in the long term (3 years), the expenses are effective in predicting future performance for all metrics, even incorporating the morningstar ratings. annex 2 shows the relationship between expenses for the year itself, and performance taking into account the information on the rating of a fund that we knew in the previous period (t-1). renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 41 we can see a negative relationship between the fund's expenses and its future performance for the performance measures of alfa, sharpe and sortino, although it is positive and significant for the annual net return. this means that when we use ratings in the model, the ability of expenses to predict performance is lost, since ratings absorb some of that predictive power. which means that ratings help us predict performance and virtually eliminate the predictive power of expenses. annex 3 shows the relationship between the expense ratio from a year ago, and performance. we can see a negative relationship between the fund's expenses and its future performance for all risk-adjusted performance measures. this means that when we use the stars (ratings) in the model, the ability of expenses to predict the risk-adjusted return (sharpe and sortino) at 1 year is lost, since the ratings absorb part of that predictive power. still, expenses can help predict performance for alpha 4-factor adjusted returns and net returns.in table 5 we show the results of the quantile regression models, which allow estimating heterogeneous behaviours according to different levels (quantiles) of the dependent variable. quantile 0,25, median and quantile 0,75 (q25, q50, q75) were used, observing that there is a negative relationship between performance and the level of expenditure for all risk-adjusted measures and the gross annual return for quantiles 0.25 and 0.5. however, in the 0.75 quantile the situation is different, with a positive relationship between performance and the level of spending. this means that in the best funds, the higher the expenses, the better is your performance. this result is consistent with livingston, yao and zhou (2019) who also, through quantile regression techniques, showed that funds with a higher level of expenses have greater volatility in performance. in annex 4 we run a quantile regression and morningstar ratings as control variables. here we confirm the results previously achieved (a negative relationship in the 0.25 and 0.5 quantiles and a positive relationship in the 0.75 quantile). the results are consistent with the conclusions drawn without the morningstar ratings, even though by using the ratings, significance is lost in q75 in relation to alfa, sharpe and sortino, maintaining the annual performance. 4.3 robustness as a robustness test, we used a quantile regression, but with different quantiles from those used previously. in this way, we use the first decile, the median and the ninth decile (q10, q50, q90). table 6 shows the robustness of the results. it is observed for the lowest quantiles, in this case the 0.1 and 0.5 quantiles, spending has a negative relationship with the different performance measures. for the 0.9 quantile there is a positive relationship between performance and spending. these results are consistent with those found previously in table 5. annex 5 shows the estimate when we incorporate fund ratings as an explanatory variable. it is observed that expenses have a negative and significant relationship in the lower quantiles (0.10, 0.50) and a positive and significant relationship in the upper quantile (0.90). renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 42 table 5. quantile regression of the relationship between performance and expenses. variable alfa sharpe sortino annual return q25 net expenses -0.1067*** -0.0356*** -0.0438*** -0.8261*** tenure 0.0039 0.0012* 0.0007 0.0510 net assets 0.0009 0.0144*** 0.0150*** 0.0981 funds years 0.0010 -0.0040*** -0.0043*** -0.0337 flexcap 0.1602 -0.0638 -0.0403 0.6385 largecap 0.2255 -0.0248 -0.0142 1.4982 midcap 0.1994 0.0965** 0.1842*** 2.3506 smallcap 0.2150 -0.0824 -0.0889 1.0160 const -0.6271** -0.5372 -0.6436*** 9.0485** q50 net expenses -0.0208 -0.0241** -0.0262** -0.2419 tenure 0.0006 0.0010 0.0008 0.0110 net assets 0.0018 0.0057* 0.0057 -0.0095 funds years -0.0040*** -0.0020 -0.0022 -0.0445 flexcap 0.3272*** 0.1063* 0.1160* 1.9252 largecap 0.2596*** 0.0439 0.0566 0.5439 midcap 0.5121*** 0.2788*** 0.3776*** 5.6020* smallcap 0.5766*** 0.1945** 0.1810 3.9711* cons -0.3512* -0.4582 -0.5763*** 12.6173*** q75 net expenses 0.0174 -0.0102* -0.0109 0.6517** tenure -0.0026 -0.0004 -0.0013 0.0202 net assets -0.0053 0.0021 0.0041 -0.0043 funds years -0.0046 -0.0011 -0.0014 -0.0485 flexcap 0.3798* 0.1790*** 0.2323*** 4.6076 largecap 0.0397 0.0017 0.0217 -0.0687 midcap 0.4172** 0.2886*** 0.4871*** 9.4906** smallcap 0.5675*** 0.2503** 0.3850*** 8.6049** cons 0.0113 -0.2822 -0.4368** 13.9186*** n 1411 1185 1185 1388 note: q25, q50, q75 refer to the quantile estimate made for the 25th quantile, the median and the 75th quantile, respectively. renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 43 table 6. quantile regression of the relationship between performance and expenses. variable alfa sharpe sortino annual return q10 net expenses -0.1508*** -0.0568*** -0.0649*** -0.9561*** tenure 0.0041 0.0021** 0.0023 -0.0021 net assets 0.0122 0.0249*** 0.0244*** 0.3866*** funds years 0.0087*** -0.0018** -0.0026* -0.0353 flexcap 1.0371** -0.0220 0.0059 -0.2857 largecap 1.0281** 0.0229 0.0457 1.7556 midcap 0.8306 0.1355 0.1827 2.1185 smallcap 0.8512* -0.1164 -0.0275 -2.7582 const -1.8438*** -0.8137*** -0.9105* 3.1581 q50 net expenses -0.0208 -0.0241*** -0.0262** -0.2419* tenure 0.0006 0.0010 0.0008 0.0110 net assets 0.0018 0.0057** 0.0057 -0.0095 funds years -0.0040** -0.0020* -0.0022* -0.0445 flexcap 0.3272*** 0.1063* 0.1160* 1.9252 largecap 0.2596** 0.0439 0.0566 0.5439 midcap 0.5121** 0.2788*** 0.3776*** 5.6020*** smallcap 0.5766*** 0.1945*** 0.1810* 3.9711 const -0.3512 -0.4582*** -0.5763*** 12.6173*** q90 net expenses 0.0541** -0.0005 -0.0054 0.7930*** tenure -0.0060 -0.0022 -0.0005 -0.0906* net assets -0.0209 0.0065 0.0056 0.0249 funds years -0.0104*** -0.0019 -0.0020 -0.0313 flexcap 0.5107*** 0.2191*** 0.3139** 6.3175*** largecap 0.1018 0.0390 0.0467 0.5299 midcap 0.7697** 0.3080*** 0.5209*** 10.7623** smallcap 0.7123** 0.4793*** 0.7060*** 12.3191*** const 0.5300 -0.1870 -0.1827 16.3357*** n 1411 1185 1185 1388 there are more than a hundred metrics in the performance literature (see cogneau and hübner 2009a, 2009b). we have performed as a robustness test the analysis of the results of using other frequently used metrics such as the treynor ratio (1965) and the jensen alpha (1968). this study reports results very similar to those obtained previously. 5. conclusions this study focuses on understanding the relationship between the level of expenses of mutual funds in the euro zone and their current and future performance. in our study, there appears to be evidence, similar to daniel, grinblatt, titman and wermers (1997) or malkiel (1995), that the renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 44 funds' expenditures are related to their performance. on average, we can say that there is a negative relationship between both variables. however, the results are consistent with livingston, yao, and zhou (2019), showing that the best performing funds may have a high expense value. given that, through quantile regression techniques, we demonstrate the existence of a negative relationship of performance expenses in the worst-performing funds and a positive relationship in the best performing funds. second, we conclude that the past spending level of mutual funds can be a good indicator for selecting funds, with the objective of obtaining a better future performance, with a stronger relationship at 3 years than at 1 year. finally, by incorporating the morningstar star ratings, we conclude in line with morey and gottesman (2006), müller and weber (2014), meinhardt (2014), antypas, caporale, kourogenis and pittis (2009) and otero, durán and domingues (2019) that these can add useful information to select funds. in particular, we find that the rating has strong predictive power in the short term. however, in the long term, spending has greater explanatory power. in this sense, if investors use morningstar star ratings and the fund's expense levels, they can improve the selection process for short and long-term mutual funds. therefore, we can conclude that spending is an important variable in selecting the most profitable funds in the short and medium term. however, other variables need to be added to understand the determinants of their performance better. as we discussed earlier, there are multiple metrics that investors can use. in this way, as a limitation of this work, we have to indicate, it could be that other metrics not analysed did not maintain the results achieved in this article, since not all have been analysed. the use of dynamic models and incorporating other variables into the models are proposed for future research. renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 45 references alexander, g., jones, j., & nigro, p. (1998). mutual fund shareholders: characteristics, investor knowledge, and sources of information. financial services review, 7 (4), 301-316. https://doi.org/10.1016/s1057-0810(99)00023-2 antypas, a., caporale, g. m., kourogenis, n., & pittis, n. (2009). selectivity, market timing and the morningstar star-rating system. https://doi.org/10.2139/ssrn.1360637 armstrong, w., genc, e., & verbeek, m. (2018)). going for gold: an analysis of morningstar analyst ratings. management science. 65, 5. https://doi.org/10.1287/mnsc.2017.2884 baltagi, b. (2010). econometric analysis of panel data (fourth ed ed.): john wiley london. berk, j. & green, r. (2004). mutual fund flows and performance in rational markets. journal of political economy, 112, 1269-1295. https://doi.org/10.1086/424739 blake, c., & morey, m. (2000). morningstar ratings and mutual fund performance. journal of financial and quantitative analysis, 35(3), 451-483. https://doi.org/10.2307/2676213 cahart, m. (1997). on persistence in mutual fund performance. journal of finance, 52( 1) ,5782. https://doi.org/10.1111/j.1540-6261.1997.tb03808.x capon, n., fitzsimons, g., & prince, r. (1996). and individual level analysis of the mutual fund investment decision. journal of financial services research, 10, 59-82. https://doi.org/10.1007/bf00120146 chalmers, j., edelen, r., & kadlec, g. (1999). an analysis of mutual fund trading costs. ssrn. https://doi.org/10.2139/ssrn.195849 chen, c. r., & huang, y. (2011). mutual fund governance and performance: a quantile regression analysis of morningstar’s stewardship grade. corporate governance: an international review, 19(4), 311–333. https://doi.org/10.1111/j.1467-8683.2011.00858.x chotivetthamrong, c. (2015). the performance of thai mutual funds: a 5-star. journal of economics, business and management, 3 (1). https://doi.org/10.7763/joebm.2015.v3.163 corporation, f. r. (2002). predicting mutual fund performance ii: after the bear.boston. financial research corporation publications. cremers, m., & petajisto, a. (2009). how active is your fund manager? a new measure that predicts performance. review of financial studies, 22(9): 3329–3365. https://doi.org/10.1093/rfs/hhp057 cogneau, p., & hübner, g. (2009a). the (more than) 100 ways to measure portfolio performance. part 1: standardized risk-adjusted measures. journal of performance measurement, 13, 5671 cogneau, p., & hübner, g. (2009b). the (more than) 100 ways to measure portfolio performance: part 2: special measures and comparison. journal of performance measurement, 14, 56-69. daniel, k., grinblatt, m., titman, s., & wermers, r. (1997). measuring mutual fund performance with characteristic‐based benchmarks. journal of finance, (53)2,1035-1058. https://doi.org/10.1111/j.1540-6261.1997.tb02724.x díaz-mendoza, a. c., lópez-espinosa, g., & martínez, m. a. (2012). the efficiency of performance-based fee funds. european financial management, 20(4), 825–855. https://doi.org/10.1111/j.1468-036x.2012.00654.x duret, a., hereil, p., mitaine, p., moussavi, n., & roncalli, t. (2008). fund rating systems and performance predictability. https://doi.org/10.2139/ssrn.1121383 efron, b. (1979). bootstrap methods: another look at the jackknife. annals of statistics, 7(1), 1– 26. https://doi.org/10.1214/aos/1176344552 ferreira, m., keswani, a., miguel, a., & ramos, s. (2012). the flow-performance relationship around the world. journal of banking & finance, 36 (6), 1759-1780. https://doi.org/10.1016/j.jbankfin.2012.01.019 ferruz, l., & alda, m. (2012). the role of fees in pension fund performance. evidence from spain. czech journal of economics and finance, 62(6), 518-535. fernandez, p., ortiz, a., (2015). rentabilidad de los fondos de pensiones en españa. 20002015. ssrn electronic journal. https://doi.org/10.2139/ssrn.2724056 flood, c (2015). high fees hurt pension saving for millions of europeans. financial times. available at (paywall): https://www.ft.com/content/864b5eca-62ca-11e5-9846-de406ccb37f2 gil-bazo, j., & ruiz-verdü , p. (2009). the relation between price and performance in the mutual fund industry. the journal of finance, 64,2153-2183. https://doi.org/10.1111/j.15406261.2009.01497.x grinblatt, m., & titman, s. (1992). the persistence of mutual fund performance. journal of finance, 47(5),1977–1984. https://doi.org/10.1111/j.1540-6261.1992.tb04692.x https://doi.org/10.1016/s1057-0810(99)00023-2 https://doi.org/10.2139/ssrn.1360637 https://doi.org/10.1287/mnsc.2017.2884 https://doi.org/10.1086/424739 https://doi.org/10.2307/2676213 https://doi.org/10.1111/j.1540-6261.1997.tb03808.x https://doi.org/10.1007/bf00120146 https://doi.org/10.2139/ssrn.195849 https://doi.org/10.1111/j.1467-8683.2011.00858.x https://doi.org/10.7763/joebm.2015.v3.163 https://doi.org/10.1093/rfs/hhp057 https://doi.org/10.1111/j.1540-6261.1997.tb02724.x https://doi.org/10.1111/j.1468-036x.2012.00654.x https://doi.org/10.2139/ssrn.1121383 https://doi.org/10.1214/aos/1176344552 https://doi.org/10.1016/j.jbankfin.2012.01.019 https://doi.org/10.2139/ssrn.2724056 https://www.ft.com/content/864b5eca-62ca-11e5-9846-de406ccb37f2 https://doi.org/10.1111/j.1540-6261.2009.01497.x https://doi.org/10.1111/j.1540-6261.2009.01497.x https://doi.org/10.1111/j.1540-6261.1992.tb04692.x renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 46 gruber, m. (1996). another puzzle: the growth in actively managed mutual funds. the journal of finance, 51(3), 783-810. https://doi.org/10.1111/j.1540-6261.1996.tb02707.x jensen, m. c. (1968). the performance of mutual funds in the period 1945-1964. the journal of finance, 23(2), 389–416. https://doi.org/10.1111/j.1540-6261.1968.tb00815.x koenker, r., & bassett, g. (1978). regression quantiles. econometrica, 46(1), 33–50. https://doi.org/10.2307/1913643 haslem, j., baker, h., & smith, d. (2008). performance and characteristics of actively managed retail equity mutual funds with diverse expense ratios. financial services review, 17(1), 49–68. https://doi.org/10.2139/ssrn.1128983 howe, t., & pope, r. (1996). equity mutual fund historical performance ratings as predictors of future performance. journal of financial and strategic decisions, 9 (1). kosowski, r., timmermann, a., white, h., & wermers, r. (2006). can mutual fund ‘stars’ really pick stocks? new evidence from a bootstrap analysis. journal of finance, 61(6), 2551–2595. https://doi.org/10.1111/j.1540-6261.2006.01015.x koenker, r., & basset, g. (1978). regression quantiles. econometrica, 46(1), 33–50. https://doi.org/10.2307/1913643 livingston, m., yao, p., & zhou, l. (2019). the volatility of mutual fund performance. journal of economics and business, 104, 1-19. https://doi.org/10.1016/j.jeconbus.2019.02.001 livingston, m., & o´neal, e. (1996). mutual fund brokerage commissions. journal of financial research, 19(2), 273-292. https://doi.org/10.1111/j.1475-6803.1996.tb00597.x malkiel, b. (1995). returns from investing in equity mutual funds 1971 to 1991. journal of finance, 50(2),1540-6261. https://doi.org/10.1111/j.1540-6261.1995.tb04795.x matallin-saez, j., soler-dominguez, a., & tortosa-ausina, e. (2012). mutual fund performance: banking versus independent managers. applied economics letters, 19(8), 755-758. https://doi.org/10.1080/13504851.2011.602008 meinhardt, c. (2014). ratings and performance of german mutual funds – a comparison of feri trust, finanztest and fondsnote. https://doi.org/10.2139/ssrn.1943577 morey, m. (2005). the kiss of death: a 5-star morningstar mutual fund rating? journal of investment management, 3(2), 41–52. https://doi.org/10.2139/ssrn.455240 morey, b., & gottesman, a. (2006). morningstar mutual fund ratings redux. ssrn, https://doi.org/10.2139/ssrn.890128 müller, s., & weber, m. (2014). evaluating the rating of stiftung warentest: how good are mutual fund ratings and can they be improved? european financial management, 20(2), 207-235. https://doi.org/10.1111/j.1468-036x.2011.00632.x nitzsche, d., cuthbertson, k., & o'sulliv, n. (2006). mutual fund performance. https://doi.org/10.2139/ssrn.955807 o’neal, e. (2004). purchase and redemption patterns of u.s. equity mutual funds. financial management, 33(1), 63–90. https://ssrn.com/abstract=520267. otero, l., duran-santomil, p., & correia-domingues, r. (2019). do investors obtain better results selecting mutual funds. spanish journal of finance and accounting, 49(3), 265-291. https://doi.org/10.1080/02102412.2019.1622066 otero, l. & durán, p. (2021). is quantitative and qualitative information relevant for choosing mutual funds? journal of business research, 123, 476–488. https://doi.org/10.1016/j.jbusres.2020.10.015 philips, c.b., & kinniry, f.m. (2010). mutual fund ratings and future performance authors. petajisto, a. (2013). active share and mutual fund performance. financial analysts journal, 69 (4), 73-93. https://doi.org/10.2469/faj.v69.n4.7 treynor j. (1965). how to rate management of investment funds. harvard business review, 43(1), 63-75. sharpe, w. (1991). the arithmetic of active management. financial analysts journal, 47, 7-9. https://doi.org/10.2469/faj.v47.n1.7. sharpe, w. (1966). mutual fund performance. the journal of business, 39,119-138 http://dx.doi.org/10.1086/294846 sirri, e., & tufano, p. (1998). costly search and mutual fund flows. journal of finance, 53(5), 1589-1622. https://doi.org/10.1111/0022-1082.00066 wermers, r. (2000). mutual fund performance: an empirical decomposition into stock-picking talent, style, transaction costs, and expenses. journal of finance, 55(4),1655–1703. https://doi.org/10.1111/0022-1082.00263 https://doi.org/10.1111/j.1540-6261.1996.tb02707.x https://doi.org/10.1111/j.1540-6261.1968.tb00815.x https://doi.org/10.2307/1913643 https://doi.org/10.2139/ssrn.1128983 https://doi.org/10.1111/j.1540-6261.2006.01015.x https://doi.org/10.2307/1913643 https://doi.org/10.1016/j.jeconbus.2019.02.001 https://doi.org/10.1111/j.1475-6803.1996.tb00597.x https://doi.org/10.1111/j.1540-6261.1995.tb04795.x https://doi.org/10.1080/13504851.2011.602008 https://doi.org/10.2139/ssrn.1943577 https://doi.org/10.2139/ssrn.455240 https://doi.org/10.2139/ssrn.890128 https://doi.org/10.1111/j.1468-036x.2011.00632.x https://doi.org/10.2139/ssrn.955807 https://doi.org/10.1080/02102412.2019.1622066 https://doi.org/10.1016/j.jbusres.2020.10.015 https://doi.org/10.2469/faj.v69.n4.7 http://dx.doi.org/10.1086/294846 https://doi.org/10.1111/0022-1082.00066 https://doi.org/10.1111/0022-1082.00263 renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 47 annex 1. regression in the panel data of the expenditure ratio (t-1) of the funds with respect to their performance, with random effects. annex 2. regression in the panel data of the fund's expense ratio with respect to its performance with star rating control variables (t-1), with random effects. variable alfa sharpe sortino annual return net expenses (t-1) -0.0369** -0.0154** -0.0268** -0.6968*** tenure (t-1) 0.0013 0.0011 0.0010 -0.0084 net assets (t-1) -0.0076 0.0075* 0.0157** -0.1399 funds years (t-1) 0.0001 -0.0037** -0.0064** -0.0320 flexcap 0.2485 0.0583 0.1223 1.9633 largecap 0.1583 -0.0037 0.0008 0.1106 midcap 0.2714* 0.2490*** 0.4068*** 5.7077*** smallcap 0.3249** 0.2271*** 0.3822*** 5.3183*** _cons 0.4959** 1.2121*** 2.2216*** 6.4395*** n 1349 1231 1231 1361 r2_w 0.7128 0.9431 0.9204 0.9444 variable alfa sharpe sortino annual return net expenses -0.0190 -0.0092 -0.0205 0.4999* tenure 0.0003 -0.0001 -0.0006 -0.0048 net assets -0.0037 0.0108*** 0.0159** 0.2073 funds years 0.0015 -0.0009 -0.0006 -0.0836* 1 star (t-1) -0.3501*** -0.4447*** -0.7626*** -0.8520 2 stars (t-1) -0.1372** -0.3160*** -0.5964*** 1.3952* 3 stars (t-1) -0.1157** -0.2356*** -0.4549*** 1.1838 4 stars (t-1) -0.1250** -0.1611*** -0.3202*** 0.4315 flexcap 0.3118** 0.0858 0.1956** 1.7596 largecap 0.3028** 0.0097 0.0080 0.7730 midcap 0.4016*** 0.1528** 0.2488** 5.8792** smallcap 0.6900*** 0.1959*** 0.2889*** 7.6662*** _cons 0.3502 1.2970*** 2.4671*** -2.1236 n 978 990 990 990 r2_w 0.7059 0.9503 0.9134 0.9493 renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 48 annex 3. regression in the panel data, of the expense ratio (t-1) of the funds with respect to their performance with star rating control variables (t-1), with random effects. variable alfa sharpe sortino annual return net expenses (t-1) -0.0399** -0.0097 -0.0166 -0.6335*** tenure(t-1) 0.0007 0.0006 0.0006 0.0022 net assets (t-1) -0.0075 0.0042 0.0096 -0.1236 funds years (t-1) 0.0027 -0.0011 -0.0025 -0.0193 1 star (t-1) -0.2097*** -0.3912*** -0.6179*** -2.5680*** 2 stars (t-1) -0.0424 -0.2968*** -0.5103*** -0.0497 3 stars (t-1) -0.0448 -0.2283*** -0.4071*** -0.4580 4 stars (t-1) -0.0253 -0.1503*** -0.2846*** -0.7123 flexcap 0.1775 0.0353 0.0857 2.1077 largecap 0.1723 -0.0126 -0.0146 0.6891 midcap 0.1683 0.1867*** 0.3049** 5.5410*** smallcap 0.3561* 0.2441*** 0.4076*** 6.7557*** _cons 0.4901* 1.4092*** 2.5632*** 5.8080** n 1057 1068 1068 1068 r2_w 0.7362 0.9492 0.9317 0.9431 renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 49 annex 4. quantile regression of the relationship between performance and expenses. variable alfa sharpe sortino annual return q25 net expenses -0.0702* -0.0240*** -0.0277*** -0.8025*** tenure 0.0022 0.0005 -0.0002 -0.0163 net assets 0.0026 0.0074*** 0.0080* 0.2224 funds years 0.0018 -0.0003 -0.0007 -0.0245 1 star (t-1) -0.0492 -0.0765*** -0.0853*** 0.5591 2 stars (t-1) 0.0297 -0.1427*** -0.1612*** 1.0409 3 stars (t-1) -0.0083 -0.2092*** -0.2617*** 0.3229 4 stars (t-1) -0.1836* -0.3979*** -0.4442*** -3.3262** flexcap 0.3285 -0.0199 0.0331 -0.8882 largecap 0.4634 -0.0402 -0.0323 -0.0239 midcap 0.3544 0.0468 0.1116 -1.3407 smallcap 0.4629 0.0257 0.1189 3.3524 cons -2.1484*** -0.2399 -0.2950 11.2904** q50 net expenses -0.0072 -0.0048 -0.0008 -0.1178 tenure -0.0003 -0.0012 -0.0013 0.0171 net assets -0.0001 0.0032 0.0080 -0.0410 funds years -0.0040 -0.0004 -0.0006 -0.0402 1 star (t-1) -0.0447 -0.1611*** -0.2567*** -0.4815 2 stars (t-1) -0.0191 -0.2339*** -0.3299*** -0.7586 3 stars (t-1) -0.0354 -0.3018*** -0.4159*** -1.2543** 4 stars (t-1) -0.2094** -0.4241*** -0.5603*** -2.9690** flexcap 0.2435 0.0353 0.0494 1.3275 largecap 0.2022 0.0009 0.0120 0.2749 midcap 0.4264** 0.1545*** 0.1784 5.0437*** smallcap 0.6319*** 0.1524*** 0.2179** 5.7617** _cons -1.3687*** -0.1087 -0.2121 15.9763*** q75 net expenses 0.0142 0.0127 0.0125 1.0721*** tenure -0.0019 0.0020** 0.0011 -0.0205 net assets -0.0219* -0.0001 -0.0014 0.0020 funds years -0.0035 0.0007 0.0012 -0.0724* 1 star (t-1) -0.1619* -0.2028*** -0.3736*** -1.4230 2 stars (t-1) -0.2081*** -0.2678*** -0.4655*** -1.3905 3 stars (t-1) -0.1937** -0.3293*** -0.5324*** -1.8922* 4 stars (t-1) -0.4484*** -0.4122*** -0.6508*** -4.0835** flexcap 0.2156 0.1274** 0.1495 5.0003* largecap 0.0231 0.0441 0.0323 0.6572 midcap 0.2512 0.2447*** 0.3048** 10.0087*** smallcap 0.7108*** 0.2875*** 0.3794** 10.4067*** _cons -0.4533 0.0570 0.2241 14.9986** n 978 990 990 990 renato correia-domingues et al. / european journal of government and economics 11(1), june 2022, 31-50 50 annex 5. quantile regression of the relationship between performance and expenses. variable alfa sharpe sortino annual return q10 net expenses -0.1388** -0.0361*** -0.0428*** -0.9359* tenure 0.0035 -0.0001 -0.0001 -0.0464 net assets 0.0370* 0.0159*** 0.0125 0.4255** funds years 0.0050 -0.0011 0.0001 -0.0052 1 star (t-1) -0.0370 -0.0923*** -0.1069*** 0.8661 2 stars (t-1) 0.1080 -0.1242*** -0.1449*** 2.2749*** 3 stars (t-1) 0.1269 -0.2003*** -0.2275*** 0.9547 4 stars (t-1) -0.1922 -0.3749*** -0.5223*** -2.1320 flexcap 0.5610 0.0474 0.1328 -0.4422 largecap 0.6909* 0.0471 0.0865 2.1836 midcap 0.5366 0.0947 0.1885 2.3780 smallcap 0.3318 0.0481 0.0757 -2.2973 _cons -3.3475*** -0.5832** -0.6522 1.4839 q50 net expenses -0.0072 -0.0048 -0.0008 -0.1178 tenure -0.0003 -0.0012 -0.0013 0.0171 net assets -0.0001 0.0032 0.0080 -0.0410 funds years -0.0040* -0.0004 -0.0006 -0.0402 1 star (t-1) -0.0447 -0.1611*** -0.2567*** -0.4815 2 stars (t-1) -0.0191 -0.2339*** -0.3299*** -0.7586 3 stars (t-1) -0.0354 -0.3018*** -0.4159*** -1.2543 4 stars (t-1) -0.2094** -0.4241*** -0.5603*** -2.9690* flexcap 0.2435 0.0353 0.0494 1.3275 largecap 0.2022 0.0009 0.0120 0.2749 midcap 0.4264* 0.1545** 0.1784*** 5.0437** smallcap 0.6319*** 0.1524** 0.2179** 5.7617** _cons -1.3687*** -0.1087 -0.2121 15.9763*** q90 net expenses 0.0578** 0.0186* 0.0220* 1.0605*** tenure 0.0047 -0.0020 -0.0024 -0.1076 net assets -0.0254 0.0062 0.0073 -0.0549 funds years -0.0028 0.0009 0.0002 0.0277 1 star (t-1) -0.1689 -0.3160*** -0.6012*** -0.7371 2 stars (t-1) -0.2990** -0.4050*** -0.7306*** -0.8581 3 stars (t-1) -0.2882** -0.4606*** -0.7774*** -0.5163 4 stars (t-1) -0.3749* -0.5503*** -0.8814*** -1.7424 flexcap 0.2929* 0.1091 0.1917 5.3523** largecap 0.0699 -0.0032 0.0378 -0.2782 midcap 0.6350*** 0.2548** 0.4390** 10.3151*** smallcap 1.2761*** 0.3646*** 0.5398 12.2550*** _cons 0.0923 0.1675 0.4179 17.9412*** n 978 990 990 990 1. introduction 2. literature review 3. sample and performance metrics 4. empirical study 5. conclusions references why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem vol.7 • no.2 2018 issn: 2254-7088 european journal of government and economics 7(2), december 2018. european journal of government and economics issn: 2254-7088 number 7, issue 2, december 2018 doi: https://doi.org/10.17979/ejge.2017.7.2 transparency and local government corruption: what does lack of transparency hide? 106-122 juan luis jiménez, daniel albalate 106-122 exchange policy credibility through the lens of the carry trade: the mexican peso and the brazilian real. 123-137 carlos fernández_herraiz, antonio javier prado domínguez, carlos pateiro-rodriguez, jesus m. garcia-iglesias path – dependence and european fisheries management 138-153 federico martín palmero, fernando gonzález laxe why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem 154-170 filipa figueira deciding on financial renegotiation in public-private partnership projects 171-199 wiston risso https://doi.org/10.17979/ejge.2017.7.2 european journal of government and economics 7(2), december 2018, 154-170. european journal of government and economics issn: 2254-7088 why the current peak in populism in the us and europe? populism as a deviation in the median voter theorem filipa figueiraa, * a university college london, united kingdom * corresponding author at: ucl school of slavonic and east european studies (ssees), university college london, gower street, london, wc1e 6bt. email: filipa.figueira@ucl.ac.uk article history. received 23 april 2018; first revision required 27 june 2018; accepted 6 october 2018. abstract. the current surge of populism in europe and the united states calls for further analysis using public choice tools. in this article, populism is modelled as a deviation from the normal state of the median voter theorem. this study adds to the public choice literature by proposing a model of populism which is suited, not only to left-wing populism, but also to other forms of populism prevalent in europe and the united states today. it is argued that, due to changes in the assumptions underpinning the median voter theorem, the operation of the model can be modified, and as a result surges of populism occur. those assumptions concern: the political spectrum; the distribution of ideological preferences; sociological, psychological and historical factors; political party competition; and extreme political preferences. it is shown that the current peak of populism in europe and the united states can be explained through a simultaneous change in all of these aspects, leading to a “perfect storm” of populism. keywords. populism; median voter theorem; public choice; ideologies; rational choice doi. https://doi.org/10.17979/ejge.2018.7.2.4423 1. introduction with the election of donald trump, the brexit vote, the new populist government in italy, and the gains in votes and seats of populist parties in several other european countries, we are clearly witnessing a surge in populism throughout europe and the united states (inglehart and norris, 2016, oliver and rahn, 2016, rooduijn, de lange and van der brug, 2014)1. however, scholars are still divided as to what may have caused it and how it can be addressed (inglehart and norris, 2016). while there is a vast body of literature on populism in political science, the public choice 1 this article is focusing on the peak in populism occurring in western-style democracies, with a particular focus on the united states and european countries. latin america and other developing countries are not the main focus here, as it is argued that populism in those countries has different characteristics, which benefit from being modelled differently (kaltwasser, 2012). mailto:filipa.figueira@ucl.ac.uk https://doi.org/10.17979/ejge.2018.7.2.4423 f. figueira / european journal of government and economics 7(2), 154-170. 155 literature on populism is scarce2, and so far has focused only on left-wing populism and excessive government spending (see, for example, acemoglu et al, 2013, and dornbusch and edwards, 1991) neglecting to study the types of populism which we are currently experiencing in the united states and europe, where government spending is no longer the crucial factor. the article takes as its starting point the median voter theorem (downs, 19573) which, it is argued, can provide us with the necessary tools to understand the multiple aspects of this phenomenon. the median voter theorem predicts that competition between political parties forces them to converge on the centre ground. this article argues that the rise of populism can be modelled as an alteration in this theorem, as voters are now increasingly rejecting parties, politicians and political positions that are at the centre of the political spectrum, and instead electing proponents of anti-consensual politics. populist politicians emphasise their outsider status, and position themselves at the end-points, or even claim to be outside, the traditional political spectrum. this article therefore argues that this can be modelled as a situation where the median voter bliss point, which would be reached through consensual policies, is being rejected by both voters and populist politicians.4 downs (1957) provides us with a set of five assumptions that need to be met for the theorem to hold. in a nutshell, those assumptions concern: the political spectrum; the distribution of ideological preferences; sociological, psychological and historical factors; political party competition; and extreme political preferences. framing populism as an alteration in the median voter theorem allows us to explore the causes of the current wave of populism, by analysing why several of those assumptions have changed simultaneously. this article therefore argues that we are currently witnessing a “perfect storm” of populism, where different factors have converged to create a peak. while there have been references to peaks in populism being caused by a combination of factors which need to be present simultaneously (oliver and rahn, 2016), the literature has not yet, to the knowledge of this author, offered a comprehensive explanation of what those factors are, in relation to the current peak. the use of the median voter theorem offers a framework in which to model and understand the 2 the public choice literature is, here, defined as the body of literature which applies economic tools to political issues, in the tradition of james m. buchanan, anthony downs and gordon tullock or, more recently, alberto alesina and dariu acemoglu. 3 the theorem is traditionally attributed to downs, in his 1957 seminal article. however, as downs acknowledges, the theorem builds on insights from a variety of scholars, including namely harold hotelling and duncan black. this article is, however, particularly interested in the version of the theorem offered by downs, as that author spells out in detail the assumptions that need to be met for the theorem to hold. 4 this analysis relies on a wider interpretation of the median voter theorem, in which the political spectrum depicts, not only the variations in policy preferences between left and right, but also the different options available to a policymaker, where the options closer to the median satisfy the preferences of a majority of voters. this wider interpretation is based on the literature which conceptualises ideologies, not as a set of specific policy views, but rather as a “constraint” on policy positions, which makes it possible to order those positions along the single dimension of a policy spectrum (converse, 1964, gabel and huber, 2000). f. figueira / european journal of government and economics 7(2), 154-170. 156 parallel factors that are, together, feeding the current surge in populism. given the multi-faceted nature of the concept of populism, this article will not use one single definition, but will instead draw on different definitions offered in the literature. this will include the definition offered by mudde (2007), of populism as a political party or politician that meets the three following characteristics: anti-establishment (opposing the “elites” and depicting the establishment as corrupt); authoritarian (emphasising personal power of a charismatic leader and/or direct democracy instead of representative democracy); nativist (against multiculturalism, and often displaying xenophobia). the article takes into account taggart’s (2004) definition of populism. according to him, populism is a combination of five elements: opposition to representative democracy; representing a heartland of “the people”; lacking core values; linked to a sense of crisis; unsustainable in the long-run. this article therefore aims to contribute to the public choice and politics literature on populism in the following ways. firstly, by modelling populism as an alteration in the equilibrium state of the median voter theorem, it aims to propose a framework on which theoretical insights on populism can be included and systematically analysed. secondly, it attempts to expand the public choice analysis of populism. to the knowledge of this author, the public choice literature so far has focused exclusively on left-wing populism, and its impact on macroeconomic policymaking, with a particular emphasis on excessive government spending, leading to unsustainable public debt and/or inflation (dornbusch and edwards, 1991).5 this article seeks to fill that gap, by using public choice in a way that is also suitable to right-wing and non-partisan populism. thirdly, this study contributes towards the analysis of why there is a peak in populism in europe and the united states in the 2010s, through the consideration of political, sociological, psychological, temporally dynamic and economic factors. the remainder of this article is organised as follows. the next section presents the model offered in this article. the following sections address, in turn, each of the assumptions of the model and how they are relevant to ongoing populist developments. the last section concludes. 2. the model: an alteration in the median voter theorem the median voter theorem states that, if a certain number of conditions are met, a democratic setting will result in policies which respect the preferences of the median voter. models and their assumptions are, by definition, an over-simplification of reality, and the median voter theorem is no exception – but it provides an optimal starting point or benchmark, from which to analyse what 5 the concept of populism has also been used with yet another different meaning in the public choice literature. in riker’s seminal “liberalism versus populism” (1982), populism is understood as the view that democracy should be primarily a mechanism for ensuring that people’s views are represented – in opposition to liberalism, where democracy should be primarily concerned with ensuring that people’s best interests are represented. this understanding of the term “populism” is not the one being used in this article. f. figueira / european journal of government and economics 7(2), 154-170. 157 is different from the model in the real world, and why. holcombe (1989) compares the median voter theorem to the assumption of perfect competition in economics – although an unrealistic assumption, it has been crucial to that discipline, in that it provides a benchmark against which real-world situations can be assessed. similarly, the median voter theorem will be used in this article as the benchmark – a perfectly functioning democratic system exempt of populism – against which the real word situation, in which populism is almost always present, but sometimes to a much greater degree than others, will be analysed. naturally, none of downs’ five assumptions is fully met at any time. however, what this article focuses on is whether the current wave of populism is linked to the fact that some of the assumptions are currently failing even more than usual.6 downs (1957) provides us with a set of five assumptions that need to be met for the theorem to hold: 1. political space is one-dimensional, and can be ordered from left to right; 2. preferences are single-peaked and have a normal, bell-shaped, distribution; 3. voters’ preferences are not altered by historical, sociological or psychological factors; 4. political parties can move within the spectrum (except beyond the nearest party); 5. extremist voters may abstain if they perceive the two main parties as too similar. despite the vast literature that followed, it is argued that downs’ five assumptions, in their simplicity, remain the most useful analysis of the pillars upon which the theorem holds. this is reinforced by the fact that subsequent literature mainly focused on either proving or disproving the theorem, rather than on exploring the five above assumptions (holcombe, 1989). this article will therefore analyse how the breakdown of each of these assumptions contributes to an alteration in the equilibrium state of the theorem. in its normal equilibrium state, the theorem predicts that the preferences of the median voter will be adopted by policymakers. at first sight this may appear not to be directly related to populism, as populism can be placed anywhere in the political system. however, the wider interpretation of the political spectrum used in this article, whereby (following converse, 1964) it depicts, not only the variations in policy preferences between left and right, but also the different options available to a policymaker, makes it possible to model populism as a deviation from the median voter’s bliss point. that equilibrium therefore represents, not only the middle of a left-right political spectrum, but also balanced policymaking which, on average, meets the best interests of a representative citizen. populism, by contrast, does not seek to offer balanced solutions that satisfy all citizens, but instead extreme and/or unusual solutions, aimed at satisfying only the interests of a “heartland that represents an idealised conception of the community they serve” (taggart, 2004, p. 274). 6 the “failure” of an assumption is understood as any deviation from the normal or equilibrium state of the model, rather than as a criticism of the model. f. figueira / european journal of government and economics 7(2), 154-170. 158 to the knowledge of this author, the only existing study which connects the median voter theorem with populism is the study of acemoglu, egorov and sonin (2013), but the authors offer a model focused on left-wing populism in latin america, which, it is argued, cannot be applied to the current wave of populism in the united states and europe. using the median voter theorem in a radically different way from the one proposed in this article, the authors propose that populism can be modelled as a situation where voters willingly chose politicians to the left of the median voter bliss point. this is because they fear that politicians may be captured by business lobbies, and will therefore chose politicians who clearly signal that they are not captured, by offering leftwing policies. the authors also allow for the same phenomenon happening to the right of the median voter, where voters are primarily afraid of politicians being too left-wing (for example, being “closet communists”). the authors therefore assume that the median voter theorem is functioning, but that populism will cause a bias to the left (or right). however, it can be argued that their model, while very useful to explain the type of populism that happens in latin america, is not helpful to analyse the type of populism taking place in europe and the united states. in particular, current populism in europe and the united states is mainly right-wing (mudde, 2004), and cannot realistically be explained by voters fearing that politicians have been captured by lobbies on the left or are “closet communists”. ongoing concerns which are key to populism, such as opposition to immigration or a perception of having lost out to globalisation (inglehart and norris, 2016), cannot be explained by this model. moreover, recent forms of populism have been disassociated from left or right-wing ideologies (for example, the election of donald trump and the brexit vote cannot be seen as biases towards either the left or the right). therefore, this article adopts a different perspective, by seeing populism, not as a bias in the median voter theorem, but as an alteration in its equilibrium state, due to changes in the assumptions that underpin the theorem. 3. assumption 1: the political spectrum is one-dimensional “1. the political parties in any society can be ordered from left to right in a manner agreed upon by all voters” (downs, 1957, p.142) the assumption of there being a one-dimensional political spectrum, along which all political issues could be neatly placed, was always an over-simplification, and has arguably become even less relevant with time (albright, 2010, giddens, 1994). despite the counter-argument that the left-right dichotomy remains relevant, as the values represented by left and right change over time to reflect new issues and concerns (bobbio, 1996, budge et al., 2001), it is impossible not to acknowledge that there is currently a multiplicity of policy dimensions in european and american politics, which cannot easily be placed along a single spectrum. today, a political compass has therefore replaced the one-dimensional spectrum in the work of several scholars (see, for example, inglehart and norris, 2016, and bakker et al, 2015). f. figueira / european journal of government and economics 7(2), 154-170. 159 is the current peak in populism partly caused by the fact that this assumption is breaking down more than usual? this article argues that this is the case, due to a combination of long-term and short-term factors. over the long-term, we can identify a gradual dismantling of this assumption over the 20th century, for the following reasons. firstly, an improvement in living standards made economic issues less relevant to voters, as opposed to social and cultural issues, which are less easily modelled in a left-right spectrum (inglehart, 1990). secondly, growing social mobility and narrowing social cleavages gradually reduced individuals’ identification with an ideology (franklin, mackie and valen, 2009). thirdly, a convergence in political views made left less distinguishable from right (albright, 2010). it is argued that this contributes to the long-term trend towards more populism, as voters are increasingly interested in political parties which do not fit within traditional ideologies, and therefore lose their identification with mainstream centre-left and centre-right parties (franklin, mackie and valen, 2009). it is also possible to identify short-term factors leading to a breakdown of this assumption nowadays. in particular, issues which do not fit easily into the left-right spectrum have become predominant in europe and the us over the past decade. one of those is immigration – although anti-immigration rhetoric is often associated with the right, the anti-immigrant sentiment is usually captured, not by mainstream right-wing parties, but rather by populist single-issue parties, which are often categorised as right-wing but do not actually share any other right-wing views on other issues (mudde, 2013). moreover, as noted by inglehart and klingemann (1976), any issues which are not purely economic or religious fit with difficulty in the left-right axis. immigration worries have become prominent in europe due to the ongoing refugee crisis, whereby conflicts in the middleeast led to a sudden surge in numbers of refugees trying to enter europe (hansen, 2016). in the united states, immigration patterns and terrorist incidents have led to xenophobic feelings against mexicans and muslims (inglehart and norris, 2016). this has distorted the traditional concerns of the left-right axis. another short-term factor is euroscepticism. euroscepticism has emerged as a populist trend, linked to the perception that the european project was led by the elites, without proper consultation of the people (taggart, 2004). here also, it is possible to identify a combination of factors, which, over the past decade, has led to rising euro-scepticism and questioning of the eu. those include economic difficulties, which historically have always been correlated with euroscepticism, and in particular the 2008 financial crisis (serricchio, tsakatika, and quaglia, 2013). they also include the exponential pace of eu integration following the adoption of the 1992 maastricht treaty, and associated concerns about a loss of national identity (eichenberg and dalton, 2007). euroscepticism can be categorised as an issue which does not fit into the left-right spectrum, and is instead better suited to the authoritarian-libertarian spectrum (marks and steenerberg, 2002). bakker et al (2015) identify three policy dimensions in europe: the traditional left-right spectrum; the libertarian-authoritarian spectrum; and the eu integration spectrum. the issues of immigration and euroscepticism can be said to fit, respectively, into the second and third spectrums. a long-term trend in the break-down of the left-right spectrum is therefore currently f. figueira / european journal of government and economics 7(2), 154-170. 160 coinciding with a short-term rise in the importance of political issues not directly related to the leftright spectrum. this is leading to a more than usual breakdown in the assumption of onedimensional politics, contributing to a more than usual deviation in the equilibrium of the median voter theorem. as people lose their identification with mainstream centre-left and centre-right parties, they are more likely to look for alternatives, and therefore more likely to become interested in populist movements. 4. assumption 2: preferences are single-peaked and normally distributed “2. each voter’s preferences are single-peaked at some point on the scale and slope monotonically downward on either side of the peak (unless it lies at one extreme of the scale).” (downs, 1957, p.142) according to this assumption, for the median voter theorem to hold, voters’ preferences need to follow a normal, bell-shaped distribution, whereby the majority of voters is located in the middle, and the number of voters becomes smaller the further away one moves from the middle7. this assumption is reasonable for a traditional left-right spectrum, when we assume that voters prefer a more left-wing or right-wing set of policies depending on their level of income (as naturally more redistributive policies benefit directly the poor, while being directly unfavourable to the rich, who will need to pay more taxes). this leaves the middle classes as the majority in the middle. once we introduce additional spectra, as discussed in the previous section, this assumption is still reasonable. in a spectrum going from liberalism to authoritarianism, it can still be expected that a majority will have a balanced view, away from the extremes. the same can be assumed of a spectrum going from complete eu integration to opposing the eu altogether. however, the assumption breaks down when voters are willingly asking for radical or extreme policies in very large numbers (demand side) or politicians are willingly offering extreme policies instead of balanced policies (supply side). this will happen when there is a polarisation of society, where groups of people have very different views from other groups. inglehart and norris (2016) show that society in the united states and europe is becoming increasingly polarised, between a group of liberal-minded voters who will usually be more well educated and younger, and a group of conservative-minded voters who will tend to be less educated, older and often male. the authors show that this demographic split is relevant both in the case of the trump election, and in the case of brexit. in such a polarised society, as predicted by downs, the normal bell-shaped distribution will be replaced by two peaks, leaving a gap at the middle. in such conditions, a politician with balanced views will be less successful in attracting votes than a politician with more extreme, populist, views. 7 the median voter theorem can be understood, both as a representation of one individual’s preferences, and as a representation of a group’s preferences. i am focusing mainly on the latter when analysing assumption 2, as that is the most relevant aspect for my analysis. f. figueira / european journal of government and economics 7(2), 154-170. 161 5. assumption 3: voters’ preferences are not altered by historical, sociological or psychological factors “the distribution of voters along the scale is variable from society to society but fixed in any one society. [reference to footnote 16:] actually, this distribution may vary in any one society even in the short-run, but i assume it to be fixed in order to avoid discussing the complex of historical, sociological, psychological and other factors which cause it to change” (downs, 1957, p. 142) downs is here assuming that voters are perfectly rational, by ignoring any historical, sociological and psychological factors that may disturb such rational behaviour. downs therefore did not argue that such factors were irrelevant to voters’ decisions, but only that they were beyond the scope of his analysis. in the absence of historical, sociological and emotional factors, voters’ decisions would be purely rational, and voting behaviour would maximise each voter’s utility. we could then expect voters to select politicians whose views correspond to theirs, and/or whose policies are beneficial to their selfish interests. such a voting behaviour will, if the other assumptions are also met, ensure that the theorem holds. once we make those factors endogenous to the analysis, voting decisions are no longer purely rational. voters are no longer only maximising their utility, and are no longer voting for the party closest to them in political space. instead, their behaviour may be guided by emotions, influenced by societal pressures or simply hampered by incomplete information. again, this assumption is never fully met in the real world, but this article argues that it is being met less than usual at the moment. there is a wide literature on the sociological, historical and psychological causes of populism, and in particular its association with crises (taggart, 2004, laclau, 2005) but there is a lack of research on why these causes are particularly acute now. this section builds on that literature, to examine why the current “perfect storm” of populism is due to a greater than usual variability in voters’ preferences. in particular, i will focus on sociological, historical and psychological reasons which lead voters to be currently particularly attracted by populism. sociological and psychological factors inglehart and norris (2016) argue that a crucial reason for populism is dissatisfaction on the side of the voters, for both economic and cultural reasons. the economic reason is a dissatisfaction emerging from changing economic trends with a significant impact on society, including globalisation, job insecurity and inequality. the cultural reason is a “backlash” against the rise in progressive values since the 1970s, in relation to multiculturalism, gender equality and human rights (inglehart, 1990). the authors find that there is more conclusive evidence to support the cultural thesis, but also admit that the two theses may be too interrelated to be entirely separated, indicating that both are valid. f. figueira / european journal of government and economics 7(2), 154-170. 162 elchardus and spruyt (2016) examine the link between support for populism and psychological factors, including declinism. declinism is defined as the feeling that society is getting worse, due to globalisation, multiculturalism, as well as changes in regulations, the environment and moral principles. this concept is clearly very close to economic and cultural dissatisfaction identified by inglehart and norris. the authors find a strong link between declinism and populist voting, and conclude that a perception that society is getting worse, is the main psychological factor that can be associated with populist voting. crucially for this article, inglehart and norris (2016) find that economic and cultural dissatisfaction have been rising gradually since the 1970s, and are now at an all-time high, particularly among the sections of the population that have predominantly voted for populist candidates in recent american and european elections (male, white and older). this trend supports this article’s thesis that voters’ tendency to act in an emotional rather than rational manner has reached a peak. it is argued that this long-term trend is now coinciding with a short-term peak, following the 2008 financial crisis. crises, in particular of an economic nature, are often at the source of populism (taggart, 2004). taggart further argues that what is relevant for populism is not crisis itself but instead the perception of crisis. although the great recession peaked in 2008, the sense of insecurity and agitation that it brought about has arguably not yet waned, particularly as many of the jobs lost during the recession were replaced with less permanent and secure positions, and the so-called euro-crisis prolonged the recession throughout europe. inglehart and norris (2016) are puzzled by the fact that economic dissatisfaction is higher, not among the lower classes, but among the middle-classes. it is suggested that this phenomenon can be explained by an economic trend: in both europe and america, growth has been relatively higher for the rich and the poor than it has been for the middle classes – this is known as the “elephant curve” (milanovic, 2013). elchardus and spruyt (2016) indeed emphasise the link between “feelings of relative deprivation” and attraction to populism, which they measure through surveys asking respondents to indicate how they perceive their economic situation in relation to that of others, as opposed to seeing it in isolation. the refugee crisis and the rise in euroscepticism, which were shown above to contribute towards a less one-dimensional political spectrum, are also shorter-term issues that feed into a sense of crisis. the short-term trends of the financial crisis, the refugee crisis and euroscepticism, combined with the long-term trend of the exponential increase in globalisation of the past 30 years (inglehart, 1990) may therefore have contributed to the current peak in populism. historical factors as mentioned above, this article assumes that the median voter theorem never holds completely, but that it breaks down more significantly at times when there is a peak in populism. the f. figueira / european journal of government and economics 7(2), 154-170. 163 perspective offered is that of recurrent break-downs in the theorem, leading to peaks in populism. the historical, cyclical and temporally dynamic nature of populism is therefore important. the cyclicality of populism has been extensively studied for the case of latin america’s left-wing populism (see, for example, dornbusch and edwards, 1991). however, there is a dearth of research on the cyclicality and temporal dynamics of right-wing and non-partisan populism, which is currently prevalent in the united states and europe. while there are studies focusing on the cyclicality of the support for individual populist parties (see, for example, anderson, 1996, for denmark and norway), comparative studies on the cyclicality of populism worldwide, and in its various forms, are lacking8. this article does not aim to offer such a study, but it does aim to contribute towards the study of cyclicality and temporal dynamics of populism, by basing the analysis of the current peak in populism on the view that it is in great part caused by the convergence of cyclical trends in europe and the united states. it is argued that populism, while always an element of democratic systems (taggart, 2004), does become more acute at certain periods in time. i argue that this evolution can most usefully be seen as a succession of peaks and troughs, and that it is possible to detect a predictable pattern similar to that of business cycles in economics. in the united states, it is possible to identify regular peaks of populism every 20 or 30 years (see, for example, kazin, 1998). those are visible in the 1890s (the people’s party, pitting small farmers against industrial interests), the early 20th century (socialist and prohibitionist movements), the 1940s (anti-communism and resurgent conservatism), the 1960s (george wallace and the white backlash), the 1980s (reagan and the christian right) and the 2010s (rise of the tea party movement, followed by the election of donald trump in 2016). in europe, the dynamics have been very different, particularly as the post-war period saw comparatively little populism (taggart, 2004) and was significantly impacted by the historical upheavals of two world wars, followed by a cold war. following the well documented peak in populism of the 1930s, involving the rise of hitler and mussolini, and culminating with the second world war, populism became less attractive (rydgren, 2005). from the 1940s until the end of the 20th century, populism in europe was less prevalent than it is today (mudde, 2004), with the traumatic effect of two wars, combined with the ideological complications of the cold war that followed, leading to a distrust of populism in both western and eastern europe. rydgren (2005) argues similarly that the stigmatization of nazism and anything associated with it led to a fall in attraction for populism in europe from the 1940s to the 1970s, boosted by the strong post-war economic development in europe. while populist movements remained present throughout those years9, the literature agrees that they were less prevalent than in the current times (see, for example, mudde, 2004, taggart, 2004), and it is difficult to identify any peak during this period.10 with memories of the (real and cold) wars fading with the end of the 8 there is a body of literature on the cyclicality of party support (for example, miller and mackie, 1973), but it is not focused on populism, which can be expected to have different dynamics. 9 examples include giannini’s common man’s front in italy (1940s) and the dutch “farmers’ party (1960s). 10 it is argued that the student protests of 1968 should be seen as a peak in civic society activism, rather f. figueira / european journal of government and economics 7(2), 154-170. 164 20th century, the 21st century is seeing the re-emergence of populism. this trend first became a concern in eastern europe (bugaric, 2008), but has now become significant in western europe as well. therefore, a cyclical peak of populism in the united states is coinciding with both a culmination of the gradual rise of populism in europe since world war 2, and a smaller peak in populism in europe caused by recent events, including the refugee crisis and the aftermath of the 2008 great recession. this coincidence between a european peak and an american peak contributes to the perfect storm of populism. 6. assumption 4: parties can move freely within the spectrum “4. once placed on the political scale, a party can move ideologically either to the left or to the right up but not beyond the nearest party towards which it is moving.” (downs, 1957, p. 142) all the assumptions so far focused on the demand-side (the voters); this assumption now focuses on the supply-side (the politicians and political parties). this assumption implies that political parties will be able to move within the spectrum freely, as to position themselves in a point where they can maximise their potential number of votes. in other words, it implies that the supply-side is functioning well in the democratic system, and that political parties are able to offer policies that match the demand from voters. if that assumption fails, voters will not be supplied with a good variety of policies to choose from, and this may explain the attraction to populism. to assess whether this is happening now, it is useful to extend the analogy with market supply and demand, to find out whether the variables that can affect supply in a market (cost, price, demand, level of competition and behaviour of producers) can explain the current situation, either through a higher supply of populism, or, indirectly, through a dysfunction of the supply side which can be used by populists in their favour11. the previous section about assumption 3 has already shown that there was an increase in demand, so this section will focus on the other factors. it can be argued that the “cost” of supplying populism has been lowered by changes in the media, and in particular by the prevalence of social media. new forms of media are more affordable, making it easier for politicians who don’t have the financial backing of mainstream parties to catch the attention and interest of voters (hong, 2013). the use of social media has been an important factor in the rise of populist candidates and movements, as evidenced most notably by donald trump’s use of twitter. the “price” can be said to change if there is an alteration in the benefit that suppliers derive from an extra voter. this happens if there is a greater opportunity to attain office, for example than a populist movement. 11 if we assume that populism is inherently negative (taggart, 2002), this section is showing a malfunction in the “supply side”, since the “defective product” (populism) is taking a higher share of the market. however, if we do not assume that populism is always suboptimal, the factors explained in this section do not necessarily show a breakdown in the assumption, but rather a change in its operation. f. figueira / european journal of government and economics 7(2), 154-170. 165 because populist parties which before would not have had that chance are now able to have it, as populism has become more acceptable and “electable”. the past years have indeed witnessed a greater electability of populist politicians; for example, donald trump’s 2000 campaign was unsuccessful, while the same politician achieved victory in 2016 (oliver and rahn, 2016). it can therefore be concluded that the price of populism has risen, leading to higher supply. another factor that be said to differ is the behaviour of “suppliers”. under perfect market conditions, suppliers need to have as their main objective the maximisation of profit or, in this case, votes. however, populist parties will often not be vote-seeking parties, but rather policyseeking parties (strom, 1990), which are not aiming to maximise their votes but instead to have an influence on policymaking. as noted in the above paragraph, several populist movements and politicians were seen as unelectable until recently, and indeed many still have not achieved office, even though the current wave of populism has provided them with the opportunity to reach objectives far beyond what was within their reach beforehand. starting with objectives other than vote-maximisation leads to populist politicians positioning themselves elsewhere than at the median. indeed, extreme-left (such as syriza or podemos) or extreme-right (such as front national or alternatief fuer deutschland) positions are often present. this, together with voters’ interest in such extreme positions, contributes towards a malfunction of the median voter theorem. more generally, populists can also benefit if there is a dysnfunction in the supply side, which will make them more attractive to voters who feel that they are not being adequately represented. oliver and rahn (2016) show that trump and other populist successes in the us were caused by a representation gap, whereby mainstream parties were not adequately representing the views of the voters. mainstream parties struggle to address divisive issues such as immigration and eu integration, leaving a vacuum in certain spots of those spectrums which can be taken by populists. 7. assumption 5: extremist voters may abstain if they perceive the main parties as too similar “5. in a two-party system, if either party moves away from the extreme nearest it toward the other party, extremist voters at its end of the scale may abstain because they see no significant difference between the choices offered them.”(downs, 1957, p. 142) this assumption foresees that extremist voters will abstain because their extreme views are not represented by any of the parties with a realistic chance of winning. this assumption can be said to break down when extremist voters and extreme views, which are normally not represented in the political system, start to be represented. indeed, populist movements often include the political mobilisation of sections of the electorate which would otherwise have felt alienated by the political system (oliver and rahn, 2016). extreme views are therefore legitimised and become part of the political discourse. these include views that would normally be seen as xenophobic (such as an opposition to immigration or targeting specific origins by religion or ethnicity), paranoid f. figueira / european journal of government and economics 7(2), 154-170. 166 (such as believing in a world-wide conspiracy between banks and large companies) or unviable (such as complete opposition to capitalism). i argue that this legitimisation is happening now more than usual, because the current populist climate has led to views which are normally seen as unacceptable being legitimised. this is a self-fulfilling prophecy: as populism rises, populist views become more acceptable, which in turn fosters more populism (mudde, 2004). the literature on “contagion” effects of populism (roduijn et al, 2014, rydgren, 2005) is useful to explain how populist gains in one party/ country can lead to populist gains somewhere else. however, so far the term “contagion” in the populism literature has been used in relation to programmatic contagion, this happening when mainstream parties are influenced by populist parties (see, for example, roduijn et al, 2014). instead, this article is concerned with the international contagion of populism itself, or the fact that a peak of populism in one country can spread to other countries. a case in point occurred in january 2017, when the leaders of several european populist parties met in germany to celebrate the election of donald trump, and told the press that trump’s victory would spur their own parties (connolly, 2017). rydgren (2005) studies the contagion between right-wing populist parties in europe, and argues that those parties mutually legitimize each other, in particular through the creation of an ideological master frame based on nativism and authoritarianism. i argue that this contagion effect goes beyond europe (as can be seen from the above mentioned references to trump by european populist parties) and is extended by the increasing globalisation of media and exchange of ideas. it can therefore be argued that one of the forces contributing towards the current rise in populism is a snow-ball effect, whereby the gradual rise in populism has in turn led to a legitimization of extremist political views, which in turn promotes further gains in populism. 8. conclusion this article has proposed a theoretical framework for the analysis of populism, based on an alteration of the equilibrium state of the median voter theorem. it has shown that populism can be directly linked with basic assumptions related to the functioning of the political system. in an ideal world where those assumptions were fully met, the theorem would apply fully, and policymaking would fully meet the preferences of the median voter. at the other extreme, when several of those assumptions break down, this results in outcomes that appear to run counter the logic of democracy, such as the seemingly irrational voting for populist politicians, resulting in outcomes that clearly do not meet median, average or balanced preferences. while none of the theorem’s assumptions is ever fully met in the real world, this article found that all the assumptions were failing more than usual at present, and that this can be associated with the current peak in populism. in particular: the left-right spectrum has lost its relevance; the distribution of ideological preferences is not normal due to a polarisation of society; sociological, f. figueira / european journal of government and economics 7(2), 154-170. 167 psychological and historical factors are affecting voters’ preferences more than usual; political party competition has been disrupted; and extreme political preferences have been legitimised. the combination of all these factors leads to the current perfect storm of populism. the article builds on existing research on the causes of populism (for example, inglehart and norris, 2016), and extends it to explain the current peak in populism. rather than attempting to find one key narrative to explain populism, this article takes a broader approach by looking at different elements of the political system, and assessing how each of them contributes to populism. in doing so, importance is given to the phenomena of contagion and cyclicality, and the role that they play in bringing together the different factors. for example, the way in which demand side changes feed into supply side changes is analysed, as well as how changes in one country carry through to another. it is important to have a theoretical framework under which to model populism because this provides scholars with a tool to examine and compare this phenomenon, both between countries, and over time. such a comparative approach makes it possible to understand why populism varies across time and regions, and to analyse what causes those variations. it also contributes towards understanding the ongoing peak in populism, which in turn helps analyse how can it be addressed optimally. the main limitation of this study is that the model proposed, as with any model, is imperfect and needs to be used with caution to avoid over-simplification. the phenomena described are highly complex, and to model them inevitably involves making simplifying assumptions. it is therefore important to bear in mind that not all the complexities of the phenomena have been captured by the model, as indeed no model can fully reflect reality. in particular, populism is not the only reason why the median voter theorem’s assumptions are challenged, and not all of those challenges result in populism. this article uses the median voter theorem as a benchmark a model that describes an unrealistic situation of perfect competition in the political system, against which the real world is measured. this contrasts with the usual approach to the theorem in the literature, whereby it is either assumed to hold perfectly, or rejected outright (holcombe, 1989). this approach may be extended to other areas of research, and could be useful to analyse other aspects of the political system. for example, if the median voter theorem is seen as the benchmark of perfect democratic system, deviations from that benchmark can also be useful to understand other issues that affect the democratic system, such as declining interest and participation in the political system by the voters, or changes in the media and how this affects politics. the wider definition of the theorem proposed in this paper makes it relevant to analyse not only the left-right political spectrum, but also any other issues which potentially destabilise the political system. f. figueira / european journal of government and economics 7(2), 154-170. 168 references acemoglu, d., egorov, g. and sonin, k. (2013). a political theory of populism. the quarterly journal of economics, 128(2), pp.771-805. doi: https://doi.org/10.1093/qje/qjs077 albright, j.j. (2010). the multidimensional nature of party competition. party politics, 16(6), pp.699719. doi: https://doi.org/10.1177/1354068809345856 anderson, c.j. (1996). economics, politics, and foreigners: populist party support in denmark and norway. electoral studies, 15(4), pp.497-511. doi: https://doi.org/10.1016/s02613794(96)00030-3 bakker, r., de vries, c., edwards, e., hooghe, l., jolly, s., marks, g., polk, j., rovny, j., steenbergen, m. and vachudova, m. a. (2015). measuring party positions in europe: the chapel hill expert survey trend file, 1999–2010. party politics, 21(1), 143-152. doi: https://doi.org/10.1177/1354068812462931 bobbio, n. and cameron, a. (1996). left and right: the significance of a political distinction. university of chicago press. budge, i., klingemann, h.d., volkens, a., bara, j. and tanenbaum, e. (2001), mapping policy preferences: estimates for parties, electors, and governments, 1945–1998. new york: oxford university press. bugaric, b. (2008). populism, liberal democracy, and the rule of law in central and eastern europe. communist and post-communist studies, 41(2), 191-203. doi: https://doi.org/10.1016/j.postcomstud.2008.03.006 connolly, k. (2017). « after the us, far right says 2017 will be the year when europe wakes up ». the guardian, january, available at: https://www.theguardian.com/world/2017/jan/21/koblenzfar-right-european-political-leaders-meeting-brexit-donald-trump (accessed 5 june 2017). converse, philip. e. (1964). the nature of belief systems in mass publics. in ideology and discontent, ed. d. e. apter. new york: free press. dornbusch, r. and edwards, s. (1991). the macroeconomics of populism. in the macroeconomics of populism in latin america (pp.7-13). university of chicago press. doi: https://doi.org/10.7208/chicago/9780226158488.001.0001 downs, a. (1957). an economic theory of political action in a democracy. journal of political economy, 65(2), pp.135-150. doi: https://doi.org/10.1086/257897 eichenberg, r.c. and dalton, r.j. (2007). post-maastricht blues: the transformation of citizen support for european integration, 1973–2004. acta politica, 42(2-3), pp.128-152. doi: https://doi.org/10.1057/palgrave.ap.5500182 elchardus, m., & spruyt, b. (2016). populism, persistent republicanism and declinism: an empirical analysis of populism as a thin ideology. government and opposition, 51(1), 111-133. franklin, m. n., mackie, t. t., & valen, h. (2009). electoral change: responses to evolving social and attitudinal structures in western countries. ecpr press. gabel, m., & huber, j. (2000). putting parties in their place: inferring party left-right ideological positions from party manifestos data. american journal of political science, 44(1), 94-103. https://doi.org/10.1093/qje/qjs077 https://doi.org/10.1177/1354068809345856 https://doi.org/10.1016/s0261-3794(96)00030-3 https://doi.org/10.1016/s0261-3794(96)00030-3 https://doi.org/10.1177/1354068812462931 https://doi.org/10.1016/j.postcomstud.2008.03.006 https://doi.org/10.7208/chicago/9780226158488.001.0001 https://doi.org/10.1086/257897 https://doi.org/10.1057/palgrave.ap.5500182 f. figueira / european journal of government and economics 7(2), 154-170. 169 doi: https://doi.org/10.2307/2669295 giddens, a. (1994). beyond left and right: the future of radical politics. stanford university press. hansen, r. (2016). making immigration work: how britain and europe can cope with their immigration crises (the government and opposition/leonard schapiro lecture, 2015). government and opposition, 51(2), pp.183-208. doi: https://doi.org/10.1017/gov.2015.40 holcombe, r. (1989). the median voter model in public choice theory. public choice, 61(2), pp.115-125. doi: https://doi.org/10.1007/bf00115658 hong, s. (2013). who benefits from twitter? social media and political competition in the us house of representatives. government information quarterly, 30(4), pp.464-472. doi: https://doi.org/10.1016/j.giq.2013.05.009 inglehart, r. (1990). culture shift in advanced industrial society. princeton: princeton university press. inglehart, r. and norris, p. (2016). ‘trump, brexit and the rise of populism: economic have-nots and cultural backlash.’ american political science association annual meeting, 1-4 september, philadelphia. doi: https://doi.org/10.2139/ssrn.2818659 inglehart, r. and klingemann, h.d. (1976). party identification, ideological preference and the leftright dimension among western mass publics. party identification and beyond: representations of voting and party competition, pp.243-273. kaltwasser, c.r. (2012). the ambivalence of populism: threat and corrective for democracy. democratization, 19(2), pp.184-208. doi: https://doi.org/10.1080/13510347.2011.572619 kazin, m. (1998). the populist persuasion: an american history. cornell university press. laclau, e. (2005). on populist reason. verso. marks, g. and steenerberg, m. (2002). understanding political contestation in the european union, comparative political studies, 35(8), pp.879–892. doi: https://doi.org/10.1177/001041402236297 milanovic, b. (2013). global income inequality in numbers: in history and now. global policy, 4(2), pp.198-208. doi: https://doi.org/10.1111/1758-5899.12032 mudde, c. (2013). ‘three decades of populist radical right parties in western europe: so what?’ european journal of political research 52: 1-19. doi: https://doi.org/10.1111/j.14756765.2012.02065.x mudde, c. (2007). populist radical right parties in europe. cambridge: cambridge university press. doi: https://doi.org/10.1017/cbo9780511492037 mudde, c. (2004). the populist zeitgeist. government and opposition, 39, pp.541–563. doi: https://doi.org/10.1111/j.1477-7053.2004.00135.x oliver, j.e. and rahn, w.m. (2016). rise of the trumpenvolk: populism in the 2016 election. the annals of the american academy of political and social science, 667(1), pp.189-206. doi: https://doi.org/10.1177/0002716216662639 rydgren, j. (2005). is extreme right-wing populism contagious? explaining the emergence of a new party family. european journal of political research, 44(3), pp.413-437. doi: https://doi.org/10.1111/j.1475-6765.2005.00233.x https://doi.org/10.2307/2669295 https://doi.org/10.1017/gov.2015.40 https://doi.org/10.1007/bf00115658 https://doi.org/10.1016/j.giq.2013.05.009 https://www.dropbox.com/s/aufvq80d5ymtakp/ipsa2016%20inglehart%20and%20norris%20populism%2019-7-2016.pdf?dl=0 https://www.dropbox.com/s/aufvq80d5ymtakp/ipsa2016%20inglehart%20and%20norris%20populism%2019-7-2016.pdf?dl=0 https://doi.org/10.2139/ssrn.2818659 https://doi.org/10.1080/13510347.2011.572619 https://doi.org/10.1177/001041402236297 https://doi.org/10.1111/1758-5899.12032 https://doi.org/10.1111/j.1475-6765.2012.02065.x https://doi.org/10.1111/j.1475-6765.2012.02065.x https://doi.org/10.1017/cbo9780511492037 https://doi.org/10.1111/j.1477-7053.2004.00135.x https://doi.org/10.1177/0002716216662639 https://doi.org/10.1111/j.1475-6765.2005.00233.x f. figueira / european journal of government and economics 7(2), 154-170. 170 rooduijn, m., de lange, s. l., & van der brug, w. (2014). a populist zeitgeist? programmatic contagion by populist parties in western europe, party politics, 20(4), pp.563-575. doi: https://doi.org/10.1177/1354068811436065 serricchio, f., tsakatika, m. and quaglia, l. (2013). euroscepticism and the global financial crisis. jcms: journal of common market studies, 51(1), pp.51-64. doi: https://doi.org/10.1111/j.14685965.2012.02299.x strom, k. (1990). a behavioural theory of competitive political parties. american journal of political science, 34(2), pp.565-598. doi: https://doi.org/10.2307/2111461 taggart, p. (2004). populism and representative politics in contemporary europe. journal of political ideologies, 9(3), pp.269-288. doi: https://doi.org/10.1080/1356931042000263528 taggart, p. (2002). “populism and the pathology of representative politics” in mény, y. and surel, y. (eds.), democracies and the populist challenge, pp.62–80. basingstoke: palgrave. doi: https://doi.org/10.1057/9781403920072_4 https://doi.org/10.1177/1354068811436065 https://doi.org/10.1111/j.1468-5965.2012.02299.x https://doi.org/10.1111/j.1468-5965.2012.02299.x https://doi.org/10.2307/2111461 https://doi.org/10.1080/1356931042000263528 https://doi.org/10.1057/9781403920072_4 ejge_front_7_2 ejge oficial 7-2 content page 7-2-4 abstract. the current surge of populism in europe and the united states calls for further analysis using public choice tools. in this article, populism is modelled as a deviation from the normal state of the median voter theorem. this study adds to th... keywords. populism; median voter theorem; public choice; ideologies; rational choice doi. https://doi.org/10.17979/ejge.2018.7.2.4423 references restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case vol.8 • no.2 2019 issn: 2254-7088 european journal of government and economics 8(2), december 2019. european journal of government and economics issn: 2254-7088 number 8, issue 2, december 2019 doi: https://doi.org/10.17979/ejge.2018.8.2 why regions fail (or succeed). the role of government institutions in the long-run 114-144 doi: https://doi.org/10.17979/ejge.2019.8.2.4989 filippo bonanno restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case 145-160 doi: https://10.17979/ejge.2019.8.2.5478 ricardo de moraes e soares and joão ricardo catarino fdi in selected developing countries: evidence from bundling and unbundling governance 161-188 doi: https://doi.org/10.17979/ejge.2019.8.2.4970 simplice asongu and, nicholas odhiambo the relationship between corporate tax rate and economic growth during the global financial crisis: evidence from a panel var governanceand domestic investment in africa 189-202 doi: https://10.17979/ejge.2019.8.2.5074 gamze oz-yalaman the impact of crd iv on bank lending 203-217 doi: https://doi.org/10.17979/ejge.2019.8.2.4656 matias huhtilainen https://doi.org/10.17979/ejge.2018.8.2 https://doi.org/10.17979/ejge.2019.8.2.4989 https://10.0.70.59/ejge.2019.8.2.5478 https://doi.org/10.17979/ejge.2019.8.2.4970 https://10.0.70.59/ejge.2019.8.2.5074 https://doi.org/10.17979/ejge.2019.8.2.4656 european journal of government and economics 8(2), december 2019, 145-160 european journal of government and economics issn: 2254-7088 restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model a study based on the portuguese case ricardo de moraes e soaresa* joão ricardo catarinob a institute of social and political sciences, university of lisbon, portugal b institute of social and political sciences, university of lisbon, portugal * corresponding author at: rsoares@iscsp.ulisboa.pt article history. received 5 july 2019; first revision required 8 october 2019; accepted 30 november 2019. abstract. the common vat system adopted by eu member states comprises a set of various rates, which differentiate the goods and services subject to this tax. however, from a technical and management point of view, it would be preferable to adopt a single tax rate as it reduces distortions and facilitates tax compliance and management. this research seeks to analyse the benefits and disadvantages of adopting a single vat as a means of simplifying the european vat model. it takes portugal as a case study. the main objective of this research is to contribute to the academic debate around the theme of the single rate of vat versus differentiated rates, through the achievement of a series of analyses and statistical tests to revenues and percentages of gdp that they correspond, in the three scenarios considered in the study: differentiated rates, single rate of 17% and single rate of 21%.in the empirical part, hypotheses were developed, the effects on tax revenue of a differential collection system were analyzed and compared to a possible single rate regime on consumption to verify which one would be more efficient. data comparison permits to verify that the estimated revenue of the single rate of vat is higher than the values obtained by the system of differentiated rates, in any of the proposed models (17% or 21%). the results of this research are valid for all countries that adopt vat or general tax transactions. keywords. consumer tax, fiscal policy, single vat rate, value-added tax, tax equity, tax system. jel codes. e62; h21; h25; h30; o23 doi. https://10.17979/ejge.2019.8.2.5478 1. introduction not only doctrine and tax authorities agree on the need to simplify tax systems, at the same time, increasing revenues1. that represents, besides, a form to develop the taxpayer’s warranties (catarino, 2009). one former finance minister of portugal has already admitted: "the tax structure needs to be simplified. the tax reform, which will begin to be debated in 2014, should take place in the vat, the largest source of current revenue: it gives more than 14 billion euros of annual revenue" (ribeiro, 2012). as a result, portugal has already reformed income 1 see the study on reduced vat rates applied to goods and services in the eu member states (organization for economic co-operation and development (2009). economic perspectives of latin america 2009. paris: oecd publishing). mailto:rsoares@iscsp.ulisboa.pt https://10.0.70.59/ejge.2019.8.2.5478 r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 146 taxes on legal entities, through law no. 2/2014 of january 16 (portugal, 2014a), and on the income of individuals, through law no. 82-e/2014, of december 31 (portugal, 2014b), with evident gains of simplification. the changes to vat, which is a european union common model tax basically had the objective of tax neutrality, the simplification of accessory operations as well as important changes at the level of the location of taxable transactions, largely adopted in the vat directive (european union, 2006/112/ce directive, november, 28). however, scarcely anything has been done the past years, to make the tax simpler, less costly and more accessible to taxpayers. the adoption of a single rate to tax all economic operations subject to vat would greatly simplify tax management firstly by clarifying the rates applicable to all transactions subject to tax. to the economist peter weiss, "portugal and other countries have exhausted the efficiency margin and must move towards a single vat rate" (ribeiro, 2012). and, according to imf economists ruud mooji and michael keen, "there is scope for more effective vat development as a tool for consolidation, especially where [...] the normal rate is already at such a high level that additional increases are problematic", therefore, there is a "strong argument in the advanced economies for a single vat rate" (ribeiro, 2012). in the green book about the future of vat, the european commission "questioned the effectiveness of this recovery method" and "in 2009 release a viability study on way to improve and simplify vat collection" (european union, 2010). nevertheless, the european commission in 2010, makes clear that it wants to make vat charging more efficient, to increase tax revenue within the common market, and at the same time, to design a vat system that is considerably simplest, to reduce “operating costs for taxpayers and for tax administrations by increasing net income" (european union, 2010). the main objective of the european commission in the area of indirect taxation is to implement a "vat system with a broad incidence base, preferably with a single rate" which may allow "to minimize costs" (european union, 2010). and, of course, increase revenue and respond, this way, to the challenge of international tax competitiveness (catarino, 2015). with regard to portugal, the efficiency2 of charging consumption tax in 2011 was only 51% (ribeiro, 2012). this can be justified by the inefficiency of tax charging, mainly due to the existence of a strong presence of the parallel economy and marked tax evasion (portugal, 2011). however, in the opinion of the european commission, the inefficiency of charging result from the absence of "modern technologies and/or financial intermediaries" (european union, 2010) with the ability to effectively charge vat. as an instrument to combat the parallel economy and tax evasion, the portuguese government, in the year 2012, conceived and implemented a series of fiscal measures which aim at promotion of efficiency in tax revenue collection. the principle tax measures of the strategic plan to combat tax and customs evasion and avoidance for the 3-year period 2012-2014, were aimed to combat tax evasion, 2 the efficiency of a tax is the relationship that is established between the taxable facts occurred and those that a given tax effectively taxes. the higher this ratio is, the more the tax is said to be effective. we assume, based on a. mitchell polinsky's teaching, that "efficiency" is the "relationship between the benefits and the aggregate costs of a situation." r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 147 fraud and the parallel economy. the government wanted to end tax evasion and reinforce sustainability of public finances (catarino et al., 2013). thus, a plan was approved for the first time, reinforcing operational means to monitor taxpayers' compliance with tax obligations. regarding vat, three major measures have been taken: 1. the compulsory use of billing programmes certified by tax authority for all economic operators. this measure was intended “to ensure greater transparency when paying for goods and services, particularly as regards the clearance and settlement of vat" (portugal, 2011). 2. the mandatory communication of the invoicing to tax and customs authority. it was established the obligation to communicate by computerised means in the portal e-invoice of all invoices issued by the economic agents. the reported data are crossed with the vat declarations issued by the companies, in order to "detect differences, situations of underinvoicing and parallel economy" (portugal, 2013). 3. the mandatory communication by electronic means of the transport documents issued to accompany the goods in circulation before the start of the transport operation. the regime of goods in circulation has been amended to establish new rules to ensure the integrity of transport documents and ensure more effective control of the documents by the tax and customs authority, hindering subsequent adulteration or concealment (catarino et al., 2015; gomes et al., 2013). the documents communicated can be consulted through the official einvoice website. however, despite the creation and implementation of a strategic plan against the parallel economy and tax evasion, no vat fund reform has yet been carried out. the aim of this study is to observe the possible adoption of a single vat rate by analysing statistical data on the evolution of revenue and the advantages/disadvantages of such a solution. 2. methodology as mentioned, the european commission argues that the single vat rate is the best economic option. however, for this research, it is important to verify the economic consequences of the application of a single vat rate in public revenue. this study uses the statistical data provided by the tax and customs authority of portugal3. however, it should be noted that there is no official statistical data that identifies the total amounts of vat revenue by rates, reduced, intermediate and normal in each year, being the figures available on the website of the finance portal corresponding to amounts accumulated. as a consequence, approximate consumption breakdowns of 29%, 11% and 60%, respectively, were applied to the calculation (tomaz, 2012). taking into consideration the mentioned above percentage values, the tax values produced for each of the differentiated rates were calculated. subsequently, these were compared to the estimated revenues resulting from two single rate scenarios: 17% and 21%. for each year considered, this estimate was obtained by means of the following expression: 3 available at http://info.portaldasfinancas.gov.pt/pt/dgci/divulgacao/estatisticas/estatisticas_iva. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 148 𝑉𝑉𝑉𝑉𝑉𝑉 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆𝑆 𝑅𝑅𝑅𝑅𝑅𝑅𝑆𝑆 = 𝑐𝑐×𝑏𝑏𝑟𝑟 𝑅𝑅𝑟𝑟 + 𝑐𝑐×𝑏𝑏𝑖𝑖 𝑅𝑅𝑖𝑖 + 𝑐𝑐×𝑏𝑏𝑛𝑛 𝑅𝑅𝑛𝑛 [1] where “c” corresponds to the proposed single rate, br, bi and bn correspond to the total revenues from differentiated rates (reduced, intermediate and normal), and ar, ai and an correspond to the differentiated rates of vat (reduced, intermediate and normal, also respectively). the main objective of this research is to contribute to the academic debate around the theme of the single rate of vat versus differentiated rates, through the achievement of a series of analyses and statistical tests to revenues and percentages of gdp that they correspond, in the three scenarios considered in the study: differentiated rates, single rate of 17% and single rate of 21%. data on 18 years of revenue and its share of gdp were used, corresponding to the period 1996 to 2013, for each of the three scenarios considered. in order to make a comparison between the models, the values of the differentiated rates are insufficient since it is necessary to estimate the possible revenue values that the state could have obtained had he chosen for the single rate model, and then, comparisons between actual and estimated values were made. to test the presence, or not, of equality of revenue averages and their respective percentages of gdp in the three scenarios under study, the anova test (fisher, 1918) was used. this test of variance analysis was preceded by the validation of suppositions required for its application: normality and homoscedasticity of revenue distribution and percentages of gdp in each of the scenarios, respectively, using the shapiro-wilk normality test (1965) and the levene test (1960). subsequently, scheffé's test (1959) was used to determine if the average levels of revenue and percentage of gdp of the various scenarios which may or may not be considered all different. the significance level used was, in all statistical tests, 0,05 (5,00%). 3. results based upon the data on vat revenue and the methodology described, it is possible to find in table 1 the amounts of tax revenue actually collected for each of the differentiated rates, as well as estimated revenues at a single rate of 17% and 21%, between the years 2000 and 2017. data in table 1 allows concluding that there is growth in vat revenue between the years 2000 and 2003 and between the years 2004 and 2008. the value of the tax collected reduce in the years 2004 and 2009, compared to the years 2003 and 2008, respectively. from that moment on the revenue remained, approximately, the same as in 2008. the data show that the peak in vat revenue collection occurred in 2017, reaching a total value of 16.001,40 m€. however, the maximum amount of revenue collected as a percentage of gdp took place in 2012, in value 8,28%. but, as mentioned above, in order to make a comparison between the models, the values of the differentiated rates are insufficient since it is necessary to estimate the possible revenue values that the state could have obtained had he chosen for the single rate r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 149 model, and then, comparisons between actual and estimated values were made. in order to compare the calculated values and the actual values, the vat revenue was estimated based first on a single general rate of 17% and later, a single rate of 21%. figure 1 allows immediate comparison of the estimated surplus value for single rates of 17% and 21%, in light of the differentiated rates model. as can be seen, the annual values of the estimated revenue, applying single rates of 17% and 21%, exceeded, in each year, the values of the effective revenue. therefore, it seems to be evidence of a greater capacity for revenue collection, i.e., evidence of potential4 revenue that the state would have raised if he had if already opted for a single vat rate system. figure 1. effective vat revenue with differentiated rates and estimated vat revenues with a single rate of 17% and 21% (in m€). source: own elaboration. by keeping the system of differentiated rates in the vat instead of the single rate model, the opportunity cost associated is the loss of collection efficiency of tax revenue. in this case, the state loses net value of revenue by the option of a model based on several tax rates aiming to alleviate tax burden on basic consumptions (with a lower tax rate), to the detriment of a model strictly based on the tax revenue. it should be noted that one of the underlying assumptions of this study is that single rate models of 17% and 21% cannot cause revenue losses for the state coffers. these models should, at least, ensure the amount of revenue that the state would collect with the application of differentiated rates. the implicit idea is based on the search for simplification of the system of taxation on consumption in the european union and portugal, without causing revenue losses to the public coffers (oliveira, 2010). 4 note: potential revenues are the difference between the estimated total value and current revenues. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 150 table 1. effective vat revenue with differentiated rates and estimated vat revenues with single rates of 17% and 21%. effective revenue with differentiated rates estimated revenue with a single rate of 17% estimated revenue with a single rate of 21% year gdp (m€) reduction rate (29%) (m€) intermediate rate (11%) (m€) normal rate (60%) (m€) vat revenue (m€) % do gdp vat revenue (m€) % do gdp vat revenue (m€) % do gdp 2000 128.466,30 2.515,10 954,00 5.203,66 8.672,77 6,75 15.106,52 11,87 18.661,00 14,53 2001 135.827,52 2.600,14 986,26 5.379,60 8.965,99 6,60 15.050,99 11,08 18.592,40 13,69 2002 143.631,42 2.887,40 1.095,22 5.973,93 9.956,56 6,93 16.713,82 11,64 20.646,49 14,37 2003 146.158,28 3.062,99 1.161,82 6.337,21 10.562,02 7,23 17.730,20 12,13 21.902,01 14,99 2004 152.371,56 2.998,81 1.137,48 6.204,43 10.340,72 6,79 17.358,71 11,39 21.443,12 14,07 2005 158.652,59 3.384,76 1.283,87 7.002,94 11.671,57 7,36 18.996,04 11,97 23.465,70 14,79 2006 166.248,72 3.596,31 1.364,12 7.440,64 12.401,07 7,46 20.183,33 12,14 24.932,35 15,00 2007 175.467,72 3.826,96 1.451,60 7.917,84 13.196,40 7,52 21.477,77 12,24 26.531,36 15,12 2008 178.872,58 3.893,98 1.477,03 8.056,50 13.427,50 7,51 22.179,99 12,40 27.398,81 15,32 2009 175.448,19 3.156,19 1.197,17 6.530,04 10.883,40 6,20 17.977,56 10,25 22.207,58 12,66 2010 179.929,81 3.527,10 1.337,86 7.297,44 12.162,40 6,76 17.945,78 9,97 22.168,31 12,32 2011 176.166,58 4.128,06 1.565,82 8.540,82 14.234,70 8,08 20.056,57 11,39 24.775,76 14,06 2012 168.397,97 4.045,50 1.534,50 8.370,00 13.950,01 8,28 19.655,43 11,67 24.280,24 14,42 2013 170.269,33 3.842,24 1.457,40 7.946,46 13.249,10 7,78 18.667,86 10,96 23.060,30 13,54 2014 173.079,06 4.006,09 1.519,55 8.288,46 13.814,10 7,98 19.463,94 11,25 24.043,70 13,89 2015 179.809,06 4.304,85 1.632,87 8.906,58 14.844,30 8,26 20.915,49 11,63 25.836,78 14,37 2016 186.480,45 4.373,93 1.659,08 9.049,50 15.082,50 8,09 21.251,11 11,40 26.251,37 14,08 2017 194.613,47 4.640,41 1.760,15 9.600,84 16.001,40 8,22 22.545,83 11,58 27.850,73 14,31 accumulated value ------223.416,51 --343.276,96 --424.048,01 -- difference in the revenue: estimated at a single rate of 17% and effective at differentiated rates 119.860,45 m€ +53,65% difference in the revenue: estimated at a single rate of 21% and effective at differentiated rates 200.631,50 m€ +89,80% difference in the revenue: estimated at a single rate of 21% and estimated at a single rate of 17% 80.771,05 m€ +67,39% source: own elaboration, based on statistics from pordata. european journal of government and economics 8(2), december 2019, 145-160 in this respect, it should be noted that the management of vat rates by the state always presupposes a planned action in order to prevent the possible risks and deviations that may affect the balance and consolidation of public accounts. the fiscal planning of revenues throughout the year is fundamental in the formulation and implementation of fiscal policies. plan previously the effects of the application of a uniform rate makes its execution more precise and simple for the tax and customs authority, allowing the identification of the resources necessary for effective control and evaluation of the desired results (pereira, 2003). the surplus values obtained through the proposed models can be confirmed by checking the variation of the revenue values at the bottom of table 1. any of the single rate options obtains higher revenues than the system of differentiated rates along the 18 years under analysis: a difference in order of 119.860,45 m€ (or + 53,65%) in the single rate model with vat at 17% and a difference in the order of 200.631,50 m€ (or + 89,80%, almost twice the value) in the single rate model with vat at 21%. the application of a single rate of 21% allows for a 67,39% increase in revenue compared to the 17% rate, corresponding to a total value of over 80.771,05 m€ along the 18 years. the question, therefore, arises as to whether the differences in revenue and the respective percentages of gdp for the three scenarios under study are statistically significant or whether they are purely random. to do this, and checking the applicability, through the assumptions of normality and homoscedasticity of each of the data sets, we used the anova test (fisher, 1918)5. table 2 shows that for both revenue amounts and percentages of gdp, the p-value observed in the anova test (fisher, 1918) was equal to 0,000 (less than 0,05), implying the rejection of the null hypotheses and the acceptance of the alternative hypotheses according to which the average values of the revenue amounts and the average values of the percentages of gdp for the three scenarios considered are not all the same. table 2. anova test results. sum of the squares degrees of freedom average squares f statistic p value revenue betweengroup 1132286800,39 2 566143400,19 100,05 0,000 intra-group 288601473,73 51 5658852,43 ---- total 1420888274,12 53 ------ % gdp betweengroup 417,12 2 208,56 428,70 0,000 intra-group 24,81 51 ---- total 441,93 53 ------ source: own elaboration. 5 p-value of the shapiro-wilk normality tests (shapiro and wilk, 1965) greater than 0.05 in all data groups, denoting that each data group follows a normal distribution (p-value = 0.200 in all cases). p-value of the levene homoscedasticity test (levene, 1960) greater than 0.05 in all data groups, denoting that in each data group there is homogeneity of the variances (p-value = 0.230 for data on vat and p-value = 0.651 for data on the percentage of gdp). r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 152 table 3. results of the scheffé test. source: own elaboration. note: the size of the categories is equal. the harmonic mean of the category size (18,000) is used. models sample size subgroup for alpha = 0,05 1 2 3 revenue differentiated rates 18 70,521 ---- single rate of 17% 18 --197,591 -- single rate of 21% 18 ----32,024 scheffé critical point --6,358 6,358 6,358 % gdp differentiated rates 18 304,758 ---- single rate of 17% 18 --135,260 -- single rate of 21% 18 ----3728,112 scheffé critical point --6,358 6,358 6,358 the subsequent application of scheffé's (1959) test allowed us to conclude statistically that not only the averages of revenue and the percentage of gdp are not all the same, but are indeed all different among the three scenarios investigated (table 3). at the revenue level, the average annual rate for the differentiated rate model is around €12,412.03 million, for the single rate model of 17% is around €19 070,94 million and for the rate model of 21% is around €23 558,22 million euros. at the level of the percentage that this revenue represents in the gdp of each year, the annual average for the model of differentiated rates the average is about 7,43% of gdp, for the single rate model of 17% is about 11,40% of gdp and for the single rate model of 21% is about 14,20% of gdp. consequently, from the comparison of the data, it appears that the estimated revenue from the single rate of vat is higher than the values obtained by the differentiated rate system, in any of the proposed models (17% or 21%), in accordance with suggested by the doctrine. the same is to say that both rates are fiscally practicable. 4. discussion the simplification of portuguese vat system would allow greater efficiency gains, higher revenue levels, greater simplification of incidence and collection rules, and avoiding fraud resulting from the differentiation of tax (rates). the same seems to be possible as regards the common vat model in use in the eu. to these essential aspects added the idea that one must take into account not only what "is spent on taxes, should keep in mind what is spent, as well, to pay them" (nabais, 2010)6. the vat should generically be efficient but also the least onerous possible for companies. this point of view is supported by recent studies, which advocate the application of a single rate would enable to produce higher income levels and reasonably reduce the management fees by the tax and customs authority, as well as business costs7. we could, therefore, say that vat 6 the author refers: "above all, to simplify the companies’ taxation, so that they pay less and in an easier way, therefore economic competition, which is developed on a global scale, does not tolerate with ultracomplex systems as they are today." 7 see, for example: sweden (2006). compliance costs of value added tax in sweden report 2006: 3b. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 153 simplification brings a competitive advantage to the tax, i.e., the single tax on the goods and services provided is neutral, eliminating distortions in competition between equivalent goods or services neutrality is the characteristic of tax whose operation does not promote changes in the behaviour of taxpayers subject to it (basto, 1991)8. another clear advantage of the single rate in portugal as well as in the eu is to reduce reducing management and tax collection costs by the easiest application of the (single) rate to all economic operations. with the reduction of management and collection costs, the tax and customs authority could reallocate part of its human capital to other more deficit areas. simultaneously, companies would no longer devote about 30 per cent of their administrative staff to fiscal requirements (albuquerque, 1991). on the other hand, the final consumer would also take advantage of this model. the first advantage, by reducing tax burden on the final price of goods and services, would lead to an increase in demand without adversely affecting public revenue. second is due to a greater consumption-incentive. with the decrease of administrative burden and tax management, the government can displace financial resources for new public investments and enhance economic growth. the persons liable for vat they can allocate resources to new investments, modernise their production system and expand the marketing of its products through resources spent in fulfilment of their tax obligations. therefore, it becomes clear that there are many associated advantages to the adoption of a single vat rate for all involved, with obvious gains of an economic nature (siqueira, 2001). clotilde palma (2011) of the opinion against a possible application of a uniform rate in portugal with the argument that "the pernicious regressive effects of the tax, which weighs more on classes with the lowest incomes, are well known". the weight of iva is higher for low-income families, as they cannot make replacement consumptions nor decide not to consume as they mainly consume essential goods and services. this author also expresses against the "generalized mechanism of reverse charge, which would misrepresent the characteristics of the tax" (palma, 2012). however, it also recognizes the complexities of multiplicity of taxes, by noting that differentiated rates originate different readings and that the understanding given “to certain concepts vary the scope of tax, causing significant distortions of competition, namely in cross-border situations [...] at consumption of food and fuel "(palma, 2012). in strictly academic terms, the single rate model has some disadvantages. fernando albino (1991) mention that "the ideal of a single tax is twofold [...] it represents a 25 years setback, to the times of the stamp duty [...] the opposition to such a tax [...] was that he did not know the economic reality underlying the transaction”. this author raises the question whether the application of the single consumption tax model takes into account the economic reality of the skatteverket: national tax board. on the other hand, see the study carried out by gastaldi f, liberati p, pisano, e, tedeschi, s (2017). regressivity reduction of vat reform. international journal of microsimulation, 10(1): 29-72. 8 as xavier de basto teaches, “with vat, in fact, the tax content of goods depends neither on the extent of the production processes in which they were obtained nor on the distribution of value added by the different operators. there is no encouragement or discouragement of vertical integration.” this tax does not produce any cumulative effect, but gives neutrality to the different transactions occurring in that cycle without being distorted. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 154 transactions carried out by agents. it should be noted, however, that vat is not a personal tax such as the irs, but rather a real tax that charges the whole demonstration of wealth by consumption, regardless of the personal characteristics of those who support it. another disadvantage of the single rate model is the "constitutional status" of the tax (albuquerque, 2003). albuquerque (2010) points out that some are "against the search for solutions that walk towards perfecting the tax system", there is a clear preference to keep everything as it is with the fear of losing revenue or withdraw the reduced rate status of certain goods and services. further then the status question against tax there is another disadvantage linked with economic theory, which consists in the absence of stimulation to savings on the part of consumer (braz e cunha, 2009). the reduction of the tax from 23% to a lower rate does not lead the consumer to feel the need to save since in principle, will be able to consume more goods, satisfying other needs and desires. according to conceição nunes and pedro nunes (2007), which used the example of the estonian tax reform, the impact assessment of proportional taxes on tax revenues (irs and vat) allows us to conclude that "the reduction of taxes has generated an inflation process, encouraging consumption in detriment of investment" (castro et al., 2011). apart of the disadvantages of the single rate tax model, we have the disadvantages pointed out by the tax theory: if there is a higher revenue collection efficiency, an excess of human capital can be generated and allocated to the management of the tax, which in turn may lead to redundancies, either in the tax and customs authority, or in taxable persons. however, the major drawback pointed out to this model of indirect taxation on consumption is related to the social impact of the measure. indeed, the tax systems have been called for redistributive role, variable and often postponed role, especially in times of financial crisis. the amendment or repeal of reduced and intermediate vat rates leads to greater inequality in the burden of tax on lower-income classes, as they spend a proportionately larger slice of their income on essentials goods. nevertheless, the genesis of the single rate of consumption model overlooks the social problems. vat is, in a way, unrelated to that phenomenon, because it taxes the manifestations of wealth through the consumption, regardless of the economic capacity of the consumer. the social costs may be compensated with corrective measures from the regressive effects of consumption taxation on the irs. therefore, while these income inequalities between individuals are understandable, they must be addressed through social policies or within the scope of the individual income tax and not necessarily within the scope of vat. the economist arthur laffer conceived in 1974, a curve to explain the elasticity of tax revenue, and how tax rates influence the amount of revenue. according to the author, when the tax rate is too high, a decrease in the rate leads to an increase in tax collection. the curve thereby demonstrated how tax revenue values progress as the state raises the rate, independently of the type of tax (carvalho, 2005). according to laffer, when the tax onus is low, the ratio between the tax and the tax revenue is direct, i.e. each additional unit of percentage increase in tax burden will provide a rise in total tax revenue. however, as successive increases r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 155 in the rate occur, it will increasingly approach the maximum point of tax collection and the curve undergoes circumflexion. if the tax continues to rise, there is a reversal in the evolution of revenue for each additional unit of rate increase. laffer defends that the phenomenon of curve inflexion has as cause the fact that high tax rates result in tax evasion behaviour and incentive to parallel economies, which causes successive increments of discouragement in the formal economy, decreasing in this way, the base of tax incidence (lima et al., 2006). for laffer, the main objective of the curve was to incite to a "critical reflection on distortions and the loss of incentives that tax rates create in the economy, and these rates, as they are increased, can reduce tax revenue" (lima et al., 2006) the curve has proved the existence of a limit for the collection of tax revenue, also depict, through a simple design chart, the possible consequences of fiscal policies on economic growth. the high levels of tax on goods and services tend to discourage the consumption activity of economic agents, and the impacts of changes in rates on tax revenues are strongly influenced by the effects of substitution and income, as a result in the behavioural adaptation of the subject to the taxation policies (nunes, 2007). the laffer effect depends on the variation of the taxable income declared by the taxpayers and the level of the tax rate, being that, as that rate increases, individuals tend to substitute the goods and services heavily taxed with those which taxation is more favourable (substitution effect). however, a decrease in disposable income will cause an increase in labour supply in the subjects, in order to achieve their initial disposable income levels (income effect). it should also take into consideration that in laffer's effect the impact of taxation on consumer choice is an important factor, because the existence of worsening vat rates has direct consequences not only in the short term but also in the long term, mainly in the reduction of saving levels. the decline in savings will affect the available income of the economy and, consequently, will have impacts on tax revenues (nunes, 2007). this effect, however, is not direct and immediate, i.e. it is not enough to reduce the tax burden for revenue to increase, and complementary growth policies are also required, such as stimulus to consumption and labour. nevertheless, as a study tool and analysis of the evolution of tax revenue, the laffer curve is a useful tool. as it is well known, the laffer curve presupposes the existence of an inversion point, from which any increase in the tax rate by effect decrease in revenue: the maximum tax efficiency of the tribute. from this point, any additional unit of rate increase causes a decrease in total revenue. if the marginal rate is higher than the curve maximum, there is inefficiency in revenue collection. the way to turn the situation around, according to laffer, is to reduce the marginal rate, in order to stimulate labour supply and create an increase in production. on the other hand, if the marginal rate is inferior to the maximum curve, there is also fiscal inefficiency, in which case, can be enough to increase the tax rate to generate an increase in total revenue. it should be noted that for a zero rate the state obtains zero revenue, and at the opposite end, where the rate is 100%, revenue is also zero because individuals are not willing to purchase goods or services in which the value of vat is equal to the price. note that when r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 156 rates are too high, the natural behaviour of individuals alters in the search for informal markets, since rates are the discouraging element of economic activity. therefore, beyond the point of fiscal efficiency, fiscal revenue decreases and, even if rates increase, revenue continues to decline. at the limit, when a rate of 100% is applied, economic activity ceases to generate revenue due to the fact that producers are unable to dispose of their goods and services since consumers are no longer willing to buy those goods or services, because they believe the final price of the goods too high. in terms of effectiveness of the proposed models, it is important to compare their fiscal efficiency with the tax efficiency of the uniform rate. figure 2 represents the annual vat revenue values of the three models. revenue under the differentiated rate regime is real and estimated for single rate models. the laffer curve constructed from the effective values of vat revenue in the years 2000 to 2017 shows its peak of maximum efficiency in the year 2017. that is, the year 2017 was the year in which the current model was able to generate a greater level of tax revenue. figure 2 reveals that 2009 was the worst year for vat revenues and that the years 2012 and 2013 are points of fiscal inefficiency because the increase in the tax rate generated a decrease in total revenue. this means that any additional unit increase to the tax rate will generate revenue losses for the state coffers. figure 2. laffer curve for effective vat revenue with differentiated rates and for estimated vat revenues with single rates of 17% and 21% (in m€). source: own elaboration the decrease in revenue by lowering the maximum vat rate from 21% to 20% in the differentiated rate model, in 2009, indicates that the fiscal efficiency point had not yet been reached and the reduction was, in fact, counterproductive to revenue. in turn, the laffer curves for estimated revenue amounts at 17% and 21% suggest that the peak of fiscal efficiency would be reached in the year 2008. this means that uniform rates would have reached their maximum value of revenue three years earlier than the model of differentiated rates. this anticipation of three years in achieving maximum fiscal efficiency could have been decisive for the way in which portugal handled the economic and financial crisis of recent years. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 157 5. conclusions in the empirical part of this study, we performed tests on the general tax model on single-rate consumption, and simulated hypotheses were placed use official statistical data, available by the tax and customs authority of portugal, during the analysed period. the importance of official data for the confirmation and/or refutation of models and hypotheses have been demonstrated. the proposal to adopt a single vat rate, with the consequent elimination of reduced and intermediate rates, constitutes an important transformation of the tax, generating significant impacts on revenue level, consumption and the social fabric. this hypothesis is confirmed in several scientific studies, referred to in this research, as a technically possible solution, able to achieve further efficiency gains in tax management and from the point of view of public revenue, necessary for public policy financing. some other studies conclude in the opposite direction. thus, the following conclusions are drawn from the research: 1. although studies have pointed out that a simpler vat model would be more efficient combating fraud, tax evasion and the parallel economy, in practice no fundamental reform was carried out with a view to simplifying the tax, in particular through single rate adoption. as this system leads to higher revenues and it is simpler, the obstacles to its implementation seem to be social and political in nature. 2. although all the member states of the european union adopt multiple vat rates, the european commission advocates the application of a single rate for an economic (increase in revenue) and an administrative reason (simplification of incidence of tax and its collection), emphasising that the biggest obstacle to the application of a single rate within the common market is political. this problem also seems to affect the portuguese model. 3. the adoption of a single rate for vat is technically feasible and has considerable financial and administrative advantages for all agents involved (state and economic operators). 4. non-adoption of the single rate model generates a loss of tax revenue. besides the economic advantages resulting from this model, there are also advantages related to the simplification of the tax system in portugal, as well as in the eu. the tests to the proposed models allowed concluding statistically that these would provide for higher tax revenue collection with greater ease of administration and compliance (gains of efficiency). 5. since the single rate system cannot lead to revenue losses or, at least, it should be able to maintain revenue levels, otherwise it will not be feasible in the light of the principles of prudence and financial sustainability, which is essential for consolidation of public accounts, the choice of 21% vat rate seems to be more appropriate (in line with that advocated by catarino and fonseca (2013)). 6. the data suggest that the proposed single rate model of 21% can generate more revenue, about 89,80% above that obtained with the differentiated rates of vat adopted in portugal. testing for a single rate model of 17% shows that this rate would be sufficient not only to guarantee revenue levels but also to increase it by 50% to 60% above the differentiated rate system. the rate of 21% offers higher revenue levels, 67,39% higher than the rate of 17%. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 158 considering that portugal faces excessive levels of public debt, with an increase in financing costs in the financial markets, it is necessary to achieve a considerable reduction of the debt burden in gdp and create financial surpluses to handle future shocks (andrade, 2012, p. 138). evidence suggests that a single vat rate seems to be a possible solution to address these needs partly. the effective adoption of a model endowed with one of the rates used in this study is, however, an economic and fiscal policy issue, and it is up to the government to choose the level of tax rates that it considers to be the most appropriate, with due respect for the principle of legality, fiscal transparency, contributory capacity and relative equality (gomes, 2008). if the vat model adopted in portugal is simpler and clearer, compliance and management costs are lower, when comparing to any other eu country applying several tax rates. thus, the possible application of a single vat rate would simplify the process of applying the tax, settlement and collection laws (gomes, 2008). the choice of a rate of 21%, 17% or another rate should take into account the likely risks of deviations in revenue that may result from the application of the proposed model. the single rate of vat can generate negative social effects not insignificant, due to the known propensity for the regressivity of the tax with respect to the taxpayers of smaller incomes, since the tax of the consumption is proportionally heavier for these. these effects will have to be studied in autonomous and future research work. it is also suggested that similar studies be replicated, for an equivalent period of time, in european countries and in countries with historical and cultural proximity to portugal, as is the case in latin american countries. acknowledgement this work is financed by national funds through the fct foundation for science and technology, i.p., under project uid / cpo / 00713/2019. references albino, f. (1991). imposto único. in albuquerque, m. (eds.), imposto único sobre transações (prós e contras) (pp. 32-34). são paulo: câmara brasileira do livro. albuquerque, m. (1991). por uma revolução tributária. in albuquerque, m. (eds.), imposto único sobre transações (prós e contras) (pp. 18-19). são paulo: câmara brasileira do livro. albuquerque, m. (2003). a verdade sobre o imposto único: eficiência, comodismo, sonegação, obsolescência, interesses corporativos, cumulatividade, fatos e mitos da reforma tributária no brasil. brasil: lcte editora. albuquerque, m. (2010). tributação no brasil e o imposto único. são paulo: mcgraw-hill. andrade, f. (2012). a limitação constitucional do défice orçamental e sua circunstância. studia juridica 107, 125-152. basto, j. g. x. (1991). a tributação do consumo e s sua coordenação internacional, centro de estudos fiscais,. ministério das finanças: lisboa. pp. 52-53. braz, c., cunha, j. c. (2009) os efeitos redistributivos do iva em portugal. boletim económico. banco de portugal. inverno 2009. carvalho, r. (2005). compreender áfrica: teorias e práticas de gestão. brasil: editora fgv. r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 159 castro, c., nunes, p. (2011). pedro. o impacto da introdução de impostos proporcionais nas receitas fiscais e performance económica: as experiências de novos estados membros da ue27. estado, gobierno, gestión pública 18, 5-35. catarino, j. (2009). para uma teoria política do tributo. cadernos de ciência e técnica fiscal 184. ministério das finanças: lisboa. catarino, j. (2019). finanças públicas e direito financeiro (5th edition). coimbra: almedina. catarino, j., fonseca, j. (2013). sustentabilidade financeira e orçamental em contexto de crise global numa europa de moeda única. revista sequência 34(67), 21-51. catarino, j., guimarães, v. (2015). lições de fiscalidade, vol. 2, gestão e planeamento fiscal internacional. coimbra: almedina. european commission taxation & customs union (2007). study on reduced vat applied to goods and services in the member states of the european union. danmark: copenhagen economics. european union. (2006). diretiva 2006/112/ce do conselho, de 28 de novembro de 2006, relativa ao sistema comum do imposto sobre o valor acrescentado. retrieved from https://eur-lex.europa.eu/legal-content/pt/txt/?uri=celex%3a32006l0112 european union. (2010). livro verde sobre o futuro do iva – rumo a um sistema de iva mais simples, mais sólido e eficaz {sec(2010) 1455 final} / com (2010) 695 final. retrieved from https://eur-lex.europa.eu/lexuriserv/lexuriserv.do?uri=com:2010:0695:fin:pt:pdf fisher, r. (1918). the correlation between relatives on the supposition of mendelian inheritance. philosophical transactions of the royal society of edinburgh 52(2), 399-433. doi: https://doi.org/10.1017/s0080456800012163 gastaldi, f., liberati, p., pisanom, e. & tedeschi, s. (2017). regressivity reduction of vat reform. international journal of microsimulation 10, 29-72. gomes, j. (2008). a simplicidade e a instabilidade no sistema fiscal português – prioridade da política fiscal. revisores e auditores 40, 55-62. gomes, j. & pires, j. (2013). novas regras aplicáveis aos bens em circulação. boletim do contribuinte, revista de informação fiscal 81(10), 360-364. levene, h. (1960). robust tests for equality of variances. in olkin, i., ghurye, s., hoeffding, w., madow, w., mann, h. (eds.). contributions to probability and statistics: essays in honor of harold hotelling. stanford: stanford university press. lima, e. & rezende, a. (2006). um estudo sobre a evolução da carga tributária no brasil: uma análise a partir da curva de laffer. encontro de administração pública e governança, são paulo: anpad. nabais, j. (2010). reflexões sobre quem paga a conta do estado social. revista da faculdade de direito da universidade do porto 7, 51-83. nunes, m. (2007). experiências supply-siders na ue15: a redução de impostos como política de melhoria da posição orçamental intertemporal. revista de estudos politécnicos 5(8), 113-138. organisation for economic co-operation and development. (2009). perspetivas económicas de américa latina 2009. paris: oecd publishing. oliveira, a. (2010). iva – imposto sobre o valor acrescentado: um imposto neutro. porto: universidade do porto. palma, c. (2011). o orçamento do estado para 2012: a proposta de orçamento do estado para 2012 e as taxas do iva. revista de finanças públicas e direito fiscal 4(3), 77-98. palma, c. (2012). a recente comunicação da comissão sobre o futuro do iva. revista toc 12(144), 48-55. pereira, j. (2003). finanças públicas: a política orçamental no brasil. são paulo: atlas. ministério das finanças de portugal. (2011). plano estratégico de combate à fraude e evasão fiscais e aduaneiras, 2012-2014. retrieved from http://apit.pt/public/scaffold_documentos/pedcafeefea_2012-2014.pdf governo de portugal. (2013). relatório do grupo de trabalho interministerial sobre a avaliação da situação económico-financeira específica dos custos de contexto dos sectores da hotelaria, restauração e similares. retrieved from http://issuu.com/psdleiria/docs/relat__rio_situa____o_econ__mico-fi https://eur-lex.europa.eu/legal-content/pt/txt/?uri=celex%3a32006l0112 https://eur-lex.europa.eu/lexuriserv/lexuriserv.do?uri=com:2010:0695:fin:pt:pdf https://doi.org/10.1017/s0080456800012163 http://issuu.com/psdleiria/docs/relat__rio_situa____o_econ__mico-fi r. de moraes e soares et al. / european journal of government and economics 8(2), december 2019, 145-160 160 polinsky, a. mitchell, (1989) an introduction to law and economics. aspen: aspen law & business. portugal. (2014a). lei n.º 2/2014, de 16 de janeiro. retrieved from https://dre.pt/pesquisa//search/571007/details/maximized portugal. (2014b). lei n.º 82-e/2014, de 31 de dezembro. retrieved from https://dre.pt/home//dre/66022085/details/maximized?p_auth=hg07h3xg ribeiro, l. (2012). peritos do fmi e de bruxelas querem que portugal caminhe para iva único. retrieved from http://www.clipquick.com/files/imprensa/2012/0414/0/1_1801432_c36f3ed054e26786605ce3d62abca4fc.pdf scheffé, h. (1959). the analysis of variance. new york: wiley. shapiro, s., & wilk, m. (1965). an analysis of variance test for normality (complete samples). biometrika 52(3-4), 591-611. doi: https://doi.org/10.1093/biomet/52.3-4.591 siqueira, r. (2001). iva uniforme com renda básica: uma proposta de reforma da política tributária e social no brasil. instituto brasileiro de economia. retrieved from http://www.anpec.org.br/encontro2001/artigos/200102109.pdf sweden. (2006). compliance costs of value-added tax in sweden – report 2006:3b. retrieved from http://www.skatteverket.se/download/18.906b37c10bd295ff4880002550/rapport200603b tomz, j. j. a. (2012). uma taxa única para o iva em portugal. revista de finanças públicas e direito fiscal 3(2), 19-29. https://dre.pt/pesquisa/-/search/571007/details/maximized https://dre.pt/pesquisa/-/search/571007/details/maximized https://dre.pt/home/-/dre/66022085/details/maximized?p_auth=hg07h3xg https://dre.pt/home/-/dre/66022085/details/maximized?p_auth=hg07h3xg http://www.clipquick.com/files/imprensa/2012/04-14/0/1_1801432_c36f3ed054e26786605ce3d62abca4fc.pdf http://www.clipquick.com/files/imprensa/2012/04-14/0/1_1801432_c36f3ed054e26786605ce3d62abca4fc.pdf https://doi.org/10.1093/biomet/52.3-4.591 http://www.anpec.org.br/encontro2001/artigos/200102109.pdf http://www.skatteverket.se/download/18.906b37c10bd295ff4880002550/rapport200603b ejge oficial 8-2 contents-8-2 number 8, issue 2, december 2019 8-2-2 vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 181-199 european journal of government and economics issn: 2254-7088 analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain marcos pradas garcía a*, maría josé maciá andreu b, marta garcía-tascón c, ana maría gallardo guerrero b a universidad de sevilla, spain b universidad católica de murcia, spain c universidad pablo olavide, spain * corresponding author at: marcos.pradas@gmail.com article history. received 13 december 2019; first revision required 27 january 2020; accepted 20 february 2020. abstract. spain is a european leader as a golf tourism destination, and andalusia is the region that receives the most tourism in this sector, boosting not only the golf industry but also the percentage of income overall. thus, user loyalty and knowing the future intentions of users is a matter of vital importance in these sports organizations. this study analyses 636 users of 17 golf courses in andalusia –73.43% men and 26.42% women– and with an average age of 50.2 ± 15.6. the results show that more than 70% of users would encourage their families to play on the golf course and recommend it in more than 75% of them. in conclusion, this study emphasises the need for the use of the adapted tool, as it is a valid and reliable instrument that guides on the aspects demanded by the user as well as how to build loyalty. keywords. sports management, loyalty, golf courses jel codes. d7, d71, d9, d91, l83, z2, z21 doi. https://doi.org/10.17979/ejge.2020.9.2.5841 1. introduction a million golf tourists have visited spain since 2012, according to the international association of golf tour operators (iagto). this association ranks spain second most popular in the world as a destination for foreign tourists whose travel is motivated by golf. golf tourism generates an expenditure of around 1.2 billion euros (iagto, 2013, 2019). spanish golf federation (rfeg) said that spain is considered a leading country in golf growth, and has 271,170 players as of january 1st, 2019 (rfeg, 2019). spain is the first choice of golf destination for 29% of european tourists, with british tourists being the largest group (kpmg, 2017). favourable weather conditions and the lack of geopolitical problems compared to direct competitors such as north africa, egypt and turkey are some of the factors that contribute to this prominence (pinero, 2017); which has led to a rapid increase in this type of facility, not without creating a broad social debate about the environmental impacts involved (vargas-sánchez and riquel-ligero, 2015). golf is an economic engine that also drives other industries (barciela, 2017), such as tourism. according to aymerich and anabitarte (2016), 41% of the rounds played in spanish mailto:marcos.pradas@gmail.com https://doi.org/10.17979/ejge.2020.9.2.5841 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 182 golf courses are by foreign players. according to a study carried out by global business partners in 2016, golf generates more than 2 billion euros per year in revenue in spain, as well as 11,000 jobs and 1.1 million tourists. andalusia has become the most important destination for golf tourism in spain, with exponential growth in the number of golf courses that complete the tourism and leisure offer (riquel-ligero and vargas-sánchez, 2012). currently, there are 349 golf courses in spain, of which 93 (26.65%) are located in andalusia (rfag, 2019). although golf tourism generates significant profits, the golf courses are frequently questioned about their close relationship with the environment and, consequently, their possible adverse impact on it. the golf courses have adopted an environmental regulation more rigorous in recent years by the andalusian government (lópez-bonilla, reyes-rodriguez and lópez-bonilla, 2018). the golf tourism in andalusia is characterised by a high degree of loyalty among users, low seasonality and expenditure over 100€ per day (pradas and garcía-tascón, 2019a) and the scale ‘big-five’ model, on the brand personality scale, can identify attributes that stakeholders consider to be essential for a golf destination (pereira, correia and schutz, 2015). one of the most important purposes of sports services is to meet the user’s future intentions. consequently, it is of paramount importance to analyse the user’s loyalty, as it has become one of the main objectives in sports management (cáslavová, pecinová, ruda and šíma, 2018; garcía, alguacil, and molina, 2020); and examine how loyalty influences the relationship between customer satisfaction and repurchase intention (woo, 2017; yi and la, 2004) or the relationship between team loyalty, sponsorship awareness, attitude toward the sponsor, and purchase intentions (biscaia, correia, rosado, ross, and maroco, 2013; wakefield, 2016 ). loyalty for liat, mansori and huei (2014) is defined as commitment toward preferred products or services. also, loyalty is defined as the repeating purchases and/or a recommendation to the third party by consumers (kuhn, bendesa, wiranatha, and okasuryawardani, 2019a) therefore, considering golf as an industry that offers products and/or services, this concept could be applied. the loyalty of users in any type of organisation is vitally important, being one of the elements that guarantee not only the benefits, but also the permanence and growth of the same (crosby and johnson, 2008; van asperen, de rooij, and dijkmans, 2018). customer loyalty has many benefits, economic and non-economic, so the implementation and continuous improvement of loyalty programs are becoming a strategy to maximise the benefits to companies (chen and quester, 2006; pradas and garcía-tascón, 2019b) or knowing the degree of customer loyalty with low-quality perceptions (garcía-fernández, sánchez-oliver, grimaldi-puyana, fernégavira, and gálvez-ruíz, 2017). there is an agreement regarding the causal relationship between quality and loyalty (theodorakis, howat, ko, and avourdiadou, 2014). some studies establish that knowing the degree of customer loyalty allows facility managers to identify customers with low perceptions of quality and act accordingly to improve their loyalty (baena-arroyo, gálvez-ruiz, sánchez-oliver, and bernal-garcía, 2016; barragán-codina, castillo-villarreal, and guerra-rodríguez, 2009; prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 183 garcía-fernández, sánchez-oliver, grimaldi-puyana, ferné-gavira, and gálvez-ruíz, 2017; granja-dueñas, 2013). studies on the perceived quality of services are intended to find out what factors are related to the retention of users, as well as to know what reasons lead the user to leave a sports centre (martínez and martínez 2009; nuviala, grao, pérez and nuviala, 2012; tsitskari, tsiotras, and tsiotras, 2006). a large number of studies have focused on responding to customer loyalty problems at the facilities (garcía-fernández, bernal-garcía, fernández-gavira and vélez-colón, 2014; garcíafernández and pires-veja, 2010; macintosh and law, 2015), as well as knowing the attitudes, profile and behaviour of users of sports centres (martínez and martínez, 2009; molina, mundina, and gómes-tafalla, 2018; nuviala et al., 2012). loyalty in the sports sector has been evaluated by means of instruments that reflect both the behavioural approach (pinillos, 2004; triadó and aparicio, 2004) and the attitudinal approach (bodet, 2012; wei, hung, yang, and jui, 2010), as well as a scale in spanish (nuviala et al., 2014). regarding studies related to perceived quality and golf, several studies have been carried out internationally (crilley, murray, howat, march, and adamson, 2002; kuhn et al., 2019a; kuhn, bendesa, wiranatha, and oka-suryawardani, 2019b; woo, 2017; wu and ai, 2015). in spain, studies related to golf and quality, highlight the research of serrano (2013), iglesias (2015), pradas (2016) and iglesias and lara (2019). likewise, it tackles the study on the situation of golf and tourism in andalusian region (paniza, 2005, 2010), golf as a tourist product (garcía-fernández et al., 2013), or the analysis of loyalty of users of a golf course in the region of murcia (cavas-garcía, díaz-suárez, and martínez-moreno, 2018). studies have also been conducted regarding the design and implementation of an internal control system to optimise the use of human, material and financial resources in a golf club (pincay-quimiz and romero-maquilon, 2015). for these reasons, the main objective of this study is to focus on analysing future behavioural intentions and loyalty of golf users in the spanish region of andalusia; based on the sociodemographic profile (sex and country of residence). 2. method it is a descriptive, non-experimental and cross-cutting/transversal study which aims to analyse the loyalty of golf course users in andalusia. 2.1. participants to calculate the sample size, a standard error of 5%, a confidence level of 95%, and p = 0.5 were initially set, obtaining a corrected sample size equal to 381. it was then decided to increase the accuracy of the study by fixing a standard error of no more than 4%, resulting in a final sample of 636 users from 17 golf courses in andalusia. a total of n = 16 users were prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 184 eliminated due to experimental mortality by not responding to one or more of the items that made up the questionnaire. respondents voluntarily agreed to carry out the study, provided they met the following inclusion criteria: 1. over 18 years old. 2. user of the golf course. 3. handicap (a license granted by the rfeg to allow the player to go out to play the course and measure their level of play). regarding the main characteristics of the sample, the average age of the user was 50.2 ± 15.6 years and 73.43% of the users were men and 26.42% of women. in addition, 59.51% of the users were spanish, 11.81% were from the united kingdom, and 9.05% from sweden. 2.2. instrument the instrument used to carry out this study is based on the zeithaml, berry and parasuraman (1996) behaviour intentions scale, translated and adapted to tourism services by setó (2003), and validated by nuviala et al. (2014) for users of sports services. pradas (2016) adapted the tool for golf courses by using the validation of nuviala et al. (2014) to replace the terms ‘sports organisations’ and/or ‘sports services’ with ‘golf courses’ or ‘sports facilities’ in the final version of the tool. the questionnaire (see appendix) consisted of a battery of 13 items that were intended to measure a wide range of behavioural intentions based on the user's opinions related to loyalty to the golf course, through a likert 1–5 scale (1. totally disagree, 2. disagree, 3. neither agree nor disagree, 4. agree, 5. totally agree). furthermore, for a better understanding, the variables have been grouped into three categories: ‘loyalty’ refers to the intention to repurchase and recommend the golf course (v1v5), ‘price sensation’ (v6-v9), and ‘response’ (v10-v13). these dimensions make it possible to relate the concept of fidelity and the service quality, satisfaction, image and/or trust of users based on setó (2003). also, these block model has been used in cavas-garcía et al. (2018). 2.3. procedure after arrangements were made with the managers of each of the golf courses, data collection was carried out between february and december 2015. the questionnaire was selfadministered, always in the presence of the interviewer. the objectives of the research work were explained to the participants, and they were instructed to complete all the responses to reduce experimental mortality and information regarding the confidentiality of the study. it was thought to be convenient to complete between 20 and 80 questionnaires in each course, according to their dimensions, and on different days and time slots to increase the sample representativeness, with an approximate duration per participant of between 5–10 minutes. prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 185 2.4. data analysis the data analysis was carried out through the statistical package spss 20.0 for windows. first, a descriptive analysis of the sample was carried out for the different variables grouped into three main dimensions: loyalty, price sensation and response. subsequently, a bivariate analysis of these same variables was carried out with respect to the country of residence and sex of the respondents, using the mann-whitney u test. it ends with a two-stage cluster/conglomerate, to identify and classify participants in user groups with similar opinions on loyalty to the golf course. 3. results a total of 636 users of golf courses in andalusia were interviewed. the analysis was done on 13 variables in three categories: ‘loyalty’ (intention to repurchase and recommend the golf course, v1-v5), ‘price sensation’ (v6-v9), and ‘response’ (v10-v13). furthermore, aspects related to the sociodemographic profile, such as sex and place of residence, were also analysed for a better understanding of the variables. regarding the results of the ‘loyalty’ block, table 1 shows that 47.16% of users agreed with the statement: i will tell other people about this golf course (v1); 31.95% ‘totally agreed’ and only 1.69% of users ‘totally disagreed’. 42.64% agreed to recommend the golf course (v2) and 32.82% ‘totally agreed’, while a negative response was given by fewer than 10% of users. more than 70% of users agreed (37.48%) or ‘totally agreed’ (34.56%) to encourage family and friends to play on this golf course (v3). on the contrary, 4.45% totally disagree. regarding if they consider the golf course as the first option for any service they might need (v4), 35.74% agreed. almost 50% was divided between fully agree (23.77%) and neither agree nor disagree (24.85%) and only 6.13% of users disagreed. approximately two-thirds of the users believed that they would play more on the golf course in the following years (v5) (totally agree 32.62% and agree 34.46%). table 1. descriptive analysis of loyalty variables (v1-v5). totally disagree disagree neither agree nor disagree agree totally agree 1. i will tell other people about this golf course 1.69 6.30 12.90 47.16 31. 95 2. i will recommend this golf course to anyone seeking my advice 4.29 4.60 15.65 42.64 32.82 3. i will encourage my family and friends to play on this golf course 4.45 6.30 17.21 37.48 34.56 4. for any service i might need i consider this golf course as my first choice 6.13 9.51 24.85 35.74 23.77 5. in the next few years i plan to play more on this golf course 6.00 9.85 17.07 34.46 32.62 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 186 table 2. variables related to ‘price sensation’ (v6-v19) and ‘response’ (v10-v13). totally disagree disagree neither agree nor disagree agree totally agree 6. in the next few years i will play less on this golf course. 34.51 29.28 17.88 9.24 9.09 7. i will recommend this golf course to anyone seeking my advice. 16.44 18.89 38.09 17.36 9.22 8. i will encourage my family and friends to play on this golf course 19.48 27.61 37.19 15.72 14.50 9. for any service i might need i consider this golf course as my first choice. 30.52 23.47 25.15 14.11 6.75 10. in the next few years i plan to play more on this golf course. 14.57 16.10 34.36 19.17 15.80 11. if i have a problem. i will tell other customers/people. 11.50 11.66 34.66 23.62 18.56 12. if i have a problem with this golf course i will make a claim with external entities such as the consumers and users association. 25.23 15.38 29.38 20.15 9.85 13. if i have a problem with the service, i will complain to the golf course manager 9.26 8.95 25.93 31.33 24.54 in reference to price sensation (v6-v9) and response (v10-v13), it is observed in table 2 that less than 20% of users agree (9.24%) or totally agree (9.09%) that they would play less in the following years on the golf courses (v6). in this sense, a large percentage of users (38.10%) did not agree or disagree with the statement that they may play on another golf course that offers better service (v7). more than half of the users claim that although the prices are higher, they will continue playing in this course, (v8), as almost 50% say they would be willing to pay a higher price for playing on this golf course for the service received (v9). for the response change to another golf course, if i have a problem with the service (v10), users did not show a strong agreement or disagreement. the highest percentage of responses was that 34.36% said they did not agree or disagree. the users showed a behaviour similar to the previous one (v12). the most chosen option for both statements was not to agree or disagree, with 34.66% and 29.38%, respectively. in v13, approximately 55% of users fully agreed (24.54%) or agreed (31.33%) that they would do so. table 3 shows a bivariate analysis of the dimensions of loyalty, price sensation and response according to the country of residence, performing the mann-whitney u test for sociodemographic variables with two categories of responses. it is observed that, in general, users residing outside of spain had a higher rating than residents in spain. however, these differences are mostly insignificant, with only the following variables being significant: even if prices go up, i will continue to play in this golf course (u = 37054.00; p = 0.042), i am willing to pay a higher price for services that i receive (u = 35841.50; p = 0.009) and if i have a problem i will complain to the director of the golf course (u = 30042.00; p <0.001). prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 187 table 3. bivariate analysis of the loyalty (v1-5), price sensation (v6-9) and response (v10-13) dimensions according to the country where they resided. country mean sd confidence interval 95% mean comparison test p-value 0.05 upper limit lower limit 1. i tell positive features other people about this golf course. spain 3.972 0.985 3.879 4.055 u=39385.50 0.373 outside spain 4.114 0.742 4.006 4.220 2. i recommend this golf course to anyone seeking my advice. spain 3.907 0.985 3.800 4.006 u=39339.50 0.345 outside spain 4.080 0.798 3.968 4.195 3. i encourage my family and friends to play on this golf course. spain 3.868 1.150 3.758 3.962 u=39600.00 0.445 outside spain 4.029 0.874 3.900 4.151 4. for any service you might need consider this golf course as the first choice. spain 3.561 1.205 3.446 3.682 u=38826.00 0.239 outside spain 3.754 0.898 3.623 3.886 5. in the next few years i will play more on this golf course. spain 3.804 1.195 3.699 3.903 u=38041.00 0.156 outside spain 3.701 1.134 3.525 3.855 6. in the next few years i will play less son this golf course. spain 2.267 1.320 2.147 2.383 u=37288.50 0.081 outside spain 2.379 1.165 2.195 2.549 7. i may play in another golf course that offers better services. spain 2.896 1.192 2.791 3.004 u=37018.00 0.051 outside spain 2.684 1.111 2.506 2.856 8. even if the prices were higher, i would play on this golf course. spain 2.732 1.284 2.615 2.857 u=37054.00 0.042* outside spain 2.909 1.068 2.761 3.072 9. i am willing to pay a higher price for playing on this golf course for the service i receive. spain 2.378 1.299 2.268 2.495 u=35841.50 0.009* outside spain 2.600 1.077 2.440 2.750 10. i would switch to another golf course if i have a problem with the service. spain 3.089 1.303 2.962 3.212 u=39423.50 0.382 outside spain 2.989 1.104 2.828 3.153 11. if i have a problem, i tell other customers/people. spain 3.236 1.274 3.121 3.350 u=40311.00 0.659 outside spain 3.309 1.070 3.161 3.468 12. if i have a problem with this golf course, i will make a claim with external entities such as the consumers and users association. spain 2.772 1.326 2.653 2.886 u=38765.00 0.297 outside spain 2.649 1.230 2.463 2.827 13. if i have a problem with the service claim from the golf course manager. spain 3.679 1.199 3.571 3.788 u=30042.00 0.000* outside spain 3.149 1.165 2.969 3.333 table 4 shows the results of the variables analysed according to gender. only significant differences between genders are observed in the variables encourage my family and friends to play in this golf course (u = 37495.00; p = 0.024), in which women had a higher rating; in the next few years play more in this golf course (u = 37796.50; p = 0.048), in which women also showed better evaluation; and change the golf course if i have problems with the service (u = 37614.00; p = 0.029), which in this case the assessment was higher among men. prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 188 table 4. bivariate analysis of the dimension of loyalty (v1-5), price sensation (v6-9) y response (v10-13) by sex. sex mean sd confidence interval 95% mean comparison test; p-value 0.05 upper limit lower limit 1. i share positive features of this golf course with other people. men 3.98 0.91 3.91 3.91 u=38665.00; 0.086 woman 4.09 0.94 3.95 3.95 2. recommend this golf course to anyone seeking my advice. men 3.92 0.91 3.82 3.82 u=39322.50; 0.159 women 4.03 1.02 3.88 3.88 3. encourage my family and friends to play on this golf course. men 3.87 1.06 3.78 3.78 u=37495.00; 0.024* women 4.04 1.08 3.88 3.88 4. for any service you might need consider this golf course as the first choice. men 3.58 1.10 3.47 3.47 u=38230.50; 0.056 women 3.72 1.184 3.55 3.55 5. in the next few years, i will play more on this golf course. men 3.75 1.13 3.64 3.64 u=37796.50; 0.048* women 3.87 1.27 3.67 3.67 6. in the next few years, i will play less son this golf course. men 2.29 1.24 2.18 2.18 u=40093.50; 0.420 women 2.26 1.35 2.07 2.07 7. i may play in another golf course that offers better services. men 2.88 1.12 2.78 2.78 u=38621.50; 0.094 women 2.71 1.28 2.52 2.52 8. even if the prices were higher, i would play on this golf course. men 2.75 1.22 2.64 2.64 u=40743.00; 0.497 women 2.82 1.24 2.64 2.64 9. i am willing to pay a higher price for playing on this golf course for the service i receive. men 2.45 1.24 2.34 2.34 u=41081.50; 0.605 women 2.39 1.24 2.21 2.21 10. i would switch to another golf course if i have a problem with the service. men 3.12 1.26 3.00 3.00 u=37614.00; 0.029* women 2.87 1.18 2.70 2.70 11. if i have a problem, i tell other customers/people. men 3.30 1.20 3.19 3.19 u=39518.50; 0.202 women 3.15 1.26 2.95 2.95 12. if i have a problem with this golf course, i will make a claim with external entities such as the consumers and users association. men 2.76 1.30 2.64 2.64 u=40479.00; 0.560 women 2.69 1.29 2.50 2.50 13. if i have a problem with the service, i will complain to the golf course manager. men 3.52 1.21 3.41 3.41 u=40664.50; 0.581 women 3.57 1.21 3.39 3.39 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 189 very small cluster sizes 79 (13.1%) very large cluster sizes 314 (52.15%) ratio of sizes: largest cluster to smallest cluster 3.97 figure 1. cluster sizes obtained by a biestatic cluster analysis. after the descriptive and bivariate analysis of the sociodemographic and loyalty variables, a two-stage cluster/conglomerate statistical analysis is performed, which identifies and classifies the participants in groups of people with similar opinions about loyalty to the course. three clusters (different groups of people with similar opinions) are evidenced according to the average of the discriminatory variable (figure 1). thus, cluster 1 represents users of the opinion regarding the chances that they will leave the golf course. they are called unfaithful clients and represent 13.1% of the participants and respond to the variables of playing in another course with better services or if they would change because they had a problem with the course (v6), (v7) and (v10). cluster 2 represents users with higher opinions or in accordance with the fact that they would claim to manager or consumers association if they had a problem (v12), (v13) and (v11). the clients are probably unfaithful and represent 52.1% of the participants. finally, cluster 3 was formed by customers who would remain on the golf course. that is, they are called loyal customers and represent 34.8% (n=of the participants (v3) (v4), (v1), (v8) and (v9). 4. discussion knowing the future intentions of user behaviour regarding sports organisations is vital because it influences their permanence and growth (chen and quester, 2006; crosby and johnson, 2008; van asperen et al., 2018). this study of customer loyalty to golf courses seeks to identify low perceptions in the variables analysed in order to establish areas of improvement in service quality to improve loyalty and seek greater loyalty (baena-arroyo et al., 2016; garcía-fernández, et al., 2017; theodorakis et al., 2014). in the age variable, it is observed that half of the golf players (50.94%) were aged between 50–69 years, the average age being 50.17 years; similar values (50 years) have been recorded in the murcia region study (cavas-garcía et al., 2018) or the australian study of crilley et al. (2002). but in this case, it accounted for 36% of the participants. in the serrano study (2013) there is a slightly lower average age with 47.58 years and the study analysed people from the ages of 16 to 79 years; more than half (57.75%) was a range of age 41 and 60 years. paniza (2005) said that the profile of the andalusian golf player is between the ages of 41 and 50 but does not show an average numerical value that endorses it. there are other studies where the average age is very low compared to these studies. the average age of 16.14 years is recorded, and they analyse the performance of amateur golf players (pinto and vázquez, 2013). most of the participants (72.24%) resided in spain, followed by a small percentage of the prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 190 total residing in other countries such as sweden (7.67%), iceland (4.75%) and denmark (3.22%). regarding nationality, about 60% were spanish participants (59.51%), followed by other nationalities such as the united kingdom (11.81%) and sweden (9.05%). in the study by crilley et al. (2002) only 5% of users were born in english-speaking countries, and the study by navarro-vera and ortuño-padilla (2010) records that of the homeowners in a residential golf complex, 60% are foreigners, of which two-thirds come from the united kingdom, and more than 20% come from germany and the netherlands. with respect to the loyalty variables of the users according to their sex, only two variables stand out in which there were significant differences in the loyalty block and where a slightly higher assessment was shown in women (v3) and (v5) concerning encouraging family members to sign up for the course, as well as playing more in the course where they were; and with respect to the variable (v10) of the response block, the assessment by men was higher in terms of changing the course if there is any problem with the service. if the results are reviewed with the study by serrano, rial, garcía and gambau (2011) there are differences between the sexes, but only relative to specific aspects of the management and adaptation of the golf service. most of the golf courses users were men (73.43%), compared to 26.42% of women. although the incorporation of women into the sport has been later, participation in golf were recorded as 58% men and 42% women (hosteltur, 2012). however, other studies show data of 74.1% of male tourists compared to 25.9% of women (junta de andalucía. ministry of tourism and commerce, 2012). higher differences were found in the serrano study (2013) where the percentage of men (82.2%) was much higher than the percentage of women (17.8%); values that are maintained in the australian work of crilley et al. (2002) were 83% of male users. in courses other than golf, such as fitness (garcía et al., 2014; garcía-ferrando and llopis, 2011; molina et al., 2018) have a greater influx of women to these sports services. it is relevant to know the global opinion that users have about the performance of the golf club, so 47.16% of users agreed with the statement of telling positive aspects to other people (v1), 31.95% totally agreed and only 1.69% of users strongly disagreed. the study by cavasgarcía et al. (2018), given the same statement, 45.2% of users agree and 47.6% totally agree. on the one hand, 42.64% agreed to recommend the course (v2) and 32.82% totally agreed, while a negative response was only given by less than 10% of users, while the percentages are higher in cavas-garcia et al. (2018) made in murcia. on the other hand, these values are similar to those in the study by paniza (2005), where 45% of the users of the andalusian courses expressed positive opinions. in the study by crilley et al. (2002), 68% of users recommend the course and 24% would strongly recommend it, in the present study carried out in andalusia, higher values are recorded (32.82%). more than 70% of users agreed (37.48%) or totally agreed (34.56%) in (v3) about encouraging the family to attend the golf course. conversely, only 4.45% strongly disagreed. in the works of hennessey, macdonald and maceachern (2008) and hutchinson, wang and lai (2010), the overall satisfaction has a significant influence on word-of-mouth references and the prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 191 intentions of using a service again. some of the users (17.63%) agreed that they would play in another golf course that offered better services (v7), and only 9.22% registered total agreement. some studies indicate that the state of the golf course and facilities are the attributes that best explain user preferences (hwang and won, 2010; won, hwang, and kleiber, 2009). the loyalty category is the one that best expresses the intention to repurchase and recommend the service (setó, 2003). in the work done in murcia by cavas-garcía et al. (2018), similar results were obtained to those found in this work, highlighting that the opinions in murcia have been in all cases superior to the clients of the golf courses in andalusia. approximately two-thirds of the clients stated that in the next few years they would play more on the golf course (v5) where they were a member (67.08%), ‘totally agree’ (32.62%) and ‘agree’ (4.46%). these terms are related to loyalty (sanz and ponce de león, 2005, 2006), while van asperen et al. (2018) indicates that this loyalty is related to social networks. the category of price sensation and being willing to pay more for the service offered (v9) in the course expressed ‘totally agree’ (6.75%) and ‘agree’ (14.11%), values slightly higher than the work from of cavas-garcía et al. (2018) in murcia. serrano (2013), confirms that users are willing to pay more for better customer service in the human factor, considerations that are shared by other studies in other types of sports services (bodet, 2006; dorado, 2004; rial, valera, rial, and real, 2010). regarding the dimension of response and change to another golf course in case of a problem (v10), they expressed ‘totally agree’ (15.80%) and ‘agree’ (19.17%). in the study in murcia by cavas-garcía et al. (2018), much better results have been obtained (doubling) in the affirmation of ‘totally agree’ (37.70%), while maintaining similar values in that of ‘agree’ (21.4%). relevant in this investigation is the contribution of the cluster statistical analysis identifying and classifying the participants in groups of people with similar opinions about loyalty with the course. data guides managers of the golf courses to establish strategies based on the results of clusters 1 and 2. it is evident that cluster 1 (v6-v7-v10) represents the so-called unfaithful clients (13.1%) who would easily switch to another golf course in the event of dissatisfaction or any problem. cluster 2 (v11-v12-v13), represent half of the participants with 52.1% and show customers unfaithful ones identified with opinions related to the fact that they would claim. undoubtedly, this business model based on the loyalty of golfers should take into account these data and establish strategies to reduce these percentages (valenzuelafernández, 2010). the galician autonomous community jointly work the galician golf courses in marketing to the french golf tourist (clúster turismo galicia, 2016). in general, and not only on golf courses, there must be a joint effort between manager and rest of the entity's employees (ramos-farroñán, valdivia-salazar, and vidaurre-sandoval, 2019; rodríguez, 2016), they must plan actions to retain the client faithful (valenzuela fernández, 2010), in the case of cluster 3 (v1-v3-v4-v8-v9) is formed by the clients that would remain on the golf course and represents a high percentage, 34.81% of the participants. prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 192 5. conclusions and practical implications the study presents a tool to predict future intentions of golf course users' behaviour, a simple scale that can be used by managers to measure customer loyalty to their golf course. other aspects to highlight about the opinion that customers show are: 1. a positive general opinion (v1-v2-v3) of the clients is observed on the golf courses they attend, highlighting in v1, that almost 50% agree (47.16%) to spread the good aspects from the countryside to friends and family. 2. managers of the golf courses must pay special attention to variables (v2) and (v3), which are the ones with which the clients feel most identified and can guide the establishment of loyalty policies. 3. given the claim that although the prices are higher, the client will continue playing on the golf course (v8), only 30.22% positioned themselves in the ‘totally agree’ and ‘agree’ responses. these values invite to the reflection since almost half (47.09%) of the users have positioned themselves in the affirmation of ‘totally disagree’ and ‘disagree’. 4. users who reside outside of spain collect an assessment of the golf courses higher than those provided by residents in spain (v8-v9) and obtain a better score in these variables related to loyalty. in the variable (v13) related to the variable block response, the clients residing in spain show a higher assessment and indicate that if they had any problem with the service, they would complain to the director of the golf course. 5. the discriminatory variable most valued in the clusters was ‘encourage family and friends to attend the club’ (v3). these results affirm that the study is a reliable and valid instrument for management because it provides very useful information about the future intentions of golf course clients, and the client base is one of the greatest assets of a golf course. in that way, this tool is very suitable for managers. it helps them identify the future intentions of the client and allows them to plan actions to hold those that are potentially included in the disloyal and first cluster group to leave the club. in conclusion, managers are advised that they must offer better service and adapt their offerings in order to build customer loyalty. prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 193 references aymerich, f., and anabitarte, j. (2016). el impacto económico del golf en españa. retrieved from http://www.rfegolf.es/artculosdocumento/turismo%20e%20impacto%20econ%c3%b3 mico/turismo%20e%20impacto%20econ%c3%b3mico%202016/2016%20impacto%20econ %c3%b3mico%20del%20golf%20en%20espa%c3%b1a.pdf baena-arroyo, m.j., gálvez-ruiz, p., sánchez-oliver, a.j., and bernal-garcía, a. (2016). la relación entre la experiencia de servicio, el valor percibido y las intenciones de comportamiento en clientes de actividades dirigidas. revista de psicología del deporte, 25(3), 89–92. barciela, f. (2017). la industria del golf sale del hoyo. la crisis y las nuevas regulaciones para el sector como las relativas al consumo de agua han propiciado la creación de empresas de servicios. retrieved from https://elpais.com/economia/2017/11/30/actualidad/1512058937 _849515.html barragán-codina, j., castillo-villareal, j., and guerra-rodríguez. p. (2009). la retención de empleados eficientes: importancia estratégica de la fidelización de los empleados. daena (international journal of good conscience), 4(2), 145-159. retrieved from http://www.spentamexico.org/v4-n2/4%282%29%20145-159.pdf biscaia, r., correia, a., rosado, a., ross, s.d., and maroco, j. (2013). sport sponsorship: the relationship between team loyalty, sponsorship awareness, attitude toward the sponsor, and purchase intentions. journal of sport management, 27(4), 288–302. https://doi.org/10.1123/jsm.27.4.288 bodet, g. (2006). investigating customer satisfaction in a health club context. european sport management quaterly, 6, 149-165. https://doi.org/10.1080/16184740600954148 bodet, g. (2012). loyalty in sport participation services: an examination of the mediating role of psychological commitment. journal of sport management, 26(1), 30-42. https://doi.org/10.1123/jsm.26.1.30 cáslavová, e., pecinová, m., ruda, t., and šíma, m. (2018). service quality in sport: a case study of golf resorts in the liberec region. acta universitatis carolinae. kinanthropologica, 54(2), 137-148. https://doi.org/10.14712/23366052.2018.11 cavas-garcía, f., díaz-suárez, a., and martínez-moreno, a. (2018). lealtad de los clientes de un campo de golf. un estudio de caso. journal of sports economics and management, 8(2), 100-106. chen, s.c., and quester, p.g. (2006). modeling store loyalty: perceived value in market orientation practice. journal of services marketing, 20(3), 188-198. https://doi.org/10.1108/08876040610665643 clúster turismo de galicia. (2016). la coordinación entre hoteles y campos de golf, clave para la comercialización del producto en galicia. retrieved from http://clusterturismo galicia.com/la-coordinacion-entre-hoteles-y-campos-de-golf-clave-para-la-comercializaciondel-producto-en-galicia-el-hotel-los-abetos-de-santiago-de-compostela-acogio-una-jornadaen-la-que-se-dieron-cita-las-di/ http://www.rfegolf.es/artculosdocumento/turismo%20e%20impacto%20econ%c3%b3mico/turismo%20e%20impacto%20econ%c3%b3mico%202016/2016%20impacto%20econ%c3%b3mico%20del%20golf%20en%20espa%c3%b1a.pdf http://www.rfegolf.es/artculosdocumento/turismo%20e%20impacto%20econ%c3%b3mico/turismo%20e%20impacto%20econ%c3%b3mico%202016/2016%20impacto%20econ%c3%b3mico%20del%20golf%20en%20espa%c3%b1a.pdf http://www.rfegolf.es/artculosdocumento/turismo%20e%20impacto%20econ%c3%b3mico/turismo%20e%20impacto%20econ%c3%b3mico%202016/2016%20impacto%20econ%c3%b3mico%20del%20golf%20en%20espa%c3%b1a.pdf https://elpais.com/economia/2017/11/30/actualidad/1512058937_849515.html https://elpais.com/economia/2017/11/30/actualidad/1512058937_849515.html http://www.spentamexico.org/v4-n2/4%282%29%20145-159.pdf https://doi.org/10.1123/jsm.27.4.288 https://doi.org/10.1080/16184740600954148 https://doi.org/10.1123/jsm.26.1.30 https://doi.org/10.14712/23366052.2018.11 https://doi.org/10.1108/08876040610665643 http://clusterturismogalicia.com/la-coordinacion-entre-hoteles-y-campos-de-golf-clave-para-la-comercializacion-del-producto-en-galicia-el-hotel-los-abetos-de-santiago-de-compostela-acogio-una-jornada-en-la-que-se-dieron-cita-las-di/ http://clusterturismogalicia.com/la-coordinacion-entre-hoteles-y-campos-de-golf-clave-para-la-comercializacion-del-producto-en-galicia-el-hotel-los-abetos-de-santiago-de-compostela-acogio-una-jornada-en-la-que-se-dieron-cita-las-di/ http://clusterturismogalicia.com/la-coordinacion-entre-hoteles-y-campos-de-golf-clave-para-la-comercializacion-del-producto-en-galicia-el-hotel-los-abetos-de-santiago-de-compostela-acogio-una-jornada-en-la-que-se-dieron-cita-las-di/ http://clusterturismogalicia.com/la-coordinacion-entre-hoteles-y-campos-de-golf-clave-para-la-comercializacion-del-producto-en-galicia-el-hotel-los-abetos-de-santiago-de-compostela-acogio-una-jornada-en-la-que-se-dieron-cita-las-di/ prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 194 crilley, g., murray, d., howat, g., march, h., and adamson, d. (2002). measuring performance in operational management and customer service quality. a survey of financial and nonfinancial metrics from the australia golf industry. journal of leisure property, 2(4), 369-371. https://doi.org/10.1057/palgrave.rlp.5090157 crosby, l.a., and johnson, s.l. (2008). en busca de experiencias que generen fidelidad. harvard deusto márketing y ventas, 85, 38-48. dorado, a. (2004). análisis de la satisfacción de los usuarios: hacia un nuevo modelo de gestión basado en la calidad para los servicios deportivos municipales (tesis doctoral). universidad de castilla-la mancha, españa. garcía, f., alguacil, m., and molina, n. (2020). predicción de la lealtad de los usuarios de centros deportivos privados. variables de gestión y bienestar subjetivo. retos, 38, 16-19. retrieved from https://recyt.fecyt.es/index.php/retos/article/view/73345/46263 garcía-fernández, j., bernal-garcía, a., fernández-gavira, j., and vélez-colon, l. (2014). analysis of existing literature on management and marketing of the fitness centre industry. south african journal for research in sport, physical education and recreation, 36(3), 7591. garcía-fernández, j., fernández-garcía, j., and bernal-garcía, a. (2014). la percepción de calidad y fidelidad en clientes de centros de fitness low-cost. suma psicológica, 21(2), 123130. https://doi.org/10.1016/s0121-4381(14)70015-3 garcía-fernández, j., and pires-veja, f. (2010). fidelización de usuarios mayores en centros de fitness: gestión de clientes por programas de actividad física. retos. nuevas tendencias en educación física, deporte y recreación, 17, 103-106. garcía-fernández, j., sánchez-oliver, a.j., grimaldi-puyana, m., fernández-gavira, j., and gálvez-ruíz, p. (2017). la calidad y la fidelidad del cliente: un análisis de segmentación en centros de fitness low-cost. journal sport psychology, 26, 17-22. garcía-ferrando, m., and llopis, r. (2011). encuesta sobre los hábitos deportivos en españa 2010. madrid, españa: consejo superior de deportes – centro de investigaciones sociológicas. granja-dueñas, t.a. (2013). guía para la aplicación de store environment y field marketing a través del uso de neuromarketing en almacenes de ropa deportiva. caso marathon sports (tesis doctoral). universidad de las américas, ecuador. hennessey, s.m., macdonald, r., and maceachern, m. (2008). a framework for understanding golfing visitors to a destination. journal of sport and tourism, 13(1), 5–35. https://doi.org/10.1080/14775080801972023 hosteltur (2012). especial golf el oro verde el turismo. retrieved from https://issuu.com/ hosteltur_2012/docs/especial_golf_hosteltur_2012-el_oro_verde_del_turi hutchinson, j., wang, y., and lai, f. (2010). the impact of satisfaction judgment on behavioral intentions: an investigation of golf travelers. journal of vacation marketing, 16(1), 45-59. https://doi.org/10.1177/1356766709356203 https://doi.org/10.1057/palgrave.rlp.5090157 https://recyt.fecyt.es/index.php/retos/article/view/73345/46263 https://doi.org/10.1016/s0121-4381(14)70015-3 https://doi.org/10.1080/14775080801972023 https://issuu.com/hosteltur_2012/docs/especial_golf_hosteltur_2012-el_oro_verde_del_turi https://issuu.com/hosteltur_2012/docs/especial_golf_hosteltur_2012-el_oro_verde_del_turi https://doi.org/10.1177/1356766709356203 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 195 hwang, s., and won, d. (2010). a conjoint analysis regarding influencing factors of golfers’ preferred driving ranges in korea. international journal of leisure and tourism marketing, 1(3), 227-237. https://doi.org/10.1504/ijltm.2010.029586 iglesias, m.a. (2015). el deporte del golf: análisis sectorial y aplicación del modelo efqm a la gestión de campos de golf en castilla y león (tesis doctoral). universidad de valladolid, españa. iglesias, m.a., and lara, a.j. (2019). turismo de golf en países del sur de europa. un mercado potencial. estudios de economia aplicada, 37(1), 64-78. https://doi.org/ 10.25115/eea.v37i1.2573 junta de andalucía. consejería de turismo y comercio. (2012). turismo de golf en andalucía. informe anual 2012. retrieved from http://www.juntadeandalucia.es/turismoycomercio /publicaciones/143324771_2012.pdf kpmg (2017). golf participation report for europe 2017. retrieved from https://www.golfbenchmark.com/golf_participation_report_for_europe_2017 kuhn, p., bendesa, k., wiranatha, a.s., and oka-suryawardani, i.g.a. (2019a). the antecedents of golfers, satisfaction and its effects on their loyalties: a structural equation model for golf tourism in bali, indonesia. journal of tourism and hospitality management, 7(1), 66-73. https://doi.org/10.15640/jthm.v7n1a7 kuhn, p., bendesa, k., wiranatha, a.s., and oka-suryawardani, i.g.a. (2019b). does motivation and trust important for tourists’ loyalties? a quantitative perspective from golf tourism in bali of indonesia. e-journal of tourism, 6, 88-101. https://doi.org/10.24922/eot.v6i1.47481 liat, c.b., mansori, s., and huei, c.t. (2014). the associations between service quality, corporate image, customer satisfaction, and loyalty: evidence from the malaysian hotel industry. journal of hospitality marketing and management, 23(3), 314-326. https://doi.org/10.1080/19368623.2013.796867 lópez-bonilla, j.m., reyes-rodríguez, m.d.c., and lópez-bonilla, l.m. (2018). the environmental attitudes and behaviours of european golf tourists. sustainability, 10, 1-16. https://doi.org/10.3390/su10072214 macintosh, e., and law, b. (2015). should i stay or should i go? imploring the decisión to join, maintain, or cancel a fitness membership. managing sport and leisure, 20(3), 191-210. https://doi.org/10.1080/23750472.2015.1025093 martínez, j.a., and martínez, l. (2009). a customer management model in sports services; a system dynamics approach. revista internacional de medicina y ciencias de la actividad física y del deporte, 9(36), 431-453. molina, n., mundina, j., and gómes-tafalla, a. (2018). perfil del usuario de centros deportivos privados, según género, edad y nivel de antigüedad. sport tk: revista euroamericana de ciencias del deporte, 8(1), 23-28. https://doi.org/10.6018/sportk.362281 navarro-vera, j.r., and ortuño-padilla, a. (2010). impacto de los campos de golf en levante. ciudad y territorio. estudios territoriales, 163, 35-48. https://doi.org/10.1504/ijltm.2010.029586 https://doi.org/10.25115/eea.v37i1.2573 https://doi.org/10.25115/eea.v37i1.2573 http://www.juntadeandalucia.es/turismoycomercio/publicaciones/143324771_2012.pdf http://www.juntadeandalucia.es/turismoycomercio/publicaciones/143324771_2012.pdf https://www.golfbenchmark.com/golf_participation_report_for_europe_2017 https://doi.org/10.15640/jthm.v7n1a7 https://doi.org/10.24922/eot.v6i1.47481 https://doi.org/10.1080/19368623.2013.796867 https://doi.org/10.3390/su10072214 https://doi.org/10.1080/23750472.2015.1025093 https://doi.org/10.6018/sportk.362281 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 196 nuviala, a., garcíafernández, j., bernal-garcía, a., grao-cruces, a., teva-villén, m.r., and pérez-ordás, r. (2014). adaptación y validación de la escala de intenciones futuras de comportamiento en usuarios de servicios deportivos. universitas psychologica, 13(3), 10711082. https://doi.org/10.11144/javeriana.upsy13-3.avei nuviala, a., grao-cruces, a., pérez, j.a., and nuviala, r. (2012). quality, satisfaction and perceived value in groups of users of sports organisations in spain. kinesiology, 44(1), 94103. retrieved from https://www.redalyc.org/pdf/1153/115324888002.pdf paniza, j.l. (2005). la percepción social del golf en andalucía. málaga, españa: consejería de comercio, turismo y deporte. paniza, j.l. (2010). empleo y formación en los campos de golf de andalucía. sevilla, españa: consejería de turismo, comercio y deporte. pedreño, a.d. (2015). perfil sociodemográfico y características del viaje del turista de golf no residente en españa: 2010-2013. anuario de jóvenes investigadores, 8, 75-77. pereira, r.l., correia, a.h., and schutz, r.l. (2015). towards a taxonomy of a golf-destination brand personality: insights from the algarve golf industry. journal of destination marketing and management, 4(1), 57-67. https://doi.org/10.1016/j.jdmm.2014.12.003 pinillos, j. (2004). winning the retention battle: librando la batalla por la retención. investigación y marketing, 83, 39-42. pincay-quimiz, v., and romero-maquillón, m.g. (2015). diseño e implementación de un sistema de control interno para optimizar el uso de los recursos humanos, materiales y financieros en la urbanizacion mocoli golf club de la ciudad de guayaquil (tesis doctoral). universidad de guayaquil, ecuador. pinto, m.f., and vázquez, n. (2013). ansiedad estado competitiva y estrategias de afrontamiento: su relación con el rendimiento en una muestra argentina de jugadores amateurs de golf. revista de psicología del deporte, 22(1), 47-52. pradas, m. (2016). análisis de la calidad de los servicios prestados de los clientes internos y externos de los campos de golf en andalucía (tesis doctoral). universidad pablo de olavide, españa. pradas, m., and garcía-tascón, m. (2019a). satisfacción con el turismo de golf en andalucía y su impacto económico. revista española de educación física y deportes, 426, 227-233. retrieved from http://www.reefd.es/index.php/reefd/article/download/783/663 pradas, m., and garcía-tascón, m. (2019b). análisis de intenciones futuras de comportamiento en usuarios de campos de golf de andalucía. revista española de educación física y deportes, 426, 533-539. retrieved from http://reefd.es/index.php/reefd/article /viewfile/826/701 pinero, j.m. (2016). el turismo y la industria del golf en españa. un nuevo año de record trabajo. retrieved from https://www.auren.com/es-es/blog/turismo/2016/11/07/el-turismo-yla-industria-del-golf-en-espana https://doi.org/10.11144/javeriana.upsy13-3.avei https://www.redalyc.org/pdf/1153/115324888002.pdf https://doi.org/10.1016/j.jdmm.2014.12.003 http://www.reefd.es/index.php/reefd/article/download/783/663 http://reefd.es/index.php/reefd/article/viewfile/826/701 http://reefd.es/index.php/reefd/article/viewfile/826/701 https://www.auren.com/es-es/blog/turismo/2016/11/07/el-turismo-y-la-industria-del-golf-en-espana https://www.auren.com/es-es/blog/turismo/2016/11/07/el-turismo-y-la-industria-del-golf-en-espana prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 197 ramos-farroñán, e.v., valdivia-salazar, c.a., and vidaurre-sandoval, j.e. (2019). el geomarketing como estrategia para el posicionamiento de una óptica en chiclayo 2018. tzhoecoen, 11(1), 24-33. real federación española de golf (2019). evolución de licencias desde 1967. retrieved from http://www.rfegolf.es/artculosdocumento/licencias/evolucionlicencias.pdf rial, j., varela, j., rial, a., and real, e. (2010). modelización y medida de la calidad percibida en centros deportivos: la escala qsport-10. revista internacional de ciencias del deporte, 18(6), 57-73. https://doi.org/10.5232/ricyde2010.01804 riquel-ligero, f., and vargas-sánchez, a. (2012). the institutional framework of environmental character on andalusian golf courses: a factor analysis. cuadernos de turismo, 29, 287-289. rodríguez, m.a. (2016). análisis de las estrategias de fidelización de los clientes internos (tesis doctoral). universidad nacional de la plata, argentina. sanz, e., and ponce de león, a. (2005-2006). valoración de los recursos del ocio físicodeportivo: análisis y selección de atributos. contextos educativos, 8-9, 201-209. retrieved from https://dialnet.unirioja.es/descarga/articulo/2091399.pdf serrano, v. (2013). diseño y elaboración de una herramienta de gestión para evaluar la calidad percibida de los usuarios en clubes de golf (tesis doctoral). universidad de la coruña, españa. serrano, v., rial, a., garcía, o., and gambau, v. (2011). el perfil de usuario en clubes de golf sociales-mixtos con campos de 9 hoyos en galicia: un intento de segmentación desde la perspectiva de género. cuadernos de psicología del deporte, 11(2), 131-138. retrieved from https://revistas.um.es/cpd/article/view/133921/123281 setó, d. (2003). la fidelidad del cliente en el ámbito de los servicios: un análisis de la escala intenciones de comportamiento. investigaciones europeas de dirección y economía de la empresa, 9(2), 189-204. the global golf tourism organisation [iagto] (2013). golf tourisme report 2013. retrieved from http://www.engorussia.ru/upload/2883.pdf the global golf tourism organisation [iagto] (2019). the global golf tourism organisation. retrieved from http://www.iagto.com theodorakis, n.d., howat, g., ko, y.j., and avourdiadou, s. (2014). a comparison of service evaluation models in the context of sport and fitness centres in greece. managing leisure, 19(1), 18-35. https://doi.org/10.1080/13606719.2013.849505 triadó, x., and aparicio, p. (2004). el tipo de cuota como variable de fidelización de usuarios en entidades deportivas: un reto actual de consecuencias futuras. investigación y marketing, 83, 31-38. tsitskari, e., tsiotras, d., and tsiotras, g. (2006). measuring service quality in sport services. total quality management and business excellence, 17(5), 623-631. https://doi.org/10.1080/14783360600588190 valenzuela-fernández, l. (2010). marketing orientado al cliente y ética empresarial: efectos sobre el valor de la cartera (tesis doctoral). universidad de chile, chile. http://www.rfegolf.es/artculosdocumento/licencias/evolucionlicencias.pdf https://doi.org/10.5232/ricyde2010.01804 https://dialnet.unirioja.es/descarga/articulo/2091399.pdf https://revistas.um.es/cpd/article/view/133921/123281 http://www.engorussia.ru/upload/2883.pdf http://www.iagto.com/ https://doi.org/10.1080/13606719.2013.849505 https://doi.org/10.1080/14783360600588190 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 198 wakefield, k. (2016). using fan passion to predict attendance, media consumption and social media behaviors. journal of sport management, 30(3), 229-247. https://doi.org/10.1123/jsm.2015-0039 van asperen, m., de rooij, p., and dijkmans, c. (2018) engagement-based loyalty: the effects of social media engagement on customer loyalty in the travel industry, international journal of hospitality and tourism administration, 19(1), 78-94. https://doi.org/10.1080/15256480.2017.1305313 vargas-sánchez, a., and riquel-ligero, f. (2015). golf tourism, its institutional setting, and environmental management: a longitudinal analysis. european journal of tourism research, 9, 41–56. wei, p., hung, h., yang, h., and jui, y. (2010). examination of the influence of service quality on membership renewal in fitness centers in san francisco bay area. journal of service science, 3(2), 13-24. https://doi.org/10.19030/jss.v3i2.366 won, d., hwang, s., and kleiber, d.a. (2009). how do golfers choose a golf course? aconjoint analysis of influencing factors. journal of park and recreation administration, 27(2), 1-16. woo, b. (2017). testing the relationships among motivation, service quality, customer satisfaction and loyalty: a case of recreational golfers in south korea. international journal of applied sport sciences, 29(1), 64-76. https://doi.org/10.24985/ijass.2017.29.1.64 wu, h.c., and ai, c.h. (2016). synthesizing the effects of experiential quality, excitement, equity, experiential satisfaction on experiential loyalty for the golf industry: the case of hainan island. journal of hospitality and tourism management, 29, 41-59. https://doi.org/10.1016/j.jhtm.2016.05.005 yi, y., and la, s. (2004). what influences the relationship between customer satisfaction and repurchase intention? investigating the effects of adjusted expectations and customer loyalty. psychology and marketing, 21(5), 351–373. https://doi.org/10.1002/mar.20009 zeithaml, v.a., berry, l., and parasuraman, a. (1996). the behavioral consequences of service quality. journal of marketing, 60(2), 31-46. https://doi.org/10.1177/002224299606000203 https://doi.org/10.1123/jsm.2015-0039 https://doi.org/10.1080/15256480.2017.1305313 https://doi.org/10.19030/jss.v3i2.366 https://doi.org/10.24985/ijass.2017.29.1.64 https://doi.org/10.1016/j.jhtm.2016.05.005 https://doi.org/10.1002/mar.20009 https://doi.org/10.1177/002224299606000203 prada garcía et al. / european journal of government and economics 9(2), july 2020, 181-199 199 appendix table a.1. questionnaire adapted of future behavioural intentions (loyalty) of the golf courses (pradas, 2016). 1. i tell positive features other people about this golf course. 1 2 3 4 5 2. i recommend this golf course to anyone seeking my advice. 3. i encourage to my family and friends to play on this golf course. 4. for any service you might need consider this golf course as the first choice. 5. in the next few years i will play more on this golf course. 6. in the next few years i will play less son this golf course. 7. i may play in another golf course that offers better services. 8. even if the prices were higher, i would play on this golf course. 9. i am willing to pay a higher price for playing on this golf course for the service i receive. 10. i would switch to another golf course if i have a problem with the service. 11. if i have a problem, i tell other customers/people. 12. if i have a problem with this golf course, i will make a claim with external entities such as the consumers and users association. 13. if i have a problem with the service claim from the golf course manager. ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5841_v2_g evaluation of the perceived social impacts of the formula e grand prix of santiago de chile vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 155-169 european journal of government and economics issn: 2254-7088 evaluation of the perceived social impacts of the formula e grand prix of santiago de chile david parra-camacho a*, daniel michel duclos bastías b, frano giakoni ramírez c, samuel lópez-carril a a universidad de valencia, spain b pontificia universidad católica de valparaíso, chile c universidad católica de murcia, spain * corresponding author at: david.parra-camacho@uv.es article history. received 18 december 2019; first revision required 17 march 2020; accepted 17 april 2020. abstract. the aim of this study is to analyse residents' perception of the social impact of the formula e grand prix of santiago de chile. a sample of 414 residents was collected through a questionnaire made up of 46 items regarding possible positive and negative impacts. descriptive analysis, confirmatory factorial analysis and cluster analysis were performed. the results showed two groups with different perceptions of this sporting event: realistic (n=152) and favourable (n=162). sociodemographic variables referring to education level, civil status, income level, location of the family residence and political orientation contribute to significantly differentiating the clusters. the variables related to interest in the event, support for the celebration, attendance, participation and frequency of contact also contributed to significantly differentiating the groups. citizens’ high degree of support for holding the sporting event could be identified by a positive social representation of the event. keywords. clusters; resident’s perception; social impact; social representations; sport event jel codes. m14; l8; z2 doi. https://doi.org/10.17979/ejge.2020.9.2.5850 1. introduction sporting events are analysed from the perspective of different fields: economic, sociological or environmental. many cities often use the rhetoric associated with "sustainable development" or "sustainable regeneration" to justify the hosting of sporting events (smith, 2009). within sporting events, the analysis of the impact on society and the quality of life of residents has gained special relevance in recent years, as the lack of support for the holding of sporting events can generate devastating effects on the host community, increasing social and political tensions (gursoy, yolal, ribeiro and panosso, 2017). the social impact of sporting events has been analysed mainly through the study of residents' perception of various variables of interest (impacts on diverse areas, support for the celebration, identification with the event, linkage with the event, etc.). papers that explore the perceptions of subgroups of host residents can facilitate a better understanding of the social impact of these events by identifying potential winners and losers (ma, ma, wu and rotherham, 2013). these kinds of works are few (ma et al., 2013), and many insist on the need to make mailto:david.parra-camacho@uv.es https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 156 more contributions along this line to identify the factors that may define these groups (chen and tian, 2015). in this regard, several authors have pointed to the importance of assessing the reactions of subgroups of hosts at sporting events as a way of better understanding stakeholders (chen and tian, 2015; fredline and faulkner, 2000; ma and rotherham, 2015; zhou and ap, 2009). the formula e championship is made up of various events held on urban circuits in which electric single-seat race cars take part. this championship was born in the 2014-2015 season with eleven races, increasing the number of races to 13 in the 2018-2019 season. because this is a sporting event with a little history, the contributions to the analysis of the impact on the host localities are limited. therefore, this paper analyses the social impact of the formula e grand prix held in santiago de chile in 2018 by identifying groups of residents with different perceptions of the impacts associated with it. in addition, the variables that define these groups are identified to be able to determine the characteristics of the residents. 2. literature review 2.1. social impact of motor sport events the social impact of sporting events can be defined in terms of changes in the quality of life of residents as a consequence of hosting a sporting event (parra, calabuig, núñez and crespo, 2017). studies analysing the social impact of sporting events tend to consult the perception of stakeholders affected by the event, with residents being one of the groups on which research in this area has focused the most (parra, alonso-dos-santos and duclos, 2018a). many contributions in this area have focused their interest on the impact of mega-sports events (olympic games and football world cups) (e.g., chi, ouyang and xu, 2018; kim and kaplanidou, 2019; koenigstorfer and preuss, 2019). nevertheless, the social impact of other major sporting events has also been analysed, such as the tour de france (balduck, maes and buelens, 2011), the fiba world cup spain 2014 (gonzález-garcía, parra, calabuig and añó, 2016), the perth america's cup (soutar and mcleod, 1993) and valencia’s (llopis and gil, 2011; parra, añó, calabuig and ayora, 2016), america's cup soccer chile 2015 (parra, alonsodos-santos and duclos, 2018a), the barcelona world race (parra, elasri, triadó and aparicio, 2016), the australian open tennis (fredline, 2005), kaohsiung world games (ma et al., 2013) and the super bowl (kim and walker, 2012; lee and krohn, 2013). in this sense, motorsports events have also been the subject of analysis in various studies (añó, calabuig and parra, 2012; calabuig, parra, añó and ayora, 2014; cegielski and mules, 2002; cheng and jarvis, 2010; fredline and faulkner, 2000; fredline and faulkner, 2002; fredline, deery and jago, 2013; kim, jun, walker and drane, 2015; mackellar, 2013; mao and huang, 2016; zhou, 2010). motor sport events with cars, with the exception of rallies, in recent years, have tended to be held on urban circuits built specifically for the event or adapted for it (lefebvre and roult, 2011). for this reason, urban circuits tend to use public roads and resources such as parks, trails and david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 157 public areas that are not usually used for this purpose, all connected as circuits specifically designed for the motor sport events (mackellar, 2013). these events often generate a series of positive and negative impacts that can affect the host community. in this sense, most of the studies that have analysed the social repercussion of their celebration in urban circuits agree in highlighting the noise problems generated by singleseaters, traffic and the agglomeration of people in the areas in which they are celebrating as egregious negative impacts (añó et al., 2012; cegielski and mules, 2002; cheng and jarvis, 2010; fredline et al., 2013). other significant negative impacts are related to economic costs (kim et al., 2015) or concerns about the maintenance of public facilities (mackellar, 2013). on the other hand, several works highlight the positive impacts related to the improvement of the city image and its promotion as a tourist destination (añó et al., 2012; cegielski and mules, 2002; cheng and jarvis, 2010; kim et al., 2015; mao and huang, 2016; zhou, 2010) or the stimulation of the local economy (fredline et al., 2013). other psychosocial and intangible aspects are related to the community's increased pride in hosting the event (fredline et al., 2013; mao and huang, 2016), entertainment opportunities (fredline et al., 2013; kim et al., 2015), or the opportunity to meet new people (cheng and jarvis, 2010). 2.2. explanatory theories of the social impact of sporting events there are several theories that attempt to explain changes in residents' perceptions of the impacts of sporting events. in this paper, we will use the theory of social representations (moscovici, 1984) as a theoretical frame of reference to analyse residents' perceptions of formula e. this theory suggests that residents' feelings towards tourism and sporting events may be affected by their direct experiences, social interactions, personal and social values, and other sources of information, such as the media (fredline, 2005, ritchie, shipway and cleeve, 2009). social representations are difficult to change because they provide a framework through which new information is interpreted (fredline, 2005). this theory can help explain how various groups of people understand and respond to social issues. it is particularly appropriate when the topic of study involves multiple social perspectives or accompanies conflicts associated with possible change, such as the hosting of a sporting event (zhou and ap, 2009). in the research on the social impact of sporting events, several works have used this theoretical framework of reference (chen and tian, 2015; cheng and jarvis, 2010; fredline, 2005; fredline et al., 2013; fredline and faulkner, 2000; zhou, 2010; zhou and ap, 2009). 2.3. cluster analysis of residents cluster analysis has been used to categorise residents' perceptions of the impacts of sporting events (e.g., calabuig et al., 2014; chen and tian, 2015; fredline et al., 2013; fredline and faulkner, 2000, 2002; parra, añó, et al., 2016; parra, calabuig, añó, ayora and núñez, 2014; david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 158 zhou, 2010; zhou and ap, 2009). cluster analysis helps to identify the community and the differences between them, as well as provide the foundation for event planning and management (chen and tian, 2015). prior studies have identified two groups (e.g., ma and rotherham, 2015; ma et al., 2013; zhou, 2010), three groups (e.g., calabuig et al., 2014; chen and tian, 2015; ma et al., 2013; parra, añó et al., 2016; parra et al., 2014; zhou and ap, 2009) and five groups (e.g., fredline and faulkner, 2000; fredline et al., 2013). 3. method 3.1. participants in this study, 414 people living in chile were interviewed about their perception of the impacts generated by the formula e race in santiago de chile. the event being studied is the santiago eprix, which was attended by 25,000 spectators at the inaugural race held on february 3, 2018 at the parque forestal urban circuit and was the fourth race of the 2017-2018 season. table 1 shows the sociodemographic characteristics of the sample. the majority of those interviewed are residents of the metropolitan region of santiago de chile (55.1%), aged between 25 and 54 (81.6%), men (79.2%) and have university-level education (88.4%). table 1. sociodemographic characteristics of the sample. n % sex man 328 79.2 woman 86 20.8 age 18-24 years 42 10.1 25-54 year 338 81.6 more than 55 years 34 7.9 occupation employed 294 71.0 unemployed 20 4.8 part-time employee 50 12.1 student 38 9.2 other (housekeeper, retired, etc.) 12 2.9 level of studies high school 16 3.9 technical 32 7.7 university 366 88.4 civil status single 270 65.2 married 122 29.5 divorced/separated 22 5.3 income level less than 10.000.000 clp 212 51.2 10.000.001-15.000.000 clp 66 15.9 15.000.001-20.000.000 clp 64 15.5 20.000.001-25.000.000 clp 36 8.7 more than 25.000.001 clp 36 8.7 family residence metropolitan region of santiago de chile 228 55.1 other regions 186 44.9 political orientation left 142 34.3 centre 170 41.1 right 102 24.6 david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 159 3.2. instrument the instrument is composed of 46 items adapted from previous studies about positive and negative impacts associated with the celebration of sport events (añó et al., 2012; calabuig et al., 2014; fredline et al., 2013; gonzález-garcía et al., 2016; parra, aguado and núñez, 2015; parra, añó, et al., 2016). the items proposed were classified by categories of positive and negative impacts. to make this classification, the impact categories proposed by preuss and solberg (2006) and fredline (2004) were taken into account. in addition, the different contributions made to the dimensions proposed by the case studies are referred to at the beginning of this section. thus, the proposed items were clustered, on the one hand, into five initial dimensions related to positive impacts (33 items in total), which were denominated as follows: socio-economic impact (7 items), impact on urban development and infrastructure (6 items), political and administrative impact (5 items), psychosocial impact (3 items), sports impact (6 items) and sociocultural impact (6 items). on the other hand, the negative impacts (13 items in total) were grouped into three dimensions: socioeconomic impact (7 items), environmental impact (3 items) and sociocultural impact (3 items). a five-point likert scale was used, where 1 means totally disagree and 5 means totally agree. the following sociodemographic variables were included: age, gender, education level, occupation, family residence, income level, civil status and political orientation. other variables of interest were also included in the questionnaire to define the characteristics of the resident groups: interest in the sport, attendance at the event, support for the celebration, involvement in tourism or events, public participation as volunteers or workers in sporting events and frequency of contact with the area in which the event is held. 3.3. procedure a convenience sampling method such as that used in other work in this area was used (e.g., garcía-pascual, parra and gonzález-garcía, 2019; gursoy and kendall, 2006; oshimi and harada, 2018; parra, alonso-dos-santos and duclos, 2018b, 2019). the questionnaires were distributed and collected after the event (between february and june 2018). 3.4. statistical analyses first, two confirmatory factorial analyses (cfa) for the scale of positive impacts (33 items) and negative impacts (13 items) were carried out through the program for structural equation models eqs 6.2. several recommended goodness-of-fit indexes were used to check the overall fit of the model through the cfa (kline, 2005): the normalised chi-square (χ2/df), the root mean square error of approximation (rmsea), the non-normalised fit index (nnfi), the comparative fit index (cfi) and the incremental fit index (ifi). the reliability of the scale was checked through cronbach’s alpha, composite reliability (cr) and average variance extracted (ave). convergent david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 160 validity was also analysed through the significance of factor charges (p<.05) in their respective dimensions and the associated t-test values. to contrast discriminant validity, the method suggested by fornell and larcker (1981) was used, which consists of checking whether the square root of the ave value of a given factor is greater than the correlation coefficients between the factor and any other factor of the proposed scale. second, an analysis of clusters was carried out using the statistical program spss version 24.0 with the items derived from the cfa. two methods of estimation (hierarchical and nonhierarchical) of the cluster solution were combined to optimise the results. the hierarchical cluster analysis was performed using the ward method grouping process and as a measure of similarity, the euclidean distance squared. based on the groups proposed in the previous analysis, a non-hierarchical analysis was applied using the k-means method, using as initial centres the means of the variables obtained for each cluster solution of the hierarchical analysis. to define the characteristics of group profiles and to evaluate predictive validity, anovas and chi-square tests were performed with sociodemographic variables and variables of interest that were not included in the initial analysis. the value of the contingency coefficient (c) was also used to check the intensity of the association or the size of the effect of the related variables. 4. results 4.1. confirmatory factor analysis of scales the cfa results on the scale of positive impacts showed an adequate overall fit after several re-specifications of the initial model in which six indicators were removed from the initial 33: [(sbχ2=716.99, df=309, p<.01); (χ2=1367.85, df=309) (χ2/df=4.42); (rmsea=.080; ic=.07-.08); nnfi=.95; cfi=.95; ifi=.95)]. the final model consists of 27 indicators grouped into the following factors: socioeconomic impact (6 items), impact on infrastructure and urban development (6 items), political and administrative impact (5 items), psychosocial impact (3 items), sociocultural impact (4 items) and sports impact (3 items). the scale reliability indicators showed adequate values, oscillating between .95 and .96 for cronbach's alpha, between .94 and .96 for the cr, and between .78 and .85 for the ave. convergent validity was tested by observing the values of factorial loads of items oscillating between .82 and .96, being significant (p<.05) and higher than the .60 recommended by kline (2005). additionally, t values oscillating between 16.05 and 39.71 were significant at the level of .05 (t>1.96). these data allowed us to ensure convergent validity. finally, to assess the discriminant validity, it was proven that all the correlations between the factors (ranging from .30 to .91) were lower than the square root of the ave. this criterion was fulfilled in all pairs of factors except between the socioeconomic impact dimension and the sociocultural impact dimension of the positive impact scale. nevertheless, the model with the combination of the two factors did not improve the global fit and the validity of the scale, so it was decided to maintain the initial factor structure. david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 161 in the case of the scale of negative impacts, it was also necessary to eliminate three indicators from the initial 13 to obtain an adequate global adjustment: [(s-bχ2=76.87, df=32, p<.01); (χ2=102.91, df=32) (χ2/df=3.21); (rmsea=.083; ic=.05-.10); nnfi=.95; cfi=.97; ifi=.97)]. the final model consists of 10 indicators clustered into the following factors: negative socioeconomic impact (4 items), negative sociocultural impact (3 items) and negative environmental impact (3 items). scale reliability indicators showed adequate values, ranging from .84 to .90 for cronbach's alpha; .87 to .92 for cr; and .70 to .75 for the ave. for this scale, the convergent validity criteria were also fulfilled by observing the significance of the values of t-tests (t>1.96) associated with the factorial charges of the items of every factor, oscillating between .73 and .98. similarly, correlations between pairs of factors (oscillating between .15 and .76) showed values lower than the square root of the ave. 4.2. cluster analysis of differences between population groups cluster analysis allowed the identification of two clusters that presented a suitable theoretical interpretation for the study (see table 2). the significance level for the anova results was set at p<.000, indicating significant differences and supporting the statistical validity of the research. the two identified clusters were named "favourable" (36.7%) and "realistic" (63.3%): 1. residents labelled as favourable showed a clear positive trend in almost all the possible positive impacts analysed, as they have values higher than 4 or close to this value (see table 1): socioeconomic impact (m=4.25), impact on infrastructure and urban development (m=3.82), political and administrative impact (m=4.56), psychosocial impact (m=4.18), sports impact (m=3.67) and sociocultural impact (m=4.33). on the other hand, they tended to minimise negative impacts: negative socioeconomic impact (m=2.78), negative sociocultural impact (m=1.76) and environmental impact (m=1.79). 2. residents labelled as realistic have variable scores depending on the dimension analysed, e.g., impact on infrastructure and urban development (m=1.57), psychosocial impact (m=1.56) and sports impact (m=1.87) show reduced scores. nevertheless, higher scores (above 3) are observed in the factors of socioeconomic impact (m=3.30), political and administrative impact (m=3.62) and sociocultural impact (m=3.32). in the case of negative impact dimensions, the negative socioeconomic impact factor score (m=4.18) should be highlighted. this group of residents was called "realistic", along the lines highlighted by other authors such as fredline and faulkner (2000), due to their more realistic and neutral perception of the impacts associated with events, as they view some impacts with a positive trend, others with a negative trend and others with a neutral trend. this term has also been used in other studies that presented groups with similar characteristics (calabuig et al., 2014; chen and tian, 2015; parra, añó et al., 2016). david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 162 table 2. average scores for each variable in the two clusters obtained through the k-means method. realistic (n = 152) favourable (n = 262) positive impacts factor 1: socioeconomic impact 3.30 (1.36) 4.25 (.73) formula e brings economic benefits to the city of santiago 3.50 (1.45) 4.59 (.69) due to formula e tourism in santiago has increased throughout the year. 2.92 (1.62) 3.83 (1.16) formula e increases local commerce and business. 3.38 (1.48) 4.37 (.90) formula e increases private investment and business opportunities. 3.61 (1.40) 4.29 (.96) formula e increases consumption in the city. 3.23 (1.51) 4.18 (.93) formula e provides great benefits in the neighbourhood where it is held. 3.17 (1.61) 4.24 (.96) factor 2: impact on urban development and infrastructure 1.57 (.77) 3.82 (.88) formula e has improved infrastructure in the city (accesses, streets, street furniture...). 1.69 (1.09) 4.13 (.85) formula e has improved the public transport system in santiago. 1.31 (.74) 3.29 (1.23) the infrastructures created by formula e are beneficial for citizens. 1.63 (1.06) 4.03 (1.04) formula e has regenerated the zone's infrastructures (streets, urban furniture, ...). 1.77 (1.17) 3.99 (1.07) formula e has improved social services in the city. 1.37 (.70) 3.55 (1.21) the infrastructures created by formula e are useful for other activities. 1.63 (.91) 3.93 (1.07) factor 3: political/administrative impact 3.62 (1.30) 4.56 (.73) formula e has improved santiago's image in the world. 3.56 (1.43) 4.57 (.85) the celebration of formula e has facilitated the opening of santiago to the world. 3.66 (1.42) 4.63 (.72) santiago's formula e increases chile's international recognition. 3.70 (1.36) 4.55 (.84) formula e promotes the city as a tourist destination. 3.56 (1.44) 4.53 (.84) formula e shows the capacity of santiago's inhabitants to host and organise major sporting events. 3.63 (1.42) 4.50 (.84) factor 4: psychosocial impact 1.56 (.73) 4.18 (.74) the celebration of formula e makes santiago a more attractive city to live in. 1.73 (1.04) 4.30 (.83) the celebration of formula e makes santiago a safer city. 1.41 (.71) 3.89 (1.04) formula e makes me proud to live in santiago. 1.54 (.81) 4.36 (.84) factor 5: sports impact 1.87 (.60) 3.67 (.84) i think that thanks to formula e the sport practice of the citizens of santiago has increased. 1.37 (.69) 3.54 (1.13) thanks to formula e, the number of sports facilities has increased. 1.32 (.70) 3.34 (1.26) formula e has increased grants and sports support to the city's clubs. 1.37 (.72) 3.28 (1.36) factor 6: socio-cultural impact 3.32 (1.36) 4.33 (.66) formula e is designed for the entertainment of santiago's people. 3.37 (1.54) 4.05 (1.08) formula e provides citizens with an opportunity to meet new people. 3.21 (1.56) 4.33 (.73) formula e improves the solidarity and hospitality of citizens with visitors. 3.01 (1.65) 4.33 (.82) formula e promotes cultural exchange and understanding of other cultures. 3.14 (1.59) 4.45 (.72) negative impacts factor 1: negative socioeconomic impact 4.18 (1.01) 2.78 (1.02) formula e disrupts residents' daily lives excessively. 4.07 (1.25) 2.58 (1.31) i think formula e distorts and hinders the normal functioning of the city. 4.13 (1.21) 2.47 (1.26) during the celebration of formula e many citizens avoid the area in which the event was held. 4.25 (1.10) 3.14 (1.26) formula e causes restrictions on access to public facilities and services. 4.26 (1.07) 2.91 (1.26) factor 2: negative sociocultural impact 1.82 1.76 formula e encourages inappropriate behaviour (drug use, alcohol, prostitution, etc.). 1.88 (1.15) 1.80 (1.15) formula e encourages the development of dangerous driving behaviours. 1.97 (1.30) 1.78 (1.05) formula e increases levels of crime and vandalism in the city. 1.61 (.92) 1.70 (1.09) factor 3: negative environmental impact 2.27 (1.23) 1.79 (.94) formula e causes damage to the environment and natural areas. 2.25 (1.41) 1.72 (1.00) formula e increases the pollution of the city. 2.00 (1.23) 1.64 (1.09) formula e increases the volume of waste in the zone. 2.57 (1.49) 1.99 (1.12) david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 163 4.3. profile and characteristics of the groups table 3 shows the results of the sociodemographic variables and of interest according to the cluster of belonging. from the point of view of sociodemographic variables, the following groups contribute to significantly differentiating the identified groups: those related to the education level of studies, civil status, income level, location of the family residence and political orientation. variables of interest related to the event were the following: interest in formula e, support for the celebration, attendance at the event, participation as a volunteer or worker and frequency of contact with the area in which it is held. the group of realist residents is characterised by a greater proportion of men (76.34%), those aged between 25 and 54 (82.44%), those with university studies (88.55%), those employed (70.23%), those who are single (67.18%), those with an annual income of less than 10.000.000 clp (61.07%), those with a political orientation towards the centre (45.80%) and left (32.82%) and those with a family residence in santiago de chile (60.31%). this group has a higher proportion of people who are interested in formula e (50.38%). in addition, they show their support for the celebration of this event in the city during the next race editions (67.54%) and have a lower proportion of residents who say they have attended the event (15.27%), and most have participated as volunteers or workers in sports events (63.36%). the frequency of contact with the zone in which the event is held is very high, as most indicate that they pass through the zone several days a week (70.99%). the group of favourable residents is characterised by a greater proportion of men (84.21%), those aged between 25 and 54 (80.26%), those with university studies (88.16%), those employed (72.37%), those who are single (61.84%), those with an income of less than 10.000.000 clp per year (34.21%), those with a political orientation towards the left (36.84%) and the centre (32.89%) and those with a family residence in other regions (53.95%). this group has a higher proportion of people who are interested in formula e (81.58%). they also clearly show their support for the celebration of this event in the city during the next race editions (95.89%) and have a lower proportion of residents who say they have attended the event (42.11%), and the majority have participated as volunteers or workers in sports events (51.32%). the frequency of contact with the zone in which the event is held is high as a large percentage indicates that they pass through the zone several days a week (40.79%) or several days a month (21.05%). david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 164 table 3. sociodemographic profile of the different groups (clusters). variable alternative of response 1 realistic (n = 152) 2 favourable (n = 262) sex χ2(2)=3.62, p=.06 c =.09 man 76.34% 84.21% woman 23.66% 15.79% age χ2(2)=5.11, p=,08 c =.11 18-24 years 11.45% 7.89% 25-54 years 82.44% 80.26% more than 55 years 6.11% 11.84% (1) educational level* χ2 (4)=6.47, p=.04 c =.12 high school education 5.34% 1.32% technical 6.11% (2) 10.53% university 88.55% 88.16% occupation χ2 (8)=8.98, p=.06 c =.15 employed 70.23% 72.37% unemployed 4.58% 5.26% part-time employee 9.92% 15.79% student 11.45% (2) 5.26% other (housekeeper, retired, etc.) 3.82% 1.32% civil state*** χ2 (2)=12.97, p=00 c =.17 single 67.18% 61.84% married 30.53% 27.63% separated/divorced /widower 2.29% 10.53% (1) annual income level *** χ2 (4)=33.67, p=<.001 c =.27 less than 10.000.000 clp/year 61.07% (2) 34.21% 10.000.001-15.000.000 clp 12.98% 21.05% (1) 15.000.001-20.000.000 clp 10.69% 23.68% (1) 20.000.001-25.000.000 clp 9.16% 7.89% more than 25.000.001 clp 6.11% 13.16% (1) family residence** χ2 (1)=7.90, p=,01 c =.14 metropolitan region of santiago de chile 60.31% (2) 46.05% other regions 39.69% 53.95% (1) political orientation** χ2 (2)=7,44, p=.02 c =.13 left 32.82% 36.84% centre 45.80% (2) 32.89% right 21.37% 30.26% (1) interest in formula e*** χ2 (2)=42.15, p=<.001 c =.30 yes 50.38% 81.58% (1) no 20.61% (2) 3.95% indifferent 29.01% (2) 14.47% support for the celebration of the event*** χ2 (1)=42.53, p=<.001 c =.32 yes 67.54% 95.89% (1) no 32.46% (2) 4.11% attendance to the event*** χ2 (1)=36.83, p=<.001 c =.29 yes 15.27% 42.11% (1) no 84.73% (2) 57.89% connection with events or associated activity χ2 (1)=.19, p=.66 c =.02 yes 28.24% 30.26% no 71.76% 69.74% participation as a volunteer or worker* χ2 (1)=5.76, p=.02 c =.12 yes 63.36% (2) 51.32% no 36.64% 48.68% (1) frequency of contact with the zone in which it is held*** χ2 (4)=39.98, p=<.001 c =.30 several days a week 70.99% (2) 40.79% several days per month 12.98% 21.05% (1) several weeks per year 7.63% 13.16% a few days a year 3.82% 10.53% (1) never 4.6% 14.5% (1) note. * indicates statistically significant relationship or statistically significant mean differences *p<.05; **p< .01; ***p< .001; c= contingency coefficient; (1) (2) the results are based on bilateral tests with a level of significance .05. in the results table, it is shown for each significant pair the key of the group of residents with the lowest column proportion below the group of residents with the highest column proportion. david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 165 5. discussion and conclusions this study contributes to analysing residents' perceptions of sporting events based on segmentation into groups with different reactions to the impacts of these types of events. first, the validity and reliability of the scales of positive and negative impacts were assessed, identifying 6 factors and 3 factors, respectively. in line with previous investigations (calabuig et al., 2014; chen and tian, 2015; ma and rotherham, 2015; zhou and ap, 2009), in this work, two groups were found consisting of residents with different profiles. in the two groups identified (realistic and favourable), they show high percentages of support for the celebration of formula e (67.5% and 95.9%, respectively). this is similar to what was observed in studies on other motorsports events, such as the formula 1 grand prix in australia or the gold coast indy car (fredline and faulkner, 2000, 2002). nevertheless, in other works on this type of event, the inverse trend has been observed, as in the case of the formula 1 european grand prix (añó et al., 2012; calabuig et al., 2014). the residents classified within the group of realists seem to show a higher level of appreciation of the real impact of the event, highlighting some positive socioeconomic, political, administrative and sociocultural impacts. nevertheless, there are also negative socioeconomic costs as a consequence of hosting the event. thus, this group shows a more circumspect social representation of the event and perhaps are less conditioned or influenced by direct experiences or personal factors. in other studies, this group has been identified as realists (chen and tian, 2015; fredline and faulkner, 2000; parra and duclos, 2013), ambivalent (fredline and faulkner, 2002) or unconcerned (fredline et al., 2013), indifferent (parra, añó et al., 2016) or moderate (parra et al., 2014). on the other hand, the residents designated as favourable show an optimistic trend in all the impact areas analysed: socioeconomic impact, impact on infrastructure and urban development, political and administrative impact, psychosocial impact, sporting impact and sociocultural impact. on the other hand, they tended to minimise negative impacts at the socioeconomic, sociocultural and environmental levels. this group presents a positive social representation of the event that is characterised by its high degree of support for the celebration of the event and interest in the sport. this group of residents has also been identified in all of the studies that analyse the perception of the impacts of sporting events (calabuig et al., 2014; chen and tian, 2015; fredline and faulkner, 2000; fredline and faulkner, 2002; fredline et al., 2013; parra, añó et al., 2016; parra et al., 2014; parra and duclos, 2013; zhou, 2010; zhou and ap, 2009). as cheng and jarvis (2010) point out, this type of event provides an opportunity for the enjoyment and entertainment of the members of this collective; the benefits and costs associated with the event are not perceived in the same way by other groups with lesser identification with the event. sociodemographic variables such as gender, age, level of education or occupation show little or no effect in determining the subgroup of each resident. this finding is consistent with previous studies, which found that gender, age, and level of education (chen and tian, 2015; ma et al., 2013; parra et al., 2014; zhou, 2010; zhou and ap, 2009) do not significantly affect david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 166 such conglomerates. nevertheless, other variables such as the level of income create significant effects in the differentiation of the groups, coinciding with what has been observed in other studies (ma and rotherham, 2015; ma et al., 2013). additionally, the civil status and the origin of the family residence generated significant effects in the differentiation of the groups, with the majority of the group with a more positive attitude towards the event having their family residence in regions other than santiago de chile. finally, the political orientation variable generated significant effects in the differentiation of the groups, which is in line with what was observed in other works (parra, añó, et al., 2016). regarding the variables of interest related to the event, a significant effect was observed in the differentiation of the groups in those variables related to support, interest in sports, attendance, participation as a worker or volunteer in sports events and frequency of contact with the zone where the event is held. this result is in line with what was observed in other studies in which it was found that the variables of support, interest and attendance (calabuig et al., 2014; fredline et al., 2013; ma et al., 2013; ma and rotherham, 2015; parra and duclos, 2013; zhou, 2010), participation as a worker or volunteer and frequency of contact with the zone (fredline and faulkner, 2002) had a significant effect on group differences. with respect to the relationship with tourism or the economic activity derived from the events, these variables did not show a significant effect in the differentiation of the clusters, coinciding with what was observed in other works on diverse sport events (ma and rotherham, 2015; parra, añó, et al., 2016). nevertheless, in other studies, it has been observed that these variables have an effect on the distinction between groups (fredline and faulkner, 2002; parra et al., 2014). 5.1. practical implications from the point of view of the analysis of the social impact of events, it is important to consider residents as a heterogeneous group since the perception of a sporting event can vary according to different factors and variables (calabuig et al., 2014). in this sense, the identification of groups of residents with different perceptions allows the organisers and the administration to improve the understanding of the different segments of citizens with the aim of better managing the impacts and trying to satisfy the needs of these groups (deery and jago, 2010). accordingly, organisers and administrations must bear in mind the characteristics and behaviours that define these groups to maximise the benefits of hosting this type of event (parra, añó, et al., 2016). this work shows that there is a high degree of citizens’ support for the celebration of the event that could be identified by a positive social representation of the sporting events. nevertheless, it is necessary for the organisers to pay attention to some of the costs highlighted by residents, such as those related to the inconvenience caused by the disruption of daily life, the distortion of the normal functioning of the city or restrictions on access to public facilities and services in the area. according to ma et al. (2013), event organisers and local authorities should consciously plan events related to improving the quality of life in the host areas if they want a positive long-term relationship. david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 167 5.2. limitations and future research lines to conclude, it should be noted that these results should not be generalised to all city residents, as a consequence of the limitations of using a convenience sampling frame. it is therefore necessary to interpret them with caution and to increase the representation of some demographic groups in future studies. since the perceptions of the population may change as a consequence of various social conditioning factors, it is worthwhile to carry out consultations over time to observe possible changes in the configuration of clusters. in the same way, it would be advisable to compare perceptions about the impacts of the event in different cities that host formula e events. in addition, future studies could change the means of addressing some variables in a dichotomous way or with alternative responses on a likert scale to better understand the behaviour of the population. references añó, v., calabuig, f., and parra, d. (2012). impacto social de un gran evento deportivo: el gran premio de europa de fórmula 1. cultura, ciencia y deporte, 7(19), 53-65. https://doi.org/10.12800/ccd.v7i19.23 balduck, a., maes, m., and buelens, m. (2011). the social impact of the tour de france: comparisons of residents' pre-and post-event perceptions. european sport management quarterly, 11(2), 91-113. https://doi.org/10.1080/16184742.2011.559134 calabuig, f., parra, d., añó, v., and ayora, d. (2014). análisis de la percepción de los residentes sobre el impacto cultural y deportivo de un gran premio de fórmula 1. movimento, 20(1), 261-280. https://doi.org/10.22456/1982-8918.40260 cegielski, m., and mules, t. (2002). aspects of residents' perceptions of the gmc 400canberra's v8 supercar race. current issues in tourism, 5(1), 54-70. https://doi.org/10.1080/13683500208667908 chen, f., and tian, l. (2015). comparative study on residents' perceptions of follow-up impacts of the 2008 olympics. tourism management, 51, 263-281. ttps://doi.org/10.1016/j. tourman.2015.05.029 cheng, e., and jarvis, n. (2010). residents' perception of the social-cultural impacts of the 2008 formula 1 singtel singapore grand prix. event management, 14(2), 91-106. https://doi.org/10.3727/152599510x12766070300849 chi, c. g. q., ouyang, z., and xu, x. (2018). changing perceptions and reasoning process: comparison of residents’ pre-and post-event attitudes. annals of tourism research, 70, 39-53. https://doi.org/10.1016/j.annals.2018.02.010 deery, m., and jago, l. (2010). social impacts of events and the role of anti-social behaviour. international journal of event and festival management, 1(1), 8-28. https://doi.org/ 10.1108/17852951011029289 fornell, c., and larcker, d. f. (1981). evaluating structural equation models with unobservable variables and measurement error. journal of marketing research, 18(1), 39-50. https://doi.org/10.2307/3151312 fredline, e. (2004). host community reactions to motorsport events: the perception of impact on quality of life. in b. w. ritchie, and d. adair (eds.), sport tourism: interrelationships, impacts and issues (pp. 155-173). clevedon, reino unido: channel view publications. fredline, e. (2005). host and guest relations and sport tourism. sport in society, 8(2), 263-279. https://doi.org/10.1080/17430430500087328 fredline, e., and faulkner, b. (2000). host community reactions: a cluster analysis. annals of tourism research, 27(3), 763-784. https://doi.org/10.1016/s0160-7383(99)00103-6 https://doi.org/10.12800/ccd.v7i19.23 https://doi.org/10.1080/16184742.2011.559134 https://doi.org/10.22456/1982-8918.40260 https://doi.org/10.1080/13683500208667908 https://doi.org/10.1016/j.tourman.2015.05.029 https://doi.org/10.1016/j.tourman.2015.05.029 https://doi.org/10.3727/152599510x12766070300849 https://doi.org/10.1016/j.annals.2018.02.010 https://doi.org/10.1108/17852951011029289 https://doi.org/10.1108/17852951011029289 https://doi.org/10.2307/3151312 https://doi.org/10.1080/17430430500087328 https://doi.org/10.1016/s0160-7383(99)00103-6 david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 168 fredline, e., and faulkner, b. (2002). variations in residents' reactions to major motorsport events: why residents perceive the impacts of events differently. event management, 7(2), 115-125. https://doi.org/10.3727/152599501108751524 fredline, l., deery, m., and jago, l. (2013). a longitudinal study of the impacts of an annual event on local residents. tourism planning and development, 10(4), 416-432. https://doi.org/10.1080/21568316.2013.779314 garcía-pascual, f., parra-camacho, d., and gonzález-garcía, r. j. (2019). validación de una escala para el análisis de las percepciones de los residentes sobre el impacto social de un centro deportivo. movimento, 25. https://doi.org/10.22456/1982-8918.86497 gonzález-garcía, r. j., parra, d., calabuig, f., and añó, v. (2016). percepción de los residentes sobre el impacto del mundobasket 2014 en gran canaria y apoyo a la celebración de eventos deportivos. revista iberoamericana de psicología del ejercicio y el deporte, 11(2), 279-288. gursoy, d., and kendall, k. w. (2006). hosting mega events: modeling locals' support. annals of tourism research, 33(3), 603-623. https://doi.org/10.1016/j.annals.2006.01.005 gursoy, d., yolal, m., ribeiro, m.a., and panosso, a. (2017). impact of trust on local residents’ mega-event perceptions and their support. journal of travel research, 56(3), 393-406. https://doi.org/10.1177/0047287516643415 kim, w., jun, h., walker, m., and drane, d. (2015). evaluating the perceived social impacts of hosting large-scale sport tourism events: scale development and validation. tourism management, 48, 21-32. https://doi.org/10.1016/j.tourman.2014.10.015 kim, c., and kaplanidou, k. (2019). the effect of sport involvement on support for mega sport events: why does it matter. sustainability, 11(20), 5687. https://doi.org/ 10.3390/su11205687 kim, w., and walker, m. (2012). measuring the social impacts associated with super bowl xliii: preliminary development of a psychic income scale. sport management review, 15(1), 91-108. https://doi.org/10.1016/j.smr.2011.05.007 kline, r. b. (2005). principles and practice of structural equation modeling (2nd ed.). new york: the guilford press. koenigstorfer, j., and preuss, h. (2019). olympic games-related values and host country residents’ pre-event evaluations in the run-up to the 2016 olympic games. journal of global sport management, 1-26. https://doi.org/10.1080/24704067.2019.1669065 lee, s., and krohn, b. d. (2013). a study of psychological support from local residents for hosting mega-sporting events: a case of the 2012 indianapolis super bowl xlvi. event management, 17(4), 361-376. https://doi.org/10.3727/152599513x137693924445 85 lefebvre, s., and roult, r. (2011). formula one’s new urban economies. cities, 28(4), 330339. https://doi.org/10.1016/j.cities.2011.03.005 llopis, m.p., and gil, i. (2011). un gran evento deportivo: perspectiva de los residentes de la ciudad de acogida. gran tour, 4, 32-61. ma, s. c., ma, s. m., wu, h. j., and rotherham, i. d. (2013). host residents’ perception changes on major sport events. european sport management quarterly, 13, 511–536. https://doi.org/10.1080/16184742.2013.838980 ma, s. c., and rotherham, i.d. (2015). residents’ changed perceptions of sport event impacts: the case of the 2012 tour de taiwan. leisure studies, 1-22 https://doi.org/10.1080/02614367.2015.1035313 mackellar, j. (2013). world rally championship 2009: assessing the community impacts on a rural town in australia. sport in society: cultures, commerce, media, politics, 16(9), 1149-1163. https://doi.org/10.1080/17430437.2013.790893 mao, l.l., and huang, h. (2016). social impact of formula one chinese grand prix: a comparison of local residents’ perceptions based on the intrinsic dimension. sport management review, 19, 306–318. https://doi.org/10.1016/j.smr.2015.08.007 moscovici, s. (1984). the phenomenon of social representations. in r. farr and s. moscovici (eds.), social representations (pp. 3–69). cambridge: cambridge university press. oshimi, d., and harada, m. (2018). host residents’ role in sporting events: the city image perspective. sport management review. https://doi.org/10.1016/j.smr.2018.04.002 https://doi.org/10.3727/152599501108751524 https://doi.org/10.1080/21568316.2013.779314 https://doi.org/10.22456/1982-8918.86497 https://doi.org/10.1016/j.annals.2006.01.005 https://doi.org/10.1177/0047287516643415 https://doi.org/10.1016/j.tourman.2014.10.015 https://doi.org/10.3390/su11205687 https://doi.org/10.3390/su11205687 https://doi.org/10.1016/j.smr.2011.05.007 https://doi.org/10.1080/24704067.2019.1669065 https://doi.org/10.3727/152599513x13769392444585 https://doi.org/10.3727/152599513x13769392444585 https://doi.org/10.1016/j.cities.2011.03.005 https://doi.org/10.1080/16184742.2013.838980 https://doi.org/10.1080/02614367.2015.1035313 https://doi.org/10.1080/17430437.2013.790893 https://doi.org/10.1016/j.smr.2015.08.007 https://doi.org/10.1016/j.smr.2018.04.002 david parra-camacho et al. / european journal of government and economics 9(2), july 2020, 155-169 169 parra, d., aguado, s., and núñez, j. m. (2015). los costes de la celebración de un evento deportivo: percepción de la comunidad de acogida. journal of sports economics and management, 5(1), 17-36. parra, d., alonso-dos-santos, m., and duclos. d. (2018a). residents' perceptions of the negative impacts of the copa américa de fútbol in chile: preand post-event comparison. journal of physical education and sport, 18(3), 1290-1297. https://doi.org/10.7752/jpes.2018.s3192 parra, d., alonso-dos-santos, m., and duclos. d. (2018b). residents' perception of the negative impacts of a sporting event: scale validation. journal of physical education and sport, 18(3), 1298-1303. https://doi.org/10.7752/jpes.2018.s3193 parra-camacho, d., alonso-dos-santos, m., and duclos. d. (2019). the relationship between factors that contribute to support and future intentions in relation to a major sporting event. academia revista latinoamericana de administración, 32(4), 442-54. https://doi.org/10.1108/arla-02-2019-0042 parra, d., añó, v., calabuig, f., and ayora, d. (2016). percepción de los residentes sobre el legado de la america’s cup. cuadernos de psicología del deporte, 16(1), 309-322. parra, d., calabuig, f., añó, v., ayora, d., and núñez, j.m. (2014). el impacto de un evento deportivo mediano: percepción de los residentes de la comunidad de acogida. retos: nuevas tendencias en educación física, deporte y recreación, 26, 88-93. parra, d., calabuig, f., núñez, j.m., and crespo, j. (2017). the relevance of the social impact of sports events in the context of public financing of sport. en: ratten, v. ferreira j.j. (eds.), sport entrepreneurship and innovation (pp. 118-141). new york: routledge. parra, d., and duclós, d. (2013). percepción de los residentes sobre el impacto socioeconómico de un evento deportivo: análisis de segmentos y perfil del residente. journal of sports economics and management, 3(1), 4-32. parra, d., elasri, a., triadó, x.m., and aparicio, p. (2016). análisis de la relación entre los beneficios y los costes percibidos y la satisfacción del residente con la celebración de un evento deportivo: efecto mediador del valor percibido. revista de psicología del deporte, 25(3), 59-63. preuss, h., and solberg, h. a. (2006). attracting major sporting events: the role of local residents. european sport management quarterly, 6(4), 391-411. https://doi.org/ 10.1080/16184740601154524 ritchie, b. w., shipway, r., and cleeve, b. (2009). resident perceptions of mega-sporting events: a non-host city perspective of the 2012 london olympic games. journal of sport and tourism, 14(2), 143-167. https://doi.org/10.1080/14775080902965108 smith, a. (2009) theorising the relationship between major sport events and social sustainability. journal of sport and tourism, 14(2-3), 109-120. https://doi.org/ 10.1080/14775080902965033 soutar, g., and mcleod, p. (1993). residents' perceptions on impact of the america's cup. annals of tourism research, 20(3), 571-582. https://doi.org/10.1016/0160-7383 (93)90010-z zhou, j. y. (2010). resident perceptions toward the impacts of the macao grand prix. journal of convention and event tourism, 11(2), 138-153. https://doi.org/10.1080/15470 148.2010.485179 zhou, j. y., and ap, j. (2009). residents' perceptions towards the impacts of the beijing 2008 olympic games. journal of travel research, 48(1), 78-91. https://doi.org/10.1177 /0047287508328792 https://doi.org/10.7752/jpes.2018.s3192 https://doi.org/10.7752/jpes.2018.s3193 https://doi.org/10.1108/arla-02-2019-0042 https://doi.org/10.1080/16184740601154524 https://doi.org/10.1080/16184740601154524 https://doi.org/10.1080/14775080902965108 https://doi.org/10.1080/14775080902965033 https://doi.org/10.1080/14775080902965033 https://doi.org/10.1016/0160-7383(93)90010-z https://doi.org/10.1016/0160-7383(93)90010-z https://doi.org/10.1080/15470148.2010.485179 https://doi.org/10.1080/15470148.2010.485179 https://doi.org/10.1177/0047287508328792 https://doi.org/10.1177/0047287508328792 ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5850_g the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach vol.9 • no.1 2020 issn: 2254-7088 european journal of government and economics 9(1), june 2020. european journal of government and economics issn: 2254-7088 number 9, issue 1, june 2020 doi: https://doi.org/10.17979/ejge.2020.9.1 the politics of renewable power in spain 5-25 doi: https://doi.org/10.17979/ejge.2020.9.1.5231 john s. duffield voting turnout in greece: expressive or instrumental? 26-45 doi: https://doi.org/10.17979/ejge.2020.9.1.5426 irene daskalopoulou the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach 46-73 doi: https://doi.org/10.17979/ejge.2020.9.1.5714 luzius stricker and moreno baruffini do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey 74-94 doi: https://doi.org/10.17979/ejge.2020.9.1.5948 esra n. kilci a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 95-114 doi: https://doi.org/10.17979/ejge.2020.9.1.5847 barbara pawełek https://doi.org/10.17979/ejge.2020.9.1 https://doi.org/10.17979/ejge.2020.9.1.5231 https://doi.org/10.17979/ejge.2020.9.1.5426 https://doi.org/10.17979/ejge.2020.9.1.5714 https://doi.org/10.17979/ejge.2020.9.1.5948 https://doi.org/10.17979/ejge.2020.9.1.5847 european journal of government and economics 9(1), june 2020, 46-73 46 european journal of government and economics issn: 2254-7088 the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach luzius stricker a*, moreno baruffini a a università della svizzera italiana, institute for economic research, switzerland * corresponding author at: luzius.stricker@usi.ch article history. received 23 august 2019; first revision required 1 february 2020; accepted 28 february 2020. abstract. this paper examines the impact of the fourth partial revision of the law of unemployment insurance (avig) on unemployment dynamics in switzerland at a cantonal level. the authors apply the synthetic control method (scm), a matching method for comparative case studies. a counterfactual analysis of the cases studied is performed by combining a control group of several untreated units, which provides a better comparison to the treatment group than a single unit. the control unit is designed as a weighted average of the available cantons in the donor pool, taking into account the similarities between the chosen controls and the treated unit. once policy changes are controlled, the results suggest a significant effect on the unemployment rate at a cantonal level: the reform had a discernible impact on lowering the unemployment rate in the italianand french-speaking cantons in switzerland. keywords. labour market reforms; unemployment; treatment effects; synthetic control method. jel codes. c52; e24; j65 doi. https://doi.org/10.17979/ejge.2020.9.1.5714 1. introduction the present research attempts to contribute to the extensive literature on labour economics and social welfare, studying the effect of a revision of the unemployment law (avig) on the return on job of unemployed workers. the focus of the paper is on the fourth partial revision of the law of unemployment insurance in switzerland in 2011. this law eliminated a previous measure to lengthen the maximum unemployment duration in disadvantaged cantons affected by the economic downturn. what we want to analyse is whether this shorter unemployment duration had an effect on the unemployment rate in ticino, the most affected of the five cantons subject to the law’s revision. the dataset consists of observations on the 26 swiss cantons, and we analyse the unemployment rate, defined according to the standards of the swiss state secretariat for economic affairs (seco)1. this standard is valid for the definition of the number of unemployed persons and the computation of the corresponding unemployment rate. our 1 registered unemployed are defined as persons who are enrolled at the regional placement service (rav) and are immediately placeable, whether or not they benefit from unemployment compensations. the unemployment rate is calculated as the share of the unemployed divided by the total labour force, collected form the swiss federal statistical office (fso) in the annual structural survey (https://www.arbeit.swiss/secoalv/de/home/menue/institutionen--medien/statistiken/definitionen.html). mailto:luzius.stricker@usi.ch https://doi.org/10.17979/ejge.2020.9.1.5714 https://www.arbeit.swiss/secoalv/de/home/menue/institutionen---medien/statistiken/definitionen.html https://www.arbeit.swiss/secoalv/de/home/menue/institutionen---medien/statistiken/definitionen.html stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 47 treated unit is the canton of ticino, which applied this measure during the pre-treatment period. other cantons that partially or entirely applied this optional extension were geneva, neuchâtel, jura and vaud. since they experienced the same cut of the measure as the treated canton of ticino, these four cantons were excluded from the control donor pool; thus, only 21 cantons remained eligible in this pool. following the considerations of abadie, diamond and hainmueller (2015), the synthetic control method (scm) was chosen because it works best with a dataset that contains few registrations on an aggregate level that are potential controls and in cases where the impact on the variable takes place on a regional level. moreover, in a recent paper, ferman and pinto (2019) have proven that a modified synthetic control (sc) estimator is generally lower biased than the difference-in-difference (did) estimator. for the study design, we considered seven different scenarios. we included and excluded various predictors and tested the composition of sc, as well as inferences for the post-treatment effects, in a graphic way for each specification. we applied an additional placebo test, following the approach proposed by abadie, diamond and hainmueller (2015) and mcclelland and gault (2017). our results show that the impact of the law revision led to a reduction in the unemployment rate in ticino, while the unemployment rate for the sc displayed a similar trend in the years following the pre-treatment period. we can clearly see that the adoption of the law, which reduced the maximum unemployment insurance duration, had a substantial impact on reducing the unemployment rate in ticino, reaching the lowest value since 1996. the article is organised as follows. section 2 reviews the previous works in the literature. section 3 provides background information on the introduction of the fourth partial revision of the avig law and the economic conditions in switzerland. section 4 describes the scm approach in detail from a theoretical point of view. it also describes the model and dataset as well as the seven different specifications applied in this research. section 5 presents the results of the specifications and the placebo test for ticino and the remainder of the treated cantons. in section 6, robustness tests check for the results’ validity. section 7 comments on the results obtained in the canton of ticino, and finally, section 8 concludes. 2. literature review in evaluating policy interventions, scientists refer to estimations in comparative case studies with aggregate databases. this implies an examination of a control group with the group of interest. one of the most famous studies with this setup is card and krueger (1994), who analysed the impact of minimum wages on employment in the fast-food industry, comparing the treated state of new jersey to the control state of pennsylvania. for the part concerning the rule of the maximum length of the beneficiary period of the unemployment insurance on the exit rate from unemployment, cahuc, carcillo, and zylberberg, (2014, p. 296) refer to a number of previous studies that confirm the outcome of their job search model. in particular, meyer (1990) focuses on the behaviour of the unemployed in the period just before they reach the maximum duration of unemployment and notes an increase in the unemployment exit rate towards its end. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 48 dormont, fougère, and prieto (2001) distinguish by qualification of workers and point out the opportunistic behaviour of higher qualified french workers when leaving unemployment. in the austrian context, card, chetty, and weber (2007) identify a difference between the exit rate from unemployment and the entry rate of employment, which diverges when reaching the maximum unemployment duration, this is a result of unemployed transiting into other labour market states, as training programs or leaving the active labour force completely. last, pellizzari (2006) attributes the scarce impact of policies reducing the unemployment benefit programs generosity on unemployment to the coexistence of other social assistance programmes, which makes recipients react in a less sensitive way on policy changes. for the analysis of such problem sets, comparative case studies have significant potential, as many introduced policies take place on an aggregate level. nevertheless, two weaknesses can be identified. first, there is ambiguity surrounding how the comparison groups are chosen. second, in employing data on an aggregate level, estimating the effect without error is not always possible, as the counterfactual fit of the control group in reproducing the outcome variable might be biased by uncertainty. the idea behind applying a sc is that combining a control group of several untreated units provides a better comparison to the treatment group than a single unit (abadie, diamond, & hainmueller, 2010, p. 493). the clear advantage of constructing a sc to address this kind of problem is that the dataset is composed of few aggregate units, and therefore, a weighted average of all potential comparison units reproduces the characteristics of interest in the best way (abadie, diamond, & hainmueller, 2010, p. 496). the essential features of the scm are that the control unit is designed as a weighted average of the available non-treated units and that it takes into account the similarities between the chosen controls and the treated unit. the weighting technique also ensures that the control does not need to be checked for extrapolation problems. a second advantage of the method is that researchers do not have to know the impact on the variable of the treatment ex-ante, but any variable can be chosen to analyse a possible effect of an introduced policy on it (abadie, diamond, & hainmueller, 2010, pp. 493-494). comparative case studies have been broadly applied in various fields, such as political science (tarrow, 2010); labour market effects of migration, as described in card (1990); and even economic costs of terrorist conflicts in spain (abadie & gardeazabal, 2003). the potential of comparative case studies has grown with the availability of aggregated data. as described in abadie, diamond and hainmueller (2015, p. 500), the scm applies best in studies with a limited number of untreated, eligible units for the control and in studies on impacts on the aggregate level, such as regions. furthermore, a well-functioning model requires a sizable number of preintervention periods to reach a proper fitting and to ensure that the outcome of interest tracks the same trend as the scm over a more extensive period. in the introduction to their paper, kaul, klößner, pfeifer and schieler (2017) state that the “scm involves the comparison of outcome variables between a unit representing the case of interest, i.e. a unit affected by the intervention, and otherwise similar but unaffected units reproducing an accurate counterfactual version of the unit of interest in the absence of the stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 49 intervention. an algorithm-derived combination of precisely weighted comparison units is supposed to better depict the characteristics of the unit of interest than either any single comparison unit alone or an equally weighted combination of all or several available control units” (p. 2). the scm has thus been applied to diverse research topics, rapidly becoming an intuitive alternative for constructing counterfactual units. athey and imbens (2017) state that the scm “is arguably the most important innovation in the policy evaluation literature in the last 15 years” (p. 9). applications of the scm in an economic and political contexts include nannicini and billmeier (2011) as well as billmeier and nannicini (2013) (economic growth), hosny (2012) (free trade), jinjarak, noy and zheng (2013) (capital inflows), acemoglou, johnson, kermani, kwak and mitton (2016) (political connections), eren and ozbeklik (2016) (right-to-work laws). possebom (2017) as well as gobillon and magnac (2016), in particular, performed impact evaluations on enterprise zones. concerning theoretical contributions to the sc literature, gardeazabal and vega-bayo (2017) find that the scm estimator performs well in comparison to alternative panel approaches. klößner and pfeifer (2017) extend the scm to the forecasting context, and klößner, kaul, pfeifer and schieler (2017) provide a critical note on the applied cross-validation technique applied in abadie, diamond and hainmueller (2015). developments of the scm include cavallo et al. (2013) as well as dube and zipperer (2015) (pooling multiple synthetic case studies), li (2017) (modified scm to estimate ate), galiani and quistorff (2016) (a development of the synth_runner package to conduct placebo and inference) and quistorff (2016) (corresponding instruction on the possible placebo tests). finally, firpo and possebom (2018) analysed sensitivity and confidence intervals for the scm, while ferman and pinto (2019) and arkhangelsky et al. (2019) considered a demeaned scm and synthetic dids. 3. the institutional context in switzerland, starting in 2003, the debt of the unemployment insurance system reached a level that demanded control in order for the system to prosper in the future. to address the debt problem, policymakers developed a recovery plan. in 2011, the accumulated debt reached 7.8 billion swiss francs. consequently, the federal council raised insurance contributions. additionally, a solidarity percentage was introduced for medium to high earning workers. these measures were enacted starting in april 2011 and were a part of the fourth partial revision of the avig law. the primary purpose was the achievement of the balance of the revenues and expenditures of the insurance and the clearance of the structural debt and was achieved within one year. later, in the period following the application of the new policy, the accumulated debt was reduced significantly. the austerity measures of the law’s revision foresee a stronger connection of the contribution period to the possible beneficiary period. additionally, several latencies have been augmented, and numerous individual cases for free treatment of beneficiaries have been reduced. the consequence is that the new measures have had impacts stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 50 on various worker segments. specifically, the expiration of the beneficiary period has been shortened for some profiles of unemployed persons. the policy change raised the expiration rate in the year of application, while in the longer run, the rate converged to a slightly higher level compared to the period before 2011. the main issue of the revision was to design the revenues and expenditures of the unemployment insurance to achieve a balance, which should be independent of conjuncture. cost cuts are achieved mainly through the adaptation of the beneficiary system of coverage that reduces beneficiary periods as well as the long wait time for compensation. at its core, the insurance principle is reinforced, and the labour market reintegration measures are designed more efficiently (seco staatssekretariat für wirtschaft, 2013). one of the main concerns of the debate was the effects on the unemployment rate. the revision could lead to a lower unemployment rate, as a higher share of unemployed persons would leave the workforce and no longer be statistically registered. this reduction in the unemployment rate would be artificial and distorted and not due to an actual improvement in economic conditions in the labour market. however, data show that the exit rate of unemployment did not change between the periods preceding and following the introduction of the new law. the effect of a temporarily higher exit rate from unemployment was inconsistent and observed only in the first months. even the duration of unemployment did not fall below the level of the preceding years. the duration increased only temporarily, as a momentarily higher concentration of long-term unemployed persons exiting from the labour force was registered (seco staatssekretariat für wirtschaft, 2013, pp. 40-43). no significant differences among workers with different demographic profiles were identified. the only observation to mention is that younger workers were slightly more affected than elderly workers. from a geographic point of view, the italianand french-speaking parts of switzerland were more strongly affected by the law’s revision than the german-speaking part; this was due to the omission of the measure to optionally lengthen the beneficiary period in cases of locally high unemployment rates, which affected those regions in particular. the seco foresaw that the effects of the law’s revision in the short run would reduce the unemployment duration and the unemployment rate marginally. at that time, no significant, long-term impact on these two variables was expected (seco staatssekretariat für wirtschaft, 2013, pp. 5-6). 4. methods and empirical framework 4.1. data and sample table 1 describes the included variables for ticino and the cantons that, in the following computation, compose the sc for ticino. (the complete table for all cantons and variable definitions is reported in appendix 1). stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 51 table 1. descriptive statistics for ticino and cantons composing the sc for ticino. unemployment rates employment social indexes education seco ilo swiss foreign -ers annual growth rate tertiary sector share assist -ance poverty median taxable income maturit y quota unit rate rate rate rate rate quota quota quota chf quota canton baselstadt mean 3.738 4.038 2.814 5.672 0.007 0.718 6.144 11.067 52175 31.571 sd 0.904 0.585 0.639 1.811 0.013 0.019 0.445 1.316 3022 8.651 schaffhausen mean 2.978 3.444 1.993 6.172 0.009 0.680 2.367 5.144 49293 29.512 sd 0.916 0.498 0.559 2.350 0.012 0.015 0.194 0.340 2819 7.377 thur-gau mean 2.467 3.444 1.724 4.942 0.009 0.680 1.656 3.144 48495 23.141 sd 0.740 0.498 0.488 1.769 0.012 0.015 0.124 0.159 3280 7.629 ticino mean 4.283 5.869 3.325 6.517 0.014 0.752 2.078 8.833 44508 41.269 sd 1.328 1.045 0.960 2.141 0.018 0.012 0.273 0.296 4279 7.898 valais mean 3.811 6.356 2.800 7.532 0.016 0.809 1.456 3.156 41840 29.296 sd 1.595 0.979 1.308 2.946 0.014 0.012 0.159 1.063 4166 6.627 zürich mean 3.417 4.025 2.514 6.146 0.011 0.830 3.344 5.333 55540 29.130 sd 0.921 0.512 0.679 1.908 0.019 0.018 0.235 0.458 3766 6.031 to develop the model, a number of data series from different sources were taken into account. in order to control for the seasonality of the data, we chose to take the second quarter of every year as a reference and to compare these quarterly data in the constructed cantonal panel data set. the reason for this choice is that each of the swiss cantons has a different seasonality (most of them reaching the annual maximum during winter or summer months) in unemployment, and in fact, it is the second quarter, which fits best in comparing the realities for most of the cantons. the dependent variable is the unemployment rate, which is computed as an average of the three monthly datasets for the second quarter of each year (simple average of april, may and june), in order to be consistent with the remainder of the data. the unemployment rate is determined from data published monthly by the seco. furthermore, we included annual tax data from the federal tax administration office for each of the 26 swiss cantons as covariates. we also list the quarterly unemployment rate, measured according to the definition of the international labour organization (ilo) from the federal statistical office (fso). other data refer to employment levels from the swiss labour force survey (slfs), education indicators (high school degree quota), social assistance statistics (poverty index, social assistance), job statistics (labour force composition and unemployment structure) and the national account (gross domestic product per canton). all data come from various fso statistics, on either an annual or quarterly basis. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 52 4.2. computational issues: constructing a synthetic control for ticino there are several challenges in constructing a sc for this case. in observing the time trends of the unemployment rate of each of the donor cantons and the treated canton of ticino, the first issue is that the unemployment rates in ticino were higher than those in the donor pool for most of the years in the pre-treatment period. this is clearly visible in figure 1. figure 1. unemployment rates (seco) of ticino (treated) and cantons in the donor pool. figure 1 helps us understand that the synthetic version of ticino, combining different cantons from the donor pool, will not achieve the same high unemployment rate and therefore formulates a constraint to the scm. in order to achieve a parallel trend for the outcome variable in the pre-treatment period of the synthetic ticino we need to extend the standard scm procedure described by abadie, diamond and hainmueller (2015, p. 498), who state that the outcome variable should produce similar trajectories over an extended period. we therefore demeaned the data on unemployment for all units. this scm extension was proposed by ferman and pinto (2019) and helps to reduce the bias and variance of the scm. furthermore, as introduced by doudchenko and imbens (2016), through the demeaning of the unemployment data we can fulfil the no-intercept rule of scm, making the treated and control cantons producing a feasible trajectory. for the further sc computation, we excluded data from the cantons of geneva, neuchâtel, jura and vaud from the sample, since in the years preceding the avig law revision, thy applied the option of increasing the unemployment duration by 120 days in situations of high unemployment as well. this step is required for a clear distinction between treated and untreated cantons. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 53 4.3. specification of the synthetic control estimator we define seven different specifications and linear combinations of the pre-treatment outcome variables to be included in the estimation of the sc. we mainly follow the five specifications proposed by ferman, pinto and possebom (2017, p. 23), adapting them to the context of our research. we add the following variables to the predictors of the original specification: (1) original specification: no pre-treatment outcome values, only predictors (2) pre-treatment outcome mean: 𝑋𝑋𝑗𝑗 = [∑ 𝑌𝑌𝑗𝑗,𝑡𝑡/𝑇𝑇0 𝑇𝑇0 𝑡𝑡=1 ]; (3) all pre-treatment outcome values: 𝑋𝑋𝑗𝑗 =[𝑌𝑌𝑗𝑗,1 …𝑌𝑌𝑗𝑗,𝑇𝑇0]’; (4) the first half of the pre-treatment outcome values: 𝑋𝑋𝑗𝑗 =[𝑌𝑌𝑗𝑗,1 …𝑌𝑌𝑗𝑗,𝑇𝑇0 2� ]’; (5) the first three-fourths of the pre-treatment outcome values: 𝑋𝑋𝑗𝑗 =[𝑌𝑌𝑗𝑗,1 …𝑌𝑌𝑗𝑗,3𝑇𝑇0 4� ]’; (6) pre-treatment outcome values of even-numbered years: 𝑋𝑋𝑗𝑗 =[𝑌𝑌𝑗𝑗,1 𝑌𝑌𝑗𝑗,3 …𝑌𝑌𝑗𝑗,𝑇𝑇0−1]’; (7) pre-treatment outcome values of odd-numbered years: 𝑋𝑋𝑗𝑗 =[𝑌𝑌𝑗𝑗,2 𝑌𝑌𝑗𝑗,4 …𝑌𝑌𝑗𝑗,𝑇𝑇0−2]’ where 𝑻𝑻𝟎𝟎 = 2011 and 𝒀𝒀𝒋𝒋,𝟏𝟏 the outcome of the year 1994. 5. results and discussion ferman, pinto and possebom (2017) recommend testing the composition of sc by considering different scenarios of predictors. in the results summary in table 2 and the following sections, the results for each of the seven scenarios are listed and compared. depending on the chosen scenarios, the synthetic version of ticino is constructed as a mix of the cantons of valais, basel-stadt, zürich, schaffhausen and thurgau. the highest relative importance for the trend is the canton of valais, whose labour market is similar to the one of ticino, based on the chosen predictors. in section 6, placebo tests will determine the validity of this composition. for all scenarios, we obtain a rather high post/pre rmspe ratio (comparing the postand pretreatment ratios of the mean squared prediction errors (rmspe2)). this indicates that for each specification, as the treatment was not assigned randomly, at least the same value could have been achieved with a probability of 60% (for the entire post-treatment period). this is unsurprising, as, in this period, the unemployment rate of ticino continues to rise in the first two years and decreases only later. only in this second step, its values are closer to the rates of the remainder of the cantons in the donor pool. more importantly for the overall post-treatment period are the p-values3 of the probability of obtaining the unemployment rate by chance in every single year of the post-treatment period. 2 (ferman, pinto, & possebom, 2017) follow a proposal of (abadie, diamond, & hainmueller, 2015) who adopt a straightforward placebo test, as a possible inference procedure. furthermore, they compute the test statistic, computing the rmspe (ratio of the mean squared prediction errors): 𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑗𝑗: = ∑ (𝑌𝑌𝑗𝑗,𝑡𝑡−𝑌𝑌𝚥𝚥,𝑡𝑡 𝑁𝑁� )2/(𝑇𝑇−𝑇𝑇0)𝑇𝑇 𝑡𝑡=𝑇𝑇0+1 ∑ (𝑌𝑌𝑗𝑗,𝑡𝑡−𝑌𝑌𝚥𝚥,𝑡𝑡 𝑁𝑁� )2/𝑇𝑇0 𝑇𝑇0 𝑡𝑡=1 3 (ferman, pinto, & possebom, 2017, p. 10) propose to calculate a p-value to reject the null hypothesis of no effect. the test is conducted at a pre-specified significance level. in the specification, 𝟙𝟙 is the indicator function of the event: 𝑝𝑝 ≔ ∑ 𝟙𝟙[𝑅𝑅𝑀𝑀𝑆𝑆𝑆𝑆𝑆𝑆𝑗𝑗≥𝑅𝑅𝑀𝑀𝑆𝑆𝑆𝑆𝑆𝑆1]𝐽𝐽+1 𝑗𝑗=0 𝐽𝐽+1 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 54 table 2. summary of the sc and placebo effects (p-values). summary synthetic control unemployment insurance reform scenario 1 scenario 2 scenario 3 scenario 4 scenario 5 scenario 6 scenario 7 rmspe 0.4260 0.4661 0.4846 0.4494 0.5019 0.4692 0.523 w-weights controls control zürich 0.430 0.072 0.232 control valais 0.524 0.534 0.570 0.565 0.568 0.591 0.552 control thurgau 0.298 control schaffhausen 0.103 0.349 0.363 0.224 0.166 0.157 control basel-stadt 0.075 0.117 0.208 0.011 0.291 post/pre rmspe ratio ticino 1.61 1.53 1.53 1.79 1.53 1.39 1.54 p-values years after treatment 1 0.857 0.667 0.650 0.571 0.600 0.700 0.650 2 1 0.952 1 0.857 0.950 1 1 3 0.095 0.095 0.150 0.095 0.150 0.150 0.150 4 0 0 0 0 0 0 0 5 0 0 0 0 0 0 0 6 0 0 0 0 0 0 0 summarizing, we can state that the probability of a randomly obtained value is high in the first two years after the avig revision, which is the period it took to have full impact on the unemployment rate. therefore, in the first two years after the law’s revision (time required to be fully adopted in ticino) the probability of obtaining the value of the synthetic ticino by chance was almost 1. starting with the third year, when the law’s revision had a full impact, the p-values are lower and undermine that the unemployment rate was not decreasing by chance. the second important point is to consider how the cantons that comprise the sc of ticino are chosen from the cantons in the donor pool. in each of the scenarios, a different set of predictor variables defines how and how strongly the predictors influence the relative importance of a canton in composing the sc. in the composition of the sc version of ticino, the characteristics that define the quality of the matching of every single variable predicting the sc out of the cantons in the donor pool are given by the v-weights matrix. this predictor weights of the scm are calculated as a vector over the combination of the included predictor variables (weights are listed by scenarios in table 3). stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 55 table 3. predictor v-weights in defining the sc from the cantons in the donor pool. variables matrix v-weights – quality of the prediction predictors scenario 1 scenario 2 scenario 3 scenario 4 scenario 5 scenario 6 scenario 7 unempl. % (ilo) 0.0029216 0.0126819 0.0251016 0.0003355 0.0178546 0.0608614 0.1030930 unempl. % swiss 0.0079965 0.0058136 0.0305801 0.0040307 0.0000000 0.0000061 0.0000427 unempl. % foreigners 0.0017593 0.1575096 0.1602247 0.2563659 0.0522683 0.4462761 0.0174129 empl. growth 0.0524550 0.0104868 0.0050203 0.0910613 0.0101024 0.0882374 0.1116625 % of 3rd sector firms 0.5339663 0.1133374 0.0001609 0.0835999 0.0585871 0.0114713 0.0542441 % of social aid 0.2806241 0.5113206 0.0054987 0.0505141 0.1710640 0.1558218 0.2433129 poverty index 0.0000026 0.0861157 0.0012954 0.0080217 0.0758434 0.0001324 0.1494127 median taxable income 0.1195161 0.0272772 0.0063893 0.0397639 0.1014102 0.0549717 0.0596940 % of high school grad. 0.0007581 0.0201676 0.0541417 0. 0166663 0.0036179 0.0000641 0.0155404 seco unempl. % (average 1994-2011) 0.0552891 seco unempl. % (1994) 0.1174882 0.0810530 0.0820655 0.0428916 seco unempl. % (1995) 0.0000000 0.0005544 0.1738635 0.1474132 seco unempl. % (1996) 0.0000000 0.0000004 0.0071515 0.0036269 seco unempl. % (1997) 0.0806469 0.0095650 0.0006242 0.0054000 seco unempl. % (1998) 0.1625513 0.3171984 0.0928863 0.0373949 seco unempl. % (1999) 0.0811219 0.0090562 0.1093570 0.0293314 seco unempl. % (2000) 0.0002206 0.0084711 0.0241066 0.0013137 seco unempl. % (2001) 0.0035450 0.0035092 0.0048304 0.0016268 seco unempl. % (2002) 0.0663867 0.0202324 0.0012570 0.0627973 seco unempl. % (2003) 0.0363450 0.0017574 0.0108765 seco unempl. % (2004) 0.0518272 0.0071014 0.0287362 seco unempl. % (2005) 0.0000000 0.0008485 0.0497369 seco unempl. % (2006) 0.0406707 0.0034017 0.0035184 seco unempl. % (2007) 0.0000000 0.0000000 seco unempl. % (2008) 0.0059918 0.0003590 seco unempl. % (2009) 0.0430760 0.0011992 seco unempl. % (2010) 0.0217150 0.0015189 in table 4, we further compare the predicted scm predictor averages for each of the seven scenarios to the sample average of the predictors in ticino. this simple matching helps to understand the quality of the matching and prediction of the unemployment rate. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 56 table 4. predictor weights in defining the sc from the cantons in the donor pool. synthetic control predictor averages compared to ticino (predictor balance) scm scm scm scm scm scm scm ticino scenario 1 scenario 2 scenario 3 scenario 4 scenario 5 scenario 6 scenario 7 unempl. % (ilo) 5.25 4.78 4.82 5.14 4.89 4.96 5.07 4.97 unempl. % swiss 3.19 2.16 2.22 2.41 2.17 2.32 2.30 2.40 unempl. % foreigners 5.98 5.59 5.78 6.05 5.84 5.82 5.96 5.80 empl. growth 0.01 0.01 0.01 0.01 0.01 0.01 0.01 0.01 % of 3rd sector firms 0.75 0.75 0.76 0.82 0.77 0.76 0.79 0.77 % of social aid 1.92 1.92 2.28 2.25 1.88 2.60 2.06 2.94 poverty index 8.70 3.64 4.45 3.77 3.77 4.83 3.77 5.30 median taxable income 46925 47342 47624 49827 47587 47634 48463 47946 % of high school grad. 43.87 29.74 30.98 30.96 30.86 31.09 30.92 31.19 seco unempl. % (average 1994-2011) 4.01 3.59 seco unempl. % (1994) 5.97 6.12 6.03 6.33 6.16 seco unempl. % (1995) 6.07 5.88 5.76 6.04 6.09 seco unempl. % (1996) 7.17 5.55 5.54 5.65 5.61 seco unempl. % (1997) 7.27 6.25 6.18 6.14 6.09 seco unempl. % (1998) 6.03 4.54 4.46 4.39 4.50 seco unempl. % (1999) 3.83 3.17 3.08 3.02 2.99 seco unempl. % (2000) 2.60 1.82 1.75 1.85 1.79 seco unempl. % (2001) 2.10 1.67 1.64 1.76 1.81 seco unempl. % (2002) 2.93 2.41 2.19 2.29 2.29 seco unempl. % (2003) 3.87 3.69 3.31 3.43 seco unempl. % (2004) 4.07 3.89 3.67 3.71 seco unempl. % (2005) 4.47 3.79 3.60 3.67 seco unempl. % (2006) 4.50 3.23 3.20 3.16 seco unempl. % (2007) 1.00 2.76 2.90 seco unempl. % (2008) 3.67 2.57 2.52 seco unempl. % (2009) 4.47 3.51 3.51 seco unempl. % (2010) 4.33 3.56 3.47 the following section graphically shows and discusses the results for the third of the listed scenarios (the remainder are in appendix 3). the only variation between the scenarios consists of the included predictors, which define the relative similarity of the cantons in the donor pool and the treated ticino. we chose scenario 3 as a benchmark specification since the scm goal is to fit the levels of the outcome variable and this works best if the entire pre-treatment period is completely included. 5.1. specification including all pre-treatment outcome values singularly (scenario 3) the third specification includes, in addition to the basic scenario, a vector with all the pretreatment outcome values from every single year. this helps to create an additional weight that searches for similarities in the single unemployment rates of the pre-treatment period among all stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 57 the included cantons. the sc in this scenario is composed of only two cantons, valais and basel-stadt, with a stronger weight of valais (w-weights in table 2). in this specification, no great difference in the slope of the sc unit can be noticed in figure 2. with regard to the effect of the introduction of the law’s revision in the post-treatment period, we note two observations. first, the previous parallel trend of unemployment rates changes. initially, this translates into a smoother growth in the unemployment rate for ticino compared to its sc. most workers work more than two years before losing their jobs and therefore have a maximum of 400 workdays covered by insurance (five workdays per week). the main effect of the omission of the possibility of lengthening the beneficiary period by 120 days in ticino has the greatest effect with a delay of 400 workdays after the revision of the law. this means, as our reference period is the second quarter of the year, that the main effect of the omission of this additional measure would be visible within two years of the law’s revision (april 1st, 2011). thus, the 2013 data should make the effect visible for the first time. the second observation is that starting in 2013, ticino’s unemployment rate drops significantly and falls below the rate of its sc for the first time since 2005. this provides evidence that a part of the chronically higher unemployment rate in ticino is due to the potentially longer beneficiary period compared with the control cantons in the donor pool. the graphical illustrations of the results for the remainder of the scenarios contained in the results table can be found in appendix 3. in addition to the specifications, further robustness tests are described in the following section 6. first, a did model with a linear fit for the controls and the treated canton was computed. second, the scm was used to analyse the remainder of the cantons that applied for the additional measure of extending the maximum unemployment duration. the results are discussed in comparison with ticino. figure 2. demeaned sc and ticino unemployment rate, including all pre-treatment variables as predictors. -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 58 6. robustness tests to test the validity of the results, two placebo tests following the literature were performed. first, the data were analysed in a did context, allowing for a differentiated view of the problem set. second, sc were constructed for the remainder of the cantons affected by the law’s revision to identify common tendencies and differences among those treated. as shown, in appendix 3, all different scenarios introduced in section 4.3 were calculated for ticino, checking for the robustness of the results by varying the relative importance of the pre-treatment outcome variables as predictors of the sc of ticino. this different scenarios do not show any variation of the results from the previously discussed scenario 3 and therefore undermine the obtained results for ticino. 6.1. difference-in-difference computation of ticino and the donor pool average to verify the diverging trend of the unemployment rates in ticino and the weighted average of the donor pool cantons, a did analysis was conducted. the data illustrated in figure 3 compares the time trend of ticino, represented by the blue line, with a weighted average of all donor pool cantons. the weights correspond to the respective size of the labour market in the remaining cantons. the data displayed in figure 4 suggests a linear fit across the years in the pre-treatment period from 2003 to the revision of the unemployment law in 2011, identifying a common trend in the time before the revision. we defined two scenarios: the first one was a cut-off in the fitted values in 2011, when the actual law revision took place, and the second one was a time lag of two years when the last unemployed person benefitting from the additional beneficiary periods left unemployment. this is graphically illustrated in figure 4. figure 3. unemployment rates (seco) of ticino (treated) and weighted average (by employees) of the donor pool. second quarter of the year. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 59 figure 4. linear fits for ticino and donor pool with cut-offs in 2011 (left) and 2013 (right). the numerical results for the did analysis are illustrated in table 5. in the pre-treatment period, ticino systematically displays a higher unemployment rate compared to the donor pool average (1.74% for 2011 cut-off and 1.77% for 2013). the did result, assuming the continuity of the common trend of decreasing unemployment rates over time from the pre-treatment period, suggests a reduction in unemployment in ticino with the revision of the unemployment insurance law. this reduction on average was as large as -0.33% in the first scenario and as large as -0.66% in the second scenario. table 5. difference-in-difference outcomes for ticino and donor pool canton average unemployment rates. (1) (2) variables cut-off 2011 cut-off 2013 time -0.130 -0.029 [0.171] [0.148] treated 1.738*** 1.765*** [0.197] [0.164] did -0.326 -0.663*** [0.271] [0.220] constant 2.425*** 2.381*** [0.159] [0.132] observations 26 26 r-squared 0.861 0.881 robust standard errors in brackets *** p<0.01, ** p<0.05, * p<0.1 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 60 6.2. results for other cantons affected by the additional measure a second placebo test aimed to compare and test the validity of the results by researching the effect the reform had in the remainder of the cantons that applied the measure to lengthen the maximum number of days of unemployment in the period preceding the revision of the unemployment law. for each of these, a sc unit was constructed and compared graphically to their unemployment rate trends. table 6 summarizes the main results of these computations. for three of the four remaining cantons that had applied the optional lengthening of the maximum unemployment duration, it was possible to construct a complete sc analysis and to compare the trend of their unemployment rate with their respective sc. with the available data for the canton of geneva, it was not possible to reproduce a complete sc analysis that followed a similar trend in the pre-treatment period, so only partial results are available for this case. the first canton analysed is geneva. unlike ticino, geneva did not implement the optional lengthening of the maximum unemployment duration in the six months preceding the revision of the unemployment insurance law. consequently, the full effect is observed earlier in geneva than in ticino. figure 5 shows that the difference between geneva’s demeaned unemployment rate and that of its sc is larger after the law’s revision. this is visible in figure 5, where the difference between geneva and its sc is larger after the law’s revision, but different from ticino the effect of a shorter unemployment duration is noticed earlier. table 6. summary of the sc computation of other cantons affected by the new policy. summary synthetic control unemployment insurance reform canton geneva canton vaud canton jura canton neuchâtel rmspe 0.512 0.282 0.495 0.641 w-weights controls control valais 0.580 0.143 0.242 control basel-stadt 0.641 0.420 0.377 0.758 control zürich 0.359 control solothurn 0.480 p-value (post/pre rmspe ratio) n o co nv er ge nc e ac hi ev ed in sy nt h_ ru nn er (d er iv at iv es ca nn ot b e ca lc ul at ed ) 0.547 0.823 1.577 p-values years after treatment 1 0.75 0.15 0.15 2 0.70 0.55 0 3 0.85 0.90 0 4 0.75 0.45 0 5 0.30 0.10 0 6 0.50 0 0 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 61 figure 5. outcome for geneva: demeaned sc and unemployment rate, including all pre-treatment variables as predictors. the second analysed canton is vaud. the results for this canton (see figure 6) suggest that starting with 2015 the demeaned unemployment rate started to be lower than its sc. the effect is less visible, a possible explanation being that vaud already had a lower unemployment rate than geneva in the pre-treatment period, so the impact of the measure was less evident. furthermore, analogously to the case of ticino, vaud adopted the measure up to the last possible date in 2011, which delays the effect slightly compared to the geneva case. the unemployment rate in the canton of jura fluctuates more than that of the previously analysed cantons. jura is a small canton that is more strongly affected by cyclical unemployment than many other cantons. the analysis, in this case, does not indicate any clear results (figure 7), especially as the high unemployment rate in 2009-2010 could not be reproduced in its demeaned sc. nevertheless, starting in 2011, the unemployment rate of jura falls below its sc for the first time since 2001. figure 6. outcome for vaud: demeaned sc and unemployment rate, including all pre-treatment variables as predictors. -3 -2 -1 0 1 2 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit -2 0 2 4 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit -3 -2 -1 0 1 2 3 1995 2000 2005 2010 2015 year treated donors stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 62 figure 7. outcome for jura: demeaned sc and unemployment rate, including all pre-treatment variables as predictors. the last of the treated cantons is neuchâtel. it was one of the last adaptors of the measure as well. similar to jura it is has a small and volatile labour market. for this, it is hard to obtain a meaningful sc in this canton since it is impossible to reproduce the volatility of the demeaned unemployment rate of this canton with the ones of the donor pool (results in figure 8). figure 8. outcome for neuchâtel: demeaned sc and unemployment rate, including all pre-treatment variables as predictors. 7. comment on results the results for different specifications of the sc, as previously proposed by ferman, pinto and possebom (2017), help us gain insight into the sensitivity of the computation. the seven scenarios include different predictor variables of the seco unemployment rate of the pretreatment period in addition to the independent predictors composing the basic scenario. with this procedure, we want to test the sensitivity of the construction of the sc of ticino. as we noticed, the two or more (depending on the scenario) cantons from the donor pool constructing the sc for ticino remain the same across some of the specifications. even if their relative weight changes across the scenarios. -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit -3 -2 -1 0 1 2 3 1995 2000 2005 2010 2015 year treated donors -3 -2 -1 0 1 2 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit -3 -2 -1 0 1 2 3 1995 2000 2005 2010 2015 year treated donors stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 63 for the fact that the control cantons are the five individuated in all scenarios, we can exclude the possibility that one canton is part of the sc only because of a defined predictor or a unique event. the specifications confirm that the weighted composition of the synthetic ticino, composed up of valais, and depending on the scenario on one or more additional cantons, is the most similar to the real ticino. the results show that in the years preceding the revision of the unemployment law, a parallel trend, with the demeaned unemployment rate for ticino, is registered. with the revision of the avig law, the previous parallel trend approaches its sc. the greatest effect, when the unemployment rate in ticino falls below that of its sc, is registered two years after the revision of the law. this can be explained by the effect of the annulment of the additional measure to lengthen the beneficiary period for unemployment in ticino beyond 400 days. in summary, the reform of the unemployment law and the standardisation of the unemployment duration caused the chronically higher unemployment rate of ticino to fall to a lower level than it would otherwise have been. in the short period, as described in the (seco staatssekretariat für wirtschaft, 2013) report, the movement from unemployment to social assistance increased, but the effect did not persist over time. the lower unemployment rate in the longer term, therefore, is also the result of the changing duration of the maximum beneficiary period and the faster reintegration of the unemployed workforce into the labour market. moreover, the achieved results are clearly supported in two of the four analysed cantons in addition to ticino. geneva and vaud, which propose similar to but less accentuated results than ticino. nevertheless, both show a reduction in their demeaned unemployment rate in comparison with their sc. 8. conclusions this research attempts to identify the effect of the introduction of the partial revision of the unemployment insurance law (avig) on the actual unemployment rates in the ticino region in southern switzerland. this area was particularly affected by the revision of this law, which eliminated the possibility of increasing the maximum unemployment duration in cases of high regional cyclic unemployment. the data indicate that for the entire period observed, the unemployment rate of the canton of ticino lies above the average of the cantons included in the donor pool. in the years preceding the reform, the unemployment rate of ticino was, on average, approximately 1.8% higher than the weighted average of the cantons in the donor pool. descriptive statistics show that this difference was reduced starting with the introduction of the revised law in 2011, reaching a difference of 0.74% in 2016. nevertheless, the unemployment rate in ticino remained higher than the average rate in the donor pool cantons. the main goal is to compare the trend of the unemployment rate in ticino with the best-fitting control cantons in order to identify the effect of the measure, which eliminated the possibility for ticino to lengthen the unemployment duration. the sc identifies a weighted average of the unemployment rate of the most similar cantons regarding a chosen number of predictors in stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 64 seven different scenarios. each of the seven scenarios includes two or more cantons that construct their respective sc. in the years preceding the revision of the unemployment insurance law, ticino and its sc followed a similar trend, and ticino’s unemployment rate was approximately 1% higher than that of its control. this difference decreased with the introduction of the law’s revision, and starting in 2014, the common trend ended. in 2015, for the first time, the unemployment rate in ticino lay below that of its sc and remained so until the end of the observed period. we can conclude that the application of the revised unemployment insurance law had a strong effect on reducing the unemployment rate in ticino, while in the sc, constructed of similar cantons, no such similarity can be identified. most workers entering unemployment in ticino contributed two or more years to the insurance and therefore can benefit from insurance coverage for a maximum of 1.5 years. once this period expires, an additional 120-day benefit prolongation could have been applied in ticino before the law’s application. with the new legislation in ticino, the unemployment rate decreased. as the seco staatssekretariat für wirtschaft (2013) report shows, in terms of unemployed persons, the french and italian areas of switzerland (in large part applying the additional measure) were disproportionally affected by the law’s revision, experiencing a greater effect than the german-speaking part of the country. this, in combination with a reduction in the general duration of unemployment, led to a more than proportional and persistent reduction in ticino’s unemployment rate compared with its sc, constructed with cantons from the remainder of switzerland. these results are in contrast to the forecast of the seco staatssekretariat für wirtschaft (2013) in a report studying all of switzerland, where, in the long run, no persistent reduction in unemployment rates resulted from the partial revision of the unemployment law. in conclusion, this means that in the period before the revision of the unemployment law, the return on job of unemployed workers lasted longer than afterwards. a shorter insurance period makes pressure on the job seekers and accelerates their search for a new position. the previous policy adapted especially in the italian and french-speaking part of switzerland reduced the intensity of the job search of unemployed as their insurance covered a longer period. in this sense, the law’s revision helped especially in those regions to fight the high unemployment rate, by limiting the maximum length of the insured beneficiary period. acknowledgements. we thank our three anonymous reviewers and the editor for constructive feedback and suggestions, which led to a much-improved paper. we would also like to thank prof. rico maggi who reviewed an earlier version of this research. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 65 references abadie, a. and gardeazabal, j. (2003). ‘the economic costs for conflict: a case study of the basque country’, american economic review, 93(1), 112-132. https://doi.org/10.1257/000282803321455188 abadie, a., diamond, a. and hainmueller, j. (2010). ‘synthetic control methods for comparative case studies: estimating the effect of california's tobacco control program’, journal of the american statistical association, 105(490), 493-505. https://doi.org/10.1198/jasa.2009.ap08746 abadie, a., diamond, a. and hainmueller, j. (2015). ‘comparative politics and the synthetic control methods in comparative case studies’, american journal of political science, 59(2), 495-510.https://doi.org/10.1111/ajps.12116 acemoglou, d., johnson, s., kermani, a., kwak, j. and mitton, t. (2016). ‘the value of connections in turbulent times: evidence from the united states’, journal of financial economics, 121, 368-391. https://doi.org/10.1016/j.jfineco.2015.10.001 arkhangelsky, d., s. athey, d. a. hirshberg, g. w. imbens, and s. wager (2019, february). ‘synthetic difference in differences’. nber working paper n. 25532. https://doi.org/10.3386/w25532 athey, s. and imbens, g. w. (2017). ‘the state of applied econometrics: causality and policy evaluation’, journal of economic perspectives, 31(2), 3-32. https://doi.org/10.1257/jep.31.2.3 billmeier, a. and nannicini, t. (2013). ‘assessing economic liberalization episodes: a synthetic control approach’, the review of economics and statistics, 95(3), 983-1001. https://doi.org/10.1162/rest_a_00324 cahuc, p., carcillo, s., and zylberberg, a. (2014). labor economics (second edition). the mit press, cambridge. card, d. (1990). ‘the impact of the mariel boatlift on the miami labour market’, industrial and labor relations review, 44, 245-257. https://doi.org/10.1177/001979399004300205 card, d., chetty, r., and weber, a. (2007). ‘the spike at benefit exhaustion: leaving the unemployment system or starting a new job?’. american economic review, 97, 113-118. https://doi.org/10.1257/aer.97.2.113 card, d. and krueger, a. (1994). ‘minimum wages and employment: a case study of the fastfood industry in new jersey and pennsylvania’, american economic review, 84, 772-793. https://doi.org/10.3386/w4509 cavallo, e., galiani, s., noy, i. and pantano, j. (2013). ‘catastrophic natural disasters and economic growth’, the review of economics and statistics, 95(5), 1549-1561. https://doi.org/10.1162/rest_a_00413 dormont, b., fougère, d., and prieto, a. (2001). ‘the effect of the time profile of unemployment insurance benefits on exit from unemplyoment’ (crest working paper). paris, www.crest.fr. doudchenko, n. and imbens, g. (2016). ‘balancing, regression, difference-in-difference and synthetic control methods: a synthesis’, nber working paper series (22791). https://doi.org/10.3386/w22791 dube, a. and zipperer, b. (2015). ‘pooling multiple case studies using synthetic controls: an application to minimum wage policies’, iza discussion paper no. 8944. eren, o. and ozbeklik, s. (2016). ’what do right-to-work laws do? evidence from a synthetic control method analysis’, journal of policy analysis and management, 35(1), 173-194. https://doi.org/10.1002/pam.21861 ferman, b., pinto, c. and possebom, v. (2017). ‘cherry picking with synthetic controls’, mpra paper (78213). ferman, b.and pinto, c. (2019, january). ‘synthetic controls with imperfect pre-treatment fit’. working paper. firpo, s. and v. possebom (2018). ‘synthetic control method: inference, sensitivity analysis and confidence sets. journal of causal inference 6 (2), pp. 126. https://doi.org/10.1515/jci-2016-0026 https://doi.org/10.1257/000282803321455188 https://doi.org/10.1198/jasa.2009.ap08746 https://doi.org/10.1111/ajps.12116 https://doi.org/10.1016/j.jfineco.2015.10.001 https://doi.org/10.3386/w25532 https://doi.org/10.1257/jep.31.2.3 https://doi.org/10.1162/rest_a_00324 https://doi.org/10.1177/001979399004300205 https://doi.org/10.1257/aer.97.2.113 https://doi.org/10.3386/w4509 https://doi.org/10.1162/rest_a_00413 http://www.crest.fr/ https://doi.org/10.3386/w22791 https://doi.org/10.1002/pam.21861 https://doi.org/10.1515/jci-2016-0026 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 66 galiani, s. and quistorff, b. (2016). the synth_runner package: utilities to automate synthetic control estimation using synth. college park: university of maryland. https://doi.org/10.1177/1536867x1801700404 gardeazabal, j. and vega-bayo, a. (2017). ‘an empirical comparison between the synthetic control method and hsiao et al.'s panel data approach to program evaluation’, journal of applied econometrics, 32(5), 983-1002. https://doi.org/10.1002/jae.2557 gobillon, l. and magnac, t. (2016). ‘regional policy evaluation: interactive fixed effects and synthetic controls’, review of economics and statistics, 98(3), 535-551. https://doi.org/10.1162/rest_a_00537 hosny, a. s. (2012). ‘algeria's trade with gafta countries: a synthetic control approach’, transition studies review, 19, 35-42. https://doi.org/10.1007/s11300-012-0228-4 jinjarak, y., noy, i. and zheng, h. (2013). ‘capital controls in brazil stemming a tide with a signal?’, journal of banking & finance, 37, 2938-2952. https://doi.org/10.1016/j.jbankfin.2013.04.007 kaul, a., klößner, s., pfeifer, g. and schieler, m. (2017, july 29). ‘synthetic control methods: never use all pre-intervention outcomes together with covariates’ available at https://mpra.ub.uni-muenchen.de/83790/1/mpra_paper_83790.pdf (accessed march 19, 2018). klößner, s. and pfeifer, g. (2017). ‘outside the box: using synthetic control methods as a forecasting technique’, applied economics letters, 25(9), 615-618. https://doi.org/10.1080/13504851.2017.1352071 klößner, s., kaul, a., pfeifer, g. and schieler, m. (2017). ‘comparative politics and the synthetic control method revisited: a note on abadie et al. (2015)’, swiss journal of economics and statistics. https://doi.org/10.1186/s41937-017-0004-9 kreif, n., grieve, r., hangartner, d., turner, a., nikolova, s. and sutton, m. (2016). ‘examination of the synthetic control method for evaluating health policies with multiple treated units’, health economics, 25(12), 1514-1528. https://doi.org/10.1002/hec.3258 li, k. t. (2017, june). ‘the wharton school, the university of pennsylvania’, available at https://faculty.wharton.upenn.edu/wp-content/uploads/2017/06/6_kathleen-t-li-jobmarket-paper-modified-synthetic-control-method-1.pdf (accessed february 05, 2018). meyer, b. (1990). ‘unemployment insurance and unemployment spells’. econometrica, 58(4), 757-782. https://doi.org/10.2307/2938349 mcclelland, r. and gault, s. (2017). ‘the synthetic control method as a tool to understand state policy’, research report, urban insitute, washington, available at https://www.urban.org/sites/default/files/publication/89246/the_synthetic_control_method_a s_a_tool_0.pdf (accessed march 15, 2018). nannicini, t. and billmeier, a. (2011). ‘economies in transition: how important is trade openness for growth?’, oxford bulletin of economics and statistics, 73(3), 287-314. https://doi.org/10.1111/j.1468-0084.2010.00626.x possebom, v. (2017). ‘free trade zone of manaus: an impact evaluation using the synthetic control method’. revista brasileira de economia 71 (2), pp. 217-231. https://doi.org/10.5935/0034-7140.20170011 pellizzari, m. (2006). ‘unemployment duration and the interaction between unemployment insurance and social assistance’. labour economics, 13(6), 773-798. https://doi.org/10.1016/j.labeco.2005.10.003 quistorff, b. (2016, march 22). ‘inter-american development bank iadb – blog’, the synth_runner package: utilities to automate synthetic control estimation using synth:, available at https://blogs.iadb.org/desarrolloefectivo_en/2016/03/22/synth_runner/ (accessed february 06, 2018). seco staatssekretariat für wirtschaft. (2013). ‘bericht zur 4. teilrevision des arbeitslosenversicherungsgesetzes (avig)‘, available at https://www.seco.admin.ch/dam/seco/de/dokumente/arbeit/alv/grundlagen/bericht_zur_4 .teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf.download.pdf/d_bericht _zur_teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf (october 17, 2017). tarrow, s. (2010). ‘the strategy of paired comparison: toward a theory of practice’, comparative political studies, 43(2), 230-259. https://doi.org/10.1177/0010414009350044 https://doi.org/10.1177/1536867x1801700404 https://doi.org/10.1002/jae.2557 https://doi.org/10.1162/rest_a_00537 https://doi.org/10.1007/s11300-012-0228-4 https://doi.org/10.1016/j.jbankfin.2013.04.007 https://mpra.ub.uni-muenchen.de/83790/1/mpra_paper_83790.pdf https://doi.org/10.1080/13504851.2017.1352071 https://doi.org/10.1186/s41937-017-0004-9 https://doi.org/10.1002/hec.3258 https://faculty.wharton.upenn.edu/wp-content/uploads/2017/06/6_kathleen-t-li-job-market-paper-modified-synthetic-control-method-1.pdf https://faculty.wharton.upenn.edu/wp-content/uploads/2017/06/6_kathleen-t-li-job-market-paper-modified-synthetic-control-method-1.pdf https://doi.org/10.2307/2938349 https://www.urban.org/sites/default/files/publication/89246/the_synthetic_control_method_as_a_tool_0.pdf https://www.urban.org/sites/default/files/publication/89246/the_synthetic_control_method_as_a_tool_0.pdf https://doi.org/10.1111/j.1468-0084.2010.00626.x https://doi.org/10.5935/0034-7140.20170011 https://doi.org/10.1016/j.labeco.2005.10.003 https://blogs.iadb.org/desarrolloefectivo_en/2016/03/22/synth_runner/ https://www.seco.admin.ch/dam/seco/de/dokumente/arbeit/alv/grundlagen/bericht_zur_4.teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf.download.pdf/d_bericht_zur_teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf https://www.seco.admin.ch/dam/seco/de/dokumente/arbeit/alv/grundlagen/bericht_zur_4.teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf.download.pdf/d_bericht_zur_teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf https://www.seco.admin.ch/dam/seco/de/dokumente/arbeit/alv/grundlagen/bericht_zur_4.teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf.download.pdf/d_bericht_zur_teilrevision_des_arbeitslosenversicherungsgesetzes_(avig).pdf https://doi.org/10.1177/0010414009350044 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 67 appendix 1 – descriptive statistics table 7. summary of descriptive statistics for all cantons. unemployment rates employment social indexes education seco ilo swiss foreigners annual growth rate tertiary sector share assistence poverty median taxable income maturity quota unit rate rate rate rate rate quota quota quota chf quota canton appenzell mean 1.681 3.444 1.297 3.394 0.009 0.680 1.778 3.456 48005.000 28.870 outerrhodes sd 0.508 0.498 0.389 1.096 0.012 0.015 0.199 0.313 3051.570 7.727 appenzell mean 1.015 3.444 0.749 2.702 0.009 0.680 1.111 2.078 46375.000 23.000 innerrhodes sd 0.427 0.498 0.332 1.159 0.012 0.015 0.169 0.120 3660.152 7.511 aargau mean 2.807 4.038 1.944 5.960 0.007 0.718 1.967 3.422 52830.000 25.071 sd 0.782 0.585 0.542 1.816 0.013 0.019 0.071 0.192 3702.076 6.218 basel district mean 2.722 4.038 2.058 5.406 0.007 0.718 2.467 4.100 56765.000 31.944 sd 0.657 0.585 0.497 1.662 0.013 0.019 0.206 0.235 4134.231 7.592 basel stadt mean 3.738 4.038 2.814 5.672 0.007 0.718 6.144 11.067 52175.000 31.571 sd 0.904 0.585 0.639 1.811 0.013 0.019 0.445 1.316 3021.654 8.651 bern mean 2.479 3.825 1.858 6.229 0.010 0.722 4.144 6.478 47480.000 27.494 sd 0.850 0.478 0.655 2.256 0.012 0.014 0.133 0.130 4362.713 8.179 fribourg mean 2.881 3.825 2.100 6.615 0.010 0.722 2.422 4.556 48180.000 33.678 sd 1.193 0.478 1.012 2.523 0.012 0.014 0.130 0.274 4917.809 8.415 geneva mean 5.989 6.356 5.404 6.800 0.016 0.809 4.389 11.167 56275.000 36.940 sd 1.078 0.979 1.034 1.199 0.014 0.012 0.807 1.112 4962.902 4.777 jura mean 3.974 3.825 3.307 7.607 0.010 0.722 2.111 5.800 44140.000 33.743 sd 1.275 0.478 1.198 2.159 0.012 0.014 0.333 0.361 3726.350 7.640 glarus mean 1.964 3.444 1.357 3.863 0.009 0.680 2.022 3.922 46070.000 22.785 sd 0.592 0.498 0.381 1.381 0.012 0.015 0.120 0.222 2800.724 5.206 grisons mean 1.860 3.444 1.288 4.281 0.009 0.680 1.233 2.800 44980.000 28.756 sd 0.485 0.498 0.465 1.057 0.012 0.015 0.132 0.100 3253.274 8.818 lucerne mean 2.378 2.875 1.631 6.031 0.016 0.729 2.189 4.322 47935.000 24.582 sd 0.818 0.617 0.546 2.476 0.020 0.020 0.162 0.179 4038.013 7.879 neuchâtel mean 4.575 3.825 3.565 7.236 0.010 0.722 6.544 8.911 48035.000 37.332 sd 1.230 0.478 1.004 1.893 0.012 0.014 0.550 0.662 4370.508 7.733 nidwalden mean 1.367 2.875 1.103 3.493 0.016 0.729 0.900 1.967 51860.000 26.096 sd 0.669 0.617 0.582 1.581 0.020 0.020 0.050 0.141 4272.082 8.330 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 68 unemployment rates employment social indexes education seco ilo swiss foreigners annual growth rate tertiary sector share assistence poverty median taxable income maturity quota unit rate rate rate rate rate quota quota quota chf quota canton obwalden mean 1.196 2.875 0.822 3.738 0.016 0.729 1.122 2.578 45795.000 22.498 sd 0.549 0.617 0.405 1.816 0.020 0.020 0.097 0.156 3994.269 8.038 st gallen mean 2.500 3.444 1.660 5.406 0.009 0.680 2.133 4.200 47570.000 23.927 sd 0.696 0.498 0.466 1.656 0.012 0.015 0.100 0.240 2959.392 6.335 schaffhausen mean 2.978 3.444 1.993 6.172 0.009 0.680 2.367 5.144 49292.500 29.512 sd 0.916 0.498 0.559 2.350 0.012 0.015 0.194 0.340 2819.027 7.377 solothurn mean 2.992 3.825 2.083 6.942 0.010 0.722 3.111 5.289 49755.000 23.410 sd 1.015 0.478 0.715 2.630 0.012 0.014 0.355 0.948 3423.214 5.666 schwyz mean 1.682 2.875 1.160 4.174 0.016 0.729 1.500 2.733 51385.000 22.689 sd 0.588 0.617 0.375 1.880 0.020 0.020 0.100 0.141 5023.343 6.931 ticino mean 4.283 5.869 3.325 6.517 0.014 0.752 2.078 8.833 44507.500 41.269 sd 1.328 1.045 0.960 2.141 0.018 0.012 0.273 0.296 4278.597 7.898 thurgau mean 2.467 3.444 1.724 4.942 0.009 0.680 1.656 3.144 48495.000 23.141 sd 0.740 0.498 0.488 1.769 0.012 0.015 0.124 0.159 3279.999 7.629 uri mean 1.111 2.875 0.740 4.030 0.016 0.729 1.144 2.467 46370.000 23.696 sd 0.442 0.617 0.334 1.693 0.020 0.020 0.053 0.112 3229.975 7.354 valais mean 3.811 6.356 2.800 7.532 0.016 0.809 1.456 3.156 41840.000 29.296 sd 1.595 0.979 1.308 2.946 0.014 0.012 0.159 1.063 4165.700 6.627 vaud mean 4.844 6.356 3.764 7.174 0.016 0.809 4.878 7.922 50515.000 30.335 sd 1.334 0.979 1.114 1.909 0.014 0.012 0.130 0.672 6148.451 5.281 zug mean 2.471 2.875 1.765 4.957 0.016 0.729 1.722 3.689 60025.000 30.932 sd 0.755 0.617 0.531 1.841 0.020 0.020 0.083 0.298 5591.241 9.337 zurich mean 3.417 4.025 2.514 6.146 0.011 0.830 3.344 5.333 55540.000 29.130 sd 0.921 0.512 0.679 1.908 0.019 0.018 0.235 0.458 3766.109 6.031 total mean 2.815 3.906 2.109 5.528 0.012 0.727 2.536 4.944 49315.190 28.527 sd 1.515 1.236 1.282 2.343 0.015 0.046 1.511 2.632 5802.191 8.811 stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 69 appendix 2 – variables list table 8. variables included in the sc computation. variable definition seco unemployment rate number of registered unemployed workers (last day of the month enrolled at the regional labour agency) divided by the number of active labour force members. labour force is revealed by the structural census and kept fixed over the years. seco_dm seco unemployment rate demeaned by the canton pre-intervention average seco unemployment rate. ilo unemployment rate unemployed workers in this sense are all persons form 15-74 years, who were unemployed in the reference week and were actively searching for a job in the previous four weeks and who are available to start a new job. unemployment rate of swiss calculated as the seco unemployment rate, but considering swiss workforce only. unemployment rate of foreigners calculated as the seco unemployment rate, but considering foreign workforce only. annual growth rate of employed persons this includes all persons aged 15 and over, who in the reference week worked at least one hour for payment and although temporarily absent from the workplace had a job as an employee or were self-employed or who worked in a family business without payment. share of tertiary sector employees number or employees defined according to the above specification who worked in a firm of the tertiary sector, divided by the number of employees in the secondary sector in the same period. social assistance quota all persons enrolled in social assistance (as part of the permanent residential population) divided by the permanent residential population in december of the previous year. poverty quota share of persons with an equivalent income below the relative poverty threshold median of taxable income median of the taxable income in the canton of natural persons that are contributing to the direct federal tax. taxable income is defined as the perceived income minus the admitted deductions (on average 30% reduction). maturity quota percentage of youth with a highschool equivalency compare to the total of youth that have finished any degree of mandatory schooling. stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 70 appendix 3 – scenarios 1, 2, 4-7 for ticino 1. basic specification with no pre-treatment outcomes (scenario 1) this first computation is the basic model that does not include any value of the dependent variable (seco unemployment rate) from the pre-treatment period as an additional predictor. as previously explained, ticino by construction already has a higher unemployment rate than the remainder of the cantons in the donor pool. a similar trend at the respective level of unemployment is clearly achieved applying the demeaned variable for the scm through which we can obtain a parallel trend for this case and draw valid results. in this specification, the sc of ticino is composed of the cantons of valais, thurgau, schaffhausen and basel-stadt. the effect is graphically illustrated in figure 9. figure 9. outcome specification 1. 2. specification with pre-treatment outcome mean (scenario 2) the second specification, in addition to the predictors of the basic scenario, includes a predictor variable that is made up of the average unemployment rate of all the pre-treatment years for the specific canton. the advantage of this additional variable is that it provides a more stable predictor, which is not strongly affected by extraordinary shocks in an examined canton. only three of the cantons composing the sc of ticino are the same as in the basic scenario, the relative importance being slightly changed to more weight for basel-stadt, schaffhausen and valais and compared to scenario 1 it excludes thurgau from the donor pool. graphical results are illustrated in figure 10. -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 71 figure 10. outcome specification 2. 3. specification with the first half of the pre-treatment outcome values singularly (scenario 4) the fourth specification includes a vector of single pre-treatment unemployment rates for the first half of the observed years starting with 1994. this scenario adds additional weight to the early years of the observed period, which are further from the current treatment. compared to the results of the previous scenario, including the values from all pre-treatment years, the weights do not change, and the sc is again made up of valais and schaffhausen with similar weights, but substitutes basel-stadt with zürich in the composition of the sc. zürich even contributes to the sc in our benchmark scenario 3. figure 11 shows the comparison of the sc and ticino demeaned unemployment rate. figure 11. outcome specification 4. -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 72 4. specification with the first three quarters of the pre-treatment outcomes (scenario 5) the fifth specification (figure 12) follows the same idea as the previous specification. it includes the first three quarters of the unemployment rate in the pre-treatment period as predictor variables. the weights for the cantons composing the sc shift back and again exclude zürich from the controls, adding basel-stadt in comparison with the previous scenario 4. the weights for the cantons are similar to scenario 2, with slightly more weight to basel-stadt. figure 12. outcome specification 5. 5. specification with even-numbered years of the singular pre-treatment outcome values (scenario 6) scenario 6 includes even-numbered years as predictors, starting in 1994; seefigure 13. this helps to remove the weight of extraordinary shocks happening in a single year and strong cyclic downand upturns. for example, the importance of fitting situations with very low unemployment as in 2001 or very high unemployment in 1997. the combination of the sc is made of the cantons of valais, zürich, schaffhausen and basel-stadt. figure 13. outcome specification 6. -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit stricker and baruffini / european journal of government and economics 9(1), june 2020, 46-73 73 6. specification with odd-numbered years of the singular pre-treatment outcome values (scenario 7) the last specification is based on the definition of scenario 6, but shifts to include the values of the unemployment rate of the odd-numbered years in the pre-treatment period as predictor variables in addition to the basic setup. the cantons composing the sc are valais, basel-stadt and schaffhausen. results are illustrated in figure 14. figure 14. outcome specification 7. list of abbrevations avig law of unemployment insurance (arbeitslosenversicherungsgesetz) did difference-in-difference fso federal statistical office ilo international labour organization rmspe ratio of the mean squared prediction errors sc synthetic control scm synthetic control method seco swiss state secretariat for economic affairs slfs swiss labour force survey -2 -1 0 1 2 3 se c o _d m 1995 2000 2005 2010 2015 year treated unit synthetic control unit number 9, issue 1, june 2020 the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach 1. introduction 2. literature review 3. the institutional context 4. methods and empirical framework 5. results and discussion 6. robustness tests 7. comment on results 8. conclusions acknowledgements references european journal of government and economics 11(1), june 2022, 51-72 european journal of government and economics issn: 2254-7088 can a corporate well-being programme maintain the strengths of the healthy employee in times of covid-19 and extensive remote working? an empirical case study. josé manuel núñez-sáncheza, ramón gómez-chacónb, carmen jambrino-maldonadoa, jerónimo garcíafernández* a universidad de málaga, spain b ceu cardenal spínola, spain c universidad de sevilla, spain * corresponding author at: jeronimo@us.es abstract. the covid-19 pandemic and the increase of working-from-home have drastically changed many aspects of work life, causing very negative effects on employees' physical and psycho-social well-being. healthy organisations have healthy employees, who have at least five psycho-social strengths of engagement, self-efficacy, resilience, optimism and hope, which are reinforced by physical activity, relating to each other in a positive way and leading to numerous benefits for the company. these strengths are being weakened by the pandemic, and the aim of this empirical study is to analyse through a case study the effects of an updated corporate wellness programme in times of pandemic on these strengths of the healthy employee. the sample was of 251 employees, 91 women and 160 men. the instruments used were the international physical activity questionnaire and the adaptation of the healthy and resilient organization questionnaire. the results indicated that workers with high physical activity, higher seniority, well guided by supervisors, as well as a comprehensive (multi-component) well-being programme, not only physical but also psycho-social, and with the use of different digital tools (an app is not enough), can mitigate these negative effects. whereas companies are grappling with reduced employee engagement among other harmful psychosocial and physical effects, this case study suggests that a good corporate well-being programme could help mitigate these detrimental consequences for their workforce and be helpful for the company to adapt to this rapidly changing workplace. keywords. covid-19, corporate well-being, engagement, resilience, healthy employee. doi. https://doi.org/10.17979/ejge.2022.11.1.8978 1. introduction the covid-19 pandemic has put huge stress on companies and, therefore, their employees, bringing about changes to working conditions. the impacts of covid-19 in workplaces and workers worldwide have been dramatic (kniffin et al., 2021). the lockdown, social distancing and safety measures have an impact on overall well-being (eurofound, 2020). as a result of covid19, individuals are facing continuous changes in various aspects of their lives, such as health, employment, and family life (xiao et al., 2020). the need for working from home for millions of workers has accelerated recent remote work trends facilitated by communication technologies and the rise of connectivity (kniffin et al., 2021). with the intrusion of the pandemic, workers experienced a complete change in their workplace © 2022. this work is licensed under a cc by-nc 4.0 license. mailto:jeronimo@us.es https://doi.org/10.17979/ejge.2022.11.1.8978 https://www.frontiersin.org/articles/10.3389/fpsyg.2020.580702/full#b128 https://creativecommons.org/licenses/by-nc/4.0/ josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 52 that disrupted their work experience (kumar, 2021), with many people reporting greater fear, worry, and psychological stress (li et al., 2020). vander elst et al. (2017) suggested that the disadvantages inherent in extensive telework may exceed the associated advantages. therefore, wang et al. (2020) underlined the importance of improving mental health and psychological resilience during the covid-19. in fact, depending on the country, lockdown included the closure of many facilities and restrictions on leaving home for non-essential work, so organisations had to find ways to keep their businesses afloat and employees productive, resulting in a marked shift to work-from-home, not only in 2020, but in 2021 when employees in most countries were still required or encouraged to work from home (de klerk et al., 2021). the challenges of extensive remote working include increased stress and decreased life satisfaction (kazekami, 2020), and a rise in isolation (vander elst et al., 2017), health and social-psychological risks (kniffin et al., 2020). covid-19 has also led to physical inactivity, as the pandemic is making the world move even less than before (hall et al., 2021) while the immune system is very responsive to exercise, this reducing the risk of disease (nieman & wentz, 2019), stress, and improving sleep quality (altena et al., 2020). the pandemic is proving to be very detrimental to the health and well-being of employees. the many harmful effects of the pandemic on the well-being of the workforce, both physical and psycho-social, can be observed. managers should understand the importance of mitigating psycho-social risks just as seriously as other safety risks, especially in times of change and uncertainty (mathisen et al., 2017). companies that want to take care of their employees should adapt their workplace wellness programmes to the new situation, otherwise the health of their employees could be affected and, consequently, that of the company (núñez-sánchez et al., 2021). healthy organisations are supposed to introduce measures and processes in the workplace to promote and maintain well-being among workers (wilson et al., 2004). as a consequence of the growing concern observed in companies in recent years for the wellbeing of employees, the concept of healthy and resilient organization (hero) arises, defined as an organisation, integrated by healthy employees with psycho-social strengths, that makes systematic, planned, and proactive efforts to have a positive impact on the health of the organisation, fostering healthy organisational resources and practices to improve the work environment, especially during times of change and crisis (llorens et al., 2013), such as covid19 times. despite the importance of the issue for companies and their workforce, it has not been possible to find case study research on the consequences that the implementation of a corporate wellness programme in a company would have on the strengths of healthy employees. due to this absence of a scientific literature, there is a clear need to determine the effect of enforced and exclusive remote work on employee engagement and employee experience, to plan for future work modes (de klerk et al., 2021) and working conditions. furthermore, research in times of covid-19 may be very important to generate resilience-building resources that can maintain and even increase engagement (salanova, 2020). therefore, this research aims to show, through a real case, the effects that a corporate welljosé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 53 being programme, applied in times of a pandemic, has on each of the five strengths of the healthy employee: engagement, resilience, self-efficacy, positive emotions and competence. in addition, the authors will try to find out if these five strengths of the healthy employee are influenced by six different factors: the physical activity levels (high, moderate or low) of the workers of this company, their use of a fitness app, the age of the employees, their seniority, the supervisor's guidance and, finally, the type of work they have done during this period: in-office or at-a-distance. based on the results, the company will be able to know the factors that most influence these five strengths of the healthy employee, and, in this way, will be able to design action plans accordingly. the extreme covid-19 context provides a rare opportunity to investigate the benefits and obstacles of working from home in a real-life situation (de klerk et al., 2021), therefore this opportunistic investigation, based on a real case study, may shed light on how to maintain or even improve employee´s engagement, resilience, self-efficacy, positive emotions, and competence. 2. literature review healthy organisations and employees healthy organisations, defined as those that make systematic, planned, and proactive efforts to improve employee health through good practices related to task improvement (e.g., job design and redesign), social environment (e.g., better open communication channels), and organisation (e.g., work-life balance strategies), are those that attract and retain the most talented and productive workers, while their management is successful and results in healthy outcomes for employees and the organisation (salanova & schaufeli, 2009). healthy organisations are integrated by healthy employees who have psycho-social strengths such as engagement, resilience, self-efficacy, optimism and hope (salanova et al., 2012). these strengths are associated with employees who are more satisfied and engaged (bakker & demerouti, 2017), and more productive (gómez-chacón et al., 2021). according to schaufeli et al. (2002), work engagement is defined as a positive, fulfilling, workrelated state of mind that is characterised by vigour, dedication, and absorption, and it refers to a more persistent (rather than momentary) and pervasive affective-cognitive state. resilience is a personal resource that supports adaptation to adverse conditions (caniëls et al., 2018), reflects one’s capacity to recover after undergoing negative emotions, and flexibly adapt to changing demands of stressful experiences (niitsu et al., 2017). self-efficacy is the belief in one's own ability to organise and execute the courses of action necessary to obtain certain achievements (bandura, 1997), helping people to cope in complicated contexts and protecting health (parker et al., 2013). optimism reflects the extent to which people have generalised favourable expectancies for their future, being related to better subjective well-being in times of adversity (carver et al. 2010) and a personal ability as the state of positive psychological development towards obtaining an ascription of being successful (luthans et al., 2007). finally, hope can be described as a josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 54 positive motivational state that contributes to leaders and followers expending the requisite energy necessary to pursue and attain organisational goals, existing along with motivation, leadership and goal pursuit (helland & winston, 2005) there is sufficient evidence in the scientific literature showing that each of the five strengths of the healthy employee is positively related to numerous factors that benefits the organisation. furthermore, each of these strengths is positively interrelated with one or more of the other strengths, which are also positively related to physical activity (see table 1). relationship between physical activity and the strengths of the healthy employee physical activity has a positive impact on different physical and psycho-social variables in employees (bezner et al., 2018), mental health and sleep (lancet, 2021), well-being (kim et al., 2017), and business performance (pronk & kottke, 2009), among others (see table 1). furthermore, gómez-chacón et al. (2021) concluded in a study before the covid-19 pandemic, that employees who engaged in both vigorous and moderate physical activity had a significantly higher average score on the five healthy employee strengths. physically active employees are associated with healthier and more resilient employees (gerber et al., 2014), and with higher levels of self-efficacy (salanova et al., 2005). with respect to engagement, the healthier the employee is, the more engaged he/she is in their job (bakker & demerouti, 2017). finally, physical activity leads to optimism and life satisfaction and positive emotions (kim et al., 2017). in times of covid-19, home confinement and telework, a worrying increase in physical inactivity is observed (hall et al., 2021), while chen et al. (2020) stated that physical activity is the best natural medicine to prevent the consequences of confinement, teleworking and stress. the world will recover from the covid-19 pandemic. however, the physical inactivity pandemic will remain and, more worryingly, there is a risk of this pandemic worsening because of covid-19. therefore, the world should realise the need to address the physical inactivity/sedentary pandemic by improving health outcomes under normal conditions and improving humanity’s resilience during future pandemics (hall et al., 2021). european journal of government and economics 11(1), june 2022, 51-72 55 table 1. strengths of the healthy employee and their consequences for the organisation, the interrelationships between these strengths and the influence of physical activity on these strengths. main consequences of each strength is interrelated to influence of physical activity en ga ge m en t engagement increases task performance, organisational performance (bailey et al., 2017), business outcomes (wang et al., 2015), commitment (schaufeli et al., 2002), health and well-being (bailey et al., 2017). improves self-efficacy in work teams (salanova & schaufeli, 2009), resilience and enthusiasm (bakker et al., 2008). workers who practice vigorous or moderate physical activity have a significantly higher average level of engagement (gómez-chacón et al, 2021). the healthier the employee, the more engaged he/she is (bakker & demerouti, 2017). r es ili en ce resilient employees can develop a sense of meaning from difficult and challenging situations (grant & kinman, 2013), positively impacting on job satisfaction and happiness at work (youssef & luthans, 2007), positively relating to personal and work well-being and negatively related to stress, depression, anxiety and mood (liossis et al., 2009). boosts engagement (malik & garg, 2020), self-efficacy (robertson et al., 2015) and optimism and hope (grant & kinman, 2013). physically active workers are associated with healthier and more resilient workers (gerber et al., 2014), are more satisfied and happier (cohn et al., 2009), produce a better business performance (pronk & kottke, 2009) and a greater well-being (liossis et al., 2009). se lfef fic ac y high levels of professional self-efficacy cause people to strive to improve strategies and make sound decisions coping with stress (heslin & klehe, 2006), better performance and productivity, favouring their job satisfaction and dedication (salanova et al., 2000), and leads people to work hard and persist in the face of setbacks, but is negatively related to occupational burnout and anxiety (heslin & klehe, 2006). increases resilience (bandura, 1998), engagement (simbula et al., 2011), optimism and hope (heslin & klexe, 2006). physical activity enhances self-efficacy with employees being less stressed at work (salanova et al., 2005), more motivated (vera et al., 2014), improving their health, job satisfaction and performance (luthans et al., 2007). o pt im is m a nd po si tiv e em ot io ns this promotes satisfaction and happiness (youssef & luthans 2007), better performance (luthans et al. ,2007), motivation and creativity (isen, 2002) and better mental health and well-being in times of adversity (carver et al., 2010). improves resilience (cohn et al., 2009), engagement (carver et al., 2010), self-efficacy (ouweneel et al., 2013), and hope (grant & kinman, 2013). moderate to high employee physical activity will enhance positive emotions leading to being more creative (isen, 2002), satisfied and happier (youssef & luthans, 2007), with a higher performance (luthans et al., 2007) and leads to optimism (kim et al., 2017). h op e hope is the characteristic most likely to improve health, job satisfaction and performance (nelson & simmons, 2006). employees with high hopes possess more goal-related strategies and are more motivated to achieve their goals (peterson et al., 2008). positively correlated with resilience (grant & kinman, 2013). workers with vigorous or moderate physical activity have significantly higher rates of optimism and positive emotions (gómez-chacón et al., 2021). european journal of government and economics 11(1), june 2022, 51-72 56 healthy employee in covid-19 times covid-19 has been associated with undesirable mental health and well-being outcomes (yildirim et al., 2021), including burnout, causing numerous negative effects on workers around the world and could weaken the strengths of the healthy employee, with detrimental consequences for the health of the workers and thus their companies. the covid-19 pandemic is likely to have profound socio-psychological, physical and technical implications for employees trying to adapt to their drastically altered work environments (carnevale & hatak, 2020). prior to covid-19, organisations focused on employee engagement strategies. but the intrusion of covid-19 led to the emasculation of engagement strategies, as organisations were more concerned about their survival. human resource managers are, therefore, grappling with reduced employee engagement (kumar, 2021). work engagement has profound implications for employee performance, and psychological and physical well-being (bakker et al., 2008), being a key indicator for employee health. contributing to personal health and company survival, engagement is therefore a relevant factor to consider in an extreme and disrupted context, such as the covid-19 pandemic (reinwald et al., 2021), but extensive telecommuting, triggered by this pandemic, can also produce disadvantages, such as social isolation and reduced employee engagement (vander elst et al., 2017). an exclusive work-from-home arrangement that extends over a long period tends to generate adverse and negative outcomes and can be demotivating and could hamper employee engagement to the point of burnout (de klerk et al., 2021). resilience, as a positive resource, enables people to navigate a stressful workplace and unexpected events like the covid-19 pandemic (ojo et al., 2021), maintaining or even increasing engagement in times of crisis (salanova, 2021). resilience is also pivotal to coping with stress and vital to staying in balance, most especially during the covid-19 period (vinkers et al., 2020). resilient employees present enthusiasm, optimism and hope, a high degree of autonomy, selfawareness and emotional literacy, flexibility and adaptability, a sense of purpose, and the ability to develop a sense of meaning from difficult and challenging situations (grant & kinman, 2013) such as pandemic times. fostering employee resilience can play a key role in building a highly engaged workforce (malik & garg, 2020). hope is also an important driver of resilience and life satisfaction. moreover, by developing and fostering hope and resilience, organisations can help mitigate unwelcome behaviours during the covid-19 pandemic (rivera et al., 2021). optimism has also been affected by the pandemic, as the stress it has caused not only has had a direct effect on increased burnout but also an indirect effect on it through reduced optimism and social connectedness, while these psychological resources may promote positive mental health and well-being because of facilitating dealing with stress in the face of adversity like the covid-19 pandemic (yildirim et al., 2021). the common understanding of how important health support in the workplace is for the company will lead employees to show more self-care behaviour and it is expected that employees’ self-care behaviour will foster their work engagement and reduce their exhaustion (kaluza et al., josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 57 2021). therefore, if the company were able to adapt its corporate well-being programmes to the times of pandemic and thus combat its harmful consequences on the five strengths of the healthy employee, it could not only avoid these effects, but also reinforce some of these strengths, such as engagement and resilience, which are so important in times of crisis. 3.method the case study method has been selected following bell et al. (2011) as it is appropriate for understanding a phenomenon and the subsequent collection of the multiple data points necessary to enable the corroboration and triangulation of data (creswell, 2002). furthermore, the case study research methodology has been widely used in management research as a source of knowledge in various areas, as a sample of what can happen and as a source of experience (mariotto et al., 2014). following bartunek et al. (2006), when theory is built from case studies these are often reflected as one of the most interesting research methods. a quantitative methodology has been used in the present research. scope of study the study will be carried out in times of covid-19 on the corporate well-being programme of the leading brewery company in spain, with a worldwide presence, and over four thousand employees. their corporate wellbeing programme is a benchmark in spain. at the onset of the pandemic, the programme had to be adapted to the new situation very quickly, once the risks of confinement had been analysed, and multiple actions were implemented, thanks to the benefits of new technologies. the aim of the company is to make all professionals feel safer, healthier and happier, ensuring that all employees are prepared to face current and future challenges in sustainable working and organisational environments. it is about putting people at the centre of the strategy, accompanying people 24 hours a day, during these difficult times, to take care of their well-being and health, and gathering information to take advantage of learning during the pandemic and incorporate it into the future company’s day-today work. to achieve these goals, the programme is comprehensive and cross-cutting, addressing not only physical, but psycho-social aspects, to respond to the many existing challenges, carrying out action in different areas: actions for physical well-being: physical activity programme with the fitness app. personalised plans, on-line and recorded classes. a nutrition programme with tips and personal plans via the well-being website and app. including the option of contacting one’s personal trainer or nutritionist via the app, emails, video calls, telephone, or whatsapp. actions for emotional well-being: emotional coach service with coaches from the company josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 58 itself. daily mindfulness sessions in the morning. motivational talks with both in-house and external staff. actions for psycho-social well-being: psycho-social survey to measure different aspects and act accordingly. communications and recommendations for psycho-social well-being. challenges among workers to promote the idea of belonging, talks on themes such as the importance of sleep. with a 24-hour emergency telephone number to help employees and make them always feel accompanied. due to the broad focus of the programme and its actions, covering physical and psycho-social aspects, it has been considered an ideal programme to analyse the effects that it could have on the strengths of the healthy employee, in times of covid-19. participants a self-administered questionnaire was sent to all 695 employees at the head office. 251 responses (36.4%) were obtained. the estimated average time to complete the survey is ten minutes. it should be noted that in some tables, some respondents have been excluded because they answered incorrectly, which could distort the results. of this sample, 160 (63.7%) were men and 91 (36.2%) women, 71.9% were mostly teleworkers and 68.3% had been with the company for more than 5 years, of whom 45.8% had been with the company for more than 10 years (see table 2). this last figure is particularly important, as the corporate programme started in the company more than 20 years ago. table 2. seniority and gender. seniority men women total % < 1 year 12 4 16 5.9% 1-4 years 28 38 66 25.8% 5-10 years 34 22 56 22.5% >10 years 86 27 113 45.8% total 160 91 251 100.0% instruments two measurement instruments were used to collect quantitative data. first, the international physical activity questionnaire (ipaq) was employed to analyse the level of physical activity. several authors have carried out studies in which the psychometric characteristics of the ipaq have been studied and confirmed its validity with reliability measured by a cronbach’s alpha of 0.88 for the short version in spanish (brown et al., 2004). the short version of the ipaq, with seven items, was used, to classify employees into three different levels: low or no physical activity, josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 59 moderate physical activity, and vigorous physical activity. this short version was validated by mantilla-toloza and gómez-conesa (2007). second, to measure the healthy employee strengths, the adaptation of the healthy and resilient organization (hero) questionnaire (gómez-chacón et al., 2020), was used, analysing five psycho-social strengths: engagement, resilience, self-efficacy, positive emotions and competence, instead of the eight original strengths by salanova et al. (2012). this five-dimension questionnaire, with 40 questions, measured with a 7-point likert-type scale, was used, as it shows a good fit, as the incremental indices (cfi and nnfi) and the absolute smr were above .90 and below .08, respectively (gómez-chacón et al., 2020). it is worth noting that the optimism strength is measured in this questionnaire by the concept of positive emotions, while the hope scale is measured by the concept of competence, so from now on, they will be referred to as such. procedure prior to the collection of data for the study, a meeting was held with those responsible for the programme in the company, in which the format and contents of the self-administered survey were presented, discussed and subsequently approved, which the company would then send out by e-mail. therefore, data were collected by means of a self-administered questionnaire to all employees at the madrid headquarters. data analysis the normality test was performed between the healthy employee variables of the sample with respect to the employees who participated in the physical activity programme, and the different work study variables. the sample is of 251 people, and glass et al. (1972) concluded that when a sample is large (n > 30) parametric tests can be performed since such tests are more powerful than non-parametric tests. an anova analysis and a student’s t-test were carried out to relate the strengths of the healthy employee to the different variables of the study. finally, post hoc tests were performed to determine whether there are significant differences between the different groups. in this case, the α bonferroni correction is used, the most widely employed and wellknown, although more conservative, method. all the analyses were done with the spss 24 statistical software. josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 60 4. results the results obtained, analysing the relationship between the five strengths of the healthy employee (horizontal axis of table 3, table 4, and table 5) with the variables physical activity, use of the fitness app, age, seniority, supervisor guidance and working in person or remotely, are presented as follows. the employees present higher averages in high physical activity with respect to physical activity in all the strengths of the healthy employee, with no significant differences (table 3). for example, in competence, employees who practice high physical activity have a median of 4.88, and in low physical activity their median is 4.59. no significant differences were found in the relationship between the strengths of the healthy employee and the use or a digital fitness app (table 3). table 3. healthy employee according to physical activity levels and use of the fitness app. engagement resilience self-efficacy positive emot. competence n av sd av sd av sd av sd av sd physical activity high 41 4.7 0.672 4.68 0.65 5.14 0.781 3.61 0.976 4.88 0.745 moderate 157 4.45 0.763 4.41 0.81 4.86 0.916 3.42 1.055 4.58 0.871 low 53 4.45 0.78 4.54 0.76 4.96 0.805 3.51 1.029 4.59 0.861 total 251 4.49 0.755 4.48 0.78 4.93 0.876 3.47 1.036 4.63 0.854 use of fitness app yes 131 4.48 0.789 4.41 0.719 4.83 0.929 3.39 1.057 4.56 0.855 no 42 4.41 0.758 4.45 0.9 5.14 0.675 3.44 1.073 4.83 0.731 do not participate 78 4.57 0.697 4.63 0.806 4.89 0.864 3.63 0.975 4.66 0.904 total 251 4.5 0.756 4.48 0.782 4.93 0.876 3.48 1.036 4.64 0.854 notes: n: sample; av: average. sd: standard deviation. in relation to age, employees show a disparity of means in the strengths of the healthy employee (table 4). after using the bonferroni correction, significant differences can be found in engagement and resilience. for example, in engagement between employees aged 30-39 years (m = 4.21) compared to those aged 40-49 years (m = 4.62) and 50-59 years (m = 4.59), and in resilience between employees aged 30-39 (m = 4.21) compared to employees aged 40-49 (m = 4.56), and employees aged 50-59 (m = 4.80). with respect to employee seniority (table 4), after using the bonferroni correction, there are significant differences in the resilience strength, between 5-10 years (m = 4.27), and more than 10 years (m = 4.66). josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 61 table 4. healthy employee according to age and seniority. engagement resilience self-efficacy positive emotions competence n av sd av sd av sd av sd av sd age 20-29 years 10 4.36 0.56 4.45 0.67 4.76 0.916 2.95 1.024 4.66 0.842 30-39 years 68 4.21* 0.898 4.21* 0.96 4.69 0.928 3.3 1.074 4.4 0.947 40-49 years 87 4.62** 0.708 4.56* 0.69 5 0.889 3.63 1.001 4.78 0.788 50-59 years 59 4.59* 0.634 4.8*** 0.65 5.08 0.764 3.48 1.038 4.69 0.862 60-69 years 11 4.42 0.799 4.46 0.55 4.87 0.734 3.04 1.176 4.59 0.554 total 235 4.47 0.766 4.51 0.79 4.91 0.873 3.44 1.05 4.63 0.857 seniority <1 year 16 4.88 0.649 4.71 1.01 4.9 0.909 4.08 1.245 4.53 1.092 1-4 years 66 4.31 0.824 4.36 0.81 4.74 0.969 3.28 1.064 4.49 0.889 5-9 years 56 4.47 0.847 4.27** 0.87 4.86 0.894 3.54 0.949 4.61 0.868 >10 years 113 4.52 0.68 4.66** 0.66 5.03 0.815 3.4 1.043 4.75 0.802 total 251 4.48 0.764 4.49 0.79 4.91 0.882 3.44 1.051 4.64 0.86 notes: * p < .05, ** p < .01, *** p < .001; n: sample; av: average, sd: standard deviation. in relation to the form of work, presential or working-from-home, no significant differences were found (table 5). regarding the supervisor guidance, there are significant differences in all the strengths. the better the supervisor guidance, the higher the scores obtained in all the strengths of the healthy employee (table 5). in the following table (table 5), when using the bonferroni correction, the significant differences are established between a very well-guided attention from the supervisor with respect to well-guided / guided / poorly guided / very poorly guided, i.e., in positive emotions very well guided (m = 3.86) and poorly guided (m = 2.5). table 5. employee strengths according to supervisor guidance and type of work during covid-19. engagement resilience self-efficacy positive emotions competence n av sd av sd av sd av sd av sd supervisor guidance very badly guided 3 2.9*** 1.945 2.09*** 1.8 3.88 1.539 1.94** 0.855 3.61 0.787 badly guided 7 3.26*** 1.098 3.34*** 1.05 3.23*** 1.329 2.5** 1.326 3.42* 1.731 guided 21 4.23* 0.635 4.12*** 0.88 4.82 0.916 2.96** 0.939 4.33 0.892 well guided 125 4.45* 0.695 4.45** 0.61 4.87 0.82 3.35** 0.938 4.66 0.786 very well guided 94 4.74*** 0.607 4.76*** 0.67 5.18*** 0.697 3.86** 1 4.78* 0.752 total 250 4.49 0.757 4.48 0.78 4.92 0.875 3.47 1.038 4.63 0.855 type of work during covid-19 mainly telework 179 4.49 0.776 4.45 0.76 4.93 0.875 3.5 1.022 4.59 0.838 mainly presential 70 4.48 0.705 4.55 0.85 4.93 0.878 3.43 1.07 4.73 0.897 notes: * p < .05, ** p < .01, *** p < .001; n: sample; av: average, sd: standard deviation. josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 62 5. discussion the health crisis caused by covid-19 is creating considerable uncertainty among workers, which is compromising their engagement and well-being. organisations need to actively support the health and well-being of employees (kniffin et al., 2021). given the positive impact of employee engagement and well-being on efficiency, productivity, and organisational performance, managers should address these factors to maintain the firm’s competitive edge (de-la-calledurán et al., 2021) even more during the covid-19 pandemic. the strengths of the healthy employee, engagement, resilience, self-efficacy, positive emotions and competence are being affected by the pandemic. aware of the health risks derived from the increasing physical inactivity caused by the pandemic (hall et al., 2021), the world health organisation (who) has been insisting with its recommendations related to physical activity, diet, tobacco and mental health during confinement (world health organization, who 2020). the aggravation of physical inactivity emerges as a relevant adverse effect of covid-19 (hall et al., 2021). however, if the company gets workers to engage in moderate or high levels of physical activity, it will have more satisfied and happier workers (cohn et al., 2009), who provide a better business performance (pronk & kottke, 2009) are more motivated (vera et al., 2014), and have greater well-being (liossis et al., 2009) and engagement (gómez-chacón et al., 2021). based on scientific evidence, maintaining a regular exercise routine is a key strategy for physical and mental health during the current coronavirus emergency (maugeri et al., 2020). there is a significantly higher increased risk for chronic disease if one is physically inactive and leads a sedentary lifestyle (hall et al., 2021). furthermore, physical activity is strongly associated with a reduced risk for severe covid-19 outcomes (sallis et al., 2021). for these reasons, among others, it is so important to develop a good corporate wellness programme, even more so in times of covid-19. bouziri et al. (2020) stated that maximising the health benefits of telework, while minimising its negative effects, constitutes a continuation of a company’s duty to preserve the health of its employees. it is imperative to invest in employee engagement in a more meaningful way to take account of the “new normal”, and be sure that all employees, during these times, trust the company and its care concerning them (kumar, 2021). as a positive resource, resilience is also very important, as it allows people to navigate a stressful workplace and unexpected events (ojo et al., 2021). wang et al (2020) highlighted the importance of improving mental health and psychological resilience during the covid-19 epidemic. in this case study, workers with higher physical activity present better data on all strengths of the healthy employee compared to the other two types of physical activity, but there are no significant differences. this positive relationship between physical activity and healthy employee strengths is in line with gómez-chacón et al. (2021) in terms of engagement, enhancing selfefficacy (salanova et al., 2005), being associated with healthier and more resilient workers (gerber et al., 2014), and positive emotions (isen, 2002). it also agrees with chen et al. (2020) who asserted that physical activity is the best natural medicine to prevent the consequences of confinement and telework stress. likewise, these results are also in line with the research carried josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 63 out in the same company by núñez-sánchez et al. (2021), which highlighted that during this period 57.9 % have increased or maintained their physical activity prior to the pandemic, which contrasts with hall et al. (2021) who stated that covid-19 is increasing physical inactivity in the world, and in spain (lópez-bueno et al., 2020). the promotion of home-based physical activities is particularly recommendable in this specific context to reach the recommended pa (physical activity) levels set out by the world health organisation (lópez-bueno et al., 2020). when studying the relationship between the strengths of the healthy employee and the use of the digital fitness app, no significant differences were found. this is due to the fact that within the actions established by the company, there can be found not only actions for physical wellness, with the support of the digital app, but actions for psycho-social and emotional well-being, with coaching and mindfulness sessions, among others. it should be noted that the digital application was only one of the various digital tools used in this period. among the different digital tools used, there can be found: a fitness app, direct communication through emails, microsoft teams, telephone and whatsapp, the uploading of fitness and wellness content to their corporate wellness website, a new private youtube channel for online activities, subsequently uploaded for on-demand use and, finally, training routines for employees, adapted to these pandemic times. this strategy is in line with iglesias-sánchez et al. (2020), who indicated that the only way to adapt to the new confinement and telework environment is through social media and digital ecosystems. best practices include establishing a culture of health and using strategic communications designed to educate, motivate, and build trust, also tailored and targeted, multichannelled, bidirectional, with optimum frequency (kent et al., 2016). according to peláez et al. (2020), coaching can be a valuable applied positive psychology intervention to increase wellbeing and optimal functioning in organisations, while good et al. (2016) observed that mindfulness can foster performance and work well-being by increasing personal resources, such as resilience. these psycho-social and emotional actions are in line with trougakos et al. (2020), who stated that organisations should help employees mitigate anxiety by offering training in effective emotional coping methods, as well as strategies to ensure that they meet their psychological needs by offering webinars in covid-19 times on topics such as resilience, stress management and work-life balance. multi-component interventions, such as this one, foster participation (robroek, 2019). regarding age, employees show a disparity of means in the strengths of the healthy employee. however, engagement and resilience show significant differences between the youngest and the groups between 40 and 59 years, which show better averages. in terms of seniority, employees show significant differences, with employees having over 10 years seniority presenting better resilience. resilience is an important strength, especially in times of a pandemic, as according to malik and garg (2020), it enhances engagement employees with a seniority of between 1 and 10 years perform worse on all strengths of the healthy employee (except for positive emotions) compared to employees with more than 10 years and those with less than 1. in this research, it is observed that employees with more than 10 years in the company have an above-average level of engagement, being also the seniority band with the highest number of josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 64 employees. in the authors’ opinion, these conclusions are logical given that the corporate wellness programme at this company began in 2001, so workers with more than 10 years’ seniority are fully familiar with the programme and with the importance of health and well-being for the company. it is worth highlighting the good results of employees with less than 1 year of seniority, which speaks very well of the current onboarding process and the acceptance of all the welfare plans by new employees. the importance of the years of implementation of the programme is in line with basinska-zych and springer (2021) who observed in different research works that the financial effectiveness of the workplace health promotion intervention, needs at least three years to obtain positive results. in relation to the way employees have performed their job functions during the pandemic (work-from-home or face-to-face), no significant differences were found. in the authors’ opinion, this demonstrates the overall good functioning of the programme and is in line with ten brummelhuis and bakker (2012), who stated that organisational and managerial support is of great importance in the work-from-home experience to increase levels of personal resources. this support, in turn, promotes engagement and well-being (wood et al., 2020). in addition, coaching can be a valuable applied positive psychology intervention to increase well-being and optimal functioning in companies (peláez et al., 2020). regarding the supervisor’s guidance, significant differences were observed in all strengths of the healthy employee. when the guidance is better, higher scores are obtained. this finding is in line with kaluza et al. (2021), as employees desire leaders who value their well-being. furthermore, their explicit commitment and consideration of employee well-being has a significant impact on employees’ psychological and physical well-being over and above other forms of such positive leadership behaviours (vincent-höper et al., 2019). health-oriented leadership improves employees’ health and well-being through employees’ own self-care behaviour, i.e., their concern for their own health, how they manage demands and resources at work and how they take care of their own well-being (franke et al., 2014). according to salanova et al. (2021), groups and individuals value having a management team committed to both holistic health and the development and promotion of health at work. finally, the authors compared the averages obtained in this research with that carried out by gómez-chacón et al. (2021), before the pandemic, in three different companies with a corporate well-being programme. it was observed that the averages obtained in research during the covid19 pandemic are better in engagement, resilience, competence and self-efficacy, only being lower in positive emotions which, in the opinion of the authors, makes sense. this is also in accordance with yildirim et al. (2021), who observed that the stress caused by the pandemic, increased burnout and reduced performance. josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 65 limitations and future lines of research this research is pioneer in the analysis of the effect of corporate well-being programmes in the healthy employee concept, in covid-19 times. the study reflects the outcomes of the corporate wellness programme adaptation to these uncertain times but presents limitations. the first one is that this company has more than 20 years of experience in the management of corporate well-being programmes, being a benchmark in spain, and also a large company and therefore has more experience and resources than other smaller companies. this could limit the adoption of their model, at least in its entirety, by other companies presenting different characteristics. another limitation is that the research has taken place during the covid-19 pandemic and this extraordinary new situation may have influenced some of the outcomes. it is true that the covid-19 pandemic is still active, and therefore it should be borne in mind that the company can continue to create and adopt new measures and participation levels, and physical activity may also vary. it would therefore be of interest to researchers and practitioners to update the findings of this study as well as to provide more quantitative and qualitative assessments of the implementation of the programme in the company once the pandemic is over. it would also be of interest to develop similar studies in a different type of companies. the grand challenge we currently face constitutes a “new reality” that offers new opportunities to which organisational scholars and practitioners alike will need and want to remain attentive (carnevale & hatak, 2020). although working from home has become the standard of working for millions, there is little previous research on how this influences employee experience and engagement (masuda et al., 2017). in accordance with kniffin et al. (2021), the impact of covid-19 on social, psychological, health-related and economic costs for workers should also be investigated and updated after the pandemic is over. research into the most appropriate strengths of the healthy employee is alive and is changing in parallel with the progress of society and companies. in this sense, trust and leadership are variables that should be studied in the future due to their impact on organisations. furthermore, it has been observed that when supervisor guidance is better, a higher score is obtained in the strengths of the healthy employee. for this reason, it would be interesting to investigate healthy leadership models and skills since, according to de klerk et al. (2021), leaders will need to develop skills to manage their employees working remotely and provide physical and mental support. finally, conducting future studies based on the position held by the employees, could also be of interest to test if this position influences any of the healthy employee strengths. managerial perspectives this research could help companies in their adaptation of well-being programmes, as through this case study they could draw conclusions and obtain practical ideas. according to carnevale and hatak (2020) further guidance is needed on how companies can adapt their human resources josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 66 practices in ways that can alleviate the aforementioned issues and enhance employees’ ability to thrive during such dynamic and uncertain times. organisations have a duty to protect and care for the well-being of employees (grant & kinman, 2013), thus improving psycho-social strengths such as resilience or engagement, and this becoming a competitive advantage (schneider et al, 2018). this research might be helpful in mitigating the harmful effects of this pandemic. increased efforts are needed to disseminate lessons learned from employers who have built cultures of health and excellent communications strategies applying these insights more broadly in workplace settings (kent et al, 2016). in this regard, salanova, (2020) underlined the importance of pandemic research in generating resources to foster resilience and engagement, while de klerk et al. (2021) highlighted the opportunity to study the effect of this forced telework on the employee experience in order to plan future ways of working. furthermore, it should be underlined that as the physical activity level is an important factor that can influence some of the strengths of the healthy employee, it is highly advisable for employers to implement effective corporate wellness programmes to trigger their employees to maintain or even increase their physical activity. organisations have found that they can function effectively with a remote workforce and, according to experts, this trend will remain. this research suggests that corporate wellness programmes, adapted to times of covid-19 and to work-from-home, could be helpful in combating the negative psycho-social and physical consequences for the workforce. this empirical work may enlighten companies to face the challenges of this new age of work and not only fight against reduced strengths of healthy employees but maintain or even improve them. 6. conclusions the aim of the research is to show, through a real case study, the effects that a multi-component corporate well-being programme, implemented in covid-19 times, and extensive remote working, has on the healthy employee´s strengths. future interventions in organisations are needed to improve mental health during this pandemic, building self-efficacy and resilience (hu et al., 2021) and this case study may be helpful in doing so. physical activity programmes have positive effects on the well-being and the strengths of the healthy employee. however, they must be complemented with other psycho-social programmes, being multi-component, highlighting the importance of coaching and mindfulness actions. on the other hand, the importance of developing a culture of well-being, the use of different digital tools (a fitness app is not enough) as well as the importance of health-oriented leaders and companies for the employees is also demonstrated. the results in two of the most important strengths of the healthy employee for the company, engagement and resilience, which have also suffered a good deal in times of the pandemic, offer better averages compared with other companies in a precovid situation. whereas companies are grappling with reduced employee engagement among other harmful consequences, this case study suggests that a good corporate wellness josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 67 programme could help mitigate these negative effects. work engagement has profound implications for employees’ performance and their psychological and physical well-being (bakker et al., 2008), being a key indicator for employee health (salanova, 2021). knight et al. (2019) have shown that there are benefits of workplace interventions to improve engagement and this research shows that it is also possible to achieve these results during a pandemic and telework. it is highly recommended for organisations to actively support the well-being and health of their workforce to fight against the adverse effects of this pandemic. funding. this research received funds from a plan propio of the university of málaga and from andalusian government sej-628. uma18 feder ja-148. acknowledgments: the authors thank the company for its commitment and generosity in sharing this corporate well-being programme information. josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 68 references altena, e., baglioni, c., espie, c. a., ellis, j., gavriloff, d., holzinger, b., ... & riemann, d. (2020). dealing with sleep problems during home confinement due to the covid‐19 outbreak: practical recommendations from a task force of the european cbt‐i academy. journal of sleep research, 29(4), e13052. https://doi.org/10.1111/jsr.13052 bailey, c., madden, a., alfes, k., & fletcher, l. (2017). the meaning, antecedents and outcomes of employee engagement: a narrative synthesis. international journal of management reviews, 19(1), 31–53. https://doi.org/10.1111/ijmr.12077 bakker, a. b., schaufeli, w. b., leiter, m. p., & taris, t. w. (2008) work engagement: an emerging concept in occupational health psychology. work stress, 22, 187–200. https://doi.org/10.1080/02678370802393649 bakker, a. b., & demerouti, e. (2017). job demands–resources theory: taking stock and looking forward. journal of occupational health psychology, 22(3), 273. https://psycnet.apa.org/doi/10.1037/ocp0000056 bandura, a. (1997). self-efficacy: the exercise of control. new york: freeman. bandura, a. (1998). health promotion from the perspective of social cognitive theory. psychology and health, 13, 623-649. bartunek, j. m., rynes, s. l., & ireland, r. d. (2006). what makes management research interesting, and why does it matter? academy of management journal, 49(1), 9-15. https://doi.org/10.5465/amj.2006.20785494 basinska-zych, a., & springer, a. (2021). organizational and individual outcomes of health promotion strategies-a review of empirical research. international journal of environment research and public health, 18, 383. https://doi.org/10.3390/ijerph18020383 bell, e., bryman, a., & harley, b. (2018). business research methods. oxford university press. bezner, j. r., franklin, k. a., lloyd, l. k., & crixell, s. h. (2020). effect of group health behaviour change coaching on psychosocial constructs associated with physical activity among university employees. international journal of sport and exercise psychology, 18(1), 93-107. bouziri, h., smith, d. r., descatha, a., dab, w., & jean, k. (2020). working from home in the time of covid-19: how to best preserve occupational health? occupational and environmental medicine, 77(7), 509-510. brown, w. j., trost, s. g., bauman, a., mummery, k., & owen, n. (2004). test-retest reliability of four physical activity measures used in population surveys. journal of science and medicine in sport, 7(2), 205-215. caniëls, m. c., semeijn, j. h., & renders, i. h. (2018). mind the mindset! the interaction of proactive personality, transformational leadership and growth mindset for engagement at work. career development international, 23(1), 48-66. carnevale, j. b., & hatak, i. (2020). employee adjustment and well-being in the era of covid19: implications for human resource management. journal of business research, 116, 183187. https://doi.org/10.1016/j.jbusres.2020.05.037 carver, c. s., scheier, m. f., & segerstrom, s. c. (2010). optimism. clinical psychology review, 30(7), 879-889. https://doi.org/10.1016/j.cpr.2010.01.006 chen, p., mao, l., nassis, g. p., harmer, p., ainsworth, b. e., & li, f. (2020). coronavirus disease (covid-19): the need to maintain regular physical activity while taking precautions. journal of sport and health science, 9(2), 103-104. https://doi.org/10.1016/j.jshs.2020.02.001 cohn, m. a., fredrickson, b. l., brown, s. l., mikels, j. a., & conway, a. m. (2009). happiness unpacked: positive emotions increase life satisfaction by building resilience. emotion, 9(3), 361–368. https://doi.org/10.1037/a0015952 creswell, j. w. (2002). educational research: planning, conducting, and evaluating quantitative. upper saddle river, nj: prentice hall. de klerk, j. j., joubert, m., & mosca, h. f. (2021). is working from home the new workplace panacea? lessons from the covid-19 pandemic for the future world of work. sa journal of industrial psychology, 47(1), 1-14. https://doi.org/10.4102/sajip.v47i0.1883 de-la-calle-durán, m.-c., & rodríguez-sánchez, j. l. (2021). employee engagement and wellbeing in times of covid-19: a proposal of the 5cs model. international journal of environment research and public health, 18, 5470. https:// doi.org/10.3390/ijerph18105470 eurofound (2020). living, working and covid-19: first findings-april 2020; publications office of the european union: luxembourg. https://doi.org/10.1111/jsr.13052 https://doi.org/10.1111/ijmr.12077 https://doi.org/10.1080/02678370802393649 https://psycnet.apa.org/doi/10.1037/ocp0000056 https://doi.org/10.5465/amj.2006.20785494 https://doi.org/10.3390/ijerph18020383 https://doi.org/10.1016/j.jbusres.2020.05.037 https://doi.org/10.1016/j.cpr.2010.01.006 https://doi.org/10.1016/j.jshs.2020.02.001 https://doi.org/10.1037/a0015952 https://doi.org/10.4102/sajip.v47i0.1883 josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 69 franke, f., felfe, j., & pundt, a. (2014). the impact of health-oriented leadership on follower health: development and test of a new instrument measuring health-promoting leadership. german journal of human resource management, 28(1-2), 139-161. https://doi.org/10.1177%2f239700221402800108 gerber, m., jonsdottir, i. h., lindwall, m., & ahlborg, g. (2014). physical activity in employees with differing occupational stress and mental health profiles: a latent profile analysis. psychology of sport and exercise, 15(6), 649-658. https://doi.org/10.1016/j.psychsport.2014.07.012 gómez-chacón, r., garcía-fernández, j., morales sánchez, v., & hernández mendo, a. (2020). adaptación y validación del cuestionario del empleado saludable del modelo hero. anales de psicología / annals of psychology, 36(2), 361–369. https://doi.org/10.6018/analesps.395431 gómez-chacón, r., garcía-fernández, j., morales sánchez, v., & hernández mendo, a. (2021). healthy employee’s psychosocial strengths in workers according to their physical activity level. revista mexicana de psicología, 38(1), 30–44. good, d. j., lyddy, c. j., glomb, t. m., bono, j. e., brown, k. w., duffy, m. k., … lazar, s. w. (2016). contemplating mindfulness at work: an integrative review. journal of management, 42(1), 114–142. https://doi.org/10.1177%2f0149206315617003 grant, l., & kinman, g. (2013). bouncing back? personal representations of resilience of student and experienced social workers. practice, 25(5), 349-366. https://doi.org/10.1080/09503153.2013.860092 helland, m. r., & winston, b. e. (2005). towards a deeper understanding of hope and leadership. journal of leadership & organizational studies, 12(2), 42-54. https://doi.org/10.1177%2f107179190501200204 hall, g., laddu, d. r., phillips, s. a., lavie, c. j., & arena, r. (2021). a tale of two pandemics: how will covid-19 and global trends in physical inactivity and sedentary behavior affect one another? progress in cardiovascular diseases, 64, 108. https://dx.doi.org/10.1016%2fj.pcad.2020.04.005 heslin, p. a., & klehe, u. c. (2006). self-efficacy. in s. g. rogelberg (ed.), encyclopedia of industrial/organizational psychology (vol. 2, pp. 705-708). thousand oaks: sage. hu, d., kong, y., li, w., han, q., zhang, x., zhu, l. x., & he, h. g. (2020). frontline nurses’ burnout, anxiety, depression, and fear statuses and their associated factors during the covid19 outbreak in wuhan, china: a large-scale cross-sectional study. the lancet 100424. https://doi.org/10.1016/j.eclinm.2020.100424 iglesias-sánchez, p. p., vaccaro witt, g. f., cabrera, f. e., & jambrino-maldonado, c. (2020). the contagion of sentiments during the covid-19 pandemic crisis: the case of isolation in spain. international journal of environmental research and public health, 17(16), 5918. isen, a. m. (2002). missing in action in the aim: positive affect’s facilitation of cognitive flexibility, innovation, and problem solving. psychological inquiry, 13(1), 57–65. kaluza, a. j., weber, f., van dick, r., & junker, n. m. (2021). when and how health‐oriented leadership relates to employee well‐being—the role of expectations, self‐care, and lmx. journal of applied social psychology, 51(4), 404-424. https://doi.org/10.1111/jasp.12744 kazekami, s. (2020). mechanisms to improve labour productivity by performing telework. telecommunications policy, 44(2), 1–15. https://doi.org/10.1016/j.telpol.2019.101868 kent, k., goetzel, r. z., roemer, e. c., prasad, a., & freundlich, n. (2016). promoting healthy workplaces by building cultures of health and applying strategic communications. journal of occupational and environmental medicine, 58(2), 114-122. https://doi.org/10.1097/jom.0000000000000629 kim, j., lee, s., chun, s., han, a., & heo, j. (2017). the effects of leisure-time physical activity for optimism, life satisfaction, psychological well-being, and positive affect among older adults with loneliness. annals of leisure research, 20(4), 406-415. https://doi.org/10.1080/11745398.2016.1238308 kniffin, k. m., narayanan, j., anseel, f., antonakis, j., ashford, s. p., bakker, a. b., ... & vugt, m. v. (2021). covid-19 and the workplace: implications, issues, and insights for future research and action. american psychologist, 76(1), 63. https://psycnet.apa.org/doi/10.1037/amp0000716 knight, c., patterson, m., & dawson, j. (2019). work engagement interventions can be effective: a systematic review. european journal of work and organizational psychology, 28(3), 348372.: https://doi.org/10.1080/1359432x.2019.1588887 https://doi.org/10.1177%2f239700221402800108 https://doi.org/10.1016/j.psychsport.2014.07.012 https://doi.org/10.6018/analesps.395431 https://doi.org/10.1177%2f0149206315617003 https://doi.org/10.1080/09503153.2013.860092 https://doi.org/10.1177%2f107179190501200204 https://dx.doi.org/10.1016%2fj.pcad.2020.04.005 https://doi.org/10.1016/j.eclinm.2020.100424 https://doi.org/10.1111/jasp.12744 https://doi.org/10.1016/j.telpol.2019.101868 https://doi.org/10.1097/jom.0000000000000629 https://doi.org/10.1080/11745398.2016.1238308 https://psycnet.apa.org/doi/10.1037/amp0000716 https://doi.org/10.1080/1359432x.2019.1588887 josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 70 kumar, p. (2021). v-5 model of employee engagement during covid-19 and post lockdown. vision, 0972262920980878. https://doi.org/10.1177%2f0972262920980878 lancet, t. (2021). a sporting chance: physical activity as part of everyday life. lancet (london, england), s0140-6736. li, s., wang, y., xue, j., zhao, n., & zhu, t. (2020). the impact of covid-19 epidemic declaration on psychological consequences: a study on active weibo users. international journal of environmental research and public health, 17(6), 2032. https://doi.org/10.3390/ijerph17062032 liossis, p. l., shochet, i. m., millear, p. m., & biggs, h. (2009). the promoting adult resilience (par) program: the effectiveness of the second, shorter pilot of a workplace prevention program. behaviour change, 26(2), 97-112. https://doi.org/10.1375/bech.26.2.97 llorens, s., salanova, m., torrente, p., & acosta, h. (2013). interventions to promote healthy & resilient organizations (hero) from positive psychology. in salutogenic organizations and change (pp. 91-106). dordrecht: springer. lópez-bueno, r., calatayud, j., andersen, l. l., balsalobre-fernández, c., casaña, j., casajús, j. a., ... & lópez-sánchez, g. f. (2020). immediate impact of the covid-19 confinement on physical activity levels in spanish adults. sustainability, 12(14), 5708. https://doi.org/10.3390/su12145708 luthans, f., youssef, c. m., & avolio, b. j. (2007). psychological capital: investing and developing positive organizational behavior. positive organizational behavior, 1(2), 9-24. malik, p., & garg, p. (2020). learning organization and work engagement: the mediating role of employee resilience. the international journal of human resource management, 31(8), 10711094. https://doi.org/10.1080/09585192.2017.1396549 mantilla-toloza, s. c., & gómez-conesa, a. (2007). international physical activity questionnaire. an adequate instrument in population physical activity monitoring. revista iberoamericana de fisioterapia y kinesiologia, 10(1), 48-52. mariotto, f. l., zanni, p. p., & moraes, g. s. m. (2014) what is the use of a single-case study in management research? revista de administración de empresas, 54, 358–369. masuda, a. d., holtschalg, c., & nicklin, j. m. (2017). why the availability of telecommuting matters: the effects of telecommuting on engagement via goal pursuit. career development international, 22(2), 200–219. https://doi.org/10.1108/cdi-05-2016-0064 mathisen, g. e., brønnick, k., arntzen, k. j., & bergh, l. i. v. (2017). identifying and managing psychosocial risks during organizational restructuring: it’s what you do and how you do it. safety science, 100, 20-29. https://doi.org/10.1016/j.ssci.2016.12.007 maugeri, g., castrogiovanni, p., battaglia, g., pippi, r., d'agata, v., palma, a., ... & musumeci, g. (2020). the impact of physical activity on psychological health during covid-19 pandemic in italy. heliyon, 6(6), e04315. https://doi.org/10.1016/j.heliyon.2020.e04315 nelson, d. l., & simmons, b. l. (2003). health psychology and work stress: a more positive approach. in j. c. quick & l. e. tetrick (eds.), handbook of occupational health psychology (pp. 97–119). american psychological association. https://doi.org/10.1037/10474-005 nieman, d. c., & wentz, l. m. (2019). the compelling link between physical activity and the body’s defence system. journal of sport health science, 8(3), 201–217. https://doi.org/10.1016/j.jshs.2018.09.009 niitsu, k., houfek, j. f., barron, c. r., stoltenberg, s. f., kupzyk, k. a., & rice, m. j. (2017). a concept analysis of resilience integrating genetics. issues in mental health nursing, 38(11), 896-906. núñez-sánchez, j. m., gómez-chacón, r., jambrino-maldonado, c., & garcía-fernández, j. (2021). corporate well-being programme in covid-19 times. the mahou san miguel case study. sustainability, 13(11), 6189. https://doi.org/10.3390/su13116189 ojo, a. o., fawehinmi, o., & yusliza, m. y. (2021). examining the predictors of resilience and work engagement during the covid-19 pandemic. sustainability, 13(5), 2902. https://doi.org/10.3390/su13052902 ouweneel, e., le blanc, p. m., & schaufeli, w. b. (2013). do-it-yourself: an online positive psychology intervention to promote positive emotions, self‐efficacy, and engagement at work. career development international, 18(2), 173-195. parker, s. l., jimmieson, n. l., & johnson, k. m. (2013). general self-efficacy influences affective task reactions during a work simulation: the temporal effects of changes in workload at different levels of control. anxiety, stress & coping, 26(2), 217-239. https://doi.org/10.1177%2f0972262920980878 https://doi.org/10.3390/ijerph17062032 https://doi.org/10.1375/bech.26.2.97 https://doi.org/10.3390/su12145708 https://doi.org/10.1080/09585192.2017.1396549 https://doi.org/10.1108/cdi-05-2016-0064 https://doi.org/10.1016/j.ssci.2016.12.007 https://doi.org/10.1016/j.heliyon.2020.e04315 https://doi.org/10.1037/10474-005 https://doi.org/10.1016/j.jshs.2018.09.009 https://doi.org/10.3390/su13116189 https://doi.org/10.3390/su13052902 josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 71 peláez, m. j., coo, c., & salanova, m. (2020). facilitating work engagement and performance through strengths-based micro-coaching: a controlled trial study. journal of happiness studies, 21,1265–1284. https://doi.org/10.1007/s10902-019-00127-5 peterson, s. j., & byron, k. (2008). exploring the role of hope in job performance: results from four studies. journal of organizational behavior: the international journal of industrial, occupational and organizational psychology and behavior, 29(6), 785-803. https://doi.org/10.1002/job.492 pronk, n. p., & kottke, t. e. (2009). physical activity promotion as a strategic corporate priority to improve worker health and business performance. preventive medicine, 49(4), 316-321. https://doi.org/10.1016/j.ypmed.2009.06.025 reinwald, m., zimmermann, s., & kunze, f. (2021). working in the eye of the pandemic: local covid-19 infections and daily employee engagement. frontiers psychology, 12, 654126. doi: 10.3389/fpsyg.2021.654126 rivera, m., shapoval, v., & medeiros, m. (2021). the relationship between career adaptability, hope, resilience, and life satisfaction for hospitality students in times of covid-19. journal of hospitality, leisure, sport & tourism education, 29, 100344. https://doi.org/10.1016/j.jhlste.2021.100344 robertson, i. t., cooper, c. l., sarkar, m., & curran, t. (2015). resilience training in the workplace from 2003 to 2014: a systematic review. journal of occupational and organizational psychology, 88(3), 533–562. https://doi.org/10.1111/joop.12120 robroek, s. j., van lenthe, f. j., van empelen, p., & burdorf, a. (2009). determinants of participation in worksite health promotion programmes: a systematic review. international journal of behavioral nutrition and physical activity, 6(1), 1-12. doi:10.1186/1479-5868-6-26 salanova, m. (2020). how to survive covid-19? notes from organisational resilience (¿cómo sobrevivir al covid-19? apuntes desde la resiliencia organizacional). international journal of social psychology, 35(3), 670-676. https://doi.org/10.1080/02134748.2020.1795397 salanova, m. (2021). work engagement: a key to heros–healthy and resilient organizations. in a research agenda for employee engagement in a changing world of work. edward elgar publishing. salanova, m., grau, r., cifre, e., & llorens, s. (2000). computer training, frequency of use and burnout: the moderating role of computer self-efficacy. computers in human behaviour, 16(6), 575-590. salanova, m., grau, r., & martinez, i. (2005) demandas laborales y conductas de afrontamiento: el rol modulador de la autoeficacia profesional. psicothema, 17(3), 390-395. salanova, m., & schaufeli, w. (2009). el engagement en el trabajo: cuando el trabajo se convierte en pasión. alianza editorial: madrid. salanova, m., llorens, s., cifre, e., & martínez, i. m. (2012). we need a hero! towards a validation of the healthy & resilient organization (hero) model. group & organization management, 37(6), 785–822. doi: 10.1177/1059601112470405. sallis, r., young, d. r., tartof, s. y., sallis, j. f., sall, j., li, q., smith, g. n., & cohen, d. a. (2021). physical inactivity is associated with a higher risk for severe covid-19 outcomes: a study in 48,440 adult patients. british journal sports medicine, 1-8, doi:10.1136/bjsports2021-104080. schaufeli, w. b., salanova, m., gonzalez-roma, v., & bakker, a. b. (2002). the measurement of engagement and burnout: a two-sample confirmatory factor analytic approach. journal of happiness studies, 3(1), 71-92. https://doi.org/10.1023/a:1015630930326 schneider, b., yost, a. b., kropp, a., kind, c., & lam, h. (2018). workforce engagement: what it is, what drives it, and why it matters for organisational performance. journal of organisational behaviour, 39(4), 462–480. https://doi.org/10.1002/job.2244 simbula, s., guglielmi, d., & schaufeli, w.b. (2011). a three-wave study of job resources, selfefficacy, and work engagement among italian schoolteachers. european journal of work and organizational psychology, 20(3), 285–304. ten brummelhuis, l. l., & bakker, a. b. (2012). a resource perspective on the work-home interface: the work-home resources model. american psychologist, 67(7),545–556. https://doi.org/10.1037/a0027974 trougakos, j. p., chawla, n., & mccarthy, j. m. (2020). working in a pandemic: exploring the impact of covid-19 health anxiety on work, family, and health outcomes. journal of applied psychology, 105(11), 1234-1245. https://doi.org/10.1037/apl0000739 https://doi.org/10.1007/s10902-019-00127-5 https://doi.org/10.1002/job.492 https://doi.org/10.1016/j.ypmed.2009.06.025 https://doi.org/10.1016/j.jhlste.2021.100344 https://doi.org/10.1111/joop.12120 https://doi.org/10.1080/02134748.2020.1795397 https://doi.org/10.1177/1059601112470405 https://doi.org/10.1023/a:1015630930326 https://doi.org/10.1002/job.2244 https://doi.org/10.1037/a0027974 https://doi.org/10.1037/apl0000739 josé manuel núñez-sánchez et al. / european journal of government and economics 11(1), june 2022, 51-72 72 vander elst, t., verhoogen, r., servu, m., van den broeck, a., baillien, e., & godderis, l. (2017). not extent of telecommuting, but job characteristics as proximal predictors of work-related well-being. journal of occupational and environmental medicine, 59(10), 180–186. https://doi.org/jom.0000000000001132 vera, m., le blanc, p. m., taris, t. w., & salanova, m. (2014). patterns of engagement: the relationship between efficacy beliefs and task engagement at the individual versus collective level. journal of applied social psychology, 44(2), 133–144. vincent-höper, s., & stein, m. (2019). the role of leaders in designing employees’ work characteristics: validation of the healthand development-promoting leadership behavior questionnaire. frontiers in psychology, 10, 1049. https://doi.org/10.3389/fpsyg.2019.01049 vinkers, c. h., van amelsvoort, t., bisson, j. i., branchi, i., cryan, j. f., domschke, k., ... & van der wee, n. j. (2020). stress resilience during the coronavirus pandemic. european neuropsychopharmacology, 35, 12-16. https://doi.org/10.1016/j.euroneuro.2020.05.003 wang, h.-j., lu, c.-q., & siu, o.-l. (2015). job insecurity and job performance: the moderating role of organizational justice and the mediating role of work engagement. journal of applied psychology, 100(4), 1249–1258. https://doi.org/10.1037/a0038330 wang, c., pan, r., wan, x., tan, y., xu, l., ho, c. s., & ho, r. c. (2020). immediate psychological responses and associated factors during the initial stage of the 2019 coronavirus disease (covid-19) epidemic among the general population in china. international journal of environmental research and public health, 17(5), 1729. https://doi.org/10.3390/ijerph17051729 wilson, m. g., dejoy, d. m., vandenberg, r. j., richardson, h. a., & mcgrath, a. l. (2004). work characteristics and employee health and well-being: test of model of healthy word organization. journal of occupational and organizational psychology, 77, 565–589. https://doi.org/10.1348/0963179042596522 wood, j., oh, j., park, j., & kim, w. (2020). the relationship between work engagement and work–life balance in organizations: a review of the empirical research. human resource development review, 19(3), 240–262. https://doi.org/10.1177/1534484320917560 world health organization (who) (2020). connecting the world to combat coronavirus, 2020. available online: https://www.who.int/campaigns/connecting-the-world-to-combatcoronavirus/healthyathome (accessed on 20 november 2021) xiao, h., zhang, y., kong, d., li, s., & yang, n. (2020). the effects of social support on sleep quality of medical staff treating patients with coronavirus disease 2019 (covid-19) in january and february 2020 in china. medical science monitor: international medical journal of experimental and clinical research, 26, e923549–e923541. https://dx.doi.org/10.12659%2fmsm.923549 yıldırım, m., çiçek, i̇., & şanlı, m. e. (2021). coronavirus stress and covid-19 burnout among healthcare staffs: the mediating role of optimism and social connectedness. current psychology, 1-9. youssef, c. m., & luthans, f. (2007). positive organizational behavior in the workplace: the impact of hope, optimism, and resilience. journal of management, 33(5), 774-800. https://doi.org/10.1177%2f0149206307305562 https://doi.org/jom.0000000000001132 https://doi.org/10.3389/fpsyg.2019.01049 https://doi.org/10.1016/j.euroneuro.2020.05.003 https://psycnet.apa.org/doi/10.1037/a0038330 https://doi.org/10.3390/ijerph17051729 https://doi.org/10.1348/0963179042596522 https://doi.org/10.1177/1534484320917560 https://www.who.int/campaigns/connecting-the-world-to-combat-coronavirus/healthyathome https://www.who.int/campaigns/connecting-the-world-to-combat-coronavirus/healthyathome https://dx.doi.org/10.12659%2fmsm.923549 https://doi.org/10.1177%2f0149206307305562 1. introduction 2. literature review 3. method 4. results 5. discussion 6. conclusions references european journal of government and economics 10(1), june 2021, 80-104 european journal of government and economics issn: 2254-7088 governance and net-import dependency on food and agricultural products in sub-saharan africa: does any causality exist? esther n. mwangia,b*, fuzhong chena, daniel m. njorogec a school of international trade and economics, university of international business and economics, beijing, china b department of business and economics, karatina university, karatina, kenya c institute of food bioresources technology, dedan kimathi university of technology, nyeri, kenya * corresponding author at: esthermwangi2013@gmail.com abstract. though most countries in sub-saharan africa (ssa) are agricultural-based, the region is a net importer of food and agricultural products and experiences the highest level of food insecurity globally. the government have a joint goal of achieving a favourable balance of trade and food security; hence this study examines the causal relationship between quality of governance and net-import dependency on food and agricultural products for 25 ssa countries during the period 1995-2015. principal component analysis is employed to develop a governance index based on the six worldwide governance indicators and a multivariate panel vector error correction framework applied to infer causality in the short and long run. the results reveal that a higher governance index is correlated with a lower net-import dependency ratio and the relationship is statistically significant. evidence of unidirectional causality running from governance to net-import dependency is reported in 14 ssa countries, mainly in the long run. in conclusion, improving governance quality could support reduced food and agricultural net-import dependency through promoting agriculture production, exports and consequently reduced trade deficits in the long run. hence, governance reforms in the region should be placed at the heart of the agricultural development agenda. keywords. agriculture; causality; governance; net-import dependency; sub-saharan jel codes. c59; h11; q17 doi. https://doi.org/10.17979/ejge.2021.10.1.5947 1. introduction global food demand is expected to increase with the increasing world population which is projected to be over 9.1 billion by 2050 with much of the population increase expected to be seen in ssa (parker, 2011; oecd and fao, 2016; eiu, 2018). governments and international organizations are working aggressively by employing policies and innovating ways tailored towards eliminating hunger and malnutrition for all in the context of the sustainable development goal (sdg) agenda 2030. most countries rely on food and agricultural imports to cover their demand deficits and feed their growing population. the number of countries depending on food imports and the rate of import dependency have been growing gradually as population and national income increases. however, most developing countries listed as net food-importing developing countries (nfidcs) by world trade organization (wto) suffer from food trade deficit (wto, 2016) and more often than not find difficulty in covering their food import bills. less developed countries (ldcs) and nfidcs receive wto agriculture mailto:esthermwangi2013@gmail.com https://doi.org/10.17979/ejge.2021.10.1.5947 esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 81 agreement special treatment1, which may be further driving them to fall into the net-importer trap. the 2007-2008 global financial crisis revealed that international trade on food is not always efficient in delivering food security (christiaensen, 2009; gilbert, 2011; murphy, 2015). this view led most economies to re-examine their food supply agenda with much consideration on food self-sufficiency2 and sustainability. food self-sufficiency is now open to debate where some authors hold the view that international agricultural trade promotes food security by stabilizing food supply and food prices (clapp, 2015; fao and ishrat, 2016; clapp, 2016; eiu,2018), while others advocate domestic production to meet local food demand (christiaensen, 2009; murphy, 2015). more importantly, advocates of international agricultural trade are cognizant of the fact that imports dependency makes a country vulnerable to external shocks such as food price spikes, volatility, food riots and export bans imposed by exporting countries (clapp, 2016; martin, 2017; eiu, 2018). moreover, trade deficits have undesirable implication to the overall economic performance of a country. despite ssa being well endowed with fresh water, labour and arable land for agriculture production, the region is a net importer of food and agricultural products, having the highest percentage cereal import dependency of 18.59 percent on average over the years from 2001 to 2013. moreover, it is the most food-insecure region globally and has the highest prevalence of undernourishment which has been increasing from 2014-2017. the percentage of the population suffering from chronic food deprivation was 23.2 in 2017 (fao et al., 2018). apparently, the underlying problem in ssa and africa, in general, is beyond a resource endowment one. besides, the underdevelopment problem and food insecurity in the region is partly due to failures in local policy, institutions, local governance and international governance regimes (world bank, 1989; chhotray and stoker, 2008; ifpri, 2018). moreover, the frequent food and agriculture scandals in african economies pertaining to food supply, food safety and health standards, and embezzlement of funds meant for agriculture and rural development is clear evidence of deficiency in governance. governance failure portrayed by political and economic instability, limited voice and accountability, low government effectiveness, poor regulatory quality, corruption, and the poor rule of law, is among the major challenges facing the implementation of agriculture policies for development agenda (world bank, 2008). according to ifpri (2018), a few african countries have improved their food and nutrition status significantly as a result of their governments’ commitment and reforms for agriculture development and food security. the big question, therefore, is how governance in ssa positions itself in reference to a country’s food and agriculture net-import dependency amidst high levels of food and nutrition insecurity. basically, there is limited literature on the relationship between governance and agricultural trade despite the critical role of government in regulating trade and ensuring food and nutrition security. as a matter of fact, much of the existing literature is qualitative and is characterized by policy responses to major global food crises such as the world food crisis of the early 1970s, the 1 ldcs and nfidcs are eligible to request financial assistant under aid programme to facilitate their food imports see wto (2016). 2 for a broader discussion on food self sufficiency, see clapp j. (2016). esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 82 2007-08 food crisis and 2010-11 price shocks (margulis, 2017). this study examines the causal effect of the quality of governance on food and agricultural products import dependency in ssa. the study is quantitative and focuses on the disaggregated dimensions of governance. moreover, we develop a composite governance index based on the six worldwide governance indicators using the principal component analysis (pca). we compare the effect of each governance indicator and the overall governance index on net-import dependency. consequently, the causality between the composite governance index and import dependency on food and agricultural products is examined. the study contributes to the existing agricultural international trade literature by showing that governance plays a significant role in the food and agricultural import dependency status of a country. the results reveal that a higher governance index is correlated with lower net-import dependency. as such, strengthening corruption control measures, government effectiveness, voice and accountability, and the rule of law significantly reduces net-import dependency on food and agricultural products. hence, increasing governance quality in ssa could support reduced net-import dependency in the long run. these results provide a basis for formulating policies designed to promote international agricultural trade for economic development and food security. the rest of this paper is organized as follows; section two provides a review of literature on agricultural international trade and governance. empirical application and procedure of analysis are explained in section three, while section four presents the results and further discussions on the findings. finally, section five provides the conclusion and recommendations of the study. 2. literature review governance and agricultural trade governance and agricultural trade interact in different pathways that promote the four pillars of food security, namely, food accessibility, availability, utilization and stability. agriculture production facilitates food supply and availability while agricultural trade ensures the distribution of food and global food stability. government has a role in solving market failure and promoting competition which in turn lowers commodity prices making food affordable and accessible to consumers. in addition, through the labor market, households earn income which increases their purchasing power which enables them to diversify their dietary intake and improve their nutritional status. furthermore, trade policies affect government services which in turn impact either positively or negatively on the various food security dimensions and the general performance of the national economy (fao and ishrat, 2016). furthermore, stable governance reduces the uncertainty that hinders investment and pollutes the business environment, thereby promoting international trade. according to martin (2017), trade liberalization reduces poverty rates and improves nutritional outcomes. however, export subsidies in developed countries lead to dumping in developing countries. dumping gradually destroy local agricultural industries which cannot esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 83 compete in the international market, causing more harm to the importing countries. logically, if the money used to cover imports bills were spent within the domestic economy, it would have a multiplier effect, thereby boosting agriculture and economic growth. sadler and magnan (2011) provide an overview of strategies which if adopted by importing governments, could lead to a reduction of risks associated with imports dependency. the government ought to and can create an enabling environment for the smooth functioning of the different pathways and agencies actively involved in food policy to support agriculture for development (gupte and longhurst, 2018). in addition, some studies (fanzo et al., 2014; nisbett and barnett, 2017; kohli et al., 2017) found that a high level of political leadership is an important driver and creates an enabling environment for improved food security and nutrition in a country. however, most national governments in developing countries face difficulty in providing public goods such as the rule of law, civil peace, infrastructure and public research necessary for promoting agricultural productivity (paarlberg, 2002). while on the one hand, governance affects international trade, on the other, a feedback effect exists whereby a country’s governance is shaped by international trade (eichengreen and leblang, 2008). governance at the national level should be redesigned to effectively provide public goods and services in order to promote agricultural productivity, international agricultural trade and accelerate the realization of sustainable development goal number 2 (sdg2). trends on food and agricultural import dependency about 16 percent of the world population depends on international trade to meet their food and agricultural products demand. it is expected that most low-income countries will continue to depend on external land and water resources (fader et al., 2013). trade, whether domestic or international, increases household income and government revenue. by this, it increases diet diversity and promotes the stability of food supply (burnett and murphy, 2014; brooks and matthews, 2015). in 1993, latin america and the caribbean were the leading grain importer in the world with an import dependency of 36.5 percent followed by ssa with 13.6 percent (paarlberg, 2002). africa, asia and the caribbean have been net importers of food and agricultural products on average and have been experiencing trade deficit for all periods between 1990 and 2016. moreover, they are net importers of cereals and pulses, which are the major foods that they depend on for their daily meals. the trade deficits have been increasing over the years, and the figures more than doubled for the period 2010-2016. the increase in trade deficits could partly be explained by the rising import dependency coupled with the rising international food prices. for all these periods, most developed countries had a food and agricultural trade surplus. interestingly, europe has a trade deficit on food and agricultural products for all the periods but, the deficits decreased significantly for the period 2010-2016, contrary to what was happening in many developing countries. these world trade balances on agricultural products, food and selected crops are illustrated in appendix a, table a.1. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 84 according to valdés and foster (2012), the number of developing countries who are both net-agricultural importers and net-food importers increased from 74 in 1999 to 89 in 2009, with an increase from 25 to 31 out of 51 ssa countries. most developing countries are net-importers of staple food grains, and their dependency on imports has been increasing over the recent years (murphy, 2015; eiu, 2017). all african countries depend on food and agricultural products imports though the degree of dependency varies across countries. fao (2012), using data for the period 1960-2007, reported that africa, despite its agricultural potential, has been a net importer of food and of agricultural products. during 2007-2011, 37 african countries were net importers of food while 22 countries were net importers of agricultural raw materials (blein et al., 2013). due to arable land and water constraints, the middle east and north africa (mena) mostly depend on imported food and is the largest grain importer in the world (sadler and magnan, 2011). conceptual framework for the purpose of discussions in this paper, the study adopts the united nations development programme (undp) (1997) definition of governance as; “the exercise of political and administrative authority at all levels to manage a country’s affairs. it comprises the mechanisms, processes and institutions, through which citizens and groups articulate their interests, exercise their legal rights, meet their obligations and mediate their differences”3. this definition consolidates three world bank definitions of governance. that is, “the exercise of political power to manage a nation’s affairs” (world bank, 1989), “the manner in which power is exercised in the management of a country’s economic and social resources” (world bank, 1994), and “the manner in which public officials and institutions acquire and exercise the authority to shape public policy and provide public goods and services” (world bank, 2007). world bank uses six indicators, namely, control of corruption, government effectiveness, political stability and absence of violence, regulatory quality, voice and accountability, and the rule of law to assess the quality of governance (kaufmann et al., 2004). good governance is built on three basic principles, namely, participation and inclusion, accountability and rule of law, and non-discrimination and equality (undp, 2011). based on these principles, good governance is participatory, consensus-oriented, accountable, transparent, responsive, effective and efficient, equitable and inclusive, and respects the rule of law. it takes into account the interests and the most vulnerable in society in decision-making and minimizes corruption. in addition to these components of good governance, organization for economic cooperation and development (oecd) states that good governance is also forward-looking in the sense that it is able to make predictions about future trends and develop policies to deal with anticipated future changes. any compromise on the stated principles and components of good governance results in poor governance and its consequences. good governance is claimed to promote international trade for sustainable economic development (un, 1998; kaufmann and kraay, 3 see more discussion on governance in undp strategy note on governance for human development, 2000. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 85 2002; resnick and birner, 2006). according to knack and keefer (1995), the quality of governance significantly affects the investment rate in an economy. moreover, donor agencies tag their aid disbursement on the quality of a country’s governance (gisselquist, 2012). 3. data and methodology data the study used panel data for 25 ssa countries for the period 1995-2015. the countries were selected based on data availability. since countries are heterogeneous in reference to their levels of development, environmental aspects, and consequently imports demand, they were classified into middle income, low-income countries, oil producers, and non-oil producers based on world bank classification. the countries included in the sample are shown in table a.2. data were mainly collected from world bank, food and agriculture organization (fao), the food and agriculture organization corporate statistical database (faostat) and international monetary fund (imf). net imports dependency ratio (nidr) was computed based on fao definition of import dependency ratio (idr)4. nidr on food and agricultural products was computed using data on total agriculture, which includes cultivation of crops for food and feed, cash crops, livestock production, plus forestry, hunting, and fishing as described by fao. it was calculated as follows: nidr= (imports-exports)*100/(production+imports-exports) (1) the quality of governance was measured using the worldwide governance indicator (wgi) scores on the six governance dimensions, namely voice and accountability, corruption control, government effectiveness, political stability and absence of violence, regulatory quality, and the rule of law (kaufmann et al., 2004; 2010). we developed a composite governance index that comprehensively captures the six dimensions of governance using pca. table 1 presents the description of the variables and the corresponding data sources. the dependent variable is nidr, while the predictor variables of interest in this study are the six governance dimensions and the composite governance index. control variables added in our analysis include agriculture productivity indicators such as agriculture value added per worker and total factor productivity (tfp). other control variables are the exchange rate, inflation, foreign reserves, population, economic growth and natural resource endowment indicators, including fresh water and agricultural land. 4idr is computed as; imports * 100/ (production + imports exports). esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 86 table 1. variable description and data sources. variable code definition data source net-import dependency ratio nidr ratio of net imports to domestic food utilization computed governance index govindex composite governance indicator computed control of corruption corrpcont governance indicator 1 world bank government effectiveness goveff governance indicator 2 world bank political stability and absence of violence polstab governance indicator 3 world bank regulatory quality regqlty governance indicator 4 world bank rule of law ruoflaw governance indicator 5 world bank voice and accountability voiceacc governance indicator 6 world bank agriculture value added per worker agrivapw productivity indicator 1 faostat total factor productivity tfp productivity indicator 2 fao exchange rate exrate real exchange rate imf inflation cpi consumer price index imf foreign currency reserves foretodebt foreign reserves to debt ratio world bank population lnpopu log (population) world bank economic growth gdppc real gdp per capita world bank fresh water endowment lnfreshh20 log(fresh water) faostat agricultural land lnland log(agricultural land) faostat methodology model identification to examine the impact of the quality of governance on food and agricultural import dependency, panel data framework was used due to the flexibility it allows in modelling differences in behaviour across entities and its ability to solve sample selection bias (greene, 2012). panel regression models, namely, pooled ordinary least squares (ols), which assumes homogeneity, fixed effect (fe) and random effects (re), were estimated in order to select the appropriate model for the data. the general modelling framework is expressed as: itiititiitit cxzxy εβεαβ ++=++= ''' (2) where ic capture the heterogeneity or individual effect and itε is the idiosyncratic errors. for the case of pooled regression, iz contains only a constant term. however, if iz is unobserved but correlated with itx then fixed effect would be the appropriate model. finally, if ic is uncorrelated with itx , then random effect would be evident. first, the pooled ols and re were estimated using stata software. breusch-pagan lagrange multiplier (lm) test developed by breusch and pagan (1980) was done to make a decision between a random effects regression and a pooled ols regression. the null hypothesis of variances across entities is zero, h0: σ²µ=0; was tested against the alternative hypothesis that variances across entities is not equal to zero; h1:σ²µ≠0. the lm test statistic is calculated by: esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 87 2 1 1 2 1 2 1 )( )1(2           − − = ∑ ∑ ∑ = = = − n i t t it n i i e et t ntlm (3) the estimated pooled ols and re equations are given in model 4 and 5, respectively. itititit xgnidr εβλα +++= '' (4) where nidr is the net import dependency ratio, g is a vector of governance indicators and x is a vector of control variables. in the case of pooled ols, λ and β include constant terms. itiititit xgnidr εµβλα ++++= '' , (5) where, µi is a group specific random element. consequently, re model and fe model were estimated and hausman specification test devised by hausman (1978) was done in order to make a choice between re model and fe model. under the null hypothesis, the preferred model is re, while for the alternative hypothesis, fe is the preferred model. the hausman test statistic is computed as: )( ∧ −= refebh β ′ )(][ 1 ∧ − −− referefe bvv β (6) where b and ^ β are the coefficient vectors, v is the covariance matrix, fe is the consistent estimator and re is the efficient estimator. the fe models which account for entity fixed effect only and for both entity and time fixed effects are as follows: itiititit xgnidr εγβλ +++= '' (7) ittiititit xgnidr εσγβλ ++++= '' (8) where γi is the country fixed effect and σt is the year fixed effect. cross-section dependence and panel unit root test pesaran (2006) show substantial bias and size distortions in estimates when cross-section dependence in panel data is overlooked. hence we test the data for cross-section dependence using lm test devised by breusch and pagan (1980). the test has a null hypothesis of no esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 88 cross-section dependence 0)cov(:0 =jtith εε for all t and ji ≠ and an alternative hypothesis of cross-section dependence 0)cov(:0 ≠jtith εε . to compute the lm statistic, fixed effect model (7) is first estimated then the lm is calculated as: ∑∑ − = += = 1 1 1 2ˆ n i n ij ijcd tlm ρ 0 2/)1(2 −nnχ (9) where ijρ̂ is the estimate of the pair-wise correlation of the residuals from the estimation of the model (7). given that causality tests require the variables to be stationary, we tested our data for nonstationary using panel unit root tests proposed by maddala and wu (1999), and pesaran (2007). the fisher-type test by maddala and wu (1999) denoted as mw hereafter is a first-generation panel unit root test and assumes cross-section independence across panels. the assumption of cross-section independence is very restrictive; hence we also apply the cross-sectional augmented im, pesaran, and shin (cips) test proposed by pesaran (2007). the cips test is a second-generation panel unit root test that allows for cross-sectional dependency. both tests can be applied in unbalanced panels and series having different lags. the tests are based on the following model: it pi z ztizitiiiti yyy εβρα +∆++=∆ ∑ = −− 1 ,,1., (10) where ∆ is the first difference operator, iρ and iβ are the autoregressive coefficients, t is the time span of the panel, and n is the number of cross-sections. the null hypothesis, 0:0 =ih ρ for all i, all panels contain unit root is tested against the alternative hypothesis; 0:1 <ih ρ for at least one i, some panels are stationary. both tests allow for heterogeneity, since iρ is allowed to vary across the cross-sections. panel cointegration test having established the order of integration and non-stationary of our variables from the unit root tests, the next step is to test whether governance is cointegrated with net-import dependency on food and agricultural products. cointegration implies long-run equilibrium relationship between the two variables. the two widely used panel cointegration tests in economic literature proposed by pedroni (1999, 2004) and westerlund (2005) are used. pedroni (1999, 2004) test is residualbased and allows for heterogeneity in the slopes and intercepts of the cointegrating equation. fixed effects and individual specific deterministic trends in the panels are allowed. westerlund (2005) is based on error correction model hence have a higher power of test compared to the test proposed by pedroni (1999, 2004). it allows for heterogeneity and accounts for crosssection correlation. the null hypothesis in both tests is that of no cointegration among variables in individual panels. the alternative hypothesis in pedroni (1999, 2004) test is that all panels are cointegrated while in westerlund (2005) test, the alternative hypothesis is that some panels are esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 89 cointegrated. granger causality test by definition, if the past values of a stationary variable x can aid in the prediction of another stationary variable ty conditional on having accounted for the effects of past values of y on ty then, x is said to “granger cause” ty (granger, 1969). we established that governance and net-import dependency are integrated of order one, i (1). hence, a vector error correction model (vecm) was used to test whether governance index “granger cause” net-import dependency. two steps procedure of estimating vecm followed by engle and granger (1987) were adopted. the first step involves the estimation of the following model in order to obtain the residuals. titiiiti govnidr ,,1, εγα ++= , (11) where iα is the cross-sections fixed effect. the first lags of the residuals are used as a proxy for vector error correction term, which represents any deviation from long-run equilibrium between governance and import dependency. in the second step, a multivariate panel vecm with dynamic error correction term (ect) is estimated. specifically, the model estimated is as follows: 2,1,2, 1 ,22, 1 ,212, 1,1,1, 1 ,12, 1 ,111, ittikti k k kkti k k kiti ittikti k k kkti k k kiti ectnidrgovgov ectgovnidrnidr µλθθθ µλθθθ ++∆+∆+=∆ ++∆+∆+=∆ −− = − = −− = − = ∑∑ ∑∑ (12) where, niθ for n = 1,2 are country-specific fixed effects and knn,θ are the coefficients corresponding to the kth lag of the endogenous variables. nλ are the coefficients of the error correction terms and nit ,µ are idiosyncratic errors. the vecm is used to test both short-run and long-run causality. to test short-run causality, the significance of the coefficients knn,θ is tested using the standard wald f-test. the null hypothesis 0: ,0 =knnh θ for k=1,..., k is tested against the alternative hypothesis 0: ,0 ≠knnh θ since the focus of this study is on whether governance “granger cause” imports dependency, rejecting the null hypothesis 0: ,120 =kh θ imply that governance “granger cause” import dependency in the short run. for long-run causality, the significance of the coefficient nλ is also tested using the standard wald test. the null hypothesis is 0:0 =λh where 21,.λλλ = and the alternative hypothesis is 0:0 ≠λh . in this study, rejecting the null hypothesis implies governance “granger cause” import dependency in the long run. the coefficients of the ect, nλ represents the speed of adjustment to the long-run equilibrium in case a shock occurs. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 90 4. results and discussion table 2 presents the results of the preliminary analysis in the model specification. the results of breusch-pagan lm for random effect provide strong evidence for rejecting the null hypothesis for no random effect. in other words, strong evidence for significant difference across units and within group’s correlation was found. table 2. summary of preliminary tests results. test statistic p-value lm for randon effects 1509.54 0.000 hausman specification 22.95 0.018 time-fixed effect 1.49 0.079 lmcd 998.267 0.000 notes. the data used for these tests covers the full sample period from 1995 to 2015. the results for hausman test were in favour of fe. therefore, fe model was selected as the consistent model with the data and was used for the analysis. furthermore, the results for the presence of time-fixed effects test provided evidence for not rejecting the null hypothesis that the coefficients for all the years’ dummies are jointly equal to zero. therefore, in this case, no time fixed effects were needed. the test for cross-section dependence gave strong evidence to reject the null hypothesis of no cross-sections dependence. in other words, the residuals across entities are correlated. table 3 summarises the results of the estimated regression using fixed effect model, where the nidr is the dependent variable is presented. at first, we used the six governance indicators as our variables of interest and later added control variables. column 1 of table 3 shows the result of the short model without the control variables. here, the coefficients for corruption control, government effectiveness and, voice and accountability are negative and statistically significant at the 0.01 significance level. the coefficient for the rule of law is also negative and statistically significant at the 0.05 significance level. however, the coefficients for political stability and regulatory quality are positive and are both not significant at the 0.05 and 0.01 significance levels. 84.8% of the variation in the net-import dependency ratio is explained by the six governance indicators. this value of r-squared is considerably high. column 2 of table 3 shows the estimation results when agricultural productivity indicators, namely agricultural value-added per worker and total factor productivity are added to the model. the signs of all the governance indicators remain the same as before. the coefficients for corruption control, government effectiveness, and voice and accountability are significant at the 0.01 level. when exchange rate, inflation and foreign reserves to debt ratio are added to the base model, the signs of the coefficients for the governance indicators remain the same as before. the coefficients for regulatory quality and rule of law are significant at the 0.05 significance level. these results are shown in column 3 of table 3. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 91 table 3. regression results for net-import dependency and governance indicators. furthermore, addition of more control variables into the model, that is; log population, gdp per capita, log fresh water and log agricultural land, does not change the sign and the significance of the coefficients for all the governance indicators. all the control variables have the expected signs. (see column 4 and 5 of table 3). finally, fixed effect robust errors are used to solve possible heteroscedasticity, and the results are illustrated in column 6 of table 3. these results imply that an increase in corruption control is expected to reduce food and agricultural import dependency, holding all other things constant. the same case applies to government effectiveness, voice and accountability, and rule of law. however, an increase in political stability is expected to increase food and agricultural import dependency. similarly, an increase in regulatory quality is expected to increase food and agricultural import dependency, holding all other things constant. corruption undermines agricultural productivity by eroding national and farmers’ income. it creates market failures such as monopoly and enhances selfish arrangements that address personal interest gains (eiu, 2018). campos et al. (1999) and dahlström et al. (2012) showed (1) (2) (3) (4) (5) (6) variables nidr nidr nidr nidr nidr nidr corrpcont -0.416*** -0.436*** -0.438*** -0.361*** -0.357*** -0.357 (0.089) (0.093) (0.097) (0.098) (0.099) (0.266) polstab 0.073 0.059 0.071 0.0822 0.086* 0.0861 (0.047) (0.048) (0.049) (0.052) (0.052) (0.101) goveff -0.293*** -0.306*** -0.313*** -0.325*** -0.325*** -0.325 (0.085) (0.086) (0.086) (0.085) (0.085) (0.196) regqlty 0.131 0.149 0.193** 0.221** 0.219** 0.219 (0.091) (0.092) (0.094) (0.093) (0.094) (0.174) ruoflaw -0.229** -0.212* -0.279** -0.287** -0.289** -0.289 (0.110) (0.111) (0.116) (0.116) (0.117) (0.212) voiceacc -0.264*** -0.263*** -0.242*** -0.241*** -0.242*** -0.242 (0.089) (0.089) (0.091) (0.090) (0.090) (0.155) agrivapw -0.001 -0.002* -0.003** -0.003** -0.003 (0.001) (0.001) (0.001) (0.001) (0.003) tfp 0.017 -0.020 -0.122** -0.121** -0.121 (0.043) (0.049) (0.054) (0.058) (0.089) exrate -0.002 -0.001 -0.001 -0.001 (0.001) (0.001) (0.001) (0.002) cpi 0.0377* -0.046 -0.048* -0.048 (0.020) (0.028) (0.028) (0.041) foretodebt 0.005 0.004 0.004 0.004 (0.004) (0.004) (0.004) (0.004) lnpopu 20.49*** 20.85*** 20.85* (5.826) (7.183) (12.00) gdppc 0.003** 0.003** 0.003 (0.001) (0.001) (0.002) lnfreshh2o -7.258 -7.258 (11.380) (9.234) lnagril -0.302 -0.302 (5.192) (7.936) observations 525 525 525 525 525 525 r-squared 0.848 0.849 0.851 0.856 0.856 0.278 number of country 25 25 25 25 25 25 notes: standard errors in parentheses. ***, **, and * denote statistical significance at 1%, 5% and 10%, respectively. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 92 that reducing corruption encourages productive investment. in addition, corruption diverts public funds meant for boosting agriculture development to personal gains. as such, corruption reduces public spending on agriculture leading to low agricultural output. public expenditure on agriculture in ssa has been stagnating at relatively low levels in comparison to other regions in the world (vos, 2015; fao, 2012). corruption control, therefore, can promote investment in the agriculture sector which in turn would increase domestic agricultural production and supply. this could reduce dependency on imports and also provide a surplus for export. conflict and political instability hamper economic and trade integration (blein et al., 2013). in addition, conflicts and political unrest negatively affect the production and distribution of food and agricultural products, both domestically and internationally. political stability, therefore, creates a good environment for trade, the creation of bilateral agreements and agricultural fdi flows. furthermore, it promotes the distribution from areas of high supply-low demand to areas of high demand-low supply. increases in income further boost consumption demand. according to eiu (2018), political stability is essential for agricultural productivity. since the agriculture sector in africa has been stagnating, much of the increased consumer demand is met by imports. hence, if governments do not support agricultural production to meet local demand and promote exports, import dependency will be inevitable. improving government effectiveness promotes the provision of public goods such as infrastructure, agricultural research and other public expenditures in agriculture which promote agricultural productivity and reduce import dependency. according to paarlberg (2002), east and southeast asia experienced reduced hunger and malnutrition through the provision of public goods such as internal peace, the rule of law, infrastructure, agricultural research and investment by the government. however, in ssa where the governments have not effectively provided these public goods, the citizens continue to suffer from hunger and rural poverty. to reduce import dependency in ssa, the agriculture sector needs to be revived. this calls for a paradigm shift of governance at the local, institutional and national level. therefore, governance reforms should be placed at the heart of the agricultural development agenda. since regulatory quality captures the perceptions of the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development, the positive sign of regulatory quality can be explained by the fact that the few large companies and importing cartels are able to corrupt their operations in order to maximize their gains from imports at the expense of the public interest. therefore, improving regulatory quality without reducing corruption, promotes private sector development and increases import dependency. moreover, corruption threatens the respect for the rule of law and voice and accountability. consequently, deficiency in the rule of law encourages corruption. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 93 table 4. regression results for net-import dependency and the composite governance index. (1) (2) (3) (4) (5) variables nidr nidr nidr nidr nidr lgovindex -2.606*** -2.592*** -1.903*** -1.903 (0.575) (0.618) (0.622) (1.337) agrivapw -0.0003 -0.003* -0.003 (0.001) (0.001) (0.003) tfp 0.003 -0.154** -0.154 (0.055) (0.061) (0.109) exrate 0.0004 0.001 0.001 (0.001) (0.001) (0.001) cpi 0.0105 -0.101*** -0.101* (0.022) (0.030) (0.058) foretodebt 0.004 0.004 0.004 (0.004) (0.004) (0.004) lnpopu 26.24*** 26.24* (6.324) (15.29) gdppc 0.005*** 0.005** (0.001) (0.002) govindex -2.893*** (0.561) observations 525 500 500 500 500 r-squared 0.811 0.816 0.817 0.828 0.102 number of country 25 25 25 25 25 notes: standard errors in parentheses. ***, **, and * denote statistical significance at 1%, 5% and 10%, respectively. to allow a comparison of the effect of the different governance indicators and the composite governance index on net-import dependency, we examined the relationship between the composite governance index and import dependency on food and agricultural products. lag one of governance index is used to control potential endogeneity. the results in table 4 shows that the coefficients for governance index and lag governance index are negative and significant at the 0.01 significance level. in line with the results presented in table 3, 81.1% of the variation in net-import dependency is explained by governance index. adding control variables does not change the sign and significance of the coefficient of lag governance index, but the magnitude decreases, as shown in column 3 and 4 of table 4. robust standard errors are used, and the results are shown in column 5 of table 4. these results imply that governance and lag of governance are negatively related to the netimport dependency. robustness checks in order to gain more insight, in this section, we provide robustness checks on the aforementioned estimation results. the sample was divided into sub-samples with more homogeneous characteristics. at first, the sample was divided into middle income and lowincome countries then into non-oil producing and oil-producing countries. thereafter, the esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 94 analysis was repeated using fe model. since adding more control variables into our base model does not change the sign of our variables of interest, we focus on the results presented in column 1 of table 3. table 5 and 6 present the results for the robustness checks. table 5 shows that in all the sub-samples, the results are robust for most of the six governance indicators coefficients. all the coefficients of governance indicators have similar signs but different magnitude as observed in the main estimation results (column 1 of table 3) except for the case of government effectiveness in the oil producer sub-sample and the case of the rule of law in the middleincome sub-sample. otherwise, the coefficient of government effectiveness is negative in middle income, low income and non-oil producer sub-samples and is significant at the 0.01 significance level. the coefficient for the rule of law is negative in low income, non-oil and oil producer subsamples but is only significant at the 0.01 significance level in low-income subsample. table 5. regression results for governance indicators and net-import dependency in different country groups. (middle income) (low income) (non-oil producer) (oil producer) variables nidr nidr nidr nidr corrpcont -0.765*** -0.061 -0.307** -0.223*** (0.155) (0.116) (0.137) (0.085) polstab 0.036 0.136** 0.131* 0.0891** (0.086) (0.064) (0.076) (0.045) goveff -0.126 -0.392*** -0.679*** 0.189*** (0.160) (0.103) (0.125) (0.068) regqlty 0.434*** 0.372*** 0.319** 0.069 (0.141) (0.121) (0.148) (0.076) ruoflaw 0.333* -0.657*** -0.185 -0.038 (0.191) (0.144) (0.173) (0.102) voiceacc -0.228 -0.431*** -0.657*** -0.221*** (0.142) (0.111) (0.149) (0.066) agrivapw -0.003*** -0.014** -0.004 -0.004*** (0.001) (0.006) (0.006) (0.001) tfp -0.303*** -0.064 -0.147* -0.179*** (0.080) (0.069) (0.079) (0.046) lnpopu 22.91*** 43.19*** 12.75* 29.26*** (8.014) (7.837) (6.827) (3.861) gdppc 0.003*** -0.029*** -0.002 0.003*** (0.001) (0.007) (0.002) (0.001) observations 189 336 315 210 r-squared 0.943 0.764 0.849 0.950 notes: standard errors in parentheses. ***, **, and * denote statistical significance at 1%, 5% and 10%, respectively. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 95 the coefficient of corruption control is negative in all sub-samples and significant at the 0.01 level in middle income and oil producer sub-samples, while it is significant at the 0.05 level in the non-oil producer sub-sample. coefficients for political stability and regulatory quality are positive in all sub-samples. political stability coefficient is significant at the 0.05 level in lowincome countries and oil producers sub-samples, while regulatory quality is significant at the 0.01 level in middle income and low-income sub-samples and is significant at the 0.05 level in the non-oil producer sub-sample. the coefficient of voice and accountability is negative in all the sub-samples and is significant at the 0.01 significance level in low income, non-oil producer and oil producer sub-samples. table 6. regression result for the composite governance index and net-import dependency in different country groups. (middle income ) ( low income) (non-oil producer) (oil producer) variables nidr nidr nidr nidr lgovindex -0.569 -3.733*** -4.646*** -1.401*** (0.801) (0.846) (0.950) (0.451) agrivapw -0.002 -0.008 0.0002 -0.004*** (0.001) (0.007) (0.007) (0.001) tfp -0.196** 0.0536 -0.006 -0.046 (0.085) (0.081) (0.091) (0.046) lnpopu 1.023 -11.35* -8.604*** 8.511*** (2.700) (6.650) (3.010) (1.550) gdppc 0.005*** 0.021 0.001 0.003*** (0.001) (0.014) (0.002) (0.001) observations 180 320 300 200 r-squared 0.928 0.656 0.794 0.942 notes: standard errors in parentheses. ***, **, and * denote statistical significance at 1%, 5% and 10%, respectively. the analysis of the relationship between the composite governance index and import dependency in the sub-samples reveals that the lag of the governance index is negative in all the sub-samples and significant at the 0.01 level in low income, non-oil producing and oilproducing countries. the results presented in table 6 are robust in consideration to the results presented in table 4. panel granger causality after we have established the relationship between the quality of governance and net-imports dependency on food and agricultural products, the next step is to determine the existence or non-existence of causality between the two variables. in this section, we present the results of the panel unit root, cointegration and causality tests. the variables of interest are the nidr and the composite governance index. mw and cips unit root tests were applied to test nonstationary in our data. the results of both panel unit root tests are presented in table 7. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 96 table 7. panel unit root tests results. (a) cips test without trend with trend variable lags z-bar statistic p-value z-bar statistic p-value nidr 0 -4.514 0.000 -3.838 0.000 1 -3.036 0.001 -2.073 0.019 2 -1.018 0.154 0.274 0.610 govindex 0 -0.642 0.261 1.254 0.895 1 -2.057 0.020 -0.533 0.297 2 -0.751 0.226 2.245 0.988 (b) mw test 2χ p-value 2χ p-value nidr 0 108.629 0.000 112.946 0.000 1 77.940 0.007 82.671 0.002 2 46.662 0.608 44.634 0.688 govindex 0 47.15 0.588 49.295 0.502 1 60.814 0.141 80.659 0.004 2 43.459 0.732 44.442 0.695 note: the data used for these tests covers the full sample period from 1995 to 2015. the test for cross-sectional dependence gave strong evidence to reject the null hypothesis of no cross-section dependence (see table 2). however, the results of cips test that allows for cross-section dependency are considered. lag 2 was selected when carrying out the test though lag one is believed to be optimal. the results show that nidr is stationary at level and at the first difference, both with trend and without trend. the governance index is not stationary at level but becomes stationary after the first difference without trend. hence nidr and governance index are integrated of order i (1). the results for the two-panel cointegration tests reveal strong evidence to reject the null hypothesis for no cointegration. table 8 presents the results for pedroni (1999, 2004) residual based test and the westerlund (2005) error corrected panel cointegration test. the test statistics for pedroni test are negative and significant except for the v. the p-values for the test statistics in westerlund test are less than 0.05. therefore, in both tests the null hypothesis for no cointegration is rejected. these results imply that net-import dependency and governance have a long-run relationship. table 8. panel cointegration tests results. pedroni cointegration test westerlund cointegration test statistic panel group statistic value z-value p-value ѵ 0.836 gt -2.194 -2.255 0.012 ρ -3.92*** -2.002** ga -10.982 -3.488 0.000 t-pp -4.726*** -4.699*** pt -12.798 -5.355 0.000 t-adf -1.674* -2.561*** pa -10.044 -6.198 0.000 notes: the data used for these tests covers the full sample period from 1995 to 2015. ***, **, and * denote statistical significance at 1%, 5% and 10%, respectively. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 97 to determine parameter estimates of the long-run cointegrating relationship between governance and net-import dependency, panel dynamic ordinary least squares (dols) were estimated using panel group means. dols involves including lags and leads in order to eliminate potential feedback effects and to control for potential endogeneity and serial correlation in the residuals. lag one and two were included in the model (11). the estimated coefficient of governance index is -0.699 and significant with a t-statistic of -2.729. this implies that a higher governance index is correlated with lower import dependency. for robustness check, and in order to deal with endogeneity problem and ensure consistent estimators are obtained, generalized method of moments (gmm) estimator proposed by arellano and bond (1991), which uses lagged differences of the dependent variable as instruments were estimated. the first lags of the residuals were used as a proxy for vector ect in the estimation of the vecm specified in model (12). the model was estimated by country in order to better infer the existence and directions of the causality. table 9 presents the summary of the coefficients of the lag of the independent variable and the ect. significance of the coefficient of the ect implies rejection of the null hypothesis in determining causality in the long run while the significance of the coefficient of the independent variable determines causality in the short run. the table shows evidence of causality running from governance to net-import dependency in angola, botswana and congo in the short run. evidence of unidirectional causality from governance to net-import dependency, in the long run, is seen in angola, cameroon, guinea, rwanda, zimbabwe, benin, burkina faso, mali, mozambique, niger, senegal and togo. in angola and botswana, there is evidence of unidirectional causality from governance to netimport dependency both in the short run and in the long run. a unidirectional causality running from net-import dependency to governance is evident in the case of guinea and kenya. in addition, net-import dependency granger cause governance in the long run in congo, ghana, kenya, malawi, nigeria, uganda, zambia, zimbabwe and niger. in kenya, there is evidence of causality both in the short run and in the long run. there is a presence of bidirectional causality in the long run between governance and net-import dependency in zimbabwe and niger. out of the 25 ssa countries analyzed, we find unidirectional causality from governance to net-import dependency in 14 countries. wherein, 13 out of the 14 countries, the causality is in the long run. unidirectional causality from net-import dependency to governance is evident in 10 countries in which 9 out of the 10 countries, the causality is in the long run. in conclusion, causality between governance and net-import dependency in these countries is mainly in the long run. good governance creates an agricultural investment climate while poor governance strengthens the grabbing hand leading to misappropriation of funds meant for agriculture investment. this hinders agricultural development and leads to low productivity, which in turn creates a supply deficit as the population and demand for more food increases. the resulting effect is increased imports to meet the local demand upon which the imports grow more, particularly in the case of dumping. if domestic agriculture production is not boosted in order to meet the demand and provide for exports, import dependency is inevitable. esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 98 table 9. causality between governance and net-import dependency results. h0:governance does not grangercause net-import dependency h0: net-import dependency does not granger-cause governance country short-run long-run short-run long-run lgovindex lect lnidr lect angola -10.975*** -1.834*** -0.009 -0.464 botswana 3.385* -0.767*** -0.026 -0.350 chad 0.549 -0.099 0.020 -0.287 cameroon 0.362 -0.436*** -0.024 -0.202 congo 3.370* -0.074 0.042 -1.014*** ethiopia -5.951 -0.380 -0.004 -0.057 guinea 0.602 -0.558** -0.159* -0.002 ghana 0.782 -0.002 0.014 -1.04*** kenya 1.862 -0.137 0.039* -0.328* malawi -0.478 -0.373 0.047 -1.058** nigeria 0.518 0.000 0.036 -0.634* rwanda 1.218 -1.122** 0.375 0.011 sierra leone -0.498 -0.302 0.047 -0.101 south africa 1.319 -0.532 0.016 -0.038 uganda -0.831 -0.268 0.007 -0.405*** zambia -4.027 -0.003 0.003 -0.586* zimbabwe 21.534 -0.265* -0.006 -0.004* central africa re -0.353 -0.314 0.144 -0.244 benin 2.481 -0.727* 0.017 -0.13 burkina faso 2.307 -0.550* -0.019 -0.007 mali -3.333 -0.593** -0.014 -0.057 mozambique 1.253 -2.021*** 0.657 -0.131 niger -0.458 -0.307* -0.096 -0.763*** senegal -3.065 -0.644* 0.170 -0.156 togo -5.311 -0.946* -0.008 -0.062 note: the data used covers the full sample period from 1995 to 2015. ***, **, and * denote statistical significance at 1%, 5% and 10%, respectively. 5. conclusions this research investigated the causal relationships between the quality of governance and netimport dependency on food and agricultural products for 25 ssa countries during the period 1995-2015. the study was motivated by several factors; firstly, despite most ssa countries being agricultural-based and depend on agriculture for their livelihood and economic development, the region has the highest level of food insecurity and undernourishment globally. secondly, the persistent increase in food and agricultural net-imports and consequently increasing food and agricultural trade deficits in the region and thirdly, governance and institution failures have been cited as some of the challenges facing implementation of agricultural development agenda and economic development in the region (world bank, 2008). we developed a governance index based on the six worldwide governance indicators using the principal component analysis. subsequently, a fixed effect model was applied to analyze the relationships. panel dols and gmm were applied to deal with the endogeneity problem. the empirical results reveal that corruption control, government effectiveness, the rule of law and, voice and accountability are negatively related to food and agricultural net-import dependency and are statistically significant. therefore, governments should employ measures to improve these governance dimensions in order to promote agricultural productivity. in addition, political stability and regulatory quality are positively related to food and agricultural import dependency esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 99 in ssa. this could be explained by the fact that political stability creates a conducive environment for investment and trade. moreover, it encourages regional integration and bilateral agreements, thereby promoting international trade in food and agricultural products. given the high levels of corruption in ssa, government regulations favour private sector development at the expense of public interest. this scenario could explain the positive and significant relationship between regulatory quality and net-import dependency. overall, a higher governance index is correlated with a lower net-import dependency ratio. on causality between governance and net-import dependency on food and agricultural products, a multivariate panel vector error correction framework was applied to investigate the long-run and short-run causality. out of the 25 countries studied, evidence of unidirectional causality running from governance to net-import dependency in 14 countries is found. in addition, unidirectional causality running from net-import dependency to governance is evident in 10 countries. we also find evidence of bidirectional causality in 2 countries, i.e. in zimbabwe and niger. the causality between governance and net-import dependency in these countries is mainly in the long run. based on these findings, we conclude that increasing governance quality could support reduced net-import dependency on food and agricultural products through promoting agricultural production, exports and consequently reduced trade deficit in ssa in the long run. governance reforms considerations in the region are pivotal for sustainable agriculture, food security and overall economic development. therefore, to meet the desired results, measures to reduce corruption and strengthen government effectiveness and efficiency should be put in place. moreover, governance needs to be inclusive and democratic so as to factor in the interests of all stakeholders in the food and agriculture sector and, more importantly, those of producers and the vulnerable population. aknowledgements i wish to thank the editor–in-chief and the reviewers for their valuable contribution in this study. references arellano, m., bond, s. (1991). some tests of specification for panel data: monte carlo evidence and an application to employment equations. the review of economic studies 58(2), 277297. https://doi.org/10.2307/2297968 blein, r., bwalya, m., chimatiro, s., faivre-dupaigre, b., kisira, s., leturque, h., wamboyamdjeu, a. (2013). african agriculture transformation and outlook. johannesburg, south africa: nepad. breusch, t., pagan, a. (1980). the lm test and its applications to model specification in econometrics. review of economic studies 47, 239–254. https://doi.org/10.2307/2297111 https://doi.org/10.2307/2297968 https://doi.org/10.2307/2297111 esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 100 brooks, j., matthews, a. (2015). trade dimensions of food security. oecd food, agriculture and fisheries papers 77. paris: oecd. burnett, k., murphy, s. (2014). what place for international trade in food sovereignty? j. peasant stud. 41 (6), 1065–1084. https://doi.org/10.1080/03066150.2013.876995 bush, g.w. (2001). remarks to the national future farmers of america organization. retrieved from http://www.presidency.ucsb.edu/ws/?pid=63838 campos, j., edgardo, d., sanjay p. (1999). the impact of corruption on investment: predictability matters. world development 27, 1059–67. https://doi.org/10.1016/s0305750x(99)00040-6 chhotray, v., stoker, g. (2008). governance theory and practice: a cross-disciplinary approach. new york, us. palgrave macmillan, 256pp. christiaensen, l. (2009). revisiting the global food architecture. lessons from the 2008 crisis. review of business and economics 54, 345-61. clapp, j. (2015). food self-sufficiency and international trade: a false dichotomy? state of agricultural commodity markets in depth. retrieved from http://www.fao.org/3/a-i5222e.pdf clapp, j. (2016). food self-sufficiency: making sense of it, and when it makes sense. food policy 66, 88–96. https://doi.org/10.1016/j.foodpol.2016.12.001 dahlström, c., lapuente v., teorell j. (2012). the merit of meritocratization: politics, bureaucracy, and the institutional deterrents of corruption. political research quarterly 65, 656–68. https://doi.org/10.1177/1065912911408109 department for environment, food and rural affairs. (2008). ensuring the uk’s food security in a changing world (a defra discussion paper). london: defra. department for environment, food and rural affairs. (2010). export restrictions on food over 2007/08: an analysis of their impact and evaluation of policy options (annex 5 of agricultural price spikes 2007/2008: causes and policy implications.). london: defra. retrieved from http://archive.defra.gov.uk/foodfarm/food/pdf/ag-price-annex%203.pdf economist intelligence unit. (2017). global food security index 2017: measuring food security and the impact of resource risks. london: the economist intelligence unit ltd. economist intelligence unit. (2018). the global food security index 2018: building resilience in the face of rising food-security risks. london: the economist intelligence unit ltd. eichengreen b., leblang d., 2008. democracy and globalization, economics and politics 20, 289-334. https://doi.org/10.1111/j.1468-0343.2007.00329.x engle, r. f., granger, c. w. j. (1987). co-integration and error correction: representation, estimation and testing. econometrica 55(2), 251-276. https://doi.org/10.2307/1913236 fader, m., gerten, d., krause, m., lucht, w. & cramer, w. (2013). spatial decoupling of agricultural production and consumption: quantifying dependences of countries on food imports due to domestic land and water constraints. environ. res. lett. 8(1), 014046 (15pp). https://doi.org/10.1088/1748-9326/8/1/014046 fanzo, j., lachat, c., sparling, t., olds, t. (2014). the nutrition sensitivity of agriculture and food policies: a summary of eight country case studies. scn news 40, 19–25. food and agriculture organization & ishrat (2016). how does agricultural trade impact food security? the state of commodity markets 2015-2016 (trade policy brief no.17). rome: fao. food and agriculture organization. (2012). the state of food and agriculture 2012: investing in agriculture for a better future. rome: fao. fao, ifad, unicef, wfp, who. (2018). the state of food security and nutrition in the world 2018. building climate resilience for food security and nutrition. rome: fao. retrieved from www.fao.org/publications gilbert, l. (2011). food reserves in developing countries: trade policy options for improved food security. ictsd programme agricultural trade and sustainable development. (issue paper no. 37). https://doi.org/10.7215/ag_ip_20110923a gisselquist, m. (2012). good governance as a concept, and why this matters for development policy. working paper no. 2012/30. finland: unu-wider. granger, c. w. (1969). investigating causal relations by econometric models and cross-spectral methods. econometrica 37(3), 424-438. https://doi.org/10.2307/1912791 https://doi.org/10.1080/03066150.2013.876995 https://doi.org/10.1016/s0305-750x(99)00040-6 https://doi.org/10.1016/s0305-750x(99)00040-6 https://doi.org/10.1016/j.foodpol.2016.12.001 https://doi.org/10.1177/1065912911408109 http://archive.defra.gov.uk/foodfarm/food/pdf/ag-price-annex%203.pdf https://ideas.repec.org/a/bla/ecopol/v20y2008i3p289-334.html https://ideas.repec.org/s/bla/ecopol.html https://doi.org/10.1111/j.1468-0343.2007.00329.x https://doi.org/10.2307/1913236 https://doi.org/10.1088/1748-9326/8/1/014046 http://www.fao.org/publications https://doi.org/10.7215/ag_ip_20110923a https://doi.org/10.2307/1912791 esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 101 greene, w. (2012). econometric analysis (7th ed.). prentice hall, nj: pearson education. hans, p. (2013). global governance and food security as global public good. center on international cooperation: new york university. hausman, j. a. (1978). specification tests in econometrics. econometrica 46, 1251–1271. https://doi.org/10.2307/1913827 international food policy research institute. (2018). global food policy report., washington, dc: ifpri. https://doi.org/10.2499/9780896292970 kaufmann, d., aart k., massimo m. (2004). governance matters iii: governance indicators for 1996, 1998, 2000, and 2002. world bank economic review 18, 253-287. https://doi.org/10.1093/wber/lhh041 kaufmann, d., aart k. (2002). governance without growth. economia, 3(1). https://doi.org/10.1353/eco.2002.0016 kaufmann, d., kraay, a., mastruzzi, m. (2010). the worldwide governance indicators: methodology and analytical issues. world bank policy research working paper no. 5430. washington, dc: the world bank. knack, s., keefer, p. (1995). institutions and economic performance: cross-country tests using alternative institutional measures. economics & politics 7, 207–27. https://doi.org/10.1111/j.1468-0343.1995.tb00111.x kohli, n., avula, r., van den bold, m., becker, e., nisbett, n., haddad, l., menon, p. (2017). what will it take to accelerate improvements in nutrition outcomes in odisha? learning from the past. global food security 13, 38–48. https://doi.org/10.1016/j.gfs.2017.04.002 maddala, g., wu, s. (1999). a comparative study of unit root tests with panel data and a new simple test. oxford bulletin of economics and statistics 61(1), 631-652. https://doi.org/10.1111/1468-0084.0610s1631 margulis, e. (2017). the global governance of food security. in j. koops & r. biermann (eds.), handbook of inter-organizational relations (pp 503-525). london: palgrave macmillan. https://doi.org/10.1057/978-1-137-36039-7_24 martin, w. (2017). agricultural trade and food security. adbi working paper 664. tokyo: asian development bank institute. retrieved from https://www.adb.org/publications /agricultural-trade-and-food-security murphy, s. (2015). food security and international trade: risk, trust and rules. university of british columbia cfs/rcéa special issue 2(2), 88–96. https://doi.org/10.15353/cfsrcea.v2i2.133 nisbett, n., barnett, i. (2017). explaining the reduction in child under nutrition in the indian state of maharashtra between 2006 and 2012: an analysis of the policy processes. food policy 70, 27–39. https://doi.org/10.1016/j.foodpol.2017.05.006 oecd, fao (2016). oecd-fao agricultural outlook 2016-2025. paris: oecd. http://dx.doi.org/10.1787/agr-outlook-2016-en paarlberg, r. (2002). governance and food security in an age of globalization. food, agriculture, and the environment discussion paper 36. washington, dc: international food policy research institute. parker, j. (2011). the 9 billion-people question: a special report on feeding the world. london: the economist intelligent unit. pedroni, p. (1999). critical values for cointegration tests in heterogeneous panels with multiple regressors. oxford bulletin of economics and statistics 61, 653-670. https://doi.org/10.1111/1468-0084.61.s1.14 pedroni, p. (2004). panel cointegration: asymptotic and finite sample properties of pooled time series tests with an application hypothesis. econometric theory 20, 597-625. https://doi.org/10.1017/s0266466604203073 pesaran, m. h. (2007). a simple panel unit root test in the presence of cross section dependence. journal of applied econometrics 22, 265-312. https://doi.org/10.1002/jae.951 pesaran, m. h. (2006). estimation and inference in large heterogeneous panel with a multifactor error structure. econometrica 74(4), 967–1012. https://doi.org/10.1111/j.1468-0262.2006 .00692.x sadler, m., magnan n. (2011). grain import dependency in the mena region: risk management options. food sec. 3 (suppl 1), s77–s89. https://doi.org/10.1007/s12571-010-0095-y https://doi.org/10.2307/1913827 https://doi.org/10.2499/9780896292970 https://doi.org/10.1093/wber/lhh041 https://doi.org/10.1353/eco.2002.0016 https://doi.org/10.1111/j.1468-0343.1995.tb00111.x https://doi.org/10.1016/j.gfs.2017.04.002 https://doi.org/10.1111/1468-0084.0610s1631 https://doi.org/10.1057/978-1-137-36039-7_24 https://www.adb.org/publications%0b/agricultural-trade-and-food-security https://www.adb.org/publications%0b/agricultural-trade-and-food-security https://doi.org/10.15353/cfs-rcea.v2i2.133 https://doi.org/10.15353/cfs-rcea.v2i2.133 https://doi.org/10.1016/j.foodpol.2017.05.006 http://dx.doi.org/10.1787/agr-outlook-2016-en https://doi.org/10.1111/1468-0084.61.s1.14 https://doi.org/10.1017/s0266466604203073 https://doi.org/10.1002/jae.951 https://doi.org/10.1111/j.1468-0262.2006%0b.00692.x https://doi.org/10.1111/j.1468-0262.2006%0b.00692.x https://doi.org/10.1007/s12571-010-0095-y esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 102 united nations. (1998). annual report of the secretary-general on the work of the organization. retrieved from http://www.un.org/docs/sg/report98/ch2.htm united nations development programme. (1997). governance for sustainable human development. new york, usa: undp. united nations development programme. (2011). towards human resilience: sustaining mdg progress in an age of economic uncertainty. new york, usa: undp. valdés, a., foster, w. (2012). net food-importing developing countries: who they are, and policy options for global price volatility. ictsd programme on agricultural trade and sustainable development issue paper 43. geneva, switzerland. retrieved from www.ictsd.org vos, r. (2015). thought for food: strengthening global governance of food security. united nations committee for development policy (cdp). background paper no. 29. new yolk, usa. retrieved from http://www.un.org/en/development/desa/papers/ westerlund, j. (2005). new simple tests for panel cointegration. econometric reviews 24, 297– 316. https://doi.org/10.1080/07474930500243019 world bank. (1989). sub-saharan africa: from crisis to sustainable growth (a long-term perspective study). washington, dc: world bank. world bank. (1994). governance, the world bank’s experience. washington, dc: world bank. https://doi.org/10.1596/0-8213-2804-2 world bank. (2007). strengthening world bank group engagement on governance and anticorruption. washington, dc: world bank. world trade organization. (2016). the wto agreements series; agriculture, third edition. ch 1211.geneva, switzerland: wto. http://www.un.org/docs/sg/report98/ch2.htm http://www.un.org/en/development/desa/papers/ https://doi.org/10.1080/07474930500243019 https://doi.org/10.1596/0-8213-2804-2 esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 103 appendix a table a.1. world food and agricultural products trade balances 1990-2016 (average value in million us$). source: authors’ calculation using fao data,1990-2016. 1990-1999 2000-2009 2010-2016 import value export value deficits (netimports) import value export value deficits (netimports) import value export value deficits (netimports) agricultural products africa 18507.6 13540.2 4967.3 34887.7 21488.8 13399.0 79558.5 45712.4 33846.1 asia 112195.2 63712.8 48482.3 188843 111521 77322 451983.8 266848.4 185135.4 caribbean 3027.1 2804.1 222 5371 2263.3 3107.7 8953.3 3229.1 5724.2 europe 215245 190612.6 24632.4 342629.6 318530.7 24098.9 560424.6 553711.4 6713.3 north america 43026 66334 -23308 80690.7 97725.2 -17034.5 142997 182606.5 -39609.5 south america 11357.9 30693.6 -19335.8 18066.4 67618.3 -49552 39639.9 152372.2 112732.3 oceania 4055.5 19350.9 -15295.4 8506.4 29982.6 -21476.2 17935.1 54367.1 -36432 food excl. fish africa 14880.6 7861.3 7019.2 28801.4 13699.8 15101.6 66486.7 31300.0 35186.7 asia 74293.5 38120.2 36173.3 132577.5 74614.2 57963.3 326903.7 180188 146715.7 caribbean 2494 2262.8 231.2 4240 1240.3 2999.7 7154.1 1824.3 5329.8 europe 145688.7 133615.4 12073.3 239453.4 220738.4 18715 397819.3 390644.9 7174.5 north america 26573.4 46119.5 -19546.1 51540.1 74639.7 -23099.6 93216 144452.4 -51236.4 south america 8375 18797.3 -10422.2 13548.1 48063.7 -34515.7 29901.4 110558.7 -80657.3 oceania 2776.5 13521.6 -10745.1 5882.3 22445.6 -16563.2 12286.3 43589.3 -31303 cereals, total africa 6160.5 496.4 5664.1 11826.9 764.5 11062.4 24396.4 1587.1 22809.2 asia 18473.8 5815.2 12658.6 26232.3 11100.3 15132 53278.3 22821.3 30457 caribbean 759.4 56.1 703.3 1144.8 31.5 1113.3 1929 55 1874 europe 11987.2 12059.9 -72.7 14248.7 17040.9 -2792.2 24170.3 37916.7 -13746.4 north america 1026.8 15571.6 -14544.8 2017.2 18872.8 -16855.6 3852.7 28865.1 -25012.4 south america 2918.3 2388.2 530.1 4214.7 4939 -724.3 8287.1 14212.7 -5925.6 oceania 240.5 2626.2 -2385.8 438.5 3230.9 -2792.4 788.4 6946.1 -6157.8 pulses, total africa 333.3 70.5 262.9 607 154.1 452.8 1144.2 679.2 465 asia 806.9 857.8 -50.9 2076.7 1513.4 563.3 5944.2 2857.1 3087.1 caribbean 88.3 1.3 87.0 132.7 4.3 128.3 150.1 6.9 143.2 europe 1161.1 624.6 536.4 1033.6 511.2 522.4 1516.5 1042.9 473.7 north america 85.2 627.8 -542.6 231.1 1367.1 -1136 485.1 3440.3 -2955.1 south america 215.3 185.2 30.1 277 230.9 46.1 600.4 499.7 100.8 oceania 17.7 155.1 -137.4 22.6 287.8 -265.2 39.2 991.6 -952.3 coffee, tea, cocoa+ sp, total africa 830.8 4091.5 -4922.3 1194.8 5976.8 -7171.6 2938.5 11512.3 -14450.8 asia 4507.5 5214.8 -707.3 7870.9 8629.7 -758.8 20088.9 21260.8 -1171.9 caribbean 52.2 196.3 -144.1 112 175.8 -63.8 205.3 280.4 -75.1 europe 17200.1 9669.6 7530.5 26284.7 19085.5 7199.2 52198.2 39656.3 12541.9 north america 5422.7 1154.1 4268.6 7953.5 2347.3 5606.2 15985.6 4912.7 11072.9 south america 438.4 4835 -4396.6 720.3 5654.9 -4934.6 1782 12189.5 -10407.4 oceania 432.3 342.7 89.6 789.9 459.8 330.1 1804 732.5 1071.5 fruits and vegetables africa 1076 2337.5 -1261.5 2229.1 4335 -2106 5838.4 11242 -5403.6 asia 12893.8 12273.5 620.2 21455 23091.5 -1636.5 55673.7 54074.9 1598.8 caribbean 306.2 334 -33.7 574.6 331.1 243.5 843.7 601.7 242 europe 43460.1 29873.1 13587 70781.1 51289.2 19491.9 113218 82236.7 30981.3 north america 10640.9 7890.4 2750.5 20378 13958.1 6419.9 37481.5 28364.1 9117.4 south america 1392.4 5395.4 -4003 1818.4 9378.9 -7560.5 4461.3 18768.2 -14306.9 oceania 643.9 1398.3 -754.4 1322.4 2137.3 -814.9 2651.5 4171.2 -1519.7 esther n. mwangi et al. / european journal of government and economics 10(1), june 2021, 80-104 104 table a.2. list of countries by sub-samples. middle income angola, botswana, cameroon, congo, ghana, kenya, nigeria, zambia, south africa low income chad, ethiopia, guinea, malawi, sierra leone, rwanda, uganda, zimbabwe, central africa re, benin, burkina faso, mali, mozambique, niger, senegal, togo non-oil producer botswana, ethiopia, kenya, rwanda, sierra leone, uganda, zambia, zimbabwe, central africa re., benin, burkina faso, mali, mozambique, senegal, togo. oil producer angola, chad, cameroon, congo, guinea, ghana, malawi, nigeria, south africa, niger middle income low income botswana, ethiopia, kenya, rwanda, sierra leone, uganda, zambia, zimbabwe, central africa re., benin, burkina faso, mali, mozambique, senegal, togo. non-oil producer angola, chad, cameroon, congo, guinea, ghana, malawi, nigeria, south africa, niger oil producer ©theauthor(s)2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 1 (2025), pages 5-20 https://doi.org/10.17979/ejge.2025.14.1.10797 submitted: may 29, 2024 accepted: march 27, 2025 published: june 20, 2025 article analysis of industrial transformation and upgrading in resource-exhausted cities: evidence from city-level and enterprise-level panel data yiwen peng 1,* 1 xi'an jiaotong university, china *correspondence: tokio13love@163.com abstract. massive and irrational exploitation of natural resources promoted the industrialization process, which formed different categories of resource-based cities. although these cities have made dramatic contributions to socio-economic development in the last two decades, they are now facing grave problems due to resource depletion. industrial transformation and upgrading are the only ways to cope with this situation. therefore, this study utilizes the city-level and enterprise-level data to analyze the industrial transformation and upgrading of resource-exhausted cities in china. both the macroand micro-empirical results demonstrate that the central policy of the state council has a positive feedback on industrial transformation. then the findings of baseline regression are validated by robustness and heterogeneity tests. what’s more, the policy facilitates industrial upgrading by promoting transition to technologyintensive and high-value chain industries. finally, we propose relevant implications for industrial transformation on the basis of the main findings. keywords: resource-exhausted city; industrial transformation; industrial upgrading; optimization of industrial structure. jel classification: o25, q32, r11 1. introduction china is a vast territory with many cities where resource industry is their most important economy’s mainstay. the type of city is a resource-based city, which played a prominent role in the industrialization process. they contributed to advancing regional economic development and solving the social problem of employment (zhang et al. 2018). however, resource-based cities have a distinct character of the industry life cycle. they develop rapidly in the early stages due to their resource advantages, but eventually, they turn into resource-exhausted cities owing to resource depletion. when the resource advantages no longer exist, those cities struggle to overcome many difficulties of economy and environment (wang et al. 2019). this situation makes a resourceexhausted city the top priority in the regional development. https://creativecommons.org/licenses/by-nc/4.0/ 2 yiwen peng in recent years, a small minority of resource-exhausted cities made some headway in industrial transformation after unceasing exploration. however, many resource-exhausted cities still confront difficult situations, such as insufficient internal impetus for transformation and development, sluggish economy, severe ecological damage, and so on. how to solve these problems is a major concern of the whole country. the stata council issued an official document 1 and announced 69 regions as resource-exhausted cities in 2008, 2009, and 2011. once identified as a resource-exhausted city, the central government will provide general transfer payments to support this city. the official document is intended to support industrial restructuring, upgrading, and sustainable development. on the one hand, the experience of transformation of those cities could provide practical implications for regional development; on the other hand, it could also provide experience for high-quality economic growth of the whole country. it is not only a crucial factor of economic development but also the top priority for a harmonious society. the industrial transformation and upgrading are the only correct solutions to the problems of resource-exhausted cities, which is emphasized in the official guidance document. industrial development is associated with the grand macroeconomic plans of the state economy (lin and zhu 2021). analyzing the industrial development has important implications for exploring the targeted supporting strategies. constant changes in the leading industry have promoted the transformation and optimization of the industrial structure in the process of economic development. the most important task of transformation is to optimize the proportion of various industries and promote the transition from labor-intensive and resource-intensive to technology-intensive industry (wang et al. 2019). there are numerous researchers who concentrate on industrial transformation and upgrading because of their great prospects, generating many valuable results. some researchers have employed metrics such as the share of manufacturing in gdp or the percentage change in clean and polluting industries to measure the level of industrial upgrading (cheng, li, and liu 2018; cole 2000; du, cheng, and yao 2021). others have concentrated on technological progress and its role in upgrading. technological advancement serves as a vital driver of industrial upgrading, directly influencing productivity (ngai & pissarides 2007; raymond et al., 2015; wang et al., 2022). empirical analyses can be categorized based on data sources: one category involves micro-level provincial or city-level data, while the other involves micro-enterprises. after comprehensively studying and fully clarifying the existing literature, we found these two obvious characteristics. (1) the majority of existing studies use city-level data or provincial-level data to explore the current situation and future development trends of resource-exhausted cities. however, as industrialization advances, the development of industrial enterprises should be a better indicator of the national economy. (2) among all studies concerning the transformation of resource-exhausted cities, many scholars have utilized various traditional methods (e.g., solow residual) to measure the industrial development (solow 1957). however, this traditional method is more suitable for macroeconomic analysis. thus, in order to overcome the shortcomings of these two aspects, we make corresponding improvements. (1) first, this paper intends to explore the industrial transformation and upgrading from both macro city-level and micro enterprise-level in resource-exhausted cities. on the one hand, we can observe whether the transformation strategies of micro enterprises are in line with the 1 the state council issued the official document named opinions for promoting the sustainable development of resource-based cities. industrial transformation and upgrading in resource-exhausted cities overall direction of macro industrial structure adjustment by combining macro and micro level data; on the other hand, we can deeply analyze the essence behind macro industrial phenomena by combining macro and micro level data. (2) second, using firm-level data from 2005-2013, this paper applies the total factor productivity calculated by lp method (levinsohn and petrin 2003) to assess the level of industrial transformation. the lp method is well known for its introduction of intermediate input as the proxy variable and therefore would effectively mitigate the data missing problem. for the remainder of this study, we will conduct a comprehensive and detailed analysis. the specific arrangements are as follows: the second section recapitulates related literature. section 3 expounds the empirical model and the data resource. in section 4, we provide a detailed analysis of preliminary regression results. section 5 further validates the baseline regression by robustness and heterogeneous tests. the mechanism analysis of industrial transformation is conducted in section 6. the last section sums up the main conclusions and proposes corresponding implications. 2. literature review and hypothesis 2.1 literature review a lot of scholars conducted extensive research on industrial transformation. in this part, we comprehensively classify and illustrate the existing literature on the development status, future prospects, and transformation of resource-based cities. (1) resource-based city. numerous studies have been conducted on the relationship between resource endowment and economics since the 1990s. in 1994, auty proposed a counterintuitive theory named the resource curse theory. he found that the countries with poorer resource endowments have industrialized more successfully than resource-abundant countries. thereafter, many scholars committed to analyzing the issue of the resource curse. based on the definition of the resource curse, sachs and warner published articles on the resource curse in 1995, 1997, and 1999, which laid an important foundation for subsequent research by many scholars(sachs and warner, 1995; 1997; 1999). cockx and francken (2014) wondered if natural resource dependence would have any impact on a country’s public health expenditures. ultimately, based on the results of countries around the world, they found a significant inverse relationship between resource dependence and public health expenditure. furthermore, they argued that the natural resources in one country should be planned by the government. smith (2015) found a long-run sustained positive impact on gdp per capita after resource exploitation with a fixed-effect empirical model. in addition, this effect exhibits obvious heterogeneous characteristics between oecd and non-oecd treatment countries. also, many researchers have analyzed the major determinants of development in resourcebased cities. according to yu et al. (2008), the development of 78 mining cities in china was found to be related to the geographical location and the natural resource types. li et al. (2013) conducted an analysis and concluded that resource-based cities usually had a slower pace of development than other cities. they argued that the possible cause was the consequence of a planned economy, specifically, an unreasonable taxation system, an unsuitable plan, and an inappropriate policy of 4 yiwen peng resource development. apart from the influencing factors, the serious consequences also involved. ploeg (2010) investigated whether resource has economic and social impacts. he identified a number of negative impacts, including exchange rate appreciation, slower growth prospects, internal conflicts, corruption, and so on. there are also studies on the transformation of resource-based cities abroad. galgóczi, b. (2014) selected the ruhr region in germany as the research object. by combining the research and survey data on germany's economic development, it was found that the green transformation practices in the ruhr region mainly adopted the following measures: establish a specialized regional leading institution to scientifically plan the regional development goals, make efforts to increase financial support and actively introduce advanced technologies, and strengthen the coal industry in terms of intensive transformation to promote optimization and upgrading of traditional industries. houston has accelerated the implementation of its urban transformation strategy (klineberg, 2020). firstly, it has improved the efficiency of resource utilization and intensified efforts to optimize and upgrade traditional industries. then, it has made full use of national policies to vigorously develop the aerospace industry and the medical technology sector. in addition, it has cultivated emerging industries and increased the proportion of the tertiary industry. driven by these measures, houston has broken away from its reliance on resources and achieved the green transformation of a resource-based city. based on the chinese national condition and the official document about resource-exhausted cities, it is crucial to examine the validity of government policy. using empirical methods and data in resource-exhausted cities, zhang et al.(2018) made a profound study on the impact of government incentives on transformation and found that the impact depends on whether the authority of the city is the municipal secretary or mayor. li et al. (2021) explored the policy impact on secondary industries in resource-based cities using data from 282 prefecture-level cities. research results showed that the supportive policy reduced the share of the secondary sector in gross domestic product. moreover, this effect was more noticeable in the central and western cities. sun, lu, and cheng (2020) utilized two indicators of gdp per capita and employment to test the effectiveness of the policy in resource-exhausted cities. some sustainable indicators were utilized in follow-up research, however. yu et al. (2022) explored the impact of a supportive strategy in resourceexhausted cities from a sustainable development perspective. they utilized the energy efficiency indicators from 284 cities in china and found that the policy has improved the energy efficiency and promoted sustainable development in those cities. (2) industrial transformation and upgrading. many scholars conducted extensive research on industrial transformation at the macro provincialor city-level. utilizing city-level data and multiple econometric models, zhu and lin (2022) explored the intrinsic relationship between resource dependence and industrial transformation under the circumstances of the market reform in china. the results showed that dependence on natural resources is a major cause of the industrial structure that is dominated by the secondary sector in the process of modernization in china. li and guan (2022) conducted an in-depth study around the industrial transformation of the state-owned sector, which revealed that state-owned enterprises could provide a boost for value-enhancement. lai et al. (2021) evaluated the industrial transformation performance of 30 chinese provinces and explored the impact of market segmentation on this performance. the empirical regression findings showed an inverted u-shaped correlation between them. liu et al. (2020) chose 10 resourceindustrial transformation and upgrading in resource-exhausted cities abundant localities in shanxi province to comprehensively assess their level of sustainable development, not just the economic transformation level, but the coordination level. they suggested that the uppermost priority of the sustainable transformation is the unemployment problem. also, the government should think highly of the quality of basic education. it might be necessary to introduce the high technology industry and fundamentally raise awareness of sustainable development. meanwhile, studies have also been conducted from the micro-enterprise aspect to analyze the industrial transformation. li and chen (2022) conducted a series of empirical studies using data from 2009 to 2015 to explore whether green credit policy would affect the operation and production of heavily polluting enterprises. ultimately, the findings confirmed that this policy has facilitated the diversification and transformation of heavily polluting firms. luo et al. (2022) studied the industrial development of real enterprises using the tfp of 2064 listed enterprises in china. in summary, it was found that finance technology innovation would facilitate the transformation of these firms, thereby achieving the goal of sustainable development. to verify the actual effect of technological innovation on industrial restructuring and upgrading, xie and teo (2022) selected enterprise data from 35 industrial sectors for empirical analysis. the findings suggested that green technological innovation can effectively promote more efficient and cleaner production, but this finding did not hold for the low-value-added industrial sector and the clean sector. through the analysis of the existing literature, we find two distinct features of the studies on this topic. (1) most existing studies at home and abroad use city-level data or county-level data to analyze the industrial transformation of resource-exhausted cities. (2) among all studies concerning the transformation of resource-based cities, most of them use traditional methods to measure the industrial development. therefore, corresponding improvements are made in this study to address the shortcomings of the above statements. (1) we will comprehensively discuss the effect of industrial transformation on resource-exhausted cities from both the macro city level and the micro enterprise level. (2) using enterprise data from 2005-2013, this research chooses the lp method to measure the industrial transformation and upgrading level. 2.2 research hypothesis referring to all the above, we suggest that this place-based policy of the state council may promote the industrial transformation and upgrading both at the city-level and enterprise-level. what’s more, industrial upgrading has succeeded in promoting the transition to technology-intensive and highvalue chain industries. thus, this study proposes the following hypotheses: hypothesis 1. the implementation of the central policy has a positive impact on the optimization of industrial structure (ois) in resource-exhausted cities. hypothesis 2. the implementation of central policy could promote the total factor productivity of enterprises in resource-exhausted cities. hypothesis 3. the implementation of central policy promotes firms’ tfp by facilitating the transition of industrial production to technology-intensive and high-value chain industries. 6 yiwen peng 3. models and data 3.1 empirical models as an effective tool to assess the experimental effect, the traditional difference-in-difference (did) model is often used in evaluating the intertemporal effect of public policy. one of the most important advantages of did model is that it can obtain the genuine effect of policy by modeling and analyzing the data from quasi-natural experiments. according to the official document of the state council, there are three groups of resource-exhausted cities published in 2008, 2009, and 2011, respectively. thus, this study uses multi-period did models to analyze the effect of policy on industrial transformation. (1) assess the policy impact on industrial transformation at the city level. this paper first examines the effect of policy on industrial structure at the city level. the multi-period did model is specified as follows: 𝑂𝑂𝑂𝑂𝑂𝑂𝑖𝑖𝑖𝑖 = 𝛼𝛼 + 𝛽𝛽 ∙ 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖 + 𝛾𝛾𝑋𝑋𝑖𝑖𝑖𝑖 + 𝜇𝜇𝑖𝑖 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖 [1] where 𝑂𝑂𝑂𝑂𝑂𝑂𝑖𝑖𝑖𝑖 is the optimization of the industrial structure of city i in year t. 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖 represents the core explanatory variable that reflects whether city i is listed as a resource-exhausted city in year t. this interaction term equals to 1 if city i is listed as a resource-exhausted city in year t, otherwise equals to 0. in this model, we are mainly concerned with the magnitude of the coefficient 𝛽𝛽1 and its significance level. 𝜀𝜀𝑖𝑖𝑖𝑖 is the error term. 𝑋𝑋𝑖𝑖𝑖𝑖 represents a set of control variables. drawing from research of mao et al. (2021), yuan and zhu (2018), this study selects fiscal expenditure (fis), gdp per capita (gdp), foreign direct investment (fdi), environmental regulation (er), human capital (hc), marketization (mar), information level (ifm), infrastructure level (ifa) as control variables. 𝜇𝜇𝑖𝑖 and 𝜐𝜐𝑖𝑖 denote individual and time fixed effects. (2) assess the policy impact on industrial transformation from the enterprises level. based on the empirical analysis above, if the government policy promotes the optimization of city’s industrial structure, to a certain extent, it should also promote the transformation of enterprises. to prove this hypothesis, this paper applies the firm data to further explore the impact of the policy on enterprises. details of the analysis model are as follows: 𝑇𝑇𝑇𝑇𝑃𝑃𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼2 + 𝛽𝛽2𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖𝛾𝛾2 + 𝜓𝜓𝑖𝑖 + 𝜂𝜂𝑖𝑖 + 𝜆𝜆𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖 [2] where 𝑇𝑇𝑇𝑇𝑃𝑃𝑖𝑖𝑖𝑖𝑖𝑖 is utilized to evaluate the transformation level of enterprises, which is the total factor productivity of firm i in city l in year t calculated by the lp method. 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 is the core explanatory variable that reflects whether firm i belongs to a resource-exhausted city in year t. when city l where firm i is located is selected as a resource-exhausted city in year t, the interaction term equals to 1, otherwise it equals to 0. in this model, we are mainly concerned with the magnitude and significance of the coefficient 𝛽𝛽2, as it denotes the actual impact of the policy on firms in resourceexhausted cities. 𝑋𝑋𝑖𝑖𝑖𝑖 represents a set of control variables. with reference to (li, yue, and chen 2016; wang and liu 2014; yu, sun, and xuan 2020), this study chooses enterprise scale (es), enterprise age (age), capital density (cd), export share (ex), operating profit margin (pm), financial expenses industrial transformation and upgrading in resource-exhausted cities (fan) and so on as control variables. 𝜓𝜓𝑖𝑖 , 𝜂𝜂𝑖𝑖 and 𝜐𝜐𝑖𝑖 denote location, individual and time fixed effects. 𝜀𝜀𝑖𝑖𝑖𝑖 is the error term. 3.2 variables and data sources to analyze the policy impact on industrial transformation from the macro city level, this paper will use city-level data from 284 cities in china from 2004 to 2016. all data in the city-level regression are collected and processed from the china city statistical yearbook. the dependent variable is the optimization of industrial structure (ois), which is measured by the ratio of the tertiary sector to the secondary sector output value. according to the current situation in china, the process of industrial structure upgrading is mainly about the transition to the tertiary industry. what’s more, the growth rate of the tertiary industry is usually higher than that of the secondary industry. thus, compared with using the output value of secondary or tertiary industries, it is more appropriate to use the ratio of these two indicators to measure the changes in industrial structure. a larger ois value signifies a more rational industrial structure. an increase in ois value indicates an economic shift towards the service industry and an upgrade of the industrial structure. to analyze the impact of policy on industrial transformation from the micro-enterprises level, this paper selects firm-level data from a database conducted by the national bureau of statistics. this database includes all types and natures of industrial enterprises with detailed information about the identity and address, as well as detailed economic information such as indicators of gross output, employment, profits, and taxes. in addition, to prevent biased estimation, we follow the instruction of brandt et al. (2012) to process the sample data with a series of treatments to systematically remove the following enterprises: (1) lost key geographical location information, (2) the number of employees less than 8, (3) the operating revenue does not locate in the interval from 0.5 percentage to 99.5 percentage of the distribution, (4) annual sales revenue less than 20 million rmb. the dependent variable in enterprise-level regression is total factor productivity (tfp) of each enterprise. by combining the previous research literature, we find that the methods of calculating tfp can be classified into three categories. the first category is parametric methods, and the representative ones are stochastic frontier analysis (sfa) and generalized method of moments (gmm). however, the sfa method has to satisfy the assumption about the tfp distribution. this means the final calculation results will be overly dependent on the specific model establishment. meanwhile, the gmm method needs to ensure that the data period is long enough. the second category is the non-parametric methods, and the representative one is the data envelopment analysis method. one of the biggest drawbacks of this method is that it needs the data to be complete, which means the sample cannot have missing values. it is unrealistic in the data collecting process. the third category is semi-parametric methods, the representative ones are the op (olley and pakes, 1996) method and the lp method (levinsohn and petrin, 2003). in this paper, the lp method is used to calculate the tfp of enterprises in china. the main reason for not utilizing the op method is that it requires the investment of enterprises cannot equal to 0. however, not every enterprise makes a certain amount of investment every year. therefore, it would generate a large number of missing 8 yiwen peng values by adopting the op method. to address this drawback, the lp method makes an improvement by using intermediate inputs as a proxy variable. this improvement makes the data more complete and therefore, obtains a more comprehensive tfp value. at present, the lp method is widely applied to measure the tfp of enterprises (peng et al. 2021; tang, liu, and wu 2020; zheng, wu, and nepal 2022). hence, with reference to previous studies, we also adopt the lp method. table 1 presents basic statistical information about the data used in the empirical analysis. the table includes the variable name, number of observations, mean value, and standard deviation. table 1. summary statistics of variables. city-level obs mean std.dev enterpriselevel obs mean std.dev ois 3692 0.83 0.47 y 1727013 8.53 1.33 second 3692 48.94 10.99 k 1727013 6.65 1.70 tertiary 3692 36.86 8.86 l 1727013 5.08 1.03 fis 3692 14.05 1.03 m 1727013 8.34 1.36 gdp 3692 3.43 2.91 tfp 1727013 7.29 1.15 fdi 3392 9.74 1.85 es 1727013 1.99 16.71 er 3692 11.10 0.98 age 1726794 2.01 0.77 hc 3692 1.61 2.17 cd 1727013 2.33 107.85 mar 3663 16.39 1.40 fan 1727013 2.14 27.71 ifm 3673 1.05 3.23 ex 1404504 3.43 62.23 ifa 3584 10.02 7.46 pm 1725475 0.045 0.13 4. analysis of empirical results 4.1 baseline regression results (1) the regression results of the policy impact on industrial transformation at the city level. the column (2) in table 2 reports the impact of the supporting policy with full control variables, the coefficient of 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖 is 0.0645, which is significantly positive at 1% level. this finding denotes that, on average, the resource-exhausted cities have a higher ois of 0.0645 compared to the non-resource-exhausted cities. to be brief, this result confirms hypothesis 1, which also sets the stage for further research of enterprises transformation. (2) the regression results of the policy impact on industrial transformation from the enterprise level. based on the city-level regression results, we conduct further enterprise-level analysis. the enterprise-level results are presented in table 3. in theory, the coefficient with all control variables could be more precisely estimate the policy effect. as we can see, the coefficients of 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖 in column (2) is significantly positive (0.4076) at 1% level, implying the tfp of enterprises in resource-exhausted cities increases by 0.4076 compared to enterprises in non-resource-exhausted cities. therefore, this result not only proves hypothesis 2 but also supports the regression result of city-level analysis. industrial transformation and upgrading in resource-exhausted cities table 2. baseline regression results from macro city level. (1) ois (2) ois 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖 0.0210 (0.0218) 0.0645*** (0.0242) 𝜇𝜇𝑖𝑖 yes yes 𝜐𝜐𝑖𝑖 yes yes 𝑋𝑋𝑖𝑖𝑖𝑖 no yes observations 3692 3295 𝑅𝑅2 0.6912 0.7079 note: the asterisks ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses.2 table 3. baseline regression results from micro enterprise level. (1) tfp (2) tfp 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 0.3894*** (0.0140) 0.4076*** (0.0211) 𝜓𝜓𝑖𝑖 yes yes 𝜂𝜂𝑖𝑖 yes yes 𝜆𝜆𝑖𝑖 yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 no yes observations 1606993 1291280 𝑅𝑅2 0.8732 0.8935 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses. 4.2 parallel trend test the parallel trend assumption is an important precondition for did model in empirical paper. in order to utilize the did model, the target variables in the resource-exhausted cities and nonresource-exhausted cities must satisfy the parallel trend assumption before the policy implementation. we conduct the following regression: 𝑇𝑇𝑇𝑇𝑃𝑃𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼3 + ∑ 𝜃𝜃𝑗𝑗𝐷𝐷𝑖𝑖𝑖𝑖𝑖𝑖 𝑗𝑗 + 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖𝛾𝛾3 + 𝜓𝜓𝑖𝑖 + 𝜂𝜂𝑖𝑖 + 𝜆𝜆𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖5 𝑗𝑗≠0,𝑗𝑗≥−3 [3] where 𝐷𝐷𝑖𝑖𝑖𝑖 𝑗𝑗 is a set of dummy variables. if the city where firm i located is selected as a resourceexhausted city in year t, 𝐷𝐷𝑖𝑖𝑖𝑖 𝑗𝑗 equals to 1, otherwise it equals to 0. the meaning of the remaining variables is the same as equation 2. in this test, we focus on the coefficient 𝜃𝜃𝑗𝑗 , which reflects the difference in tfp between resource-exhausted cities and non-resource-exhausted cities in the year t of the policy implementation. the results of the parallel trend test in figure 1 indicate that the coefficient estimates for each period before the implementation of the support policy are not significant. therefore, there is no significant difference between firms in treatment group and control group before the policy implementation. thus, the research sample passed the parallel trend test. 2 the robustness and heterogeneity tests of city-level baseline regression are not presented in the paper due to the limited space. interested readers could contact the author for further detailed results. 10 yiwen peng figure 1. time trend of the policy implementation (city and enterprises). 5. robustness tests and heterogeneous tests 5.1 robustness tests (1) alternative measurement of tfp as mentioned in section 3.2, there are many other methods to calculate the tfp. to further ensure the robustness of baseline regression, we re-estimate the tfp with ols and fixed effects (fe) methods. regression results in this part are displayed in table 4. the coefficients of 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖 are still significantly positive. this finding demonstrates the robustness of the primary regression results. table 4. tfp calculated by the ols and fe methods. (1) ols (2) ols (3) fe (4) fe 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 0.4159*** (0.0145) 0.4189*** (0.0217) 0.4173*** (0.0146) 0.4193*** (0.0219) 𝜓𝜓𝑖𝑖 yes yes yes yes 𝜂𝜂𝑖𝑖 yes yes yes yes 𝜆𝜆𝑖𝑖 yes yes yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 no yes no yes observations 1606993 1291288 1606993 1291288 𝑅𝑅2 0.8872 0.9058 0.8887 0.9071 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses. (2) dropping the sample in the year of policy implementation in the baseline regression model, the treatment group includes enterprises in three groups of resource-exhausted cities in 2008, 2009, and 2011. in order to obtain a more precise difference between treatment and control groups, we drop the sample in these three years to re-estimate the policy effect. the regression results are presented in column 1 & 2 of table 5. as we can see, the absolute values of the coefficients are a bit smaller than those of baseline regression. this is because, through dropping the samples from the year of policy implementation, we could obtain a more realistic policy effect. still, the re-estimation results confirm the robustness of the primary regression. industrial transformation and upgrading in resource-exhausted cities (3) filtering the regression sample in order to avoid the effect of extreme samples on the regression results, we can truncate the research samples that do not fall in the interval from 1 percent to 95 percent of the distribution. after the truncation process, the estimation results in column 3 & 4 of table 5 are still significantly positive at 1% level. this finding confirms the robustness of the regression results of did model. table 5. results of robustness tests. (1) tfp (2) tfp (3) tfp (4) tfp 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 0.3723*** (0.0142) 0.2321*** (0.0182) 0.3880*** (0.0152) 0.4191*** (0.0233) 𝜓𝜓𝑖𝑖 yes yes yes yes 𝜂𝜂𝑖𝑖 yes yes yes yes 𝜆𝜆𝑖𝑖 yes yes yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 no yes no yes observations 1498570 1209291 1345578 1125914 𝑅𝑅2 0.8429 0.9110 0.8772 0.8967 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses. (4) placebo test in order to avoid some unobservable omitted variables of baseline regression result, we conduct placebo test by randomly selecting different cities 300 times (cai et al. 2016; liu and lu 2015). firstly, 59 cities are randomly selected from the overall group of cities involved in the research as the placebo treatment group. the placebo treatment group here means that these cities are actually not really affected by the actual policy we are concerned with, but in the process of the test, we will conduct subsequent analyses as if they were affected by the policy. this random selection method can simulate a random, placebo-like treatment state, thus providing a basis for subsequent comparative analyses. secondly, for the selected 59 cities as the placebo treatment group, statistical models and methods similar to those used in the benchmark regression analysis are applied to calculate the coefficient of the impact of the policy on tfp under this placebo treatment situation. this coefficient reflects the quantitative manifestation of the virtual policy impact degree under this random setting. thirdly, in order to make the results more general and stable and avoid the contingency that may be brought about by a single random selection, we repeat the entire process of randomly selecting cities as the placebo treatment group and obtaining the corresponding coefficients 300 times. each repetition will yield a different coefficient of the impact of the policy on tfp, and at the same time, the corresponding p-value will also be obtained. finally, the 300 regression coefficients and the corresponding 300 p-values obtained during the 300 repetitions are used to draw a kernel density distribution diagram through professional statistical plotting methods. the kernel density distribution diagram can intuitively show the distribution patterns and central tendencies of these coefficients and p-values. when observing the drawn kernel density distribution diagram, it can be seen that the 300 estimated coefficients present certain distribution rules. most of them are concentrated around 0, and on the whole, they show the characteristics of a normal distribution. this means that under this random, placebo-like placebo treatment situation, most of the obtained coefficients tend to be close to 0, that is, there is no obvious and systematic impact. meanwhile, the p-values corresponding to most of the coefficients show that they are not significant, 12 yiwen peng further indicating that these randomly generated impacts are statistically unreliable and are most likely caused by random fluctuations. then, we compare the preliminary regression coefficient obtained from the actual benchmark regression analysis with the distribution of these 300 coefficients. as can be seen from the dashed line in figure 2, the preliminary regression coefficient is located at the high-end tail position of the distribution of the 300 coefficients. in such a placebo test, being in such an extreme position is a low-probability event. in other words, there is a significant difference between the actual benchmark regression coefficient and the coefficient distribution obtained under the random simulation of the placebo situation. therefore, we have reason to believe that the benchmark regression result is not caused by those unobservable and random omitted variables, but truly reflects the real impact of the studied policy on tfp. figure 2. the placebo test. (5) applying the balanced panel data according to the data cleaning process in section 3.2, we finally obtained unbalanced panel data of 1.73 million observations. if there were some inconsistencies in the imbalanced panel data, the primary estimation would be biased. therefore, this part chooses to apply the balanced panel data for noise reduction and then verify the robustness of the baseline estimation. all of the coefficients in table 6 are still positive at 1% significance level, which suggests that the policy does promote enterprises’ tfp in resource-exhausted cities. table 6. selecting balanced panel data. (1) tfp (2) tfp 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 0.2415*** (0.0138) 0.1982*** (0.0152) 𝜓𝜓𝑖𝑖 yes yes 𝜂𝜂𝑖𝑖 yes yes 𝜆𝜆𝑖𝑖 yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 no yes observations 563635 502160 𝑅𝑅2 0.8589 0.8758 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses. industrial transformation and upgrading in resource-exhausted cities 5.2 heterogeneous tests in the following analysis, we further examine the heterogeneous effect of the central supporting policy on enterprises across regions, industry categories, and ownership types. thereby, we could comprehensively understand the specific influence mechanism of the policy. the first dimension compares the western regions and the non-western regions. as we could see in column 1 and 2 of table 7, the policy effects are all significantly positive. however, the estimated coefficients in non-western regions are larger and more significant than in western regions. according to previous literature, superior geography would create a better business environment, such as stricter intellectual property laws and more sufficient public finances (acemogl et al., 2012; rodrik, subramanian et al., 2004). as a result, producers are more likely to respond to government policy in a relatively superior business environment. our results are consistent with that previous research, which also validates the baseline regression results. the second dimension compares the different industry sectors. the industry sectors include the mining industry, the manufacturing industry, and the electricity, heat, and gas industry. the production of these industries all involves resource consumption. according to columns 3 and 4 of table 7, we can see that all coefficients are still significantly positive. however, the policy has a greater impact on tfp of the manufacturing industry. one possible reason is that the transformation of the manufacturing industry is more easily successful than the mining industry and the electricity, heat, and gas industry. because the production of these two industries is so dependent on natural resources, transformation and upgrading are still hard for them (j. li et al. 2022). the last dimension compares the ownership types: state-owned enterprises, private enterprises, and foreign companies. the impact of the three columns in table 8 is different. the private and foreign enterprises react much more greatly than state-owned enterprises and the coefficients of these two types of enterprise are positive. however, the coefficient of state-owned company is small and negative. institutional differences may be responsible for this regression result (brandt and li, 2003; kornai, 2001). during the production start-up stage, state-owned enterprises usually have easier access to bank loans. this is because state-owned enterprises often have a special status and a close connection with the government, which makes banks and other financial institutions more inclined to them when granting loans. meanwhile, the government often uses financial institutions to support state-owned enterprises, taking financial institutions as a means to enable state-owned enterprises to obtain the financial assistance needed in the production process more conveniently. however, this relatively easy access to funds has brought about some negative impacts. due to the relatively easy access to funds, state-owned enterprises may lack sufficient motivation to achieve efficient production. unlike private enterprises and foreign-funded enterprises, they do not face greater financial pressure and thus do not need to continuously optimize the production process and improve production efficiency to ensure the survival and development of the enterprises. over time, this lack of motivation for efficient production has ultimately led to the failure of industrial transformation and upgrading, making it difficult for state-owned enterprises to achieve ideal results in aspects such as industrial structure adjustment and moving towards industrial links with higher added value. consequently, they have gradually shown a different situation from private enterprises and foreign-funded enterprises in overall market competition and 14 yiwen peng industry development, which also reasonably explains the differences in the regression results presented in table 8 mentioned earlier. in conclusion, the differences in ownership types, through institutional differences and different performances in aspects such as production and credit, jointly result in different impact results and reaction degrees when enterprises face the same research factors. table 7. heterogeneous treatment effect across regions and industries. (1) west regions (2) non-west regions (3) manufacturing industry (4) other industry 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 0.1506** (0.0691) 0.4229*** (0.0221) 0.4299*** (0.0235) 0.1361*** (0.0486) 𝜓𝜓𝑖𝑖 yes yes yes yes 𝜂𝜂𝑖𝑖 yes yes yes yes 𝜆𝜆𝑖𝑖 yes yes yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 yes yes yes yes observations 100544 1190733 1218331 72137 𝑅𝑅2 0.9187 0.8909 0.8916 0.9237 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses. table 8. heterogeneous treatment effect across ownership types. (1) state-owned (2) private (3) foreign 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 -0.0382 (0.0387) 0.4344*** (0.0263) 0.3341** (0.0744) 𝜓𝜓𝑖𝑖 yes yes yes 𝜂𝜂𝑖𝑖 yes yes yes 𝜆𝜆𝑖𝑖 yes yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 yes yes yes observations 74846 691874 198754 𝑅𝑅2 0.9624 0.8958 0.9193 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are presented in parentheses. 6. mechanism analysis the supportive policy in resource-exhausted cities is promulgated to increase employment, eliminate poverty and maintain socio-economic stability. industrial transformation is the only way to achieve these goals. industrial transformation and upgrading can be summarized in three evolutionary procedures. first, transforming from the primary sector to the tertiary sector industry; second, transforming from a resource-intensive and labor-intensive industry to a technologyintensive industry; third, transforming from low-value chain to high-value chain. in this section, we conduct two tests to analyze the transformation mechanism channel from the perspective of average wages. we expect the implementation of the central policy to lead to a shift in industrial production to technology-intensive and high-value chain industries. this transition would be reflected in more high-skilled labor and less low-skilled labor in the market. however, we cannot obtain the data of employment by skill level of enterprises. to solve this problem, we calculate the indicator of average wages (𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖) by dividing the total cost of employment by the total number of employees. here we industrial transformation and upgrading in resource-exhausted cities utilize an important assumption that wages of high-skilled labor are higher than wages of low-skilled labor (s. li, wang, and zhang 2022). in a relatively well-developed labor market, wage levels are often determined by the supply and demand relationship of labor. generally speaking, high-skilled labor forces, with their relatively scarce professional knowledge, complex skills and strong problemsolving abilities, etc., can create higher value for enterprises. therefore, under market competition, enterprises are willing to pay higher wages to this type of labor to attract and retain them. in addition, considering china's national conditions and the availability of data, it is quite appropriate to select the average wage to represent the level of labor skills. the first test is a difference-indifferences (ddd) estimation model. based on lessons from cai et al. (2016), we introduce a mechanism variable and construct an interaction term of 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 × 𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 to validate the mechanism analysis from an average wage perspective. the ddd model specification is: 𝑇𝑇𝑇𝑇𝑃𝑃𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼4 + 𝛽𝛽4𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 × 𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖𝛾𝛾4 + 𝜓𝜓𝑖𝑖 + 𝜂𝜂𝑖𝑖 + 𝜆𝜆𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖 [4] where 𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 is the average wage of each enterprise. all the other variables in equation 4 remain consistent with those in the baseline regression. in this mechanism analysis section, our major concern is the coefficient 𝛽𝛽4. the mechanism regression results are presented in columns (1) and (2) in table 9. the second test is a two-stage mediating effects model. baron and kenny (1986) provided a useful method for us; we constructed the following model to empirically identify the mechanism analysis from an average wage perspective. the two-stage mediating effects model specification is: 𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼5 + 𝛽𝛽5𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖𝛾𝛾5 + 𝜓𝜓𝑖𝑖 + 𝜂𝜂𝑖𝑖 + 𝜆𝜆𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖 [5] 𝑇𝑇𝑇𝑇𝑃𝑃𝑖𝑖𝑖𝑖𝑖𝑖 = 𝛼𝛼6 + 𝛽𝛽6𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 + 𝜏𝜏𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 + 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖𝛾𝛾6 + 𝜓𝜓𝑖𝑖 + 𝜂𝜂𝑖𝑖 + 𝜆𝜆𝑖𝑖 + 𝜀𝜀𝑖𝑖𝑖𝑖𝑖𝑖 [6] where mediating variable remains average wage of enterprises (𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖). all the other variables in equation 5 and 6 remain consistent with those in the baseline regression. in this part, our major concern is 𝛽𝛽5 and 𝛽𝛽6. the results of mediating effects model are presented in columns (3) and (4) in table 9. table 9. results for mechanism analysis. ddd model mediating effect model (1) 𝑻𝑻𝑻𝑻𝑻𝑻𝒊𝒊𝒊𝒊𝒊𝒊 (2) 𝑻𝑻𝑻𝑻𝑻𝑻𝒊𝒊𝒊𝒊𝒊𝒊 (3) 𝑨𝑨𝑨𝑨𝒊𝒊𝒊𝒊𝒊𝒊 (4) 𝑻𝑻𝑻𝑻𝑻𝑻𝒊𝒊𝒊𝒊𝒊𝒊 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 × 𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 0.0328*** (0.0012) 0.0334*** (0.0018) 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 × 𝑃𝑃𝑃𝑃𝑃𝑃𝑇𝑇𝑖𝑖𝑖𝑖𝑖𝑖 0.0796*** (0.0256) 0.3989*** (0.0216) 𝐴𝐴𝑊𝑊𝑖𝑖𝑖𝑖𝑖𝑖 0.1370*** (0.0017) 𝜓𝜓𝑖𝑖 yes yes yes yes 𝜂𝜂𝑖𝑖 yes yes yes yes 𝜆𝜆𝑖𝑖 yes yes yes yes 𝑋𝑋𝑖𝑖𝑖𝑖𝑖𝑖 no yes yes yes observations 1559540 1270221 1270221 1270221 𝑅𝑅2 0.8750 0.8942 0.8125 0.8974 note: ***, **, and * indicate 1%, 5%, and 10% significance levels, respectively. detailed standard errors are 16 yiwen peng presented in parentheses. according to columns (1) and (2) of table 9, the coefficients of the triple difference term are the main areas of concern. it is clear that the coefficients are positive and significant whether the regression involves control variables or not. this result suggests that supportive policies promote an increase in the highly skilled labor force. based on the previous hypothesis, we can infer that central policy promotes the shift of industrial production to technology-intensive and high-value chain industries. what’s more, the coefficients in columns (3) and (4) of table 9 provide further evidence of the mediating effect of this mechanism channel. the findings denote that the central policy has a positive effect on increasing the wages of highly skilled labor. moreover, the central policy promotes enterprises’ tfp by increasing the wages of high-skilled labor. the findings in the mechanism analysis are also consistent with previous studies (larrain 2015). in summary, we may conclude that the policy increases enterprises’ tfp by facilitating the transition of industrial production to technology-intensive and high-value chain industries, as reflected in the increase of average wages. all the findings in this section are compatible with hypothesis 3. 7. conclusion and implications since the 18th cpc national congress, china has regarded ecological civilization as an important content of the construction of a harmonious society. analyzing the development of resourceexhausted cities is not only promoting transformation of these cities, but also offering experiences for transformation of other cities in china. therefore, our research applies did model to examine the effects of transformation and upgrading of resource-exhausted cities from both city-level and enterprise-level. further robustness tests and heterogeneous tests verified the baseline regression results. according to the research above, the main conclusions are summarized as follows: first, the policy has substantially increased the tfp of enterprises located in resource-exhausted cities. specifically, enterprises located in resource-exhausted cities have an increased tfp of 0.4076 compared with enterprises located in non-resource-exhausted cities. this result is also validated through follow-up robustness and heterogeneous tests. second, the impact of the policy is different among three types of enterprises. according to the heterogeneous test in section 5.2, the policy did not favor the state-owned enterprises because they lack the motivation of transformation due to their institutional advantages. third, based on the results of the mechanism analysis, we may conclude that the policy promotes firms’ tfp by facilitating the transition of industrial production to technology-intensive and high-value chain industries. finally, in view of the above-mentioned findings, we wish to draw the following implications. (1) enterprises of different ownership types, different scales, and in different industries have diverse problems and advantages in the process of transformation. therefore, policy formulation needs to fully consider these differences and formulate exclusive support policies for different types of enterprises. for example, policies for state-owned enterprises can be designed to help break institutional barriers, stimulate internal transformation motivation, and provide policy support for industrial transformation and upgrading in resource-exhausted cities private enterprises in aspects such as more convenient financing channels and market access opportunities. (2) the economic development levels, industrial foundations, and resource endowments in different regions vary. for economically developed regions, policies can focus on guiding enterprises to transform towards higher value-added fields such as high-end manufacturing and high-tech industries, and encourage the research and development and innovative application of cutting-edge technologies. in resource-based regions, it is necessary to formulate policies to promote the transformation of traditional resource industries towards green and sustainable industries, deep processing, and related emerging industries around the needs of the transformation of resourceexhausted cities. for example, offer tax incentives and special subsidies to enterprises in resourceexhausted cities to help them break away from their dependence on a single resource and achieve industrial upgrading. (3) the government can set up special transformation funds and provide financial subsidies to enterprises that are actively carrying out transformation according to a certain proportion of their transformation investments, which can be used for aspects such as technological research and development, equipment renewal, and personnel training. meanwhile, by reducing relevant taxes and fees of enterprises, such as lowering enterprise income tax and value-added tax, the cost pressure on enterprises during the transformation period can be alleviated, enabling enterprises to invest more funds into transformation and upgrading. (4) formulate attractive talent introduction policies. for example, provide high-end talents with favorable treatments such as housing subsidies, preferential policies for children's education, and start-up funds for scientific research to attract talents with advanced technologies and management experience to work in transforming enterprises and make up for the shortage of professional talents in the process of enterprise transformation. support enterprises in carrying out cooperation in running schools, targeted training, and other projects with universities and vocational colleges. according to the actual talent needs of enterprise transformation, jointly formulate training programs to cultivate skilled and innovative talents that meet the requirements of enterprise technological upgrading and business expansion. meanwhile, the government can fund free skill training programs for enterprise employees to improve the overall quality of enterprise employees and lay a solid talent foundation for enterprise transformation. there is still much room for further study. this study only covers the short-term data from 2008 to 2013. when long-term data is available, further studies can follow up and analyze the longterm effect of the policy. 18 yiwen peng references acemoglu, d., johnson, s., & robinson, j. a. (2012). the colonial origins of comparative development: an empirical investigation: reply. american economic review, 102(6), 3077–3110. https://doi.org/10.1257/aer.102.6.3077 auty, r. m. (1994). industrial policy reform in six large newly industrializing countries: the resource curse thesis. world development, 22(1), 11–26. https://doi.org/10.1016/0305-750x(94)90165-1 baron, r. m., & kenny, d. a. (1986). the moderator-mediator variable distinction in social psychological research: conceptual, strategic, and statistical considerations. journal of personality and social psychology, 51(6), 1173–1182. https://doi.org/10.1037/0022-3514.51.6.1173 brandt, l., & li, h. (2003). bank discrimination in transition economies: ideology, information, or incentives? journal of comparative economics, 31(3), 387–413. https://doi.org/10.1016/s01475967(03)00080-5 brandt, l., van biesebroeck, j., & zhang, y. (2012). creative accounting or creative destruction? firm-level productivity growth in chinese manufacturing. journal of development economics, 97(2), 339– 351. https://doi.org/10.1016/j.jdeveco.2011.02.002 cai, x., lu, y., wu, m., & yu, l. (2016). does environmental regulation drive away inbound foreign direct investment? evidence from a quasi-natural experiment in china. journal of development economics, 123, 73–85. https://doi.org/10.1016/j.jdeveco.2016.08.003 cheng, z., li, l., & liu, j. (2018). industrial structure, technical progress and carbon intensity in china's provinces. renewable and sustainable energy reviews, 81, 2935–2946. https://doi.org/10.1016/j.rser.2017.06.103 cockx, l., & francken, n. (2014). extending the concept of the resource curse: natural resources and public spending on health. ecological economics, 108, 136–149. https://doi.org/10.1016/j.ecolecon.2014.10.013 cole, m. a. (2000). air pollution and 'dirty' industries: how and why does the composition of manufacturing output change with economic development? environmental and resource economics, 17(1), 109–123. https://doi.org/10.1023/a:1008388221831 du, k., cheng, y., & yao, x. (2021). environmental regulation, green technology innovation, and industrial structure upgrading: the road to the green transformation of chinese cities. energy economics, 98, 105247. https://doi.org/10.1016/j.eneco.2021.105247 galgóczi, b. (2014). the long and winding road from black to green: decades of structural change in the ruhr region. international journal of labour research, 6(2), 217–240. klineberg, s. l. (2020). prophetic city: houston on the cusp of a changing america. new york, ny: simon and schuster. lai, a., yang, z., & cui, l. (2021). market segmentation impact on industrial transformation: evidence for environmental protection in china. journal of cleaner production, 297, 126607. https://doi.org/10.1016/j.jclepro.2021.126607 larraín, m. (2015). capital account opening and wage inequality. the review of financial studies, 28(6), 1555–1587. https://doi.org/10.1093/rfs/hhu088 levinsohn, j., & petrin, a. (2003). estimating production functions using inputs to control for unobservables. the review of economic studies, 70(2), 317–341. https://doi.org/10.1111/1467937x.00246 li, b., yue, y., & chen, t. (2016). export and innovation: empirical evidence from patent data at firm level. the journal of world economy, 39(12), 72–94. li, h., long, r., & chen, h. (2013). economic transition policies in chinese resource-based cities: an overview of government efforts. energy policy, 55, 251–260. https://doi.org/10.1016/j.enpol.2012.12.007 li, j., tian, y., zhang, y., & xie, k. (2022). assessing spatially multistage carbon transfer in the life cycle of energy with a novel multi-flow and multi-node model: a case of china's coal-to-electricity chain. journal of cleaner production, 339, 130699. https://doi.org/10.1016/j.jclepro.2022.130699 li, m., & guan, s. (2022). does china's state-owned sector lead industrial transformation and upgrading? journal of cleaner production, 338, 130412. https://doi.org/10.1016/j.jclepro.2022.130412 li, q., liu, s., yang, m., & xu, f. (2021). the effects of china's sustainable development policy for resourcebased cities on local industrial transformation. resources policy, 71, 101940. https://doi.org/10.1016/j.resourpol.2020.101940 li, r., & chen, y. (2022). the influence of a green credit policy on the transformation and upgrading of heavily polluting enterprises: a diversification perspective. economic analysis and policy, 74, 539–552. https://doi.org/10.1016/j.eap.2022.03.009 https://doi.org/10.1257/aer.102.6.3077 https://doi.org/10.1016/0305-750x(94)90165-1 https://doi.org/10.1037/0022-3514.51.6.1173 https://doi.org/10.1016/s0147-5967(03)00080-5 https://doi.org/10.1016/s0147-5967(03)00080-5 https://doi.org/10.1016/j.jdeveco.2011.02.002 https://doi.org/10.1016/j.jdeveco.2016.08.003 https://doi.org/10.1016/j.rser.2017.06.103 https://doi.org/10.1016/j.ecolecon.2014.10.013 https://doi.org/10.1023/a:1008388221831 https://doi.org/10.1016/j.eneco.2021.105247 https://doi.org/10.1016/j.jclepro.2021.126607 https://doi.org/10.1093/rfs/hhu088 https://doi.org/10.1111/1467-937x.00246 https://doi.org/10.1111/1467-937x.00246 https://doi.org/10.1016/j.enpol.2012.12.007 https://doi.org/10.1016/j.jclepro.2022.130699 https://doi.org/10.1016/j.jclepro.2022.130412 https://doi.org/10.1016/j.resourpol.2020.101940 https://doi.org/10.1016/j.eap.2022.03.009 industrial transformation and upgrading in resource-exhausted cities li, s., wang, m., & zhang, d. (2022). total factor productivity growth at the firm-level: the effects of capital account liberalization. journal of international economics, 139, 103676. https://doi.org/10.1016/j.jinteco.2022.103676 lin, b., & zhu, r. (2021). energy efficiency of the mining sector in china: what are the main influence factors? resources, conservation & recycling, 167, 105321. https://doi.org/10.1016/j.resconrec.2020.105321 liu, b., wang, j., jing, z., & tang, q. (2020). measurement of sustainable transformation capability of resource-based cities based on fuzzy membership function: a case study of shanxi province, china. resources policy, 68, 101739. https://doi.org/10.1016/j.resourpol.2020.101739 liu, q., & lu, y. (2015). firm investment and exporting: evidence from china's value-added tax reform. journal of international economics, 97, 392–403. https://doi.org/10.1016/j.jinteco.2015.07.003 luo, s., sun, y., yang, f., & zhou, g. (2022). does fintech innovation promote enterprise transformation? evidence from china. technology in society, 68, 101821. https://doi.org/10.1016/j.techsoc.2021.101821 mao, w., wang, w., sun, h., yao, p., wang, x., & luo, d. (2021). urban industrial transformation patterns under natural resource dependence: a rule mining technique. energy policy, 156, 112383. https://doi.org/10.1016/j.enpol.2021.112383 ngai, l. r., & pissarides, c. a. (2007). structural change in a multisector model of growth. american economic review, 97(1), 429–443. https://doi.org/10.1257/aer.97.1.429 olley, g. s., & pakes, a. (1996). the dynamics of productivity in the telecommunications equipment industry. econometrica, 64(6), 1263–1297. https://doi.org/10.2307/2171831 peng, j., xie, r., ma, c., & fu, y. (2021). market-based environmental regulation and total factor productivity: evidence from chinese enterprises. economic modelling, 95, 394–407. https://doi.org/10.1016/j.econmod.2020.03.006 raymond, w., mairesse, j., mohnen, p., & palm, f. (2015). dynamic models of r&d, innovation and productivity: panel data evidence for dutch and french manufacturing. european economic review, 78, 285–306. https://doi.org/10.1016/j.euroecorev.2015.06.002 rodrik, d., subramanian, a., & trebbi, f. (2004). institutions rule: the primacy of institutions over geography and integration in economic development. journal of economic growth, 9(2), 131–165. https://doi.org/10.1023/b:joeg.0000031425.72248.85 sachs, j. d., & warner, a. m. (1995). natural resource abundance and economic growth. national bureau of economic research. https://doi.org/10.3386/w5398 sachs, j. d., & warner, a. m. (1997). sources of slow growth in african economies. journal of african economies, 6, 335–376. https://doi.org/10.1093/oxfordjournals.jae.a020932 sachs, j. d., & warner, a. m. (1999). the big push, natural resource booms and growth. journal of development economics, 59(1), 43–76. https://doi.org/10.1016/s0304-3878(99)00005-x smith, b. (2015). the resource curse exorcised: evidence from a panel of countries. journal of development economics, 116, 57–73. https://doi.org/10.1016/j.jdeveco.2015.04.001 solow, r. m. (1957). technical change and the aggregate production function. the review of economics and statistics, 39(3), 312–320. https://doi.org/10.2307/1926047 sun, t., lu, y., & cheng, l. (2020). implementation effect of resource-exhausted cities' supporting policies, long-term mechanism and industrial upgrading. china industrial economics, 07, 98–116. tang, h., liu, j., & wu, j. (2020). the impact of command-and-control environmental regulation on enterprise total factor productivity: a quasi-natural experiment based on china's 'two control zone' policy. journal of cleaner production, 254, 120011. https://doi.org/10.1016/j.jclepro.2020.120011 van der ploeg, f. (2010). challenges and opportunities for resource rich economies. cepr discussion papers, 5688. https://cepr.org/publications/dp5688 wang, j., & liu, b. (2014). environmental regulation and enterprises' tfp—an empirical analysis based on china's industrial enterprises data. china industrial economics, 03, 44–56. wang, k., wu, m., sun, y., shi, x., sun, a., & zhang, p. (2019). resource abundance, industrial structure, and regional carbon emissions efficiency in china. resources policy, 60, 203–214. https://doi.org/10.1016/j.resourpol.2019.01.001 wang, x., zhang, t., nathwani, j., yang, f., & shao, q. (2022). environmental regulation, technology innovation, and low carbon development: revisiting the ekc hypothesis, porter hypothesis, and jevons' paradox in china's iron & steel industry. technological forecasting and social change, 176, 121471. https://doi.org/10.1016/j.techfore.2022.121471 xie, r., & teo, t. s. h. (2022). green technology innovation, environmental externality, and the cleaner upgrading of industrial structure in china: considering the moderating effect of environmental regulation. technological forecasting and social change, 184, 122020. https://doi.org/10.1016/j.techfore.2022.122020 https://doi.org/10.1016/j.jinteco.2022.103676 https://doi.org/10.1016/j.resconrec.2020.105321 https://doi.org/10.1016/j.resourpol.2020.101739 https://doi.org/10.1016/j.jinteco.2015.07.003 https://doi.org/10.1016/j.techsoc.2021.101821 https://doi.org/10.1016/j.enpol.2021.112383 https://doi.org/10.1257/aer.97.1.429 https://doi.org/10.2307/2171831 https://doi.org/10.1016/j.econmod.2020.03.006 https://doi.org/10.1016/j.euroecorev.2015.06.002 https://doi.org/10.1023/b:joeg.0000031425.72248.85 https://doi.org/10.3386/w5398 https://doi.org/10.1093/oxfordjournals.jae.a020932 https://doi.org/10.1016/s0304-3878(99)00005-x https://doi.org/10.1016/j.jdeveco.2015.04.001 https://doi.org/10.2307/1926047 https://doi.org/10.1016/j.jclepro.2020.120011 https://doi.org/10.1016/j.resourpol.2019.01.001 https://doi.org/10.1016/j.techfore.2022.121471 https://doi.org/10.1016/j.techfore.2022.122020 20 yiwen peng yu, j., zhang, z., & zhou, y. (2008). the sustainability of china’s major mining cities. resources policy, 33, 12–22. https://doi.org/10.1016/j.resourpol.2007.10.001 yu, w., peng, y., & yao, x. (2022). the effects of china's supporting policy for resource-exhausted cities on local energy efficiency: an empirical study based on 284 cities in china. energy economics, 112, 106165. https://doi.org/10.1016/j.eneco.2022.106165 yu, y., sun, p., & xuan, y. (2020). do constraints on local governments' environmental targets affect industrial transformation and upgrading? economic research journal, 55(08), 57–72. yuan, h., & zhu, c. (2018). do national high-tech zones promote the transformation and upgrading of china's industrial structure. china industrial economics, 08, 60–77. zhang, h., xiong, l., li, l., & zhang, s. (2018). political incentives, transformation efficiency and resourceexhausted cities. journal of cleaner production, 196, 1418–1428. https://doi.org/10.1016/j.jclepro.2018.06.093 zheng, x., wu, c., & nepal, r. (2022). did the administrative approval reform in china affect the productivity of energy firms? a quasi-natural experimental approach. energy economics, 105, 105564. https://doi.org/10.1016/j.eneco.2021.105564 zhu, j., & lin, b. (2022). resource dependence, market-oriented reform, and industrial transformation: empirical evidence from chinese cities. resources policy, 78, 102914. https://doi.org/10.1016/j.resourpol.2022.102914 https://doi.org/10.1016/j.resourpol.2007.10.001 https://doi.org/10.1016/j.eneco.2022.106165 https://doi.org/10.1016/j.jclepro.2018.06.093 https://doi.org/10.1016/j.eneco.2021.105564 https://doi.org/10.1016/j.resourpol.2022.102914 1. introduction 2. literature review and hypothesis 3. models and data 4. analysis of empirical results 5. robustness tests and heterogeneous tests 6. mechanism analysis 7. conclusion and implications references european journal of government and economics 10(1), june 2021, 65-79 european journal of government and economics issn: 2254-7088 the financialisation of the fishing industry: galicia as a case study federico martin palmero, fernando gonzález laxe a* a department of economics, universidade da coruña, spain * corresponding author at: f.g.martin@udc.es abstract. successive technical and organizational innovations have modelled the current structure and composition of the fishing industry. the market structure varies considerably between countries and the fish species they catch. there is a generalised global convergence towards an industrial-outsourced model, in which companies occupy a central and basic position in economic strategies. two predominant key trends are identified: financialisation and the monopolisation of fishing areas. this paper begins with an analysis of the fishing sector from the perspective of the presence of investment funds that hold stakes in companies specialising in a principal region of europe. keywords. fisheries sector; financialisation; investment funds; business concentration. doi. https://doi.org/10.17979/ejge.2021.10.1.7444 1. economic globalisation and the framework of financialisation we live in a world ruled by the financial sector. the expansion of this process is interpreted as a new dynamic that covers changes in the behaviour and values of the economy, politics and society, renewing, as stated by chesnais (1994), old paradigms. financialisation coincides with the greater role assumed by financial capital within economic globalisation and commercial liberalism, giving rise to a new regime of accumulation (lapavitsas, 2011). in our opinion, financialisation has a degree of independence within globalisation, depending on a set of elements, systems and technical objectives, as well as laws and regulations that each country can develop and encourage. under these parameters, economic stakeholders and institutional agents—such as investors, pension funds, investment funds, financial companies, and insurance companies, among others—take on an increasing role in determining local and regional policies, as each country establishes its own state-finance nexus. thus, it is easy to allow for many geographical variants, depending on agreements and alliances within each territory (pike & pollard, 2010). financialisation is a new manifestation of current phenomena. it can be defined as the growing importance of the financial sector in economic operations. this is mainly due to the increased liberalisation of markets and the growth of information and technology. it is, therefore, associated with surges in international capital flows. it thereby means two things: a change in the strategic orientation of the company, given that the profit margins are significant in both the financial and productive spheres; and the management highlighting a new business dynamic that is responsible for the companies operating both as financial agents and production units. https://doi.org/10.17979/ejge.2021.10.1.7444 martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 66 the fishing industry, like the food and agriculture sector, has quickly transformed following recent globalisation trends. they have all adapted depending on their capacity to extract and produce, to increase their marketing offers, and to adjust to new demands. under these circumstances, the fishing industry has not been excluded or alienated from the dynamics of the financialisation of the world economy. a large part of the sector turns to the financial market for financing, above all, on the stock markets, and, as a result, investment funds demand both high and secure returns in the short term. such growth has been in parallel to a process of “ocean grabbing”, as the sea is a strong and attractive asset for investors, who wish to secure their investments through increased possibilities of obtaining fishing access rights (ducastel & anseeuw, 2011). this study explains the changes experienced by the fishing industry from the perspective of the acquisition of shares from financial companies with the aim of obtaining very significant returns, and under regulatory policies that favour and provide incentives for property rights and access to certain fishing areas. secondly, a specific area has been studied. in this case, the most important fishing region in europe, where, in the last decade, the most internationalised and specialised companies have experienced a consistent increased presence of investment funds and financial capital. 2. the effects of financialisation on the fishing industry: sustainability, privatise or perish? in recent years, there have been growing concerns about fishing resource management, its control and access conditions. insofar as fishing resources are commercialised, fishing rights can be purchased, rented or disposed of in specially created markets. in addition, they can be regulated by the national fisheries legislation itself, either for certain areas and/or for certain species. thus, it is possible for a small or significant group of people or companies to accumulate a substantial number of fishing rights for a certain species or obtain privileged access to a fishing area, by virtue of having the option of concentrating fishing rights in one company. it can also occur that this concentration of fishing rights results in a productive specialisation around a limited number of ports, highlighting the concentration processes in the fishing industry. without a doubt, the fishing industry is an income generator and, at the same time, a means of livelihood. according to the fao (2017), between fishermen and fish farmers, the industry encompasses 56 million people, and it is multiplied threefold if we consider upstream and downstream activities (processing, marketing, manufacturers of fishing gear and equipment, shipbuilding). fishery and aquaculture products represent approximately 17% of animal-protein intake amongst people worldwide. likewise, fishing activity is at the centre of important economic and geostrategic interests. not only due to being considered a key activity for the most traded food products in the world (40% of production is sold in international markets compared to only 5% of rice, or 20% of wheat), but furthermore its estimated value (146,000 million dollars) is more than the total income from coffee, bananas, cocoa, tea, sugar and tobacco (pnud, 2016). therefore, the continuous rise in fish consumption makes the fishing sector an attractive activity martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 67 for industry and finance. the industrialisation of fishing has driven a globalised market for such products. the technical innovations of canning and transport processes have made it possible to maintain the quality of the products, reduce distances between supply and demand, and make them more accessible to consumers, both physically and by addressing the effects of seasonality. on the other hand, the multiplying of free trade agreements and the proliferation of fisheries agreements between countries has accelerated the dynamics of integration of most of the supply markets, reducing and eliminating barriers to the free import-export of fishery products and investment flows themselves. the continuous pressure on resources highlights the current situation. conflicts are multiplying around the exploitation, management and distribution of resources, on the one hand, and, on the other, the existence of territorial inequalities based on the different characteristics of the fishing segments (noël, 2013). this asymmetry is reflected in the world trade of fisheries products and their consumption levels. regarding trade, a considerable proportion of total fishing production (35%) is exported, which means a high degree of accessibility and strong integration of the fisheries sector in world trade. in this respect, developed countries still dominate imports of fishing products, although their proportion has reduced slightly (71% of world imports in 2016, compared to 80% in 2006 and 84% in 1996). these imports condition the market, as they encourage many producers to produce, process and export. developing countries, for their part, continue supplying the markets of advanced countries, and exports represent a very significant part of their incomes. foreign sales of fishing products by developing countries rose to 70,000 million dollars in 2016. the continuous improvements in distribution systems have encouraged regional trade by expanding the range of species and forms of products that reflect differences in tastes and consumer preferences. the species most traded globally are prawns (15% of exports in value terms), salmon (14%), white fish (such as hake and cod, which represent 10%) and tuna (9%). regarding consumption, a growing proportion of the use of fisheries products towards human consumption has been recorded, playing, therefore, a fundamental role in global nutrition and food safety. the apparent global fish consumption per capita has grown to 20.4 kg/person/year, primarily due to population growth, increased income, urban development and more efficient distribution channels. in a context of greater international competition, it is easy to deduce that the investors will be attracted depending on comparative advantages arising from access conditions, fishing rights, capacity to acquire and dispose of fishing quotas, and rules for the granting of fishing rights (barbersgaard, 2018). in this regard, many countries have not hesitated to include the possibility of transferring said access rights in their regulations (mansfield, 2004, barbesgaard, 2018)—a point that justifies some researchers suggesting trends towards ocean grabbing, towards the privatisation of fisheries resources and water territories (carothers & chambers, 2012); or predicting the triumph of neoliberalism in the sector (bennet et al., 2015). the new trends encourage investment funds in fishing activities. according to recent studies (österblom et al., 2015), it is noted that 13 transnational companies control between 11% and martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 68 16% of total global fishing catches (that is, between 9 and 13 million tonnes) and between 19% and 40% of the most important fishing stocks in international markets in commercial terms. such companies control all the segments of production, benefiting, similarly, from subventions and exercising a decisive influence on international decisions in fishing and aquaculture. in this sense, investment funds take on all the risks, and experience shows that takeovers are easier than mergers. furthermore, fisheries investments are focused on attracting industrial companies and the emergence of venture capital funds. reality shows, similarly, that the use of individual transferable fishing quotas has contributed (in chile, for example) to four companies controlling 90% of the quotas; or in iceland, to ten companies holding 50% of the fishing rights. this means that for fishing, there is a need for a trade law that obviously generates concentration. in this context, companies pool fishing rights; namely, they seek to define a production model that spans everything from the selection of the species to be caught to the terms to use. the conclusions are that: a) production focuses on a specific number of species; b) trade is concentrated on a few species with international impact and acceptance, and c) the prices of these species record high fluctuations. on the other hand, we can also observe the establishment of conservation areas or marine protected areas in which access is limited; or the establishment of trends for the privatisation of coastal and marine areas in the interest of other users (hydroelectric and thermal power projects, residential projects or the promotion of new urban guidelines) (gonzález-laxe, 2008). 3. discussion on the dynamics of financialisation most of the fishing industry has been integrated into what we could call the extractivist model. it is characterised by aiming to reduce control and management by the institutions, searching, through this, to divest the productive activities of the current regulatory and management frameworks. due to this, some authors (cóccaro et al., 2000) state that we are faced with a concept similar to that of “fishery mining”; where the fishing catches resemble a pattern aimed at capturing a substantial proportion of the income or surplus generated by natural resource extraction in the light of the international commodity price cycle. that is, a pattern of accumulation, in which the state plays the role of guarantor and where the company (whether national or international) is responsible for management. this model can have three direct effects: promote a dismantling of social cohesion and generate significant changes in cultural and identity patterns; contribute to the growth of productive individualism, and lead to greater and more intense competition between its members, which in its day was called the olympic race to fish. the theoretical framework of this new pattern is therefore based on the following considerations. fishing uses technically renewable biomass, so it is necessary to redefine the criteria of economic rationality threefold: as far as natural goods are concerned; as far as territories are involved; and as far as profitability is affected. in this regard, fishing, being an extractive and random activity, requires an approach based on area (the sea), on work processes martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 69 (fishing methods and gear), on the workforce (availability and work conditions), and on legitimisation (regulation and enforcement by the state). the new global financial context has encouraged many changes in the reality of the fishing industry and its behaviour. it does so both at the institutional level, through new international conventions and agreements, and also at the corporate level, through new forms of organisation and financing. its greatest impact involves the modification of the classic paradigms of fishing exploitation, highlighting the above-mentioned process of ocean grabbing by certain fleets, managed by various forms and structures of ownership (carothers & chambers, 2012; joufrray et al., 2019). as fishing resources have the status of collective goods, the fisherman’s interest is oriented towards maximising his individual interests; that is, above any collective behaviour, an attitude that in most cases could lead to suicidal actions. hence, since 1989, fisheries economists have basically introduced the systems of property rights aimed at guaranteeing the respect of optimal or adequate use of fishing resources (neher et al, 1989). the individual transferable quotas (itq) applied by some countries are considered to be subsidised monetary securities (paper-fish) or a source of capital that allows a company's value to expand in a stock market. it is what arnason (2008) defined stating that “itq are living capital”; to the extent that they can be used as a basis for increasing financial capital and thus contribute to other areas of economic activity. arnason himself warned of the dual role of the itq: guarantee for loans and large-scale privatisation. and in parallel, he omitted other consequences and effects such as those arising from an accumulation of debts and fishing rights together with a transfer of uncertainty to the financial institutions that granted the above-mentioned loans (gylfason el al, 2009). the commercialisation of fishing rights affected all the fisheries and fuelled a dynamic of concentration of ownership in a few hands. these circumstances have led to the establishment of what is known as fishing neoliberalism (pinkerton & davis, 2015). there is also a growing commitment to further liberalisation of markets, encouraging foreign investment and allowing sales and consumption decisions to remain in the hands of free capital and people or companies who hold it (mansfield, 2004). at times, a growing privatisation of the seas is recommended as a formula to prevent gaspillage (overfishing), or undeclared and unregulated illegal fishing (appleby, 2013), putting into context the defence of human rights and greater democratic control in other situations (franco et al., 2015). the changes that have taken place have led to several transformations at the same time: a) the work process is diversified, both in its technical dimension (infrastructures, fishing gear, detection techniques) and in its social dimension (references to employment contracts, working conditions, migrant workers); b) there are changes in production relations (regarding organisation, representativeness and governance); c) the decision-making process is modified (multi-level management, and bottom-up/top-down strategies); and d) there is an increase in private investment and concentration of ownership in the sector, replacing the traditional process of social cooperative domain. in this regard, a distinction is made between high added value fisheries, strongly focused on fleets that are very specialised in terms of production and technologically martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 70 advanced, compared to small-scale fisheries or coastal fishing (gonzález-laxe, et al., 2018). the sea is becoming progressively deregulated and property rights start to be the cornerstone of the new international fisheries framework. the access, use, handling and availability of fishery resources depend on these circumstances. in addition, they are considered decisive factors in the functioning of economies, as they provide the basic set of incentives that encourage or discourage economic activity (north, 1990). in the fishery sector, property rights have four essential characteristics: universality (they should belong to somebody and it facilitates long-term planning); exclusivity (that is, the capacity to exclude others from their use and consumption); transferability (possibility of trade and profit guarantee); and security (legal protection by the state). their use has served as an excuse for “incentives must be created to prevent the overexploitation of the fishery resources”; and, on this issue, property rights broaden the perspective of public policy analysis through new regulations. this scenario confirms the statement: “powered by capital and its desire for profit, the current wave of enclosures targeting the world’s fisheries and ocean and inland water resources is taking place within the same context as global land grabbing” (tni, 2014). the common denominator in the situation is: a) the exclusion of small-scale fishermen to fishing and market access; b) the emergence of new legal frameworks that undermine the position of small-scale fishing systems and producers; and c) the strengthening of the positions of the business agents and other powerful economic stakeholders. thus, the mechanisms applied are limited to withholding or relinquishing the legal right to fish or capture resources due to changes in legal frameworks; to depriving small-scale fishermen of their customary fishing rights through the privatisation of their coastal and marine protected areas through concessions with private companies; and to facing declining catches due to overfishing, destruction of fishing grounds and the new status of marine habitats. the rights-based-fisheries management systems are based on the logic of the economic system; that is, capital accumulation is progressively linked to increasing access to resources (allison et al., 2012). 4. business trends financing trends have been significantly altered. the direct repercussions have been: a) changes in ownership structures and, directly, in the field of productive investment decisions and dividend distribution. it affects the shareholder composition. it is therefore necessary to analyse the shareholder composition and the strategies of the groups expecting profits and/or dividends from listed companies. trends are aimed at obtaining capital liquidity both to be able to enter and exit the capital of the group/company, and, at the same time, retain the investment and therefore maintain control of the firm; b) encouraging actions on a short-term basis or within the framework of the current circumstantial paradigm. they aim to stop hostile threats, limiting the intent of the competitors regarding stock options or the drop in prices of specialised products; martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 71 c) stimulating the processes of offshoring and consolidation of higher value-added activities. that is, they are focused on marketing or product-design innovations, creating market niches to achieve a monopoly regarding suppliers and an oligopoly in terms of customers; and d) promoting the transformation of many productive agents into financial rentiers, seeking havens in the financial sphere (krippner, 2005). consequently, the financing processes do not show a standard model, but rather several modalities can be observed depending on the different organisations and ownership structures. by way of example, we show four types: first, those that are of a family or local nature, corresponding to the small-scale or local fisheries that apply for loans for their investments or to modernise their fishing structures or activity. secondly, new corporate structures, corresponding to those operating in national fishing grounds or long-distance fishing. the integration models in the financial sphere focus on loans, on risk capital, or in the placement of private bonds. thirdly, the consolidation of large corporations, normally linked to an industrialised or internationalised fishing, specialising in commodities, which mainly rely on bonds, public share tenders or specific placements. and finally, the processes of company mergers, consolidation of parent companies or expansion trends through an increase in subsidiaries. if the 100 largest fishing companies are analysed, just over half (53%) are private, and we are witnessing intensive merger and acquisition operations (seaman et al., 2017). likewise, several processes are achieved: a trend towards production diversification, based on new species and market niches; and a process of setting up subsidiaries to serve different geographies and markets. the challenges of both approaches are related to the supply chains and the consolidation of the so-called seafood business. jouffray et al., (2019) stress the role of shareholders in view of the successive capital increases in order to protect themselves financially, giving prevalence to the positioning of the markets and their internationalisation. in this regard, the following effects stand out: a) the presence of international corporations is highlighted (through the creation of joints-ventures and ute [temporary business associations] with local capital); b) the trends of the foreignisation of companies is intensified, on the basis of the possibility of transferability and marketability of resources, via experimental exploitation quotas or licence transfers; c) trade liberalisation trends are strengthened, and trade openness and economic deregulation are increasing, mainly with regard to international trade rules and foreign direct investment; and, finally, d) a new financial and income logic is emerging, associated with the privatisation and foreign exploitation of fisheries resources, contributing to the development of an extractive export circle. the current changes concerning productive supply patterns contribute to accelerating those processes closely related to natural resources, insofar as internal factors (those linked to martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 72 technological changes and shifting demand) are well controlled by the sector itself. the elements of change are centred on three points. first, in the changes that concern the international fishery system (definition of exclusive economic zones; regionalisation of fisheries, protected marine areas; property rights). secondly, in those concerning global food and consumption patterns (products, preparations, expenditure and demand). and thirdly, in the new role of the ngos and their influence on productive sectors. 5. the case of galicia galicia is the most important fishing region in europe. this is where the largest concentration of industrial fishing and processing and distribution industries in spain is located (see table1). it stands out for its extensive technological developments and historical know-how. it excels for its strategic importance; for extensive biodiversity of coastal ecosystems, which support the development of various extractive activities; for its multi-species fisheries, operated by some 4,000 vessels that alternate fishing gear and methods depending to access conditions and fishing periods; and for the different fishing strategies, which combine both diversification and flexibility in their activities. table 1. the galician fishing industry in figures (2018). category indicator galicia spain (%) eu (%) fishing fleet vessels (no.) 4,400 49.0 5.3 tonnage (grt) 137,830 41.6 8.7 power (kw) 276,418 35.5 3.4 production fishing (t) 222,515 24.5 4.2 aquaculture (t) 289,772 92.0 21.1 processing industry production (€ million) 2,820 60.0 14.7 consumption per capita (kg/inhabitant) 32.0 25.5 24.5 fishing employment awu 35,373 46.1 9.7 foreign trade fishing exports (€ million) 1,472 46.3 imports (€ million) 2,283 37.0 revenue added value of the sector / gdp 1.90 0.19 0.05 source: based on official data from the european commission; ministry of agriculture, fisheries and food; ine; consellería do mar and ige. the last few years have been prosperous regarding the entry of funds and investment managers into the sector, acquiring and absorbing certain companies. focus has been placed mainly on large companies (leaders or the top group of their segments) looking for a prominent and significant sectoral presence in terms of supply niches and product ranges. it can be said that, in the galician fishing industry, distressed debts had no bearing on the purchase of businesses in danger of liquidation and at knock-down prices, but private equity has had a greater martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 73 and more constant presence in companies with high growth potential and as alternatives to traditional financing. similarly, also notable is the emergence of management funds, acting as administrators of companies owned by second or third-generation family members who have hesitated between selling them or seeking a dividend. therefore, the presence of financial capital cannot be conceived as a one-off, but instead as a structural commitment in a sector and in a specific territorial space where the turnover is over three billion euros (see table 3). it responds, thus, to a financial option that combines characteristics linked to territorial proximity and profitability parameters. table 2. presence of foreign companies, investment funds and financial entities in galician fishing companies. management fund / company purchased company production area and year portobello (spain) iberconsa frozen fish (2015) bolton group (italy) conservas garavilla canning (2015) platinum equity (usa) iberconsa-portobello frozen fish (2019) stator management (spain) paquito (60%) boiled mussels (2018) banks and funds pescanova frozen fish (2013) scandia food (romania) thenaise provote canning (2018) kerry group (ireland) hasenosa manufacturing of fish-mixes and sauces (2018) shanghai kaichuang ocean resources (bright food group) (china) hijos de carlos albo canning (2016) ged capital (usa) purchases 71% of discefa octopus distribution (2016) sherpa capital (spain) isidro1952. (proposed takeover) sushi processing plant and six fish farms producing trout, turbot and sea bream (2019) abanca (80%) controls the company. nueva pescanova frozen fish. bankruptcy proceedings (2013); successive purchases between funds and banks (2019) profand caladero frozen and processed fish (2019) source: developed by authors. we differentiate between four different modalities with respect to financing options. first, those that, with local ties, group together family businesses, and their production is oriented towards national markets until they have very significant market shares. we are referring to specialised industrial fisheries in the commodities of hake (merluccius merluccius) and tuna (thunnus alalunga, thunnus albacares, thunnus obesus, katsuwonus pelamis), such as absa and portoceleiro, or family canning companies, with facilities outside spain and with production oriented towards international markets (jealsa). the second group of companies, based locally, is more oriented towards international markets, developing a strong global expansion, on the structure of a local financial institution (nueva pescanova). a third group comprises companies targeting national and international markets, but with majority participation of equity funds (iberconsa). finally, a fourth example is a canning company with facilities and presence in several markets, involving cross-capital participation with companies in the same sector (grupo calvo). martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 74 chart 1. diagram of galician fishing companies according to their levels of ownership and scale of operations. in short, the flow of galician fishing activity is carried out under a process that ranges from family-owned extractive companies with local ties to companies more integrated into supply chains and with very close links to so-called international commodity products, thus forming part of the financial assets of large marine-industry-finance complexes. below, we quantify the degree of financialisation of the selected companies. we define them as the result of subtracting the ratio of gross fixed capital formation to gross added value from the operating result / gross added value ratio. its meaning is as follows. if tf>0, i.e. positive values, it means a high rate of uninvested profits in the institution's own gfcf. they would be distributed under the financial benefits formula. it is the traditional payment method for shareholders and company owners. subsequently, the greater the value of the degree of financialisation, the higher the profit-taking at the expense of fixed capital investments. in contrast, if tf<0; i.e., negative values, the interpretation can be twofold depending on the rates that make up the indicator. that is, it could be due to losses in the year; and/or at the same time, the need to continue to invest in fixed capital to maintain business activity in the medium and long term in subsequent periods. martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 75 table 3. degree of financialisation (figures in thousands of euros). nueva pescanova canning albo iberconsa portoceleiro absa burela canning calvo canning jealsa a net turnover (nt) 1,008,201 93,008 343,747 77,677 69,479 582,656 639,404 b purchasing (procurement) 588,448 39,943 118,530 70,920 67,183 354,401 452,286 gva (a b) 419,753 53,065 225,217 6,757 2,296 228,255 187,118 operating profit 45,791 5,793 54,975 -1,160 353 25,889 32,410 c non-current assets for 2018 371,343 2,133 183,954 37,696 3,962 115,038 283,987 d non-current assets for 2017 370,730 1,806 110,942 42,813 4,337 116,272 269,613 gfcf (c d) 613 327 73,012 -5,117 -375 -1,234 14,374 degree of financialisation 10.76 10.30 -8.01 58.56 31.69 11.88 9.64 source: developed by authors using the accounts presented by the companies in the trade registry. they correspond to 2018. a first overview of the accounts arising from the use of the operating results reveals very enlightening conclusions. first, low operating results are particularly noteworthy in the freshhaulage sector (porto-celeiro and absa-burela). an analysis of the operating account shows that in 2018 divestments were carried out (burela, -375 thousand euros and celeiro, -5,117 thousand euros). such trends are a response to the situation of uncertainty arising both from the allocation of quotas for fishing in community waters and from the prospects raised by brexit, variables that condition the expectations of this fleet. secondly, in reference to the industrial fishing industry (nueva pescanova and iberconsa) the operating results are very positive (nueva pescanova, +45 million euros; and iberconsa, +54 million euros). however, a different position is observed in investments. iberconsa reinvests profits (73 million euros) while nueva pescanova, involved in the final process of constituting the nucleus of its shareholders, records no reinvestment of profits. finally, both companies report high added value and depend on the financial markets. the third hypothesis is related to the canning sector (conservas albo, calvo and jealsa). their accounts reflect a disparity arising from the time lags incurred by these companies within the sector’s extensive cycle characterised by phases of adjustment and of mergers and acquisitions in order to reach an adequate size so as to be able to deal with an offer other than that of the distribution circuits and respond to the business fluctuations of the sector. thus, all three record very positive numbers in their operating accounts (jealsa, 32.4 million euros; calvo, 25.8 million euros; and albo, 5.7 million euros). similarly, two of them hardly reinvest profits (calvo and albo), while the third (jealsa) does. 6. conclusions in recent years, a transition towards a hybrid food model has been detected. namely, on the one hand, there is a generalisation of the outsourced agro-industrial model, within a context of greater liberal capitalism; and on the other, the consolidation of alternative models based on grassroots relations and close circuits is witnessed, under the hypothesis of strengthening policies in the interest of sustainable local development. the first scenario described highlights a group of large companies—linked to supply sectors, food industries, and logistics and distribution companies— martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 76 that ensure the bulk of the extracted production and marketing of food. the resulting market is awash with powerful publicity campaigns and an intense lobby by multinationals, where crosscutting issues in terms of quality standards, consumer information and taxation, for example, show high levels of control in terms of global food governance. in the second model, there is a predominance of medium-sized companies with limitations in regard to their capacities; either in terms of capital or markets. they have difficulties accessing traditional financial markets and adopt classic legal status (such as public limited companies). there is a clear predominance of family businesses, smaller in size, and with technologies adapted to small-production formats. companies are progressively taking control of fishing resources. most of them are increasingly related to production chains. the main fishing commodities are the first to be affected by such trends. the salmon industry is no exception to this trend, nor are the white-fish, farmed-shrimp or tuna sectors. such production concentration enables the big companies to define their production models based on the selection of species and the techniques used. there is no doubt, as claimed by olivier de schutter (2014), that this industrial fishing system shapes and stimulates the growing demand for certain fishery products, both in advanced and in less-developed countries. hence, when considering an allocation of fishing quotas (access and fishing rights), the dichotomy of privatise or perish is raised. in a nutshell, the trajectories are being conditioned by investment, by productive adaptation in terms of resources, by expanding non-seasonal markets and by managing surplus production. galicia is a good showcase for this dual behaviour, which also shows certain variations, depending on the different levels of specialisation and presence in the markets. we consider four models based on a division according to ownership levels and positioning levels. references allison, e., ratner, b., asgard, b., willman, r., pomery, r., kurien, j. (2012). rights-based fisheries governance: from fishing rights to human rights. fish and fisheries, 13(1), 14-29. https://doi.org/10.1111/j.1467-2979.2011.00405.x appleby, t. (2013). privatising fishing rights: the way to a fisheries wonderland? public law, 58, 481-497. arnason, r. (2008). iceland´s itq system creates new wealth. the electronic journal of sustainable development, 1(2), ejsd.org. barbersgaard, m. (2018). blue growth: saviour or ocean grabbing? the journal of peasant studies, 45(1), 130-149. https://doi.org/10.1080/03066150.2017.1377186 bennett, n.j., govan, h., satterfield, t. (2015). ocean grabbing. marine policy, 57, 61-68. https://doi.org/10.1016/j.marpol.2015.03.026 carothers, c., chambers, c. (2012). fisheries privatization and the remaking of fishery systems. environment and society, 3(1), 39-59. https://doi.org/10.3167/ares.2012.030104 chesnais, f. (1994). la mondialisation du capital. paris. ed. syros. cóccaro, j., le bail, j., gómez, o., benito, a. (2000). la minería pesquera argentina. mar de plata. minutes from the ii humboldt international conference. de schutter, o. (2014). the transformative potential of the right to food. report of the special rapporteur on the right to food. un general assembly. a/hrc/25/57. https://doi.org/10.1111/j.1467-2979.2011.00405.x https://doi.org/10.1080/03066150.2017.1377186 https://doi.org/10.1016/j.marpol.2015.03.026 https://doi.org/10.3167/ares.2012.030104 martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 77 ducastel, a., anseeuw, w. (2011). la production grabbing et la transnationalisation de l´agriculture (sud)-africaine. transnationales, 10/11. https://doi.org/10.4000/ transcontinentales.1080 fao/cofi (2017). report on fish trade-related activities. fao sub-committee on fish trade. cofi: ft/xvi/2017/2. june 2017. franco, j., monsalve, s. borrás, s (2015). democratic land control and human rights. current opinion in environmental sustainability, 15, 66-71. https://doi.org/10.1016/j.cosust.2015 .08.010 gylfanson, p. (2009). nordics in global crisis: vulnerability and resilience. the research institute of the finnish economy. etla. helsinki. gonzález-laxe, f. (2008). territorialisation process in fisheries management. ocean and coastal management, 51(1), 259-271. https://doi.org/10.1016/j.ocecoaman.2007.07.001 gonzález-laxe, f., martin-palmero, f, martin-bermúdez, f., novo. i. (2018). governance of the fishery industry: a new global context. ocean and coastal management, 50, 73-82. https://doi.org/10.1016/j.ocecoaman.2017.12.009 jouffray, j.b., crona, b., wassénius, bebbington, j., scholtens, b. (2019). leverage points in the financial sector for seafood sustainability. science advances, 5, eaax3324. https://doi.org/10.1126/sciadv.aax3324 krippner, g.r. (2005). the financialisation of the american economy. socio-economic review, 3, 173-208. https://doi.org/10.1093/ser/mwi008 lapavitsas. c. (2011). theorizing financialization. work, employment and society, 25(4), 611636. https://doi.org/10.1177/0950017011419708 mansfield, b. (2004). neoliberalism in the oceans: rationalization, property rights and commons question. geoforum, 35, 313-326. https://doi.org/10.1016/j.geoforum.2003.05.002 neher, p. a., arnason, r., mollet, n. (1989). rights based fishing. kluwer academic pub. dordrecht. https://doi.org/10.1007/978-94-009-2372-0 noël, j. (2013). la mondialisation des activités halieutiques: breve analyse géo-historique. m@ppemonde 109-1, 1-26. north, d. (1990). institutions, institutional change and economic performance. cambridge university press. https://doi.org/10.1017/cbo9780511808678 österblom, h., jouffray, j-b., folke, c., crona, b., troell, m., merrie, a., rockström, j. tsikliras, a. (2015). transnational corporations as ´keystone actors´. marine ecosystems. plos one, 10(5), 1-15. https://doi.org/10.1371/journal.pone.0127533 pike, a., pollard, j. (2010). economic geographies of financialization. economic geography, 86, 29-51. https://doi.org/10.1111/j.1944-8287.2009.01057.x pinkerton, e., davis, r. (2015). neoliberalism and the politics of enclosure in north american small-scale fisheries. marine policy, 61, 303-312. https://doi.org/10.1016/j.marpol. 2015.03.025 pnud (2016). trade and environment review 2016: fish trade. unctad-ditc-ted. united nations. seaman, t., tallaksen, e., ramsden, n. (2017). seafood´s top delmakers (undercurrent news, 2017). tni (2014). the global ocean grab: a primer. agrarian justice programme, masifundise and afrika kontakt. https://doi.org/10.4000/transcontinentales.1080 https://doi.org/10.4000/transcontinentales.1080 https://doi.org/10.1016/j.cosust.2015.08.010 https://doi.org/10.1016/j.cosust.2015.08.010 https://doi.org/10.1016/j.ocecoaman.2007.07.001 https://doi.org/10.1016/j.ocecoaman.2017.12.009 https://doi.org/10.1126/sciadv.aax3324 https://doi.org/10.1093/ser/mwi008 https://doi.org/10.1177/0950017011419708 https://doi.org/10.1016/j.geoforum.2003.05.002 https://doi.org/10.1007/978-94-009-2372-0 https://doi.org/10.1017/cbo9780511808678 https://doi.org/10.1371/journal.pone.0127533 https://doi.org/10.1111/j.1944-8287.2009.01057.x https://doi.org/10.1016/j.marpol.2015.03.025 https://doi.org/10.1016/j.marpol.2015.03.025 martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 78 annex absa. founded in 1997 by a group of ship-owners from the port of burela (north of lugo). annual average sales amount to 24,500 tonnes for a sum of 80.5 million euros. the group manages 48 boats (of which 28 vessels operate in waters of grand-sole and 20 in the cantabrian-northwest fishing grounds). the main target species are fresh hake (11,000 tonnes and 52 million euros in 2018) and albacore (1,600 tonnes and 6.2 million euros in 2018). porto-celeiro. group of companies located in celeiro (north of lugo). the group is formed by nine companies. it covers 30% of all fishing capacity for hake from community waters. in 2018, 14,132 tonnes were marketed for a sum of 49.3 million euros, of which 9,856 tonnes were fresh hake with a value of 43 million euros. they own 50 boats, of which 20 involve bottom-set longlines, and 19 vessels fish in the cantabrian-northwest fishing grounds. grupo nueva pescanova. spain's largest fishing group both in turnover and in marketed fish products. founded in 2015 as the result of a corporate restructuring process. currently, the financial institution abanca owns the largest share capital (80.4%). it is present in 80 countries, with a turnover in 2018 of 1,008 million euros. the group owns 25 factories distributed over several countries and its fleet consists of around 70 boats that fish mainly in the southern hemisphere. it has nearly 10,000 employees. the main species caught and marketed are hake, shrimps, european flying squid, prawns and pink cusk-eel. in 2018, its sales reached 177,000 tonnes of fish products and 62,000 tonnes from aquaculture. iberconsa. second largest fishing group in spain in turnover and tonnes. founded in 1981 by a group of galician ship-owners. today it is an international group formed by 20 companies, with a turnover of 350 million euros and 3,600 workers. its fleet consists of 45 boats and five processing plants located in argentina, namibia, south africa and spain. production in 2018 was 124,000 tonnes, of which 75,500 tonnes were processed in argentina. the main species are prawns, red shrimp and pink cusk-eel. portobello in 2015 paid out 30 million for 55% of the group. later, in 2019, the north american venture capital fund, platinum equity, completed the purchase of the entire group for a volume of over 500 million euros. grupo calvo. founded in 1940, it specialises in canning. it is currently a global food company encompassing 25 trading companies. in 1993, it began its international expansion by acquiring the italian company nostromo; in 2003, it opened a canning factory in el salvador, and in 2004, it purchased the brazilian company gomes da costa. according to 2019 figures, it has a turnover of 593 million euros, has a presence in 74 countries and between its three factories (a coruña, galicia; la unión, el salvador; and itajai, brazil) it produces 101,000 tonnes of finished products. it owns seven large tuna boats, two merchant boats and two support vessels, and it serves a hundred tuna suppliers. the ownership structure is composed of two groups of shareholders: the calvo family, owning 60%, and the italian group bolton, with 40%. corporación jealsa. founded in 1958 as a family group, it is the leading manufacturer of canned fish in spain, and the second largest in europe. it is made up of 26 societies divided into four activities: food; fishing and related services, circular economy, and energy. its turnover in 2018 amounted to 639 million euros, with a workforce of 4,600 people. it has factories and martin palmero and gonzález laxe / european journal of government and economics 10(1), june 2021, 65-79 79 branches in puerto quetzal (guatemala), puerto montt (chile) and fortaleza (brazil). it owns two tuna boats for catching skipjack, bigeye and yellowfin tuna. it is also a supplier of products for third-party brands. hijos de carlos albo. founded in 1869, it started out as an artisanal producer of salted and pickled anchovies. two of the three workplaces are located in galicia (celeiro and vigo). it has a presence in 25 markets, distributed in all five continents. it was purchased in 2016 for 61 million euros by shanghai kaichung marine international, part of the bright food holding company with the aim of internationalising the tuna trade. in 2017, it had a turnover of 92 million euros. the financialisation of the fishing industry: galicia as a case study 1. economic globalisation and the framework of financialisation 2. the effects of financialisation on the fishing industry: sustainability, privatise or perish? 3. discussion on the dynamics of financialisation 4. business trends 5. the case of galicia 6. conclusions voting turnout in greece: expressive or instrumental? vol.9 • no.1 2020 issn: 2254-7088 european journal of government and economics 9(1), june 2020. european journal of government and economics issn: 2254-7088 number 9, issue 1, june 2020 doi: https://doi.org/10.17979/ejge.2020.9.1 the politics of renewable power in spain 5-25 doi: https://doi.org/10.17979/ejge.2020.9.1.5231 john s. duffield voting turnout in greece: expressive or instrumental? 26-45 doi: https://doi.org/10.17979/ejge.2020.9.1.5426 irene daskalopoulou the effect of reduced unemployment duration on the unemployment rate: a synthetic control approach 46-73 doi: https://doi.org/10.17979/ejge.2020.9.1.5714 luzius stricker and moreno baruffini do confidence indicators have an impact on macro-financial indicators? an analysis of the financial service and real sector confidence indexes: evidence from turkey 74-94 doi: https://doi.org/10.17979/ejge.2020.9.1.5948 esra n. kilci a comparative analysis of the european union member states in terms of public spending on environmental protection in 2004-2017 95-114 doi: https://doi.org/10.17979/ejge.2020.9.1.5847 barbara pawełek https://doi.org/10.17979/ejge.2020.9.1 https://doi.org/10.17979/ejge.2020.9.1.5231 https://doi.org/10.17979/ejge.2020.9.1.5426 https://doi.org/10.17979/ejge.2020.9.1.5714 https://doi.org/10.17979/ejge.2020.9.1.5948 https://doi.org/10.17979/ejge.2020.9.1.5847 european journal of government and economics 9(1), june 2020, 26-45 26 european journal of government and economics issn: 2254-7088 voting turnout in greece: expressive or instrumental? irene daskalopouloua* a university of peloponnese, department of economics, greece * corresponding author at: daskal@uop.gr article history. received 25 june 2019; first revision required 12 november 2019; accepted 2 january 2020. abstract. the study analyses the micro-level determinants of voting turnout rates in greece. in particular, we test for the effects of citizens’ socio-economic features, political participation, activism and trust as pointing to either an expressive or instrumental voting decision process. the analysis involves bootstrap logistic regression techniques and ess data covering the 2002-2011 period. evidence is found of instrumental voting in greece as suggested by the effects of absolute and relative income and the effect of civic participation and trust variables. in addition, the profile of voters is differentiated in the preand during the crisis periods. the study makes a twofold contribution. first, the suggested analysis is unique for greece, and thus it provides important information regarding citizens’ motives towards electoral participation. the second contribution relates to the study’s relevance to policy analysis and design. results suggest that not only ideology but also the economic agenda might be an important predictor of electoral participation and consequently, legitimization and the quality of democracy in greece. keywords. voting turnout; political participation; activism; trust; economic crisis; greece jel codes. d72 doi. https://doi.org/10.17979/ejge.2020.9.1.5426 1. introduction voting is perhaps the single most important evidence of the legitimacy pertaining to a democratic regime. as an essential element of participation, voting turnout constitutes the sine qua non of democratic elections and a means for people to legally take part in collective decision making processes and change their governing officials (lipset, 1959; dahl, 1982; lijphart 1999; schmitter and karl, 1991). in that sense, political democracy is about regulating the political power held by elites in contrast to non-elites (bollen, 1980), i.e. it is a synthesis of political freedom and political equality (munck, 2016). the quality of democracy is subject to the existing nexus between the political system of a society and other characteristics, e.g. the modernization process, social justice and a market–based economic system (lipset, 1959). these characteristics are actually societal choices and phenomena with multifaceted causes and consequences (bollen, 1990; munck and verkuilen, 2002). furthermore, they account for the difference between democracy and democratic standards, i.e. the difference between the formal rule of law and socio-political and economic outcomes (lipset, 1959; hewitt, 1977; dahl, 1984; gastil, 1987; munck, 2016). mailto:daskal@uop.gr https://doi.org/10.17979/ejge.2020.9.1.5426 irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 27 participation in that sense is a critical element of democracy as operationalized in practice, since it practically safeguards political freedom and political equality from turning into mere formalities (munck, 2016). when the majority of citizens have the power to change the status quo, then democracy carries the most desirable properties of stability, legitimacy and effectiveness (lipset, 1959; dahl, 1982; lijphart 1999; schmitter and karl, 1991; munck, 2016). these properties contribute to socio-economic welfare as different individual preferences are matched via elections (lipset, 1959; schmitter and karl, 1991). given that electoral participation may fluctuate alongside national contexts (e.g. compulsory voting, political system), types of elections, and time, important insights might be provided via country-level evidence regarding the individuals’ decision to participate in national elections. from the early 1980s onwards, voting abstention rates in greece constantly increase at a slow, albeit standard, rate. this fact indicates the presence of a possibly persistent trend that merits deeper analysis. political rights and political liberties in greece are sufficient to characterize the country as a fair and stable political democracy (danopoulos, 2017). nevertheless, political accountability in the country is weak, in all its aspects (vertical, horizontal, and social) a fact that hampers the quality of democracy (danopoulos, 2015). increasing abstention rates point to a legitimization crisis that most probably relates to the wider civic culture qualities of the greek society (daskalopoulou, 2018a). the analysis of the various political, social and economic factors as predictors of voting turnout might offer important insights to the topic. to that extent, we analyze the effect of civic culture features on the probability of voting turnout in greece. in particular, the present study has a twofold aim. first, we are interested in sketching the profile of voters (compared to non-voters) in order to identify the socio-economic and demographic characteristics of people who are more likely to participate in this crucial democratic legitimization process. the second aim relates to identifying the possible effect that formal and latent political participation forms, activism and trust might exercise upon an individual’s decision to participate in national elections in greece. taken together these voting determinants will allow us to differentiate between expressive participation (acts motivated by sense of identity and obligation to neighbors or community, for example) and instrumental participation (acts motivated by the functional and political concerns of people such as protect personal investments and promote local businesses, for example) (dahl, 1984; talo and mannarini, 2015) as the underlying motive of voting turnout in greece. as regards the study’s contribution, two points need to be made. first, the suggested analysis is unique for greece as no previous study has been performed in this area and will thus provide us with important information regarding people’s motives towards electoral participation. the second contribution relates to the study’s relevance for policy analysis and design. during the past decades, greece has made important achievements with regard to its integration in the european union (eu) regulation framework and procedures. nevertheless, the country’s socio-economic and institutional basis has proven unable to handle the impact of the financial crisis while the governmental authorities and political organizations have largely failed to gain widespread support for the necessary structural changes that might ensure greece’s irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 28 sustainable socio-economic development (bitros, 2013; bitros and karayiannis, 2013). the decade of crisis was thus a period of profound asymmetry between individuals’ motives and policy-makers’ objectives. a clearer understanding of the motives underlying political participation procedures and the content of consent attributed to voting is crucial as for the country to be able to build stronger institutions that will help her address future challenges (baltas, 2013; bitros, 2015). here we assume that knowledge on the potentially expressive or instrumental voting character of features such as economic status, formal and informal participation, activism and trust might enhance our understanding of how to build social consensus via actively supporting the key societal decision-making mechanisms of voting participation. the empirical analysis is based on ess data referring to the 2002-2011 period for greece. analysis differentiates between the pre-crisis (2002-2008) and the crisis period (2011) and yields important evidence with regard to the profile of voters and the instrumental nature of their electoral participation. the rest of the paper is organized as follows: part 2 is devoted to a brief presentation of the study’s theoretical context. part 3 presents the model and data. part 4 presents the results and part 5 concludes the paper with a discussion of the study’s findings. 2. theoretical context: the democracy-participation relationship political democracy is a synthesis between political freedom and political equality rules that are set forth in order to facilitate collective decision making in the presence of different preferences (lipset, 1959; munck, 2016). it is a political system ‘… which supplies regular constitutional opportunities for changing the governing officials’ (lipset, 1959: 71), and regulates the difference in political power held by elites and non-elites (bollen, 1980). furthermore, constitutional democracy carries stability, legitimacy and effectiveness because it is the vehicle for achieving the wider socio-economic goals of society (lipset, 1959; dahl, 1982; lijphart 1999; schmitter and karl, 1991). to that extent, political democracy is inexorably linked to social and economic goals, but it is not identical to social democracy and/or economic democracy (lipset, 1959; hewitt, 1977; dahl, 1982; gastil, 1987). socio-economic concerns often enter the discussion regarding the quality of a democracy. this relates directly to the nexus between the political system and other societal choices such as the modernization process of societies, social justice and a market–based economic system (bollen, 1990; munck and verkuilen, (2002). according to schmitter and karl (1991: 83), we might identify political democracy through the presence of key democratic institutions such as: a) consensus, i.e. people’s degree of agreement with substantive political actions and the role of the state, b) participation, i.e. rules supporting active and equal participation in politics should one wishes to, c) access, i.e. equal opportunities of groups to express their preferences, d) responsiveness, i.e. rulers must be held accountable for their actions through regular and fair processes, and e) parliamentary sovereignty, i.e. the legislature must not be the only body that makes rules or even be the only irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 29 body with the final authority to decide which laws are binding. these democratic institutions appear through an immense variety of empirical manifestations (munck, 2016). the crucial role of the social environment of politics is clearly evidenced in the cross-national differences in democracy. we might categorize the origins of these differences into four wide areas. the first one relates to cross-national differences in the political system and the legitimacy of a country’s institutions (e.g. confidence in a country’s government and parliament) (klingemann, 1999; karp et al., 2003; aarts and thomassen, 2008; ariely, 2015). the second one relates to the type and stock of social capital and in particular trust and solidarity as key societal features (fukuyama, 2001; 2014; putnam, 1995; newton, 1997; marozzi, 2015). the third area of differences among countries relates to the role of mass media and their interaction with society (newton, 1997; fukuyama, 2014; ceron and memoli, 2016). finally, the fourth area relates to differences in what is known as abstract types of civic participation and engagement, or else disengagement, which is thought to be a genuine and active style of participation in modern economies (ekman and amnå, 2012; talò and mannarini, 2015). thus, citizenship and the decision-making standards in a democracy evolve through societal characteristics such as mutual trust, fairness and the willingness to compromise, trust in institutions, civil organizations and social movements, and so on, or else, ‘civic culture’ (dahl, 1984; schmitter and karl, 1991; bollen, 1990; newton, 1997; norris 2001; fukuyama, 2001; yamagishi, 2001). given the democratic legitimization power of political participation, the developed countries view the increasingly declining turnout rates in their national (and supranational) electorates as an unexpected ‘paradox’ (powell, 1986; flickingerand studlar, 1992). cross country studies have come to analyze the phenomenon, and their evidence suggest that electoral participation is affected by the quality of institutions underlying a democratic regime, e.g. the role of mass media freedom and political representation, socio-demographic characteristics and political preferences, economic conditions and the political system (matsusaka, 1995; feddersen and pesendorfer, 1996; sobbrio and navarra, 2010; lewis-beck and nadeau, 2012; birch, 2018). summarizing the common ground in the field, a number of studies suggest that voting turnout is influenced by a great number of factors which might be distinguished into three wide sets namely, the socio-economic environment, institutions, and party systems (powell, 1986; blais and dobrzynska, 1998; franklin, 2001; grönlund and setälä, 2007). while acknowledging these three sets of factors as crucial in determining turnout, their variation across national and supranational contexts suggests that we are still far from a thorough understanding of why people vote. at the theoretical level, macro approaches lack a plausible theory of human motivation that might be used to provide comprehensive explanations of electoral participation thus leading to a general aggregate level theory (lane and ersson, 1990). indeed, differences in institutional arrangements and cultural factors account for crossnational variation in voter turnout rates (jackman and miller, 1995). through a meta-analysis that assesses the empirical evidence of 83 aggregate-level studies, geys (2006) argues that we indeed lack a ‘core’ model of voter turnout. on the other hand, micro level studies increasingly stress the need to analyze further the role of differences in political preferences, institutions and irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 30 the socio-economic environment in order to acquire more comprehensive knowledge of such phenomena as voting turnout. indicative are the findings of sobbrio and navarra (2010) who stress that political preferences and education seem to play a significant role in the likelihood of ‘communicating voting’ and this expression is different between left-wing and right-wing voters. similarly, in their individual level study grönlund and setälä (2007), analyze institutional trust, and in particular trust in parliament, as a key determinant that increases the likelihood of voting. in the present study, we analyze voting turnout in greece using individual level data that will allow us to determine the role of socio-demographic and economic conditions as well as the role of individual level preferences over institutions and the political system of the country. available knowledge includes very few studies regarding the determinants of voting in greece. at the macro level of analysis, indicative is the study of alogoskoufis and philippopoulos (1991) who extend the ‘rational partisan model’ to introduce the role of inflation and unemployment dynamics as voter determinants. more recently, the study of nezi (2012) uses individual level data to test for the ‘grievance asymmetry’ hypothesis in relation to support for the incumbent party while kosmidis (2013) analyses the link between voting intentions and the state of the economy during the crisis period. other studies deal with the role of mass media in voting and voting intentions in greece (papagiannidis et al., 2012) and the role of subjective individual perceptions in economic voting (freire and costa lobo, 2005). here, we follow dahl (1984) and talo and mannarini (2015) and we try to differentiate between expressive participation (acts motivated by sense of identity and obligation to neighbors, community etc.) and instrumental participation (acts motivated by the functional and political concerns of people such as protect personal investments and promote local businesses etc.) as determinants of voting turnout in the case of greece. to do so we set a twofold aim that consists of: 1) sketching the socio-demographic and economic profile of voters (compared to non-voters) and, 2) identifying the possible effect that formal and latent political participation, activism and trust might exercise upon an individual’s decision to participate in elections in greece. taken together, these two sets of voting determinants will allow us to draw more informed conclusions regarding the individuals that are more likely to participate in such a crucial democratic legitimization process such as parliamentary elections. assessing the quality of democracy in post-1974 greece1, danopoulos (2015; 2017) concludes that the country’s quality of democracy is fair, but is in need of improvement. daskalopoulou (2018a) reports low individual level rates of satisfaction with democracy in greece that depend largely upon the perceived quality of civil institutions in the country. with constantly decreasing voting turnout rates in greece, concern has grown over key aspects of our democracy, namely legitimization and representation. voting turnout in greece is compulsory. this is a quite important characteristic of the greek democracy since, in terms of political democracy, mandatory electoral participation is a fair institution, an equitable and effective coordination device to support for the provision of democracy as a public good (birch, 2018). 1 in 1974, democracy has been restored in the country after the collapse of the dictatorship that ruled the country from 1967. irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 31 figure 1. voting abstention in parliamentary elections in greece since 1974. source: own calculations based on parliamentary elections data available at: https://www.hellenicparliament.gr/en/vouli-ton-ellinon/to-politevma/ekloges/eklogika-apotelesmata-new/ to that extent, both political and socio-economic equality are to be expected as the outcomes of a democracy (birch, 2018). nevertheless, greece’s aim of preserving electoral participation as compulsory is rather vague. the compulsory voting clause is practically invalid since penalties to non-voters were rarely applied ever since the enactment of compulsory voting in the country while, in 2001 the interpretive act allowing for the introduction of penalties to non-voters was withdrawn (malkopoulou, 2014). to that extent, we might relate compulsory voting to citizens’ sense of duty towards the democratic regime, but the expected legitimization outcome of such a regime quality rule might not be fully realized. as figure 1 shows, voting abstention in parliamentary elections in greece has more than doubled in the last four decades, from 20.46% in 1974 to 43.43% in 2015. in the latest parliamentary elections held on july 2019, the abstention rate was again very high, reaching 42.09%. these percentages might be somewhat overestimated due to old records cataloguing a bigger eligible to vote population. nonetheless, the voting abstention trend is recorded as important and persistent. within this context, we analyze the individual level motivation and mobilization determinants of voting turnout in greece by means of testing the following hypotheses: h1. the socio-demographic and economic characteristics of respondents affect their voting turnout decision. 20.46 18.88 18.50 19.82 19.67 19.30 20.76 21.77 23.65 25.03 23.50 25.85 29.05 34.88 37.51 36.38 43.43 0 5 10 15 20 25 30 35 40 45 https://www.hellenicparliament.gr/en/vouli-ton-ellinon/to-politevma/ekloges/eklogika-apotelesmata-new/ irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 32 h2. formal political participation will exercise a statistically significant and positive (negative) effect on the probability of voting. h3. latent political participation will exercise a statistically significant and positive (negative) effect upon the probability of voting if it operates as complementary (substitute) to formal political participation. h4. activism will exercise a statistically significant and positive (negative) effect upon the probability of voting if it operates as complementary (substitute) to formal political participation. h5. trust will exercise a statistically significant and positive (negative) effect upon the probability of voting. the presence and the sign of the above-described effects are expected to provide us with important insights as regards the individuals’ expectations, perceptions, and the overall motivation and mobilization factors that underlie their decision to vote. h1 is considered the benchmark model that controls for the socio-demographic and economic characteristics of individuals which together with h2-h5 will provide us with important information about the profile of voters and the importance of economic or expressive considerations in this decision. 3. empirical model and data 3.1. the model as explained in the previous part, the aim is to identify those factors that will enhance the probability that an individual participates in elections. thus, a person’s decision to vote may be modelled as a binary (dichotomous) dependent variable of the form: 𝑦𝑦 = �1, 𝑖𝑖𝑖𝑖 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝 𝑖𝑖 𝑣𝑣𝑝𝑝𝑣𝑣𝑝𝑝𝑣𝑣 0, 𝑝𝑝𝑣𝑣ℎ𝑝𝑝𝑝𝑝𝑒𝑒𝑖𝑖𝑝𝑝𝑝𝑝 , [1] where y is the dependent variable denoting voters and non-voters amongst those that are eligible to vote. in principle, any continuous probability distribution defined over the real line will suffice to obtain consistent predictions of the probability of the outcomes expressed in equation [1] (greene, 1997). either a normal distribution (probit model) or a logistic distribution (logit model) can be used to model the above outcomes. the two distributions are expected to give similar predictions unless the sample contains very few responses/non-responses (i.e. very few values of y equal to 1 or y equal to 0) and/or there is wide variation in an important independent irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 33 variable (greene, 1997; amemiya, 1981). in our case, there is a very large difference in the percentage of voters compared to that of non-voters (see table 1) so we have chosen to use a logistic distribution. in that case we get a logit model of the form: ( ) ( ) ' ' 'prob 1 1 x x ey x e b b b= = = λ + , [2] where ( ).λ indicates the logistic cumulative distribution function, x is a vector of explanatory variables, and b is a set of corresponding parameters that reflect the impact of changes in x on the probability of y*. bootstrap logistic regression has been performed in order to obtain robust estimations and account for the lower (compared to the estimated voting abstention rate during the period of analysis) percentage of non-voters in our sample. our estimation procedures involve stratified resampling using 50,000 samples and 95% bias corrected and accelerated (bca) confidence intervals (efron and tibshirani, 1986). the goodness of fit measures usually reported are the percent correctly predicted and various pseudo-r squared measures, the most often cited being the likelihood-ratio test statistic suggested by mcfadden (1974). here, we report the χ2 value of the omnibus test of the null hypothesis that the regression slopes for all predictors in the model are equal to zero, the log likelihood value (-2ll) which also tests for the significance of the explanatory variables model (full model) compared to the null model, two pseudo-r2 values2 that tell us approximately how much variation in the outcome is explained by the model and, the hosmer and lemeshow test of the goodness of fit, which is a chi-square (χ2) test of whether or not the model is an adequate fit to the data3 (pituch and stevens, 2016). it should be noted though that goodness-of-fit measures in the case of binary response models is not as important as statistical and economic significance of the explanatory variables (wooldridge, 2002; estrella, 1998). a final note refers to the interpretation of the logit model coefficients. the estimatedb ’s indicate the amount of increase (or decrease, if the sign of the coefficient is negative) in the predicted log odds of y = 1 that would be predicted by a 1 unit increase (or decrease) in the predictor, holding all other predictors constant. thus, in the case of the logit model the slope coefficient b is interpreted as the rate of change in the "log odds" of the dependent variable (y) as an independent variable (x) changes. because this explanation is not very intuitive, it is accustomed to compute the more intuitive expb, which is the effect of the independent variable on the odds ratio4. 2 the versions are the cox & snell and the nagelkerke tests which are again used as approximations since they vary significantly depending on sample size and specification (cox and snell, 1989; nagelkerke, 1991). 3 the null hypothesis is that the model is a ‘good enough’ fit to the data (p=>.05) and we will only reject this null hypothesis, i.e. the model is a ‘poor’ fit, if p<.05. the test is subject to sample size and the inclusion of interactions in the data so again it should be considered as an approximation (hosmer and lemeshow, 2013). 4 the odds ratio is the probability of the event divided by the probability of the nonevent. for example, if expb1 =2, then a one unit change in the independent variable x1 would make the event twice as likely (.67/.33) to occur. for more details see: ucla, scg. available at: https://stats.idre.ucla.edu/sas/modules/sas-learning-moduleintroduction-to-thefeatures-of-sas/. https://stats.idre.ucla.edu/sas/modules/sas-learning-moduleintroduction-to-the-features-of-sas/ https://stats.idre.ucla.edu/sas/modules/sas-learning-moduleintroduction-to-the-features-of-sas/ irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 34 3.2. data and variables the sample consists of a total number of 9,740 observations obtained from the four ess waves that are available for greece (waves 1-2002, 2-2006, 4-2008 and 5-2011). a usable sample of 9,135 observations consisting of the respondents that are eligible to vote, has been selected. voters represent 81.6% of the sample (7,863 obs.) and non-voters represent 13.9% of the sample (1,272 obs.) (table 1). our dependent variable is a binary one taking the value of 1 if the respondent has voted in the last national elections and 0 if he/she hasn’t. the independent variables have been divided into five sets of factors referring to: 1st) the socio-economic and demographic profile of the respondents (lnage, lneducation, gender, household size, children, lives with husband/wife/partner at household grid, household income, income satisfaction), 2nd) their pattern of formal political participation (worked in political party or action group last 12 months, member of political party, contacted politician or government official last 12 months), 3rd) their pattern of latent political participation (feel closer to a particular party than all other parties, how interested in politics, tv watching, news/politics/current affairs on average weekday, placement on left to right scale), 4th) their pattern of activism (worn or displayed campaign badge/sticker last 12 months, signed petition last 12 months, taken part in lawful public demonstration last 12 months, boycotted certain products last 12 months, worked in another organization or association last 12 months), 5th) the individuals’ level of generalized and institutional trust (most people can be trusted or you can't be too careful, trust in country's parliament, trust in the legal system, trust in the police, trust in the european parliament, trust in the united nations). finally, wave dummies have been used to test for the presence of time structural breaks in our model. table 2 presents the definition and measurement of variables as well as basic descriptive statistics for the whole sample and the voters and the non-voters sub-samples. table 1. distribution of voters and non-voters in the sample. wave 1 wave 2 wave 4 wave 5 total voters 2,139 2,050 1,692 1,982 7,863 non-voters 248 229 252 543 1,272 not eligible to vote 167 126 125 187 605 total 2,554 2,405 2,069 2,712 9,740 total usable sample 2,387 2,279 1,944 2,525 9,135 source: own calculations using ess data for greece. irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 35 table 2. definition, measurement and basic descriptive statistics for the whole sample and the voters / non-voters subsamples. variable definition and measurement basic descriptive statistics all voters non-voters mean sd mean sd mean sd socio-demographic and economic characteristics age, lnage of respondent, calculated (13-98) 48.20 18.63 50.57 17.48 43.17 20.70 education, lnyears of full-time education completed (0-28) 10.56 4.49 10.44 4.63 11.06 4.21 gender (dummy, 1=male) .44 .50 .44 .50 .42 .49 children (dummy, 1=yes) .41 .49 .43 .50 .30 .46 household size, lnnumber of people living regularly at the household 2.69 1.31 2.71 1.30 2.55 1.34 married (dummy, 1=lives with husband/wife/partner at household grid) .60 .49 .65 .48 .42 .49 household income, in twelve income categories (1-12) 4.82 2.21 4.89 2.21 4.50 2.21 income satifaction, (0-3, 3 = living comfortably on present income) 1.35 .89 1.37 .89 1.26 .88 formal political participation worked in political party or action group last 12 months (dummy, 1=yes) .04 .21 .05 .22 .02 .12 member of political party (dummy, 1=yes) .06 .24 .07 .25 .01 .12 contacted politician or government official last 12 months (dummy, 1=yes) .12 .32 .13 .34 .04 .20 latent political participation feel closer to a particular party than all other parties (dummy, 1=yes) .51 .41 .58 .49 .23 .42 how interested in politics (0-3, 0 = not at all interested) 1.05 .95 1.12 .95 .78 .90 tv watching, news/politics/current affairs on average weekday (0-7, 0 = no time at all) 1.81 1.47 1.90 1.47 1.47 1.43 placement on left to right scale (0-10, 10 = right) 5.42 2.17 5.47 2.18 5.02 2.05 activism worn or displayed campaign badge/sticker last 12 months (dummy, 1=yes) .03 .17 .03 .17 .02 .13 signed petition last 12 months dummy, 1=yes) .04 .21 .05 .21 .03 .17 taken part in lawful public demonstration last 12 months (dummy, 1=yes) .06 .25 .07 .25 .05 .23 boycotted certain products last 12 months (dummy, 1=yes) .10 .30 .10 .31 .10 .31 worked in another organisation or association last 12 months (dummy, 1=yes) .05 .22 .05 .23 .02 .15 generalized and institutional trust most people can be trusted or you can't be too careful (0-10, 10 = complete trust) 3.87 2.40 3.81 2.40 4.05 2.39 trust in country's parliament (0-10, 10 = complete trust) 3.74 2.74 3.82 2.74 3.01 2.64 trust in the legal system (0-10, 10 = complete trust) 5.07 2.90 5.10 2.88 4.55 2.92 trust in the police (0-10, 10 = complete trust) 5.51 2.87 5.61 2.84 4.84 2.97 trust in the european parliament (0-10, 10 = complete trust) 4.43 2.82 4.48 2.80 3.71 2.78 trust in the united nations (0-10, 10 = complete trust) 3.74 2.79 3.72 2.76 3.36 2.70 source: own calculations using ess data for greece. 4. results table 3 presents summary statistics for the estimated models. our benchmark model (m1) is the model including only the socio-economic and demographic and the time variables (control variables model). m1 is improved with the inclusion of the proposed participation and trust variables, and thus the full model (m2) is preferred. irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 36 model 2 referring to the whole period under study presents a very satisfactory fit to the data and can thus be used as a reference point in the analysis of the predictors of voting turnout in greece (table 4). however, since there is a structural break in the model, evidenced by the statistically significant sign of the 2011 wave dummy, we have split the sample into the pre and the crisis periods and estimated models 3 and 4, respectively. after robust estimation procedures were applied the structural break in our model is estimated as negative suggesting that the probability of voting decreased in the crisis period (table 4). other important changes in the estimated coefficients refer to the effect of education and the income variables (table 4). comparing the results of models 3 and 4 (table 5), we see that the socio-economic and demographic profile of voters in the two periods is different (h1 is confirmed). in the pre-crisis period, the probability of voting turnout increases with age, children and income. in contrast, during the crisis, the probability of voting turnout increases with age, marriage and household size. as regards the economic condition of respondents, it is important that income satisfaction is a positive voting predictor while income levels turn to a negative predictor. the difference in the profile of voters in the two periods is important evidence. in the pre-crisis period voters are older, higher income people with children in their family. in the crisis period, we see that voting turnout is more likely to occur for older, married people with larger household size who are satisfied with their relative income position but unsatisfied with their absolute income level. table 3. model statistics. model summary m1 m2 m3 m4 n 9,135 9,135 6,610 2,525 x2 (p) 388.581 (p<.001) 357.609 (p<.001) 232.778 (p<.001) 119.190 (p<.001) -2ll 4,376.094 2,566.537 1,615.719 916.336 cox & snell r2 .062 .078 .070 .093 nagelkerke r2 .114 .160 .160 .163 hosmer & lemeshow χ2test (p) 9.029 (p=.340) 6.031 (p=.644) 6.577 (p=.583) 7.590 (p=.475) classificationa overall percentage correct 85.7 89.4 91.3 84.9 sensitivity 97.4 98.4 98.8 96.2 aclassification cutoff point 0.65. irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 37 table 4. bootstrap logistic regression estimates: benchmark model and full model. benchmark model (m1) full model (m2) b wald 2χ p exp(b) b wald 2χ p exp(b) constant -4.089 44.800 .000 .017 2.710 3.282 .070 15.033 socio-demographic & economic controls age 1.260 103.744 .000 3.524 .935 29.372 .000 2.547 education .270 8.226 .004 1.310 .222 2.523 .112 1.249 gender .073 .798 .372 1.076 .073 .446 .504 1.076 married .461 21.370 .000 1.586 .327 6.035 .014 1.387 household size .374 11.862 .001 1.454 .281 3.831 .050 1.324 children .066 .340 .560 1.069 .254 2.778 .096 1.289 income satisfaction .165 10.391 .001 1.180 .107 2.309 .129 1.113 household income .000 .000 .997 1.000 .005 .031 .861 1.005 wave 1 2002 .086 .446 .504 1.089 .027 .026 .871 1.028 wave 2 2006 .066 .252 .616 1.068 .175 .989 .320 1.191 wave 5 2010 -.777 45.567 .000 .460 -.348 4.832 .028 .706 formal political participation member political party -.551 2.287 .130 .576 contacted politician / official -.405 3.415 .065 .667 worked political party .017 .002 .966 1.017 latent political participation tv watching, news / politics / current affairs -.003 .007 .935 .997 closer to particular party 1.031 76.478 .000 .357 interest in politics -.170 6.498 .011 .844 left right placement .074 7.070 .008 1.077 activism worked organization / association -.229 .458 .499 .795 campaign badge / sticker -.324 .456 .500 .723 signed petition -.483 2.001 .157 .617 public demonstration -.295 1.621 .203 .745 boycotted products .515 9.005 .003 1.674 trust generalized trust .006 .054 .816 1.006 country's parliament .024 .676 .411 1.024 legal system -.075 6.452 .011 .928 police .049 3.338 .068 1.050 european parliament .046 1.685 .194 1.048 united nations -.067 4.632 .031 .935 irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 38 table 5. bootstrap logistic regression estimates: full model for the preand during the crisis periods. full model: pre-crisis (m3) full model: crisis (m4) b wald 2χ p exp(b) b wald 2χ p exp(b) constant .402 .048 .827 1.495 7.636 5.227 .022 2071.747 socio-demographic & economic controls age 1.152 27.135 .000 3.165 .642 5.048 .025 1.900 education .116 .396 .529 1.123 .369 2.617 .106 1.446 gender .165 1.379 .240 1.180 -.081 .199 .655 .922 married .238 1.961 .161 1.269 .404 3.262 .071 1.497 household size .263 2.183 .140 1.301 .450 3.000 .083 1.568 children .370 3.581 .058 1.448 .062 .058 .809 1.064 income satisfaction -.022 .056 .814 .979 .329 8.127 .004 1.390 household income .078 3.966 .046 1.081 -.095 4.137 .042 .910 wave 1 2002 .014 .006 .937 1.014 wave 2 2006 .155 .730 .393 1.168 formal political participation member political party -.445 1.105 .293 .641 -.873 1.330 .249 .418 contacted politician / official -.414 2.414 .120 .661 -.664 2.622 .105 .515 worked political party .233 .275 .600 1.262 -.727 .437 .509 .483 latent political participation tv watching, news / politics / current affairs .003 .004 .952 1.003 .012 .032 .859 1.012 interest in politics -.201 5.053 .025 .818 -.133 1.640 .200 .876 closer to particular party -.945 42.842 .000 .389 -1.289 34.292 .000 .276 left right placement .085 5.625 .018 1.088 .064 1.971 .160 1.066 activism worked organization / association -.168 .175 .676 .845 -.476 .518 .472 .621 campaign badge / sticker .126 .055 .815 1.134 -1.210 1.217 .270 .298 signed petition -1.093 4.672 .031 .335 .250 .257 .612 1.284 public demonstration .218 .474 .491 1.244 -.657 3.651 .056 .518 boycotted products .526 6.028 .014 1.693 .603 4.098 .043 1.828 trust generalized trust .013 .192 .661 1.013 -.007 .027 .869 .993 country's parliament .026 .526 .468 1.027 .023 .196 .658 1.024 legal system -.066 2.922 .087 .936 -.078 2.753 .097 .925 police .045 1.650 .199 1.046 .049 1.359 .244 1.051 european parliament .056 1.596 .207 1.057 .032 .243 .622 1.033 united nations -.090 5.731 .017 .914 -.046 .551 .458 .955 irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 39 in the two periods, important changes are observed with regard to the effect of the social capital variables analyzed. formal political participation exerts a statistically significant but negative effect on the probability of voting (h2 is confirmed). more specifically, formal political participation in the form of political contacts decreases the probability of voting. this effect, however, is not confirmed when the two periods are analyzed separately (h2 is not confirmed in m3, m4). mixed results are presented in the case of latent political participation (h3 is confirmed). as shown, the probability of voting decreases for those respondents who feel closer to a particular party and show higher levels of interest in politics while it increases as placement on the left to right political scale increases. in the crisis period, feeling closer to a particular party is the only statistically significant voting predictor that again exerts a negative effect. activism also presents mixed effects as a determinant of voting turnout (h4 is confirmed). in the precrisis period, the probability of voting increases for respondents which have boycotted certain products and decreases for the respondents who have signed petitions. in the crisis period, voting turnout increases for respondents who have boycotted certain products and decreases for those who have participated in public demonstrations. finally, as regards the effect of trust on the probability of voting turnout we see that generalized trust does not affect the decision to vote while institutional trust is a voting predictor (h5 is confirmed). it is important, however, that trust in the legal system is found a negative voting predictor, and this finding is observed in both periods. trust in supranational institutions such as the un is also a negative voting predictor but the effect of this variable is reported only in the pre-crisis period. it could be associated with greek citizens’ increased awareness over the global role and interventions of such institutions, albeit it is a finding that merits further research in the future. given ess data availability, and bootstrap estimation, other variables that could be used as explanatory variables in the current context were also tested for their possible effect. in particular, we have tested for the possible sensitiveness of our results with regard to the respondents’: a) employment status and type of employment; b) political beliefs (trust in politicians, trust in political parties); c) use of other sources of information about politics (newspaper reading, politics/current affairs on average weekday, and/or radio listening, news/politics/current affairs on average weekday); and d) abstract forms of engagement (feelings about politics, e.g. politics too complicated to understand, difficulty in making mind up about political issues) (mitsopoulos and pelagidis, 2009; ceron and memoli, 2016; daskalopoulou, 2018b; talò and mannarini, 2015; ekman and amnå, 2012). none of these variables has been found to exert a statistically significant effect on the probability of voting turnout. excluding them from the estimated models did not affect their fit and the corresponding classification rates. it is possible that the effect of these variables is captured by the ones already included in the analysis. an additional possible explanation might relate to the high nonresponse rate of some of these variables (e.g. the news variables, trust in politicians, trust in political parties). in any case however, their possible effect should be tested in future research studying voting turnout (and abstention) in the post-crisis era in the country. irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 40 5. conclusion and discussion the present study aims at identifying the determinants of voting turnout in greece using european social survey data for the 2002-2011 period and a binary dependent variable model. five empirical hypotheses are formed and tested in the context of the study using bootstrap logistic regression techniques. through the estimation of several models we test for the sociodemographic and economic characteristics that affect the respondents’ voting turnout decision, and the effect of formal political participation, latent political participation, activism, and trust (generalized and institutional), on the probability a person decides to vote. the analysis controls for the structural break observed in the model after the onset of the economic crisis in the country and particularly after the enforcement of the first financial consolidation measures. empirical results sketch two different profiles that are compatible with a backward turn, or perhaps a delay, in the democratic modernization process of greece. more specifically, in the pre-crisis period, we see that the voters are people who value contemporary forms of civic engagement (latent political participation and activism) and tend to distrust formal institutions such as the legal system and supranational institutions. taken together, these effects seem to suggest the presence of a political distrust trend (or a more apolitical stance). this trend involves older higher income respondents having children and who consider other forms of civic engagement, e.g. petitions, as a substitute to standard civil participation processes, and distrust public institutions (legal system) and supranational institutions like the un. in the crisis period, the socio-demographic and economic profile of voters is different. results show that age, marriage and household size are strong positive predictors of voting in that period. an interesting finding relates to the effect of the income variables. voting turnout increases as income satisfaction increases and decreases with absolute income level. thus, we might argue that there is a differentiated stance after the onset of the financial consolidation measures which has caused those satisfied with their income to increase their voting rates and those of higher absolute income to decrease their voting rates. taken together, the evidence for the two periods support the argument that citizens vote in accordance with instrumental voting. people decide on the basis of income, while ‘traditional’ forms of formal engagement and political attachment seem to cause negative effects. having in mind two crucial contextual factors, namely (1) the time of the analysis (bailout programs and fiscal measures, radical political changes) and (2) the no-penalties compulsory voting clause, evidence is provided that pecuniary interests are related to strongly motivated voters. the present findings are important also in terms of policy analysis in the field. the importance of economic considerations for voting participation clearly indicates that citizens’ support to state regime is interlinked with growth and prosperity prerequisites. this, in turn, adds complexity to an existing backward spiral that commenced with the onset of the financial crisis, and the measures employed to address it, and continuous to exacerbate in the presence of other socio-economic challenges and phenomena such as unemployment and exploitation in irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 41 the workplace, migration, rising crime and insecurity etc. (pantazidou, 2013; voulgarellichristidou, 2016). furthermore, the economic and democratic depression experienced in greece (bellucci et al., 2012) coincides with the citizens’ deeper knowledge and understanding of how country level decisions are transferred to european union and taken therein in favor of an enlarged but not necessarily integrated eu community (baltas, 2013; bitros, 2015). to the extent that greece has still important work to do in terms of implementing the measures agreed under its bailout package and forwarding the deepening of structural reforms, widespread societal consensus is required as these measures will be coming in a ‘tired’ socio-economic context that is difficult to provide consent to inefficient, unfair and unproductive measures such as the ones implemented in the last years. to that extent it is important that the country builds strong institutions that might provide the societal consensus that is detrimental for the sustainability of measures that are taken in order to deal with the crisis effectively, and perhaps avoid a future one. in modern democracies, abstract types of trust prevail and thus a challenge is at hand to identify the cognitive mobilization mechanisms that will be at work in this phase of a society’s overall development process (newton, 1997). fukuyama (2001; 2014) makes similar observations regarding the way in which quality institutions enhance trust in democratic procedures. he suggests that a society’s stock of trust can be destroyed by a state that is inefficient in the provision of necessary public goods, and particularly property rights, public safety, control of the state’s involvement in market activities (fukuyama, 2001). albeit a stable and mature constitutional democracy that sustains fair political rights and liberties to her citizens (danopoulos, 2017) greece faces a democratic quality and depth challenge. increasing abstention rates point to a legitimization crisis that most probably relates to the wider civic culture qualities of the greek society (bitros, 2013; daskalopoulou, 2018a). in particular, rent seeking activities, government inefficiency and partisan politics have built change resistant barriers (bitros, 2013). to that extent, it is important to verify that in the case of greece, the voting decision seems to have a strong instrumental character. income matters and citizens seem to move away from political institutions and turn to alternative forms of engagement either as a complement or a substitute to formal political participation. to that extent the future might bring about a combination of reactions/trends towards voting which will involve an increase in commitment to vote as the ultimate instrument of participation; a trend towards a more apolitical (pathetic) stance; and a trend towards alternative forms of civic engagement (non-standard, abstract forms of engagement). the current findings suggest that wider alternative political engagement/disengagement mechanisms are present in greek society, and they merit attention and future research. the analysis of the potentially long run effects that the crisis’ measures and developments might have on citizens’ motivation to vote is an additional issue for future research in the field. acknowledgments the insightful and constructive comments of the journal’s editor and of two anonymous journal reviewers are highly appreciated. irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 42 references alogoskoufis, g. and philippopoulos, a. (1991). political parties, elections and inflation in greece, cepr discussion papers 547, c.e.p.r. discussion papers. baltas, n.c. (2013). the greek financial crisis and the outlook of the greek economy. the journal of economic asymmetries, 10(1), 32-37.https://doi.org/10.1016/j.jeca.2013.09.002 bellucci, p., lobo, m.c., and lewis-beck, m.s. (2012). economic crisis and elections: the european periphery. electoral studies, 31(3), 469-471. https://doi.org/10.1016/j.electstud.2012.02.009 birch, s. (2018). democratic norms, empirical realities and approaches to improving voter turnout. res publica, 24(1), 9-30. https://doi.org/10.1007/s11158-017-9380-3 bitros, g.c. (2013). from riches to rags or what went wrong in greece. journal of economic and social measurement, 38(1), 5-39. https://doi.org/10.3233/jem-130370 bitros, g.c. (2015). european union failures in greece and some possible explanations. in: g.c. bitros & n.c. kyriazis (eds.) essays in contemporary economics (pp. 3-22). switzerland: springer international publishing. https://doi.org/10.1007/978-3-319-10043-2_1 bitros, g.c., and karayiannis, a.d. (2013). creative crisis in democracy and economy. springer science and business media. https://doi.org/10.1007/978-3-642-33421-4 blais, a., and dobrzynska, a. (1998). turnout in electoral democracies. european journal of political research, 33(2), 239-261. https://doi.org/10.1111/1475-6765.00382 bollen, k.a. (1980). issues in the comparative measurement of political democracy. american sociological review, 45, 370-390. https://doi.org/10.2307/2095172 bollen, k.a. (1990). political democracy: conceptual and measurement traps, studies in comparative international development, 25 (1), 7-24. https://doi.org/10.1007/bf02716903 ceron, a. and memoli, v. (2016), flames and debates: do social media affect satisfaction with democracy?, social indicators research, 126(1), 225-240. https://doi.org/10.1007/s11205015-0893-x cox, d.d. and snell, e.j. (1989). the analysis of binary data (2nd ed.). chapman and hall. https://doi.org/10.2307/2531476 dahl, r.a. (1982). dilemmas of pluralist democracy, new haven, yale university press. dahl, r.a. (1984). modern political analysis, 4th ed., englewood cliffs, n.j.: prentice-hall. danopoulos, c.p. (2015). accountability and the quality of democracy in greece. mediterranean quarterly, 26(4), 110-131. https://doi.org/10.1215/10474552-3425211 danopoulos, c.p. (2017). civil liberties and rights, equality and the quality of democracy in greece. journal of balkan and near eastern studies, 19(3), 225-242. https://doi.org/10.1080/19448953.2017.1267364 daskalopoulou, i. (2018a). satisfaction with democracy and social capital in greece. international journal of social economics, 45(4), 614-628. https://doi.org/10.1108/ijse-032017-0063 https://doi.org/10.1016/j.jeca.2013.09.002 https://doi.org/10.1016/j.electstud.2012.02.009 https://doi.org/10.1007/s11158-017-9380-3 https://doi.org/10.3233/jem-130370 https://doi.org/10.1007/978-3-319-10043-2_1 https://doi.org/10.1007/978-3-642-33421-4 https://doi.org/10.1111/1475-6765.00382 https://doi.org/10.2307/2095172 https://doi.org/10.1007/bf02716903 https://doi.org/10.1007/s11205-015-0893-x https://doi.org/10.1007/s11205-015-0893-x https://doi.org/10.2307/2531476 https://doi.org/10.1215/10474552-3425211 https://doi.org/10.1080/19448953.2017.1267364 https://doi.org/10.1108/ijse-03-2017-0063 https://doi.org/10.1108/ijse-03-2017-0063 irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 43 daskalopoulou, i. (2018b). civic participation and soft social capital: evidence from greece. european political science, 17(3), 404-421. https://doi.org/10.1057/s41304-017-0114-y efron, b. and tibshirani, r. (1986). bootstrap methods for standard errors, confidence intervals, and other measures of statistical accuracy. statistical science, i, 54-77. https://doi.org/10.1214/ss/1177013815 ekman, j. and amnå, e. (2012). political participation and civic engagement: towards a new typology, human affairs, 22, 283-300. https://doi.org/10.2478/s13374-012-0024-1 feddersen, t.j., and pesendorfer, w. (1996). the swing voter's curse. the american economic review, 408-424. flickinger, r.s., and studlar, d.t. (1992). the disappearing voters? exploring declining turnout in western european elections. west european politics, 15(2), 1-16. https://doi.org/10.1080/01402389208424903 franklin, m.n. (2001). how structural factors cause turnout variations at european parliament elections. european union politics, 2(3), 309-328. https://doi.org/10.1177/ 1465116501002003003 freire, a., and lobo, m. c. (2005). economics, ideology and vote: southern europe, 19852000. european journal of political research, 44(4), 493-518. https://doi.org/10.1111/j.14756765.2005.00236.x fukuyama, f. (2001). social capital, civil society and development. third world quarterly, 22(1), 7-20. https://doi.org/10.1080/713701144 fukuyama, f. (2014). political order and political decay: from the industrial revolution to the globalization of democracy. macmillan. gastil, r.d. (1987). freedom in the world: political rights and civic liberties, 1986-1987, westport, ct, greenwood press. geys, b. (2006). explaining voter turnout: a review of aggregate-level research. electoral studies, 25(4), 637-663. https://doi.org/10.1016/j.electstud.2005.09.002 greene, w. (1997).econometric analysis, prentice-hall international, ny. grönlund, k., and setälä, m. (2007). political trust, satisfaction and voter turnout. comparative european politics, 5(4), 400-422. https://doi.org/10.1057/palgrave.cep.6110113 hewitt, c. (1977). the effect of political democracy and social democracy on equality in industrial societies: a cross-national comparison. american sociological review, 42, 450464. hosmer, d.w. and lemeshow, s. (2013). applied logistic regression. new york: wiley. https://doi.org/10.2307/2094750 jackman, r.w., and miller, r.a. (1995). voter turnout in the industrial democracies during the 1980s. comparative political studies, 27(4), 467-492. https://doi.org/10.1177/001041409 5027004001 karp, j.a., banducci, s.a, and bowler, s. (2003). to know it is to love it?, comparative political studies, 36 (3), 271-292. https://doi.org/10.1177/0010414002250669 https://doi.org/10.1057/s41304-017-0114-y https://doi.org/10.1214/ss/1177013815 https://doi.org/10.2478/s13374-012-0024-1 https://doi.org/10.1080/01402389208424903 https://doi.org/10.1177/1465116501002003003 https://doi.org/10.1177/1465116501002003003 https://doi.org/10.1111/j.1475-6765.2005.00236.x https://doi.org/10.1111/j.1475-6765.2005.00236.x https://doi.org/10.1080/713701144 https://doi.org/10.1016/j.electstud.2005.09.002 https://doi.org/10.1057/palgrave.cep.6110113 https://doi.org/10.2307/2094750 https://doi.org/10.1177/0010414095027004001 https://doi.org/10.1177/0010414095027004001 https://doi.org/10.1177/0010414002250669 irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 44 klingemann, h-d. (1999). mapping political support in the 1990s: a global analysis. in: p. norris (ed.), critical citizens: global support for democratic government (pp. 31-56), oxford, uk: oxford university press. https://doi.org/10.1093/0198295685.003.0002 kosmidis, s. (2013). government constraints and economic voting in greece. hellenic observatory papers on greece and southeast europe (70). hellenic observatory, london school of economics and political science, london, uk. lane, j.e., and ersson, s. (1990). macro and micro understanding in political science: what explains electoral participation?. european journal of political research, 18(4), 457-465. https://doi.org/10.1111/j.1475-6765.1990.tb00243.x lewis-beck, m.s., and nadeau, r. (2012). pigs or not? economic voting in southern europe. electoral studies, 31(3), 472-477. https://doi.org/10.1016/j.electstud.2012.02.008 lijphart, a. (1999).patterns of democracy: government forms and performance in thirty-six countries, new haven, ct: yale university press. lipset, s.m. (1959). some social requisites of democracy; economic development and political legitimacy. the american political science review, 53 (1), 69-105. https://doi.org/10.2307/1951731 malkopoulou, a. (2014). the history of compulsory voting in europe: democracy's duty?. routledge. https://doi.org/10.4324/9781315777399 marozzi, m., (2015). measuring trust in european public institutions. social indicators research, 123 (3), 879-895. https://doi.org/10.1007/s11205-014-0765-9 matsusaka, j. g. (1995). explaining voter turnout patterns: an information theory. public choice, 84 (1-2), 91-117. https://doi.org/10.1007/bf01047803 mcfadden, d. (1974). conditional logit analysis of qualitative choice analysis, in: zarembka p. (ed.), frontiers in econometrics, academic press, ny, pp. 105-142. mitsopoulos, m. and pelagidis, t. (2009). viking in greece: kleptocratic interest groups in a closed, rent-seeking economy. cato journal, 29(2), 389-406. munck, g. l. (2016).what is democracy? a reconceptualization of the quality of democracy. democratization, 23 (1), 1-26. https://doi.org/10.1080/13510347.2014.918104 munck, g.l. and verkuilen, j. (2002). conceptualizing and measuring democracy: evaluating alternative indices. comparative political studies, 35 (1), 5-34. https://doi.org/10.1177/001041400203500101 nagelkerke, n.j.d. (1991). a note on a general definition of the coefficient of determination. biometrika, 78(3), 691-692. https://doi.org/10.1093/biomet/78.3.691 nezi, r. (2012). economic voting under the economic crisis: evidence from greece. electoral studies, 31(3), 498-505. https://doi.org/10.1016/j.electstud.2012.02.007 norris, p., (2001). making democracies work: social capital and civic engagement in 47 societies, no. rwp01-036. pantazidou, m. (2013). treading new ground: a changing moment for citizen action in greece. development in practice, 23(5-6), 755-770. https://doi.org/10.1080/09614524.2013.801399 https://doi.org/10.1093/0198295685.003.0002 https://doi.org/10.1111/j.1475-6765.1990.tb00243.x https://doi.org/10.1016/j.electstud.2012.02.008 https://doi.org/10.2307/1951731 https://doi.org/10.4324/9781315777399 https://doi.org/10.1007/s11205-014-0765-9 https://doi.org/10.1007/bf01047803 https://doi.org/10.1080/13510347.2014.918104 https://doi.org/10.1177/001041400203500101 https://doi.org/10.1093/biomet/78.3.691 https://doi.org/10.1016/j.electstud.2012.02.007 https://doi.org/10.1080/09614524.2013.801399 irene daskalopoulou / european journal of government and economics 9(1), june 2020, 26-45 45 papagiannidis, s., coursaris, c.k., and bourlakis, m. (2012). do websites influence the nature of voting intentions? the case of two national elections in greece. computers in human behavior, 28(2), 300-307. https://doi.org/10.1016/j.chb.2011.09.013 pituch, k.a., and stevens, j. (2016). applied multivariate statistics for the social sciences: analyses with sas and ibm's spss (6th ed.). new york: routledge. https://doi.org/10.4324/9781315814919 powell, g.b. (1986). american voter turnout in comparative perspective. american political science review, 80(1), 17-43. https://doi.org/10.2307/1957082 putnam, r.d. (1995). bowling alone: america's declining social capital. journal of democracy 6(1), 65-78. https://doi.org/10.1353/jod.1995.0002 schmitter, p.c. and karl, t.l. (1991). what democracy is … and is not, journal of democracy, 2(3), 75-88. https://doi.org/10.1353/jod.1991.0033 sobbrio, f., and navarra, p. (2010). electoral participation and communicative voting in europe. european journal of political economy, 26(2), 185-207. https://doi.org/10.1016/ j.ejpoleco.2010.01.002 talò, c. and mannarini, t., (2015).measuring participation: development and validation the participatory behaviors scale. social indicators research, 123(3), 799-816. https://doi.org/10.1007/s11205-014-0761-0 voulgarelli-christidou, c. (2016). dangers of an urban crisis within the european union: fuelling xenophobia and undermining democracy. perspectives on global development and technology, 15(1-2), 236-245. https://doi.org/10.1163/15691497-12341385 wooldridge, j.m. (2002). econometric analysis of cross section and panel data, the mit press, ma. yamagishi, t., (2001). trust as a form of social intelligence, in k.s. cook (ed.), trust in society (pp. 121-147), russell sage foundation, new york city, new york. https://doi.org/10.1016/j.chb.2011.09.013 https://doi.org/10.4324/9781315814919 https://doi.org/10.2307/1957082 https://doi.org/10.1353/jod.1995.0002 https://doi.org/10.1353/jod.1991.0033 https://doi.org/10.1016/%0bj.ejpoleco.2010.01.002 https://doi.org/10.1016/%0bj.ejpoleco.2010.01.002 https://doi.org/10.1007/s11205-014-0761-0 https://doi.org/10.1163/15691497-12341385 number 9, issue 1, june 2020 voting turnout in greece: expressive or instrumental? 1. introduction 2. theoretical context: the democracy-participation relationship 3. empirical model and data 4. results 5. conclusion and discussion acknowledgments references unleashing entrepreneurial potential in transition economies: a comparative analysis of the impact of macro and micro policies © the author(s) 2023. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 12, no. 2 (2023), pages 119-138 https://doi.org/10.17979/ejge.2023.12.2.9600 submitted: apr 1, 2023 accepted: nov 9, 2023 published: dec 5, 2023 article unleashing entrepreneurial potential in transition economies: a comparative analysis of the impact of macro and micro policies fadil sahiti 1, * 1 rit kosovo (auk), rochester institute of technology, prishtina, republic of kosovo *correspondence: fsahiti@mail.rit.edu abstract. this study investigates the relationship between macroeconomic and microeconomic policies and entrepreneurial dynamics in two economies transitioning from planned economies to free markets, comparing them to a developed economy. macroeconomic policies, despite not directly targeting entrepreneurship, significantly impact entrepreneurial dynamics. conversely, microeconomic policies specifically aim to promote and enhance entrepreneurial activity. the analysis links policy quality to key entrepreneurial indicators: new firm creation, incumbent firm survival, and overall firm stock. findings reveal that while transition economies often adopt entrepreneurship policies similar to developed nations, some remain country-specific. these policy variations manifest in distinct entrepreneurial dynamics across the economies. keywords: entrepreneurship policy; transition economies, start-up jel classification: l53; p20; o57; l26 1. introduction entrepreneurship is widely recognized as an important factor affecting economic growth (acs & sanders, 2013; koski & pajarinen, 2013; acs & szerb, 2007). empirical evidence suggests that reduced entrepreneurial activity is likely to lead to reduced economic growth (audretsch et al., 2007). therefore, governments around the world try to create policies that lead to a greater dynamism of entrepreneurial activities. this is more visible in developed economies, in which economic policies help not only in the growth of entrepreneurial activities, but also in the growth of those enterprises that manage to survive the market (bartelsman et al., 2013). this phenomenon is less visible in developing economies and especially in those in transition, where the growth and expansion of firms is less visible, as many of them struggle to grow, age and remain small throughout their life cycle (estrin & mickiewicz, 2011; hsieh & klenow, 2014). kosovo and north macedonia are characterized by an enterprise population consisting mainly of small firms, which employ less than five employees. in contrast, advanced economies are characterized by a pronounced "up-or-out" culture, where a large proportion of new businesses are also small, but over time, many of them https://creativecommons.org/licenses/by-nc/4.0/ 120 fadil sahiti either develop and grow or are forced to exit the market (eslava et al., 2019). this paper investigates the interaction between macro and microeconomic policies geared towards entrepreneurial activity. macroeconomic policies refer to those that have a general impact on entrepreneurship, such as fiscal and monetary policies, tax policies, business regulations, and mechanisms aimed at regulating market competition, among others. microeconomic policies refer to policies that are more directly related to enhancing entrepreneurship, such as creating an entrepreneurial culture in society, facilitating access to financial funds, focusing on the development of technology and innovation, and advisory and assistance entrepreneurship policies. indicators for measuring entrepreneurial activity include the number of new market entrants, the stock of active firms and the survival rate of firms over five years. in the focus of the analysis of this paper are three countries that were once socialist economies and part of the same state the former yugoslavia. two of these economies (kosovo and north macedonia) are still in the process of transition from a socialist to a capitalist economy, while the benchmarking economy (slovenia) is now part of the eu. the comparative analysis was carried out with the aim of identifying those entrepreneurial policies that can explain the differences in entrepreneurship indicators and thus explain the different levels of economic development. this topic is motivated by a growing body of empirical evidence on the reduced level of entrepreneurship in less developed economies (bento & restuccia, 2020). the findings presented in this paper suggest that both economic policies and entrepreneurial activities in transition economies have experienced intense dynamics. specifically, our data suggest that some economic policies and entrepreneurship indicators of transition countries resemble advanced economies, but many others are idiosyncratic to these economies. unlike the slovenian economy, the economies of the transition countries are mainly populated with very small, non-innovative enterprises that survive for many years but fail to grow. as a result, the contribution of these companies to the economic development of these countries is very low. the data presented in this paper suggest that many new entrants do not survive and exit the market too early. however, provided that they manage to survive up to year four, the probability of surviving after five years is significantly higher, leading to reduced firm dynamism and market selection. although there are empirical works that have investigated the role of economic policies in entrepreneurial activities. however, their focus is concentrated either on a single country or the study is carried out from the perspective of a single policy or institution (cole et al., 2016). in the entrepreneurship literature, there are relatively few empirical studies that investigate the relationship between government policies and level of entrepreneurial activities in the context of late transition economies. therefore, the contribution of this paper is not only to provide new empirical evidence on government policies that aim to promote higher entrepreneurship in late transition countries, but also to improve our understanding of the role of these policies in the overall development of national economies. moreover, previous research studies rarely included such comprehensive indicators of economic policies and entrepreneurial activities, so this study is also a contribution to the diversification of research methodologies available in this field. finally, the results presented in this paper should provide a basis for further research on this important topic. the paper is organized as follows. section 2 reviews the literature and provides the theoretical background of the paper, followed by section 3, where we discuss the data and the unleashing entrepreneurial potential in transition economies methodology employed in the study. in section 4, we provide a short profile of transition economies. section 5 summarizes the main findings. section 6 provides concluding remarks. 2. literature review and theoretical perspective the theories and indicators used to explain the relationship between economic policies and entrepreneurship began to develop early, especially after the publications of schumpeter (1950), or kirzner (1973). the results presented so far show that the design of good policies and especially their successful implementation in practice can have a significant impact on increasing the level of entrepreneurship (figueroa-armijos & johnson, 2016). in almost all stages of entrepreneurship, from the entry of firms during the process of expansion or exit, the design of effective and efficient policies is crucial (oecd, 2017). some combination and similarity of these policies exist in almost all economies, advanced and less advanced (estrin and mickiewicz, 2011). in fact, why is it so important for governments to design specific policies that promote entrepreneurship? promotion of entrepreneurship through economic policies began at the end of the last century, especially after the 70s. according to audretsch and thurik (2000), until the 1980s, governments were focused on designing economic policies aimed at protecting the interests of large firms. there are reasons behind this. first, through the application of economies of scale, large firms were able to produce standardized products and thus successfully meet market needs. another reason had to do with the fact that during this period, the cooperation between unions, governments and firms was greater and more effective and thus, firms offered permanent employment to their employees (audretsch & thurik, 2000). meanwhile, small firms were considered out of fashion; with a small contribution to employment; less efficient than large firms; and not very innovative (baswell, 1973; audretsch, 2002; robson & haigh, 2008). audretsch and thurik call this period the period of the 'managed economy', in which business ownership is concentrated and the main concern of firms was how to mass produce products. at the end of the 70s of the last century, the view and attitude of policymakers towards the role of entrepreneurship in the economic development of a country began to change. the world economy was undergoing a significant transformation; it was undergoing a transformation process from a "managed" economy to an "entrepreneurial" economy. governments, especially in the most advanced countries, began to see entrepreneurship as an important factor influencing economic and social development (baumol, 1990). policymakers had already begun to look for the best ways to promote entrepreneurial activities. according to storey and green (2010) there are three reasons why this transformation has occurred. first, the impact of smaller firms on employment growth began to be supported both theoretically and empirically, particularly by empirical studies published by birch in 1981 and 1987. second, in addition to employment, the contribution of small firms began to be seen in other areas, especially in the contribution that these firms make to economic development (romer, 1990; audretsch et al., 2006). findings from various studies suggested that by increasing market competition, small firms affect a country's economic growth (autio et al., 1999; disney et al., 2003; bartelsman et al., 2013). the third reason is related to other sustainable benefits. for example, the products and services of small firms have the capacity to offer customers a wider range of solutions 122 fadil sahiti and variations for their needs; forming a small firm is a good alternative and solution in situations when people lose employment from other firms; small firms can also act as agents of social change by integrating environmental and social concerns into their business operations, for example (tracey & jarvis, 2007). there are other reasons why governments seek to design and implement policies that stimulate entrepreneurship. a reason that is often mentioned in the economic and business literature has to do with what is known as 'market failures'. the need for the creation and implementation of specific economic and entrepreneurial policies arises especially when there is evidence of a lack of fair competition in the market; when customers and suppliers are not properly informed about market conditions; when there is the presence of externalities, etc. according to the political economy approach, the number of reasons why policymakers formulate policies can be larger, including government elections, power relations, bureaucracy, equity, and distributional issues, and so on (coen & dannreuther, 2002). 2.1 the theoretical perspective there are two interconnected economic perspectives which this paper is based upon: development economics and neo-institutional economics. both theories argue that regardless of the level of development of an economy, the influence of the institutional factor on entrepreneurship is fundamental. according to the perspective of development economics, the process of entrepreneurship develops even in countries with unstable institutions, i.e., the process of entrepreneurship continues despite the lack of effective protection of business property (see e.g., nenova, 2004; powell et al., 2008). meanwhile, according to the perspective of neo-institutional economics, laws, rules, and social norms have a fundamental influence on the development of entrepreneurship and, consequently on the economic development of a country (see, e.g., north, 1990; keefer & shirley, 2000; rodrik et al., 2002). proponents of the neo-institutional economics perspective believe that institutions are an important driver of the distribution of entrepreneurial talent in any society, including transition economies. the term economy in transition in this paper refers to countries that are in the process of macroeconomic reform and that are changing from a state-led economy to a more market-led economy (svejnar, 2002). this transition process is comprehensive, as it covers the main aspects of society, such as market liberalization, macroeconomic stabilization, privatization of state economic assets, as well as legal and institutional reforms (kornai, 2008; berend, 2003). the term "late transition economy" refers to kosovo and north macedonia, which, even after more than 20 years as independent states, still have not managed to complete the transition process from socialist to capitalist economies (ebrd, 2018). the concept of entrepreneurship is taken from lundström and stevenson (2005, p. 42), who define entrepreneurship "... as a process where individuals become aware of business ownership as a possible option or alternative, develop business ideas, learn business processes of business to become an entrepreneur, and undertake the start-up and development of a business.... entrepreneurship can be found in both start-ups and growing businesses”. unleashing entrepreneurial potential in transition economies figure 1. conceptual framework. source: adapted from oecd (2008) and unctad (2012). it is worth noting that the focus of this study is the policies that affect entrepreneurial activities before and after the start of the business activity. pre-entrepreneurial policies address issues such as motivation, opportunities, and skills, with the aim of creating an encouraging institutional base for people to start their own businesses. moreover, entrepreneurship policies are designed to influence a wide range of areas, including education (primary, secondary, and tertiary), promotion of entrepreneurship in the media and society, reduction of administrative, legislative and regulations for the creation of a financial firm, and support for young entrepreneurs, for example (stevenson & lundström 2007, p. 105). on the other hand, the purpose of policies after the start of business activity includes issues related to the process of survival and growth of firms. among several theoretical frameworks that are applied today to investigate policies that limit or enable entrepreneurial activities, the economic variables adapted for this study are taken from the framework provided by oecd (2008) and unctad (2012). in addition, in our framework, we have also included empirical indicators of entrepreneurial performance (see figure 2). this framework assumes that the more advanced and contextual a country's entrepreneurship policies are, especially the more efficient their institutionalization in practice, the higher the level of entrepreneurial activities is expected to be. 124 fadil sahiti 3. data and methodology the data used in this study were provided by official sources, such as the kosovo statistics agency (ksa), the kosovo business registration agency (kbra), and the state statistics office (sso) for north macedonia. data for slovenia are mainly provided by eurostat and other government agencies. the main entrepreneurship indicators used in the analysis include the population of active firms, the number of firms entering the market for the first time, the number of firms leaving the market, the distribution of firms by size, the distribution of employment across firms, and the survival of firms up to 10 years. the data provided enable the identification of indicators related to firms’ birth rates, firms exit rates, survival, and risk probabilities, as well as employment rates. the definition of these indicators is based on international business demographic standards (oecd/eurostat, 2007: p 12 and 77). the data provided by kbra contains a sample of 73,206 new entrants to the business sector during this period, and 12,354 firms that have left the market during the same period. the method of comparative analysis used in the paper is widely used in scientific studies, especially in the social sciences (collier, 1990). this method is particularly useful when applied to identify and evaluate similarities and differences between different countries, as is the case in this paper where it is intended to identify differences in the field of policies and entrepreneurship dynamics between countries with different levels of economic development. however, this type of analysis is not easy, as it is not easy to obtain relevant data from relevant countries. therefore, the data used in this study have their limitations. thus, although our data are relatively comprehensive, they nevertheless contain some limitations. for example, from the data on entrepreneurship, we cannot distinguish firms that voluntarily exit the market or have gone bankrupt. the other data limit is related to the identification of the change of ownership of the firm during the activity if the firm has changed owners. on this last point, the data provided by kbra suggest that changes in firm ownership are nevertheless very rare and, in this respect, this does not constitute a major limitation. another disadvantage associated with the data includes the inability to identify mergers, or the fact that the only reliable measure of firm size is the number of employees, i.e., we do not possess other indicators of firm size. 4. a short profile of economies in transition both kosovo and north macedonia have been centrally planned economies and emerged as independent countries from the breakup of the former yugoslavia (estrin & uvalic, 2008). due to the war that happened in these countries, the economic transition process, compared to other former socialist countries, was longer and not easy (bartlett, 2008; uvalic, 2012; ebrd, 2018). both economies are classified by the world bank in the group of upper-middle-income countries. kosovo and north macedonia have come a long way since independence (n. macedonia in 1991 and kosovo in 2008) and have achieved remarkable successes in certain areas. the latter has been an eu candidate since 2005, while the former in 1916 has signed the ‘stabilisation and association agreement’ with the eu. despite the progress made in the transition process, the problems these economies still face are different and mainly inherited from the previous economic unleashing entrepreneurial potential in transition economies system (uvalic, 2012; ebrd,2018). both economies are characterized by relatively low productivity and slower growth prospects. compared to european union (eu) members, labour productivity in both economies is lower than almost a quarter of the average of eu member countries (world bank, 2021; ilo, 2021; makstat, 2019). due to the limited functioning of market mechanisms, the reallocation of resources from less productive sectors and firms to more productive ones has decreased (european commission, 2021). this shows the impotence of market mechanisms in allocating investment through the reallocation of capital and other inputs from unproductive sectors and firms to more productive sectors and firms (world bank, 2020). furthermore, the findings presented in this paper suggest that the level of entrepreneurship in transition countries is significantly lower than in the reference country slovenia. the entry rates are lower, and the stock of incumbent firms is significantly low compared to eu countries. the lack of strong dynamism in the economy is due to numerous structural challenges and problems, including insufficient access to finance, significant gaps in infrastructure, insufficient competition, a large informal economic sector, high levels of corruption, etc. (sahiti & smith, 2017). table 1. economic indicators, 2021. svn mkd kos size of population (million) 2.1 2.06 1.7 gni per capita (us$) (thousand) 28.3 6.2 5.1 gdp in $ (billion) 61.75 13.83 9.41 rate of unemployment (%) 4.74 16.2 20.7 informal economy (%) _ 17 31.0 interest landing rates 2.13 4.4 4.7 profit tax (% of commercial profits) 9.3 11 12.7 number of active firms (thousand) 150.5 72.9 40.6 number of firm entry (thousand) 17.3 6.3 10.2 source: world bank, eurostat and official statistical agencies 5. empirical findings 5.1 macroeconomic policies in this section we examine macroeconomic policies that may not have entrepreneurs as their primary focus, but their impact at the level of entrepreneurship can be considerable. the analysis includes traditional macroeconomic policies such as fiscal and monetary policies, policies that affect macroeconomic stability, political and institutional instability in the country, as well as other highimpact policies such as business regulations and infrastructure. monetary policies, especially policies implemented by the government in setting the lending rate base (the rates at which the central bank lends), can significantly affect entrepreneurial activities 126 fadil sahiti (parker, 1996). this influence is also suggested by our findings. the cost of loans is in correlation with the rates of new market entries as well as with the stock of existing firms. in other words, the lower the cost of credit, the higher the entry rates and the number of incumbent firms see figure 2. another factor that affects entrepreneurial activity is the stability of the macroeconomics (parker, 1996; stiglitz, 2000). the level of income is an indicator of this stability, as it can affect the total demand for goods and services and thus affect the level of entrepreneurship in an economy. this correlation between the level of income and the level of entrepreneurial activity is shown in our findings see figure 2. political instability can also affect entrepreneurship. previous findings (klapper et al., 2008) show how business entry is related to political instability. data provided by international organizations suggest that in terms of the rule of law, political stability, violence, and corruption, transition countries perform significantly lower than developed countries such as slovenia. the lower taxes applied in transition countries do not seem to influence the development of entrepreneurship in countries in transition see table 1 above. even in terms of business regulations, countries in transition can be compared with developed countries. in the 2020 world bank doing business index, both economies rank quite high: north macedonia is 17th and kosovo 57th (world bank, 2022). however, both economies face several problems that continue to undermine fair competition, private investment, and the growth of firms in general. in the recent beeps survey, 43% of firms identified courts as a major barrier to doing business (world bank, 2020a). another important problem has to do with unfair competition in the market, especially with business informality. in the world bank survey (beeps), respectively 63.4% and 54.5% of managers interviewed in the two countries see economic informality as a major obstacle to the growth of their business. figure 2. macroeconomic indicators and entrepreneurship. source: world bank (2021). 0 5 10 15 20 25 30 birth of firms gnic interest rate kos nmkd svn unleashing entrepreneurial potential in transition economies figure 3. institutional and governance indicators. source: world bank (2019). the quality of infrastructure roads, telecommunications, energy, and water supply also affects the level of entrepreneurial activity. our findings suggest that, despite some progress, countries in transition still lack some key infrastructure services. this is particularly evident in kosovo, which still faces unreliable electricity supply. in one of the world bank studies (beeps), 63% of all firms and 78% of manufacturing firms consider electricity supply to be a major problem (world bank, 2020). 5.2 microeconomic policies in this section, we examine government policies aimed at promoting higher entrepreneurial activity. four specific policies are the focus of our examination, which correspond to the micropolicy components provided in the framework presented earlier in figure 1. entrepreneurship culture. it is assumed that, through specific policies, governments try to promote 'entrepreneurial culture', because the expectation is that an individual with an 'entrepreneurial mindset' is more likely to one day start their own private business (european commission, 2003). the most traditional way to develop an entrepreneurial culture in people is when entrepreneurship as a subject is included in formal education curricula. this enables the development of more entrepreneurial skills and abilities in the younger generations and prepares them for the world of work. our findings indicate that policies that focus on entrepreneurship education are not coherent. for example, data provided by gem shows that the inclusion of entrepreneurship education in school curricula in transition countries is significantly lower than in schools in developed countries see table 2. also, the number and quality of training programs aiming to prepare young people to start 0 20 40 60 80 100 kos nmkd slo voice and accountability political stability and violence government effectiveness regulatory quality rule of law control of corruption 128 fadil sahiti a private business is significantly lower compared to slovenia, whose government finances a specific program called spot point (oecd, 2017). other organizations, such as the ibrd, report that economies in transition lag behind developed countries in the provision of systematic training of the workforce and in the index of the knowledge economy, among other factors see table below. theoretically, in both transition economies, vocational schools and training (vet) have been integrated into the education system, especially at the level of upper-secondary education. about 53% in kosovo and 60.2% in north macedonia of secondary school students choose vocational education – which is above the eu average of 47.8% (oecd, 2019). however, the vet systems in both countries suffer from numerous problems, such as the mismatch of curricula with market needs, the poor quality of teaching and learning, the lack of practical teaching materials, and especially the lack of practical work (internship) (oecd, 2017; european commission, 2019). unlike countries in transition, in slovenia, in addition to theoretical education, vocational schools are obliged to ensure that each student attends practical work education internship. slovenian law defines the status of students attending vocational schools, according to which each student is obliged to spend 50% of the time in school and the rest in practical work (sba fact sheet, 2019). the employment of vet graduates in slovenia is very high 84.5%, while there is no data for kosovo and north macedonia. access and provision of finance. for firms operating in transition economies, bank loans remain the main source of external financing, as project financing through equity capital is still negligible (oecd, 2017). in addition, due to difficulties in meeting loan collateral requirements, access to bank funds for startups and small firms is significantly more difficult. to facilitate access to credit, countries in transition have created loan guarantee schemes, which are financed by public funds and external donors. however, compared to similar schemes operating in developed countries, their number and financial capacity are significantly limited. another source of funding is business grants, which enable entrepreneurs to invest in equipment or training and consulting support for their managers and employees (storey and greene, 2010). there are institutions that offer business grants to entrepreneurs in transition countries, but their number and especially the financial capacity is much smaller compared to developed countries see the table below. supporting technology and innovation. entrepreneurship is often associated with technology and innovation. policymakers may see advantages in promoting the innovative skills of entrepreneurs, as in this way, they boost their country's economic growth. acs and audretsch (2003) emphasize that policy intervention in this area is necessary due to the advantages of spillover effects. this is why policymakers try to create mechanisms and allocate specific financial funds that promote the development of technology and innovation. the expectation is that through the diffusion of innovative ideas and technology, innovative entrepreneurs will positively influence other businesses and customers. in this section, we examine two examples of government policies: business incubators, as well as budget funds allocated for randd financing. unleashing entrepreneurial potential in transition economies table 2. microeconomic policies. entrepreneurship culture kos mkd slo entrepreneurship education at basic school (1-5)*1 1.8 2.06 2.07 entrepreneurship education at post secondary levels (1 5)*1 2.87 2.51 2.66 vet (% of total secondary education enrolment) 53 60.2 70.8 tertiary education (aged 30-34) (%) *2 20.8 35.7 44.9 knowledge economy index (total score out of 10)*3 3.22 4.5 6.65 staff training 1 7 (best)*4 2.7 3.2 4.5 finance loan-guarantee schemes kosovo credit guarantee fund development bank of north macedonia slovene enteprise fund (sef) business grants the millenium foundation kosovo fund for innovation and technological deveopment sef, employment service of slovenia, spirit slovenia equity financing from business angles gjirafa lab enif supported enteprise innovation fund sef, slovenian regional development fund, sid business bank, business angles of slovenia. technology and innovation business incubators innovation center kosovo (ick), gjirafa lab, the jakova innovation center, innovation and training park (itp) yes foundation, techpark which operates within the south east european university, ceed hub in skopje, center for technology transfer and innovation abc accelerator (200 start-ups in 6 years), reveris, kovačnica business incubator kranj, hekovnik startup school, incubator sežana, ljubljana university incubator, saša incubator share of budget in r&d 0.1 0.36 1.86 entrepreneurship awareness and networking agencies enterprise support agency (kiesa) agency for the promotion of entrepreneurship (apprm) slovenian public agency for entrepreneurship, innovation, development, investment and tourism (spirit) sources: entrepreneurship culture sources: *1 source: gem for kosovo (2014), world bank for n. macedonia & eu countries. *2 source: for kosovo european training foundation, for other countries eurostat. *3 source: ebrd knowledge economy index. *4 source: for kosovo european training foundation, for other countries wef. source for finance data: national banks and world bank. source for technology and innovation: government agencies for transition economies, and eurostat for eu countries. source for entrepreneurship awareness and networking: government agencies. the purpose of business incubators is to provide various conditions and services for start-ups and small firms, such as accommodation in shared offices, shared support services, professional business support and advice, network provision, etc. (messeghem et al., 2013). incubators offering similar services have also developed in transition countries. however, the number and specifically the scope and scale of their activities, compared to developed countries, is significantly smaller (oecd, 2017). another shortcoming of incubators in these countries is the lack of mechanisms that monitor and measure their practical impact, such as the number of start-up businesses supported or the 130 fadil sahiti evaluation of their innovative products or services. investment in randd encourages the development of new products and new business opportunities. our findings suggest that the amount of funding allocated to randd in transition countries is significantly lower compared to developed economies see table 2. entrepreneurship awareness and networking. both transition countries are in the initial stage of creating policies and institutions that promote entrepreneurial culture, as well as policies and other institutional mechanisms in providing business advice to young entrepreneurs. although north macedonia seems more advanced in this regard, kosovo lacks clear policies aimed at promoting entrepreneurship. agencies involved in promoting entrepreneurship are predominantly financed by donations (usaid, eu, giz, etc.), and less by public funds (oecd, 2017). 5.3 entrepreneurship performance the previous section has shown that macroeconomic policies (e.g., interest rates, national income, etc.) as well as microeconomic policies (entrepreneurial culture, access to finance, etc.), significantly affect entrepreneurial activities. this influence is manifested in specific indicators of entrepreneurship. for example, the stock of incumbent firms in transition countries is considerably lower compared to slovenia see table 1. moreover, for 5 years (2017 2021) this stock has not evolved at all. on the contrary, the number of existing firms in slovenia has increased by 4 thousand firms. another specific entrepreneurship feature of transition economies is the firm population structure. the proportion of firms with fewer than five employees is significantly higher in transition economies see table 2. this also applies to employment rates, with many small firms accounting for a higher proportion of employment. also, the data suggest that the structure of firms entering the market for the first time in transition economies is clearly dominated by micro-firms. these figures are not surprising given that as argued by many authors small firms are more prone to external uncertainties (geroski, 1995; rajan & zingales, 2003; bartelsman et al., 2013). firms operating in transition economies face significant political and economic uncertainty, and especially unfair competition (ebrd, 2018). it is evident that the quality, continuity, and orientation of the political regimes of transition economies affect the investment climate for many businesses. with regard to business regulations, evidence indicates that these economies have made important progress. however, based on the entrepreneurship indicators, it can be inferred that their impact is overestimated, just as it can be said about the impact of taxes. in both these areas, the economies in transition are not far behind the slovenian economy see table 1 and figure 3. data on micropolicies aimed at improving the business culture among younger generations show that countries in transition have made some progress. however, despite the large number of educational programs and their inclusion in the formal education system, there is little convincing evidence showing their effectiveness in improving the level of entrepreneurship. access to finance, whether through loans or equity, has also improved. yet, the data shows that access to external finance remains one of the main challenges of startups and small firms operating in these economies. unleashing entrepreneurial potential in transition economies lack of qualified human capital can also be one of the reasons for the low entry rate of large firms. according to oecd (2017) reports, entrepreneurs in kosovo and northern macedonia see the lack of a skilled workforce as one of the biggest obstacles to business growth. it is worth noting that most business firms in both transition countries are family-owned firms and are managed by family members, characterized by limited managerial and organizational skills (bloom & van reenen 2011; riinvest 2015; sahiti 2019). one of the reasons why these firms continue to be run by family members is related to the lack of trust and the rule of law, which makes owners reluctant to delegate management tasks or hire more capable and experienced managers (akcigit et al., 2021). finally, when the challenges and problems related to physical infrastructure are added to the mosaic of constraints described above, then it can be more clearly understood why entrepreneurs in transition countries tend to enter the market as small, or why the evolution of incumbent firms is almost non-existent, and particularly why micro-firms dominate the economy. table 3. size-class breakdown and number of employed individuals. svn mkd kos active firms: proportion of each size-class in total (%) 1-4 employees 77.0 90 93.9 5-9 employees 11.8 5.0 10+ employees 11.2 10 1.1 persons employed: proportion of each size-class in total (%) 1-4 employees 16.9 31.6 75.3 5-9 employees 9.1 10.89 10+ employees 74.0 68.4 13.81 firm entry: proportion of each size-class in total (%) 1-4 employees 93.0 _ 96.4 5-9 employees 5.2 3.0 10+ employees 1.8 _ 0.6 firm exit: proportion of each size-class in total (%) 1-4 employees 94.8 _ 97.5 5-9 employees 3.5 2.3 10+ employees 1.7 _ 0.2 source: ksa for kosovo, sso for n. macedonia and eurostat for slovenia 5.4 survival patterns: cross-country comparison in this section, we provide an overview of survival rates for firms entering the market for the first time and their probability of survival over a 5-year period, i.e., for the period 2014 2019. a firm born in year t is considered that managed to survive until t + 1 only if it is active. activity is measured through turnover and employment indicators in each part of the year t +1. the results in figure 4 show that the survival rates of firms in economies in transition for a three-year period are very similar to the survival rates of the reference country. however, if firms manage to survive or are active in the market for five years, the prospects of survival in transition economies improve significantly. this is especially true for the companies operating in the kosovo 132 fadil sahiti market, where about 53% manage to survive the market pressure, this percentage is significantly higher than that of slovenia, with about 46%. these findings show that market forces in transition countries are much weaker. these findings also suggest that the purpose of business operations for several firms in transition countries is survival, and not necessarily profit seeking (de soto, 2000; naudé et al., 2014). overall, this finding confirms one of the stylized facts proposed by bartelsman et al. (2013), that firms in transition economies, tend to experience better survival rates, which confirms the hypothesis that new entrants enjoy a period of relatively low market pressure especially in new, sparsely populated markets. figure 4. firm survival, age 1-5. source: eurostat and statistical agencies for kosovo and n. macedonia. 5.5 the determinants of new entrant firm survival in kosovo this section explores what determines the post-entry performance of firms in kosovo firms. we examine patterns of survival and explore the impact of explanatory variables in a non-parametric analysis. in this section are estimated kaplan-meier survival functions and tested for significant differences among survival functions across groups of firms, according to the different values for firm size, legal status and ownership, industry, and region. we conduct log-rank tests with no assumption of a particular survival time distribution. the weights are equal to 1 at all points in time; the focus is on large time values (hosmer & lemeshow, 2000). landau and everitt (2004) argue that this method (kaplan-meier) is the most common non-parametric one used to estimate the survival function. the kaplan-meier estimator of the survival function (or survival probability) s(t) = pr(t ≥ t) is: [1] where nj indicated the number of firms ‘at risk’ immediately before the j-th exit time (for every exit, business entities are censored at or after that time) and dj is the number of firm failures. overall, observed failure age is less than or equal to t. table 4 provides descriptive statistics, as well as the results of non-parametric survival 85 89 88 73 75 73 64 65 60 55 54,1 55 48 51 53 svn mkd kos age 5 age 4 age 3 age 2 age 1 unleashing entrepreneurial potential in transition economies analysis. the whole analysis is based on 73,206 new-born firms during the period 2010-2019. in this total, 12,354 firms had exited the market by the end of the period. column 4 shows that most newborn firms are very small (97.5% have 1-5 employees), of the individual owner type (84.5%), in the services sector (80.9%) and located mainly in the capital of the transition countries (56.4%). table 4. survival determinants using the kaplan-meier survival rate function. spell length (years) kaplan-meier survival rate log rank test** spells failure restrected extended firm's age (new-born firms) (new firm's death) mean mean* 1 2 3 number % total number failure rate % total firm size (n. employees) (1 4) 5.2 25.6 0.954 0.891 0.854 15.47 71,376 97.5 12,082 17.01 97.8 (5 9) 5.1 25.9 0.939 0.890 0.799 0.006 878 1.2 161 22.01 1.3 (+ 10) 5.5 44.8 0.988 0.910 0.890 952 1.3 111 12.2 0.9 ownership sole proprietor 5.3 24.1 0.944 0.892 0.834 475.11 0.001 61,859 84.5 11,390 17.5 92.2 ltd 5.7 88.1 0.989 0.971 0.956 8,565 11.7 420 4.8 3.4 general partnership 5.1 20.09 0.938 0.854 0.810 2,123 2.9 457 21.4 3.7 foreign company 5.5 46 0.986 0.924 0.901 659 0.9 87 10.9 0.7 sector industry 5.6 32.1 0.974 0.932 0.894 105.47 10,908 14.9 1,384 12.8 11.2 service 5.4 26.1 0.959 0.891 0.846 0.001 59,224 80.9 10,476 18.1 84.8 construction 5.5 39.2 0.988 0.934 0.910 3,074 4.2 494 4.9 4 region capital 5.5 32.1 0.964 0.902 0.864 340.11 0.00 25,549 34.9 3,805 14.3 30.8 dummy 1 5.2 23.1 0.949 0.951 0.826 12,591 17.2 2,336 18.9 18.9 dummy 2 5.4 34.9 0.968 0.914 0.880 15,739 21.5 2,211 12.6 17.9 dummy 3 5.2 16.9 0.946 0.854 0.799 11,201 15.3 2,594 22.5 21 dummy 4 5.4 25.1 0.952 0.884 0.840 8,126 11.1 1,408 17.9 11.4 total 5.4 27.4 0.963 0.906 0.853 73,206 100 12,354 16.6 100 notes: *if the longest follow-up time is censored, extended mean computes the mean survival by exponentially extending the survival curve to zero, and restricted mean computes the means survival time restricted to the longest follow-up time. if the longest follow-up time is a failure, the restricted mean survival time and the extended mean survival time are equal. **log rank test for the equality of the survival functions for each explanatory variable. source: kbra, ksa and sso. at first glance, it seems that the survival rate of new-born companies in transition economies is relatively high. this may be due to an effect of the market selection process or high sunk entry costs, which become high barriers to exit if re-entry is possible. on the other hand, it may also be due to the high unemployment rate that pushes people towards entrepreneurship and motivates them to fight for survival after entering the market. reynolds et al. (2005) suggest that a lack of job 134 fadil sahiti opportunities can make creating and maintaining a firm, important for personal survival. therefore, regardless of revenue levels, the entrepreneur’s personal circumstances might be one of the main reasons for the firm continuing in the market. 6. conclusions the purpose of this paper was to examine government macroeconomic and microeconomic policies and assess the impact they have on entrepreneurial activities in two transition economies. findings from these economies are compared with the slovenian economy. the assessment was conducted through three main indicators of entrepreneurship, namely market entry rates, the stock of incumbent firms, and the firm's survival prospects. the results presented in this paper show that economies in late transition have managed to draft a large number of policies at the macro and micro level, which promote entrepreneurial activities. the findings suggest that factors related to political and economic stability and especially the quality of institutional and governance factors have a strong influence on the level of entrepreneurial activity. this is not surprising given that entrepreneurs are more sensitive to external environmental uncertainties. although not directly, our findings suggest that the quality, consistency, and direction of a country's macro policies influence entrepreneurs' business investment decisions. faced with these difficulties, the level of entrepreneurial activity in transition economies is significantly lower compared to reference countries. significant progress has also been made in the drafting of micro policies aimed especially at increasing the level of entrepreneurship. however, entrepreneurship in transition countries faces challenges and problems related to the greater inclusion of entrepreneurship in educational programs of all three levels, the improvement of policies related to the easier access of small firms to financial funds, and the creation of policies aiming to enhance the development of technology and innovation, among others. the paper shows that some indicators of entrepreneurship in transition countries are similar to those of the reference economy. however, there are differences in at least two important indicators. first, the evolution of incumbent firms in transition countries is relatively stagnant, which means that the number of incumbent firms evolves significantly slowly. second, the data suggest that the population of firms in transition countries consists of very small firms, which remain small and relatively inefficient throughout their lifetime. most new entrants are small, and most do not grow above four employees. even in slovenia, most start-ups are small, but once they enter the market, they either exit or grow. interestingly, we found that the contribution of small firms to employment in transition countries is significantly higher compared to the reference country. in conclusion, countries in transition have managed to design entrepreneurship policies that are largely in line with those of developed countries. however, their impact on increasing the level of entrepreneurship for many reasons does not seem to be very pronounced. one possible reason may be that the design of these policies is based on a "me too" approach rather than any rigorous analysis of how these policies fit a specific business environment. this shows that entrepreneurship policies, no matter how good they are on paper, if they are not adapted to the specifics of an economy, are not enough to promote higher levels of entrepreneurship. other macroeconomic factors, such as unleashing entrepreneurial potential in transition economies the political and economic stability of a country, and especially the political regime, have a strong influence on regulating the investment climate for many entrepreneurs. therefore, entrepreneurial activities are not exclusively matters of macro or micro policies but are based on the interaction they have between them. finally, public policymakers can put entrepreneurs at the centre of their interest, but this may not be enough, as the entrepreneurial community is heterogeneous and has different interests. our humble recommendation is that policymakers, in this case those in transition countries, when designing entrepreneurship policies, should try to address the specific and real problems of entrepreneurs. the best approach is to create common forums where different business issues and perspectives can be discussed and incorporated. this approach would enable more integrated entrepreneurial policy choices. references acs, z.j., & sanders, m. (2013). knowledge spillover entrepreneurship in an endogenous growth model. small business economics, 41, 775-795. https://doi.org/10.1007/s11187-013-9506-8 acs, z.j., & szerb, l. (2007). entrepreneurship, economic growth, and public policy. small business economics, 28, 109-122. https://doi.org/10.1007/s11187-006-9012-3 acs, z.j. & audretsch, d.b. (2003). handbook of entrepreneurship research, dordrecht: klüwer. https://doi.org/10.1007/b105789 akcigit, u., harun, a., & michael, p. (2021). lack of selection and limits to delegation: firm dynamics in developing countries. american economic review, 111(1), 231-75. https://doi.org/10.1257/aer.20180555 audretsch, d.b., keilbach, m.c., & lehmann, e.e. (2006). entrepreneurship and economic growth. new york: oxford university press. https://doi.org/10.1093/acprof:oso/9780195183511.001.0001 audretsch, d., & thurik, a. (2000). capitalism and democracy in the 21st century: from the managed to the entrepreneurial economy. journal of evolutionary economics, 10, 17-34. https://doi.org/10.1007/s001910050003 audretsch, d.b. (2002). entrepreneurship: a survey of literature. brussels. european commission. audretsch, d. b., grilo, i., & thurik, a. r. (eds.). (2007). handbook of research on entrepreneurship policy. cheltenham, uk: edward elgar publishing. https://doi.org/10.4337/9781847206794 autio, e., wallenius, h., & arenius, p. (1999). finish gazelles: origins and impacts. paper presented at the icsb conference, naples, 20-23 june. bartelsman, e., haltiwanger, j., & scarpetta, s. (2013). cross-country differences in productivity: the role of allocation and selection. american economic review, 103, 305-334. https://doi.org/10.1257/aer.103.1.305 bartlett, w. (2008). europe's troubled region. economic development, institutional reform and social welfare in the western balkans. london and new york: routledge. https://doi.org/10.4324/ 9780203644898 baswell, j. (1973). the rise and decline of small firms. london. george allen and unwin. baumol, w.j. (1990). entrepreneurship: productive, unproductive, and destructive. journal of political economy, 98(5), 893-921. https://doi.org/10.1086/261712 bento, p., & restuccia, d. (2020). on average establishment size across sectors and countries. journal of monetary economics, 117, 220-242. https://doi.org/10.1016/j.jmoneco.2020.01.001 berend, i.t. (2003). in history derailed: central and eastern europe in the long nineteenth century (1st ed., pp. i-iv). university of california press. https://doi.org/10.1525/9780520932098 birch, d.l. (1981). who creates jobs? the public interest, 65, 3-14. birch, d.l. (1987). job creation in america: how our smallest companies put the most people to work, free press, new york. bloom, n., & van reenen, j. (2011) human resource management and productivity. in o. ashenfelter and d. card (eds.), handbook of labour economics (vol. 4b, chapter 19, pp. 1697-1769). san diego, ca: elsevier. https://doi.org/10.1016/s0169-7218(11)02417-8 https://doi.org/10.1007/s11187-013-9506-8 https://doi.org/10.1007/s11187-006-9012-3 https://doi.org/10.1007/b105789 https://doi.org/10.1257/aer.20180555 https://doi.org/10.1093/acprof:oso/9780195183511.001.0001 https://doi.org/10.1007/s001910050003 https://doi.org/10.4337/9781847206794 https://doi.org/10.1257/aer.103.1.305 https://doi.org/10.4324/9780203644898 https://doi.org/10.4324/9780203644898 https://doi.org/10.1086/261712 https://doi.org/10.1016/j.jmoneco.2020.01.001 https://doi.org/10.1525/9780520932098 https://doi.org/10.1016/s0169-7218(11)02417-8 136 fadil sahiti coen, d. & dannreuther, c. (2002). when size matters: europeanisation of large and sme business government relations. politique européenne, 7, 116-137. https://doi.org/10.3917/poeu.007. 0116 cole, r., cumming, d., & li, d. (2016). do banks or vcs spur small firm growth? journal of international financial markets, institutions and money, 41: 60-72. https://doi.org/10.2139/ssrn.2517291 collier, d. (1990) "comparative method in the 1990s, apsa-cp", newsletter of the apsa organized section in comparative politics, 9(1), 4-5. de soto, h. (2000) the mystery of capital: why capitalism triumphs in the west and fails everywhere else, new york, ny: basic books. disney, r., haskel, j., & heden, y. (2003). restructuring and productivity growth in uk manufacturing. the economic journal, 113(489), 666-694. https://doi.org/10.1111/1468-0297.t01-1-00145 ebrd (2018). the western balkans in transition: diagnosing the constraints on the path to a sustainable market economy. available at: www.ebrd.com/news/2018/towards-higher-productivity-in-thewestern-balkans.html eslava, m., haltiwanger, j.c. & pinzón, a. (2019). job creation in colombia vs the us: "up or out dynamics" meets the "life cycle of plants". technical report. nber working paper no. 25550. https://doi.org/10.3386/w25550 estrin, s., & uvalic m. (2008). from illyria towards capitalism: did labour-management theory teach us anything about yugoslavia and transition in its successor states?. comparative economic studies, 50, 663-96. https://doi.org/10.1057/ces.2008.41 estrin, s., & mickiewicz, t. (2011). institutions and female entrepreneurship. small business economics 37, 397-415. https://doi.org/10.1007/s11187-011-9373-0 european commission (2003). entrepreneurship in europe: green paper. brussels: enterprise publications. european commission (2019), 'kosovo 2019 report', commission staff working document. strasburg. https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/20190529-kosovoreport.pdf european commission (2021). kosovo factograph. european commission, brussels. https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/near_factograph_ kosovo.pdf figueroa-armijos, m., & johnson, t.g. (2016). entrepreneurship policy and economic growth: solution or delusion? evidence from a state initiative. small business economics, 47, 1033-1047. https://doi.org/10.1007/s11187-016-9750-9 geroski, p. (1995). what do we know about entry?' international journal of industrial organization, 13(1), 421-440. https://doi.org/10.1016/0167-7187(95)00498-x hosmer, d.w., & lemeshow, s. (2000). applied logistic regression. 2nd edition, john wiley and sons, inc., new york. https://doi.org/10.1002/0471722146 hsieh, t., & klenow, p.j. (2014). the life cycle of plants in india and mexico. quarterly journal of economics, 129 (3), 1035-1084. https://doi.org/10.1093/qje/qjs044 ilo (2021). strengthening social dialogue in north macedonia. international labour organization. https://www.ilo.org/budapest/what-we-do/projects/wcms_679543/lang--en/index.htm keefer, p., & shirley, m. (2000). formal versus informal institutions in economic development. in: claude ménard (ed.), institutions, contracts and organizations, chapter 10. edward elgar publishing. https://doi.org/10.4337/9781781952764.00021 kirzner, i. m. (1973). competition and entrepreneurship. chicago: the university of chicago press. klapper, l., amit, r., guillén, m.f., & quesada, j.m. (2008). entrepreneurship and firm formation across countries (the world bank policy research working paper 4313). washington, dc: world bank. https://doi.org/10.1596/1813-9450-4313 kornai, j. (2008). the great transformation of central eastern europe: success and disappointment. in: kornai, j., mátyás, l., roland, g. (eds). institutional change and economic behaviour. international economic association series. palgrave macmillan, london. https://doi.org/10.1057/ 9780230583429_1 koski, h., & pajarinen, m. (2013). the role of business subsidies in job creation of start-ups, gazelles and incumbents. small business economics, 41, 195-214. https://doi.org/10.1007/s11187-0129420-5 landau, s., & b. s.everitt (2004). a handbook of statistical analysis using spss. boca raton, fl: crc press. https://doi.org/10.1201/9780203009765 lundström, a., & stevenson, l. (2005). entrepreneurship policy theory and practice. new york: springer https://doi.org/10.1007/b104813 makstat (2019). makstat database: basic structural business indicators, by activity sector, by nkd rev 2, state statistical office. https://doi.org/10.3917/poeu.007.0116 https://doi.org/10.3917/poeu.007.0116 https://doi.org/10.2139/ssrn.2517291 https://doi.org/10.1111/1468-0297.t01-1-00145 http://www.ebrd.com/news/2018/towards-higher-productivity-in-the-western-balkans.html http://www.ebrd.com/news/2018/towards-higher-productivity-in-the-western-balkans.html https://doi.org/10.3386/w25550 https://doi.org/10.1057/ces.2008.41 https://doi.org/10.1007/s11187-011-9373-0 https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/20190529-kosovo-report.pdf https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/20190529-kosovo-report.pdf https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/near_factograph_kosovo.pdf https://ec.europa.eu/neighbourhood-enlargement/sites/near/files/near_factograph_kosovo.pdf https://doi.org/10.1007/s11187-016-9750-9 https://doi.org/10.1016/0167-7187(95)00498-x https://doi.org/10.1002/0471722146 https://doi.org/10.1093/qje/qjs044 https://www.ilo.org/budapest/what-we-do/projects/wcms_679543/lang--en/index.htm https://doi.org/10.4337/9781781952764.00021 https://doi.org/10.1596/1813-9450-4313 https://doi.org/10.1057/9780230583429_1 https://doi.org/10.1057/9780230583429_1 https://doi.org/10.1007/s11187-012-9420-5 https://doi.org/10.1007/s11187-012-9420-5 https://doi.org/10.1201/9780203009765 https://doi.org/10.1007/b104813 unleashing entrepreneurial potential in transition economies messeghem, k., sammut, s., & bakkali, c. (2013). business incubator. in: carayannis, e.g. (eds) encyclopedia of creativity, invention, innovation and entrepreneurship. springer, new york, ny. https://doi.org/10.1007/978-1-4614-3858-8_196 naudé, w.a., amorós, e., & cristi, o. (2014). surfeiting the appetite may sicken: entrepreneurship and the happiness of nations. small business economics, 42(3), 523-540. https://doi.org/10.1007/ s11187-013-9492-x nenova, t. (2004). private sector response to the absence of government institutions in somalia, world bank, washington, dc. north, d.c. (1990). institutions, institutional change and economic performance, cambridge university press, cambridge. https://doi.org/10.1017/cbo9780511808678 oecd (2008). measuring entrepreneurship. paris. statistics directorate. oecd (2017). financing smes and entrepreneurs 2017: an oecd scoreboard. paris: oecd oecd/eurostat (2007) eurostat-oecd manual on business demography statistics, oecd publishing, paris. https://doi.org/10.1787/9789264041882-en parker, s. c. (1996). a time series model of self-employment under uncertainty. economica, 63(251), 459475. https://doi.org/10.2307/2555017 powell, b., ford, r. & nowrasteh, a. (2008). somalia after collapse: chaos or improvement? journal of economic behavior and organization, 67, 657-670. https://doi.org/10.1016/j.jebo.2008.04.008 reynolds, p., bosma, n., autio, e., hunt, s., de bono, n., servais, i., lopez-garcia, p., et al. (2005). global entrepreneurshipmonitor: data collection design and implementation 1998-2003.small business economics, 24(3), 205-231. https://doi.org/10.1007/s11187-005-1980-1 rajan, r., & zingales, l. (2003) 'the great reversal: the politics of financial development in the 20th century, journal of financial economics, 69(1), 5 50. https://doi.org/10.1016/s0304405x(03)00125-9 riinvest institute (2015). governance guide for family-owned businesses in kosovo. https://www.riinvestinstitute.org/uploads/files/2016/september/20/governance_guide_f or_family_owned_businesses1474378025.pdf (accessed on 01 july 2021). robson, s., & haigh, g. (2008). first findings from the uk innovation survey 2007. economic & labour market review, 2, 47-53. https://doi.org/10.1057/elmr.2008.58 rodrik, d., subramanian, a., & trebbi, f. (2002). institutions rule: the primacy of institutions over geography and integration in economic development. cid working paper no. 97, center for international development, harvard university, cambridge, ma. https://doi.org/10.3386/ w9305 romer, p.m. (1990). endogenous technological change. journal of political economy, 98(5), s71-s102. https://doi.org/10.1086/261725 sahiti, f. (2019). the growth of firms in less developed countries: lessons learned from kosovo. palgrave macmillan, london. https://doi.org/10.1007/978-3-030-14821-8 sahiti, f., & smith, h.l. (2017). an application of growth diagnostics on the growth of firms: with evidence from kosovo firms. journal of innovation and entrepreneurship, 6(16), 1-24. https://doi.org/10.1186/s13731-017-0078-8 sba (2019). enterprise and industry european commission, brussels. available at: https://ec.europa.eu /docsroom/documents/32581/attachments/26/translations/en/renditions/native (accessed 7 may 2022). schumpeter, j.a. (1950). capitalism, socialism and democracy. new york: harper and row. stevenson, l., & lundström, a. (2007). dressing the emperor: the fabric of entrepreneurship policy. in: audretsch, d.b., grilo, i., & thurik, a.r. (eds), handbook of research on entrepreneurship policy, chapter 6. edward elgar publishing, 94 – 129. https://doi.org/10.4337/9781847206794.00010 stiglitz, j. (2000). unraveling the washington consensus. multinational monitor, 21(4), 13-17. storey, d.j., & greene, f.j. (2010). small business and entrepreneurship. pearson education limited. svejnar, j. (2002). transition economies: performance and challenges. journal of economic perspectives, 16 (1), 3-28. https://doi.org/10.1257/0895330027058 tracey, p., & jarvis, o. (2007). toward a theory of social venture franchising. entrepreneurship theory and practice, 31(5), 667-685. https://doi.org/10.1111/j.1540-6520.2007.00194.x unctad (2012). entrepreneurship policy framework and implementation guidance. geneva. uvalic, m. (2012). chapter 14 transition in southeast europe: understanding economic development and institutional change. https://doi.org/10.1057/9780230361836_15 world bank (2022). doing business report. world bank, washington dc. https://www.doingbusiness.org /content/dam/doingbusiness/media/annual-reports/english/db2019-report_webversion.pdf. world bank (2020), world development indicators, world bank databank website. https://databank. worldbank.org/source/world-development-indicators https://doi.org/10.1007/978-1-4614-3858-8_196 https://doi.org/10.1007/s11187-013-9492-x https://doi.org/10.1007/s11187-013-9492-x https://doi.org/10.1017/cbo9780511808678 https://doi.org/10.1787/9789264041882-en https://doi.org/10.2307/2555017 https://doi.org/10.1016/j.jebo.2008.04.008 https://doi.org/10.1007/s11187-005-1980-1 https://doi.org/10.1016/s0304-405x(03)00125-9 https://doi.org/10.1016/s0304-405x(03)00125-9 https://www.riinvestinstitute.org/uploads/files/2016/september/20/governance_guide_for_family_owned_businesses1474378025.pdf https://www.riinvestinstitute.org/uploads/files/2016/september/20/governance_guide_for_family_owned_businesses1474378025.pdf https://doi.org/10.1057/elmr.2008.58 https://doi.org/10.3386/w9305 https://doi.org/10.3386/w9305 https://doi.org/10.1086/261725 https://doi.org/10.1007/978-3-030-14821-8 https://doi.org/10.1186/s13731-017-0078-8 https://ec.europa.eu/docsroom/documents/32581/attachments/26/translations/en/renditions/native https://ec.europa.eu/docsroom/documents/32581/attachments/26/translations/en/renditions/native https://doi.org/10.4337/9781847206794.00010 https://doi.org/10.1257/0895330027058 https://doi.org/10.1111/j.1540-6520.2007.00194.x https://doi.org/10.1057/9780230361836_15 https://www.doingbusiness.org/content/dam/doingbusiness/media/annual-reports/english/db2019-report_web-version.pdf https://www.doingbusiness.org/content/dam/doingbusiness/media/annual-reports/english/db2019-report_web-version.pdf https://www.doingbusiness.org/content/dam/doingbusiness/media/annual-reports/english/db2019-report_web-version.pdf https://databank.worldbank.org/source/world-development-indicators https://databank.worldbank.org/source/world-development-indicators 138 fadil sahiti world bank (2020a). enterprise surveys, world bank, washington dc. https://www.enterprisesurveys .org/en/data/exploreeconomies/2019/kosovo world bank (2021). macro poverty outlook – kosovo. world bank, washington dc. https://pubdocs .worldbank.org/en/993701492011106034/mpo-ksv.pdf https://www.enterprisesurveys.org/en/data/exploreeconomies/2019/kosovo https://www.enterprisesurveys.org/en/data/exploreeconomies/2019/kosovo https://pubdocs.worldbank.org/en/993701492011106034/mpo-ksv.pdf https://pubdocs.worldbank.org/en/993701492011106034/mpo-ksv.pdf 1. introduction 2. literature review and theoretical perspective 3. data and methodology 4. a short profile of economies in transition 5. empirical findings 6. conclusions references european journal of government and economics 11(2), december 2022, 234-250 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 multilevel governance and smart specialization in eu regions: an evidence-based critical review eristian wibisono a, b, * a doctoral school of regional policy and economics, university of pecs, hungary. b member of poliss: https://poliss.eu. this project has received funding from the european’s union h2020 research and innovation programme under grant agreement no 860887 * corresponding author at: wibisono.tian@gmail.com abstract. this paper critically reviews the literature on multilevel governance issues in support of implementing smart specialization policies in eu regions. using an evidence-based critical review approach, key literature that draws on three critical concepts is explored: multilevel governance, regional innovation policy, and smart specialization in various governance conditions and diverse regional resources. the evidence reviewed points to the critical role of multilevel governance in implementing smart specialization. effective coordination mechanisms are essential building blocks to encounter the challenges of multilevel governance for smart specialization. more consequential, however, are substantial synergies that are solid, harmonious, and balanced among multi-stakeholders within institutions and across levels of government. this paper contributes to the limited literature on multilevel governance in support of the smart specialization policy. further studies considering different types of regions are recommended to enrich future literature. keywords. multilevel governance, regional innovation policy, smart specialization, european union. jel codes. o25, o38, r58. doi. https://doi.org/10.17979/ejge.2022.11.2.9004 1. introduction the lack of linkages between local institutions and weak governance in implementing regional innovation policies emerged long before the new concept of place-based policies, smart specialization, was introduced while on the other hand, institutional and governance characteristics are vital characteristics that can distinguish one region from another (isaksen and trippl 2017; camagni and capello 2017). local government structures are complex with various fragmentation of authority, dense development activities, and limited inter-institutional communication and coordination (lawson 2003; henning kroll 2015). smart specialization policy comes with a regional approach or regional diversity (morgan 2013), which is closely related to the capacity and capability of the region in realizing innovation policies that follow regional uniqueness and regional economic transformation goals. innovation policies are implemented at various levels of government. therefore, it is crucial to consider the enabling factors of multilevel governance, especially if the innovation policy involves links between levels of government. as stated by gonzález-lópez (2019), multilevel interactions https://creativecommons.org/licenses/by-nc/4.0/ mailto:wibisono.tian@gmail.com https://doi.org/10.17979/ejge.2022.11.2.9004 eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 235 in regional innovation policies should be considered to gain policy lessons that can be useful for creating better policies in the future. the most prominent challenge in multilevel governance is the interactions between different levels to develop effective innovation policy outcomes (guimón 2018). however, the question then arises whether the policy mix resulting from interactions between levels of government or the relationship of local policy instruments with other policy instruments can affect the eventual efficiency of innovation policy. in the smart specialization policy, there is a lot of skepticism and questioning about whether the new concept of smart specialization also considers administrative, institutional, and political characteristics in different types of regions. in other words, does every region with diverse characteristics have the same or different governance problems in implementing the smart specialization policy? this paper addresses multilevel governance issues and challenges in the context of smart specialization policies and critically reviews evidence and lessons learned in implementing innovation policy strategies in several eu regions. using a literature review approach, this paper is intended to enrich the study of smart specialization place-based innovation policy from the perspective of place-based multilevel governance. by critically reviewing some evidence in several eu regions regarding their smart specialization strategy (s3) development experiences, it is indicated that the role of government in smart specialization governance at the local/regional level is closely related to the role and participation of higher levels of government, following the multilevel governance approach. while these related studies are still evolving, the key points presented at the end of this paper are expected to motivate future studies. this paper applies a critical literature review using literature sources published in reputable journals. the research protocol began with the primary literature search using the keywords "multi-level, multilevel, governance, innovation, smart specialization, europe, and region" in the journal indexing databases web of science and scopus, limiting the scope to the years of the research (2000-2021). a comprehensive background of the literature is presented in two subsections in the second part of the paper, which discusses governance challenges in implementation and how the concept of multilevel governance is appropriate and can support the implementation of smart specialization. in the third section, a review of evidence from previous research is presented to demonstrate the success of good governance close to the concept of multilevel governance in supporting the successful implementation of smart specialization (s3) strategies in several eu regions. the fourth section presents the paper's conclusions. 2. multilevel governance in smart specialization policy: theoretical background governance challenges in smart specialization implementation the concept of smart specialization is driven by the productivity and economic transformation problems of eu member states, which are lagging behind the united states. according to scholars of this concept, such as foray et al. (2011) and mccann & ortega-argilés (2014), the problems eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 236 are more due to internal operational issues of the industrial sector at various technological levels rather than the lack of high-tech industrial sectors in european regions. developers of the concept of smart specialization emphasize the importance of new domain inventions and dynamic entrepreneurship in many future fields that new and more appropriate industrial policies must support. as a result, the involvement of the state or government in this "discovery" process is crucial. this concept upholds the principle that industrial policy should be born from academic recommendations that are then translated into practical policy recommendations (henning kroll 2017). since the smart specialization policy is practically reserved for european regions, and therefore guided by the european regional policy, the involvement of national public actors in this is very limited (dominique foray, david, and hall 2011; henning kroll 2015). it is a "placebased policy" (barca 2009). hence, the old productivity-oriented industrial policy at the national level has become more oriented toward regional socio-economic issues, which are still being studied (d foray et al. 2012; h kroll 2016; boschma 2014; capello and kroll 2016). in the political agenda of smart specialization, the ex-ante requirements of the european commission require regions to set regional investment priorities obtained through identifying new domains in the entrepreneurial discovery process (edp). the implementation of smart specialization in the regions has become an essential political agenda because the european structural and investment funds (esif) have to be accountable to regional political authorities with different capacities among the member states. there are several essential factors that, according to (henning kroll 2017), can cause smart specialization policies to be absorbed differently by each region, including variations in regional economic potential, multilevel governance of local governments, and the political culture of local governments. the multilevel governance context of smart specialization is essential to a region for administrative reasons. multilevel governance implies that policy actors who play a role in a particular policy process can play a role again in another policy process (hooghe, marks, and marks 2001; kuhlmann 2001; flanagan, uyarra, and laranja 2011), for example in the case of smart specialization policies. many regions in europe (such as some federal government regions in germany) have decided not to have their regional industrial policies and continue or implement strategic national industrial policies (henning kroll 2017). the discovery of new domains and complex edp processes undertaken at the national level and then adapted at the regional level has raised major questions about what the mix of national and regional policies looks like and how best practices for the financial management of european funding are implemented at the regional level. a further issue is the structural complexity of local government. local governments have complex structures with fragmented authorities, dense development activities, and limited inter-agency communication (lawson 2003). it can be concluded that, in addition to being fragmented, the government has limited internal and external coordination capacity (henning kroll 2015). regional innovation policies long before smart specialization was often associated with a lack of linkages between local institutions and weak governance of socio-economic systems (isaksen and trippl 2017). regional innovation policies should consider regional specificities, including the eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 237 governance and institutional character of the region (camagni and capello 2017). as a new placebased innovation policy, smart specialization puts forward a territorial approach to what is referred to as territorial diversity (morgan 2013). this territorial diversity is, of course, closely related to the region's capacity and ability to identify the region's economic potential that can truly describe the future development trajectory and structural changes of the region in line with the specificities of its geography. the place-based policy instruments promoted by smart specialization have opened flexible access to structural funds according to the context of regional specificities to support the economic transformation goals of the region. gonzález-lópez (2019) argues for the importance of considering the influence of multilevel governance on innovation policy in the context of regional innovation policy. his concept is based on the two main analytical tools for assessing learning in regional innovation policies that have been raised in the studies of lundvall & borrás (1998) and nauwelaers & wintjes (2008). innovation policies are often implemented at different levels of government. regarding discussing innovation policy in a regional context, it is crucial to consider the multilevel governance structures of different levels of government (regional, national, and sub-national) and their interactions. the key to this concept is, of course, strong coordination to achieve policy coherence. it also means that other levels of government (e.g., eu national or sub-national governments) are external knowledge channels or sources that strongly influence policy learning at the regional government level. previous scholars have expressed some skepticism about smart specialization in the context of regional and local governance, arguing that the policy concept of smart specialization is insufficient and incomplete to address administrative, institutional, and political constraints in specific regions. it also means many regional operational capacity issues in implementing smart specialization (capello and kroll 2016; iacobucci 2014; pugh 2014). concretely, the implementation of s3 in more developed or urban areas has received more attention than, for instance, in less developed regions or sparsely populated areas (spas). mccann, p., ortegaagiles, r. & foray, d. (2015, p. 68) mention that the main challenge for the last instance regions in implementing s3 is the limited critical mass of economic actors in these regions. boschma (2015) argues that regions with weak innovation capacity have less potential to diversify. on the other side, tödtling & trippl (2005) argue that due to regional diversity reasons, more underdeveloped regions tend to favor the easy way of copying innovation policy best practices in more prosperous regions. the study by sörvik et al. (2019) may provide further insights into regional innovation policy issues for smart specialization in spas with limited local actors and various regional governance, political and administrative matters. the different policy levels at the regional and national levels should not be understood as interchangeable but rather as interdependent and complementary (guimón 2018). multilevel governance refers to the proportional and tiered division of responsibilities in the process of policy planning and implementation across different administrative and territorial levels (benz and eberlein 1999; koschatzky and kroll 2009). the most prominent challenge in multilevel governance is the effectiveness of interactions between different levels to create efficient eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 238 innovation policy outcomes. a further challenge relates to the policy mix resulting from interactions and interdependent relationships with other policy instruments that can affect the efficiency of innovation policies (flanagan, uyarra, and laranja 2011; martins 2016). some studies show that the combination of policy mix and multilevel governance is believed to create a more effective innovation process locally than at other levels (karo 2012; howlett and rayner 2007). however, it is crucial to bear in mind that while this combination has the potential to increase the effectiveness and efficiency of regional innovation policies, coordination failures can vandalize the expected success. possible administrative governance problems include policy duplication and overlapping, poorly agreed prioritization, and policies that are not coherent and consistent with other policies (magro, navarro, and zabala‐iturriagagoitia 2014). in the smart specialization policy context, regions are encouraged to actively participate in determining priority domains according to their regional characteristics either as participatory partners in bridging national policies or independently developing policies and strategies independent of national authorities and implementing projects according to their resources (cohen, 2019; laranja et al., 2020). smart specialization is an evolving regional policy concept that draws on established concepts of governance and regional innovation systems (del carmen sánchez-carreira, gonzález-lópez, and varela-vázquez 2021; hassink and kiese 2021; gancarczyk, ujwary-gil, and gonzález-lópez 2021). this policy places great emphasis on mechanisms for sharing responsibility and proper coordination between various local actors and at various levels of government (moodie et al. 2021; larrea, estensoro, and pertoldi 2019; guzzo and perianez-forte 2019; marinelli, fernández sirera, and pontikakis 2021; serbanica 2021; magro and wilson 2019). the smart specialization strategy (s3) faces many challenges in its implementation. developing and underdeveloped countries or regions, in particular, with decentralized innovation policy structures, face more acute challenges due to the diversity of their innovation systems (trippl, zukauskaite, and healy 2019; papamichail, rosiello, and wield 2019; mccann and soete 2020; dominique foray 2019). this phenomenon is exacerbated by the problem of institutional instability and income inequality which are still the leading cases even in developed countries such as the us, uk, and several member countries of the european union (atkinson, casarico, and voitchovsky 2018; cassiolato and lastres 2000). guimón (2018) explores several issues related to the decentralization of science and innovation policies in several developing countries. two interesting things discussed in his study are the importance of competency sharing and proper coordination mechanisms at various levels of government. the division of local government competencies in developing smart specialization policies in the regions must be adequately distinguished from the central or national government. in addition, the proper coordination mechanism between levels of government must also be regulated to address various governance issues in the implementation of s3 (valdmaa, pugh, and müür 2021; ruhrmann, fritsch, and leydesdorff 2021; ghinoi et al. 2021; dominique foray, eichler, and keller 2021). eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 239 how does multilevel governance support the implementation of smart specialization? multilevel governance (mlg) can be understood as the active involvement of local innovation actors in developing or implementing policies through formal and informal mechanisms (tödtlingschönhofer et al. 2013; caponio 2021; ongaro et al. 2019; moodie et al. 2021). the definition of mlg put forward by schmitter (2004) is: "arrangements for making binding decisions that involve multiple politically independent but interdependent actors private and public at different levels of territorial aggregation in more or less continuous negotiation/ deliberation /implementation, and that do not assign exclusive policy competence or assert a stable hierarchy of political authority to any of these levels". barca (2009) defines mlg as a system where policies are designed, implemented, and distributed across different levels of government and local authorities or institutions according to their respective objectives. furthermore, serbanica & constantin (2017) underline the role of mlg in integrating various policies and activities at different levels of government, emphasizing the importance of synergy between all actors. in the context of eu policy, the concept of mlg has been used explicitly with the concept of a network approach or interrelated relationships without paying attention to hierarchical relationships. in this approach, mlg is described as "a collective decision-making process by multi-stakeholder authorities at different levels of governance and across different policy sectors" (“spatial foresight local and regional partners contributing to europe 2020” 2015). hooghe et al. (2001) classify mlg into two types: type 1, which promotes the concept of vertical governance relationships where decisions in a process are taken based on hierarchical relationships and geography, and type 2, which promotes more fluid horizontal relationships, which seems more appropriate in the context of smart specialization innovation policy. smart specialization foregrounds the relationships of the four elements of the quadruple helix and places importance on the horizontal interactions of various regional innovation actors in generating new domains in formulating the smart specialization strategy (s3). interactions between national and local/regional governments in the s3 development process may be limited to optimizing the role of local entrepreneurial actors who best understand local conditions. therefore, considering multilevel governance (mlg) in the implementation of smart specialization that involves the role of government at several levels (national, regional, and subregional) will potentially improve the effectiveness and efficiency of collaboration between them. on the one hand, national governments may have better competencies than regional governments, or regional governments may have better competencies than sub-regional governments. but on the other hand, it is not necessarily that the government at a higher level can always be present to handle many things at once when in a region, new domain opportunities arise in various fields or sectors. it then raises the question of the optimal governance mechanism in managing s3 at various levels and interconnecting horizontally and complementing each other (larrea, estensoro, and pertoldi 2019). the rationale behind smart specialization is a place-based approach that has the potential to be implemented in any region through at least two things: collaboration between different levels eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 240 of government with similar or different levels of knowledge for learning; and capitalizing on the local government's strong understanding and knowledge of its potential. thus, quoting from larrea et al. (2019), the main concept of mlg for smart specialization can be said to be a complex collaboration process between different government authorities to make s3 open and accessible to other actors. the mlg concept can help improve the implementation of s3 in accordance with the fundamental principles of smart specialization (cohen, 2021; joint research centre, institute for prospective technological studies, goenaga beldarrain, & foray, 2013). first, linking s3 priority programs with local investments enhances innovation capacity and local knowledge (level of granularity principle). second, it positions the government as a platform that guides the entrepreneurial discovery process (edp) in identifying and integrating local assets and knowledge. thirdly, s3 emphasizes the importance of monitoring and evaluation involving all stakeholders through relevant communication mechanisms and for learning purposes. the description above confirms that governance challenges emerge as an essential issue in the implementation of s3; thus, mlg also emerges convincingly. while multilevel coordination in implementing s3 is crucial, it is also essential to understand that mlg, in this case, is more of a "joint strategy creation process" where significant central government control over cohesion projects for local interests must also be balanced (marques & morgan, 2018). based on these arguments, it can be convinced that the smart specialization framework requires a significant role of local governments and actors, which the mlg concept could potentially support. in this regard, this paper will afterward focus on how the role of government at various territorial levels with a multilevel governance approach can support the implementation of smart specialization. 3. multilevel governance and smart specialization in eu regions: an evidence-based critical review this section aims to present evidence and a critical review of the various governance-related challenges in the implementation of s3 in eu regions and how the multilevel governance (mlg) approach could then potentially help improve the implementation of s3 in different regions in the eu. this section also intends to show that according to the experience and evidence from several eu regions, the smart specialization framework essentially has features that are mutually supportive with the mlg concept as discussed in the previous section. in the first critical review, i observe what has been discussed by kroll (2017), who takes a case study of the multilevel governance of regional innovation policy in two regions in germany, north rhine-westphalia (nrw) and saxony. as in germany's federal system, all länder have a very complex set of federal-regional duties per the country's constitution. the federal government handles innovation and technology policy in germany, and meaningful projects resulting from the entrepreneurial discovery process are also financed nationally without much regional involvement, especially in terms of budget. in short, the german federal government plays the most crucial role in innovation policy at the regional level. furthermore, within a complex eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 241 governmental structure, innovation and technology policy in germany involves three prominent authorities that deal with economics, science, and the chancellery. each of these entities has diverse resources and elements of interest within its organizational structure. effective coordination is the most substantive challenge in this regard. in the first case study of north rhine-westphalia (nrw), kroll (2017) emphasizes governance aspects. critical actors in the region play a significant role in regional innovation policy even though it is integrated into the national innovation policy. as the region is autonomously wellequipped with strategic facilities and resources, its regional innovation success is less dependent on the national investment (h kroll and meyborg 2013). north rhine-westphalia (nrw) has developed specialized policies in the long term as national policies cannot always facilitate the regional transformation challenges nrw. in addition, due to the region's large geographic and economic size, another challenge is the complexity of its administration. the nrw region may have managed to overcome budgetary issues and its administration. however, there is complex coordination and communication within the internal agendas of ministries and agencies. due to the challenges of the region's transformation, nrw was fully supported by european structural funding over a long period which causes nrw's regional administrative capacity has emerged as an essential factor in supporting regional innovation policy. the following case study from kroll (2017) study is the region of saxony, which represents the eastern part of germany. saxony joined the federal republic of germany in 1990, and since then, its focused economic policies have transformed its economy and made it the strongest in east germany. due to the remarkable legacy of the previous government system, the region has many qualified human resources. information technology (it) is emerging as the region's new leading industry, while the machinery industry inherited from the past is still a strength that continues to be promoted (zenker and kroll 2014). the entrepreneurial discovery process in many fields has been successful, but these new domains generally differ from the previously existing and thriving leading fields. in the process of successful economic transformation, saxony has greatly benefited from national investments that was supported by three critical success factors: public research institutes, leading universities, and clusters of regional (zenker and kroll 2014). local governments focus european structural fund for saxony on supporting the growth of mid-cap companies. although the geographic and economic size of the saxony region is much smaller than nrw, saxony is administratively challenged in its regional organization efficiency. the organization created after the reunification process is still relatively new, replacing the previous organization. in taking responsibility for making an excellent regional innovation climate, saxony only relies on two departments in a single ministry that coordinates relatively well with other ministries and agencies (henning kroll et al. 2016). saxony's government administratively is among germany's most professional and efficient. on the other hand, saxony has succeeded in developing the best strategy document that links its public investments directly into innovation and technology and the possible socio-economic impacts. in the process, the determination of this strategy should be based on scientific evidence eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 242 and regional economies' elements. with an already efficient organizational structure, substantive improvements are not so visible. in the ris3 process, saxony may differ slightly from nrw in that it cooperates with the european commission in accessing additional funds, which they can allocate to several administrative units in its region to develop better regional innovation strategies. the two case studies discussed by kroll (2017) show the uniqueness and significant progress of regions in germany in supporting innovation and technology through public funding. public investment can be accessed after going through a substantial long process, starting with forming a strategic policy derived from the entrepreneurial discovery process and finding new domains for public investment purposes, matching the character of the smart specialization place-based innovation policy. this success also seems to reflect the effectiveness and efficiency of aspects of good governance. there are similarities in the relevance of the national budget for these two case studies, namely, both accessing federal funding. but organizationally, the complexity of the two is very different. nrw is much more complex than saxony. it is mainly due to the very diverse size of the region's geography, economy, and population. it also causes the role of eu funding in the two regions to differ. there is eu funding involvement in nrw with substantial efforts from the region. but for saxony, eu funding plays a significant leading role in regional innovation policy. second, a recent study by gonzález-lópez (2019) explored the regional innovation system's evolution, policy learning, and influencing factors over two decades in galicia (spain). using a case study approach, he interviewed seven key actor’s instrumentals in the policy-making innovation process in the mandate period 1989-2017. the integrated innovation policy that is considered the primary milestone of innovation policy in galicia is estimated to have started in 1999, while the galician ris3 of 2014 was the primary innovation policy reference used at the time of the study. in terms of creating good governance and low-cost, flexible, and efficient bureaucracy, galicia created the galician innovation agency, the most significant institutional evolution in the region's innovation policy-making process. it was demonstrated by a smaller number of policy instruments but a larger budget in the hope that it would reduce bureaucratic costs and increase the impact of innovation policy. other innovations include implementing reforms in public procurement programs that require intensive interaction with the government's internal audit and financial management agencies. on the other hand, an apparent weakness in the innovation policy evolution over two decades in galicia has been the evaluation and monitoring of policies, which, although formally on the agenda of r&d planning documents, have not been earnestly implemented. in addition, the technology assessment and forecasting process, which has often involved universities and independent consultants, has not been optimally utilized. external study reports that are useful for policymaking are taken for granted as a document or work output but are not used sustainably in the policy planning process. in the last two decades, only the most recent period involved social and political participation in the region's design and implementation of innovation policy. it was due to a requirement from eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 243 the eu commission in the entrepreneurial discovery process that required the involvement of experts and various stakeholders in supporting the creation of the region's innovation policy. in this process, the galician government also interacts with the two levels of government above it, namely spain and europe. galicia has policies and government structures that the spanish government heavily influences. arguably many things are copied for normative or political reasons, making specific changes challenging to identify and tend to bring uncertainty. it has been a constant trend in spanish governance. multilevel governance channels are needed to adapt continuously to the country's changing political and legal situation. in addition to being strongly influenced by the central government, galicia's innovation policy learning is also affected by innovation policy at the eu level. innovation policy strategies that follow the european commission's guidelines are highly relevant to the galician innovation policy system. other essential lessons also come from other eu regions or member states. simple imitation may still produce positive results if policymakers always consider the peculiarities of the region. the design and implementation of ris3 in the last period of galician governance demonstrated the efforts of many stakeholders in the innovation policy formulation process, which should be maintained in the future. this is the first time that galicia's innovation policy process has involved systematic socio-political participation. the concept of smart specialization innovation policy proved to be very advantageous in the evolution of innovation policy in galicia. third, i discuss the study by sörvik et al. (2019), which explored the application of s3 in sparsely populated areas in five regions across five european countries using a comparative case study approach. the study addresses critical issues related to s3 in spa regions in europe based on five key dimensions of the smart specialization concept. the first evidence shows that spa regions have undergone a meaningful transition in the regional innovation policy-making process since the advent of the s3 framework. this transition is embedded in the spa regions in the form of a change in the region's paradigm towards innovation, making them more supportive of the region's potential and more future-oriented. they also quickly accepted the entrepreneurial discovery process (edp) paradigm as the primary basis for smart specialization policy. the emergence of s3 as an innovation policy strategy reconstructed the mindset of local governments in spa in managing their region's resources. the s3 also increased the participation of various stakeholders in the policy-making process. for example, as happened in nordland (norway), which runs a green industry project in producing renewable energy by synergizing with various sectors or as happened in lapland (finland), where the government involves multipartners in building innovation policy strategies that focus on the birth of strategic smes as a new growth element of the region by improving ict access which they have been facing due to their geographical position. the implementation of s3 in the spa region was carried out by prioritizing innovation in the traditional primary resources of the region. however, the challenges become substantive in several respects, including multilevel governance, quality of human resources, new sources of knowledge, and expansion of channels. this issue is essential in multilevel governance, given that spa areas generally have limited administrative resources and institutional thinness. eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 244 program synergies and strategies become more efficient due to good coordination between institutions. in addition, the government is also faced with the problem of conflict of interest in multilevel governance. at the spa region level, it is pretty challenging to mobilize stakeholders at the national level to engage in their regional activities. one of the best suggestions by moodysson et al. (2015) is aligning spa region policies with other higher-level policies where there are many sectoral regulations that may be relevant for different levels of government. for example, national regulations on environmental sustainability or programs related to renewable energy are the kind of sectoral regulations that can be implemented even at the lowest level of government. this kind of integration is also critical to create synergies within the spa region and between levels of government. as sörvik et al. (2019) suggested, the european social fund could be considered to improve the competencies of spa regions in addition to aligning and enhancing synergies between rural policies and smart specialization policies. 4. conclusion governance in innovation policy is widely recognized knowledge. however, there needs to be more literature on how implementing the new place-based innovation policy, smart specialization, can be improved through multilevel governance (mlg) efforts. this paper critically examines the governance challenges in implementing smart specialization in the eu region using an evidence-based review approach. these fundamental issues are firstly raised and then followed by presenting supporting literature on how multilevel governance (mlg) is considered can support the implementation of smart specialization at the local/regional level. some of the evidence reviewed shows that, firstly, regions with a strong historical background that have managed to transform significantly show a vital role of national authorities (such as the cases of nrw and saxony in germany). in creating regional innovation strategies in line with smart specialization principles, regions with sufficient resources to discover new domains through the entrepreneurial discovery process (edp) are controlled by the national (federal) government with a balanced presence of local government authorities. in this case, mechanisms for effective coordination and communication between levels of authority are crucial. secondly, regional authorities' most frequent difficulties are related to monitoring and evaluating policy implementation in regions where innovation policies are well-established and integrated nationally (such as the case of galicia in spain). involving multi-stakeholders from the edp process to the monitoring and evaluation stage of the smart specialization strategy (s3) can advance its effectiveness. intensive interaction between institutions and levels of government can serve as a learning channel to absorb external knowledge and experience so that governance at the local/regional level can continuously adapt to changes at the country or eu level. thirdly, regions with significant limitations in governance and institutions for geographic and demographic reasons (such as sparsely populated areas (spas) in nordland-norway and lapland-finland) can still develop their innovation strategies based on their regional uniqueness through harmonized eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 245 innovation projects. effective coordination between local institutions is undoubtedly an important asset, and at the same time, substantial synergies between levels of government can optimize the impact of these projects. this evidence review clarifies that the smart specialization principles implemented at the subnational level are not without challenges. this paper intends to highlight that governance challenges in implementing smart specialization at the sub-national level are closely related to the role of government across levels. evidence also suggests that effective coordination mechanisms can help address governance issues at the sub-national level. however, concerning multilevel governance (mlg), the main emphasis in this concept is not only on the coordination perspective but also on solid, harmonious, and balanced synergies among multi-stakeholders at different levels of government. finally, it is premature to claim that the points raised in this paper conclusively show how multilevel governance is worth considering in implementing smart specialization policies. studies on this subject are still to be developed, and the critical review in this paper is expected to be one of the motivating ones. furthermore, the apparent limitation of this paper is that the selection of region types in presenting the evidence review may not be equivalent. such a composition may have led to accuracy and precision problems in drawing conclusions. the generalization of the types of regions has been applied in the presentation of the paper. future research could examine the phenomenon of stratified governance based on different types of regions, for instance, the less developed regions of the european union. acknowledgments the author gratefully acknowledges the support provided by poliss (https://poliss.eu), the funded project under european union’s h2020 research and innovation programme, grant agreement no. 860887. views and opinions are only those of the author and do not necessarily reflect those of the european union. neither the european union nor the granting authority can be held responsible for them. the author would also like to thank professor tamás sebestyén, professor ilona pálné kovács, and anonymous reviewers for their constructive criticism and suggestions. eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 246 references atkinson, a. b., casarico, a., & voitchovsky, s. (2018). top incomes and the gender divide. journal of economic inequality, 16(2), 225-256. https://doi.org/10.1007/s10888-018-9384-z barca, f. (2009). agenda for a reformed cohesion policy. european communities brussels. benz, a., & eberlein, b. (1999). the europeanization of regional policies: patterns of multilevel governance. journal of european public policy, 6(2), 329-348. https://doi.org/10.1080/ 135017699343748 boschma, r. (2014). constructing regional advantage and smart specialisation: comparison of two european policy concepts. constructing regional advantage and smart specialisation: comparison of two european policy concepts, 51-68. https://doi.org/10.3280/scre2014001004 boschma, r. (2015). towards an evolutionary perspective on regional resilience. regional studies, 49(5), 733-751. https://doi.org/10.1080/00343404.2014.959481 camagni, r., & capello, r. (2017). regional innovation patterns and the eu regional policy reform: towards smart innovation policies. in seminal studies in regional and urban economics (pp. 313-343). springer. https://doi.org/10.1007/978-3-319-57807-1_16 capello, r., & kroll, h. (2016). from theory to practice in smart specialization strategy: emerging limits and possible future trajectories. european planning studies, 24(8), 1393-1406. https://doi.org/10.1080/09654313.2016.1156058 caponio, t. (2021). governing migration through multi‐level governance? city networks in europe and the united states. jcms: journal of common market studies, 59(6), 1590-1606. https://doi.org/10.1111/jcms.13214 cassiolato, j. e., & lastres, h. m. m. (2000). local systems of innovation in mercosur countries. industry and innovation, 7(1), 33-53. https://doi.org/10.1080/713670250 cohen, c. (2019). implementing smart specialisation: an analysis of practices across europe. publications office of the european union, luxembourg, doi, 10, 759464. cohen, s. (2021). reflections guiding smart specialisation strategies impact assessment. joint research centre (seville site). del carmen sánchez-carreira, m., gonzález-lópez, m., & varela-vázquez, p. (2021). the implementation of the smart specialization approach in the peripheral region of galicia. in partnerships for regional innovation and development (pp. 49-75). routledge. https://doi.org/10.4324/9780429330261-4 flanagan, k., uyarra, e., & laranja, m. (2011). reconceptualising the 'policy mix'for innovation. research policy, 40(5), 702-713. https://doi.org/10.1016/j.respol.2011.02.005 foray, d. (2019). on sector-non-neutral innovation policy: towards new design principles. journal of evolutionary economics, 29(5), 1379-1397. https://doi.org/10.1007/s00191-018-0599-8 foray, d., david, p. a., & hall, b. h. (2011). smart specialisation from academic idea to political instrument, the surprising career of a concept and the difficulties involved in its implementation. in mtei-working_paper-2011-001. https://doi.org/10.1007/s10888-018-9384-z https://doi.org/10.1080/135017699343748 https://doi.org/10.1080/135017699343748 https://doi.org/10.3280/scre2014-001004 https://doi.org/10.3280/scre2014-001004 https://doi.org/10.1080/00343404.2014.959481 https://doi.org/10.1007/978-3-319-57807-1_16 https://doi.org/10.1080/09654313.2016.1156058 https://doi.org/10.1080/713670250 https://doi.org/10.4324/9780429330261-4 https://doi.org/10.1016/j.respol.2011.02.005 https://doi.org/10.1007/s00191-018-0599-8 eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 247 foray, d., eichler, m., & keller, m. (2021). smart specialization strategies-insights gained from a unique european policy experiment on innovation and industrial policy design. review of evolutionary political economy, 2(1), 83-103. https://doi.org/10.1007/s43253-020-00026-z foray, d., goddard, j., goenaga beldarrain, x., landabaso, m., mccann, p., morgan, k., nauwelaers, c., & ortega-argilés. r. (2012). guide to research and innovation strategies for smart specialisations. apo.org.au. https://apo.org.au/node/90736 gancarczyk, m., ujwary-gil, a., & gonzález-lópez, m. (2021). the expansion of the smart specialization concept and practice. in partnerships for regional innovation and development (pp. 1-18). routledge. https://doi.org/10.4324/9780429330261-1 ghinoi, s., steiner, b., makkonen, t., & hassink, r. (2021). smart specialisation strategies on the periphery: a data-triangulation approach to governance issues and practices. regional studies, 55(3), 402-413. https://doi.org/10.1080/00343404.2020.1791321 tödtling-schönhofer, h., radzyner, a., frangenheim, a., bachtler, j., mendez, c., charles, d., & granqvist, k. (2013). an assessment of multilevel governance in cohesion policy, 2007-2013. gonzález-lópez, m. (2019). understanding policy learning in regional innovation policies: lessons from the galician case. innovation: the european journal of social science research, 32(1), 104-118. https://doi.org/10.1080/13511610.2018.1519780 guimón, j. (2018). decentralisation of science and innovation policies in emerging countries: three key challenges that deserve further attention. international journal of technological learning, innovation and development, 10(1), 69-87. https://doi.org/10.1504/ijtlid.2018. 091801 guzzo, f., & perianez-forte, i. (2019). smart specialisation at work: evidence from the peer and exchange and learning workshops. joint research centre (seville site). hassink, r., & kiese, m. (2021). solving the restructuring problems of (former) old industrial regions with smart specialization? conceptual thoughts and evidence from the ruhr. review of regional research, 41(2), 131-155. https://doi.org/10.1007/s10037-021-00157-8 hooghe, l., marks, g., & marks, g. w. (2001). multilevel governance and european integration. rowman & littlefield. howlett, m., & rayner, j. (2007). design principles for policy mixes: cohesion and coherence in 'new governance arrangements.' policy and society, 26(4), 1-18. https://doi.org/10.1016/ s1449-4035(07)70118-2 iacobucci, d. (2014). designing and implementing a smart specialisation strategy at regional level: some open questions. scienze regionali. https://doi.org/10.3280/scre2014-001006 isaksen, a., & trippl, m. (2017). exogenously led and policy-supported new path development in peripheral regions: analytical and synthetic routes. economic geography, 93(5), 436-457. https://doi.org/10.1080/00130095.2016.1154443 joint research centre, institute for prospective technological studies, goenaga beldarrain, x., & foray, d. (2013). the goals of smart specialisation, publications office. https://data.europa .eu/doi/10.2791/20158 https://doi.org/10.1007/s43253-020-00026-z https://apo.org.au/node/90736 https://doi.org/10.1080/00343404.2020.1791321 https://doi.org/10.1080/13511610.2018.1519780 https://doi.org/10.1504/ijtlid.2018.091801 https://doi.org/10.1504/ijtlid.2018.091801 https://doi.org/10.1007/s10037-021-00157-8 https://doi.org/10.1016/s1449-4035(07)70118-2 https://doi.org/10.1016/s1449-4035(07)70118-2 https://doi.org/10.3280/scre2014-001006 https://doi.org/10.1080/00130095.2016.1154443 https://data.europa.eu/doi/10.2791/20158 https://data.europa.eu/doi/10.2791/20158 eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 248 karo, e. (2012). modernizing governance of innovation policy through 'decentralization': a new fashion or a threat to state capacities? innovation, 14(4), 495-509. https://doi.org/10.5172/impp.2012.14.4.495 koschatzky, k., & kroll, h. (2009). multilevel governance in regional innovation systems. ekonomiaz. revista vasca de economía, 70(01), 132-149. kroll, h. (2015). efforts to implement smart specialization in practice-leading unlike horses to the water. european planning studies, 23(10), 2079-2098. https://doi.org/10.1080/09654313 .2014.1003036 kroll, h. (2016). policy brief on smart specialisation. occasional report. karlsruhe: fraunhofer. https://www.isi.fraunhofer.de/content/dam/isi/dokumente/ccp/2016/kroll_2016_policy_brief_ on_smart_specialisation.pdf kroll, h. (2017). smart specialization policy in an economically well-developed, multilevel governance system. in advances in the theory and practice of smart specialization (pp. 99123). elsevier. https://doi.org/10.1016/b978-0-12-804137-6.00005-x kroll, h., böke, i., schiller, d., & stahlecker, t. (2016). bringing owls to athens? the transformative potential of ris3 for innovation policy in germany's federal states. european planning studies, 24(8), 1459-1477. https://doi.org/10.1080/09654313.2016.1159666 kroll, h., & meyborg, m. (2013). regional innovation monitor plus, regional innovation report north rhine-westphalia. technopolis group, brussels. kuhlmann, s. (2001). future governance of innovation policy in europe-three scenarios. research policy, 30(6), 953-976. https://doi.org/10.1016/s0048-7333(00)00167-0 laranja, m., edwards, j., pinto, h., & foray, d. (2020). implementation of smart specialisation strategies in portugal: an assessment. luxembourg, publications office of the european union. larrea, m., estensoro, m., & pertoldi, m. (2019). multilevel governance for smart specialisation: basic pillars for its construction. publications office of the european union. doi, 10, 898433. lawson, t. (2003). institutionalism: on the need to firm up notions of social structure and the human subject. journal of economic issues, 37(1), 175-207. https://doi.org/10.1080 /00213624.2003.11506562 lundvall, b.-å., & borrás, s. (1998). the globalising learning economy: implications for innovation policy. magro, e., navarro, m., & zabala‐iturriagagoitia, j. m. (2014). coordination‐mix: the hidden face of sti policy. review of policy research, 31(5), 367-389. https://doi.org/10.1111/ropr.12090 magro, e., & wilson, j. r. (2019). policy-mix evaluation: governance challenges from new placebased innovation policies. research policy, 48(10), 0-1. https://doi.org/10.1016/j.respol. 2018.06.010 marinelli, e., fernández sirera, t., & pontikakis, d. (2021). towards a transformative smart specialisation strategy: lessons from catalonia, bulgaria and greece. publications office of the european union, luxembourg. https://doi.org/10.5172/impp.2012.14.4.495 https://doi.org/10.1080/09654313.2014.1003036 https://doi.org/10.1080/09654313.2014.1003036 https://www.isi.fraunhofer.de/content/dam/isi/dokumente/ccp/2016/kroll_2016_policy_brief_on_smart_specialisation.pdf https://www.isi.fraunhofer.de/content/dam/isi/dokumente/ccp/2016/kroll_2016_policy_brief_on_smart_specialisation.pdf https://doi.org/10.1016/b978-0-12-804137-6.00005-x https://doi.org/10.1080/09654313.2016.1159666 https://doi.org/10.1016/s0048-7333(00)00167-0 https://doi.org/10.1080/00213624.2003.11506562 https://doi.org/10.1080/00213624.2003.11506562 https://doi.org/10.1111/ropr.12090 https://doi.org/10.1016/j.respol.2018.06.010 https://doi.org/10.1016/j.respol.2018.06.010 eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 249 marques, p., & morgan, k. (2018). the heroic assumptions of smart specialisation: a sympathetic critique of regional innovation policy. in: isaksen, a., martin, r., trippl, m. (eds) new avenues for regional innovation systems theoretical advances, empirical cases and policy lessons. springer, cham. https://doi.org/10.1007/978-3-319-71661-9_14 martins, j. t. (2016). relational capabilities to leverage new knowledge. the learning organization, 23(6), 398-414. https://doi.org/10.1108/tlo-03-2016-0022 mccann, p., & ortega-argilés, r. (2014). the role of the smart specialisation agenda in a reformed eu cohesion policy. the role of the smart specialisation agenda in a reformed eu cohesion policy, 15-32. https://doi.org/10.3280/scre2014-001002 mccann, p., ortega-agiles, r. & foray, d. (2015). smart specialization and european regional development policy. in david b. audretsch, albert n. link, and mary walshok (eds), the oxford handbook of local competitiveness, oxford handbooks. https://doi.org/10.1093 /oxfordhb/9780199993307.013.26 mccann, p., & soete, l. (2020). place-based innovation for sustainability. publications office of the european union: luxemburg. moodie, j. r., wøien meijer, m., salenius, v., & kull, m. (2021). territorial governance and smart specialisation: empowering the sub-national level in eu regional policy. territory, politics, governance, 1-21. https://doi.org/10.1080/21622671.2021.1905542 moodysson, j., trippl, m., & zukauskaite, e. (2015). policy learning and smart specialization. balancing policy change and continuity for new regional industrial paths. https://doi.org/10.1093/scipol/scw071 morgan, k. (2013). the regional state in the era of smart specialisation. ekonomiaz: revista vasca de economía, 83, 102-125. nauwelaers, c., & wintjes, r. (2008). innovation policy, innovation in policy: policy learning within and across systems and clusters. innovation policy in europe. measurement and strategy, edward elgar, cheltenham, 225-268. ongaro, e., gong, t., & jing, y. (2019). toward multilevel governance in china? coping with complex public affairs across jurisdictions and organizations. public policy and administration, 34(2), 105-120. https://doi.org/10.1177/0952076718799397 papamichail, g., rosiello, a., & wield, d. (2019). capacity-building barriers to s3 implementation: an empirical framework for catch-up regions. innovation the european journal of social science research, 32(1), 66-84. https://doi.org/10.1080/13511610.2018.1537844 pugh, r. e. (2014). 'old wine in new bottles'? smart specialisation in wales. regional studies, regional science, 1(1), 152-157. https://doi.org/10.1080/21681376.2014.944209 ruhrmann, h., fritsch, m., & leydesdorff, l. (2021). synergy and policy-making in german innovation systems: smart specialisation strategies at national, regional, local levels? regional studies, 1-12. https://doi.org/10.1080/00343404.2021.1872780 schmitter, p. c. (2004). neo-neofunctionalism. na. serbanica, c. (2021). smart specialisation, peripheries and eu cohesion policy. eu cohesion policy and spatial governance. https://doi.org/10.4337/9781839103582.00023 https://doi.org/10.1007/978-3-319-71661-9_14 https://doi.org/10.1108/tlo-03-2016-0022 https://doi.org/10.3280/scre2014-001002 https://doi.org/10.1093/oxfordhb/9780199993307.013.26 https://doi.org/10.1093/oxfordhb/9780199993307.013.26 https://doi.org/10.1080/21622671.2021.1905542 https://doi.org/10.1093/scipol/scw071 https://doi.org/10.1177/0952076718799397 https://doi.org/10.1080/13511610.2018.1537844 https://doi.org/10.1080/21681376.2014.944209 https://doi.org/10.1080/00343404.2021.1872780 https://doi.org/10.4337/9781839103582.00023 eristian wibisono / european journal of government and economics 11(2), december 2022, 234-250 250 serbanica, c., & constantin, d. l. (2017). sustainable cities in central and eastern european countries. moving towards smart specialization. habitat international. https://doi.org/10.1016/ j.habitatint.2017.03.005 sörvik, j., teräs, j., dubois, a., & pertoldi, m. (2019). smart specialisation in sparsely populated areas: challenges, opportunities and new openings. regional studies, 53(7), 1070-1080. https://doi.org/10.1080/00343404.2018.1530752 spatial foresight local and regional partners contributing to europe 2020. (2015). tödtling, f., & trippl, m. (2005). one size fits all? towards a differentiated regional innovation policy approach. research policy, 34(8), 1203-1219. https://doi.org/10.1016/j.respol. 2005.01.018 trippl, m., zukauskaite, e., & healy, a. (2019). shaping smart specialization: the role of placespecific factors in advanced, intermediate and less-developed european regions. regional studies. https://doi.org/10.1080/00343404.2019.1582763 valdmaa, k., pugh, r., & müür, j. (2021). challenges with strategic placed-based innovation policy: implementation of smart specialization in estonia and wales. european planning studies, 29(4), 681-698. https://doi.org/10.1080/09654313.2020.1767541 zenker, a., & kroll, h. (2014). regional innovation monitor plus, regional innovation report saxony-anhalt. brüssel: technopolis belgium. https://doi.org/10.1016/j.habitatint.2017.03.005 https://doi.org/10.1016/j.habitatint.2017.03.005 https://doi.org/10.1080/00343404.2018.1530752 https://doi.org/10.1016/j.respol.2005.01.018 https://doi.org/10.1016/j.respol.2005.01.018 https://doi.org/10.1080/00343404.2019.1582763 https://doi.org/10.1080/09654313.2020.1767541 abstract 1. introduction 2. multilevel governance in smart specialization policy: theoretical background 3. multilevel governance and smart specialization in eu regions: an evidence-based critical review 4. conclusion acknowledgments references european journal of government and economics 11(1), june 2022, 97-112 european journal of government and economics issn:2254-7088 gender accessibility to credit among entrepreneurs: empirical evidence from women entrepreneurs in kano metropolis rabi bashir a, abubakar hamid danlami a, * a department of economics, faculty of social sciences, bayero university kano, nigeria * corresponding author at: sadiqdanlami@yahoo.com abstract. this study was carried out to empirically assess women entrepreneurs’ access to credit in kano metropolis. the study was aimed at analyzing the maximum willingness to accept loans by women entrepreneurs and exploring the major problems hindering women entrepreneurs from fundraising for businesses in kano metropolis. the study employed a multi-stage random sampling to select a total of three hundred and seventy-six (376) women entrepreneurs as the sample of the study. the analysis of the data was carried out using frequency count and simple percentages to analyze the demographic characteristics of the respondents. furthermore, the inferential statistics were done using logit models to capture the specific objectives of the study. the findings of the study revealed that high interest and collateral requirement prevent women entrepreneurs from accessing a loan from financial institutions, while religious affiliation is a major factor affecting women entrepreneurs' access to credit facilities. based on the findings of the study, the study recommends that financial institutions should introduce and operate an islamic finance system to encourage women entrepreneurs to go for a loan as the interest rate is one of the most important constraints to women entrepreneurs seeking financial credits. additionally, collateral demand attached to a loan for women should be removed to encourage women entrepreneurs to raise capital. keywords.women, gender, entrepreneurs, finance, credit accessibility. doi. https://doi.org/10.17979/ejge.2022.11.1.8198 1. introduction entrepreneurial activity is crucial for private sector development and imperative for a country’s economic prosperity. micro, small and medium enterprises (msmes) create jobs and provide nurturing ground for private sector development. msmes play a prominent role in developing countries that are characterized by high levels of unemployment and poverty (abubakar et al., 2013). entrepreneurship has been regarded as a major contributing factor to economic growth and poverty alleviation both in urban and rural areas. the role of entrepreneurship and entrepreneurial culture in economic and social development has often been underestimated. over the years, however, it has become increasingly apparent that entrepreneurship indeed contributes to economic development. nevertheless, significant numbers of enterprises are owned by men (international labour organization (ilo), 2006). in the past, it was not common to see women entrepreneurs worldwide, especially in developing countries. the idea and practice of women's entrepreneurship is a recent © 2022. this work is licensed under a cc by-nc 4.0 license. mailto:sadiqdanlami@yahoo.com https://doi.org/10.17979/ejge.2022.11.1.8198 https://creativecommons.org/licenses/by-nc/4.0/ a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 98 phenomenon. until the 1980’s, little was known about women's entrepreneurship both in practice and research, which made its focus entirely on men. scientific discourse about women’s entrepreneurship and women-owned and run organizations is just the development of 1980s (ilo, 2006). women entrepreneurship plays an important role in the economic development of a country. ochola and okelo (2013) found that women entrepreneurs enhance livelihoods within the family and lead to wealth creation at the national level. women’s involvement in entrepreneurship is applauded as a necessary precursor to the economic growth of nations (richardson et al., 2004). women in nigeria have been known to operate smes businesses effectively in areas such as food production and preservation, craft and cottage industries, trading, farming, and services provision. the deterioration of the economic situation since the 1980s adversely affected the women economic condition globally. hence, it exposed them to a high poverty level situation that has resulted in regarding them as the poorest of the poor (cheton, 2002 & burkett, 2003). women have been recognized as the most neglected and marginalized sector as per as access to credit is concerned due to their inability to provide collateral security and other conditions required by financial institutions. this is a typical case of gender inequality, which might hinder economic growth (world bank, 2001). furthermore, programmes by governments and public agencies provide skills, start-up capital and other inputs, which are meant to support women interactions within the formal sector. despite the empowerment supports, many of these programmes have been mostly adhoc and poorly sustained. with the expansion of global capital and growing transnational trade, the informal sector in nigeria, when women’s creativity is largely ignored and left to rot through non-gender sensitive economic programmes. women in this sector are left unaided, and if so, what alternative businesses and economic arenas are available to them if they are to take advantage of the new investment environment driven by globalization. these and many other concerns have raised new questions on how women are benefiting from global trade and which environment is needed to support women’s involvement. in the case of many countries undergoing various economic reforms, other questions have been raised about how endangered these reforms were when the contributions of the main sector in which women in developing countries operate (informal sector) is yet to be fully integrated into strategies of economic reform. this has resulted in the invisibility of women’s economic contribution and their inability to access many of the benefits available within these economic packages and programmes. access to credit by women entrepreneurs is hampered by numerous factors, including a low level of education, higher interest rate by the lending institutions, socio-cultural factors that place women in a lower status in the society as compared to men in as far as running businesses is concerned. in kano state, it was well observed that women have less involvement in entrepreneurial activities. most of the women in kano engaged in-home businesses. kano is a majority muslim populated state, and the majority is hausa or hausa/fulani. there is the notion among the a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 99 women in kano, especially the muslims, that obtaining a bank loan is against the teaching of their religion. others have the notion that when loans are collected from banks, the interest on them is forbidden. thus, even if the women have met all the necessary requirements to collect loans from any financial institutions, some of the women are not willing to collect loans from banks. this may be a result of religious beliefs and, more importantly, lack of trust. in the northern part of nigeria especially kano, women do not trust financial institutions. they believe that these financial institutions have exploitative agenda and have tendencies to lead them to financial hardship if they can’t pay back the loan in due time. cultural belief is one major barrier to women's entrepreneurship in africa. the majority of cultural practices in africa is that women's main purpose is to either work in the farmlands or just stay at home. this trend serves as a major barrier for women to get involved in entrepreneurial activities, thereby reducing their access to loans from financial institutions. hence this study assesses the determinants of women entrepreneurs’ access to credit in kano metropolis, kano state. the study specifically addressed how collateral requirements, socio-cultural values, and interest rates affect women entrepreneurs' access to credit facilities in kano metropolis, nigeria. 2. literature review and theoretical framework 2.1 empirical literature review several studies were conducted to assess the relationship between gender and credit accessibility in different areas. for instance, olateji et al. (2017), examined the factors that determine women`s access to microcredit programs in lagos, the study revealed that all the variables that were used in measuring the determinants of women`s access to microcredit are significant except the variable age which is not significant. in the same vein, kweyu (2017), concluded that providing financial support for women can increase the number of new businesses, which in turn, would boost economic activity, and enable the expansion of old businesses, leading to increased productivity and growth. furthermore, eze etal. (2016), found that household size and marital status were more likely to increase access to credit. education attainment has a positive relationship with access to credit. distance is less likely to decrease access to credit. this study was supported by the findings of balogun et al. (2016), as was pointed out by esther (2016), who examined the factors affecting access of women enterprise funds by women groups in kenya. akhabonje and namusonge (2016) established that lending procedures, collateral requirements, credit bureau referencing policies and training offered by finance institutions significantly influence access to credit facilities by sme from financial institutions. moreover, studies by mutai (2015), waari and mwangi (2015) and chinonye et al. (2015), showed that women entrepreneurs are gaining recognition in entrepreneurial activities, they are however, confronted with several challenges which include inadequacy of funds for start-up and a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 100 expansion, lack of electricity/infrastructural facilities, un-conducive business environment, customers dissatisfaction and complaints, high level of competition and lack of trustworthy personnel. in addition, they pointed out that information asymmetry, business risks and transactional cost influence access to finance. karanja et al. (2014) examined how lending procedures hinder access to credit services by women entrepreneurs, using both qualitative and quantitative methods, and they found out that all the variables have a causal relationship with access to credit by women entrepreneurs. likewise, machiraetal. (2014), attempted to determine the accessibility of women's enterprise funds by women enterprise owners in tharaka south district, the study revealed that access to women enterprise fund and studied variables were closely related. this study was similar to that of gichuki et al. (2014), who investigated selected factors perceived to influence the performance of women-owned small micro-enterprises. the study showed that the selected factors income, credit and education level of the respondents influenced positive changes in the net profits and capital of small micro enterprises (smes), while age did not influence the change in the enterprise. from the literature reviewed, it was shown that not all factors have equal importance in determining the gender accessibility of credit for different areas due to differences in socioeconomic settings, environmental factors, and cultural factors as well as the average level of development in the area. this (i.e., differences in the study area) has led to the arrival of different and inconsistent conclusions in the literature on gender and credit accessibility among women entrepreneurs. 2.2 theoretical framework this study will adopt the theoretical framework used by muntai (2015). muntai (2015) used schumpeter’s (1939) theory of innovation which is in line with some of the specific objectives of this study. according to schumpeter’s theory of innovation, development implies carrying one new combination of entrepreneurship. ‘entrepreneur’ is an innovator who carries a new combination of new goods / services, new method of production, new market, new source of supply of raw materials, new organization. in political economics, entrepreneurship is the process of identifying and starting a new business venture, sourcing and organizing the required resources, while taking both the risks and rewards associated with the venture. according to schumpeter, a woman entrepreneur is willing and able to convert a new idea or invention into a successful innovation despite limitations accessibility to business finances. entrepreneurship employs what schumpeter called "the gale of creative destruction" to replace in whole or in part inferior offerings across markets and industries, simultaneously creating new products and new business models. thus, creative destruction is largely responsible for the dynamism of industry and long-term economic growth. in spite of the consensus that size has a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 101 little effect on innovation; size does have important consequences on innovation. some advantages of large firms are: a) the spread of fixed costs of innovation b) economies of scale, scope in research and development c) capital market imperfection confers advantages to large firms in securing finance for risky research and development projects as firms grow larger, efficiency in r & d is undermined either through the loss of managerial control or, alternatively, excessive bureaucratic control, which diverts the attention of the firm’s bench scientists and technologists. moreover, as a firm grows large, the incentives of individual scientists and entrepreneurs may be blunted as either their ability to capture benefits from their individual efforts dismisses or their creative impulses are frustrated by the conservatism characteristics of hierarchy of large corporations. the model presented by muntai (2015) was adapted in this study; the model is presented as in the diagram below. figure 1. conceptual framework. source: adapted from muntai (2015) socio-cultural factor age religious background tribal background educational background marital status geopolitcal zone willingness to accept/collect amount willing to collect interest rates cumbersome procedures small loan size collateral issues business plan availability of the financial institution religeous affiliation accessibility of credit/finance facilities by women collateral requirements and interest amount collected level of interest rates adequate collateral length of repayment business plan religeous affiliation a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 102 in order to estimate the willingness to accept/collect loans by women entrepreneurs, this study will adopt the logit regression framework, which was specified by epo (2012). according to epo (2012), assume that the probability of being a local female entrepreneur is determined by an underlying response variable that captures the economic activity undertaken by an individual. 3. methodology 3.1 population of the study there is no specific data to determine the number of female entrepreneurs in kano state at large and how many are within the kano municipal. this study focuses on women entrepreneurs in kano metropolitan areas, namely fagge, dala, gwale, tarauni, nassarawa, kumbotso, ungogo and kano municipal. 3.2 sample size since the exact number of women entrepreneurs is not known, the study does not have a sample frame. since there is an absence of a sample frame, the study used the standard deviation technique to estimate the sample size of the study. the sample size determination formula is given as: 𝑆𝑆 = (𝑍𝑍𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠)2 + 𝛿𝛿(1 − 𝛿𝛿)/(𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑜𝑜𝑜𝑜 𝑒𝑒𝑀𝑀𝑀𝑀𝑜𝑜𝑀𝑀)2 given; 90% confidence interval, 0.5 standard deviation and margin error of 0.05. 𝑆𝑆 = (1.645)2𝑋𝑋0.5(1.05)/(0.05)2 2.706 x 0.25 0.0025 0.6765 0.0025 =270.6 however, because of the problem of non-response bias (danlami, 2017; danlami et al., 2018) a total of 400 questionnaires were distributed 3.3 sampling technique the study used multi-stage sampling technique as used by some previous studies (danlami, 2019; kofarmata & danlami, 2019; danlami & islam, 2020; tsauni & danlami, 2016). the a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 103 proportion of each segment is calculated, and the questionnaires will be divided on the bases of such proportion estimated in each segment, and a cluster sample using multi-stage random sampling will be used to select the respondents from each segment comprising the number of women entrepreneurs in kano metropolitan, kano state. this sampling technique is employed because it has two advantages viz: (a) it is easier to administer than most single-stage designs mainly because of the fact that sampling frame under multi-stage sampling is developed in partial units. (b) a large number of units can be sampled for a given cost under multistage sampling because of sequential clustering, whereas this is not possible in most of the simple designs (kothari 2004). 3.4 method of data analysis to examine how collateral requirements and interest rate affects women entrepreneurs' access to credit facilities in kano metropolis. the study will also employ the binary logit model for this. following danlami et al. (2017) the empirical model can be derived as; 𝑃𝑃 = 𝐸𝐸 �𝑌𝑌 = 1 𝑋𝑋𝑖𝑖 � = 1 1+𝑠𝑠−�𝐵𝐵𝑖𝑖+𝐵𝐵2𝑋𝑋𝑖𝑖� [1] 𝑀𝑀𝑜𝑜 𝑧𝑧 = 𝛽𝛽1 + 𝛽𝛽2𝑋𝑋𝑖𝑖then; 𝑃𝑃𝑀𝑀 = 1 1+ 𝑠𝑠−𝑍𝑍𝑖𝑖 = 𝑠𝑠𝑧𝑧 1+𝑠𝑠𝑧𝑧 [2] if p represents the probability of occurrence (say if the respondent had collected a loan or not), the probability of not occurrence can be expressed as: 1 − 𝑃𝑃𝑀𝑀 = 1 1+𝑠𝑠𝑧𝑧𝑖𝑖 [3] hence the odds ratio between the probabilities of occurrence and non-occurrence can be expressed as: pi 1−pi = 1+ ezi 1+e−zi = ezi [4] where: pi 1−pi represents the odds ratio of if the respondent had collected loan. that is the ratio of the probability that a female entrepreneur had once collected a loan to the probability of otherwise. taking the natural log of equation (4) we obtained the following expression as: 𝐿𝐿𝑖𝑖 = 𝑙𝑙𝑀𝑀 � 𝑃𝑃𝑖𝑖 1−𝑃𝑃𝑖𝑖 � = 𝑍𝑍 = 𝛽𝛽1 + 𝛽𝛽2𝑋𝑋𝑖𝑖 [5] where l means the log of odds ratios, equation (5) represents what is known as the logit model, which is used when the defendant variable takes a binary value; 0 or 1. the empirical logit model estimated in this study can be expressed as: a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 104 ln� pi 1−pi � = βo= β0 + β1 x1+ β2 x2 + β3 x3 + β4 x4 + β5 x5 + β6 x6 + β7 x7 +δi [6] � pi 1−pi �= collected loan before (0 have not and 1 otherwise) βi = coefficients x1= huge collateral x2= high interest rates x3= accessibility to financial institutions x4= age x5= education x6= marital status x7 = local government area δi = error term 4. results the demographic and socio-economic characteristics of the respondents captured from the study area including gender distribution, age, level of income, educational qualification, marital status, type of residence, household size, occupational distribution and level of income of the respondents, were analyzed in this section using the simple descriptive statistics analysis. table 1. demographic and socio-economic analysis responses frequency percentage marital-status not married 192 51.06 married 184 48.94 education no formal education 125 33.24 primary level 40 10.64 secondary level 158 42.02 tertiary level 53 14.10 religion muslim 281 74.73 christian 95 25.27 tribe hausa/fulani 237 63.03 yoruba 70 18.62 igbo 46 12.23 igala 23 6.12 a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 105 table 1 shows the demographic distribution of the respondents. from the table, the result shows that majority of the respondents consisting of 192 (51.06%) are not married, while 184(48.94%) of the respondents are married. also, the educational qualification of the respondents shows that 125(33.24%), while 40(10.64%) of the respondents have primary level education and 158(42.02%) of the respondents have secondary level education as 53(14.10%) of the respondents have only tertiary level education. likewise, 281(74.73%) of the respondents are muslims as 95(25.27%) of the respondents are christians. based on the tribal distribution, 237(63.03%) of the respondents are hausa/fulani while, 70(18.62%) of the respondents are yorubas, and 46(12.23%) of the respondents are igbo’s as 23(6.12%) are igala’s. going by the results, it is evident that the hausa/fulani women are less involved in entrepreneurial activities compared to women from other tribes. likewise, the christian ladies are more involved in entrepreneurial activities compared to the muslim ladies. this by implication shows the cultural dimension of the involvement of women entrepreneurs in kano metropolis. this further implies most indigenes from the north-western part of nigeria hardly encourage their women in entrepreneurial activities. 4.1. collateral requirements, interest rate and women entrepreneurs' access to credit the model made used the binary dependent variable model of logit model to determine how collateral requirements and interest rate affects women entrepreneurs' access to credit facilities in kano metropolis. table 2 presents the logit estimate of the first model. table 2 shows the logit estimation examining how collateral requirements and interest rate affects women entrepreneurs' access to credit facilities in kano metropolis. the result shows that the collateral requirement by financial institutions has a significant effect on women entrepreneurs’ access to credit facilities. the results of higher demand for collateral, women entrepreneurs are less likely to collect loans from financial institutions. the result further shows that the coefficient of interest is negative and statistically significant. the result implies that with a higher interest rate, women entrepreneurs are less likely to collect loans from financial institutions. the result also shows that the availability of financial intuitions’ is a significant factor affecting women entrepreneurs access to credit facilities. the sign of the coefficient is negative, which implies that lack of availability of financial institutions, women entrepreneurs are less likely to collect loans from financial institutions. in addition, the coefficient of age is positive and statistically significant. this implies as age of women entrepreneurs increase, they are more likely to collect loan from financial institutions. a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 106 table 2. collateral and interest constraints to women entrepreneurs’ access to credit facilities. variables coefficient odd ratios huge collateral -1.176* 0.309* (0.632) (0.195) huge collateral 5.078*** 160.4*** (1.248) (200.3) high interest rate -2.774*** 0.0624*** (0.599) (0.0374) availability financial institute -0.399* 0.671* (0.241) (0.162) age 0.140*** 1.150*** (0.0196) (0.0225) primary education 3.212*** 24.83*** (0.843) (20.93) secondary education -1.166** 0.312** (0.554) (0.173) tertiary education 0.843 2.323 (0.835) (1.941) married -1.440*** 0.237*** (0.510) (0.121) fagge -2.915*** 0.0542*** (0.799) (0.0433) ungogo 1.255* 3.506* (0.730) (2.561) constant -3.814*** 0.0221*** (1.002) (0.0221) observations 368 368 note: the asterisks ***, ** and * indicate significance at 1%, 5% and 10% respectively. the figures in parenthesis () are standard errors on the education level of the respondents, the sign of the coefficient of those with primary education are more likely to collect loans from financial institutions, and it is statistically significant. while the sign for secondary education level is negative and statistically significant, which implies that women entrepreneurs with secondary educational level are less likely to collect loans from financial institutions, also, women entrepreneurs with tertiary education are more likely to collect loans from financial institutions. the sign of the coefficient of marriage is negative and statistically significant, which implies that married women entrepreneurs are less likely to collect loans from financial institutions. likewise, in the local government where the women entrepreneurs reside, the sign of the coefficient of fagge lga is negative and statistically significant. this implies that women entrepreneurs in fagge lga are less likely to collect loans from financial institutions. also, the sign of the coefficient of ungogo lga is positive and significant. this implies that women entrepreneurs in ungogo lga are more likely to collect loans from financial institutions. the result above shows that collateral requirement has hindered women from accessing credit facilities irrespective of their demographic characteristics. the issue of collateral has been a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 107 a serious impediment to accessing a loan from financial institutions and given the fact that most the women entrepreneurs are not buoyant enough to offer collateral to collect loans from financial institutions. in addition, more of the educated women, as the results shows, will be less likely to collect loans from financial institutions, this may be because, they fear that banks will come after their assets if they fail to pay the loans on time. table 3. willingness to accept loan from financial institutions by women entrepreneurs. variables coefficients odd ratios legal_status -1.718* 0.179* (0.941) (0.169) operation 0.707** 2.027** (0.301) (0.610) collateral_required 5.390*** 219.2*** (1.593) (349.2) high_interest_rates -3.137*** 0.0434*** (1.114) (0.0484) religion_affiliation -0.370 0.691 (0.767) (0.530) religion_affiliation 0.0612 1.063 (0.871) (0.926) maximum_loan_willing 5.06e-06 1.000 (4.75e-06) (4.75e-06) repayment 0.850*** 2.339*** (0.268) (0.628) accessibility_financial_institut -10.59*** 2.52e-05*** (2.637) (6.64e-05) age 0.0312 1.032 (0.0522) (0.0539) marital_status 1.457* 4.293* (0.759) (3.259) education -0.541*** 0.582*** (0.147) (0.0858) religion -4.293*** 0.0137*** (1.609) (0.0220) hausa 1.568 4.798 (1.047) (5.023) yoruba 3.990** 54.03** (1.625) (87.80) igbo 0.574 1.776 (1.858) (3.299) household_size 1.705*** 5.500*** (0.488) (2.684) constant -7.762*** 0.000425*** (2.900) (0.00123) observations 189 189 note: the asterisks ***, ** and * indicate significance at 1%, 5% and 10% respectively. the figures in parenthesis () are standard errors. a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 108 4.2the willingness to accept loan by women entrepreneurs from financial institutions this model used logit model to determine willingness to accept loans by women entrepreneurs from financial institutions in kano metropolis. the table below presents the logit estimate of the first model. table 3 shows the estimated logit model of the first model of the study. the estimated model shows the coefficients and the odds ratios of the estimated variables, which estimate the factors that determine the willingness to accept loans from financial institutions by women entrepreneurs in kano metropolis. from the results, the variable operation, which measures the length of operating the business by women entrepreneurs was found to be positive and statistically significant, with the coefficient value of 0.707 at 5% level of significance. the longer they operate, the more willing to accept a loan from financial institutions. this is in line with a priori expectations as those women entrepreneurs who work more may need a loan to maintain the length of operations. the odd ratio shows that the odds of accepting loans is higher when they operate longer than when they operate less. this finding is in line with the work of kweyu (2017), who found that years of operation have a significant influence on women's access to credit facilities. in addition, the results show the coefficient of collateral requirement by financial institutions was found to be positive with a coefficient value of 5.390 and statistically significant at 1%. collateral requirement is one factor that determines the willingness to accept a loan from financial institutions by women entrepreneurs. this doesn’t conform to a priori expectation as collateral requirements should discourage women from seeking financial assistance. this is in line with finding of akhabonje and namusonge (2016), who tried to establish the factors influencing access to credit among small and medium scale enterprises in kitale municipality, and found out that collateral requirements by finance institutions significantly influence access to credit facilitiesby sme from financial institutions. also, the coefficient of high interest rate was found to be negative and statistically significant at 1% with a value of -3.137. this result conforms to a priori expectations as higher interest will make women entrepreneurs to be less willing to accept/collect loans from financial institutions. this implies that with high interest rates, women entrepreneurs are less willing to collect loans from financial institutions. interests prove to be a significant factor in women's decision to accept loans from financial institutions or not. in addition, the coefficient of the variable length of repayment is positive and statistically significant at 1% with a value of 0.850. this implies that the length of repayment of loans is a significant factor in determining women's willingness to accept a loan from financial institutions. the result shows that women are more willing to accept loans from a financial institution when the length of repayment is higher. this should be a stochastic variable as which it can take any sign. by implication, if the time of repayment is long enough, women entrepreneurs could go for a loan, and if the length of repayment is short, women entrepreneurs will not go for a loan. also, accessibility of financial institutions has a negative coefficient value of -10.59 and found a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 109 to be statistically significant at 1%. the variable accessibility of financial institutions as proved to be a significant factor that determines women entrepreneurs’ willingness to accept loans from financial institutions. the results show that the less accessible the financial institutions the less willing women are to accept a loan from the financial institutions. the sign of the coefficient is negative, which doesn’t conform with the prior expectation because accessibility to financial institutions should encourage women entrepreneurs to accept a loan, the negative sign may be due to the fact that women entrepreneurs are less willing to accept loans because the financial institutions are close to them. also, education has a negative coefficient of -0.541 and is statistically significant at 1%, which implies that education is a significant factor that determines women entrepreneurs’ willingness to accept loans from financial institutions. the sign is a stochastic variable which may take any sign as an educated person may be willing to accept or decline a loan offer depending on other factors she may be looking at. the result shows that the higher the level of education the less willing the women entrepreneurs to accept loans from financial institutions. this finding is in line with the work of kweyu (2017), who posited that education level has a significant influence on women's access to credit facilities. likewise, the coefficient of religion is -4.292 and was found to be negative and statistically significant at 5%. the variable conforms to the a priori expectation since religion has distanced itself from loans with interest. the variable has proved to be a significant factor that determines the women entrepreneur's willingness to accept a loan from financial institutions. due to religious regulations, women entrepreneurs are less willing to accept a loan from financial institutions. the result shows that igbo women entrepreneurs are more likely to accept a loan that other tribes among the respondents. also, the sign of the coefficient of household size is positive and statistically significant at 1%. this conforms to the a priori expectation as persons with larger household size may seek for loan to maintain the house. the results show that the larger the household size of the women entrepreneur, the more willing they are to collect loans from financial institutions. this finding is in line with the findings of boniface (2012), who investigated the effects of microcredit and social capital on female entrepreneurship. he found out that household size has an influence on access to credit facilities by women entrepreneur the findings from the results above show that women are not likely to collect loans because of high interest rates, irrespective of demographic characteristics. this situation shows why all the religions and tribal backgrounds are against asking and collecting interest in business activities. also, interest increase with time, and most women will find it difficult to pay the loans given the increasing interest rate. this problem has affected women entrepreneurs accessing loans from financial institutions. a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 110 4.3 constraints facing women entrepreneurs table 4 analyzes the constraints facing women entrepreneurs in kano metropolis, kano state nigeria. table 4. constraints to financial access by women entrepreneurs in kano metropolis. constraints frequency percentages cumbersome_procedure 254 99.20% high_interest_rates 262 99.60% small_loan_size 172 57.53% lack_of_collateral 242 76.10%` business_plan 236 77.89% religion_affiliation 243 77.88% inadequate capital 242 66.85% poor patronage 120 33.15% total 1771* note: * indicates multiple responses table 4 shows the responses on the various constraints affecting women entrepreneurs in kano metropolis, kano state nigeria regarding financial access. the results show that 254(99.2%) of the respondents identify the cumbersome procedure of collecting loans from financial institutions as a major constraint affecting women entrepreneurs. this is because when the women go for a loan, they find the procedures too much to stressful. also, 262(99.6%) identify high interest rates as a major constraint affecting women entrepreneurs' access to credit facilities in kano metropolis. the level of interest has not allowed most women entrepreneurs to seek loan from financial institutions. this is due to their low returns on profit to even pay back the loan. in addition, 172(57.53%) of the respondents identify that small loan size is a constrain affecting women entrepreneurs' access to credit facilities in kano metropolis. this is because women entrepreneurs want a larger loan to expand their businesses but cannot get that amount from financial institutions. also, 242(76.10 %`) of the respondents identify the inability to get collateral as a constraint on women entrepreneurs' access to credit facilities. this is because demand for collateral when the majority of women entrepreneurs in kano metropolis is small scale businesswomen with little or no collateral. while 236(77.89%) of the respondents identify the inability to draw a business plan as another constraint on women entrepreneurs' access to credit facilities. this is because majority of the women don’t have the requisite knowledge to draw business plan. furthermore, 243(77.88%) the religious affiliation of women entrepreneurs is a major constrain to the ability of women entrepreneurs’ access to credit facilitates. the religious factors of islamic doctrine have prohibited interest likewise, the christian faith, as such most of the credit facilities provided by the financial institutions go along with interest rate requirements which makes it a problem for women entrepreneurs to gain access to credit facilities. likewise, a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 111 242(66.85%) of the respondents identify inadequate capital as a constraint affecting women entrepreneurs to expand their businesses. this is because women entrepreneurs in kano metropolis have less capital to operate and expand their businesses. as 120 (33.15%) of the respondents identify poor patronage of their commodities is a major constraint affecting women entrepreneurs in kano metropolis. another major factor affecting women entrepreneurs in kano metropolis is the low patronage of their products this is because the products they sell have low value. 5. conclusions and recommendations the study further shows that high interest and collateral requirement prevent women entrepreneurs from accessing loan from financial institutions, while religious affiliations is a major factor affecting women entrepreneurs' access to credit facilities. the study recommends that financial institutions should prepare financial provisions or packages for women entrepreneurs who are willing to accept loans based on non-interest loan financing. the study recommends that financial institutions should make provisions for training women entrepreneurs on how to raise capital to expand their businesses and also give them incentives with fewer conditions attached to them. references abubakar. s. g, shazali a. m., & fariastuti d. (2013). entrepreneurship and its determinants in nigeria: empirical evidence from micro and small enterprises in kano state.journal of entrepreneurship and management2(2), 2 -12. balogun,u.o., bustaman,u,s.a., & johari,f.b. (2016). determinant factors affecting women social enterprise performances in sokoto state, nigeria. asian journal of social sciences and humanities, 5(2), 47 -53. baumol, w.j. (1990). entrepreneurship: productive, unproductive, and destructive.journal of political economy 97: 721-739. https://doi.org/10.1086/261712 baumol, w.j. (2002). the free-market innovation machine: analyzing the growth miracle of capitalism.princeton university press: usa. https://doi.org/10.1515/9781400851638 casson, m., yeung, b., basu, a., &wadeson, w. (eds.), (2006). the oxford handbook of entrepreneurship. oxford, oxford university press: london. https://doi.org/10.1093/oxfordhb/9780199546992.001.0001 chen, c.c., greene, p.g., & crick, a. (1998). does entrepreneurial self-efficacy distinguish entrepreneurs from managers? journal of business venturing, 13(4), 295–316. https://doi.org/10.1016/s0883-9026(97)00029-3 chinonye, o., &chima, m. (2010). women entrepreneurship development in nigeria: the effect of environmental factors. petroleum-gas university of ploiesti bulletin, lxii (4); 43 52. danlami, a. h. (2019). assessment of factors influencing firewood consumption in bauchi state, nigeria. journal of sustainability science and management, 14(1), 99 – 109. danlami, a.h. (2017). an intensity of household kerosene use in bauchi state, nigeria: a tobit analysis. nigerian journal of management technology & analysis, 8(2), 1 – 13. https://doi.org/10.1086/261712 https://doi.org/10.1515/9781400851638 https://doi.org/10.1093/oxfordhb/9780199546992.001.0001 https://doi.org/10.1016/s0883-9026(97)00029-3 a. h. danlami and r. bashir / european journal of government and economics 11(1), june 2022, 97-112 112 danlami, a.h., applanaidu, s.d., & islam, r. (2017). from biomass cooking fuel to modern alternative for bauchi state households: a preliminary analysis. biofuels, 8(3): 323-331. https://doi.org/10.1080/17597269.2016.1226724 danlami, a.h., applanaidu, s.d., & islam, r. (2018). axiom of the relative income hypothesis and household energy choice and consumption in developing areas: empirical evidence using verme model. kasetsart journal of social sciences, 39, 422 – 431. https://doi.org/10.1016/j.kjss.2018.06.010 danlami, a. h., & islam, r. (2020). explorative analysis of household energy consumption in bauchi state, nigeria. in a. editor (eds.), energy efficiency and sustainable lighting –a bet for the future (page number). london: intechopen. https://doi.org/10.5772/intechopen.89597 esther, m. m. (2016). factors affecting access of women enterprise fund by women group in nakuru west sub-country, kenya. gichuki, c.n., milcah, m.m., & kinuthia, l.n. (2014). performance of women owned enterprises accessing credit from village credit and savings associations in kenya. journal of global entrepreneurship research, 4(16), 2 – 13. https://doi.org/10.1186/s40497-014-0016-1 holmes, t.j., & schmitz j.a. (1990). a theory of entrepreneurship and its application to the study of business transfers.journal of political economy 98: 265-294. https://doi.org/10.1086/261678 ilo. (2006). vulnerability and young women entrepreneurs: a case study of ethiopian informal economy. geneva: international labor organization. karanja,j.g., mwangi, a.k & nyakarimi, s.n. (2014).analysis of factors influencing access to credit services by women entrepreneurs in kenya. kofarmata, y. i. & danlami, a.h. (2021). a micro level analysis of the intensity of agricultural finance supply in nigeria: empirical evidence. sn business & economics, 1(12), 1 – 17. springer journal. https://doi.org/10.1007/s43546-020-00011-y kofarmata, y. i. & danlami, a.h. (2019). determinants of credit rationing among rural farmers in developing areas: empirical evidence based on micro level data. agricultural finance review, 79(2), 158 – 173. https://doi.org/10.1108/afr-03-2018-0023 lazear, e.p. (2004). balanced skills and entrepreneurship. american economic review 94: 208-211. https://doi.org/10.1257/0002828041301425 machira, h.k., njati, i.c., thaine, k.s. & s, g.h. (2014). accessibility of women enterprise fund and small and micro-women enterprise owners in thoraka south district, kenya. https://doi.org/10.9790/487x-161379103 muntai, m., (2015). what do we know about the patterns and determinants of female entrepreneurship across countries? european journal of development research, 22 (3), 277-293. https://doi.org/10.1057/ejdr.2010.17 ocholah, r. m., okelo, s., ojwang, c., aila, f., &ojera, p. b. (2013). literature review on the relationship between microfinance provision and women enterprise performance. greener journal of social sciences, 3(5), 278-285. https://doi.org/10.15580/gjss.2013.5.022613502 olateji, a.d., aminu, a.w., & danmola, r.a. (2017). the determinants of women entrepreneurs` access to microcredit progromme: a case study of cowrie’s microfinance bank lagos, nigeria. international journal of development and sustainability 6(8), 783-791. richardson, p., howarth, r., & finnegan, g. (2004). the challenges of growing small businesses: insights from women entrepreneurs in africa. seed working paper no.47.geneva: international labour office. schmitz, j.a. (1989). imitation, entrepreneurship, and long-run growth. journal of political economy, 97(3), 721 – 739. https://doi.org/10.1086/261624 schultz, t.w. (1975). the value of the ability to deal with disequilibria, journal of economic literature, 13, 827-846. tsauni, a.m., & danlami, a.h. (2016). social capital formation and implication for the achievement of fadama iii project development in kano state, nigeria. hyperion economic journal 1(4), 37 – 48. waari, d.n. & mwangi,w.m. (2015). factors influencing access to finance by micro, small and medium enterprises in merucounty, kenya. international journal of economics, commerce and management, iii(4), 1-15. world bank. (2012). the world bank annual report 2012: volume 1. main report. world bank annual report. https://doi.org/10.1596/978-0-8213-9568-4 wosowei, e.c., & ifelunini, i.a. (2013). constraints to women entrepreneurs` access to microfinance in south –south nigeria. research journal of finance and accounting, 4(6), 613. https://doi.org/10.1080/17597269.2016.1226724 https://doi.org/10.1016/j.kjss.2018.06.010 https://doi.org/10.5772/intechopen.89597 https://doi.org/10.1186/s40497-014-0016-1 https://doi.org/10.1086/261678 https://doi.org/10.1007/s43546-020-00011-y https://doi.org/10.1108/afr-03-2018-0023 https://doi.org/10.1257/0002828041301425 https://doi.org/10.9790/487x-161379103 https://doi.org/10.1057/ejdr.2010.17 https://doi.org/10.15580/gjss.2013.5.022613502 https://doi.org/10.1086/261624 https://doi.org/10.1596/978-0-8213-9568-4 1. introduction 2. literature review and theoretical framework 3. methodology 4. results 5. conclusions and recommendations references residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) vol.8 • no.1 2019 issn: 2254-7088 european journal of government and economics 8(1), june 2019. european journal of government and economics issn: 2254-7088 number 8, issue 1, june 2019 doi: https://doi.org/10.17979/ejge.2018.8.1 the evaluation index improvement for the relocation of research equipment 6-29 donghun yoon residents’ attitudes towards different tourist offers: maldonado-punta del este conurbation (uruguay) 30-47 josé ramón cardona, daniela álvarez bassi, maría dolores sánchez-fernández social capital and banking system profitability: a survey of european union countries 48-62 arash nayebyazdi governance and domestic investment in africa 63-80 chimere okechukwu iheonu economic analysis of supply functions, private returns to investment in education and skill mismatch in egypt 81-95 marwa shibl biltagy give me liberty or give me money: the fiscal decentralization and autonomy of regional governance in slovakia 96-109 jaroslav mihálik, peter horváth, martin švikruha https://doi.org/10.17979/ejge.2018.8.1 european journal of government and economics 8(1), june 2019, 30-47. european journal of government and economics issn: 2254-7088 residents’ attitudes towards different tourist offers: maldonadopunta del este conurbation (uruguay) josé ramón cardonaa, daniela álvarez bassib, maría dolores sánchez-fernándezc * a universitat de les illes balears, spain b universidad católica del uruguay, uruguay c universidade da coruña, spain * corresponding author at: maria.sanchezf@udc.es article history. received 8 october 2018; first revision required 21 october 2018; accepted 10 january 2019. abstract. the study of residents' attitudes is fundamental for tourist destinations managers. residents form their attitude towards tourism through a cost-benefit analysis based on their perception of tourism impacts. the objective of this paper is to analyse the relationship between the perception of the different types of impacts, the general attitude toward tourism and various types of tourism offer: sport tourism, maritime tourism, nature tourism and sun and beach tourism. the analysis of the data was performed using partial least squares on a sample of 420 residents of the maldonado-punta del este conurbation, interviewed in 2016. the strongest causal relationship is between general attitude and acceptance of sun and beach tourism, the traditional offer in the region. the main conclusion is that a more favourable attitude to tourism does not imply greater support for any type of tourism. it is not possible to generalize the support to tourism to any tourism project, and tourist destinations managers must take it into account. keywords. attitudes, residents, pls, punta del este. jel codes: c88, l83, z32 doi. https://10.17979/ejge.2019.8.1.4580 1. introduction several studies on the residents’ attitudes towards tourism have been undertaken since the end of the 70s (almeida, balbuena & cortés, 2015; sharpley, 2014). at first researchers focused their investigations on measuring attitudes and their relationship with the impacts perceived (besculides, lee & mccormick, 2002; gursoy, jurowski & uysal, 2002; jurowski, uysal & williams, 1997; lindberg & johnson, 1997; mason & cheyne, 2000; teye, sirakaya & sönmez, 2002). these investigations grouped the impacts, benefits and costs into three or four categories within the economic, social, cultural and environmental framework (andereck, valentine, knopf & vogt, 2005; ayres, 2000; gursoy & rutherford, 2004; gursoyet al., 2002; özel & kozak, 2017). mailto:maria.sanchezf@udc.es https://10.0.70.59/ejge.2019.8.1.4580 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 31 on the other hand, different types of tourism affect the residents’ attitudes (murphy, 1985; williams & lawson, 2001) when it involves different impacts on the local environment and interactions between tourists and residents. however, the vast majority of studies on residents´ attitudes analyse the tourism sector as a whole. only a few cases have focused on a particular type of tourism or activity. the only exceptions are resorts (hernandez, cohen & garcia, 1996), national parks (hammitt, bixler & noe, 1996; holladay & ormsby, 2011; jurowski, uysal, williams &noe, 1995; sekhar, 2003; walpole & goodwin, 2001), dark tourism (chen, wang & xu, 2017), hunting (mackay & campbell, 2004) and casinos (coulter, hermans & parker, 2013; kang, lee, yoon& long, 2008; lee, kang & reisinger, 2010; lockyer, 2012; mccartney & in, 2016; nichols, giacopassi & stitt, 2002; stitt, giacopassi & nichols, 2003; sutton & griffiths, 2008; vong, 2009). the objective of this investigation is to determine how the residents’ perceptions and attitudes affect the level of acceptance of different types of tourism offer (lai & hitchcock, 2017; mccartney & in, 2016; nunkoo & gursoy, 2017) in the case of maldonado-punta del este conurbation. punta del este forms part of the department of maldonado and is the main sun and beach tourism destination in the oriental republic of uruguay. there are 9,200 inhabitants in the city centre of punta del este, but there are more than 100,000 residents in the maldonado-punta del este conurbation. punta del este receives more than 689,000 tourists annually during the high season from december to february (ministerio de turismo, 2017). the different types of tourism offers analysed are sports tourism, maritime tourism (sailing and cruises), nature tourism and sun and beach tourism. the analysis of the data has been carried out through the partial least squares (pls) regression, specifically the statistical software smartpls 2.0 (ringle, wende & will, 2005). the objective of this investigation study is to analyse whether there is a relationship between the residents’ attitudes towards tourism in general and the degree of acceptance of different types of tourism offer. this article has four sections. the first section being the literature review, the second section showing the methodology used the third being the analysis of the results and the fourth section gathers the conclusions. the bibliographic references can be found at the end. 2. literature review investigations carried out on residents’ attitudes towards tourism have grouped the impacts perceived into three or four categories within the economic, social, cultural and environmental framework (andereck et al., 2005; ayres, 2000; gursoy & rutherford, 2004; gursoy et al., 2002; özel & kozak, 2017). the economic benefits stand out the most amongst, the four types of benefits analyzed in the literature review (besculides et al., 2002; bruner, 1996; gursoy et al., 2002; jurowski et al., 1997; madrigal, 1993; milman & pizam, 1988; ramón, álvarez & sánchez, 2018; teye et al., 2002); mainly the creation of job opportunities (besculides et al., 2002; bruner, 1996; gursoy et al., 2002; teye et al., 2002). however, the other benefits must j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 32 not be disregarded, which leads to the proposed hypothesis 1: hypothesis 1: the perception of the benefits generated by tourism has a positive effect on the general attitude towards tourism. hypothesis 1.1: the perception of the economic benefits generated by tourism has a positive effect on the general attitude towards tourism. hypothesis 1.2: the perception of the social benefits generated by tourism has a positive effect on the general attitude towards tourism. hypothesis 1.3: the perception of the cultural benefits generated cultural by tourism has a positive effect on the general attitude towards tourism. hypothesis 1.4: the perception of the environmental benefits generated by tourism has a positive effect on the general attitude towards tourism. as a counterpart of the benefits, the perceived costs have a significant and negative effect on attitudes towards tourism (jurowski et al., 1997; milman & pizam, 1988; prentice, 1993; ramón et al., 2018). the perception of the costs along with the perception of benefits allows residents to carry out an overall assessment of tourism, which in turn determines their attitude towards the sector. based on the existing literature review (bujosa & rosselló, 2007; jurowski et al., 1997; king, pizam & milman, 1993; long et al., 1990; milman & pizam, 1988; prentice, 1993), the perception of the existence of costs derived from tourism leads to a less favourable assessment and a more negative attitude towards tourism. the following hypothesis 2 is proposed: hypothesis 2: the perception of the costs generated by tourism has a negative effect on the general attitude towards tourism. hypothesis 2.1: the perception of the economic costs generated by tourism has a negative effect on the general attitude towards tourism. hypothesis 2.2: the perception of the social costs generated by tourism has a negative effect on the general attitude towards tourism. hypothesis 2.3: the perception of cultural costs generated by tourism has a negative effect on the general attitude towards tourism. hypothesis 2.4: the perception of environmental costs generated by tourism has a negative effect on the general attitude towards tourism. the differences in the behaviour of tourists at the destination, when studying the different types of tourism, must be taken into account as it is the element which has the most significant effect on residents´ attitudes. a tourist who shows respect and interest in the local culture and society is seen differently from a tourist seeking only freedom at low prices. plog (1974) establishes typologies as to the reason why tourists choose their destination: allocentrics (tourists who search for unknown places with no tourism development) and psychocentrics j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 33 (tourists who escape from unknown places and visit consolidated destinations). tourists with cultural and adventurous motivations tend to show greater respect for the residents and take part in more activities during their stay. on the other hand, there are tourists with no interest in the local culture and who want to rest (gómez, san martín & bertiche, 1999). this type of tourist considers that the tourist destination, including residents, should adapt to the tourists and not vice versa. some tourists have anti-tourist attitudes(bruckner & finkielkraut, 1979; doran, larsen & wolff, 2015; gration, raciti & arcodia, 2011; gustafson, 2002; jacobsen, 2000; yu, kim, chen & schwartz, 2012)and do not want to mix with tourist masses, i.e. they want to live an original experience which is not lived by mass tourism, since tourists are not considered (hennig, 1997). anti-tourist attitudes have become a symbol of prestige since this type of tourist does not want to be an ordinary tourist (bruckner & finkielkraut, 1979). their behaviour emphasizes individualism and has elements in common with allocentric tourists such as the adventurous spirit and the taste for freedom which travelling independently gives (plog, 1974, 2002). the type and intensity of contact between visitors and residents are important variables in determining the residents’ attitudes (pearce, 1996), especially when the resident works in the sector or is economically dependent on tourism (milman & pizam, 1988). the interaction between the resident population and visitors is one of the most important factors when determining the perceptions and attitudes of residents (murphy, 1985), as well as determining the levels of tourist satisfaction through the perception obtained from their destination (gómez et al., 1999; nyaupane, teye & paris, 2008; yu & lee, 2014). the degree of acceptance of a type of tourism on behalf of the residents depends on the benefits and costs generated and the tourists’ attitudes towards the destination they have visited (milman & pizam, 1988; pearce, 1996). it is expected that the residents with a more positive attitude towards tourism have higher levels of acceptance of different types of tourism although these causal relationships may depend on the type of offer analysed. the following hypothesis 3 is proposed: hypothesis 3: the general attitude towards tourism increases the acceptance of the different tourist offers. sports tourism includes multiple types of sports. golf stands out due to its impact on the environment and the level of expenditure created by tourists who tend to practice this sport. hypothesis 3.1: the general attitude towards tourism increases the acceptance of sports tourism. along with montevideo, punta del este is one of the two ports in uruguay with an important presence of cruiser liners (ministerio de turismo, 2017). it is also a destination of great interest for nautical tourism due to its international relevance and its geographical location. both types of offers make punta del este a destination with great potential for tourist activities linked to the sea and can be included within the denomination of maritime tourism. hypothesis 3.2: the general attitude towards tourism increases the acceptance of maritime j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 34 tourism. rural tourism and nature-based tourism base their offer on the natural environment and activities involving physical exercise in natural surroundings. hypothesis 3.3: the general attitude towards tourism increases the acceptance of nature tourism. sun and beach tourism is the main offer in punta del este, attracting a high volume of tourists who stay in hotels and rented houses. hypothesis 3.4: the general attitude towards tourism increases the acceptance of sun and beach tourism. the causal model proposed on the basis of the hypotheses proposed is shown in figure 1. they will be subject to analysis for the case of punta del este. figure 1. proposed structural model. source: own elaboration. 3. methodology the data, which was obtained between february and september 2016, is made up of a sample of 420 residents from maldonado-punta del este conurbation. to achieve the highest level of representativeness of the sample and to solve potential bias in the composition of the sample, the demographic parameters of the collated questionnaires were checked. the questionnaire used contained items with likert scales for responses and a set of socio-demographic questions (table 1). the maximum margin of error allowed was 4.88% given a level of confidence of 95%. in the case of the constructs measuring residents’ perceptions and attitudes, the causal analysis used a set of items whose possibility of response consisted of a five-point likert scale; 1 being "totally disagree", 3 "irrelevant" and 5 "strongly agree". for the constructs measuring the j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 35 degree of acceptance of the tourist offers, the likert scale ranged from 1 "unacceptable" to 5 "totally acceptable". based on table 1, a slightly larger presence of men (55%) than women (45%) can be seen in the sample; however, it is a very common difference in this type of field work. the most representative age range is between 25 and 54, the vast majority having secondary and university qualifications. the main profile is characterized by working in the tourism sector and being born outside the studied region. table 1. socio-demographic profile of the samples. frequency % sex: man. 230 54.76 woman. 190 45.24 age less than 25. 104 24.76 from 25 to 34. 67 15.95 from 35 to 44. 81 19.29 from 45 to 54. 80 19.05 from 55 to 64. 49 11.67 65 or more. 39 9.29 level of studies: no studies. 8 1.90 primary studies. 39 9.29 secondary studies. 177 42.14 university studies. 196 46.67 birthplace: in the region. 181 43.10 outside the region. 239 56.90 works in tourism: yes. 243 57.86 no. 177 42.14 source: own elaboration. the research model (figure 1) has been tested using partial least squares (pls) technique (gursoy et al., 2002; lindberg & johnson, 1997; nunkoo & so, 2016; ramón et al., 2018), a variance-based structural equation modelling (sem) method. partial least squares (pls) technique is especially suitable for predictive research and theoretical developments. more precisely, this study uses smartpls 2.0 software for the pls analysis (ringle et al., 2005). while the results differ little for the alternative weighting schemes, path weighting is the recommended approach and the one used in this study. this weighting scheme provides the highest r² value for endogenous latent variables and is generally applicable for all kinds of pls path model specifications and estimations (henseler, ringle & sinkovics, 2009). j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 36 4. results before proceeding to the analysis of the proposed structural model it is necessary to previously analyse the measurement model. individual reliability is considered adequate when an item has a factor loading that is greater than 0.707 on its respective construct (carmines & zeller, 1979). construct reliability is usually assessed using cronbach’s alpha (cronbach, 1970) and composite reliability (anderson & gerbing, 1988; bagozzi & yi, 1988). it can be observed in table 2 that values for cronbach’s alpha and composite reliability are acceptable. table 2. reliability and convergent validity ave composite reliability r2 cronbach’s alpha communality economic benefits 0.599 0.882 0.000 0.832 0.599 social benefits 0.562 0.865 0.000 0.804 0.562 cultural benefits 0.738 0.893 0.000 0.828 0.738 environmental benefits 0.515 0.809 0.000 0.687 0.515 economic costs 0.743 0.850 0.000 0.744 0.743 social costs 0.615 0.757 0.000 0.696 0.615 cultural costs 0.554 0.832 0.000 0.734 0.554 environmental costs 0.596 0.880 0.000 0.835 0.596 general attitude 0.627 0.871 0.363 0.802 0.627 sport tourism 0.767 0.868 0.067 0.699 0.767 maritime tourism 0.747 0.856 0.050 0.662 0.747 nature tourism 0.828 0.906 0.010 0.800 0.828 sun and beach tourism 0.589 0.811 0.103 0.659 0.589 source: own elaboration. to assess convergent validity must be examined the average variance extracted (ave). ave values should be greater than 0.50 (baggozi & yi, 1998) and are greater (table 2). there are two approaches to assess discriminate validity (anderson & gerbing, 1988) in pls: no item should load more highly on another construct than it does on the construct it intends to measure; the square root of the ave of each latent variable should be greater than its correlations with any other latent variable in the assessment (chin, 1998). table 3 illustrates the final measurement model used. table 3. loadings of structural models. arithmetic average standard deviation loadings economic benefits: tourism generates many job opportunities for residents. 4.350 0.780 0.776 tourism generates numerous business opportunities for residents and small businesses. 4.195 0.837 0.838 tourism generates greater opportunities for investment in the town. 4.264 0.798 0.780 tourism generates revenue for the administration and local organisms. 4.293 0.818 0.734 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 37 arithmetic average standard deviation loadings tourism significantly increases residents’ levels of income. 4.048 0.940 0.737 social benefits: thanks to tourism residents have a better and wider range of leisure and entertainment. 3.740 1.072 0.685 thanks to tourism basic services are better. 2.967 1.276 0.777 thanks to tourism there are better public services. 2.855 1.225 0.786 tourism promotes the restoration and conservation of historical heritage. 3.274 1.150 0.738 tourism improves the quality of infrastructure and public works. 3.555 1.062 0.757 cultural benefits: tourism promotes the preservation of local culture. 3.062 1.081 0.914 tourism promotes the preservation of local traditions. 3.000 1.123 0.911 tourism promotes the cultural identity of citizens. 3.167 1.100 0.741 environmental benefits: tourism encourages the protection of natural areas. 3.379 1.101 0.762 tourism converts natural resources into a source of income for residents. 3.652 1.009 0.688 tourism is less polluting than other economic activities. 3.429 1.118 0.714 tourism promotes respect for the environment. 2.936 1.101 0.705 economic costs: tourism has led to an increase in prices and the cost of living. 4.076 1.007 0.716 tourism has led to an increase in the cost of housing and land. 4.136 0.945 0.987 social costs: tourism has increased the levels of public insecurity. 3.119 1.149 0.892 tourism has increased the discomfort to residents. 3.548 1.106 0.659 cultural costs: tourism hinders the enjoyment of public spaces by overcrowding them. 3.269 1.164 0.697 tourism has made residents feel like strangers in their own town. 2.817 1.145 0.826 tourism has generated a negative effect on local culture. 2.588 1.023 0.772 tourism has generated conflicts between visitors and residents. 2.969 1.103 0.674 environmental costs: tourism causes serious environmental pollution problems. 3.057 1.052 0.782 tourism leads to the loss of local ecosystems. 3.002 1.025 0.841 tourism consumes resources in excess. 3.319 1.064 0.747 tourism has contributed to the degradation of the natural environment of the town. 3.062 1.063 0.806 tourism has caused the saturation of some natural areas. 3.362 1.077 0.675 general attitude: tourism development has been very beneficial to the town and its inhabitants. 4.093 0.836 0.772 tourism must continue to be promoted as an essential part of the town. 4.300 0.802 0.828 tourism is beneficial for residents’ day to day lives. 3.988 0.930 0.802 there is a better quality of life thanks to tourism. 3.848 0.996 0.765 sport tourism: golf tourism. 3.781 1.012 0.849 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 38 arithmetic average standard deviation loadings sport tourism. 4.067 0.934 0.902 maritime tourism: cruise ship tourism. 3.976 1.095 0.874 nautical tourism. 3.964 0.964 0.855 nature tourism: nature-based tourism. 4.012 1.006 0.872 agro-tourism, rural tourism, etc. 3.848 1.065 0.946 sun and beach tourism: holiday houses and apartments for rent. 4.150 0.822 0.702 second home tourism. 3.990 0.902 0.754 family sun and beach tourism. 4.533 0.788 0.841 source: own elaboration. after testing the measurement model, proceeded to the analysis of the structural model and the proposed causal relationships. pls-sem does not assume that the data is normally distributed, which implies that parametric significance tests cannot be applied to test whether coefficients such as outer loadings and path coefficients are significant. instead, pls-sem relies on a nonparametric bootstrap procedure (davison & hinkley, 1997; efron & tibshirani, 1993) to test the significance of estimated path coefficients. through bootstrapping, 5,000 subsamples are created with randomly drawn observations from the original set of data (with replacement). the subsample is then used to estimate the pls path model by calculating the average values of the parameters obtained in the 5,000 samples and compared with those obtained from the original set of data. the parameters estimated from the subsamples are used to derive standard errors for the estimates. with this information, t-values are calculated to assess each estimate's significance (hair, hult, ringle & sarstedt, 2014).to determine the critical values, a student’s t distribution with 4,999 degrees of freedom and one tail has been used (as the direction of the relationship was defined). significance analysis results for the different direct causal relationships, both through the use of student's t values and using nonparametric techniques (henseler et al., 2009), are detailed in table 4. based on the results (table 4), only the economic benefits (hypothesis 1.1) have a significant and important effect on the general attitude of the residents of maldonado-punta del este conurbation. this reminds us of the importance that the economic improvements most directly perceived by residents have on their attitude towards the sector (gursoy et al., 2002; jurowski et al., 1997; lankford & howard, 1994; teye et al., 2002). social benefits (hypothesis 1.2) do not have a significant effect on residents’ general attitude, therefore differing from the results of some previous studies (gursoy et al., 2002; lankford & howard, 1994). cultural benefits (hypothesis 1.3) and environmental benefits (hypothesis 1.4) did not have significant effects on the general attitude towards tourism either. table 4. path coefficients. j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 39 path coefficients standard error t-statistic p-value economic benefits general attitude (h1.1) 0.373*** 0.115 3.245 0.001 social benefits general attitude (h1.2) 0.098ns 0.114 0.857 0.196 cultural benefits general attitude (h1.3) 0.036ns 0.095 0.382 0.351 environmental benefits general attitude (h1.4) 0.067ns 0.109 0.615 0.269 economic costsgeneral attitude (h2.1) 0.134ns 0.133 1.015 0.155 social costsgeneral attitude (h2.2) 0.018ns 0.098 0.184 0.427 cultural costsgeneral attitude (h2.3) -0.228* 0.111 2.053 0.020 environmental costsgeneral attitude (h2.4) -0.073ns 0.105 0.697 0.243 general attitude sport tourism (h3.1) 0.258** 0.090 2.863 0.002 general attitude maritime tourism (h3.2) 0.224* 0.107 2.098 0.018 general attitude nature tourism (h3.3) 0.102ns 0.124 0.823 0.205 general attitude sun and beach tourism (h3.4) 0.321*** 0.090 3.589 0.000 *p<0.05; **p<0.01; ***p<0.001; ns not significant. source: own elaboration. economic costs (hypothesis 2.1), social costs (hypothesis 2.2) and environmental costs (hypothesis 2.4) do not have a significant effect on the general attitude towards tourism (table 4), not coinciding with the results of previous studies (jurowski et al., 1997; long et al., 1990; milman & pizam, 1988; prentice, 1993). only cultural costs (hypothesis 2.3) have a negative and significant effect on the general attitude. the general attitude of residents towards tourism has significant and positive effects (table 4) on the sun and beach tourism (hypothesis 3.4), sports tourism (hypothesis 3.1) and maritime tourism (hypothesis 3.2), but there is no significant effect on nature tourism (hypothesis 3.3). the two most important (β > 0.250) and significant (p > 0.01) effects from these are on the sun and beach tourism and sports tourism. it should be considered if the explanatory power of the degree of acceptance of the various analysed tourist offers would improve by implementing a direct relation between the perception of benefits and costs generated by tourism and the degree of acceptance of different types of tourism. in order to do so, the analysis was repeated with the new structural models. table 5 shows that the vast majority of causal relations raised are not significant: • sport tourism: only cultural costs show a negative and significant effect. • maritime tourism: there is a positive and significant effect of economic benefits and a negative and significant effect of environmental costs. • natural tourism: no causal relation is significant in this case. • sun and beach tourism: the economic benefits have a positive and significant effect and cultural costs have a negative and significant effect. the few significant causal relations have significance levels that do not reach the level 0.01, indicating that they are possibly causal relations of little importance and not therefore not possible to generalize. table 5. alternative models. j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 40 path coefficients standard error t-statistic p-value economic benefits sport tourism 0.089ns 0.125 0.715 0.237 social benefits sport tourism 0.051ns 0.139 0.367 0.357 cultural benefits sport tourism 0.080ns 0.133 0.599 0.275 environmental benefits sport tourism 0.104ns 0.135 0.768 0.221 economic costs sport tourism 0.053ns 0.108 0.487 0.313 social costs sport tourism 0.007ns 0.143 0.049 0.481 cultural costs sport tourism -0.197ᶲ 0.135 1.454 0.073 environmental costs sport tourism 0.036ns 0.128 0.279 0.390 economic benefits maritime tourism 0.165ᶲ 0.128 1.289 0.099 social benefits maritime tourism -0.005ns 0.141 0.035 0.486 cultural benefits maritime tourism 0.128ns 0.115 1.108 0.134 environmental benefits maritime tourism 0.089ns 0.127 0.706 0.240 economic costs maritime tourism 0.078ns 0.111 0.702 0.241 social costs maritime tourism 0.019ns 0.146 0.127 0.449 cultural costs maritime tourism 0.053ns 0.136 0.389 0.349 environmental costs maritime tourism -0.237* 0.121 1.966 0.025 economic benefits nature tourism 0.145ns 0.142 1.021 0.154 social benefits nature tourism -0.062ns 0.165 0.376 0.353 cultural benefits nature tourism 0.098ns 0.136 0.720 0.236 environmental benefits nature tourism 0.038ns 0.143 0.267 0.395 economic costs nature tourism 0.041ns 0.132 0.313 0.377 social costs nature tourism 0.034ns 0.134 0.253 0.400 cultural costs nature tourism -0.097ns 0.134 0.724 0.235 environmental costs nature tourism 0.038ns 0.128 0.301 0.382 economic benefits sun and beach tourism 0.173ᶲ 0.128 1.352 0.088 social benefits sun and beach tourism 0.087ns 0.136 0.639 0.261 cultural benefits sun and beach tourism -0.079ns 0.123 0.640 0.261 environmental benefits sun and beach tourism 0.009ns 0.140 0.064 0.474 economic costs sun and beach tourism 0.100ns 0.115 0.870 0.192 social costs sun and beach tourism -0.061ns 0.137 0.442 0.329 cultural costs sun and beach tourism -0.218* 0.130 1.685 0.046 environmental costs sun and beach tourism -0.081ns 0.120 0.673 0.250 ᶲp<0.1; *p<0.05; ns not significant. source: own elaboration. the alternative analysis proposed makes it possible to note that the causal model initially proposed has a higher explanatory capacity about residents’ behaviour than the alternative approach with different types of offers posed as a variable directly dependent on the general perceptions of benefits and costs. although the initially proposed model has few significant causal relationships (table 6), the suggested alternative models have fewer significant relationships. moreover, in alternative models, the degree of significance of causal relations is very low and does not allow the existence of causal relations to be affirmed. it is very possible that there is no existence of the causal relations proposed in the alternative models. table 6. hypotheses summary table. j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 41 hypotheses accepted rejected 1: the perception of the benefits generated by tourism has a positive effect on the general attitude towards tourism. 1.1: the perception of the economic benefits generated by tourism has a positive effect on the general attitude towards tourism. x 1.2: the perception of the social benefits generated by tourism has a positive effect on the general attitude towards tourism. x 1.3: the perception of the cultural benefits generated cultural by tourism has a positive effect on the general attitude towards tourism. x 1.4: the perception of the environmental benefits generated by tourism has a positive effect on the general attitude towards tourism. x 2: the perception of the costs generated by tourism has a negative effect on the general attitude towards tourism. 2.1: the perception of the economic costs generated by tourism has a negative effect on the general attitude towards tourism. x 2.2: the perception of the social costs generated by tourism has a negative effect on the general attitude towards tourism. x 2.3: the perception of cultural costs generated by tourism has a negative effect on the general attitude towards tourism. x 2.4: the perception of environmental costs generated by tourism has a negative effect on the general attitude towards tourism. x 3: the general attitude towards tourism increases the acceptance of the different tourist offers. 3.1: the general attitude towards tourism increases the acceptance of sports tourism. x 3.2: the general attitude towards tourism increases the acceptance of maritime tourism. x 3.3: the general attitude towards tourism increases the acceptance of nature tourism. x 3.4: the general attitude towards tourism increases the acceptance of sun and beach tourism. x source: own elaboration. the results of this work indicate that attitudes towards different types of tourism are created in two phases. firstly, tourism is valued as a whole taking into account the economic benefits it generates and the cultural costs it causes, mainly being the inconvenience and discomfort, it generates (table 6). although it is very likely to take the current tourism model of the destination as a reference to create attitudes towards tourism, it is not possible to exchange a particular type of tourism with the global tourism sector as a dependent variable. in the second phase, the residents decide the combination of offers they prefer for the tourism sector; therefore, the overall valuation of tourism has different causal effects on the four types of offer analysed. these causal effects show recognition of the possibilities of each type of tourism in the region and its quantitative importance. the greater the potential of the destination to develop a higher type of tourism, the greater the causal effect. in the case of maldonado-punta del este, a conurbation located on the coast, there are significant effects on the sun and beach tourism (the main in punta del este) and maritime and sports tourism, but not on nature tourism. the latter has more limited possibilities in this type of tourist destination. j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 42 5. conclusions when analysing the effect of residents’ attitude on the degree of acceptance of the different tourist offers it can be seen that there are no significant effects on nature tourism. in the other three types of tourism, the relations are significant, although the effects are more important and significant in the case of sun and beach tourism. as a result of the causal analysis carried out, the proposed model was reduced to the causal relationships shown in figure 2, once the nonsignificant causal relationships were eliminated. nature tourism has a high degree of acceptance by residents but it is a response which has little to do with the tourism analysis process carried out by residents. it is possibly a more "automatic" response and caused less by a critical analysis of tourism. figure 2. causal relations preserved. source: own elaboration. the main limitation of the study is that it is a case study: the analysis has been carried out on a sample of a specific tourist destination. another limitation is that other types of tourism offers have not been taken into account, although they are more than in previous studies, and the measurement scales used may have influenced the result. even so, the analysis provides the main result of the importance of separating the general attitude toward tourism from the concrete attitudes towards a specific type of offer. this implies that the managers of the tourist destinations must carry out two complementary analyses: measure the residents’ attitudes towards the tourism sector in general, to determine the support towards this type of activity; determine the degree of acceptance of the various types of tourism offer and whether this acceptance depends on the general attitude towards the sector or is independent. if it is independent of the general attitude, it is possible that it is a "learned" or "politically correct" support but not based on a rational analysis of tourism activity. when it depends on the general j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 43 attitude, it would be showing a preference for the short and medium term development of tourism in the region. but these are only hypotheses of interpretation that should be developed in future studies. based on the results, managers from the tourism sector are recommended to maximize the economic benefits which have a direct impact on the residents. this would involve prioritizing the endogenous development of the sector (workers, entrepreneurs and local investors), and minimizing the negative consequences of contact between tourists and residents: calculating the capacity of the region and not exceeding it; prioritizing tourists who are respectful of the places visited and who have similar cultures to that of the residents. when tourists are from countries which are culturally different from the local society, it is important to diversify tourism so that local culture remains the main culture in the destination and reduces the alienation of the resident in their own land. future studies should repeat the analysis with new types of tourism to try to determine when the acceptance of a type of tourism is the result of the attitude towards tourism or a response with no connection to the residents’ cost-benefit evaluation. in relation to this point, it should be analysed if the degree of acceptance of offers with no connection to the general attitude towards tourism is due to a learned response or a perception of respectful tourism with the local inhabitants. references almeida, f., balbuena, a., & cortés, r. (2015). resident's attitudes towards the impacts of tourism. tourism management perspectives, 13, 33-40. doi: https://doi.org/10.1016/j.tmp.2014.11.002 andereck, k. l., valentine, k. m., knoff, r. c., & vogt, c. a. (2005). residents’ perceptions of community tourism impacts. annals of tourism research, 32(4), 1056-1076. doi: https://doi.org/10.1016/j.annals.2005.03.001 anderson, j. c., & gerbing, d. w. (1988). structural equation modeling in practice: a review and recommended two-step approach. psychological bulletin, 103, 411–423. doi: https://doi.org/10.1037/0033-2909.103.3.411 ayres, r. (2000). tourism as a passport to development in small states: reflections on cyprus. international journal of social economics, 27(2), 114-133. doi: https://doi.org/10.1108/03068290010308992 bagozzi, r. p.,& yi, y. (1988). on the evaluation of structural equation models. journal of the academy of marketing science, 16(2), 74-94. doi: https://doi.org/10.1007/bf02723327 besculides, a., lee, m., & mccormick, p. (2002). resident's perceptions of the cultural benefits of tourism. annals of tourism research, 29(2), 303-319. doi: https://doi.org/10.1016/s01607383(01)00066-4 https://doi.org/10.1016/j.tmp.2014.11.002 https://doi.org/10.1016/j.annals.2005.03.001 https://doi.org/10.1037/0033-2909.103.3.411 https://doi.org/10.1108/03068290010308992 https://doi.org/10.1007/bf02723327 https://doi.org/10.1016/s0160-7383(01)00066-4 https://doi.org/10.1016/s0160-7383(01)00066-4 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 44 bruckner, p., & finkielkraut, a. (1979). au coin de la rue. l'aventure. paris: èditions du seuil. bruner, e. (1996). tourism in ghana: the representation of slavery and the return of the black diaspora. american anthropologist, 98(2), 290-304. doi: https://doi.org/10.1525/aa.1996.98.2.02a00060 bujosa, a., & rosselló, j. (2007). modeling environmental attitudes toward tourism. tourism management, 28(3), 688-695. doi: https://doi.org/10.1016/j.tourman.2006.04.004 carmines, e., & zeller, r. (1979). reliability and validity assessment, n. 07-017. sage university paper series on quantitative applications in the social sciences. beverly hills, ca: sage. doi: https://doi.org/10.4135/9781412985642 chen, s., wang, s., & xu, h. (2017). influence of place identity on residents’ attitudes to dark tourism. journal of china tourism research, 13(4), 338-356. doi: https://doi.org/10.1080/19388160.2017.1401023 chin, w. w. (1998). issues and opinions on structural equation modeling. mis quarterly, 22(1), 7-16. coulter, r. l., hermans, c. m., & parker, r. s. (2013). religiosity and generational effects on gambling: support for and opposition to introducing casino gambling in a non-gambling tourist entertainment environment. journal of business research, 66(9), 1682-1688. doi: https://doi.org/10.1016/j.jbusres.2012.12.015 cronbach, l. j. (1970). essentials of psychological testing. 3rd ed. new york, ny: harper & row. davison, a., & hinkley, d. (1997). bootstrap methods and their application. cambridge: cambridge university press. doi: https://doi.org/10.1017/cbo9780511802843 doran, r., larsen, s., & wolff, k. (2015). different but similar: social comparison of travel motives among tourists. international journal of tourism research, 17(6), 555-563. doi: https://doi.org/10.1002/jtr.2023 efron, b.,& tibshirani, r. j. (1993). an introduction to the bootstrap. new york, ny: chapman hall. doi: https://doi.org/10.1007/978-1-4899-4541-9 gómez, l., san martín, j., & bertiche, c. (1999). a model of tourism experience and attitude change. annals of tourism research, 26(4), 1024-1027. doi: https://doi.org/10.1016/s01607383(99)00063-8 gration, d., raciti, m., & arcodia, c. (2011). the role of consumer self-concept in marketing festivals. journal of travel and tourism marketing, 28(6), 644-655. doi: https://doi.org/10.1080/10548408.2011.603631 gursoy, d., & rutherford, d. (2004). host attitudes toward tourism: an improved structural model. annals of tourism research, 31(3), 495-516. doi: https://doi.org/10.1016/j.annals.2003.08.008 gursoy, d., jurowski, c., & uysal, m. (2002). resident attitudes: a structural modeling approach. annals of tourism research, 29(1), 79-105. doi: https://doi.org/10.1016/s01607383(01)00028-7 gustafson, p. (2002). tourism and seasonal retirement migration. annals of tourism research, https://doi.org/10.1525/aa.1996.98.2.02a00060 https://doi.org/10.1016/j.tourman.2006.04.004 https://doi.org/10.4135/9781412985642 https://doi.org/10.1080/19388160.2017.1401023 https://doi.org/10.1016/j.jbusres.2012.12.015 https://doi.org/10.1017/cbo9780511802843 https://doi.org/10.1002/jtr.2023 https://doi.org/10.1007/978-1-4899-4541-9 https://doi.org/10.1016/s0160-7383(99)00063-8 https://doi.org/10.1016/s0160-7383(99)00063-8 https://doi.org/10.1080/10548408.2011.603631 https://doi.org/10.1016/j.annals.2003.08.008 https://doi.org/10.1016/s0160-7383(01)00028-7 https://doi.org/10.1016/s0160-7383(01)00028-7 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 45 29(4), 899-918. doi: https://doi.org/10.1016/s0160-7383(01)00084-6 hair, j., hult, g., ringle, c., & sarstedt, m. (2014). a primer on partial least squares structural equation modeling (pls-sem). thousand oaks, ca: sage. hammitt, w. e., bixler, r. d., & noe, f. p. (1996). going beyond importance-performance analysis to analyze the observance: influence of park impacts. journal of park and recreation administration, 14(1), 45-62. hennig, c. (1997). touristenbeschimpfung. zur geschichte des anti-tourismus. zeitschrift für volkskunde, 93(1), 31-41. henseler, j., ringle, c. m., & sinkovics, r. r. (2009). the use of partial least squares path modeling in international marketing. in r. r. sinkovics, &p. n. ghauri (ed.), advances in international marketing (pp. 277-320). bingley, wa: emerald. doi: https://doi.org/10.1108/s1474-7979(2009)0000020014 hernandez, s. a., cohen, j., & garcia, h. l. (1996). residents' attitudes towards an instant resort enclave. annals of tourism research, 23(4), 755-779. doi: https://doi.org/10.1016/01607383(95)00114-x holladay, p. j., & ormsby, a. a. (2011). a comparative study of local perceptions of ecotourism and conservation at five blues lake national park, belize. journal of ecotourism, 10(2), 118-134. doi: https://doi.org/10.1080/14724049.2010.529910 jacobsen, j. k. s. (2000). anti-tourist attitudes: mediterranean charter tourism. annals of tourism research, 27(2), 284-300. doi: https://doi.org/10.1016/s0160-7383(99)00069-9 jurowski, c., uysal, m., & williams, r. d. (1997). a theoretical analysis of host community resident reactions to tourism. journal of travel research, 36(2), 3-11. doi: https://doi.org/10.1177/004728759703600202 jurowski, c., uysal, m., williams, r. d., & noe, f. p. (1995). an examination of preferences and evaluations of visitors based on environmental attitudes: biscayne bay national park. journal of sustainable tourism, 3(2), 73-86. doi: https://doi.org/10.1177/004728759703600202 kang, s. k., lee, c., yoon, y.,& long, p. t. (2008). resident perception of the impact of limited-stakes community-based casino gaming in mature gaming communities. tourism management, 29(4), 681-694. doi: https://doi.org/10.1016/j.tourman.2007.07.011 king, b., pizam, a., & milman, a. (1993). social impacts of tourism: host perceptions. annals of tourism research, 20(4), 650-665. doi: https://doi.org/10.1016/0160-7383(93)90089-l lai, i. k. w., & hitchcock, m. (2017). local reactions to mass tourism and community tourism development in macau. journal of sustainable tourism, 25(4), 451-470. doi: https://doi.org/10.1080/09669582.2016.1221413 lankford, s. v., & howard, d. r. (1994). developing a tourism impacts attitude scale. annals of tourism research, 21(1), 121-139. doi: https://doi.org/10.1016/0160-7383(94)90008-6 lee, c. k., kang, s., & reisinger, y. (2010). community attachment in two rural gaming communities: comparisons between colorado gaming communities, usa and gangwon gaming communities, south korea. tourism geographies, 12(1), 140-168. doi: https://doi.org/10.1016/s0160-7383(01)00084-6 https://doi.org/10.1108/s1474-7979(2009)0000020014 https://doi.org/10.1016/0160-7383(95)00114-x https://doi.org/10.1016/0160-7383(95)00114-x https://doi.org/10.1080/14724049.2010.529910 https://doi.org/10.1016/s0160-7383(99)00069-9 https://doi.org/10.1177/004728759703600202 https://doi.org/10.1177/004728759703600202 https://doi.org/10.1016/j.tourman.2007.07.011 https://doi.org/10.1016/0160-7383(93)90089-l https://doi.org/10.1080/09669582.2016.1221413 https://doi.org/10.1016/0160-7383(94)90008-6 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 46 https://doi.org/10.1080/14616680903493688 lindberg, k., & johnson, r. l. (1997). modeling residents attitudes toward tourism. annals of tourism research, 24(2), 402-424. doi: https://doi.org/10.1016/s0160-7383(97)80009-6 lockyer, t. (2012). perceptions of the hamilton new zealand casino upon local residents. tourism review, 67(4), 12-25. doi: https://doi.org/10.1108/16605371211277786 mackay, k. j., & campbell, j. m. (2004). an examination of residents’ support for hunting as a tourism product. tourism management, 25(4), 443-452. doi: https://doi.org/10.1016/s02615177(03)00127-4 madrigal, r. (1993). a tale of tourism in two cities. annals of tourism research, 20(2), 336-353. doi: https://doi.org/10.1016/0160-7383(93)90059-c mason, p., & cheyne, j. (2000). resident's attitudes to proposed tourism development. annals of tourism research, 27(2), 391-411. doi: https://doi.org/10.1016/s0160-7383(99)00084-5 mccartney, g., & in, w. l. w. (2016). house of cards an analysis of macao’s resident support for tourism and casino development. tourism and hospitality research, 16(2), 144-157. doi: https://doi.org/10.1177/1467358415600215 milman, a., & pizam, a. (1988). social impact of tourism on central florida. annals of tourism research, 15(2), 191-204. doi: https://doi.org/10.1016/0160-7383(88)90082-5 ministerio de turismo (2017). anuario 2017: estadísticas de turismo. montevideo: ministerio de turismo. retrieved form http://www.mintur.gub.uy/ [september 10, 2018] murphy, p. e. (1985). tourism: a community approach. new york, ny: routledge. nichols, m., giacopassi, d., & stitt, b. g. (2002). casino gambling as a catalyst of economic development: perceptions of residents in new casino jurisdictions. tourism economics, 8(1), 59-76. doi: https://doi.org/10.5367/000000002101297990 nunkoo, r., & gursoy, d. (2017). political trust and residents’ support for alternative and mass tourism: an improved structural model. tourism geographies, 19(3), 318-339. doi: https://doi.org/10.1080/14616688.2016.1196239 nunkoo, r., & so, k. k. f. (2016). residents’ support for tourism: testing alternative structural models. journal of travel research, 55(7), 847-861. doi: https://doi.org/10.1177/0047287515592972 nyaupane, g. p., teye, v., & paris, c. (2008). innocents abroad:attitude change toward hosts. annals of tourism research, 35(3), 650-667. doi: https://doi.org/10.1016/j.annals.2008.03.002 özel, ç. h., & kozak, n. (2017). an exploratory study of resident perceptions toward the tourism industry in cappadocia: a social exchange theory approach. asia pacific journal of tourism research, 22(3), 284-300. doi: https://doi.org/10.1080/10941665.2016.1236826 pearce, d. g. (1996). tourist organizations in sweden. tourism management, 17(6), 413-424. doi: https://doi.org/10.1016/0261-5177(96)00050-7 plog, s. c. (1974). why destination areas rise and fall in popularity. cornell hotel and restaurant administration quarterly, 14(4), 55-58. doi: https://doi.org/10.1177/001088047401400409 https://doi.org/10.1080/14616680903493688 https://doi.org/10.1016/s0160-7383(97)80009-6 https://doi.org/10.1108/16605371211277786 https://doi.org/10.1016/s0261-5177(03)00127-4 https://doi.org/10.1016/s0261-5177(03)00127-4 https://doi.org/10.1016/0160-7383(93)90059-c https://doi.org/10.1016/s0160-7383(99)00084-5 https://doi.org/10.1177/1467358415600215 https://doi.org/10.1016/0160-7383(88)90082-5 https://doi.org/10.5367/000000002101297990 https://doi.org/10.1080/14616688.2016.1196239 https://doi.org/10.1177/0047287515592972 https://doi.org/10.1016/j.annals.2008.03.002 https://doi.org/10.1080/10941665.2016.1236826 https://doi.org/10.1016/0261-5177(96)00050-7 https://doi.org/10.1177/001088047401400409 j.r. cardona et al. / european journal of government and economics 8(1), june 2019, 30-47. 47 plog, s. c. (2002). the power of psychographics and the concept of venturesomeness. journal of travel research, 40(3), 244-251. doi: https://doi.org/10.1177/004728750204000302 prentice, r. (1993). community-driven tourism planning and residents' preferences. tourism management, 14(3), 218-227. doi: https://doi.org/10.1016/0261-5177(93)90023-e ramón, j., álvarez, d., & sánchez, m. d. (2018). principales causantes de la actitud de los residentes: comparativa entre dos destinos de playa. revista portuguesa de estudos regionais, 47, 37-48. ringle, c. m., wende, s., & will, s. (2005). smartpls 2.0 (beta). hamburg: smartpls. sekhar, n. u. (2003). local people's attitudes towards conservation and wildlife tourism around sariska tiger reserve, india. journal of environmental management, 69(4), 339-347. doi: https://doi.org/10.1016/j.jenvman.2003.09.002 sharpley, r. (2014). host perceptions of tourism: a review of the research. tourism management, 42, 37-49. doi: https://doi.org/10.1016/j.tourman.2013.10.007 stitt, b. g., giacopassi, d., & nichols, m. (2003). gambling among older adults: a comparative analysis. experimental aging research, 29(2), 189-203. doi: https://doi.org/10.1080/03610730303713 sutton, r., & griffiths, m. d. (2008). the casino attitudes scale: the development of a new brief psychometric instrument. international journal of mental health and addiction, 6(2), 244248. doi: https://doi.org/10.1007/s11469-007-9127-z teye, v.,sirakaya, e., & sönmez, s. f. (2002). resident's attitudes toward tourism development. annals of tourism research, 29(3), 668-688. doi: https://doi.org/10.1016/s0160-7383(01)00074-3 vong, f. (2009). changes in residents’ gambling attitudes and perceived impacts at the fifth anniversary of macao’s gaming deregulation. journal of travel research, 47(3), 388-397. doi: https://doi.org/10.1177/0047287508322787 walpole, m. j., & goodwin, h. j. (2001). local attitudes towards conservation and tourism around komodo national park, indonesia. environmental conservation, 28(2), 160-166. doi: https://doi.org/10.1017/s0376892901000169 williams, j., & lawson, r. (2001). community issues and resident opinions of tourism. annals of tourism research, 28(2), 269-290. doi: https://doi.org/10.1016/s0160-7383(00)00030-x yu, j., & lee, t. j. (2014). impact of tourists' intercultural interactions. journal of travel research, 53(2), 225-238. doi: https://doi.org/10.1177/0047287513496467 yu, x., kim, n., chen, c., & schwartz, z. (2012). are you a tourist? tourism definition from the tourist perspective. tourism analysis, 17(4), 445-457. doi: https://doi.org/10.3727/108354212x13473157390687 https://doi.org/10.1177/004728750204000302 https://doi.org/10.1016/0261-5177(93)90023-e https://doi.org/10.1016/j.jenvman.2003.09.002 https://doi.org/10.1016/j.tourman.2013.10.007 https://doi.org/10.1080/03610730303713 https://doi.org/10.1007/s11469-007-9127-z https://doi.org/10.1016/s0160-7383(01)00074-3 https://doi.org/10.1177/0047287508322787 https://doi.org/10.1017/s0376892901000169 https://doi.org/10.1016/s0160-7383(00)00030-x https://doi.org/10.1177/0047287513496467 https://doi.org/10.3727/108354212x13473157390687 front81 ejge oficial 8-1 contents81 number 8, issue 1, june 2019 4580 abstract. the study of residents' attitudes is fundamental for tourist destinations managers. residents form their attitude towards tourism through a cost-benefit analysis based on their perception of tourism impacts. the objective of this paper is to... keywords. attitudes, residents, pls, punta del este. jel codes: c88, l83, z32 doi. https://10.17979/ejge.2019.8.1.4580 1. introduction 2. literature review 3. methodology 4. results 5. conclusions references surveys assessing sports services and municipal governance vol.9 • no.2 2020 issn: 2254-7088 special issue. the role of institutions and governance in sport european journal of government and economics 9(2), july 2020. european journal of government and economics issn: 2254-7088 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport doi: https://doi.org/10.17979/ejge.2020.9.2 how the uefa financial fair play regulations affect to football clubs’ priorities and leagues’ competitive balance? 119-142 doi: https://doi.org/10.17979/ejge.2020.9.2.5842 pedro garcia-del-barrio and giambattista rossi surveys assessing sports services and municipal governance 143-154 doi: https://doi.org/10.17979/ejge.2020.9.2.5949 júlia bosch, laureà fanega, jaume garcía, núria hernández, xavier moya, and carles murillo evaluation of the perceived social impacts of the formula e grand prix of santiago de chile 155-169 doi: https://doi.org/10.17979/ejge.2020.9.2.5850 david parra-camacho, daniel michel duclos bastías, frano giakoni ramírez, and samuel lópez-carril comparative analysis of income trends and perceived value of squad of the highest turnover european football clubs (2010-2019) 170-180 doi: https://doi.org/10.17979/ejge.2020.9.2.5953 benito pérez-gonzález, luis de la riva, josé bonal, and álvaro fernández-luna analysis of loyalty and future intentions of the users of the golf courses in andalusia, spain 181-199 doi: https://doi.org/10.17979/ejge.2020.9.2.5841 marcos pradas garcía, maría josé maciá andreu, marta garcía-tascón, and ana maría gallardo guerrero female leadership in sports clubs 200-209 doi: https://doi.org/10.17979/ejge.2020.9.2.5840 alfonso martínez-moreno, francisco cavas-garcía, francisco cano-noguera, and arturo díaz-suárez the profile of leisure time sports people and their reason for doing sport in spanish sports facilities 210-219 doi: https://doi.org/10.17979/ejge.2020.9.2.5846 moisés grimaldi-puyana, pablo gálvez-ruiz, manel valcarce-torrente, and ainara bernal-garcía https://doi.org/10.17979/ejge.2020.9.2 https://doi.org/10.17979/ejge.2020.9.2.5842 https://doi.org/10.17979/ejge.2020.9.2.5949 https://doi.org/10.17979/ejge.2020.9.2.5850 https://doi.org/10.17979/ejge.2020.9.2.5953 https://doi.org/10.17979/ejge.2020.9.2.5841 https://doi.org/10.17979/ejge.2020.9.2.5840 https://doi.org/10.17979/ejge.2020.9.2.5846 european journal of government and economics 9(2), july 2020, 143-154 european journal of government and economics issn: 2254-7088 surveys assessing sports services and municipal governance júlia bosch a*, laureà fanega b, jaume garcía a, núria hernández b, xavier moya a, carles murillo a a universitat pompeu fabra, spain b l'hospitalet de llobregat city council, spain *corresponding author at: julia.bosch@upf.edu article history. received 10 january 2020; first revision required 13 march 2020; accepted 25 may 2020. abstract. this study aims to investigate the social importance of sport in l’hospitalet de llobregat. as an instrument for social cohesion and integration, sport has become a very important part of municipal activity. we use two surveys: the baròmetre, which gathers opinion about municipal services; and satisfaction surveys completed by the users of sports facilities. the sports offer (6.5) receives a better rating than overall municipal management (5.9), and counts for 20.4% of the global assessment of the municipal management. satisfaction with facilities is 7.3 out of 10, while price receives the lowest mark. detailed analysis of satisfaction surveys conducted among the population can help the municipal management of sports facilities. policies that focus only on prices are unlikely to be as effective at improving members’ perception of sports facilities as those that also take other aspects into account. keywords. local governance; satisfaction surveys; sports facilities; sports policy; sports services jel codes. c13; h41; h70; z28 doi. https://doi.org/10.17979/ejge.2020.9.2.5949 1. introduction the public management of sport facilities requires, first of all, understanding the determinants of the income and expenses these facilities entail. for this, it is essential to be able to characterise the behaviour of the population in general and of members in particular. at the request of l'hospitalet de llobregat city council (catalonia), a study was carried out to analyse the future strategy of its sports policy (bosch et al., 2018). the economic and social dimension of sport in l'hospitalet and its future economic viability cannot be analysed without the explicit recognition of its territorial uniqueness. l'hospitalet has peculiar characteristics due to its geographical location, the environment in which it is located, its size and the density and sociodemographic composition of the population. the assessment of the economic and social dimension of sports practice in the city must be viewed from the perspective of its singularity, without neglecting the reality of neighbouring municipalities (cunningham et al., 1997). the legal nature of the ownership of sports facilities and equipment, the management model (direct or indirect), and the dimensions that configure the users' experience with the service received (measuring user perceptions as an indicator of service quality), are some of the elements that help to define the framework of action for municipal sport in l'hospitalet. the aim of the study was to identify the relationship between the different items that make up municipal mailto:julia.bosch@upf.edu https://doi.org/10.17979/ejge.2020.9.2.5949 bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 144 facilities’ income and expenses, whether they are managed directly or indirectly. as well as understanding the determining factors of income and expenses, and being able to characterise the behaviour of the population in general and of the members in particular, it is important to know the opinions of the city’s inhabitants of its sports offer and its management by the city council. thus, the objective of this article is to investigate the social importance of the sport in l’hospitalet based on two surveys: the baròmetre, which gathers inhabitants’ opinions on municipal services, and the satisfaction surveys conducted on members (users) of the sports facilities. both correspond to 2017. exhaustive analysis of the results of these surveys provides the city council with relevant information that improves the management of its sports facilities and, therefore, municipal governance in the field of the sport. this article has four sections: the first describes the municipal sports policy of l’hospitalet; the second presents the results of the baròmetre; the third examines the results of the user satisfaction surveys; and the final section draws conclusions. 2. sport at l'hospitalet city council sport has an important place on l'hospitalet city council’s local agenda, and the council recognises that it is a growing phenomenon. increased citizen participation in sport has social benefits, and playing sport in free time has been a growing trend in recent years, overcoming the traditional competitive dimension. local sport is a powerful tool for creating positive changes in society, although it has often failed to achieve the social standing it deserves (ajuntament de l’hospitalet, 2013). the quantitative aspects of sport – sports facilities, people who practise sport, sports entities and members, and ultimately, expenses and revenue – are becoming increasingly important in the decision-making and political strategies of public entities. quantifying the supply and demand for local sports services must be accompanied by qualitative elements that describe and evaluate the results of the provision of services, such as the evolution of management models, their utility and efficiency, and, finally, user perceptions and measures of quality. city policy increasingly considers the sporting phenomenon. sport, either practised actively or passively as a spectator at a sport event, is an increasingly unavoidable reality (bosch et al., 2018). this does not mean that local, autonomic or state election programmes have to engage with sporting events and make them an important part of their action programme, although we believe this situation will change based on the evidence of the growing importance of sport in the future (prat et al., 2008). the ownership of sports facilities has a certain uniqueness in catalonia, with most being public property (55% of the total). the role of the city council is therefore crucial because most of these facilities belong to the cities councils. the rest of the spaces are privately owned, distributed in non-profit sports companies, trading societies dedicated to sport, and resorts. the economic and social dimension of sport in l'hospitalet and its future economic viability bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 145 cannot be analysed without the explicit recognition of territorial singularity. the city of l'hospitalet has peculiar characteristics due to its geographical location, the environment in which it is located, its size, population density and sociodemographic composition. the valuation of the economic and social dimension of sports practice in l'hospitalet must be contemplated from the perspective of its singularity, without neglecting the reality of other municipalities in the region and the demarcation of barcelona (cunningham et al., 1997). what is more, the local sports phenomenon must be contemplated within the framework of the current regulations and, especially, catalonia’s law on sport,1 and according to the study of the management of sports facilities, catalonia’s master plan for sports facilities and equipment (fábregas et al., 2005). in addition, the study by l'hospitalet city council (2013), which quantified, using a tool designed by the diputació de barcelona, the economic weight of sport in the city in order to visualise the importance of local sport in the city. the estimation of the economic weight of sport in 2011 can be compared with the local budget (this is the only work of this type carried out by l'hospitalet). the sports budget was 5.8 million euros, 2.19% of the total budget, or 15.28 euros per citizen. the study of economic weight concludes that sport has a weight of almost 18.8 million euros, representing 49.53 euros per citizen. sports facilities and activities organised represented a total of 12.5 million euros (66.8% of the total), while sports entities absorbed 13.4% of the total. in terms of the sources of funding, 23.9% of the total 18.8 million euros was subsidised by the public administrations, mainly the city council (95.4% of the total) and, to a lesser extent, the diputació de barcelona (0.7%) and the generalitat de catalunya (3.9%). the subsidy received by the local sports facilities was 13.2% of the total while in the case of sports entities, the public subsidy accounted for 47.2% of the generated weight. on the other hand, sport has representation in the planning of l’hospitalet city council’s local activities. for instance, the main objectives related to sport are specified in the section on basic rights guaranteed in the 2016–2019 legislature. strategic objective 01.06 specifies the following objectives: to put sport at the service of personal and community growth; to guarantee access to sport and physical activity to the citizens of l'hospitalet in order to improve their life quality; to support sports entities to guarantee their continuity and sustainability; and, last but not least, to develop a master plan for sport in l'hospitalet that will define the strategy of action to ensure the viability of the sports facilities in the city. the strategic objectives of sports tourism should also be taken into consideration in order to help make the city a global leader in sports tourism. 1 llei de l’esport de catalunya: http://portaljuridic.gencat.cat/ca/pjur_ocults/pjur_resultats_fitxa?action=fitxa &documentid=224696 http://portaljuridic.gencat.cat/ca/pjur_ocults/pjur_resultats_fitxa?action=fitxa&documentid=224696 http://portaljuridic.gencat.cat/ca/pjur_ocults/pjur_resultats_fitxa?action=fitxa&documentid=224696 bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 146 3. baròmetre municipal the baròmetre is a survey that gathers the opinions of the inhabitants of l'hospitalet about certain services involved in the municipality’s social life. one is sport, specifically the sports offer. in particular, we want to know what the main concerns of citizens are and the assessment they make of municipal management at a general level and by area of action. table 1 shows the average score by neighbourhood for municipal management as a whole (global) and for the sports offer (sport). it also shows the percentage in each district of scores below an average mark of 5 (fail), those which on average exceed 8 (excellent), and the percentage of participants that answered the question about the sports offer. as can be seen, the average rating of the sports offer is superior to that of overall municipal management (6.5 versus 5.9, respectively). this is the same across all neighbourhoods.2 the neighbourhood where sport is most valued, with an average mark of 7.2, is gornal, which coincides with the neighbourhood with the lowest fail (5.5%) and the highest excellent (26.5%) percentages. curiously, however, at 90.1% it is not the neighbourhood with the highest percentage of answers: 91.9% of respondents answered this question in bellvitge, where the average sport rating was 7. other neighbourhoods with an average mark equal to or above the city as a whole are: centre, sant josep, collblanc, santa eulàlia/granvia sud and can serra. by contrast, the neighbourhoods where sport is less valued, although it still receives a high pass, are sanfeliu, pubilla cases and torrassa. the last column of the table could be used to approach sports participation in each neighbourhood, in the sense that those who did not answer the survey in relation to this topic do not practice sports and, therefore, do not know the municipal facilities. table 1. average assessment of municipal management by neighbourhood (2017). global sport % fail % excellent % answers centre 6.1 6.6 10.2 13.5 85.1 sanfeliu 5.9 6.0 24.0 8.1 76.1 sant josep 5.8 6.5 14.1 10.6 83.3 collblanc 6.2 6.6 10.9 15.0 73.3 la torrassa 5.8 6.2 19.0 7.8 73.3 santa eulàlia/granvia sud 6.0 6.5 12.0 9.0 89.9 la florida 5.6 6.4 13.5 13.9 76.7 les planes 5.9 6.4 16.8 11.9 82.0 can serra 5.9 6.8 9.8 12.4 84.1 pubilla cases 5.7 6.1 14.7 8.7 82.0 el gornal 6.1 7.2 5.5 26.5 90.1 bellvitge 6.2 7.0 7.2 18.1 91.9 city as a whole 5.9 6.5 12.7 12.2 82.6 source: compiled by authors using data from the baròmetre. 2 it should be noted that the most highly valued services are the libraries and the municipal markets, while the sports offer has the same mark as the cultural and leisure offer and garbage collection. bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 147 table 2. average assessment of municipal management by gender (2017). global sport % fail % excellent % answers male 5.9 6.4 15.1 11.8 87.0 female 6.0 6.7 10.3 12.6 78.5 total 5.9 6.5 12.7 12.2 82.6 source: compiled by authors using data from the baròmetre. regarding gender assessment (table 2), women give a higher score both in global management and in the particular case of sport (0.3 higher than men's score). however, the percentage of women's responses is .8.5 points lower than that of men, which could mean a lower level of sports participation. by age group, curiously, it is the extremes – those under 25 and over 65 – who value sport most highly, although the answer rate is below 60%. it should be noted, however, that while younger people value sport far above global management (7.1 versus 6, respectively), the elderly value both categories almost equally (6.8 for sport and 6.7 overall management). the other age groups also score sport higher than global management, with the former receiving a mark above 6, while the latter falls below 6 (table 3). table 3. average assessment of municipal management by age group (2017). global sport % fail % excellent % answers 18–24 years 6.0 7.1 4.0 15.9 92.4 25–34 years 5.8 6.4 12.1 11.5 93.4 35–49 years 5.7 6.3 14.2 10.6 94.6 50–64 years 5.9 6.4 16.7 11.5 81.1 65 and over 6.7 6.8 10.0 15.3 58.8 total 5.9 6.5 12.7 12.2 82.6 source: compiled by authors using data from the baròmetre. the groups that most value sport are retirees and students, with both awarding it almost 7 (6.9). on the other hand, self-employed workers are those who value sport the least, with a mark of 6. over 17% of this group awarded a fail and just over 2% excellent. in any case, once again all people, regardless of their employment situation, rated the sports offer more highly than they did the global management of the city (table 4). the results show that for all assessment criteria (sex, age, employment status, and neighbourhood) sport scores higher than overall management (6.5 out of 10 versus 5.9). table 4. average assessment of municipal management by employment status (2017). global sport % fail % excellent % answers employee 5.8 6.4 14.3 11.0 91.9 self-employed or businessperson 5.6 6.0 17.6 2.1 94.5 unemployed 6.0 6.7 12.5 13.4 90.0 retirees 6.5 6.9 9.9 16.9 62.2 housework 6.5 6.8 6.9 17.9 67.1 student 5.7 6.9 5.3 11.7 90.6 total 5.9 6.5 12.6 12.2 82.6 source: compiled by authors using data from the baròmetre. bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 148 however, these marks, which answer direct questions, hide the importance of each service in the global assessment. to quantify this importance, weights have been obtained for the assessment of each service. they were obtained as coefficients in a regression model where the dependent variable is the global score and the explanatory variables are the scores for each of the eight services considered, following the methodology used in costa et al. (2014), where the sum of the coefficients is restricted to 1 and a constant term is not included. the most valued service, libraries, with 7.4 points, have no influence on the overall assessment since their weight is 0 (not significant). likewise, two of the services with the highest scores (municipal markets and garbage collection) also have very little weight in the overall valuation. on the contrary, the municipal police, street cleaning and the sports offer are the services that count for most in the global assessment of municipal management. with weights above 20%, between the three they account for almost 64% of the global score (table 5). table 5. weight of the different services in the assessment of municipal management. assessment weight (%) mark libraries 0.0 7.4 street cleaning 21.4 5.4 garbage collection 5.2 6.5 maintenance of parks and gardens 11.7 6.3 municipal police 21.9 6.4 cultural and leisure offer 13.9 6.5 municipal markets 5.6 6.9 sport offer 20.4 6.5 total 100.0 5.9 source: compiled by authors. 4. satisfaction survey among members of municipal sport facilities l'hospitalet city council’s servei d'esports (sports service) carries out an annual satisfaction survey among the members of the municipal sport facilities to gather their opinion on certain relevant aspects, such as cleaning and membership price, among others. this section analyses the data corresponding to 2017, looking first at the whole city, and then the data per sports centre disaggregated by sex and age group. it should be mentioned that table 6 does not include the information corresponding to two facilities (the municipal sports centres les planes and santa eulàlia) due to the low number of responses. but they have been included in the calculations in tables 7 and 8, meaning the totals differ by 0.1. table 6. assessment of sports facilities (2017). general report. cleaning attention activities monitoring price global distrib. % l'h nord 6.4 7.1 6.8 7.7 5.4 6.3 7.1 swimming pools 6.8 7.0 6.8 7.0 5.8 6.8 9.6 fum d'estampa 6.1 7.6 7.0 8.0 6.0 6.2 9.6 gornal 7.6 7.9 7.6 7.8 6.3 7.6 16.9 sanfeliu 6.8 7.1 6.8 7.1 5.5 6.6 15.3 bellvitge 7.3 8.5 8.5 9.2 7.4 7.9 21.4 centre 7.5 8.7 8.1 8.5 6.1 7.8 20.2 total 7.1 7.9 7.6 8.1 6.2 7.3 100.0 source: compiled by authors using data from the satisfaction survey. bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 149 for the facilities analysed (table 6), the score given by members was 7.3, with the highest those of the municipal sports facilities bellvitge (7.9), centre (7.8) and gornal (7.6). on the other hand, the lowest marks were given to the municipal sports facility fum d'estampa (6.2) and to the municipal sports complex l'hospitalet nord (6.3). by subject, the most valued are monitoring (8.1) and attention (7.9), with bellvitge standing out in monitoring, with the only excellent (9.2) in the whole table. it should be noted that none of the centres received a fail and that the only marks below 6 were price at l'hospitalet nord (5.4) sports complex, the sanfeliu sport facility (5.5) and the municipal pools (5.8). table 7 shows the sports facility scores by sex, where it can be seen that this variable does not affect the global assessment. attention, activities and price do not depend on the sex of the member, while women score cleaning a little lower and monitoring a little higher. table 7. assessment of sports facilities by gender (2017). general report. cleaning attention activities monitoring price global distrib. % female 6.9 7.9 7.6 8.1 6.2 7.2 52.3 male 7.2 7.9 7.6 7.9 6.2 7.2 47.7 total 7.1 7.9 7.6 8.0 6.2 7.2 100.0 source: compiled by authors using data from the satisfaction survey. by age (table 8), the youngest (under 18 years) value the facilities highest, with an overall score of 8.1, although they only represent 4.2% of the total number of respondents. the most critical are those aged 30 to 44 with lower scores than the total in all the categories. table 8. assessment of sports facilities by age group (2017). general report. cleaning attention activities monitoring price global distrib. % under 18 7.9 8.2 8.2 8.2 6.7 8.1 4.2 18 to 29 7.4 7.8 7.7 8.0 5.8 7.3 17.3 30 to 44 6.9 7.6 7.2 7.7 5.9 7.0 30.2 45 to 59 6.9 7.8 7.4 8.1 6.1 7.1 26.5 60 and over 7.1 8.4 8.0 8.4 7.0 7.5 21.9 total 7.1 7.9 7.6 8.0 6.2 7.2 100.0 source: compiled by authors using data from the satisfaction survey. as in the previous case, and using the same methodology, weights in the overall assessment have been estimated for each of the different aspects of the survey (table 9). one of the most valued aspects, monitoring, has a very low weight in the overall assessment. on the other hand, the aspect with most weight in the global score is activities, followed by attention, cleaning and, at some distance behind, membership price. table 9. weight of the different aspects in the assessment of the sport facilities. assessment weight (%) score cleaning 23.3 7.1 attention 25.7 7.9 activities 32.1 7.6 monitoring 0.1 8.1 membership price 18.8 6.2 total 100.0 7.3 source: compiled by authors. bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 150 in the analysis by facility, it is worth highlighting the importance of the price of the sports facilities fum d'estampa and l’hospitalet nord, with 30% of the total, the low weight of activities at the sanfeliu sports facility, where the most highly rated subject was the attention, as it was at the municipal pools, and the importance of cleaning at the bellvitge and fum d'estampa sports facilities (table 10). table 10. weight of the different aspects in the assessment of the sports facilities (2017). general report. cleaning attention activities monitoring price global l'h nord 12,7 17,5 39,0 0,0 30,8 100,0 swimming pools 15,8 36,7 32,0 0,9 14,6 100,0 fum d'estampa 25,2 3,5 38,6 0,0 32,7 100,0 gornal 17,1 28,1 35,1 8,6 11,1 100,0 sanfeliu 22,6 38,7 18,1 0,0 20,6 100,0 bellvitge 30,5 14,1 25,7 11,7 18,0 100,0 centre 21,0 20,8 22,2 18,6 17,3 100,0 total 23,3 25,7 32,1 0,1 18,8 100,0 source: compiled by the authors. the following tables present the information for the different sports facilities in the city by sex and by age group. table 11 corresponds to the l'hospitalet nord sports complex and shows that, as in the general case, women give higher grades than men to all the aspects considered, and rate the monitoring and attention highest and price the lowest. regarding age groups, young people (up to 29 years old) gave the best marks in all categories, while the cohort that was most critical of this facility was 30 to 44 year olds, who even gave a fail for price (4.8). table 11. assessment of sports facilities by gender and age group (2017). l’hospitalet nord sports complex. cleaning attention activities monitoring price global distrib. % gender female 6.5 6.9 7.3 8.0 5.7 6.4 38.6 male 6.4 7.2 6.5 7.5 5.2 6.2 61.4 age group under 18 8.1 9.1 7.9 9.7 7.1 8.2 6.2 18 to 29 7.3 7.2 7.4 7.9 5.1 6.7 23.5 30 to 44 5.5 6.6 6.3 7.1 4.8 5.6 29.7 45 to 59 6.2 7.0 6.4 7.6 5.5 6.0 34.5 60 and over 7.3 7.4 7.6 8.3 7.6 7.6 6.2 total 6.4 7.1 6.8 7.7 5.4 6.3 100.0 source: compiled by authors using data from the satisfaction survey. at the municipal pools, the marks for all the aspects considered are around 7, except for price, which is below 6. women, young people aged between 18 and 29, and adults between 30 and 44 years old give the lowest rating to the price of this facility (table 12). bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 151 table 12. assessment of sports facilities by gender and age group (2017). municipal swimming pools. cleaning attention activities monitoring price global distrib. % gender female 6.9 7.1 6.7 7.3 5.4 6.7 50.8 male 6.7 7.0 7.0 6.8 6.2 6.9 49.2 age group under 18 7.4 7.4 7.3 7.1 6.6 7.8 5.1 18 to 29 7.4 7.4 7.6 7.5 5.4 7.1 24.9 30 to 44 6.2 6.4 5.8 6.3 5.1 6.1 31.5 45 to 59 6.8 7.3 7.4 7.6 6.5 7.1 27.9 60 and over 7.1 7.0 6.6 6.9 6.0 6.9 10.7 total 6.8 7.0 6.8 7.0 5.8 6.8 100.0 source: compiled by authors using data from the satisfaction survey. table 13. assessment of sports facilities by gender and age group (2017). fum d’estampa sports facility. cleaning attention activities monitoring price global distrib. % gender female 5.9 7.5 6.9 8.0 5.9 6.1 70.1 male 6.5 8.0 7.3 7.8 6.2 6.6 30.0 age group under 18 5.7 5.3 5.0 5.7 5.7 5.0 1.5 18 to 29 7.5 7.5 8.2 8.5 6.3 6.7 6.6 30 to 44 5.7 6.8 6.2 7.4 5.1 5.6 26.9 45 to 59 6.3 7.3 6.5 7.9 5.8 6.0 21.8 60 and over 6.0 8.4 7.6 8.3 6.6 6.8 43.2 total 6.1 7.6 7.0 8.0 6.0 6.2 100.0 source: compiled by authors using data from the satisfaction survey. regarding the fum d'estampa municipal sports facility (table 13), women, who account for more than 70% of the responses, give worse marks than men in all categories, except for monitoring, and the difference of 0.5 stands out in the global assessment of the installation. by age group, the low marks given by the group of under 18s cannot be considered relevant, since they represent only 1.5% of the total answers. the group from 30 to 44 years old gives the lowest marks in all the characteristics considered, with ratings below the sports average. price, at only 5.1, particularly stands out. table 14. assessment of sports facilities by gender and age group (2017). gornal sports facility. cleaning attention activities monitoring price global distrib. % gender female 7.6 7.8 7.6 7.9 6.4 7.8 58.2 male 7.7 7.9 7.6 7.7 6.1 7.5 41.8 age group under 18 6.3 6.8 9.5 6.3 5.5 8.8 1.2 18 to 29 7.9 8.1 7.9 7.6 5.7 7.9 16.4 30 to 44 7.5 7.6 7.2 7.6 6.0 7.3 40.6 45 to 59 7.4 8.0 7.7 8.0 6.4 7.7 27.1 60 and over 8.2 8.3 8.0 8.4 7.6 8.2 14.7 total 7.6 7.9 7.6 7.8 6.3 7.7 100.0 source: compiled by authors using data from the satisfaction survey. bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 152 the gornal sports facility is characterised by receiving marks exceeding 7.5 in all aspects, except price (6.3). the overall assessment is almost 8, thanks mainly to the mark given by women, young people between 18 and 29 years old, and those aged 60 or older (table 14). table 15. assessment of sports facilities by gender and age group (2017). sanfeliu sports facility. cleaning attention activities monitoring price global distrib. % gender female 6.4 7.1 6.5 7.1 5.5 6.5 47.6 male 7.2 7.1 7.0 7.0 5.5 6.7 52.4 age group under 18 8.0 8.1 8.0 8.0 6.7 8.0 10.2 18 to 29 6.8 7.2 6.8 7.1 5.0 6.6 25.2 30 to 44 7.0 7.1 6.9 7.0 5.8 6.7 26.5 45 to 59 6.2 6.4 6.0 6.5 4.9 5.9 24.9 60 and over 6.8 7.2 7.0 7.3 5.9 6.5 13.1 total 6.8 7.1 6.8 7.1 5.5 6.6 100.0 source: compiled by authors using data from the satisfaction survey. table 15 corresponds to the sanfeliu sport facility, which receives average marks of between 5.5 (price) and 7.1 (attention and monitoring). by gender, men assess the facility a little better than women (0.2 more), who give a very low grade to cleaning (0.8 less than men). by age group, those under the age of 18 most value the sports facility, much more than the other age groups, with the most critical being those between 45 and 59 years old. table 16. assessment of sports facilities by gender and age group (2017). bellvitge sports facility. cleaning attention activities monitoring price global distrib. % gender female 7.0 8.5 8.7 9.3 7.3 7.8 48.6 male 7.6 8.5 8.3 9.1 7.6 8.0 51.4 age group under 18 8.6 9.4 9.1 9.3 8.5 8.7 3.4 18 to 29 7.8 8.8 8.5 9.2 7.6 8.1 18.5 30 to 44 7.4 8.3 8.4 9.2 7.4 8.1 30.1 45 to 59 7.1 8.4 8.4 9.2 6.9 7.6 28.5 60 and over 6.7 8.3 8.7 9.1 7.9 7.4 19.4 total 7.3 8.5 8.5 9.2 7.4 7.9 100.0 source: compiled by authors using data from the satisfaction survey. as mentioned earlier, the bellvitge sports facility is the most highly rated, with a global average score of 7.9 (table 16). especially noteworthy is the monitoring, which receives marks of over 9 in all age groups and for both men and women. attention and activity scores are generally above 8 and the worst rated categories are price and cleaning, this being the only installation where price is not the worst rated aspect. the centre sports facility, the second most highly rated, scores above 8 in attention, monitoring and activities, 7.5 in the case of cleaning and 6.1 for price. women rate the facility a little more highly than men and the age group that gives lower than average scores across all aspects is 30 to 44 years old (table 17). bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 153 table 17. assessment of sports facilities by gender and age group (2017). centre sports facility. cleaning attention activities monitoring price global distrib. % gender female 7.6 8.8 8.2 8.5 6.3 8.0 51.1 male 7.5 8.7 8.1 8.5 5.9 7.6 48.9 age group under 18 8.3 8.0 8.5 8.4 5.5 8.6 2.9 18 to 29 7.7 8.5 8.3 8.6 5.4 7.8 9.6 30 to 44 7.3 8.5 7.9 8.2 5.6 7.4 25.3 45 to 59 7.4 8.8 8.0 8.5 5.8 7.6 23.4 60 and over 7.7 8.9 8.3 8.7 6.9 8.2 38.8 total 7.5 8.7 8.1 8.5 6.1 7.8 100.0 source: compiled by authors using data from the satisfaction survey. 5. conclusions sport has become a very important part of municipal activity, as evidenced by the fact that one of the strategic objectives of the municipal action program of the city of l'hospitalet de llobregat is to “put sport at the service of personal and community growth”. in recent years, sport has become an instrument of social cohesion and integration, a significant fact in a city such as l'hospitalet where the percentage of the population born abroad is relatively significant. citizens of l'hospitalet value the municipal management of the sports offer (facilities), giving it a very good mark (6.5 out of 10), higher than the overall city hall management average (5.9). in addition, the sports offer is one of the services that scores best in the overall assessment of municipal management, along with the municipal police and street cleaning. although the rating can be considered to be good, there is still some room for improvement, especially in neighbourhoods such as sanfeliu, pubilla cases and torrassa, where the score is relatively lower. on the other hand, the overall assessment of sports facilities by members is good (7.3 out of 10), with price receiving the worst mark (6.2), while monitoring and attention are the most highly valued, and none of the aspects receives a fail. thus, campaigns and actions that tend to lower membership prices or raise awareness among members that what they receive is adequate for the price they pay (they receive more than they perceive), could help improve this rating. for facilities, the centres that obtain the highest marks are bellvitge, centre and gornal, while there is more room for improvement at fum d'estampa and the hospitalet nord ce. detailed analysis of satisfaction surveys of the population can help municipal management of sport facilities. in the example of l'hospitalet, despite being the lowest rated aspect, membership price is not as decisive in the global score as the activities offered, attention or cleaning. policies that focus only on prices are therefore unlikely to be as effective at improving members’ perception of sports facilities as those that also take other aspects into account. bosch et al. / european journal of government and economics 9(2), july 2020, 143-154 154 references ajuntament de l’hospitalet (2013). estudi de la dimensió económica de l’esport a l’hospitalet de llobregat. ed. ajuntament de l’hospitalet de llobregat. bosch, j., garcía, j., moya, x., and murillo, c. (2018). estratègia de futur de la politica esportiva de l'ajuntament de l'hospitalet de llobregat. mimeo, servei d'esports, ajuntament de l'hospitalet. costa, a., garcía, j., and raymond, j.l. (2014). “are all quality dimensions of equal importance when measuring the perceived quality of official statistics? evidence from spain”. journal of official statistics, 30, 547–562. https://doi.org/10.2478/jos-2014-0034 cunningham, l.f., young, c.e. and lee, m. (1997), “a customer-based taxonomy of services: implications for service marketers”, in d.e. bowen, d. iacobucci y t. swartz (eds.), advances in services marketing and management, vol. 6 (1ª ed)., 189–202. jai press inc. https://doi.org/10.1016/s1067-5671(97)06023-x fábregas, f. (coord.), bordas, m. (ed.), lópez-jurado, c., giralt, c. and martí, p. (2005), pla director d’instal·lacions i equipaments esportius de catalunya. consell català de l’esport. generalitat de catalunya. prat, m., soler, s., vallès, c., and orrit, x. (2008), política i esport, observatori crític de l’esport ed. https://doi.org/10.2478/jos-2014-0034 https://doi.org/10.1016/s1067-5671(97)06023-x ejge_special_issue_9_2 contents9_2 number 9, issue 2, june 2020 special issue. the role of institutions and governance in sport 5949_g european journal of government and economics 12(1), june 2023, 58-78 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 territorial diagnostics and citizen perception of municipal public management, towards an efficient reform of local administration manuel octavio del campo villares a *, francisco jesús ferreiro seoane b, eladio jardón ferreiro c, miguel molina picazo d a department of economics, univeridade da coruña, spain b universidad de santiago de compostela, spain c instituto internacional de marketing y comunicación, iimc, spain d instituto mediterráneo de protocolo. universidad miguel hernández, spain * corresponding author at: moctadcv@udc.es abstract. spain has a population distribution dispersed in small population centres where 60% of municipalities have less than 1,000 inhabitants. this situation generates a financial and functional incapacity to guarantee the successful and efficient provision of local public services. in order to fulfil its competences and improve services, a territorial diagnosis of the investigated area is necessary to adapt the size and structure of its administration. however, the social superstructure is reluctant to change the form of administrative action, despite being perceived as slow and bureaucratic by citizens. for this reason, this work proposes a reform to improve local economic and financial management by redesigning its administrative structure, thus achieving citizen recognition of the advantages of change. the proposal is developed through a computerised reporting model for public decision-making, the result of which is a systematic spatial report of administrative-financial decisions, which optimises decision-making and makes public management more visible. keywords. local development, municipality, public provision. jel codes. h11, h41, h73 doi. https://doi.org/10.17979/ejge.2023.12.1.9345 1. introduction although the main task of any public administration is to "administer", this cannot be understood as an end but rather as a service in the common interest. to this end, all administrations, regardless of their size, must concentrate their efforts on seeking the best possible organisation, which will enable them to provide their own services as efficiently and as fairly as possible for the benefit of their citizens. for this reason, the position put forward is the defence of the "adaptation and organisational modernisation of the administration". in this sense, the organisational structure -administrative form"must be" the one that best suits the characterisation of the territory under analysis https://creativecommons.org/licenses/by-nc/4.0/ mailto:moctadcv@udc.es https://doi.org/10.17979/ejge.2023.12.1.9345 del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 59 (rodríguez, 2006; bel, 2011). to this end, it is necessary for the parties involved (institutions, companies, citizens) to know the socio-economic state of their territory through the diagnostic of their present situation and future expectations. therefore, we speak of governance (administration) as a concept linked to the idea of innovating in order to improve. any administrative change must involve the introduction of improvements in the provision of public services, in the regulation of spaces for citizen participation and in territorial development. for this reason, innovation should be seen as synonymous with learning, adaptation and the involvement of all the agents that form part of the administration to achieve the most appropriate administrative organization for each spatial and demographic reality. so, the study is based on two questions: if the structure of the local administration in spain is the most convenient to fulfil its functions, and how does the citizen interpret the management of this administrative level closest to the citizen. these questions introduce the objective of the study. as a result, the main objective of this research is to propose a reform of the economic-financial management of local administrations, a change that will increase the efficiency of local authorities' resource management and improve the provision of public services through a redesign of their administrative structure. also, it offers a proposal based on a systematic spatial reporting of administrative-financial decisions that demonstrates the possibility of achieving a greater adaptation of municipal size to the socio-economic reality, as well as the optimisation of local decision-making and the obtention of a more effective and efficient public function model. for this reason, the contribution of the work is located in the elaboration of an automatic and objective comprehensive scorecard (cmi) that allows us to know the efficiency in the management of municipal public services. regarding the structure of the study, it begins by exposing the regulatory framework that regulates municipal economic-financial management in spain. next, the functions to be carried out by each municipality and the current municipal structure in spain are identified. the applied part of the work begins with the study of the perception of the citizen in relation to the management of municipal services. the study concludes with the presentation of the proposed automatic territorial diagnosis system in relation to the provision of local public services and the result that is expected from it, and that is materialized in a technical-administrative-financial report where the effectiveness of public management municipal is evaluated. 2. theoretical framework in this section, a specific analysis of the situation of local administrations in spain is presented. through a study of the theoretical context at a regulatory and demographic level, at the same time as a sociological analysis of the opinions of the citizens of these localities is developed. and del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 60 therefore, on the economic, administrative and efficiency impact of this public organisation and management. 2.1. regulatory framework and context. the general socioeconomic situation, the demographic dynamics as well as the current sociopolitical picture make the complexity of guaranteeing an optimal provision of local public services (lps), in time and form together with the economic-financial sustainability of the local administration (la) one of the most important challenges of the public administration in spain. this is compounded by the crisis situation in which we have found ourselves, which began in 2008, prolonged by the complexity and instability of international relations in general, the crisis resulting from covid-19 and the redistribution of economic forces on the world map. all this comes together, giving rise to a growing interest and concern for the profitability obtained through the use given to public resources and, in particular, the concern to preserve the welfare state achieved (welfare, education and health). the immediate effect of the described reality is the intensification of the search for a more efficient, accessible and financially sustainable public management as a political urgency (bel, 2011; elizalde, 2020). for the first time, we ask ourselves what the "gain" obtained from the public resources employed is. to address this dual problem: "a growing financial need and greater efficiency in the use of public resources", the reform of article 135 of the spanish constitution in 2011, endorsed by organic act 2/2012 on budgetary stability and budget sustainability, established the need to ensure "budgetary stability" as a principle that will govern the work of all administrations. the application of this constitutional precept to the local sphere requires a major regulatory adaptation to guarantee the application of the principles of budgetary stability, financial sustainability and efficiency in its management, given its heterogeneity in terms of size, population and socio-political dimension. therefore, it is necessary to adapt the basic aspects of la's functioning and organisation to space and time. the administrative adaptation of which required legal sanction through act 27/2013, on the rationalisation and sustainability of local administration, whose objectives include: 1. establish municipal powers, avoiding duplication with respect to those exercised by other administrations in the same area. 2. rationalise the local organisational structure in order to comply with the principles of efficiency, stability and financial sustainability. 3. guarantee rigorous and effective financial and budgetary control. 4. promote and favour private economic initiatives by reducing the administrative bureaucracy that presides over the management of this administrative level. the second of these objectives, set out in the preamble of act 27/2013, is the subject of this paper. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 61 2.2. status and situation. spanish municipal structure dates to a royal decree of 1833, which established the economic and political competences of the local entities together with the design of the local administration at two basic levels: province and municipality. since that date, changes in the municipal map have been extremely limited, especially when compared to those in the dynamics of settlement, distribution, management and economic production, as well as the effects of these on any territory. thus, the institutional architecture shows a growing mismatch between municipal structure and geoeconomic evolution (rodríguez, 2011; del campo and ferreiro, 2013). therefore, it is difficult to understand how the public institutions responsible for defending the common interest can remain unchanged in the face of a rapidly changing socio-economic scenario. even more, bearing in mind that the origin of the municipal structure that arose as a result of the aforementioned royal decree stems from a centralist conception where provinces and municipalities were instruments at the service of central power, their functions being to improve tax collection and control citizens' movements. thus, to the question of whether these are the functions that the la should perform in the future, the answer should be negative. simply because of the mere passage of time and the technical progress inherent to it, municipal functions have grown disproportionately from the mid19th century to the present day (rodríguez, 2011). consequently, it is obligatory to refer to what the municipal functions are: a. representative body of the democratic will and political conception for the citizen → administration that serves as the basis for the organisation of political power and generates an identification of the citizen with that territory. b. organisation that provides and offers "public" services to the citizen → a form of "company" that offers and provides services of general interest and cannot be replaced by the market. c. promoter of socioeconomic development → acting as a sociocultural, productive or touristic dynamiser and serving general interest beyond the municipal borders. from these functions, it can be deduced that a basic function of the municipality and, by extension, its governing body, the city council, is to act as a political-administrative nucleus in the lives of its neighbours. in light of this, it is worth asking whether this will be its main role in the future or whether it will be to attend to the needs and demands of citizens in a reality that is not only social but also spatial, and which is changing with unusual speed. the answer to this question requires identifying the most convenient municipal structure at a spatial, functional and political level (de la mora, 1997; del campo y ferreiro, 2013). del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 62 moreover, identifying a spatial demarcation is not only a question of geographical dimension. in fact, it is possible to identify up to three dimensions linked to the concept of territory and which must converge in the same administrative space: a. institutional: establishing the political relations of individuals and defining the capacity to decide their public representatives. b. functional: coverage in the provision of different collective services, acquiring a dimension beyond the traditional municipal framework. c. relational: sphere in which to establish social and economic relations between individuals: labour and housing. the coincidence of these three dimensions in the same physical space is a factor that generates economies of scale in local public provision (rodríguez, 2006). having reached this point, it is worth asking what the current structure of the spanish local administration is. information is presented in tables 1 and 2, where a comparative analysis of the municipal size by population and autonomous community is made, respectively, in the decade between 2021 and 2012. with regard to the population distribution of the municipalities, it is worth noting that more than 60% of them have less than 1,000 inhabitants. this reflects the fact that the spanish administrative map is a large area populated by a large number of small population centres with little capacity in terms of effective demand and socio-productive structure. a reality where municipalities with more than 50,000 inhabitants represent less than 2% of the total, and where, separating the municipalities into those with more and those with less than 5,000 inhabitants, the former have seen their weight in the total reduced over the decade analysed, with no significant process of concentration being detected. table 1. municipal distribution of spain by population group, 2012-2021. segment-population 2012 2021 nº % % accumul. nº % % accumul. <101 1.193 14,70 14,70 1.137 14,01 14,01 101-500 2.670 32,89 47,59 2.697 33,24 47,25 501-1.000 1.033 12,73 60,32 1.052 12,96 60,21 1.001-2.000 912 11,24 71,56 922 11,36 71,57 2.001-5.000 991 12,21 83,77 1.000 12,32 83,89 5.001-10.000 560 6,90 90,67 553 6,81 90,70 10.001-20.000 355 4,37 95,04 361 4,45 95,15 20.001-50.000 257 3,17 98,21 250 3,08 98,23 50.001-100.000 83 1,02 99,23 82 1,01 99,24 100.001-500.000 57 0,70 99,93 56 0,69 99,93 >500.000 6 0,07 100,00 6 0,07 100,00 note. own elaboration based on ine data. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 63 accordingly, the size of the territory and its population are related. in terms of population density, three groups of communities can be distinguished, which are shown in figure 1: 1. underpopulated: this makes the provision of spl very difficult and limited (lack of market and dispersion). these communities have an average population density of 40 inhabitants/km2, with more than 90% of their municipalities having <10,000 inhabitants and suffering from distressing population and economic desertification. these are: aragon, castile and leon, castile-la mancha, navarre, la rioja and extremadura. 2. intermediate group: with an average density of 105 inhab/km2, it maintains a high-risk status characterised by ageing and dispersed settlement. its threat is the lack of productive fabric. these are: andalusia, asturias, cantabria and galicia. 3. sufficient average population size: it has a service sector of considerable size and diversity. its average density is 400 inhabitants/km2 and its average population size is capable of sustaining the necessary spl. these are: balearic islands, canary islands, catalonia, valencia, madrid, basque country and murcia. the most revealing fact in table 2 is the large difference between the number of spanish municipalities with ≤10,000 inhabitants (91%) and their population (20%). this shows the structural weakness of the majority of local corporations in spain. in thirteen communities, the percentage of municipalities with ≤10,000 inhabitants exceeds 75%, while in ten, the population of these municipalities does not reach 25%. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 64 figure 1. population density of the autonomous communities in spain, 2021. table 2. municipal size of spain by autonomous community 2012-2021. autonomous community number mun. population/nº municipalities population density (inhab/km2) % popu. in mun. ≤ 10.000 % mun. < 10.000 popu. mun. ≤10.000 / nº mun. ≤ 10.000 2021 2012 2021 2021 2021 2021 2012 2021 andalusia 785 10.960 10.793 108 19% 81% 2.738 2.592 aragon 731 1.846 1.814 28 30% 98% 578 547 asturias 78 13.812 12.972 105 15% 76% 2.514 2.503 balearic i. 67 16.708 17.508 292 15% 64% 3.973 4.041 canary i. 88 24.072 24.693 428 10% 52% 4.805 4.747 cantabria 102 5.822 5.730 115 35% 90% 2.171 2.255 cast-león 2248 1.133 1.060 25 43% 99% 506 466 cast-man. 919 2.309 2.230 26 44% 96% 1.070 1.023 catalonia 947 7.995 8.198 259 18% 87% 1.712 1.735 valencia 542 9.464 9.332 247 17% 81% 2.027 1.986 extremad. 388 2.878 2.731 26 50% 97% 1.521 1.418 galicia 313 8.830 8.612 98 29% 82% 3.244 3.066 madrid 179 36.305 37.716 1038 6% 72% 2.861 3.027 murcia 45 32.766 33.744 216 45% 31% 4.052 4.095 navarre 272 2.370 2.432 64 42% 96% 1.121 1.067 basque c. 251 8.837 8.821 330 19% 82% 2.015 2.017 la rioja 174 1.860 1.838 64 33% 97% 700 632 total 8.131 5.824 5.828 99 20% 91% 1.343 1.301 own elaboration based on ine data. notes: mun=municipalities; popu=population del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 65 if the observed reference variable is the average municipal population of municipalities with ≤10,000 inhabitants, six communities did not reach 1,500 inhabitants/municipality in 2021, one more than in 2012. in the decade considered, eleven regions reduced their average population size. the number of autonomous communities that do not reach 10 municipalities >50,000 inhabitants is 15 in 2012 and 16 in 2021. meanwhile, the number of communities that do not reach 25 municipalities with >20,000 inhabitants in the same years is 12 and 14, respectively, with dispersed settlement increasing. in addition, this picture of the municipal structure in spain was reinforced by a series of sociopolitical factors (de la mora, 1997; ramió and salvador, 2012). these aggravated the negative effects caused by the expansion of municipal activity derived from the displacement and relocation of productive factors: a. the growth and modernisation of the country means that any municipality wants to match the competencies of other municipalities in administratively more advanced countries. b. the democratic system requires bringing the "administration vs. administered" closer together; in this line, the municipality goes from being an administration that serves the citizen demand to taking on competencies that are not its own due to its proximity to the citizenry. c. the municipal "interest" in satisfying citizen demand led to an increase in the demands of its citizens, generating a circular process of sustained growth in the demand for new spl. 2.3. study approach: the administrative perception of citizens. after the overview above, a scenario of socio-political debate on the sustainability of the basic structure of the local corporations is opening up. this debate arises between two antagonistic positions on the management of the lps: absolute globalism and maximum localism. this does little to help in identifying the desired municipal model: a system organised under a theoretical and legal nominal autonomy, although with an evident incapacity to execute it for many of the spanish municipalities; or a flexible system, capable of adapting to the needs of socio-economic and geospatial evolution, fully assuming its competences. furthermore, and in view of the magnitude of the tasks that the local administration serves, a question immediately arises: are the municipalities in spain, with their current structure, in a position to carry out these functions, or even to expand them, or on the contrary, are they unable to provide an effective response to their basic competences as entities of political representation, providers and suppliers of collective services? to answer this question, it is necessary to have adequate and sufficient information on the general state of the territory under diagnosis, a comprehensive territorial diagnosis adjusted to the characteristics of the space investigated. the technical aspect and the socio-economic reality del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 66 mean that this knowledge has a more practical than political character and origin (guy, 2011; durán, 2016, del campo et al, 2021) and, where demagogy has no place, these issues are as varied as those listed below (bonnefoy and armijo, 2005; alfaro and gómez, 2016; ine, 2018) and whose common characteristic is their continuous modification: • state, evolution, movements, demographic projection and population. • socio-economic structure, activity and productive settlement. • provision and management of collective services: distribution and accessibility. • basic demands of the population: economic, identity and social. • transfer of competences from the environment. • municipal financial and budgetary situation. therefore, if society and its economy are immersed in a continuous process of change (modernity and progress), the institutions that serve the collective interest should not remain static over time (bosch and solé-ollé, 2012). for instance, two antagonistic processes: inframunicipalism and urban mega-congestion, escape the current conceptualisation of municipality (de la mora, 1997; bel, 2011), exemplifying the mismatch between the current local administrative structure, the socio-economic dynamics and the new models of spatial occupation. however, experience shows how the historical, sociocultural and geopolitical superstructure conditions any initiative to modify either the current administrative boundaries or the way of providing lps, maintaining the primacy of the idea of the classic municipality as the "only" criterion for this purpose, regardless of its limitations. a classic municipality is understood to be one that was designed following the municipal administrative structure included in rd 1833. in fact, it is only in the current socio-economic situation of crisis and financial and social urgency to maintain the welfare state that a certain awareness seems to emerge regarding the profitability and use given to public resources and which administrative level is best able to provide certain services. but what is the public's perception of the administration's performance? this is a key question for assessing the expected aptitude in the face of a transformation of la in size and form. this information will allow a generic evaluation of many of the problems and obstacles of a social nature derived from a change in the way public services are approached and provided by the administration. this information is summarised in table 3, where up to six argumentative constructions can be distinguished regarding the population's perception of its administration. in these, it is worth noting that the general view transmitted by citizens with respect to the actions and function of government is "traditional", that is, it defends a normative conception of public action and is averse to change. the administration is seen as an instrument of guarantee and protection, while the private sector is associated with the idea of profit and inequality. the documentary basis for the results obtained comes from the information collected in the "survey on the quality of public services" and the "survey on public opinion and fiscal policy" conducted by the sociological research centre (cis) in the period 2015-2020, considering 2015 as the base year of study. the analysis was completed with information on periodicity and del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 67 compliance from the ministry of public administrations (map) and indicators of subjective wellbeing, obtained through a survey by the national statistics institute (ine). the appendix shows the list of questions considered to assess citizen perception of government activity in a broad sense, summarised schematically in table 3. table 3. citizen perception of the activity of public administrations, 2015. general perception the citizen gives a negative assessment when asked abstract or generic questions about the administration and its employees (bureauphobe citizen). perception of management areas in contrast, when asked about specific areas of government or public policy, they respond positively (bureaucophile citizen). general public-private perception better general evaluation of private management (agility + efficiency) than public management (bureaucracy + delays), it would be positive to incorporate private management techniques in the administration. perception of social effect publicprivate provision when choosing between public vs. private management for the provision of basic services, the majority choose public management, as it is associated with satisfying the common interest and being a guarantee of equality. institutional perception the la is the administrative level with the least bureaucratic connotation, its actions are more visible and it creates a sense of belonging. perception of interadministrative management results while, in terms of the management of public resources, it is the autonomous communities that are most highly valued, the city council is associated with identity more than with functionality. note. compilation based on data from: cis, map, ine, accumulated results 2015-2020. table 4. evolution of citizen perceptions of administration, 2016-2021. evolution(δvar.2021/2016) general perception constant +1,6% management areas perception increasing +9,6% general public-private perception constant +1,2% social perception public-private provision decreasing -5,6% institutional perception constant -1,0% inter-administration results perception increasing +3,2% own elaboration based on data from: cis, map, ine, accumulated results 2016-2021. moreover, perception is a subjective issue with great variability in terms of space and time, and the period analysed is 6 years. table 4 shows the evolution (trend): increasing, constant or decreasing followed by each of the six arguments described in table 3, and also includes the average variation rate observed in the 2016-2021 period with respect to 2015-2020 regarding the degree of satisfaction (trust) or dissatisfaction (distrust) conveyed by the administration in its relationship with citizens. this variation was obtained from the identification and selection of those questions (annex) with significance in relation to each of the arguments developed in the previous table. the trend observed was classified according to the following criteria: increasing ≥2%; constant <2% and >-2%; and decreasing ≤-2%. according to the data in table 4, it can be affirmed that beyond a general perception of slowness (waiting times) and administrative bureaucracy (documentary requirements), when del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 68 looking deeper into the management of public services, there is a prevailing defence of public intervention without changing its basic form of action. more management and less bureaucracy in the public sector are something desired by all, but its implementation generates doubts and mixed feelings. citizens show a general interest in an agile and efficient public activity. however, change in public management is viewed with suspicion (concern). only as social interaction grows and the comparison with other territories is visualised, the subject realises that certain changes in the provision of basic (collective) services are those that facilitate an improvement in their living conditions, whether these are of an economic nature or in the form of greater satisfaction by perceiving a higher quality in the provision of those services (lópez doblas, 2018; elizalde, 2020). for that reason, only through persuasion and practical conviction of the collective and individual benefits (making these visible) that a reform or adaptation of the municipal size to the socio-economic reality entails is it possible to achieve social and political fruition of the need for this administrative change, without this entailing a social or identity crisis in the face of a municipal merger, creation of commonwealths, consortiums, metropolitan areas or cooperation agreements (guy, 2011; durán, 2016). in this respect, political pedagogy plays a transcendental role, conveniently showing what is gained and lost with a change in the management and provision of any basic service. the complexity and dimension of the problems faced by all administrations, especially la, due to its heterogeneity, means that they cannot be solved exclusively through regulatory changes (ramió and salvador, 2012; del campo, 2014); a change of mentality is needed in search of formulas to better provide lps. the search for an effective and efficient public service model must be seen as a necessity, and for this it is necessary to start from a correct knowledge of the analysed space in all its areas (guy, 2011). communication based on knowledge and management of the treated environment constitutes the support for this change of mentality. the search for the best administrative structure for each territory should be an objective in order to provide the best possible provision of those services that are essential in the daily work of any person. except for certain general services of exclusively national scope, such as defence, justice, infrastructures or foreign relations, no administration is better than any other for exercising a given competence in terms of basic territorial services. it can only be considered as such if it is the one that best adapts to the specificity of the environment, its surroundings and the socioeconomic circumstances of the space and time analysed. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 69 table 5. outsourcing or privatisation vs. innovation or empowerment in public management. under the impression that it is impossible for the administration to have the capacity to maintain quality services in a context of spending restrictions, it is decided to defer the management of certain services to the private sector, while maintaining their public nature. outsourcing ≠ privatisation • outsource, the responsibility remains public, the administration directs and controls and the private agency manages. • privatise, the public responsibility for the service ceases to be public. is outsourcing a negative option? • it is not always synonymous with dismantling the public sector or reducing the public services provided. • it may be the only possible strategy to maintain the welfare state in a context of diminishing public resources. what services are outsourced? it is not true that only welfare state services are outsourced: care, health and education; they are also outsourced: economic, cultural and administrative services. necessary condition for the private operator to provide quality public services: that the administrations exercise their planning, control and evaluation functions. private organisations are not always more efficient than public organisations in the provision of public services, because of: • suffering from bad management. • the need for immediate profit. considering the above, might it not be thought that the most correct option at times would be to improve the functioning of public organisations from within, without resorting to outsourcing or privatisation? focusing on the updating and training of their staff in a general sense: values and functions. is it possible to see outsourcing as a sign of the failure of public organisations to provide quality public services? note. own elaboration. going deeper into the initial idea of innovation and public modernisation, table 5 shows a flexible and innovative characterisation of two ways of acting in the search for optimal administration in the provision of lps adapted to the characteristics of the territory, and whose distinction is necessary to know. on the one hand, we have the traditional formulation represented by the concepts of outsourcing and/or privatisation (working with or as the private sector outward vision); and on the other hand, administrative adaptation and modernisation (working from within to improve the public sector), where innovation and training are the references for the development of a better public function. precisely, the aim of this work is to contribute from a technical approach to disseminate and extend the effective possibility that an internal adaptation of the municipal size to the changes due to the mobility of goods and factors is possible. this would be a way of contributing to the improvement in the provision of lps (optimising citizen welfare), increasing efficiency in the management of local resources, without forgetting the cost and equity in the access and location of such services (galera, 2019). the issue of the unsustainable dimension (economic, financial, human, technical...) of local corporations has been recurrent for several decades. (bosch y solé-ollé, 2012; durán, 2016). del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 70 these deficits prevent them from fulfilling their main objectives: the provision of basic services and political representation (bosch and solé-ollé, 2012; del campo, 2014; galera, 2019). on the basis of the established objective, the work aims to serve as a guide to embark on a path aimed at optimising the local administrative structure, stressing the need to provide the necessary technical information in all the transcendental areas: demography, socio-economic structure, location, size and provision of collective services, as well as that derived from their financial-budgetary situation. in this way, to help those who are responsible for making decisions to opt for the most convenient solution in order to increase the well-being of the citizen obtained from the provision of lps. below, an automatic instrument for diagnosis and proposal for improvement in the economicfinancial management of local services is presented, based on the adaptation and redesign of the established municipal administrative structure. the starting point of the developed design is the selection of the basic and transcendent indicators to evaluate the municipal capacity in the provision of those public services that are proper (compulsory) to its administrative size. 3. diagnostic proposal to rationalise the provision of local public services through the management (ge) and knowledge (co) of the environment (me). the proposal is based on an application that will facilitate decision-making by those government structures responsible for configuring the optimal administrative size in the provision of lps (blasco et al, 2013; del campo et al, 2021). being aware that municipalities do not have a common pattern of behaviour or the same characteristics in basic aspects: size, socio-economic capacity or demographics. therefore, the objective is to offer a tool that evaluates and organises the necessary qualitative and quantitative information to facilitate decision-making on the ideal municipal size and to explain this decision. the instrument is configured as an automatic and periodic reporting model of the demographic, socio-economic, provisional and financial situation of the municipalities under study, which will allow us to detect not only financial problems but also social and sustainability problems that condition the survival and viability of these municipalities, as well as facilitating their comparability. thus, the resulting output is a management system based on the provision of relevant information on a computer tool, the ultimate result of which is a diagnostic report and a proposal for economic-financial improvement and sustainability in the provision of lps. this is based on the rationalisation of its competences and structure, based on the information obtained from the management and knowledge of the territory analysed. the informatics support used is represented graphically in figure 2, which takes as a documentary basis all the relevant technical and situational information of the municipality(ies) under study in the areas of geo-demography, labour structure, socioeconomic level, public provision and financial-budgetary situation, for the three complete years prior to the analysis. the del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 71 system is fed back each year (blasco et al, 2013; del campo et al, 2021), eliminating the oldest year and incorporating the most recent (phase 1 of action). once the information referred to in the previous paragraph has been collected and entered into the system, the system offers three types of output variables (phase 2 of action): a. individual indicators of provision or service, where the direction followed by the indicator and the intensity of the variation observed are taken into account. they are obtained through official statistics and ministerial websites, being selected based on the characteristics of the investigated territory. b. combined indicators of provision or service, made up of the integration of several indicators, or else put in reference to a dependent variable. indicators are valid for all municipalities with a minimum population size capable of generating the requested information (more than 1,000 inhabitants). c. relationships of sustainability in the provision of lps, based on the situation with its demand/supply nexus. the alerts to be considered will be obtained from these relationships. finally, the application offers qualitative results: a list of those policies that should be implemented, the most appropriate administrative categorisation for the investigated area, and what would be the recommended implementation proposal (phase 3 of action), collected in a "systematic administrative-financial report". the system is expressed schematically, by means of the following list of actions: initial data (level_1) → basic indicators and data ordering (level_2) → evolution and analysis of variation of basic indicators (level_3) → identification of alerts detecting problems (level_4) → recommended measures/policies (level_5) del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 72 figure 2. proposed model to rationalise the provision of spl. own elaboration. the information in figure 2 shows independently the three levels of action in which the information management and processing model is materialised. three interrelated levels, where the success of each one is a function of the previous one: level 1: design, analysis and selection of indicators. level 2: measurement and evaluation of indicators, it development. level 3: diagnosis, strategic proposal and organisational decision. the necessary leadership to carry out any administrative reform, guaranteeing its sustainability, as well as the undeniable combination of the double political objective of "efficiency vs. equity" requires making decisions based on relevant and responsible information in the double sense of quality and quantity. therefore, it is about political leadership and management based not only on the capacity and ability of the politician but also on knowledge and, above all, on the most convenient information in relation to the service to be provided. politicians must know the reality of the territory they govern and manage, its needs and the consequences of their activity, which is why the dissemination action based on the management (ge) and knowledge (co) of the environment (me) is essential if the aim is to modify the way in which the lps are provided. this requires a clear understanding of the relevant parameters when del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 73 making such a decision, only then will it be possible to explain to the different interest groups (citizens, social groups, etc.) the reasons and criteria that motivated them. 4. systematic administrative-financial reporting of municipal finances. a way of transmitting a public reform. according to the previous point, this report is the result of the application developed, which is configured as an automatic reporting of periodical information, supporting decision-making in the provision of spl and self-supported by the administration analysed. the report has been created on the basis of a computerised support and aims to establish an automated decision-making model based on the socio-economic and financial capacity of the analysed environment. the designed application uses excel as computer support, which through the visual basic language for applications (vba) allows us to generate custom forms with text boxes and drop-down lists. its objective is to facilitate the detection of financial problems, the control of public income and expenditure, the perspective of vital sustainability and the management of municipal services, and its comparability. its characteristics are easiness of obtaining and simplicity; completeness and relevance; transparency and agility; and updating and participation. as a result, the report allows a more systematic, comprehensible and transparent presentation, greater convenience in its layout, and its regular review will contribute to its continuous improvement. in addition, the integral vision of local problems facilitates the commitment of all the agents involved in its management, their motivation and orientation towards an efficient use of the resources available to the municipal finance. for this purpose, table 6 shows by "case" the diagnosis, results and actions to be implemented, of the "systematic municipal administrative-financial report", or output of the model. the study covers four areas of action: demography and population; socio-labour structure; living conditions and provision of spl; and financial-budgetary capacity. for each area, a partial situation proposal is obtained and the actions to be implemented are listed, distinguishing between (alfaro and gómez, 2016; del campo et al, 2021): • dac actions (direct administrative control): local capacity to carry them out. • iac actions (indirect administrative control): outside local competence. in addition, the report presents a partial proposal for each area, providing guidance on its immediate sustainability and the most desirable direction to take. the merger of the partial proposals converges in an overall proposal aimed at showing the way forward if the level of spl provision is to be improved in the medium and long term. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 74 table 6. proposal for action and improvement of local administration (hypothesis). population control and regeneration capacity result reduced survival capacity, unsustainable structure dac providing incentives to encourage births dac subsidise public transport during the busiest times for business and education. dac promote the development of the traditional local community business. proposal set out measures to encourage a merger by integration state and evolution of the labour market result small domestic market and low labour dynamism dac granting of advantageous land for new economic activities dac create an official and standardized contracting office dac reduction of the ivtm (tax on motor vehicles) for new economic activities, bonuses for creating and maintaining employment. dac promote the development of the traditional business of the area. proposal examine the viability of administrative size: merger by integration or selective merger living conditions and infrastructures report result risk of provision and equipment due to reduced density dac grant building licences adjusted to the available environment and equipment. iac aggregation of population centres, elimination of unsustainable ones due to lack of population dac develop cooperation policies with other local corporations to manage basic facilities and services. iac redistribution of students in schools according to level and installed capacity proposal examine the viability of administrative size: merger by integration or selective merger management control and per-capita co-responsibility result efficient and inequitable (inequitativa) municipal management dac increase the economic performance derived from own heritage iac renegotiate the interest rate of the debt according to market conditions dac reduce non-basic current transfers granted dac reduction of urban property tax (ibi) according to the size of the family unit. proposal reducing direct taxation vs. other own revenues general proposal: adopt measures for a municipal integration in the short and medium term. note. own elaboration. the technical maintenance and updating of the information that feeds the proposed application recommend that it be located at the administrative level with the necessary capacity and personnel (province, municipalities with more than 10,000 inhabitants, ...). 5. conclusions and the need for change. this paper discusses two different but intrinsically related issues. on the one hand, the identification of a problem in the provision of lps. the origin of this situation lies in the lack of capacity of most municipalities in spain to meet the basic needs of their neighbourhood in an efficient and accessible way in space and time, which is apparent in: • lack of internal market, lack of population and dispersed settlement. • lack of a developed productive fabric with the capacity to attract people. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 75 • dependent financial structure, preventing any management and political autonomy. on the other hand, after identifying the problem to be solved, "how to improve the provision of lps through an optimal local government in size and shape", the work moves on to how this problem should be communicated. besides, made known to all parties concerned, so that they understand and accept it as something globally good. the work moves on to how this problem should be communicated and brought to the attention of all affected parties, so that they comprehend it and take it as something globally good. so, being political pedagogy and the exposition of the expected results through the identification of the benefits and costs derived from the proposed administrative reform the way to make people see the need to modify the management framework at the local level, changing the mentality of the social agents. therefore, the starting point for the success of the proposed change is a good knowledge of the management (ge) carried out to date and the knowledge (co) of the environment (me) or territory being explored. in this sense, the next study step is to decide if the conclusions of the resulting technical report should be advisory or binding. in this sense, the spatial evolution, both demographic and socioeconomic, leaves us with little doubt that if we do not take immediate measures, we will be able to do little to avoid the abandonment and relocation of most of the national territory. finally, and as a social justification for the proposal presented, there are several arguments in favour of this administrative change, including the following: 1. the size and complexity of public administrations makes their adaptation very costly, making them obsolete in terms of the effectiveness of their proposed organisational solutions and even more so in a society in constant change. 2. the competitiveness of a country and the quality of life of its citizens is a function of the efficiency of its administration. having a public administration that does not create dysfunctions in private activity and efficiently performs its tasks: regulation, control and management of services and stimulation of socio-economic activity, is a necessary condition for the country's growth and development. 3. all public resources have the same origin: productive activity and, within this, mainly private activity. for this reason, the administration, more than anyone else, must be responsible for the use and profitability of the funds placed at its disposal. and as the first function of any administration is precisely to "administer" and not to produce, if the administrative size implies duplicities and rigidities, which are the origin of a waste of resources, this is something that we cannot consent to by sitting back and waiting. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 76 references alfaro, c., y gómez, j. (2016). un sistema de indicadores para la medición, evaluación, innovación y participación orientado a la administración pública. methaodos, social science journal, 4(2), 274-290. https://doi.org/10.17502/m.rcs.v4i2.124 bel, g. (2011). servicios locales: tamaño, escala y gobernanza. barcelona institute of economics, ieb report on fiscal federalism, ed. ieb, barcelona, 12-20. blasco, n., cuairán, r., y twal d. (2013). sistema automático de diagnóstico y propuestas de mejora económico-financiera para universidades públicas. proceedings of the xxiii conference of the economics of education association (aede). bonnefoy, j.c., y armijo, m. (2005). indicadores de desempeño en el sector público. latin american and caribbean institute for economic and social planning (ilpes). manuals series nº 45. bosch, n., y solé-ollé, a. (2012). una valoración preliminar de la reforma de la administración local en españa. barcelona institute of economics, ieb report on fiscal federalism, ed. ieb, barcelona, 80-87. centro de investigaciones sociológicas (cis) (2015-2020). calidad de los servicios públicos xiii. report, "study 3229". http://www.cis.es/cis/export/sites/default/-archivos/marginales/ 3220_3239/3229/es3229mar.pdf de la mora, l. (1997). los retos de las ciudades en el umbral del siglo xxi. los nuevos retos del sector público ante la unión monetaria europea. publishes 5 days newspaper and official credit institute foundation, 12. del campo, m.o., miguéns, v., y cid, a. (2021). el diagnóstico territorial: un instrumento en defensa de la provisión de los servicios públicos locales. administration & citizenship, galician school of public administration journal, 16(2), 161-180. https://doi.org/10.36402/ac.v16i2.4774 del campo, m.o. (2014). la reforma de la administración local: una aproximación práctica al caso español. criteria res publica fulget, journal of political and social thought, (13), 3242. del campo, m.o., y ferreiro, f.j. (2013). the reform of the local administration in spain in terms of population. journal of us-china public administration, 10(10), 945-957. durán, f.j. (2016): la fusión coactiva de municipios en españa. chronicler of the social and democratic state under the rule of law, 57, 44-51. elizalde, b. (2020). la diversidad familiar ante el reto de los cuidados. economists without borders. dossiers esf, 36, 34-38. galera, a. (2019): razones para una reforma federal del modelo territorial. administration & citizenship, galician school of public administration journal, 14(1), 29-39. guy, g. (2011). más allá de las asociaciones intermunicipales hacia municipios de dos niveles de gobierno: el caso francés barcelona institute of economics, ieb report on fiscal federalism, ed. ieb, barcelona, 32-42. real decreto sobre la división civil de territorio español en la península e islas adyacentes en 49 provincias y estableciendo los subdelegados de fomento en las provincias del reino. boletín oficial del estado, 154, de 3 de diciembre de 1833. https://www.boe.es/buscar/doc.php?id=boe-a-1833-1241 statistics national institute (2018-2020). indicadores de calidad de vida. in https://www.ine.es/ss/satellite?l=es_es&c=inepublicacion_c&cid=1259937499084&p=125 4735110672&pagename=productosyservicios%2fpyslayout&param1=pysdetallegratuit as&param2=1259944522579&param4=mostrar#top law 27/2013, of rationalization and sustainability of the local administration. lópez doblas, j. (2018). formas de convivencia de las personas mayores. spanish journal of sociological research, 161, 23-40. http://dx.doi.org/10.5477/cis/reis.161.23 ministry of public administration (mpa) (2016-2020). encuesta nacional satisfacción de la calidad de los servicios públicos en la administración pública general. file in https://observatorioserviciospublicos.gob.do/publicaciones/encuestas/institucionales/cuestio nario_encuestas.pdf organic law 2/2012, on budgetary stability and financial sustainability. ramió, c., y salvador, m. (2012). provisión de servicios públicos en el contexto de la administración local de españa: el papel de los factores políticos institucionales y la externalización de los gobiernos locales. management and public policy, 21(2), 439-474. https://doi.org/10.17502/m.rcs.v4i2.124 http://www.cis.es/cis/export/sites/default/-archivos/marginales/3220_3239/3229/es3229mar.pdf http://www.cis.es/cis/export/sites/default/-archivos/marginales/3220_3239/3229/es3229mar.pdf https://doi.org/10.36402/ac.v16i2.4774 https://www.boe.es/buscar/doc.php?id=boe-a-1833-1241 https://www.ine.es/ss/satellite?l=es_es&c=inepublicacion_c&cid=1259937499084&p=1254735110672&pagename=productosyservicios%2fpyslayout&param1=pysdetallegratuitas&param2=1259944522579&param4=mostrar#top https://www.ine.es/ss/satellite?l=es_es&c=inepublicacion_c&cid=1259937499084&p=1254735110672&pagename=productosyservicios%2fpyslayout&param1=pysdetallegratuitas&param2=1259944522579&param4=mostrar#top https://www.ine.es/ss/satellite?l=es_es&c=inepublicacion_c&cid=1259937499084&p=1254735110672&pagename=productosyservicios%2fpyslayout&param1=pysdetallegratuitas&param2=1259944522579&param4=mostrar#top http://dx.doi.org/10.5477/cis/reis.161.23 https://observatorioserviciospublicos.gob.do/publicaciones/encuestas/institucionales/cuestionario_encuestas.pdf https://observatorioserviciospublicos.gob.do/publicaciones/encuestas/institucionales/cuestionario_encuestas.pdf del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 77 rodríguez, r. (2006). el territorio local en europa. reestructuración de su base organizativa y posibilidades para españa. geography, territory and environment, nº6, 115-132. rodríguez, r. (2011). oportunidades de la crisis. debate sobre reestructuración de la administración territorial. administration & citizenship, galician school of public administration journal, 6(2), 21-34. del campo villares et al. / european journal of government and economics 12(1), june 2023, 58-78 78 appendix. questions on citizens’ perception of the administration. note. the related questions have been taken from the "quality of public services" and "public opinion and fiscal policy" surveys, both prepared by the cis. these questions were completed and corrected in those cases where the question was altered or modified in its design or configuration during the period under study (2016-2020) with the subjective welfare indicators obtained through a survey by the ine and the information published by the map on periodicity and compliance in the provision of the public services surveyed. reviving social europe: a critical assessment of eu social policy in the wake of multiple crises © the author(s) 2023. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 12, no. 2 (2023), pages 105-118 https://doi.org/10.17979/ejge.2023.12.2.9541 submitted: feb 28, 2023 accepted: oct 24, 2023 published: dec 5, 2023 article reviving social europe: a critical assessment of eu social policy in the wake of multiple crises adrian hawley 1, * 1 royal holloway, university of london, united kingdom *correspondence: adrian.hawley@rhul.ac.uk abstract. the concept of "social europe" encompasses the european union's social policy framework. this paper argues that the financial crisis of 2007-2009, the subsequent euro crisis, and the ensuing austerity measures diverted significant attention away from social europe. this neglect led to declining living standards, reductions in public services, and the emergence of critical challenges such as the gig economy. in recent years, the covid-19 pandemic has further exacerbated these issues, prompting a renewed focus on social europe by eu institutions. this paper conducts a chronological analysis of recent legislative (acquis) and non-legislative (soft power) instruments implemented to address these challenges and assesses the prospects of social europe in the near term. keywords: social europe; european union; crises; acquis; soft power jel classification: h1; n44 1. introduction a constant feature of the eu discourse is social europe. in this paper, i ask three questions. firstly, what does it mean? secondly, to what extent is it substantiated and thirdly, is it any nearer to being so in an era of polycrisis? as jean monnet famously remarked in 1978, ‘‘europe will be forged in crises, and will be the sum of the solutions adopted for those crises’’. i hypothesise that critical events over the past decade have latterly led to a sharpened focus on social policy and its mode of governance by the eu institutions, exemplified by instruments of both a legally binding (hard law) and particularly a normative (soft law) nature which have been a characteristic modus operandi of the eu in this field. achieving consensus in a policy field where the eu only (currently) has limited competence acts as a continual brake. moreover, we should not overlook that social policy entails both expenditure and regulation, the former on welfare and health, and the latter on regulation of such matters as employment rights. as we shall see throughout this paper the eu is proficient at regulation but finds it more problematic when it comes to raising and committing the funds required to implement it. what has happened in the eu as a result of the spate of crises fits with a view of historical institutionalism, which holds that exogenous factors may not directly change policy-making but lead https://creativecommons.org/licenses/by-nc/4.0/ 106 adrian hawley to self-induced agent-led endogenous transformations following a period of drift and whose eventual outcome is described as ‘layering’ (cartwright, 2019). this is evident, as i shall show, in the eu’s slowly evolving (and contested) normative and regulatory response not only to new labour practices, namely the platform economy but also to its re-orientation towards promoting the social market economy. the social policies of ms borne out of historical cultural and political traditions are many and varied, but procedures for dialogue among ms and eu institutions have increased markedly since the financial crash of 2007-9 and the ensuing eurocrisis. the eu doctrine of a ‘european social market model’ of which social policy is a component is accepted to a greater or lesser extent among ms but an idée fixe of (re)gaining economic competitiveness at union level and the protection of national sovereignty – the westphalian legacy among member states in an age of multiple crises hinders the adoption of a common response particularly through legally binding instruments. i will show, however, that a series of crises has gradually led to a renewed impetus and new ways of collaboration. a return to more stable conditions could cause it to fall back. but climate change and sustainability now represent something more like a ‘permacrisis’ where social and economic effects are inseparable. the paper takes the following form. firstly, i explain my methodology, which consists of a chronological content analysis of a spectrum of recent primary documentation relating to social policy formulation by the eu institutions. the chronology traces the stages of recent social policy formulation, adoption and, wherever the case, addition to the acquis. i explain why these documents were selected and look for patterns in the three principal actors, namely the commission, european parliament (ep) and council. secondly, i define what is referred to in these documents as ‘social europe’, an eye-catching shorthand for eu social policy. social europe is a very broad term used by politicians, academic analysts and the media and particularly the unifying vision of the social democrats’ eu agenda for several decades (shrzypek, 2023:782). but are we referring to the ‘core’ welfare state covering, as scharpf (2001:27) put it, provision and financing of means-tested social assistance, of earningsrelated social insurance covering income losses in cases of unemployment, sickness and disability, and in old age, of health care, and of social services for families with small children, for the handicapped, the sick, and the aged’? or, do we also mean social rights in a wider context which cover employment, universal health service and non-discrimination on grounds of gender, ethnicity, age or sexual orientation? it is difficult to separate them because access to all these is generally regarded by their advocates as ‘human’ rights (which was not always the case). we also hear, alongside social europe, references to the ‘social market economy’ (a fusion, possibly mythical of neo-liberal and mildly progressive social democratic positions) and latterly ‘the ‘social economy’ comprising cooperatives, mutual benefit societies, associations and charities, foundations, and social enterprises (ec, 2021). we must also include ‘social dialogue’ between the social partners relating to employment issues. all these terms add up to what is also commonly described as the ‘social dimension’ and invoke the idea (or more accurately the ideal or even a chimera?) of a specific ‘european social model‘. thirdly, i look at the evidence from the chronological analysis and find that as a specifically eu institutional enterprise, achieving social europe as far as it has been, has become a multi-layered process reviving social europe depending upon an eclectic and by no means coherent mixture of procedural practices, principles and recommendations and a paucity of hard law instruments, namely directives, regulations and decisions. 2. methodology from a triage of eu institutional documents over the past decade i have found twenty-three most salient for my hypothesis. these are presented in tables 1-3 in chronological order. table 1. ec documentary evidence subject date reference commission president juncker’s plan for the european pillar of social rights 2015 ‘setting europe in motion: main messages’, opening statement in the ep plenary session’, strasbourg, 22/10/ 2015). establishing the epsr 2017 communication (com(2017) 250 final, 26/04 2017 refers inter al. to a new social scoreboard within the european semester (es) the epsr in twenty principles 2017 proclamation at gothenburg social summit 17/11/2017 transparent and predictable working conditions in the eu 2019 directive (eu 2019/1152, 20/06/2019 revisions to work–life balance for parents and carers 2019 directive (eu) 2019/158), 20/06/2019 repealing council directive 2010/18/eu special euro barometer (509) 2021 social issues, november december 2020, pub. march 2021. eprs action plan 2021 communication on monitoring the implementation of the european pillar of social rights, 04/03/2021 porto social summit 2021 ‘taking forward the eprs and strengthening europe’s social dimension’, 07/05/2021 social economy action plan 2021 ‘to boost the social economy and create jobs,’ employment policy 2021 proposal for a council decision, 02/06/2021 adequate minimum wages in the eu 2022 directive (eu 2022/2041) 19/10.2022 the future of social protection and of the welfare state in the eu 2023 report (jan 2023) of the high-level group that met between november 2021 and december 2022 strengthening social dialogue in the eu 2023 proposal for a council recommendation com (2023) 38 �inal and communication com (2023), 40 �inal, both on 25/01/23 108 adrian hawley table 2. ep documentary evidence subject date reference the european semester 2021 economic policy coordination: employment and social aspects in the annual sustainable growth strategy, plenary sitting text, 02/03/2021 revaluation of conditions and wages for essential workers 2022 study by d-g internal policies emp committee (pe 703.344) jan 2022 evolving eu governance 2023 study requested by econ committee (pe 733.742) february 2023 adequate minimum income 2023 ep resolution (p9_a(2023)0076) text adopted) 15 march 2023 road map towards social europe 2023 ep resolution (p9_ta(2023)0203 text adopted) 11 may 2023 call for tender 2023 for external expertise in the �ields of employment, social policy and social protection and health and safety at work, (ip/a/empl/fwc 2023-028), 08/06/23 table 3. council documentary evidence subject date reference access to social protection workers and the selfemployed 2019 recommendation (c 387/01), 08/11/2019 guidelines for employment policies of ms 2021 decision (eu) 2021/1868), 15/10/2021 adequate minimum income 2023 press release (30/01/2023) on adoption of commission recommendation(28/09/2021) porto biennial forum 2023 notice (6912/23) from employment and social affairs council (epsco), to ms delegations 03/03/23 as is apparent, more relevant documents have emanated from the commission than either the parliament or the council (of ministers), as might be expected as it generally has the privilege of initiating legislative and non-legislative proposals. several conclusions can be drawn from this chronology. firstly, the impetus for greater emphasis on social policy derives from the commission. why might this be so in addition to the reason given above? what might make the commission more socially minded? the juncker commission (2014-2019) has been described as a ‘political commission’ (but not more party political) compared with its immediate forebears (kassim and laffan 2019) and juncker was the first president to be selected as the leader of the largest party (epp) in the may 2014 ep elections (the socalled spitzenkandidaten process). juncker’s election manifesto expressly ‘aimed to turn the page on austerity’ and respond to challenges that had been neglected during the financial and economic crisis. this was accompanied by flexibility in adhering to the stability and growth pact (sgp) rules. juncker’s objective was reflected from the outset by the selection of franz timmermans as first vice president responsible for several portfolios, among them, the rule of law and the charter of reviving social europe fundamental rights and marianne thyssen, commissioner for employment, social affairs, skills and labour mobility. with less fanfare, we can see from the chronology that the von der leyen commission has put ‘some flesh on the bones’ of juncker’s social initiative. in contrast to the latter’s grandiloquent ‘triple a for social europe’, von der leyen’s presidential manifesto contented itself with the only slightly less rhetorical ‘an economy that works for people: ensuring social fairness and prosperity’. more recently, we can see that such proposals have been endorsed by the ep and the council, though the former had been ‘the defender of social europe during austerity’ and the conclusions of the 2017 council ‘were the most extensive on social europe since the launching of the lisbon strategy in 2000’ (copeland, 2022:1639). secondly ,the chronology reveals mainly ‘soft law’ measures (recommendations and principles) arising out of reports, studies, opinions and resolutions, a very small number of which have resulted so far in additions to the acquis. this is hardly surprising given that social policy is mainly a national competence and the eu’s role is chiefly one of co-ordination. relating to the latter, the reanimation of the european semester (es) is apparent and specifically its engagement with social policy, though the extent of its influence is disputed and requires further research (ma, 2023:36; zeitlin & vanhercke, 2018:169). thirdly, the moving spirit and guiding force has been the european pillar of social rights (epsr), an ambitious project of the juncker commission, conceived soon after the eurocrisis and a response to electoral discontent with the effects of fiscal consolidation (i.e. austerity). the incoming von der leyen commission was soon faced with responding to the economic and social dislocation of the covid-19 crisis, which it did with massive income support (sure) and subsequently the €723 billion recovery and resilience facility (rrf) in grants and loans adopted in february 2021. ‘contributing to the implementation of the european pillar of social rights’, is mentioned no less than ten times in the regulation (eu, 2021) and also in other communications from the commission. of the rrf’s six pillars, only one refers specifically to the social dimension and another is largely devoted to health services, pillar 4: social and territorial cohesion, ‘especially contributing to the implementation of the epsr’ pillar 5: health, and economic, social and institutional resilience ‘with the aim of, inter alia, increasing crisis preparedness and crisis response capacity’ what is less clear, however, is the extent to which redistributive social policies are advantaged, let alone mainstreamed in national recovery and resilience plans (nrpps). i will return to this aspect in a later section of this paper. a further important set of documents was also consulted. these relate to precarious working conditions in the platform economy. the social protection of workers is inherent in social policy. platform work is a specific but important instance but it should not be left out. its rise has been facilitated by the digital revolution but has coincided with the financial crises as traditional forms of more reliable employment have declined. 110 adrian hawley table 4. a chronological triage of eu institutional documentary evidence on the platform economy institution and subject date reference ep the platform economy and precarious work’ 20172020 studies y by d-g internal policies (1) 07/12/ 2017 and (2) 17/09/2020 ec impact of the digital transformation on eu labour markets 2019 report of the high-level expert group 08/04/2019 council access to social protection for workers and the self-employed 2019 recommendation (2019/c 387/01) ec improving working conditions of platform workers 2020 2021 study to gather evidence s (vt/2018/032 final report), 13/03/2020 study of impact of an eu initiative, final report ), 22/10 2021). proposed directive (com(2021) 762 �inal), 09/12/2021 ep fair working conditions, rights and social protection for platform workers 2021 resolution (p9_ta(2021)0385) 16/09/2021 ec access to social protection for workers and the selfemployed 2023 report on council recommendation (com 2023)43 final), 31/01/2023 council – rules on platform work 2023 press release 12/06/23. agrees position and ready to negotiate text of proposed new directive with ep the documents in tables 1-4 were selected as milestones in the eu’s institutional discussion on social policy following the eurocrisis, which started in 2009. a number of conclusions stand out. firstly, we can see that attention, which was diverted during this period, was first re-directed by the incoming commission in 2015, specifically by the new president jean-claude juncker and his foundational initiative the epsr. secondly, it was the commission that successively ‘ran with the ball’ accelerating during successive crises, namely the covid-19 and energy crises. the latter brought the ep into the discussion when the social dimension was formally added to the existing european semester of policy coordination (the es) – whose legal foundation was itself established following the eurocrisis. the covid-19 crisis gave rise to several social policy initiatives, namely “support to mitigate unemployment risks in an emergency” (sure), and a strengthening of cohesion support, (react-eu) and most prominently the €723 billion recovery and resilience facility (the cornerstone of newgeneration eu). there is debate, however, as to the extent of allocation in member states’ national recovery plans (nrps) of rrf funds to social policies (rainone and pochet 2022:28,30). the contribution of the council has been mainly confined to a third conclusion as we can see from the chronology (as shown in tables 3 and 4), namely the rise of the platform economy, a critical change in employment conditions over the past decade. the chronological evidence moreover reveals a change of direction in the commission towards the platform economy. up to the proclamation of the epsr, the commission was repeatedly stating (e.g. bieńkowska, 2017a; bieńkowska, 2017b; bulc, 2017) that it did not intend 'at this stage' to introduce any new legislation to regulate the platform economy’, exemplifying ‘a certain regulatory reluctance’ (ep, 2017:3) on its part and a preference for ‘creating the right framework reviving social europe conditions and the right environment’, namely soft rather than hard law initiatives. the change is evident in table 4. the council’s ‘conversion’ to the issue of the platform economy is also quite recent – no previous reference on the subject having been found, rather a preoccupation with ‘mainstreaming’ competitiveness following the eurocrisis. 3. defining social europe the adjective ‘social’ and the proper name ‘europe’ do not sit easily together. let us assume that by ‘social’, is meant ‘social justice’ or ‘fairness’, perhaps in the form famously expressed in john rawls’ theory of justice (1977, 1997). by ‘europe’, in this context, is meant the eu, and specifically, its supranational policy (in so far as it exists) manifest in hard and (mainly) soft law instruments under the treaty. achieving ‘fairness’ as an objective, in our western contemporary sensibility, has gradually extended to most areas of our communal life. few would overtly dissent with the intent, but succeeding remains problematical – and divisive for two principal reasons. firstly, there is a seemingly permanent conflict of policy priorities between competitiveness (‘growth’) – some would call it the ‘economic realities’ and distribution ( denounced by some as ‘creeping ‘socialism’). the chronology presented above shows that after a period in which the former was prioritised following the financial crises there has been a (re)turn towards the latter in recent years. not that the social repercussions were not quite rapidly appreciated at least by the incoming commission president in 2014. meanwhile, the commission continued to emphasise the doctrine of innovation and competitiveness as critical in the face of globalisation. although the tide was turning with the proclamation of the epsr, it would take further crises of a wholly unanticipated character, namely covid-19, the russian invasion of ukraine and resulting high inflation (particularly food and energy) to further ‘shift the dial’. secondly, the eu is not ‘europe’ or a proxy for it. a social dimension, however, has existed since the founding treaty creating the eec in 1957 taking the form of action plans, programmes and charters. a social protocol and agreement among eleven ms was adopted with the formalisation of the union at maastricht (teu) and the social chapter, a facilitating legislative mechanism, at amsterdam in 1997 (see figure1). since then, social policy has been adopted, which is presumed to be legitimised by the ‘social clause’ in the current treaty of lisbon (tfeu art.9). according to this, “in defining and implementing its policies and activities, the union shall take into account requirements linked to the promotion of a high level of employment, the guarantee of adequate social protection, the fight against social exclusion, and a high level of education, training and protection of human health.” 112 adrian hawley figure 1. milestones in ec/eu social policy from tec to the epsr the union’s competence, however, remains opaque, resulting in ambiguity as to the binding nature of eu policy in the social field. this is exemplified by the eus’ shared competence under tfeu. the operative role of the eu is defined in art.5 (3), namely, the union may take initiatives to ensure coordination of member states' social policies. but, finally, as laid down in tfeu art.153, 4 “[the provisions]shall not affect the right of member states to define the fundamental principles of their social security systems” it is to be concluded, therefore, that the eu’s role is one of coordination and not harmonisation, of procedure rather than regulation, and that of member states one of ‘mutual adjustment’, to borrow scharpf’s (2001:2.6) term for cooperation ‘where european solutions are not available’. any semblance of a common social policy is further limited by two factors. one is provided by the principles of subsidiarity and proportionality (art.g. clause 5(3b) teu; protocols 1. and 2.tfeu), confirmed in an institutional agreement in 2016. the other, as observed by liebfried and pierson (1991:22), and later by pochet (2017) is that under art.153 2(b) tfeu, decision-making in the council on social policy, still requires unanimity (despite some changes under the social chapter) in many areas, including social security, social protection of workers and some aspects of nondiscrimination and labour law. these are ‘deeply embedded in national economic, taxation and income redistribution models which differ greatly across the eu’ (ec2019:11). despite and because of the above, social policy has converged to some extent among member states but not amounting to either de facto or a de jure commonality. streeck (1995) observed voluntarism in a nested game between national welfare states and supranational constraints in which the former were not absorbed through integration, though not unaffected by it. the growing effects of the platform economy and the resulting institutional discourse traced in the chronology have added a further dimension to the formulation of eu social policy. let us suppose the rawlsian objective of social justice, predicated on fairness meaning equal access to all the opportunities afforded by living in contemporary society such as we know it. if we think of reviving social europe essentials such as access to health care, education, housing, communication, mobility, personal safety and to the law, we can hardly exclude decent working conditions and adequate (minimum) income. all these seem to currently invoke in some way the term ‘crisis’ by which i mean radical changes taking place with adverse societal effects which seem largely out of control and whose outcomes are hard to predict. changes to working conditions that have arisen following the pandemic and the rise of the platform economy cannot be disassociated from the formulation of social policy. they are inherent in ‘social europe’ should we wish to use that term and closely associated with a further aspect in defining it, as mentioned above, namely social dialogue. engaging the social partners throughout the policy formulation procedure has now received renewed attention from the commission and council as will be shown in the next section. the chronological evidence presented above has shown that a renewed attention to social policy in the eu is correlated with an exceptional number of recent crises. only in retrospect will we be able to tell if it has been a ‘critical juncture’ meaning a relatively short space of time during which there is a substantially heightened probability that agents’ choices will affect the output of interest’ (carella & graziano, 2022:378, original emphasis). that is a question that cannot be answered now, but i turn back to the chronological evidence to make some assessment of the relative importance of recent and pending legislative changes under critical conditions. 4. substantiating social europe through crises if a ‘leap forward’ in the social dimension is now occurring through times of crisis as the chronology presented above suggests, it will be consistent with monnet’s famous dictum. nor would it be the first time that social change has followed a devastating crisis which has come to be seen, in retrospect, as a critical juncture. bringing about decisive policy changes in a quasifederal bloc such as the eu faces what scharpf (1988) called the ‘joint decision-making trap’. the question, therefore, is what developments have occurred under recent critical conditions to exit the ‘trap’ and accelerate social europe? procedures have waxed and waned and, recently, waxed again over the past two decades but i point to the significance of two, namely the open method of coordination (omc) and the european semester (es), together with country specific recommendations and the social scoreboard to which it has given rise. one of the goals of the lisbon agenda of 2000 was ’modernising the european social modern model ‘by structural reforms of domestic labour markets and welfare states in parallel with policies to tackle social exclusion and increased investment in education and training’ (hix and høyland 2011:202). in this respect it failed, nor did it reach its overall objective of turning the the eu into ‘the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion’. to avoid constraining national objections (the decision-trap) without recourse to legally binding instruments, requiring unanimity in the council in some cases, a method known as omc was adopted, which involved governments monitoring each other’s progress towards agreed goals and ‘naming and shaming those that did not honour their commitments’. the commission under barroso tried again in march 2010 with the launch of the europe 2020 114 adrian hawley strategy – just before fears of greek sovereign debt default presaged the beginning of the eurocrisis . a social policy flagship initiative (ec, 2010:18) the european platform against poverty and social exclusion was, “to transform the open method of coordination on social exclusion and social protection into a platform for cooperation, peer-review and exchange of good practice, and into an instrument to foster commitment by public and private players to reduce social exclusion, and take concrete action, including through targeted support from the structural funds, notably the esf (european social fund).” member states were enjoined to collectively take responsibility for combating poverty and social exclusion and to deploy their social security and pension systems to ensure adequate income support and access to health care. its policy content was criticised, however, ‘as it cleaves to the growth and market-first model definitive of liberalism – it underplays social rights, quality job creation or a broad-ranging social programme’ daly (2012:283). looking back, it is evident that the ensuing crisis did anything but promote its goals. instead it ushered in an era of fiscal consolidation (better known as austerity). regaining competitiveness and growth prevailed again. as can seen from the chronology (table 1.), however, the incoming juncker commission took notice, its legacy being the proclamation of the epsr in november 2017 and the addition of the social scoreboard in the es (visualisable as a series of interactive charts), promoting juncker’s ambitious objective to achieve, as he called it a ‘social triple a for the eu as a whole’. the eu’s continuing, as opposed to emergency, response to the unprecedented covid-19 crisis is the rrf. the only criteria for funding are ‘relevance, effectiveness, efficiency and coherence’ and milestones and targets being met by 1 august 2026 (eu, 2021). noteworthy is a pervasive requirement that projects support the principles of the epsr. as stated in the preamble to the rrf regulation, the semester has become the forum for reviewing ms’ performance in this respect with their progress monitored on the social scoreboard. since the rrf was established only two and a half years ago and without every national recovery and resilience plan (nrpp) having yet been submitted (or approved), it is far too early to observe any lasting effects, including those in the field of social policy. we can, however, get some idea of ms’ priorities for the grants and loans on very favourable terms which could not have been anticipated before the covid-19 crisis. what is of specific interest here is the extent of (intended) contribution of nrpps to implementation of the epsr. for this we refer to four further institutional documents (table 5). taking these assessments chronologically, the three ep committees raised concerns that nrrps should allocate the use of rrf funds for reform initiatives contributing to the implementation of the epsr and not as a substitute for recurring national budgetary expenditure. their plans would not be endorsed in that case. reviving social europe table 5. assessment of the implementation of the recovery and resilience facility. institution date ep report by: committee on budgets, committee on economic and monetary affairs 08/06/2022 ep opinion of: committee on employment and social affairs 08/06/2022 ec review report 29/07/2022 ec proposal for a council recommendation on strengthening social dialogue in the european union 25/01/2023 in terms of the two ‘social’ pillars of the rrf (see section 2) the commission’s report finds that of the twenty five plans adopted by the council, 16% of total expenditure is allocated to policies related to health, economic, social and institutional resilience ‘reflecting also the responses to the covid pandemic’ (ec, 2022a: 8,9). where social expenditure is taken to cover four categories of employment and skills, education and childcare, health and long-term care, and social policies, social spending is reported as representing about 30% of the total expenditure. ms ‘significant’ contribution to implementation of the epsr is specifically noted. whether the description proves justified during the years following the current rrf ‘window’ ending in 2026, remains to be seen. 70% of the rrf programme is allocated to nrpps for green and digital transitions with at least 37% for climate and 20% for digital objectives per ms. could the crises threatened by the latter divert attention from implementing the epsr? 5. conclusion i started with monet’s dictum that the eu moves forward with crises. the past decade or so has seen as profusion amounting to what is being described as a polycrisis. using a chronological methodology of analysing institutional documents over this period, i have hypothesised a correlation between such crises and progress towards (re)establishing social europe. i have explained what i take to mean by this polyvalent headline term, its secular constraints and the eu’s legitimacy for intervening to promote it. the latter, as i showed, is largely, but by no means entirely, restricted to decisions that are not legally binding on ms. i also showed that alternative procedures for collaboration which had existed before the age of polycrisis have now been revived and augmented. the eu’s response to the covid-19 pandemic, a crisis of unprecedented proportions since the treaty of rome, is the rrf (the centre-piece of nextgenerationeu). i showed that one of its central elements is support for the principles of the epsr. since then, the equally unanticipated russian invasion of ukraine, the energy shock, high inflation and the looming confrontations of mass immigration, ai and climate change have contributed to the sense of polycrisis. how then might its effects shape the near-term future of social europe? could even a ‘post-crisis’ europe be imagined recalling the great strides in social policy in 116 adrian hawley many parts of europe after world war 11 and starting a decade prior to the the eu’s founding treaty? on the other hand the effect of the financial crises that started in 2007 was that new modes of economic governance in the eu intensified attention to fiscal continence and diverted attention from progressive social policies until mainstreamed by the juncker presidency. an exogenous event of this magnitude failed to change the direction of eu social policy (carella & graziano, 2022). much will ride on the rrf and on the twenty principles of the epsr whose implementation by ms will achieve the commission’s goal of ‘more social europe by 2030’ (ec, 2022b). to the extent that they are, the epsr could in retrospect come to be seen as a critical juncture in decisively furthering social europe. this is, however, by no means the first time that the eu (like its predecessor the eec) has attempted to establish its credentials as the guardian of social protection (see fig 1.) with reforms to the agendas of ms. in the current case, there are, however, a number of constraining factors which can be categorised as follows. firstly, there is an asymmetry of power relations. rrf funds are only disbursable on certain conditions under the rrf regulation with nrp’s vetted in the es. based on the analysis of the country reports, the commission proposes adoption of country-specific recommendations (csrs) to the council. thus compliance is administered by the commission and the council ‘with the ep only playing a marginal role’ (rittberger, 2022:20) as an advisor. not that there is any doubt as to where its intentions lie. as resolved in its resolution on a roadmap towards a social europe (see table 2), the epsr needs to be reinforced with legislation at eu and ms level and provision made in the multi annual financial framework (maff) for social investment with convergence seen as a top priority. the reference to maff recalls the ineffectiveness of regulation without expenditure. since in the case of rrf, funds are approved or withheld, the es becomes an ‘instrument of hard rather than soft governance’. the avoidance of excessive debit procedure under the sgp (stability and growth pact) rules, even though currently relaxed under the general escape clause, still exerts a strong, possibly decisive, influence, in determining the acceptability of nrps for rrf as it did a decade ago in the context of the financial crisis (de la porte & heins, 2014:17; sabato & vanhercke, 2017:73). the eu’s conflicted relationship between economic and social policy goes back further to the late 1990’s and the legitimacy of the former has always been clearer than that of the latter (zeitlin & vanhercke, 2018:150,155). as rittberger (2022:22) says, ‘a country’s financial stability is at the core of the es’ and this may explain why the proportions of funds so far agreed and disbursed for nrp’s ‘social’ objectives is relatively small compared with digital and green transitions. secondly, there is a question of durability and reach. rrf is an essentially short-term response to the covid-19 crisis and it cannot be predicted whether or not it will be extended or renewed. and the epsr only formally applies to eurozone ms – others can join if they want to. thirdly, social policy is itself a heterogeneous mixture of objectives and priorities. it is as much orientated towards prevention a well as relief, resulting in a doctrinal conflict of priorities. the former aims at eliminating poverty and social exclusion by social investment measures, whereas the latter aims at relieving their effects through welfare (i.e. the safety net). the two poles are reflected in the ‘recovery’ and ‘resilience’ of the rrf regulation, the former towards relief and the latter towards prevention in future crises (not least in the event of another pandemic). on the other hand, there may be a greater role for direct democracy. maybe the effect of current crises will result in public discontent reflected in the rise again of the populist vote as it did reviving social europe after the financial crisis. it could embolden a future commission to undertake a new social policy project of the significance of epsr, as did the juncker commission after the financial crisis, or to make its principles more binding. that will take a degree of public consultation even wider than for the epsr, itself novel in its scale. the 2021-2022 conference of europe adopted proposals covering forty nine different policy objectives. in the social field, these comprised full implementation of the epsr, involving common frameworks for minimum income levels and minimum wage protections in each member state, affordable and accessible quality childcare across the eu, common minimum healthcare standards at the eu level and a european education area within which all citizens have equal access to quality education and life-long learning (fella, 2022:6). many of the conference’s proposals, however, would require treaty change which does not have unanimous appeal throughout ms. in 2021 the commission launched a pilot project in italy and twelve other ms for esspass, a digitally enabled system to make it easier for people to exercise their social security rights and duties when moving and working throughout the eu. finally, i return to the aspect of social dialogue alluded to earlier (section 1) as an aspect of social europe or the social dimension. the porto summit (see table 1) referred to it as ‘a structuring component of the european social model’. it is promoted by the conference of europe in its final report (proposal 13) and recognised by the treaties and underlined by principle 8 of the epsr (see table 1. commission proposal for a council recommendation and communication 25 jan 2023). the two pillars of principle 8 are consultation on economic, employment and social policies and collective bargaining, ‘according to national practices’. but therein lies a constraint as some nordic ms, notably sweden, reject imposition of an eu wide regulatory basis for a minimum wage and collective bargaining in favour of their own separately negotiated collective agreements. crises may deepen and strengthen social europe and there is evidence that the most recent have generated a turn in that direction. but as the financial adverts warn us, the past cannot be taken as a guide to future performance. references bieńkowska, e. (2017a). parliamentary answer to question e-007579/2016, given by the commissioner for the internal market and consumer protection. retrived from http://www.europarl .europa.eu/sides/getallanswers.do?reference=e-2016-007579&language=en bieńkowska, e. (2017b). parliamentary answer to question e-000245/2017', given by the commissioner for the internal market and consumer protection. retrived from http://www.europarl.europa.eu /sides/getallanswers.do?reference=e-2017-000245&language=en bulc, v. (2017). parliamentary answer to question e-004792017, given by commissioner for transport. retrived from http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2017004790&language=en carella, b. & graziano, p. (2022). back to the future in eu social policy? endogenous critical junctures and the case of the european pillar of social rights', jcms: journal of common market studies, 60(2), 374-390. https://doi.org/10.1111/jcms.13236 cartwright, m. (2019). historical institutionalism and technological change: the case of uber. business and politics, 23(1), 67-90. https://doi.org/10.1017/bap.2019.23 copeland, p. (2022). the juncker commission as a politicising bricoleur and the renewed momentum in social europe. cms: journal of common market studies, 60(6), 1629-1644. https://doi.org /10.1111/jcms.13336 http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2016-007579&language=en http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2016-007579&language=en http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2017-000245&language=en http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2017-000245&language=en http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2017-004790&language=en http://www.europarl.europa.eu/sides/getallanswers.do?reference=e-2017-004790&language=en https://doi.org/10.1111/jcms.13236 https://doi.org/10.1017/bap.2019.23 https://doi.org/10.1111/jcms.13336 https://doi.org/10.1111/jcms.13336 118 adrian hawley daly, m. (2012). paradigms in eu social policy: a critical account of europe 2020. transfer 18(3), 273-284. https://doi.org/10.1177/1024258912448598 de la porte, c. & heins, e. (2014). a new era of european integration? governance of labour market and social policy since the sovereign debt crisis. comparative european politics 13, 8-28. https://doi.org/10.1057/cep.2014.39 ec european commision (2019). more efficient decision-making in social policy: identification of areas for an enhanced move to qualified majority voting. com(2019) 186, 1-16. ec european commision (2010). europe 2020, a strategy for smart, sustainable and inclusive growth. ec european commision (2021). social economy action plan. ec european commision (2022a). review report on the implementation of the recovery and resilience facility. com(2022) 383, final 29 july 2027. ec european commision (2022b). commission welcomes member states' targets for a more social europe by 2030. press release 16 june 2022. ep european parliament (2017). online platforms: how to adapt regulatory framework to the digital age?, briefing by policy department a: economy and scientific policy, pp. 2-9. eu european union (2021). regulation 2021 /241 of the european parliament and of the council of 12 february 2021 establishing the recovery and resilience facility. fella, s. (2022). the conference on the future of europe: proposals and next steps. house of commons library research briefing number 9551, 1-8. hix, s. & høyland, b. (2011). the political system of the european union, 3rd ed., hampshire, uk and new york, ny: palgrave macmillan. kassim, h. & laffan, b.(2019). the juncker presidency: the 'political commission' in practice', jcms: journal of common market studies, 5(s1), 49-61. https://doi.org/10.1111/jcms.12941 liebfried, s. & pierson, p. (1991). the pospects for social europe. program for the study of germany and europe working paper series 1.4, 1-48. ma, y. (2023). socialising the european semester? the limited influence of danish social partners on eu policymaking. journal of contemporary european research, 19(1), 23-40. https://doi.org/ 10.30950/jcer.v19i1.1169 pochet, p. (2017). the european pillar of social rights in historical perspective. blog lse european politics and policy (europp), 14 november 2017. rainone, s. & pochet, p. (2022). the eu recovery strategy: a blueprint for a more social europe or a house of cards? etui research paper working paper 2022.18. http://dx.doi.org/10.2139/ssrn.42 70429 rawls, j. (1977). a theory of justice, original edition, (revised 1997), camb. ms: harvard university press. rittberger, b. (2022). democratic control and legitimacy in the evolving eu economic governance framework', study requested by the ep econ committee. sabato, s. & vanhercke, b.(2017). towards a european pillar of social rights: from a preliminary outline to a commission recommendation', ch.4, social policy in the european union: state of play. vanhercke, b., sabato, s. & bouget, d. (eds.),73-96. brussels: european trade union institute (etui), european social observatory (ose). scharpf, f.w. (1989). decision rules, decision styles and policy choices. journal of theoretical politics, 1(2), 149-176. https://doi.org/10.1177/0951692889001002003 scharpf, f.w. (2001). problem-solving capacity of the multilevel european polity. mpifg working paper 01/4, max planck institute for the study of societies. shrzypek, a. (2023). the progressive europe we want. progressive year book 2023. brussels: feps. streeck, w. (1995). neo-voluntarism: a new european social policy regime? european law journal, 1(1), 31-59. https://doi.org/10.1111/j.1468-0386.1995.tb00003.x zeitlin, j., & vanhercke, b. (2018). socializing the european semester: eu social and economic policy coordination in crisis and beyond. journal of european public policy, 25(2), 149-174. https://doi.org/10.1080/13501763.2017.1363269 https://doi.org/10.1177/1024258912448598 https://doi.org/10.1057/cep.2014.39 https://doi.org/10.1111/jcms.12941 https://doi.org/10.30950/jcer.v19i1.1169 https://doi.org/10.30950/jcer.v19i1.1169 http://dx.doi.org/10.2139/ssrn.4270429 http://dx.doi.org/10.2139/ssrn.4270429 https://doi.org/10.1177/0951692889001002003 https://doi.org/10.1111/j.1468-0386.1995.tb00003.x https://doi.org/10.1080/13501763.2017.1363269 1. introduction 2. methodology 3. defining social europe 4. substantiating social europe through crises 5. conclusion references european journal of government and economics 11(2), december 2022, 193-209 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 does real interest rate reduce income inequality in india? evidence from multivariate framework analysis. muhammed ashiq villanthenkodath a , mantu kumar mahalik b a school of social sciences and humanities, b.s. abdur rahman crescent institute of science and technology, tamil nadu, 600048, india b department of humanities and social sciences indian institute of technology kharagpur, west bengal-721302, india * corresponding author at: muhammedashiq@crescent.education abstract. this study empirically examines the impact of real interest rate on income inequality in india within a kuznets curve framework considering the role of economic growth, trade openness and technological innovation as the control variables. this study employs the ardl bounds test for validating the long-run relationship over the annual data period 1995 to 2019. the results reveal the long-run relationship between the series in india. the findings suggest that the initial increase in interest rate significantly reduces income inequality. but, in a later stage, a threshold exists for such an increased interest rate to revert the prior beneficial impact. this finding further shows that kuznets’ inverted u-shaped hypothesis is not valid for the relationship between income inequality and real interest rate in india. it shows that the real interest rate impedes income distribution in the long run. these findings are also found to be robust using fmols and dols estimators. we find that economic growth significantly reduces income inequality, whereas trade openness promotes it. surprisingly, technological innovation enhances income inequality, but this effect vanishes in the long-run. however, these findings suggest that policymakers in india should not ignore the impeding role of real interest rates while aiming at achieving effective income distribution between haves and have-nots in the long run. keywords. income inequality; interest rates, economic growth; india. jel codes. d63; e49; o40; o53. doi. https://doi.org/10.17979/ejge.2022.11.2.8652 1. introduction the increase in income inequality in recent years has drawn considerable attention from academic and policy experts on the relationship between economic growth and income inequality in both developed and developing countries. economic growth and income inequality, being endogenous outcomes of the economic system, are subject to common influences, with respect to both structural changes and macroeconomic policies. structural changes such as improvement in technology facilitate economic development, which is the underlying assumption of kuznet’s curve, and may result in economic inequality. however, macroeconomic policies, particularly fiscal policies are considered an important instrument for achieving goals in terms of equity and efficiency (musgrave, 1959). https://creativecommons.org/licenses/by-nc/4.0/ mailto:muhammedashiq@gmail.com https://doi.org/10.17979/ejge.2022.11.2.8652 villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 194 this paper focuses on the expenditure side of the budget and emphasizes the role of social spending in explaining the dynamics between growth and inequality1. government spending in the social sector received an impetus with the emergence of endogenous growth theory (lucas, 1988; romer, 1994), which largely focused on enhancing human capital development. such policies facilitate the process of innovation, research, knowledge creation, and information dissemination and reduce vulnerability to external shocks (gebregziabher & niño-zarazúa, 2014). thus, the government expenditure in the social sector is found to have a positive impact on longterm economic growth and development (benhabib & spiegel, 1994). in addition, such policies play an important role in poverty and inequality reduction (fiszbein et al., 2014). indeed, the millennium development goals that bring poverty reduction, equity, and risk management to the forefront of debates further drew the attention of policymakers and development agencies toward establishing a strong social welfare system, particularly for the poorest and most vulnerable section of the society (baldacci et al., 2008). despite its relevance, few studies have attempted to explore the relationship between income inequality, economic growth, and government spending in the social sector2. most of the empirical studies have analyzed either the impact of government spending in the social sector on economic growth (antonia afonso & alegre, 2011; antónio afonso & furceri, 2010; folster & henrekson, 2001) or its role in poverty and income inequality reduction (cubero & vladkova, 2010; foster, 2012; ospina, 2010; rudra, 2004). further, these studies have not considered the role of gross income inequality in determining the impact of government spending on social spending and how such an impact influences the relationship between economic growth and net income inequality. while gross income inequality is pre-tax and government transfers’ income inequality, net income inequality is post-tax and government transfers’ income inequality. thus, our aim was to study the role of social spending in determining the relationship between economic growth and inequality. first, we analyzed the impact of gross income inequality and other political and economic factors on social spending. second, we examined which categories of social spending (education, health, and social protection) are effective in reducing income inequality and the effects of these policies on economic growth. the remainder of this paper is structured as follows: section 2 provides a brief summary of our literature review, which deals with income inequality, economic growth, and government spending in the social sector. section 3 describes the database and empirical methodology. section 4 delineates the findings and results. section 5 gives the discussion and policy implications. section 6 provides the concluding remarks. 1 in this study, social spending refers to government expenditure on education, health, and social protection. 2 arjona et al. (2003) examined the income distribution and social expenditure effects on economic growth. villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 195 2. literature review the literature identifies different routes through which interest rates can influence income inequality in any country. firstly, the literature evidenced the unequal distribution of debt-toincome and debt-to-equity ratios for the households that belong to the different income distribution strata (berisha et al., 2020; martín-legendre et al., 2020). similarly, a study by saez (2017) for the united states shows that the bottom 90% of households have had no savings for the last three decades. therefore, it can be concluded that the top strata consisting of the few people in the income distribution typically have more savings than debt. in contrast, the lower strata (comprised of the majority population) of the income distribution have more debt than savings. hence, an increase in interest rate creates a harmful effect on those who are in the lower-income strata, but it gives an advantage to the higher-income strata. in other words, increasing the benefits of interest rates will accrue to those who have more savings than debt compared to people with more debt than savings, thereby fueling income inequality. secondly, monetary policy contractions (increase in interest rate) probably create a loss of jobs for the lower-income households since they are the employees of the firms. the rise in interest rate affects the employees by reducing the fresh investment of firms in new employment generation projects, which, in turn, reduces the equilibrium rate of employment and, thereby, the unemployment level increases and falls in the income of the employees (phelps, 1994). similarly, investment at the firm level decreases due to the rising cost of capital (interest rate), which, in turn, leads to lower capital accumulation. hence, the demand for labour reduces due to the given ratio of employment to capital, i.e., no new employment opportunity (blanchard, 1999). in brief, the loss is happening for the labour class but not for the capital owners who are in the upper strata of income. therefore, there is a possibility of the existence of a gap between the rich and poor in the economy. hence, it is called the monetary policy impact due to earnings heterogeneity (coibion et al., 2017). thirdly, a low-interest rate leads to increased income inequality, either due to boosting capital gains or asset prices shooting up (auclert, 2019). this peculiar property of monetary policy is called the financial segmentation channel effect. the financial segmentation channel effect starts working when the monetary policy is expansionary in nature. recently, an empirical analysis was conducted by berisha et al. (2020), aiming to unveil the nexus between income inequality and other macroeconomic variables for the brics economies, in which they observed a positive impact of the real interest rate on income inequality. similarly, husain et al. (2020) explored the association between interest rates and income inequality in indonesia. the results evolved from the dynamic ordinary least square (dols), and fully-modified ols (fmols) show the income inequality exacerbation role of interest rate after attaining a threshold level. from an empirical perspective, the study by saiki & frost (2014) shows the role of unconventional monetary policy (i.e. repairing the transmission mechanisms of monetary policy and financial markets stability) on inequality in japan. the evidence shows the widening inequality role of unconventional monetary policy due to increased asset prices than the economic villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 196 fundamentals, i.e., employment and wage during severe financial and economic stress. the upper strata of society are the holders of the asset, which, in turn, enhances their income over and above the lower strata. a similar conclusion has been reached by bivens (2015) for the united states, mumtaz & theophilopoulou (2017) for the united kingdom, o’farrell & rawdanowicz (2017) for advanced economies, and furceri et al. (2018) for 32 emerging market and advanced countries while assessing the role of monetary policy on inequality. finally, economic growth creates either a positive or negative impact on income inequality. the popular kuznets hypothesis states that in the early stage of economic progress, income inequality also rises with economic growth, whereas income inequality decreases at the later stages of economic growth due to the trickle-down effect (kuznets, 1955). the kuznets’ inverted u-shaped hypothesis pertaining to the link between economic growth and income distribution has been confirmed in certain empirical studies, which include studies carried out by ahluwalia (1976) for a sample of 60 countries, and eusufzai (1997) for 54 countries. similarly, the studies of bahmani-oskooee & gelan (2008), jha (1996), and minami (1998) confirmed the hypothesis of the kuznets curve. in contrast, certain studies also fail to establish the kuznets hypothesis while validating it either in time series or cross-country level (anand & kanbur, 1993; angeles, 2010; papanek & kyn, 1986). besides, it is observed that technology and innovation exhibit a significant role in determining inequality (aghion et al., 2018). similarly, the impact of globalization and trade openness on income inequality determination is undeniable (jaumotte et al., 2013). therefore, it is vital to control these factors while empirically modelling the income inequality function for india within a time series framework. however, the literature survey reveals that there is theoretical construction related to the nexus between income inequality and interest rate in the literature. however, there is a need for empirical validation in the country-specific context for alleviating income inequality since interest rates play a pivotal role in determining income inequality through the accumulation of capital in the hands of a few. besides this, we also found that there is no clear consensus among the studies assessing the impact of economic growth, technology, and trade on income inequality in india. hence, it is worth conducting an empirical analysis of the impact of real interest rates on income inequality while considering economic growth, trade openness and technology as a control variable since india suffers from a skewed income distribution. 3. model construction, data, and estimation strategy this study specified the model to evaluate the nexus between income inequality, real interest rate, economic growth, trade openness and technology by following berisha et al. (2020) and husain et al. (2020). 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡 = 𝛾𝛾0 + 𝛾𝛾1𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡 + 𝛾𝛾2𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡2 + 𝛾𝛾3𝐺𝐺𝐺𝐺𝐺𝐺𝑡𝑡 + 𝛾𝛾4𝑂𝑂𝐺𝐺𝐼𝐼𝐼𝐼𝑡𝑡 + 𝛾𝛾5𝑇𝑇𝐼𝐼𝑇𝑇𝑇𝑇 𝑡𝑡 + 𝜀𝜀𝑡𝑡 [1] villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 197 where, 𝜀𝜀𝑡𝑡 is the error term, 𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼 is the income inequality, which is measured by the gini coefficient index. the terms 𝑖𝑖. 𝑒𝑒.𝑅𝑅𝐼𝐼𝑅𝑅 and 𝑅𝑅𝐼𝐼𝑅𝑅2 stand for the real interest rate and its square term to capture the kuznets’ effect. further, gdp denotes the gross domestic product (gdp), which is taken as a proxy for measuring real economic growth. the notations, i.e. 𝑂𝑂𝐺𝐺𝐼𝐼𝐼𝐼 and 𝑇𝑇𝐼𝐼𝑇𝑇𝑇𝑇, stand for trade openness (i.e. globalization) and technological change, respectively. symbol 𝛾𝛾0 is the constant term, 𝛾𝛾1, 𝛾𝛾2 , 𝛾𝛾3, 𝛾𝛾4 and 𝛾𝛾5 are also the coefficients of the explanatory variables. an annual time series data from 1978-2019 has been collected for estimating the specified model. the period of inquiry was selected based on data availability for the indian case. however, all the obtained data has been converted to the natural log for estimation to avoid the heterogeneity by following (ansari et al., 2022; pal et al., 2022; shameem p et al., 2022; villanthenkodath, ansari, et al., 2022; villanthenkodath, mahalik, et al., 2022; villanthenkodath & mahalik, 2022; villanthenkodath & mohammed, 2022). the data obtained from different sources are reported in table 1 with their detailed definitions. estimation of the specified model in equation 1 was conducted using the autoregressive distributed lag (ardl) model of cointegration introduced by pesaran et al. (2001) and pesaran & shin (1999). many factors guide the choice of the ardl bounds testing approach for the empirical analysis. first, it corrects the endogeneity of independent variables by selecting the appropriate lag while simulating the model.3 thus, the application of ardl makes us deal with such an endogeneity issue in our modeling framework. second, the model can be used even for a small sample size (i.e. 30-80 observations). table1. definition of data and their sources. variables definitions data source ineq the data on post-tax/transfer income inequality called net gini coefficient swiid rir the real interest rate is the lending interest rate adjusted for inflation as measured by the gdp deflator. the terms and conditions attached to lending rates differ by country, however, limiting their comparability. we also take quadratic term. wdi gdp gdp per capita is gross domestic product divided by midyear population. gdp is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. it is calculated without making deductions for depreciation of fabricated assets or depletion and degradation of natural resources. data are in constant 2010 u.s. dollars. wdi open trade is the sum of exports and imports of goods and services measured as a share of gross domestic product. wdi tech total patent applications (direct and pct national phase entries) wipo note: swiid is the standardized world income inequality database (https://dataverse.harvard.edu/dataset), wid stands for world inequality database (https://wid.world/), wipo is world intellectual property organization (https://www3.wipo.int/ipstats) and wdi is the world development indicators (https://databank.worldbank.org/). 3 the endogendiety problem arises in the literature when causality runs from income inequality to economic growth and vice-versa (alesina and rodrik, 1994; persson and tabellini, 1994; piketty and saez, 2006). we are thankful to one of the reviewers for raising this issue at the revision stage. https://wid.world/ https://www3.wipo.int/ipstats https://databank.worldbank.org/ villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 198 third, it is possible to use ardl bound testing approach when the set of explanatory variables are either integrated of the order one i.e. i (1) or having integration of the mixed orders i.e. i (0) and i (1). hence, there are no restrictions based on the nature of the data. therefore, the ardl model is superior to conventional cointegration techniques. the long-run relationship among the variables is estimated by employing equation 2, which is a corollary with the model specified in equation 1. ∆𝐿𝐿𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡 = 𝜆𝜆0 + �𝜆𝜆1𝑖𝑖∆𝐿𝐿𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + �𝜆𝜆2𝑖𝑖∆𝐿𝐿𝐼𝐼𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + �𝜆𝜆3𝑖𝑖∆𝐿𝐿𝐼𝐼𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡−𝑖𝑖2 𝑝𝑝 𝑖𝑖=1 + �𝜆𝜆4𝑖𝑖∆𝐿𝐿𝐼𝐼𝐺𝐺𝐺𝐺𝐺𝐺𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + �𝜆𝜆5𝑖𝑖∆𝐿𝐿𝐼𝐼𝑂𝑂𝐺𝐺𝐼𝐼𝐼𝐼𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + �𝜆𝜆6𝑖𝑖∆𝐿𝐿𝐼𝐼𝑇𝑇𝐼𝐼𝑇𝑇𝑇𝑇𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + 𝜑𝜑7𝐿𝐿𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡−1 + 𝜑𝜑8𝐿𝐿𝐼𝐼𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡−1 + 𝜑𝜑9𝐿𝐿𝐼𝐼𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡−12 + 𝜑𝜑10𝐿𝐿𝐼𝐼𝐺𝐺𝐺𝐺𝐺𝐺𝑡𝑡−1 + 𝜑𝜑11𝐿𝐿𝐼𝐼𝑂𝑂𝐺𝐺𝐼𝐼𝐼𝐼𝑡𝑡−1 + 𝜑𝜑12𝐿𝐿𝐼𝐼𝑇𝑇𝐼𝐼𝑇𝑇𝑇𝑇𝑡𝑡−1 + 𝜀𝜀𝑡𝑡 [2] after obtaining and extracting the cointegration relation between the study variables and longterm coefficients, there is a need to estimate the short-run dynamics. hence, the corresponding error correction model (ecm) equation has been estimated to obtain the short-run coefficients by using equation 3. ∆𝐿𝐿𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡 = 𝜆𝜆0 + ∑ 𝜆𝜆1𝑖𝑖∆𝐿𝐿𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + ∑ 𝜆𝜆2𝑖𝑖∆𝐿𝐿𝐼𝐼𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + ∑ 𝜆𝜆3𝑖𝑖∆𝐿𝐿𝐼𝐼𝑅𝑅𝐼𝐼𝑅𝑅𝑡𝑡−𝑖𝑖2𝑝𝑝 𝑖𝑖=1 + ∑ 𝜆𝜆4𝑖𝑖∆𝐿𝐿𝐼𝐼𝐺𝐺𝐺𝐺𝐺𝐺𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + +∑ 𝜆𝜆5𝑖𝑖∆𝐿𝐿𝐼𝐼𝑂𝑂𝐺𝐺𝐼𝐼𝐼𝐼𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + ∑ 𝜆𝜆6𝑖𝑖∆𝐿𝐿𝐼𝐼𝑇𝑇𝐼𝐼𝑇𝑇𝑇𝑇𝑡𝑡−𝑖𝑖 𝑝𝑝 𝑖𝑖=1 + 𝛿𝛿𝐼𝐼𝑇𝑇𝑇𝑇𝑡𝑡−1 + 𝜀𝜀𝑡𝑡 [3] in equations 2 and 3, the first difference of the concerned variables is denoted by using ∆. the optimal lag length of variables is represented by employing 𝑝𝑝, while 𝜆𝜆0 stands for the constant term. the error correction term (𝐼𝐼𝑇𝑇𝑇𝑇𝑡𝑡−1) shows the adjustment coefficient (𝛿𝛿) where the correction of short-run disequilibrium can be judged in order to own the stable long-run relationship between the series. the second and first segments of equations represent the long-run relationship and short-run dynamics, respectively. similarly, the null hypothesis stating no cointegration among the study variables can be expressed in the form of 𝑇𝑇0:𝜑𝜑7 = 𝜑𝜑8 = 𝜑𝜑9 = 𝜑𝜑10 = 𝜑𝜑11 = 𝜑𝜑12 = 0. in contrast, the alternative hypothesis is also specified as 𝑇𝑇1:𝜑𝜑7 ≠ 𝜑𝜑8 ≠ 𝜑𝜑9 ≠ 𝜑𝜑10 ≠ 𝜑𝜑11 ≠ 𝜑𝜑12 ≠ 0. the stated hypothesis is tested using the joint f-statistics with the corresponding critical values suggested by narayan (2005) for a sample size. the null hypothesis can be rejected when the computed f-statistic is above the upper limit of critical values, whereas the acceptance of the null hypothesis occurs when the f-statistic value is below the lower limit of critical values. however, the fcalculated value falls in between upper and lower bound critical values, and then there is no conclusive evidence of cointegration in the specified income inequality model. further, the overall long-run results of the employing ardl model were cross-checked by using dols (stock and watson, 1993) and fmols (pedroni, 2001) techniques. although the villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 199 interconnection between variables can be extracted by using various econometric approaches, the asymptotic coherence of the results can be available from fmols and dols techniques. precisely, the fmols can handle the issue of serial /autocorrelation, endogeneity, and multicollinearity by undertaking a non-parametric approach. however, in dols, the same issue can be handled by using a parametric approach. therefore, these techniques are superior to ardl model. 4. empirical results and discussion of findings figure 1 shows the trend of the variables over the sample period. it allows us to understand the long-run relationship between the series. the characteristics of the variables are also reported in table 2. towards this end, unit root tests for the variables are analyzed initially, and the same is reported in table 3. 0 10 20 30 40 50 60 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 ineq 0 2 4 6 8 10 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 rir 400 800 1,200 1,600 2,000 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 gdp 20 30 40 50 60 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 open 0 10,000 20,000 30,000 40,000 50,000 60,000 1996 1998 2000 2002 2004 2006 2008 2010 2012 2014 2016 2018 tech figure 1. trend plot of study variables. villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 200 table 2. descriptive statistics. descriptive statistics ineq rir gdp open tech mean 32.955 5.612 1143.415 38.746 28489.440 median 34.200 5.682 1075.994 40.743 34287.000 maximum 52.277 9.191 1972.758 55.794 53627.000 minimum 8.833 1.318 618.368 21.929 4826.000 std. dev. 14.293 2.124 422.105 11.378 16671.550 skewness -0.359 -0.30 0.555 -0.124 -0.119 kurtosis 1.814 2.329 2.096 1.691 1.390 observations 25 25 25 25 25 the outcomes of the augmented dickey-fuller (adf) and phillips-peron (pp) test proposed by dickey & fuller (1979) and phillips & perron (1988) shown in table 3 indicate that all the variables are stationary at the first difference. however, the conventional unit root tests are unable to capture the possible structural break in the series. hence, this study employs the zivot & andrews (1992) structural break unit root test to check the integration order of the series amid a single structural break. this result is also in line with the conventional unit root with an exception for technology. therefore, additional experiments can be conducted in the ardl framework since the series consists of variables having a maximum of the first order of integration i.e. i(1). for estimating the ardl model, the selection of appropriate lag is necessary. hence, this study chooses the hannan-quinn criterion (hq) criteria for selecting the lag length as it produces more desirable outcomes in comparison to other criteria. the ardl bound test result is presented in table 4. since the estimated value of f-statistics is higher than the upper bound critical value at a 1% level of statistical significance, there is evidence of the long-term association among the study variables. it means that the series moves to long-run equilibrium amid the presence of these variables. table 3. unit root tests. dickey-fuller test pp test za level δ level δ level break δ break lnineq 3.291 -8.983* 1.159 -8.983* -0.261 2001 -9.325* 2002 rir -2.145 -5.456* -2.232 -5.449 -2.748 2003 -5.740 2014 lngdp 1.123 -3.992* 1.520 -3.972* -0.818 2003 -5.259* 2002 lnopen -1.562 -3.918* -1.546 -3.985* -2.828 2001 -5.348 2013 lntech -1.181 -4.763* -1.174 -4.786* -6.216* 2010 -7.813 2007 note. * and *** : significant at 1% and 10% levels. δ shows the first difference. ln stands for the natural log. villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 201 table 4. ardl cointegration test results. bounds testing approach to cointegration diagnostic tests estimated model lag length f-stat χ2 normal χ2 serial χ2 arch reset lnineq=f(rir,rir2,lngdp, lnopen, lntech) (3, 2, 2, 2, 1, 2) 7.168* 0.623 [0.732] 0.384 [0.722] 1.014 [0.327] 0.085 [0.790] note: * indicates the 1% level of statistical significance. the value inside [ . ] shows the probability value. n stands for the number of observation, while k is included explanatory variables in the model. further, the ardl-based long-and short-run coefficients of the estimated model are presented in table 5 (panels i & ii). the long-run results shown in panel-i of table 5 indicate that the real interest rate exhibits a downward trend in income inequality at the earlier stage. however, after attaining a threshold level, the increase in interest rate leads to an upward trend in income inequality. it means that there is evidence of the u-shaped relationship between income inequality and real interest rate in india. it shows an adverse impact of higher interest rates on income inequality in the long run. the probable interpretation for rising income inequality due to the rise in real interest rate, in the long run, has many folds. first, the increase in interest rate benefits the households who belong to the higher income strata since they have higher savings than the debt service. hence, they acquire higher returns on savings, thereby further increasing their income level. second, the households with higher debt (bottom of income strata) will keep paying more interest payments to the banks, and eventually, their surplus over consumption (i.e. savings) could be reduced in the long run. it can create an income disparity between the rich and the poor. third, monetary contraction (i.e. increasing interest rate) makes banking loans expensive for business firms. as a result, business firms reduce investment which also creates unemployment among lower-income households. even if employment is created for lower income strata of the people, they are paid lower wages. it shows that increasing interest does not help much to poor people. it is a situation of income disparity between the rich and poor due to rising interest rates (carruth et al., 1998; phelps, 1994). these findings are in line with the study of husain et al. (2020). however, an inverted u-shaped relationship between income inequality and real interest rate is found in the short run (panel ii). the result also shows that economic growth significantly reduces income inequality in the long run. it indicates that income inequality reduces as economic growth improves. alternatively, it is a sign of inclusive growth, creating employment opportunities for poor people. as a result, poor households can have a regular income not only to mitigate their basic requirements but also to increase their savings capacity. eventually, the living standards of poor households tend to be higher with further increased economic growth. our finding is not in line with (adelman & robinson, 1989; bahmani-oskooee & gelan, 2008; jha, 1996; minami, 1998). however, the impact of economic growth on income inequality is not statistically significant in the short run. it shows that economic growth does not improve income distribution in the short run. villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 202 this may be because an expansion of economic activities creating employment opportunities takes some time. if people are not employed, then it becomes difficult to increase their living standards and income capacity. moreover, trade openness enhances income inequality both in the short and long runs. the possible cause for this could be that the country may be imposing import duties on the goods and services which are more highly demanded by poor people than the rich, which, in turn, causes the increase in price and widen the income gap. besides, technological innovation increases income inequality in the long run, but it is not significant. moreover, it becomes effective in reducing income inequality in the short run. table 5 (panel ii) also portrays a significant and negative speed of adjustment coefficient from short to long-run equilibrium. it means that about 80% of disequilibrium is corrected each year. the last segment in panel iii of table 5 delineates the necessary post-estimation tests for checking the model adequacy. the included variables explain 68% of variations in inequality, the rest variations are captured in the error term. a significant f-statistics value indicates the overall goodness fit of the model. for the ardl model validity and stability, the study used the cumulative sum (cusum) and cumulative sum of squares (cusumsq) tests proposed by brown et al. (1975). figure 2 shows the plots of cusum and cusumsq, which fall within the significance of 5% level by confirming the estimation of the ardl model is stable. hence, the specified model and estimated parameters in this analysis are reliable and efficient. table 5. estimated longand short-run coefficients using the ardl approach. panel-i: long-run analysis dependent variable: lnineq regressors coefficient se t-ratio pvalue rir -0.542** 0.162 -3.351 0.029 rir2 0.238** 0.070 3.382 0.028 lngdp -2.124* 0.177 -12.011 0.000 lnopen 0.746* 0.125 5.968 0.004 lntech 0.155 0.131 1.180 0.304 c 14.161* 0.387 36.552 0.000 panel-ii: short-run analysis δ rir 0.868*** 0.401 2.163 0.097 δ rir2 -0.417** 0.147 -2.839 0.047 δ lngdp 0.415 2.504 0.166 0.876 δ lnopen 1.604*** 0.623 2.576 0.062 δ lntech -0.576*** 0.248 -2.324 0.081 cointeq(-1)* -0.801*** 0.339 -11.200 0.000 panel-iii: short-run diagnostic tests r-squared 0.96 adjusted r-squared 0.91 se of regression 0.07 sd of dependent variable 0.558 mean of dependent variable 3.292 f-statistic [prob.] 35.88 [0.001] note. *, *, ***: 1%, 5%, 10% levels of statistical significance, respectively. villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 203 -6 -4 -2 0 2 4 6 2016 2017 2018 2019 cusum 5% significance -0.4 0.0 0.4 0.8 1.2 1.6 2016 2017 2018 2019 cusum of squares 5% significance figure 2. cusum and cusumsq tests at 5% level of significance additionally, the robustness of the estimated overall long-run results of ardl was validated by using fmols and dols techniques. the reported outcomes in table 6 are also in harmony by exhibiting a significant u-shape relation of interest rate on income inequality. besides, a significant negative (positive) impact of economic growth (trade openness and technological innovation) on income inequality has been further affirmed. villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 204 table 6. dols and fmols results. panel-i: dols dependent variable: lnineq regressors coefficient se t-ratio pvalue rir -0.823* 0.236 -3.492 0.010 rir2 0.353* 0.094 3.755 0.007 lngdp -2.502* 0.161 -15.550 0.000 lnopen 0.359*** 0.190 1.890 0.101 lntech 0.534* 0.146 3.651 0.008 c 14.604* 0.351 41.572 0.000 r-squared 0.97 adjusted r-squared 0.93 panel-ii: fmols rir -0.139 0.189 -0.734 0.472 rir2 0.083 0.076 1.087 0.291 lngdp -2.102* 0.158 -13.325 0.000 lnopen 0.832* 0.159 5.221 0.000 lntech 0.065 0.109 0.598 0.557 c 14.366* 0.449 31.980 0.000 r-squared 0.93 adjusted r-squared 0.91 note. *, *, ***: 1%, 5%, 10% levels of statistical significance, respectively. 5. conclusion and policy implications this study investigates the impact of real interest rates on income distribution in india by considering the influence of economic growth, trade openness and technology. moreover, the kuznets’ hypothesis is also analyzed using the square term of real interest rate in the income inequality function. for the empirical analysis, this study used the ardl model for the data spanning from 1978 to 2019. the outcomes of the ardl show that real interest rate fosters income inequality after reaching a threshold level, especially in the long run. similarly, economic growth and trade openness also exacerbate income inequality in india, whereas technology is not effective in increasing it. the robustness of these findings is confirmed by using the dols and fmols techniques. we also observed a u-shaped connection between income quality and real interest rate. the outcome reveals that income inequality probably starts to spur if the real interest rate surpasses above 6.13% in india. based on the above findings, it can be argued that in case the monetary authority is increasing the interest rate, the households having more savings accrue the benefit, whereas households with more debt suffer a lot. it shows the consequence of such a move eventually creates income inequality in india. therefore, the policymakers in india need to be cautious while forming monetary policy since there is a trade-off between macroeconomic stabilization and income distribution if the rate of interest goes beyond a threshold level. we also find that economic growth reduces income inequality in india. it shows that rising economic growth reduces the income gap between rich and poor people in india. this is villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 205 because the trickle-down growth effect equally benefits rich and poor people. interestingly, the way the economic growth plan created in india is benefiting the labour force in the employment market. the poor people at the grass-root level of india also receive necessary benefits from the welfare scheme implementation. trade openness enhances income inequality both in the short and long runs. it suggests that policymakers in india need to provide attention to the price level of imported commodities. if oil is imported from oil-exporting countries with higher payments, then oil domestic oil suppliers will charge higher prices for the people who buy it. it may not be an issue for the rich people but it becomes expensive for the poor in india as it is a developing economy. finally, technological innovation enhances income inequality in the long run, but it is not effective. it implies that though technology usage by the business firms replaces the labour force in production activities in order to increase technical efficiency (i.e. output/technology), it does not become successful. this may be because producers integrate the labour force with technology in the production structure in order to minimize the cost of production, as labour is available at an affordable wage in india. as a result, employment level in the labour market is created, also improving income distribution rather than increasing income inequality. references africa tax and inequality report (2014). "africa rising? inequalities and the essential role of fair taxation". retrieved from https://www.taxjusticeafrica.net/resources/publications aghion, p., akcigit, u., bergeaud, a., blundell, r., & hémous, d. (2018). innovation and top income inequality. rev econ stud 86(1):1-45. https://doi.org/10.1093/restud/rdy027 aghion, p., caroli, e., & penalosa, c. g. (1999). inequality and economic growth: the perspective of the new growth theories. journal of economic literature, 37(4), 1615-1660. https://doi.org/10.1257/jel.37.4.1615 agnello, l., mallick, s. k., & sousa, r. m. (2012). financial reforms and income inequality. economics letters, 116(3), 583-587. https://doi.org/10.1016/j.econlet.2012.06.005 ahluwalia, m. s. (1976). inequality, poverty and development. journal of development economics, 3(4), 307-342. https://doi.org/10.1016/0304-3878(76)90027-4 alesina, a., & rodrik, d. (1994). distributive politics and economic growth. the quarterly journal of economics, 109(2), 465-490. https://doi.org/10.2307/2118470 anand, s., & kanbur, s. m. r. (1993). inequality and development a critique. journal of development economics, 41(1), 19-43. https://doi.org/10.1016/0304-3878(93)90035-l aneja, r., barkha, & banday, u. j. (2020). regional economic growth and inequality in india: a sector-wise decomposition analysis. arthaniti: journal of economic theory and practice, 0976747920910825. https://doi.org/10.1177/0976747920910825 https://www.taxjusticeafrica.net/resources/publications https://doi.org/10.1093/restud/rdy027 https://doi.org/10.1257/jel.37.4.1615 https://doi.org/10.1016/j.econlet.2012.06.005 https://doi.org/10.1016/0304-3878(76)90027-4 https://doi.org/10.2307/2118470 https://doi.org/10.1016/0304-3878(93)90035-l https://doi.org/10.1177/0976747920910825 villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 206 angeles, l. (2010). an alternative test of kuznets' hypothesis. the journal of economic inequality, 8(4), 463-473. https://doi.org/10.1007/s10888-009-9117-4 ansari, m. a., villanthenkodath, m. a., akram, v., & rath, b. n. (2022). the nexus between ecological footprint, economic growth, and energy poverty in sub-saharan africa: a technological threshold approach. environment, development and sustainability. https://doi.org/10.1007/s10668-022-02377-5 auclert, a. (2019). monetary policy and the redistribution channel. american economic review, 109(6), 2333-2367. https://doi.org/10.1257/aer.20160137 bahmani-oskooee, m., & gelan, a. (2008). kuznets inverted-u hypothesis revisited: a timeseries approach using us data. applied economics letters, 15(9), 677-681. https://doi.org/10.1080/13504850600749040 berisha, e., gupta, r., & meszaros, j. (2020). the impact of macroeconomic factors on income inequality: evidence from the brics. economic modelling, 91(c), 559-567. https://doi.org/10.1016/j.econmod.2019.12.007 bhuyan, b., sahoo, b. k., & suar, d. (2020). nutritional status, poverty, and relative deprivation among socio-economic and gender groups in india: is the growth inclusive? world development perspectives, 18, 100180. https://doi.org/10.1016/j.wdp.2020.100180 bivens, j. (2015). gauging the impact of the fed on inequality during the great recession. hutchins center on fiscal and monetary policy working paper no.12. blanchard, o. j. (1999). european unemployment: the role of shocks and institutions. cambridge ma. brown, j. c., pusey, p. n., goodwin, j. w., & ottewill, r. h. (1975). light scattering study of dynamic and time-averaged correlations in dispersions of charged particles. journal of physics a: mathematical and general, 8(5), 664. https://doi.org/10.1088/0305-4470/8/5/004 carruth, a. a., hooker, m. a., & oswald, a. j. (1998). unemployment equilibria and input prices: theory and evidence from the united states. in the warwick economics research paper series (twerps) (no. 496; the warwick economics research paper series (twerps)). university of warwick, department of economics. https://doi.org/10.1162/003465398557708 coibion, o., gorodnichenko, y., kueng, l., & silvia, j. (2017). innocent bystanders? monetary policy and inequality. journal of monetary economics, 88, 70-89. https://doi.org/10.1016/j.jmoneco.2017.05.005 datt, g., ravallion, m., & murgai, r. (2020). poverty and growth in india over six decades. american journal of agricultural economics, 102(1), 4-27. https://doi.org/10.1093/ajae/aaz043 demirguc-kunt, a., & levine, r. (2009). finance and inequality: theory and evidence (no. w15275). national bureau of economic research. https://doi.org/10.3386/w15275 dickey, d. a., & fuller, w. a. (1979). distribution of the estimators for autoregressive time series with a unit root. journal of the american statistical association, 74(366a), 427-431. https://doi.org/10.1080/01621459.1979.10482531 eusufzai, z. (1997). the kuznets hypothesis: an indirect test. economics letters, 54(1), 81-85. https://doi.org/10.1016/s0165-1765(96)00900-7 https://doi.org/10.1007/s10888-009-9117-4 https://doi.org/10.1007/s10668-022-02377-5 https://doi.org/10.1257/aer.20160137 https://doi.org/10.1080/13504850600749040 https://doi.org/10.1016/j.econmod.2019.12.007 https://doi.org/10.1016/j.wdp.2020.100180 https://doi.org/10.1088/0305-4470/8/5/004 https://doi.org/10.1162/003465398557708 https://doi.org/10.1016/j.jmoneco.2017.05.005 https://doi.org/10.3386/w15275 https://doi.org/10.1080/01621459.1979.10482531 https://doi.org/10.1016/s0165-1765(96)00900-7 villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 207 furceri, d., loungani, p., & zdzienicka, a. (2017). the effects of monetary policy shocks on inequality. journal of international money and finance, 85, 168-186. https://doi.org/10.1016/j.jimonfin.2017.11.004 galor, o., & moav, o. (2004). from physical to human capital accumulation: inequality and the process of development. review of economic studies, 71(4), 1001-1026. https://doi.org/10.1111/0034-6527.00312 husain, s., sohag, k., hasan, r., & shams, s. m. r. (2020). interest rate and income disparity: evidence from indonesia. strategic change, 29(6), 665-672. https://doi.org/10.1002/jsc.2381 jha, s. k. (1996). the kuznets curve: a reassessment. world development, 24(4), 773-780. https://doi.org/10.1016/0305-750x(95)00165-9 jones ci, kim j (2018) a schumpeterian model of top income inequality. j polit econ 126(5):1785-1826. https://doi.org/10.1086/699190 kar, s., & sakthivel, s. (2007). reforms and regional inequality in india. economic and political weekly, 42(47), 69-77. kuznets, s. (1955). economic growth and income inequality. american economic review, 45(1), 1-28. mallick, h., mahalik, m. k., & padhan, h. (2020). does globalization exacerbate income inequality in two largest emerging economies? the role of fdi and remittances inflows. international review of economics, 1-38. https://doi.org/10.1007/s12232-020-00350-0 martín-legendre, j. i., castellanos-garcía, p., & sánchez-santos, j. m. (2020). monetary policy and income distribution: empirical evidence for the eu-15. journal of world economy 55, 121-142. https://doi.org/10.33776/rem.v0i55.3820 minami, r. (1998). economic development and income distribution in japan: an assessment of the kuznets hypothesis. cambridge journal of economics, 22(1), 39-58. https://doi.org/10.1093/oxfordjournals.cje.a013704 mumtaz, h., & theophilopoulou, a. (2017). the impact of monetary policy on inequality in the uk. an empirical analysis. european economic review, 98, 410-423. https://doi.org/10.1016/j.euroecorev.2017.07.008 narayan, p. k. (2005). the saving and investment nexus for china: evidence from cointegration tests. applied economics, 37(17), 1979-1990. https://doi.org/10.1080/00036840500278103 o'farrell, r., & rawdanowicz, l. (2017). monetary policy and inequality: financial channels. international finance, 20(2), 174-188. https://doi.org/10.1111/infi.12108 pal, s., villanthenkodath, m. a., patel, g., & mahalik, m. k. (2022). the impact of remittance inflows on economic growth, unemployment and income inequality: an international evidence. international journal of economic policy studies, 16(1), 211–235. https://doi.org/10.1007/s42495-021-00074-1 papanek, g. f., & kyn, o. (1986). the effect on income distribution of development, the growth rate and economic strategy. journal of development economics, 23(1), 55-65. https://doi.org/10.1016/0304-3878(86)90079-9 https://doi.org/10.1016/j.jimonfin.2017.11.004 https://doi.org/10.1002/jsc.2381 https://doi.org/10.1016/0305-750x(95)00165-9 https://doi.org/10.1086/699190 https://doi.org/10.1007/s12232-020-00350-0 https://doi.org/10.33776/rem.v0i55.3820 https://doi.org/10.1093/oxfordjournals.cje.a013704 https://doi.org/10.1016/j.euroecorev.2017.07.008 https://doi.org/10.1080/00036840500278103 https://doi.org/10.1111/infi.12108 https://doi.org/10.1007/s42495-021-00074-1 https://doi.org/10.1016/0304-3878(86)90079-9 villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 208 pedroni, p. (2001). fully modified ols for heterogeneous cointegrated panels. in nonstationary panels, panel cointegration, and dynamic panels. emerald group publishing limited. perron, p. (1989). the great crash, the oil price shock, and the unit root hypothesis. econometrica, 57(6), 1361-1401. https://doi.org/10.2307/1913712 persson, t., & tabellini, g. (1994). ls lnequality harmful for growth. american economic review, 84(3), 600-621. pesaran, m. h., & shin, y. (1999). an autoregressive distributed-lag modelling approach to cointegration analysis. in s. strøm (ed.), econometrics and economic theory in the 20th century: the ragnar frisch centennial symposium (pp. 371-413). cambridge university press. https://doi.org/10.1017/ccol0521633230.011 pesaran, m. h., shin, y., & smith, r. j. (2001). bounds testing approaches to the analysis of level relationships. journal of applied econometrics, 16(3), 289-326. https://doi.org/10.1002/jae.616 phelps, e. s. (1994). structural slumps: the modern equilibrium theory of unemployment, interest, and assets. harvard university press. piketty, t., & saez, e. (2006). the evolution of top incomes: a historical and international perspective. american economic review, 96(2), 200-205. https://doi.org/10.1257/000282806777212116 saez, e. (2017). income and wealth inequality: evidence and policy implications. contemporary economic policy, 35(1), 7-25. https://doi.org/10.1111/coep.12210 saiki, a., & frost, j. (2014). does unconventional monetary policy affect inequality? evidence from japan. applied economics, 46(36), 4445-4454. https://doi.org/10.1080/00036846.2014.962229 serneels, p., & dercon, s. (2021). aspirations, poverty, and education. evidence from india. the journal of development studies, 57(1), 163-183. https://doi.org/10.1080/00220388.2020.1806242 shameem p, m., villanthenkodath, m. a., & chittedi, k. r. (2022). economic growth and sectoral level electricity consumption nexus in india: new evidence from combined cointegration and frequency domain causality approaches. international journal of sustainable energy, 0(0), 1– 18. https://doi.org/10.1080/14786451.2022.2095386 stock, j. h., & watson, m. w. (1993). a simple estimator of cointegrating vectors in higher order integrated systems. econometrica: journal of the econometric society, 783-820. https://doi.org/10.2307/2951763 villanthenkodath, m. a., & mahalik, m. k. (2022). did the covid-19 outbreak asymmetrically affect the spot electricity price in india? energy research letters, 3(2), 32615. villanthenkodath, m. a., & mohammed, s. p. (2022). examining the impact of electricity production on economic growth and environmental quality in japan: a disaggregated level analysis. environmental science and pollution research. https://doi.org/10.1007/s11356-02222125-6 https://doi.org/10.2307/1913712 https://doi.org/10.1017/ccol0521633230.011 https://doi.org/10.1002/jae.616 https://doi.org/10.1257/000282806777212116 https://doi.org/10.1111/coep.12210 https://doi.org/10.1080/00036846.2014.962229 https://doi.org/10.1080/00220388.2020.1806242 https://doi.org/10.1080/14786451.2022.2095386 https://doi.org/10.2307/2951763 https://doi.org/10.1007/s11356-022-22125-6 https://doi.org/10.1007/s11356-022-22125-6 villanthenkodath & mahalik / european journal of government and economics 11(2), december 2022, 193-209 209 villanthenkodath, m. a., ansari, m. a., kumar, p., & raju, y. n. (2022). effect of information and communication technology on the environmental sustainability: an empirical assessment for south africa. telematics and informatics reports, 7, 100013. https://doi.org/10.1016/j.teler.2022.100013 villanthenkodath, m. a., mahalik, m. k., & patel, g. (2022). effects of foreign aid and energy aid inflows on renewable and non-renewable electricity production in brics countries. environmental science and pollution research. https://doi.org/10.1007/s11356-022-22730-5 https://doi.org/10.1016/j.teler.2022.100013 https://doi.org/10.1007/s11356-022-22730-5 abstract 1. introduction 2. literature review 3. model construction, data, and estimation strategy 4. empirical results and discussion of findings 5. conclusion and policy implications references brain drain and economic growth: evidence of productivity growth european journal of government and economics 10(2), december 2021, 128-145 european journal of government and economics issn: 2254-7088 brain drain and economic growth: evidence of productivity growth from brain circulation woosik yua* a graduate school of international studies, seoul national university, seoul, republic of korea * corresponding author at: woosik@snu.ac.kr abstract. this paper analyzes the effect of the so-called ‘brain drain’ on economic growth through the channel of growth in total factor productivity. we analyze panel data that measure the severity of brain drain, which are from imd and the u.s. national science foundation. our analysis shows that middle-income countries have more brain drain compared to the group of high-income countries. also, emerging economies that grow fast tend to experience more brain drain. our results from fixed effects regression models show that that brain drain has a significant and positive impact on economic growth, and the main channel is productivity growth. this can be considered as evidence of the positive effects of ‘brain circulation’, which is one of the brain drain phenomena that settlement of the talents in advanced countries can eventually help improve the productivity of home country by the sharing of advanced technologies and skills around them with colleagues in motherland. therefore, a strategy of utilizing overseas resident talents should also be considered, alongside the brain-attraction policy. keywords. brain drain, brain circulation, economic growth, total factor productivity jel codes. o11; o15; o47; f43 doi. https://doi.org/10.17979/ejge.2021.10.1.7831 1. introduction as globalization deepens, issues with brain drain are often discussed worldwide. brain drain refers to a state or a phenomenon in which a country cannot utilize its effective labor of human resources or human capital. this phenomenon is largely divided into internal brain drain and external brain drain. internal brain drain refers to the phenomenon in which domestic personnel with specialized skills are engaged in jobs unrelated to their majors, and external brain drain refers to the state in which competent talents of a country reside in other foreign countries, and their productivity contributes to the production of other countries. the brain drain to be addressed in this study is the second concept, the external brain drain. whenever the ‘brain drain index’ is announced at the international institute for management development (imd) in switzerland every year, developing countries have continuously raised serious questions about the negative impact of brain drain on economic growth as well as national productivity and competitiveness. the argument is that the more serious the outflow of competent human resources to foreign countries, the less experts and research personnel in the home country, and this results in the weakening of the country’s research and technology competitiveness and reduction of innovation, which in turn adversely affects economic growth. mailto:woosik@snu.ac.kr https://doi.org/10.17979/ejge.2021.10.1.7831 woosik yu / european journal of government and economics 10(2), december 2021, 128-145 129 on the contrary, some argue that brain drain may have a positive effect on the national economy and competitiveness. in south korea during the 1970s, the ‘brain drain’ had even been regarded as an act of ‘traitoring’ and the students residing abroad after the completion of their studies were harshly blamed. however, in recent days, as described in saxenian (2005) and teney (2021), the practical concept of ‘brain circulation’ came up in which the students or workers who studied and were employed abroad do not return to their home country right away, but return after working abroad and acquiring advanced learning and skills, or contributing to the motherland through various human networks built up while living abroad. in this regard, brain drain is considered a positive phenomenon as the first step for the brain circulation. as such, there are not many studies that quantitatively analyze the direct relationship between brain drain and economic growth using growth regression models, although brain drain can be a major factor in economic growth, either through the leakage of human capital or through the technological progress from brain circulation. therefore, in this study, we intend to supplement the prior study by conducting an empirical analysis of the effects of external brain drain on economic growth using growth regression models, while looking at the time series trend for brain drain situations of several selected countries. 2. literature review studies on the negative effects of the flight of skilled workers on the national economy and competitiveness are conducted by grubel and scott (1966), bhagwati and hamada (1974), and mcculloch and yellen (1977). in addition, docquier and rapoport (2008) presented empirical analysis that since the 1990s, the migration of skilled workers (brain drain) has increased by a greater margin than that of ordinary workers, and explained the factors behind the brain drain. however, studies by mountford (1997), stark, helmenstein, and prskawetz (1997), beine, docquier and rapoport (2001), and cinar and docquier (2004) suggest that the migration of the talents to advanced technology-holding countries could have a positive impact on economic growth and competitiveness in the home country. according to beine et al. (2001), immigration of talents can have both a ‘brain effect’ when returning home after acquiring advanced technology and a ‘drain effect’ that occurs when not returning home, and when brain effect is large enough under the open economy model, brain drain can have a positive effect on economic growth. stark et al. (1997) also presented a research that said the dissemination of skills and knowledge gained in advanced technology-holding countries could help improve the home country’s technological prowess when migrant talent returns. cinar et al. (2004) also argued that the home remittance of competent people who moved to developed countries had a positive effect on economic growth of the home country, as is often seen in the cases of immigration of workers from developing countries. moreover, vidal (1998) explored brain drain’s positive effect on human capital formation, and saxenian (2005) described the process of brain circulation from brain drain with lowered transaction costs associated with digitization, and how it ultimately helps the development of woosik yu / european journal of government and economics 10(2), december 2021, 128-145 130 technology in home country, using the data of chineseand indian-born engineers in the united states (u.s.). more recently, kang, lim, and hwang (2013) provided empirical evidence of brain drain’s indirect but positive influence on national competitiveness through r&d investment and accumulated human capital. also, lodigiani, marchiori, and shen (2015) showed that brain drain could have both positive and negative effects on gdp per capita through the technology diffusion at origin from the high-skilled diaspora, depending on the region’s relative distance to the technological frontier. amid this widespread agreement and objection to the severity of brain drain, this study seeks to infer the effects of brain drain on economic growth and technological growth (a possible channel to economic growth) of home country, using the data that proxies the degree of brain drain. 3. motivation we were intrigued by the concept of ‘brain circulation’, one of the positive results from brain drain. as described in the above sections, talents who studied and were employed abroad can benefit their motherland through their acquisition of frontier technology and skills and sharing of their knowledge to their colleagues living in motherland, or co-workings with colleagues in home country. with an awareness that brain drain can adversely affect society in the sense of human capital flight, we hypothesized that brain drain could also have a positive effect on economic growth, as many other researchers claim. however, our main argument is more comprehensive. we hypothesize that brain drain helps achieve economic growth through productivity growth. brain drain can induce brain circulation, and brain circulation helps home country’s technological progress. technological progress ultimately improves productivity of production, and results in growth of income. particularly, the total factor productivity (tfp) is predicted to be mostly affected by the brain drain via brain circulation and technology diffusion. tfp refers to the productivity of an economy’s composite inputs that are used to produce national output, and in neoclassical growth theory, economic growth rate converges to the growth rate of tfp in the long run (steady state). thus, tfp can be a proxy for a country’s productive efficiency and technological level. tfp growth is one of the key components and channels for the economic growth in a simple growth model of 𝑌𝑌 = 𝑇𝑇𝑇𝑇 where 𝑌𝑌 is national income, 𝑇𝑇 is total factor productivity, 𝑇𝑇 is composite input, and 𝑁𝑁 is population. from the model, the income growth can be derived: 𝑌𝑌 𝑁𝑁 = 𝑇𝑇 ∙ 𝑋𝑋 𝑁𝑁 [1] then, 𝑌𝑌 𝑁𝑁 = 𝑦𝑦 is national income per capita and 𝑋𝑋 𝑁𝑁 = 𝑥𝑥 is composite input per capita. when converting to growth terms, national income per capita growth (𝑔𝑔𝑦𝑦) can be expressed as the sum woosik yu / european journal of government and economics 10(2), december 2021, 128-145 131 of growth of total factor productivity (𝑔𝑔𝑇𝑇 ) and growth of per-capita composite input (𝑔𝑔𝑥𝑥 ) as described in equation [2]: 𝑔𝑔𝑦𝑦 = 𝑔𝑔𝑇𝑇 + 𝑔𝑔𝑥𝑥 [2] this implies that there are two main channels to which economic growth can be achieved: increasing growth rate of tfp or growth of composite input per capita. as the key hypothesis of this paper is that brain drain may enhance technology and productivity of home country through technological diffusion, testing the effects of brain drain on tfp growth will be the focus of our empirical analysis, as a channel to achieve income growth. 4. empirical landscape before analyzing the data, we will look at the main explanatory variables of the paper along with the descriptive statistics (see table 1). in this study, two indicators are used as key variables that can proxy the degree of brain drain. the first data was established using the ‘brain drain index’ survey data from the world competitive executive opinion survey of the international institute for management development (imd) in switzerland. the criteria number in imd competitiveness yearbook is ‘3.2.21’ and its title is ‘brain drain’. the survey question is “brain drain ([of ]welleducated and skilled people) does not hinder competitiveness in your economy”1 with a scale from zero to ten, and is asked to influential entrepreneurs worldwide. recent years of data can be downloaded online at website of imd world competitiveness online2, and the past data can be found in imd world competitiveness yearbook series. as mentioned above, the index values of the original data are distributed from 0 to 10, and the more severe the brain drain is, the more negative impact it has on economic and management activities, and the closer it becomes to zero. in this study, the corresponding brain drain index values were subtracted from 10, so that the higher the brain drain is, the higher the values are. this newly defined imd brain drain index is referred to as the ‘brain drain index’ in this paper. the indicator covers data from 61 countries and provides a total of 22 years of time series values from 1995 to 2016. this data is important in the sense that it can give information on how the entrepreneurs feel about brain drain of the country, as the entrepreneurs are the front-line agents who put the up-to-date technologies into practical uses, and they are eager to employ the talented the most. 1 the imd world talent ranking methodology (2019) 2 https://worldcompetitiveness.imd.org/customsearch https://worldcompetitiveness.imd.org/customsearch woosik yu / european journal of government and economics 10(2), december 2021, 128-145 132 table 1. descriptive statistics of brain drain index and share of doctorate recipients staying in the u.s. sample size mean standard deviation minimum maximum no. of countries period brain drain index3 1,174 4.779 1.582 1.04 8.85 61 1995~2016 phd ratios staying in the u.s. 287 0.628 0.167 0.10 0.92 41 2010~2016 brain drain index 210 5.379 1.322 2.75 8.85 30 (common sample counties) 2010~2016 (common sample period) phd ratios staying in the u.s. 210 0.609 0.145 0.21 0.92 note. third and fourth rows are for common sample period (2010~2016) and sample country list (30 countries). source: imd world competitiveness executive opinion survey, us national science foundation table 2. rankings of imd brain drain index by country in 20164. rank country b.d. index rank country b.d. index rank country b.d. index 1 hungary 8.44 22 lithuania 5.97 43 canada 4.20 2 venezuela 8.44 23 argentina 5.90 44 israel 4.17 3 bulgaria 8.04 24 mongol 5.83 45 iceland 4.08 4 ukraine 8.00 25 jordan 5.78 46 chile 3.94 5 croatia 7.84 26 brazil 5.73 47 belgium 3.89 6 south africa 7.23 27 japan 5.58 48 indonesia 3.76 7 slovakia 7.13 28 spain 5.49 49 germany 3.64 8 kazakhstan 7.05 29 mexico 5.23 50 britain 3.58 9 slovenia 6.85 30 france 5.22 51 singapore 3.51 10 russia 6.75 31 italy 5.14 52 finland 3.46 11 greece 6.68 32 india 5.09 53 uae 3.38 12 portugal 6.61 33 ireland 4.93 54 luxembourg 3.35 13 latvia 6.58 34 rumania 4.55 55 netherlands 2.80 14 poland 6.58 35 turkey 4.49 56 hong kong 2.78 15 estonia 6.51 36 cech 4.46 57 denmark 2.75 16 south korea 6.40 37 thailand 4.41 58 sweden 2.70 17 taiwan 6.31 38 qatar 4.41 59 usa 2.67 18 philippines 6.23 39 austria 4.35 60 swiss 2.03 19 columbia 6.07 40 australia 4.31 61 norway 1.68 20 china 6.05 41 new zealand 4.30 21 peru 5.98 42 malaysia 4.28 note. source: imd world competitiveness executive opinion survey. 3 values were subtracted from 10. 4 the higher the value, the more serious the degree of brain drain is. woosik yu / european journal of government and economics 10(2), december 2021, 128-145 133 the second brain drain proxy data are from the u.s. national science foundation's ‘doctorate recipients from u.s. universities’ dataset, which can be downloaded from nsf website5. more specifically, the data we used are taken from the ‘share of doctorate recipients with temporary visas intending to stay in the u.s. after doctorate receipt’. in this paper, we will refer all of the doctorate degrees in this data as ‘doctor of philosophy (ph.d.)’, though we are aware that not all doctorate-level degrees are ph.d. degrees. thus, we will call this variable as ‘share of ph.d.’s staying in the u.s.’ for convenience. the variable takes a value of zero to one, and covers data of u.s. ph.d. recipients from a total of 41 countries over the world with the seven-year time series availability from 2010 to 2016. the u.s. universities’ ph.d. graduates, who are considered to be the frontiers of advanced technology, can be classified as the ‘top-class’ professionals in their fields, and therefore, we postulate that the higher the percentage of them remaining in the u.s. is, the higher the level of brain drain their home countries face. 4.1. imd brain drain index data first, table 2 of imd brain drain index rankings by country shows that hungary, venezuela, and bulgaria are ranked the first, second and the third as of 2016. not surprisingly, the upper-ranked countries are mostly middle-income or lower-middle-income countries, rather than the richest or poorest countries. as described in the table above, seven out of the top 10 countries, i.e., hungary, bulgaria, ukraine, croatia, slovakia, slovenia, and russia, are eastern european countries. this is likely to be the result of the recent increase in joins of these east european countries into the european union, acquiring domestic labor’s freedom of movement and employment in advanced countries such as u.k., germany, and france. we chose four countries, south korea, china, japan, and u.s. as samples to review timeseries trends of individual economies. we chose china because it is one of the fast-growing countries with largest population size in the world, u.s. because of its largest gdp and highest technology level with top-class quality of human capital, south korea because it is known as a miraculous example case where education and human capital played important roles to escape from the ‘middle-income trap’ and achieve sustained rapid growth, i.e., ‘six-percent six-decade’ growth. finally, japan was selected because it is widely known that it has become a trend in japan that japanese students do not pursue doctorate program abroad, and at the same time, the economy has been slowing down for decades despite its large size. 5 https://www.nsf.gov/statistics/doctorates/ https://www.nsf.gov/statistics/doctorates/ woosik yu / european journal of government and economics 10(2), december 2021, 128-145 134 figure 1. time series of imd brain drain index of south korea, china, japan, and u.s.a. source: imd world competitiveness executive opinion survey figure 1 shows that china’s imd brain drain index has been somewhat flat with some up and downs, but decreasing since 2013, while that of japan has been rising. moreover, in recent years, south korea’s imd brain drain index exceeded that of china (in years 2012, 2014, 2015, and 2016). in 2016 south korea was ranked 16th at 6.40, surpassing china’s 6.05 (20th). also, it is interesting to notice that south korea’s brain drain has leaped dramatically during the global financial crisis, in years 2007 and 2008, and since then, the trend is more or less flat, aside from the one-time negative shock in 2013. 4.2. nsf doctorate recipients from u.s. universities data next, table 3, the 2016 u.s. nsf’s ranking of the share of ph.d. recipients staying in the u.s. after their completion in the course, tells that iran, bangladesh and venezuela are at the top of the list. unlike the imd brain drain index, six of the top 10 countries are made up of asian countries, iran, bangladesh, nepal, india, china, and taiwan, while only two eastern european countries, bulgaria and romania, are listed in the top 10. also, venezuela, a south american country, and nigeria, an african country, are ranked third and eighth, respectively. this is attributed to the absence of the effect of eastern european countries’ membership status in the european union. in other words, there are no visa problems for u.s. ph.d. holders from eastern europe who have gone to settle down in the advanced european countries such as u.k. and germany, and they are not counted and not reflected in this ranking. this is because the top universities and research institutes in europe do not lag far behind compared to those of the u.s. in terms of research environment and the quality of co-workers. global financial crisis 1 2 3 4 5 6 7 8 im d b ra in d ra in in de x (1 0 o rig in al v al ue ) 19 95 19 98 20 01 20 04 20 07 20 10 20 13 20 16 year south korea china japan usa woosik yu / european journal of government and economics 10(2), december 2021, 128-145 135 table 3. rankings of share of new ph.d.’s staying in the u.s. by country in 20166. rank country no. of us phd recipients share of staying in the us rank country no. of us phd recipients share of staying in the us 1 iran 695 0.911 22 italy 167 0.659 2 bangladesh 185 0.903 23 kenya 61 0.656 3 venezuela 27 0.889 24 argentina 63 0.635 4 bulgaria 44 0.886 25 spain 73 0.616 5 nepal 226 0.885 26 columbia 187 0.615 6 romania 51 0.882 27 mexico 222 0.613 7 india 2,203 0.872 28 germany 183 0.612 8 nigeria 111 0.847 29 turkey 472 0.576 9 china 5,534 0.809 30 brazil 155 0.568 10 taiwan 593 0.779 31 canada 407 0.565 11 philippines 51 0.765 32 britain 115 0.548 12 vietnam 124 0.718 33 japan 166 0.530 13 greece 84 0.714 34 israel 83 0.518 14 ukraine 31 0.710 35 pakistan 102 0.451 15 lebanon 58 0.707 36 jordan 98 0.449 16 ghana 87 0.701 37 indonesia 92 0.435 17 russia 108 0.694 38 singapore 90 0.389 18 south korea 1,228 0.668 39 chile 130 0.323 19 australia 42 0.667 40 thailand 185 0.303 20 france 105 0.667 41 saudi arabia 238 0.101 21 egypt 118 0.661 note. source: u.s. national science foundation interestingly, figure 2 shows a different pattern in the number of ph.d.’s between south korea and china. korea’s share of doctorate recipients staying in the u.s. sharply increases since 2013 while that of china continuously drops since 2012. moreover, the number of new ph.d. graduates from korea decreases since 2012, while that of china has risen continuously and significantly at least from the year 2010, showing the seemingly opposite trends. in korea, a high percentage (52.4%) of doctorate holders are starting their research career as non-regular workers (song et al., 2016)7, which are often discriminated against not only by the threat of job security but also by limitations to research activities, wages, children’s education supports, and welfare systems compared to the full-time positioned workers. on the other hand, since the 1990s, chinese government started an initiative called ‘thousand talents plan (ttp)’ or so-called ‘the salmon project’, of which the goal is to bring back 1000 talents to china who have studied in advanced countries by providing exceptional benefits 6 the higher the value, the more serious the degree of brain drain is. 7 in 2016, 75.5% of south korea’s new doctorate recipients are employed or confirmed to be employed, and among them, only 63.1% are employed for full-time regular positions, and 36.9% are employed for temporary (non-regular) positions. in sum, more than half (52.4%) of south korea’s doctorate recipients are either unemployed or employed for temporary jobs. woosik yu / european journal of government and economics 10(2), december 2021, 128-145 136 including financial reward of one million yuan per person, housings, medical insurance, and even the high-quality education opportunities for their children. (oh, 2016). this plan, which is also translated as the ‘the recruiting program of global experts’, aims to recruit world-class scholars, entrepreneurs, professional technicians and manager-level talents over the next five to ten years and deploy them to state-driven research projects, state-owned enterprises and banks, and industrial technology complexes (lee, 2018). as a result of these efforts, many chinese doctorate recipients appear to be returning to their home countries to settle down, and graduate students actively seek for ph.d. study opportunities in the u.s. because they are aware of these rewards. figure 2. number of u.s. ph.d. recipients and the share of them intending to stay in the u.s. source: u.s. national science foundation on the other hand, for japan and iran, the number of ph.d. recipients and the share of ph.d.’s staying in the u.s. move in the same direction. in japan, both the number of ph.d. recipients and the share of ph.d.’s staying in the u.s. show downward patterns by 2015. in the case of iran, both variables show a pattern of upward movement, and the gradients are steep. in particular, the number of u.s. doctorate recipients from iran grows rapidly from 147 in year 2010 to 695 in 2016. iran was selected as a new sample for analysis simply because its ‘share of ph.d.’s staying in the u.s.’ was the highest among the sample countries as of 2016. 12 00 13 00 14 00 15 00 .5 8 .6 .6 2 .6 4 .6 6 2010 2012 2014 2016 south korea 35 00 40 00 45 00 50 00 55 00 .8 1 .8 15 .8 2 .8 25 2010 2012 2014 2016 china 16 0 18 0 20 0 22 0 24 0 .4 .4 5 .5 .5 5 2010 2012 2014 2016 year japan 0 20 0 40 0 60 0 80 0 .8 4 .8 6 .8 8 .9 .9 2 2010 2012 2014 2016 year iran no. of new ph.d.s (straight) share of staying (dotted) woosik yu / european journal of government and economics 10(2), december 2021, 128-145 137 table 4. pearson correlation analysis between brain drain index and selected macroeconomic variable. variables8 correlation coefficient p-value brain drain index share of staying in the us 0.519 0.000 brain drain index total factor productivity -0.477 0.000 brain drain index gdp per capita -0.452 0.000 brain drain index human capital index -0.271 0.000 share of staying in the us total factor productivity -0.283 0.000 share of staying in the us gdp per capita -0.415 0.000 share of staying in the us human capital index -0.255 0.000 4.3. correlation analysis of brain drain indicators table 4 shows the results of pearson correlation tests between the brain drain variables and key macroeconomic variables. we are presenting this correlation analysis to show the landscape of macroeconomic data and their relationships with brain drain variables before moving on to the regression analysis. first, there was a strong correlation between imd brain drain index and the share of ph.d. recipients staying in the u.s. the higher the share of ph.d.’s staying in the u.s. is, the higher the brain drain index is. in addition, both variables have negative correlation with the key macroeconomic growth variables, i.e., total factor productivity, per capita gdp, and human capital index. the more technically advanced, or the more productive the countries are, the richer the countries are, and the higher the human capital index is, the less brain drain there is. 5. empirical analysis 5.1. model specification in this study, we use these two variables described earlier, imd’s ‘brain drain index’ and nsf’s ‘share of ph.d.’s staying in the u.s.’ as key explanatory variables and explain the economic growth and total factor productivity growth using fixed-effects linear regression models. the growth regression models are represented by the following six reduced-form equations such that: 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 1: 𝑔𝑔_𝑦𝑦𝑖𝑖 ,𝑡𝑡~𝑡𝑡+5 = 𝛽𝛽0 + 𝛽𝛽1𝐵𝐵𝐵𝐵𝐵𝐵𝑖𝑖 ,𝑡𝑡 + 𝛾𝛾𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝜈𝜈𝑡𝑡 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖.𝑡𝑡 [3] 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 2: 𝑔𝑔_𝑦𝑦𝑖𝑖 ,𝑡𝑡~𝑡𝑡+5 = 𝛽𝛽0 + 𝛽𝛽1𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 + 𝛾𝛾𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝜈𝜈𝑡𝑡 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖.𝑡𝑡 [4] 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 3: 𝑔𝑔_𝑦𝑦𝑖𝑖 ,𝑡𝑡~𝑡𝑡+5 = 𝛽𝛽0 + 𝛽𝛽1𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 + 𝛾𝛾0𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖,𝑡𝑡 + 𝛾𝛾1𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝜈𝜈𝑡𝑡 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖.𝑡𝑡 [5] 8 total factor productivity index is ‘ctfp’ from penn world table (pwt) 9.1 (feenstra rc, inklaar r & timmer mp, 2015), gdp per capita is ‘rgdpe/pop’ from pwt, human capital index is ‘hc’ from pwt. woosik yu / european journal of government and economics 10(2), december 2021, 128-145 138 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 4: 𝑔𝑔_𝑇𝑇𝑖𝑖,𝑡𝑡~𝑡𝑡+5 = 𝛽𝛽0 + 𝛽𝛽1𝐵𝐵𝐵𝐵𝐵𝐵𝑖𝑖,𝑡𝑡 + 𝛾𝛾𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝜈𝜈𝑡𝑡 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖.𝑡𝑡 [6] 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 5: 𝑔𝑔_𝑇𝑇𝑖𝑖,𝑡𝑡~𝑡𝑡+5 = 𝛽𝛽0 + 𝛽𝛽1𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 + 𝛾𝛾𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝜈𝜈𝑡𝑡 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖.𝑡𝑡 [7] 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 6: 𝑔𝑔_𝑇𝑇𝑖𝑖,𝑡𝑡~𝑡𝑡+5 = 𝛽𝛽0 + 𝛽𝛽1𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 + 𝛾𝛾0𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖,𝑡𝑡 + 𝛾𝛾1𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝜈𝜈𝑡𝑡 + 𝜐𝜐𝑖𝑖 + 𝜀𝜀𝑖𝑖.𝑡𝑡 [8] where 𝑔𝑔_𝑦𝑦𝑖𝑖,𝑡𝑡~𝑡𝑡+5 and 𝑔𝑔_𝑇𝑇𝑖𝑖,𝑡𝑡~𝑡𝑡+5 are average annual growth rates of home country 𝑖𝑖 ’s income (per-capita gdp) and total factor productivity from year 𝑡𝑡 to year 𝑡𝑡 + 5, respectively. 𝐵𝐵𝐵𝐵𝐵𝐵𝑖𝑖,𝑡𝑡 and 𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 denote country 𝑖𝑖’s brain drain index and share of doctoral graduates staying in the u.s. at year 𝑡𝑡, respectively. 𝑁𝑁𝑁𝑁𝑁𝑁𝑖𝑖,𝑡𝑡 represents the number of u.s. ph.d. recipients of country 𝑖𝑖 in year 𝑡𝑡. this term is included to control for the differences in capacities of sample countries to send their students for u.s. doctoral programs. 𝑇𝑇𝑖𝑖,𝑡𝑡 is a set of external environmental and policy variables that may affect the dependent variables, namely, lagged income (lagged gdp per capita), years of schooling, fertility rate, investment to gdp ratio, government spending to gdp ratio, trade openness, and terms of trade change. these variables are often used in literature with conventional growth regression models such as barro (2016) and lee (2017). moreover, 𝜈𝜈𝑡𝑡 and 𝜐𝜐𝑖𝑖 are included to control for the yearand country-specific fixed effects, respectively. 𝜀𝜀𝑖𝑖.𝑡𝑡 represents the error term. the followings are summary statistics of variables used in regression analysis. table 5 is a summary statistics table for the regression models that use imd brain drain data, and table 6 is the table for the models that use nsf doctorate recipients from u.s. universities data. since imd data contains more country samples with a longer time span, the regression models that use this data have more observations than those that use nsf data. the correlation matrices of all used variables are presented in the appendix (see table a1 and table a2). table 5. summary statistics for model 1 and model 4 (1993~2017, 61 countries).9 variable obs. mean std. dev. min max gdp per capita growth* 1,342 0.034 0.035 -0.133 0.230 tfp growth* 1,320 0.003 0.028 -0.146 0.125 brain drain index 1,361 4.80 1.52 1.33 8.56 lagged income** (usd) 1,525 23,589 17,852 1,170 136,890 years of schooling 1,464 9.95 2.20 3.29 13.55 fertility rate 1,500 1.90 0.64 0.93 4.93 investment/gdp 1,525 0.248 0.069 0.002 0.640 government spending/gdp 1,525 0.182 0.064 0.037 0.423 trade openness 1,525 0.800 0.693 0.068 5.457 terms of trade change 1,525 0.005 0.064 -0.060 1.789 *: average annual growth rate from year t to year t+5. **: value of income for year t-5 9 gdp per capita (or income) is ‘rgdpe/pop’ from pwt, years of schooling is ‘yr_sch’ from pwt, investment to gdp ratio is ‘csh_i’ from pwt, government spending to gdp ratio is from ‘csh_g’ from pwt, trade to gdp ratio is ‘csh_x-csh_m’ from pwt, the terms of trade is ‘pl_x/pl_m’ from pwt, and fertility rate is from world development indicator’s ‘fertility rate’ variable. woosik yu / european journal of government and economics 10(2), december 2021, 128-145 139 table 6. summary statistics for models 2, 3, 5, and 6 (2008~2017, 41 countries) variable obs. mean std. dev. min max gdp per capita growth* 287 0.022 0.033 -0.133 0.101 tfp growth* 245 -0.002 0.027 -0.146 0.056 share of staying in us 410 0.626 0.160 0.122 0.904 number of us ph.d. recipients 410 333 780 28 5,454 lagged income** 410 18,570 14,789 1271 72,583 years of schooling 360 9.11 2.69 3.11 13.55 fertility rate 400 2.23 0.96 1.19 5.90 investment/gdp 410 0.238 0.075 0.076 0.501 government spending/gdp 410 0.165 0.047 0.059 0.336 trade openness 410 0.528 0.451 0.084 3.251 terms of trade change 410 0.013 0.122 -0.042 1.789 *: average annual growth rate from year t to year t+5 **: value of income for year t-5 5.2. results table 7 shows the results of linear regression analysis that explain the growth rate of gdp per capita and the growth rate of total factor productivity. all variables take five-year (plus and minus 2 years) moving average forms to rule out the short-term shocks and capture the long-term trend. as mentioned above, the dependent variables, gdp per capita growth and tfp growth, are generated by calculating growth rates from year 𝑡𝑡 to year 𝑡𝑡 + 5 to see the future (five-year-later) effects of brain drain. the control variable ‘ln (lagged income)’ took the lagged form, which uses the value of the year five years ahead (𝑡𝑡 − 5). the analysis shows that both the brain drain index and the share of ph.d.’s staying in the u.s. have significant and positive effects on economic growth and productivity growth. models that have tfp growth as the dependent variable (models 4, 5, and 6) show greater regression coefficients for key explanatory variables than the models with gdp per capita growth. this implies that brain drain more affects tfp growth and suggest that the main channel of brain drain affecting economic growth is likely to be the productivity growth. in other words, since brain drain does not directly affect income growth, but via tfp growth, the regression coefficient on tfp turns out to be stronger, and this is on the same line with our hypothesis. also, the differences of fit (rsquared and adjusted r-squared) are higher for models with ‘share of staying in u.s.’ variable (models 2, 3, 5 and 6), implying that these models better explain the effects of brain drain compared to other models. in addition, the inclusion of controls for the number of u.s. ph.d. recipients does not alter the regression result, suggesting that the possible sample selection issues related to the magnitude of doctorate recipients are well-controlled. woosik yu / european journal of government and economics 10(2), december 2021, 128-145 140 table 7. results of linear regression of economic and total productivity growth. model 1 model 2 model 3 model 4 model 5 model 6 gdp per capita growth (t+5) tfp growth (t+5) brain drain index 0.012** (0.012) 0.013*** (0.003) share of staying in us 0.062** (0.015) 0.063** (0.021) 0.080*** (0.003) 0.082*** (0.005) ln (number of us ph.d. recipients) 0.0004 0.002 ln (lagged income) -0.041*** 0.010 0.010 -0.028*** -0.003 -0.003 ln (years of schooling) -0.033 -0.003 -0.004 -0.015 0.011 0.006 ln (fertility rate) -0.045*** -0.014 -0.014 -0.021* -0.008 -0.006 ln (investment/gdp) -0.031*** -0.016 -0.016 0.013** 0.017 0.018 ln (gov. sp. /gdp) -0.003 0.070*** 0.070*** 0.014*** 0.081*** 0.081*** ln (trade openness) 0.031*** -0.073*** -0.073*** 0.030*** -0.046* -0.047* terms of trade chg. -0.096 -0.257** -0.257** 0.164* -0.072 -0.071 country & year fe included included included included included included observations 1,156 273 273 1,150 238 238 r-squared 0.406 0.801 0.801 0.308 0.735 0.735 adjusted r-squared 0.357 0.754 0.753 0.252 0.669 0.667 note. p < 0.1, ** p < 0.05, *** p < 0.01; values in parenthesis are p-values; ‘brain drain index’ and ‘share of staying in u.s.’ are in natural logarithm terms. however, in model 1, investment to gdp ratio shows a strong and negative relationship with per capita gdp growth rate, and in models 2 and 3, trade openness has strong and negative regression coefficients, and thus, these models seem to be disqualifying to be established as benchmark models and show possibilities of needs for other control variables. on the other hand, the correlation coefficients of models 4, 5, and 6 seem to be consistent with the conventional growth regression models. lastly, the authors are acknowledging the potential weakness of the result that even with the year-specific fixed effects method, due to the moving average forms of the variables, there may be a possibility of issues with autocorrelation across time. also, a more thorough analysis can be conducted with the inclusion of factors that determines the decision of the talented to study abroad in the u.s., as well as the decision to stay in the u.s. or return to the home country. for example, the following factors, which are not considered in this paper’s empirical analysis, may improve the models: the baseline living conditions of home country, job and income prospects in the u.s. compared to those of home countries, cultural differences, cost of living in the home countries. woosik yu / european journal of government and economics 10(2), december 2021, 128-145 141 6. conclusion according to our international comparisons of brain drain, the patterns of brain drain vary from country to country. particularly, south korea and china show stark differences. south korea’s human resource outflow continuously increases and deepens while china's talent outflow is showing a marked decline. korean media often argue that the deepening of this pattern will reduce the relative quality of south korea’s human capital, and that of china will continue to accumulate and surpass korea’s level soon. in general, middle-income countries rather than highincome group have more brain drain. also, emerging economies that grow fast tend to experience more brain drain. however, the results of growth regression analysis show that so-called brain drain has a significant impact on both economic growth and productivity (tfp) growth, implying that brain drain does not always affect the economy in a negative way. moreover, brain drain has shown to have a more positive association with productivity growth, suggesting that the main channel of brain drain inducing economic growth is likely to be the productivity growth. this can be considered as evidence of the positive effects of ‘brain circulation’, which is one of the brain drain phenomenon that settlement of the talents in advanced countries can eventually help improve the productivity of home country by sharing of advanced technologies and skills around them with colleagues in motherland. therefore, from a policy point of view, a two-track strategy is suggested to simultaneously pursue measures to maximize the improvement of national level of technology, which emphasizes the utilization of overseas resident talents alongside the brain-attraction policy. in this context, developing nations should try avoiding the unconditional brain-attraction policy. furthermore, the support policies to raise the quality of research institutes and universities to the world-class level, as well as policies that encourage talents-sharing with global frontier firms will also help improve the home country’s technological prowess through technology and knowledge diffusion in the long run, as these policies can attract and embrace skilled overseas brain groups. references barro, r. j. (2016). economic growth and convergence, applied to china. china and world economy. 5(24):5-19. https://doi.org/10.1111/cwe.12172 beine, m., docquier, f., & rapoport, h. (2001). brain drain and economic growth: theory and evidence. journal of development economics, 64(1), 275-289. https://doi.org/10.1016/s03043878(00)00133-4 bénassy, j. p., & brezis, e. s. (2013). brain drain and development traps. journal of development economics, 102, 15-22. https://doi.org/10.1016/j.jdeveco.2012.11.002 https://doi.org/10.1111/cwe.12172 https://doi.org/10.1016/s0304-3878(00)00133-4 https://doi.org/10.1016/s0304-3878(00)00133-4 https://doi.org/10.1016/j.jdeveco.2012.11.002 woosik yu / european journal of government and economics 10(2), december 2021, 128-145 142 bhagwati, j., & hamada, k. (1974). the brain drain, international integration of markets for professionals and unemployment: a theoretical analysis. journal of development economics, 1(1), 19-42. https://doi.org/10.1016/0304-3878(74)90020-0 cinar, d., & docquier, f. (2004). brain drain and remittances: implications for the source country. brussels economic review, 47(1), 103-118. docquier, f., & rapoport, h. (2008). brain drain and human capital formation in developing countries: winners and losers. the economic journal, 118(528), 631-652. https://doi.org/10.1111/j.1468-0297.2008.02135.x feenstra r. c., inklaar r., & timmer m. p. (2015). the next generation of the penn world table. american economic review, 105(10), 3150–3182. https://doi.org/10.1257/aer.20130954 grubel, h. b., & scott, a. d. (1966). the international flow of human capital. american economic review, 56(1/2), 268-274. imd world competitiveness executive opinion survey imd world competitiveness yearbook kang, d. k., e.-s lim, & j. hwang. (2013). an exploratory study on the relationship between brain drain and national competitiveness. iom mrtc working paper series, no. 2013-05. iom migration research and training centre. republic of korea. lee, j. w. (2017). china's economic growth and convergence. world economy. 40:2455-2474. https://doi.org/10.1111/twec.12554 lee, k. (2018). chinese science paper results exceed the u.s. the science times. url: https://www.sciencetimes.co.kr/?news=%ec%a4%91%ea%b5%ad-%ea%b3%bc%ed%9 5%99%eb%85%bc%eb%ac%b8-%ec%84%b1%ea%b3%bc-%eb%af%b8%ea%b5% ad-%ec%b6%94%ec%9b%94%ed%95%9c%eb%8b%a4 (retrieved on june 3, 2019. in korean, title translated by the author). lodigiani, e., marchiori, l., & shen, i.-l. (2015). revisiting the brain drain literature with insights from a dynamic general equilibrium world model. the world economy, 39(4), 557–573. https://doi.org/10.1111/twec.12269 mcculloch, r., & yellen, j. t. (1977). factor mobility, regional development and the distribution of income. journal of political economy, 85(1), 79-96. https://doi.org/10.1086/260546 mountford, a. (1997). can a brain drain be good for growth in the source economy? journal of development economics, 53(2), 287-303. https://doi.org/10.1016/s0304-3878(97)00021-7 oh, h. (2016). leakage of korean talents. chosun ilbo. http://news.chosun.com/site/data/html _dir/2016/08/08/2016080801521.html (retrieved on september 26, 2019. in korean, title translated by the author). saxenian, a. (2005). from brain drain to brain circulation: transnational communities and regional upgrading in india and china. studies in comparative international development, 40(2), 35-61. https://doi.org/10.1007/bf02686293 song, c. y., oh, h. y., kim, h. j., & lee, e. h. (2016). korea survey of earned doctorates, 2016. korea research institute for vocational education & training, krivet. republic of korea. https://doi.org/10.1016/0304-3878(74)90020-0 https://doi.org/10.1111/j.1468-0297.2008.02135.x https://doi.org/10.1257/aer.20130954 https://doi.org/10.1111/twec.12554 https://www.sciencetimes.co.kr/?news=%ec%a4%91%ea%b5%ad-%ea%b3%bc%ed%95%99%eb%85%bc%eb%ac%b8-%ec%84%b1%ea%b3%bc-%eb%af%b8%ea%b5%ad-%ec%b6%94%ec%9b%94%ed%95%9c%eb%8b%a4 https://www.sciencetimes.co.kr/?news=%ec%a4%91%ea%b5%ad-%ea%b3%bc%ed%95%99%eb%85%bc%eb%ac%b8-%ec%84%b1%ea%b3%bc-%eb%af%b8%ea%b5%ad-%ec%b6%94%ec%9b%94%ed%95%9c%eb%8b%a4 https://www.sciencetimes.co.kr/?news=%ec%a4%91%ea%b5%ad-%ea%b3%bc%ed%95%99%eb%85%bc%eb%ac%b8-%ec%84%b1%ea%b3%bc-%eb%af%b8%ea%b5%ad-%ec%b6%94%ec%9b%94%ed%95%9c%eb%8b%a4 https://doi.org/10.1111/twec.12269 https://doi.org/10.1086/260546 https://doi.org/10.1016/s0304-3878(97)00021-7 http://news.chosun.com/site/data/html_dir/2016/08/08/2016080801521.html http://news.chosun.com/site/data/html_dir/2016/08/08/2016080801521.html https://doi.org/10.1007/bf02686293 woosik yu / european journal of government and economics 10(2), december 2021, 128-145 143 stark, o., helmenstein, c., & prskawetz, a. (1997). a brain gain with a brain drain. economics letters, 55(2), 227-234. https://doi.org/10.1016/s0165-1765(97)00085-2 survey of earned doctorates, doctorate recipients from u.s. universities, national science foundation (nsf), url: https://www.nsf.gov/statistics/doctorates/ teney, c. (2021). immigration of highly skilled european professionals to germany: intra-eu brain gain or brain circulation? innovation: the european journal of social science research, 34(1), 69-92. https://doi.org/10.1080/13511610.2019.1578197 the imd world talent ranking methodology (2019). https://www.imd.org/globalassets/wcc/ docs/release-2019/talent/imd_world_talent_ranking_methodology_2019.pdf (retrieved on november 12, 2021). vidal, j.-p. (1998). the effect of emigration on human capital formation. journal of population economics, 11(4), 589-600. https://doi.org/10.1007/s001480050086 https://doi.org/10.1016/s0165-1765(97)00085-2 https://www.nsf.gov/statistics/doctorates/ https://doi.org/10.1080/13511610.2019.1578197 https://www.imd.org/globalassets/wcc/docs/release-2019/talent/imd_world_talent_ranking_methodology_2019.pdf https://www.imd.org/globalassets/wcc/docs/release-2019/talent/imd_world_talent_ranking_methodology_2019.pdf https://doi.org/10.1007/s001480050086 woosik yu / european journal of government and economics 10(2), december 2021, 128-145 144 appendix table a1. correlation matrix for variables of models 1 and 4. var1 var2 var3 var4 var5 var6 var7 var8 var9 var2 0.667 1.000 (0.000) var3 0.136 0.060 1.000 (0.000) (0.041) var4 -0.327 -0.219 -0.448 1.000 (0.000) (0.000) (0.000) var5 -0.170 0.039 -0.261 0.427 1.000 (0.000) (0.153) (0.000) (0.000) var6 0.041 0.008 0.163 -0.231 -0.509 1.000 (0.136) (0.779) (0.000) (0.000) (0.000) var7 -0.181 -0.085 -0.353 0.409 0.008 -0.087 1.000 (0.000) (0.002) (0.000) (0.000) (0.766) (0.001) var8 0.268 0.143 0.361 -0.405 0.094 -0.141 -0.413 1.000 (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) var9 -0.022 -0.025 -0.244 0.397 0.266 -0.305 0.391 -0.155 1.000 (0.412) (0.371) (0.000) (0.000) (0.000) (0.000) (0.000) (0.000) var10 -0.069 -0.105 0.128 -0.028 -0.035 0.038 -0.017 0.038 -0.045 (0.012) (0.000) (0.000) (0.283) (0.177) (0.139) (0.503) (0.134) (0.077) note. values in parentheses are p-values; var1: gdp per capita growth; var2: tfp growth; var3: imd brain drain index; var4: lagged income; var5: years of schooling; var6: fertility rate; var7: investment to gdp ratio; var8: government spending to gdp ratio; var9: trade openness; var10: terms of trade change woosik yu / european journal of government and economics 10(2), december 2021, 128-145 145 table a2. correlation matrix for variables of models 2, 3, 5, and 6. var1 var2 var3 var4 var5 var6 var7 var8 var9 var10 var2 0.866 1.000 (0.000) var3 0.154 0.057 1.000 (0.009) (0.378) var4 0.198 0.031 0.251 1.000 (0.001) (0.635) (0.000) var5 -0.084 0.094 -0.400 -0.098 1.000 (0.155) (0.144) (0.000) (0.048) var6 -0.032 0.209 -0.252 -0.112 0.747 1.000 (0.592) (0.001) (0.000) (0.034) (0.000) var7 -0.196 -0.294 0.082 -0.145 -0.487 -0.560 1.000 (0.001) (0.000) (0.104) (0.004) (0.000) (0.000) var8 0.157 0.003 -0.079 0.568 0.222 -0.005 -0.330 1.000 (0.008) (0.967) (0.112) (0.000) (0.000) (0.930) (0.000) var9 -0.058 0.287 -0.084 -0.156 0.233 0.523 -0.361 -0.222 1.000 (0.328) (0.000) (0.090) (0.002) (0.000) (0.000) (0.000) (0.000) var10 0.019 0.042 -0.323 -0.058 0.688 0.488 -0.361 0.320 0.009 1.000 (0.753) (0.512) (0.000) (0.246) (0.000) (0.000) (0.000) (0.000) (0.862) var11 -0.034 0.060 0.108 -0.036 -0.020 -0.004 0.002 -0.003 0.072 -0.072 (0.570) (0.349) (0.029) (0.463) (0.694) (0.934) (0.963) (0.948) (0.146) (0.144) note. values in parentheses are p-values; var1: gdp per capita growth; var2: tfp growth; var3: share of staying in u.s.; var4: number of u.s. ph.d. recipients; var5: lagged income; var6: years of schooling; var7: fertility rate; var8: investment to gdp ratio; var9: government spending to gdp ratio; var10: trade openness; var11: terms of trade change brain drain and economic growth: evidence of productivity growthfrom brain circulation 1. introduction 2. literature review 3. motivation 4. empirical landscape 5. empirical analysis 6. conclusion references european journal of government and economics 11(1), june 2022, 113-137 european journal of government and economics issn: 2254-7088 corporate governance and organizational commitment: the mediating role of organizational culture. najm a. najm a, *, abdul azez b. alnidawy a, abdul sattar. h. yousif a a al-zaytoonah university of jordan, jordan * corresponding author at: najim_abood@yahoo.com abstract. this study seeks to determine the impact of corporate governance dimensions (compliance with the corporate governance code, top management, control environment, transparency and disclosure, rights of shareholders and stakeholders) on the three main types of organizational commitment, (affective, continuance and normative). it also aims at examining the impact of organizational culture, as an intermediate variable, on the relationship between the two above mentioned variables. the sample of this study comprised 152 respondents working at five types of jordanian companies. the results of the study have confirmed the positive effect of the three corporate governance dimensions (compliance with the corporate governance code, top management, and control environment) on three types of organizational commitment (affective, continuance, and normative). the results also confirmed that there is no significant effect of t transparency and disclosure and the rights of shareholders and stakeholders on affective and normative commitment. key terms: corporate governance, top management, board of directors, management committee, control environment, transparency, disclosure, shareholders, and stakeholders. keywords. corporate governance, top management, board of directors, transparency, disclosure, stakeholders. doi. https://doi.org/10.17979/ejge.2022.11.1.7564 1. introduction the world is currently witnessing a growing interest in the principles and practices of governance in government institutions (as public governance) and business organizations (as private governance). the european union governance system for 2020 strategy has been adopted at the union level as a whole and at the individual member level, (drumaux and joyce, 2018). companies, on the other hand, adopt a corporate governance code to avoid financial crises and ethical scandals alike (stringham, 2015; pattber, 2005). the importance of governance can also be recognized through the broad international movement toward the concept of public governance to improve governmental decisions and practices (osborne, 2010) and enhance public sector integrity and assure its effectiveness and innovation (torfing, and triantafillou, 2016, p2). the global cross-border governance operates under globalization, rules of international competition, and private governance, which are all of high importance to business firms and other non-governmental companies (baer, 2013, p45). the evolution of governance could be perceived through the evolution of the concept itself as it developed from merely a set of governance rules to strategic governance, which was the © 2022. this work is licensed under a cc by-nc 4.0 license. mailto:najim_abood@yahoo.com https://doi.org/10.17979/ejge.2022.11.1.7564 https://creativecommons.org/licenses/by-nc/4.0/ najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 114 case for the development of other types of governance: public and private governance, hierarchical and market governance (aluchna, 2009), internal and external governance (daley et al., 2003, p20), company-based and personal governance. according to sam dipiazza, former ceo of pricewaterhousecoopers, it has become dramatically clear that the foundation of corporate integrity is the personal integrity (rampersad and hussain, 2014) that is necessary for senior management members in a corporate governance perspective. other types of governance were also subjected to the same process of development, such as entrepreneurial governance (barns, 2016), internet and it governance (weber et al., 2009; grembergen. 2004), and academic governance (hogler et al., 2009). governance is also likely to increase companies' interest in their social, environmental and ethical responsibilities to serve the community in which they operate. in a survey conducted by mckinsey & co. (2007), 95% of 391 ceos who participated in this survey, assert that companies are more concerned with public responsibilities than they were five years ago (cited in: hemphill, 2010, p118). good governance represents the most effective and ethical framework for facing the growing challenges in the business environment, and its associated pressure, which usually drives a company to take on completely unacceptable decisions and practices. the experiences of organizations like worldcom, im clone, vivendi, global crossing, lucent, arthur andersen and specially enron and enronism (bellingham,2003, pp xii and 1, khiari et al., 2007, p148) have revealed that companies, regardless of their sizes, when operated without good governance could start failing, become threatened with moral scandals and might be prosecuted through lawsuits that would chase their leadership even after the liquidation of that company. these disturbing experiences revealed that senior management could process data in a way that leads to improving the company’s business results in the short term, despite their knowledge that these results are incorrect and can cause severe damage to the company in the long run and lead to disputes and problems with investors (hoopes, 2003, xxxix). at the same time, good corporate governance that works to achieve a balanced distribution of powers between the company's chief executives (the president and the board of directors) and the integration of responsibilities among the various stakeholders (shareholders, employees, suppliers, customers and the community), contributes to avoiding the many shortcomings experienced by companies with poor governance. in this context, haspeslagh (2010) stated a list of persistent shortcomings that characterized the companies of failed corporate governance, such as greed, wishful thinking and linear extrapolation, addiction to capital markets, which represent causes of sustained crises to the company. it might be necessary to emphasize that the development of corporate governance has its own supporting factors represented in need to align stakeholders’ rights with shareholders’ rights, fighting financial and moral scandals (berthelot et al., 2010, p635) and corruption (especially in government corporate governance) in order to expedite the adoption of the corporate governance code and public governance regulations imposed on companies such as the corporate governance code of jordan, and corporate governance awards at national levels such as corporate governance awards by indian chamber of commerce najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 115 (https://www.indianchamber.org), and at the regional level such as corporate governance award for middle east & africa, by ethical boardroom (https://ethicalboardroom.com), and at the international level: the corporate governance awards of the world: such as world finance corporate governance awards by world finance (https://www.worldfinance.com). these factors will work continuously to promote corporate governance principles and practices in a business environment. on the other hand, corporate governance, which is an important breakthrough in the development of the structure and responsibilities of senior management and its relationship with the various stakeholders in companies, continues to face real challenges because there is a great deal of disagreement about how good or bad the current governance mechanisms are (shleifer and vishny, 1997). the corporate governance received good evaluations in terms of its positive impact on the protection of the rights of shareholders and the organizational commitment (purwanto, 2015), the organizational performance of the companies (adebayo et al., 2014). however, governance is still associated with corporate failure (lakshana and wijekoonb, 2012), poor performance of internal financial control systems (jensen, 1993), the high compensation of ceos or the “fat cat problem (lin et al., 2013), corporate scandals (pozner et al., 2010; agrawal and chadha, 2005). the us, which has the best corporate governance system, was hit by the 2007-2008 financial crisis (us housing bubble), which lasted 18 months, and the system took another long time to recover (basco, 2018; altug, 2010). these unpleasant events and undesirable results can strongly reflect the organizational commitment to the company's relationship with its employees. they also reveal the need to embody the principles of good governance and that the regulatory authorities contribute to ensuring this compliance (de villiers and dimes, 2021). in jordan, as a developing country, companies look to corporate governance as an effective system for regulating the relationships between the various parties (ceos, directors, managers, owners and other stakeholders), working according to rules and principles for making fair decisions and improving the results of their business. the current study aimed to determine the impact of the dimensions of governance (board of directors and management committee), control environment, transparency and disclosure, rights of shareholders and stakeholders) on the three main types of organizational commitment, (affective, continuance and normative) in the environment of jordanian companies. it also sought to determine the impact of organizational culture on the causal relationship between corporate governance and organizational commitment. finally, the noticeable lack of studies that focused on corporate governance and organizational commitment, reveals the contribution of this study in revealing the causal relationship between governance and organizational commitment and the impact of organizational culture on this relationship. the results of this study provide a contribution to understanding the relationship between these factors in the business environment. https://www.indianchamber.org/ https://ethicalboardroom.com/ https://www.journals.uchicago.edu/doi/10.1086/430808 https://www.journals.uchicago.edu/doi/10.1086/430808 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 116 2. corporate governance and organizational commitment corporate governance has become an essential organizational feature and an important guideline for casting the light on the nature of top management duties and responsibilities (board directors and ceo). moreover, it evolved into a base for specifying a company’s various relationships with its shareholders, stakeholders, investors, governmental institutions, and the community in which it is operating. for a considerable time, the corporate senior management has been controlling all types of activities, while they have immunity against any accountability. the formal adoption of corporate governance has led to important changes that have contributed to the acceptance of governance as a framework of essential managerial development that establishes the principles, values, rules and methods of how senior managers manage their companies. perhaps, this development clearly denotes the main reason behind the growing interest of researchers for studying and analyzing corporate governance at both business and public administration levels. based on the outcomes of many past studies, some broadly used definitions of corporate governance could be presented. handley-schachler (2007) has stated that: “corporate governance is concerned with structures and allocations of responsibilities within companies”. while, haspeslagh (2010) believes that governance is: “a system of rules, regulations and practices through which we used to hold managers and owners accountable for their failure of meeting the performance level expected by the society,”. according to this definition, there is evidence that governance is a set of rules and regulations that control and monitor the activities and relationships of managers and owners in a company. table 1 portrays a chosen set of corporate governance definitions. by reviewing the definitions shown in table 1, some considerable observations can be drawn, which are as follows: these definitions clearly affirmed that there is no consensus among researchers within this field on a united definition of corporate governance. the (1997) studies provided six definitions, whereas this study provided (17) additional definitions with the possibility of more definitions suggested by future studies. researchers differed in their formulation and adoption of the definition of corporate governance due to differences in their backgrounds, experiences and cultures. aluchna’s study (2009) has confirmed that the ownership structure affects corporate governance. it might be useful to mention that there are many differences between the american, british (the anglo-saxon business system) and european perspectives of corporate governance the american perspective of governance is based on individual owners, the concentration of ownership and shareholder value as the primary focus of a company’s strategy (eldomiaty and choi, 2006), while the european perspective is concerned with family ownership and institutional investors (monks and minow, 2004). li and harrison’s study (2008) revealed that the national culture (such as individuality/collectivity and masculinity/femininity) affects the corporate governance system differently from one country to another. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 117 table 1. corporate governance definitions. author definition oecd (2004) corporate governance is the rules and practices that govern the relationship between the managers and shareholders of corporations, as well as stakeholders like employees and creditors. du plessis et al. (2011) the system of regulating and overseeing corporate conduct and of balancing the interests of all internal stakeholders and other parties (external stakeholders, governments and local communities). haspeslagh (2010) the system of rules, regulations and practices by which we hold managers and owners accountable and responsible for whatever performance society expects. talamo (2011) the system of monitoring devices, internal and external, specific to each organization, that defines how these mechanisms are set up and how each will fulfill its monitoring role. dignam and lowry (2006) a set of processes, customs, policies, laws and institutions affecting the way a corporation is directed, administered or controlled, and its purpose is to influence directly or indirectly the behavior of the organization towards its stakeholders. handley-schachler (2007) corporate governance is concerned with structures and the allocation of responsibilities within companies. dimsdale, and prevezer (1994) corporate governance is concerned with the way in which corporations are governed the relationship between the management of a company and its shareholders. oecd (1999) a set of relationships between a company’s management, its board, its shareholders and other stakeholders. rampersad and hussain (2014) the systematic process of continuous, gradual, and routine improvement, steering, and learning that lead to sustainable high performance and ethical excellence. fahy and weiner (2004) it is the systems and processes put in place to direct and control an organization in order to increase performance and achieve sustainable shareholder value. the governance definitions can be classified into three categories: broad definitions that cover many areas within and outside the company, such as “the ways in which both suppliers of funds and corporations have to ensure their returns on investment (berthelot et al., 2010, p336), narrow definitions that have linked the concept with the company such as ”corporate governance is concerned with structures and the allocation of responsibilities within companies”, or the definitions focusing on the relationships with shareholders such as “corporate governance is concerned with the way in which corporations are governed the relationship between the management of a company and its shareholders”(dimsdale and prevezer, 1994), and finally the detailed definitions that attempt to highlight the important components of governance such as “the system of regulating and overseeing corporate conduct and balancing the interests of all internal stakeholders and other parties (external stakeholders, governments and local communities)”. according to fahy et al.,’s study (2004), good governance is closely related to the company's strategy, and therefore good governance supports the company's strategy. on the other hand, weak governance leads to a poor corporate strategy. the strategic commitment to corporate governance enables the company to build long-term internal and external relationships that would enhance its sustainable competitive advantage. as far as this study is concerned, corporate governance “is a set of rules and guidelines that are in compliance with the corporate governance code, top management structure (board of directors and management committee), robust control environment, transparency and disclosure, protecting the rights of shareholders and stakeholders to achieve company objectives effectively najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 118 and ethically”. this study does not address corporate governance from a shareholder or a stakeholder perspective (dobbin and jung, 2010, p33). it is also not concerned with governance theories such as agency theory (jensen’s theory) (pavel et al., 2012), or transaction cost economics theory which views governance in terms of designing particular mediums for supporting economic transactions (ruhanen et al., 2010). this study focuses on the importance of governance and the impact of its dimensions on organizational commitment with the mediating role of organizational culture in the relationship between the two variables. the importance of these dimensions in companies can represent important evidence of the protection of investors and owners and the extent of rationality in the decisions and practices of managers in these companies. as a result, corporate governance can positively affect the improvement of the business environment and the companies ’contribution to strengthening the economy’s strength in the country. it is necessary to note that researchers do not agree on the dimensions of corporate governance, whether in terms of number or definitional elements. table 2 shows this difference in determining these dimensions according to several studies. in 1999, the organization for economic cooperation and development (oecd) adopted the principles of corporate governance (oecd, 2004, p3) to begin a new phase of interest in good governance for both countries and business organizations. with a substantial, growing of interest in governance, a group of studies have focused on determining the dimensions of corporate governance. according to oecd, a company's governance consists of six dimensions (oecd, 219); corporate governance framework, the rights of shareholders and key ownership functions, the equitable treatment of shareholders, the role of stakeholders in corporate governance, disclosure and transparency, the responsibilities of the board. the majority of companies consider these dimensions as important focal points for establishing internal and external relationships based on, equal treatment, transparency, and disclosure of information, which are necessary to build trust with all stakeholders and society. in jordan, where this study was conducted, the jordanian corporate governance code is guided by oecd corporate governance dimensions, which comprises five dimensions; compliance with the corporate governance code, top management (board of directors and management committee), control environment, transparency and disclosure, rights of shareholders and stakeholders. the corporate governance that was adopted by this study has focused on those five dimensions. employee turnover is an ongoing challenge for business organizations and public institutions alike (iqbal, 2010, p276). however, this challenge has become a organizational and ethical quandary for companies that do not care about organizational commitment and treat their loyal and disloyal employees in the same way. companies are well aware that their most critical resources for market competition are their employees, who make up their human capitals. consequently, the continuation of employees working at the company and doing their best to meet its objectives is a very important source for achieving sustainable competitiveness to outperform its competitors. according to heneman and judge (2009, p692), there are two main categories of labor turnover, first: involuntary turnover, which includes two types: discharge najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 119 turnover (discipline and poor performance) and downsizing turnover (closing, relocation, redundancy due to a merger or acquisition), second: voluntary turnover that consists of two types as well: avoidable turnover (due to wages and job changes) and unavoidable turnover (as in the cases of death, retirement employee turnover was behind the increasing importance of organizational commitment that led to maintain the retaining of highly-qualified employees work at the company for a long time. *table 2. corporate governance dimensions. author/organization corporate governance dimensions hill, and jones, 2009 the board of directors, stock-based compensation financial statements and auditors, the takeover constraint. aishath and hassan, 2014 islamic corporate governance: independent shari’ah committee, the shari’ah committee’s decisions shall be binding on the board of directors, sub-committees of the shari’ah in all departments of the company, shari’ah audit at least once a year, the stakeholders of islamic corporations shall not only be shareholders, right to sue the shari’ah committee in case of pure negligence.. morgan et al., 2009 traditional compliance responsibilities: enterprise code of conduct, disclosure of board’s roles and functions. contemporary citizenship responsibilities: governance structure, emerging citizenship responsibilities: board discloses mechanism for engaging and protecting stakeholders, broad’s responsibility for corporate citizenship du plessis et al., 2011 the system of regulating and overseeing corporate conduct, the interests of internal stakeholders and other parties ensuring responsible behavior by corporations the maximum level of efficiency and profitability for a corporation. public governance: drumaux and joyce, 2018. infrastructure governance index: hertie school of governance and oecd, 2016: the governance report 2016 * for public governance: top-down steering, targeting of funding, voluntary alignment public and amplifying public opinion * three dimensions of infrastructure governance index infrastructure planning, infrastructure management, infrastructure outcomes. rampersad and hussain, 2014 code of corporate governance guide, governance committees: corporate governance, audit, nominating and remuneration committee, shareholders, code of conduct and ethics (and conflict of interest, whistle-blowing program) najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 120 the results of some field studies have confirmed that organizational commitment positively affects companies by considerably, reducing employee turnover, absenteeism, and other working problems. those results have, also, affirmed that organizational commitment increases job satisfaction, improving performance and productivity (singh and gupta, 2015; mosadeghrad et al., 2008; smith, 1996; allen and meyer, 1990). according to a field study conducted by allen and mayer (1990, 1997) and jaros, (2007), it was found that organizational commitment consists of three main types: affective, continuance, and normative commitment. therefore, organizational commitment can be measured using these types; affective commitment (which reflects commitment based on emotional ties that an employee develops with the company he/she is working at, primarily through his/her positive work experiences). continuance commitment (which refers to commitment based on perception, the economic and social costs, if an employee decides to leave the organization). -normative commitment (which reflects commitment based on perceived obligation towards the organization). this study seeks to determine the effect of corporate governance dimensions (according to the jordanian corporate governance code) on organizational commitment, with specifying the effect of organizational culture, as an intermediate variable, on the relationship between corporate governance and organizational commitment. accordingly, the study hypotheses were formulated as follows: h1: corporate governance dimensions have a positive impact on affective commitment. h2: corporate governance dimensions positively affect continuance commitment. h3: corporate governance dimensions have a positive impact on normative commitment. organizational culture is the set of basic assumptions, shared values, and processes, or shared set of values, customs and rituals that unify the way of thinking and behavior of employees in the company. organizational culture does not represent a single pattern in terms of effectiveness and influence. there is a healthy or unhealthy, (kotter and heskett, 1992) strong or weak (barnes et al., 2006), high or low performance (anderson, d. and anderson, 201, p191) culture that has the ability to create a common way of interpreting experiences and events in the company. on the other hand, there is the unhealthy and weak organizational culture where it has limited impact on employees and the pattern of their response to these experiences and events. in this study, the organizational culture represents the mediating variable that can affect the relationship between corporate governance and organizational commitment in jordanian companies as in the fourth hypothesis, as follows: h4: there is a positive mediating effect of organizational culture on the relationship between corporate governance and organizational commitment. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 121 3. method measures in this study, the questionnaire was developed as a data collection tool based on previous studies of corporate governance (rampersad and hussain, 2014; morgan et al., 2009; hill, and jones, 2009), organizational commitment (allan and meyer, 1990; meyer et al.,, 1993; alam, 2011), and organizational culture in jordanian context (cameron and quinn, 2006; wiewiora et al., 2013). the questionnaire was designed to cover the dimensions of governance, organizational commitment, and organizational culture. each dimension of the study was measured by a number of statements representing its components. the compliance with the corporate governance code was measured by four items (a1-a4); top management, (board of directors and management committee) was measured by six items (b1-b6); control environment by five items (c1-c5); transparency and disclosure by five items, and rights of shareholders and stakeholders by six items. also in the questionnaire, each type of organizational commitment was measured by seven items: emotional commitment (x1-x7), continuance (y1-y7), and normative (z1-z7). lastly organizational culture (5 statements) (m1-m5). a five-grade likert scale (1 for strongly disagree and 5 for strongly agree) was used to measure the participants' response to questionnaire statements and assess variables' values. participants a random sample was withdrawn from a set of jordanian companies identified by their governance code, and they were of five categories: public shareholding companies that are not listed in the stock exchange (pusc), private shareholding companies (prsc), : limited liability companies (lilc),: non-profit private companies (nppc) and: non-profit limited liability companies (nplc). the last two categories of organizations were government institutions (stateowned enterprises) plus few social associations whose number of respondents was relatively few. the study sample consisted of 152 respondents representing employees from private companies (banks, insurance, and industry) in addition to employees working at two government institutions (ministry of finance and ministry of labor). table 3 displays the study participant’s characteristics. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 122 table 3. the demographic sample characteristics. characteristics frequency % gender male female total 99 56 152 65.1 34.9 100.0 age < 30 30 -39 40-49 50-59 total 76 42 29 5 152 50.0 27.6 19.1 3.3 100.0 marital status single married total 81 71 152 53,3 46.7 100.0 education < secondary diploma bachelor master phd total 2 25 93 24 8 152 7.1 11.6 61.2 8.9 5.3 100.0 working experience (years) < 5 5-10 11-15 > 15 total 54 45 31 22 152 35.5 29.6 20.4 14.5 100.0 type of company* pusc prsc lilc nppc nplc total 46 53 40 8 5 152 30.3 34.9 26.3 5.3 3.3 100.0 job title manager/supervisor engineer/technician officer n/a total 49 19 72 12 152 32.2 12.5 47.4 7.9 100.0 study variables the statistical analysis of this study was based on three types of variables; the independent (the five dimensions of corporate governance), dependent variables (the three types of organizational commitment), and mediator variable (organizational culture). the five dimensions of corporate governance adopted by this study were developed in compliance with the jordanian corporate governance code. these dimensions are: compliance with the corporate governance code (assuring employee’s awareness of the code of principles and guidelines), top management (the board of directors/management committee), controlled environment, transparency & disclosure, najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 123 and rights of shareholders/ stakeholders, (the jordanian corporate governance code, 2009, p3). according to the three components of the organizational commitment model that were suggested by allen and meyer (1990), the dependent variables involved, affective, continuance, and normative commitment. this organizational commitment model has received considerable attention from number of researchers in this domain during the past three decades (singh and gupta, 2015; ortega-parra and sastre-castillo, 2013; cho and huang, 2012; solinger et al. 2007). table 4 shows the importance of governance dimensions according to respondents' answers. the importance of two dimensions of governance (top management and transparency and disclosure) was relatively high (3.772 and 3.738, respectively), while the importance of the other three dimensions (compliance with the corporate governance code, top management, and rights of shareholders/ stakeholders) was at a medium level (3.324, 3.542. and 3.309). also the importance of the three types of organizational commitment was at a middle level. table 4. descriptive statistics. governance dimensions n minimum maximum mean std. deviation * governance dimensions compliance with the corporate governance code 152 1.00 5.00 3.324 middle 0.857 top management (the board of directors/management committee), 152 1.00 5.00 3.542 middle 0.786 control environment 152 1.00 5.00 3.772 high 0.701 transparency and disclosure 152 1.00 5.00 3.738 high 0.777 rights of shareholders/ stakeholders 152 1.00 5.00 3.309 middle 0.735 * organizational commitment affective commitment 152 1.00 5.00 3.422 middle 0.865 continuance commitment 152 1.00 5.00 3.235 middle 0.908 normative commitment 152 1.00 5.00 3.245 middle 0.869 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 124 4. validity and reliability to ensure the questionnaire statements validity for statistical analysis, specific tests have been conducted. to examine the internal consistency between the questionnaire's statements, the reliability test was used and cronbach’s alpha was calculated for those statements as a measurement of the study variables. this test results, displayed in table 5, indicated that the cronbach’s alpha values for all variables were greater than 0.70 (hair et al., 1998, sekaran and bougie, 2016, p290). these results have clearly confirmed the internal consistency between the questionnaire statements. in order to test the collinearity problem, a multicollinearity test was carried out and its results are portrayed by table 5. these test results have plainly indicated that all values of variance inflation factor (vif) of the independent variables were less than 10 and the tolerance value is greater than 0.1. these results assured that there is no collinearity problem related to the relationship between all corporate governance dimensions. table 5: reliability and collinearity test variables* reliability collinearity test cronbach’s alpha tolerance vif independent variables gc 0.842 0.654 1.530 tm 0.837 0.469 2.177 ce 0.760 0.462 2.162 td 0.822 0.536 1.865 rs 0.805 0.627 1.595 overall reliability 0.933 the affective commitment is a dependent variable dependent variables** ac 0.903 cc 0.894 nc 0.891 overall reliability 0.953 notes: * gc: compliance with the corporate governance code, tm: top management, ce: control environment, td: transparency and disclosure, rs: rights of shareholders and stakeholders, ac, cc, and nc: affective, continuance and normative commitment respectively. ** good cronbach's alpha is 0.70 and above. *** variance inflation factor (vif= 1/tolerance) must be less than 10, so the tolerance value should be greater than 0.10. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 125 regarding the construct validity, the adequacy of the questionnaire, those were tested using factor analysis loadings and the kaiser-meyer-olkin (kmo) test. these statistical analysis results portrayed by table 6 confirms that the statements factor loadings values for all variables were greater than 0.30, and kmo values were greater than 0.50 as it was ranged from 0.775 to 0.897. the results of bartlett’s test of sphericity have clearly indicated that, significant chisquare (0.000), and eigenvalues for all variables were greater than 1. therefore, based on the factor analysis results, the appropriateness of the data for further statistical analysis has been approved. it would be meaningful to mention that the results of construct validity reviled by this study were in consistent with those of other previous studies (hackett et al., 1994; allen and meyer, 1996). table 6. factor analysis and kmo of variables. items factor 1 extraction kmo* * independent variables (corporate governance dimensions) 1. compliance with the corporate governance code (a1-a4) the company encourages its employees to be aware about the guide to the rules of the corporate governance adopted by jordanian companies. 0.777 0.604 0.775 meetings have been held at the company to introduce and explain the corporate governance rules guide. 0.829 0.687 the company encourages the compliance with the guidelines of the corporate governance code. 0.874 0.764 it can be said that the corporate governance code is part of our corporate culture. 0.817 0.668 2. top management (the board of directors/management committee), (b1-b6) the company’s board of directors is distinguished by its representation of all company's stakeholders (owners, employees, local community, etc.). 0.686 0.470 0.779 the company's board of directors holds regular periodical meetings. 0.757 0.572 the senior management (board of directors/management committee) considers that it is its responsibilities to ensure working in a cooperative and integrated manner. 0,815 0.664 senior management (the board and management committee) works to serve all stakeholders (owners, employees, customers, and the local community). 0.828 0.686 the company (board of directors and management committee)) enjoys fair salaries and bonuses at all managerial levels. 0.723 0.523 the company relies on specialized committees to nominate and select members of senior management (board of directors management committee). 0.676 0.457 3. control environment (c1-c5) the company has an administrative unit (department or division) for internal control to ensure the effectiveness of the adopted control system. 0.688 0.473 0.780 the company is distinguished by its existed risk management stated policy and it’s periodically assessing of potential risks, with taking the necessary measures to deal with. 0.771 0.595 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 126 the company enjoys maintaining the compliance system and adhering to the instructions and decisions issued by all upper, middle and lower administrative levels. 0.800 0.639 the company has appointed an external auditor to ensure the principle of independency in auditing the company's business. 0.626 0.392 the company is highly interested in using a very effective way to study and evaluate any conflicts of interest in the nomination and appointment to the important administrative centers. 0.697 0.486 4. transparency and disclosure (d1-d5) the company issues annual financial and non-financial reports on its business results. 0.781 0.610 0.812 the company is committed to disclose its financial statements (budget, income statement, changes in its capital, and its financial position) to ensure transparency in its business. 0.770 0.592 the company discloses its non-financial evidence (board meetings attendance, issued punishments, discussions of performance reports, and potential risks facing the company) transparently. 0.771 0.595 the company announces its adoption for social responsibility programs and projects that serve the community and stakeholders dealing with the company. 0.811 0.658 the company provides the necessary data and information to all concerned authorities at the appropriate time by publishing reports, advertisements in newspapers and online. 0.698 0.488 5. rights of shareholders/ stakeholders (e1-e6) the company affords shareholders the right to vote in the general assembly meetings. 0.728 0.530 0.778 as far as i know that major shareholders do not interfere in the company's business. 0.416 0.173 the company provides opportunities for shareholders to participate in the nomination and appointment of board members. 0.813 0.660 shareholders participate positively in expanding the company's interest in stakeholders who are affected by the company's business. 0.810 0.657 the company adopts its own code of conduct to improve its ethical behavior. 0.709 0.503 the company encourages employees to disclose violations committed at the company . 0.799 0.639 three kinds of organizational commitment i. affective commitment (x1-x7) i totally believe that my own goals are in line with my company goals. 0.764 0.584 0.897 i feel that i am emotionally attached with my company. 0.795 0.663 my personal values are in line with the corporate values. 0.831 0.691 my work at the company allows me to use my talents and skills. 0.839 0.704 my work and my relationships at the company make me happy. 0.794 0.630 the organizational climate in the company leads me to prefer working in my company. 0.862 0.743 my colleagues notify me that i am an important person in the workgroup. 0.676 0.457 ii. continuance commitment (y1-y7) i think i will continue to work for the company for many years. 0.822 0.676 0.891 i think my life would be uncomfortable if i left my company 0.731 0.534 i see that my continued work depends on many important factors that are more important than the material gains in the company. 0.651 0.424 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 127 i tend to accept any job in order to continue working for my company. 0.730 0.533 i still belong to my company even if the working conditions and privileges are greater in other companies. 0.844 0.713 i will remain in my job, even if my colleagues prefer to work for other companies. 0.852 0.727 i see a desire to continue with my current job until retirement age. 0.826 0.682 iii. normative commitment (z1-z7) i have an ethical and human relationship with my company that keeps me going on with it. 0.716 0.513 0.865 the company had had a distinguished influence in shaping my provisional career. 0.763 0.581 i feel that i have a strong personal commitment to continue working for this company. 0.821 0.673 for me, the company is the best place to work. 0.860 0.740 my moral commitment toward my colleagues encourages me to stay at the company. 0.776 0.603 i will stay working at this company even if i lose some material benefits. 0.791 0.626 i prefer to stay in the company even if there are better opportunities to develop my career outside it. 0.725 0.526 organizational culture (mediator variable) (m1-m5) the company is characterized by an organizational culture based on cooperation and joint action for the goals of the company and not the goals of individuals. 0.758 0.574 0.781 company culture encourages change and acceptance of the risks arising from it. 0.766 0.585 organizational culture tends to be open to rapid developments in a competitive business environment. 0.814 0.663 the company is constantly developing its organizational culture by focusing on its social responsibility to serve the community. 0.763 0.582 the company has compatibility between its organizational culture and the values of the society in which it operates. 0.788 0.620 notes: * kaiser-meyer-olkin (kmo) as a measure of sampling adequacy. ** eigen values for all variables were greater than 1. the discriminant validity was used to distinguish between variables and there is no overlap between them. the correlation matrix was implemented to ensure that there is a differentiation between the variables. table 7 results have highlighted that the inter-correlations values were ranged from 0.319–0.791. these values were acceptable and confirmed that the variables do not overlap or are interrelated (hair et al., 2020, p265). najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 128 table 7. inter-variable correlations. 5. results the study aims at examining the impacts of the five dimensions of governance on organizational commitment. (the three types that were suggested by allen and mayer (1990, 1997)). systematic processing of this type of test imposes that a correlation between the variables of each hypothesis has to be specified by both, correlation coefficient (r) and determination coefficient (𝑹𝑹𝟐𝟐), supported by f-test value at a significance level of (p<0.05) to determine the level of generality and significance of a correlation model. by testing the first hypothesis (h1), it was expected that governance improves the affective commitment of employees, as it helps a company to operate according to some balancing principles between the company’s decision centers as well as between different stakeholders. a correlation analysis usually facilitates the determination of direction and intensity of a relationship between adopted variables. it can be noted from table 8, that the correlation coefficient value was (0.649), while a determination coefficient value was (0.442) with a calculated-f value of (21.304) which was greater than the tabulated-f at a significance level of (pvalue<0.05). these results confirm a positive relationship between the dimensions of corporate governance and emotional commitment. to determine the causal relationship between these variables, regression analysis was carried out. the results of this analysis have indicated that beta values were (0.133) for the compliance with the corporate governance code, (0.545) for top management, and (0.109) for the controlled environment all at (p<0.05) level of significance. by contrast, the beta values were negative for other two dimensions, transparency and disclosure (-0.067) and the rights of stakeholders (-0.029) at an insignificant level (pvalue>0.05). these results indicate a negative impact of these two dimensions on the affective commitment in jordanian companies. these results also, highlight the need for variables gc tm ce td rs ac cc nc gc 1 tm 0.542 1 ce 0.439 0.639 1 td 0.453 0.530 0.637 1 rs 0.437 0.560 0485 0.467 1 ac 0.434 0.636 0.469 0.339 0.357 1 cc 0.336 0.490 0.334 0.467 0.319 0.730 1 nc 0.360 0.493 0.355 0.339 0.356 0.732 0.791 1 legend: gc: compliance with the corporate governance code, tm: top management, ce: control environment, td: transparency and disclosure, rs: rights of shareholders and stakeholders, ac, cc, and nc: affective, continuance and normative commitment, respectively. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 129 focusing on transparency and disclosure as well as paying extra attention to the rights of shareholders and different stakeholders. (such as employees, customers, suppliers, and the society in which the company operates). the results discussed above confirmed the acceptance of the first three alternative hypotheses, which are; there is a statistically positive significant impact of the three dimensions of governance (compliance with the corporate governance code, top management, and control environment) on affective commitment. by contrast, the null hypotheses of the other two dimensions (transparency and disclosure and rights of shareholders and different stakeholders) were accepted. these results mean that the effect of both variables on emotional commitment, is considered insignificant and can be disregarded. this result can also indicate that the government still needs to allocate significant efforts to improve the conditions of the application practices of policies, rules and strategies especially those related to transparency and disclosure and the company's relationship with stakeholders. table 8. corporate governance and affective commitment. variables (independent) r r2 f sig beta t-statistic sig gc 0.649 0.442 21.304 sig 0.133 2.715 0.019 tm 0.545 5.873 0.000 ce 0.109 3.178 0.041 td -0.067 -0.775 0.440 rs -0.029 -0.362 0.718 the second hypothesis (h2) concerns to the impact of the governance dimensions on continence commitment (the perception of costs associated with leaving organization). the results exhibited in table 9, referred to a positive relationship between governance dimensions collectively and continence commitment, as the correlation coefficient value was (0.500). the determination coefficient value was (0.250) with a calculated-f of (9.741) that was greater than the tabulated-f at (p<05) level of significance. according to the regression analysis results it was found that, each governance dimension had a different effect on continence commitment. the four dimensions of governance (compliance with the corporate governance code, top management, control environment and rights of shareholders and different stakeholders) have a significant positive impact on continence commitment, as beta values were 0.102, 0.419, 0.042 and 0.435 respectively all at level of significance (p<0.05). the results also confirmed that there was a negative impact (-0.063) of transparency and disclosure on continence commitment but at an insignificant level. therefore, alternative hypotheses of the four dimensions should be accepted. in contrast, the null hypothesis is accepted, which asserts that the impact of transparency and disclosure on continence commitment is not significant and can be disregarded. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 130 table 9. hypothesis test (h2): dependent variable: continence commitment. variables (independent) r r2 f sig beta t-statistic sig gc 0.500 0.250 9.741 0.000 0.102 2.153 0.001 tm 0.419 3.963 0.000 ce 0.042 2.401 0.009 td -0.063 -0.646 0.519 rs 0.435 2.436 0.004 to test (h3), table 10 results showed that the coefficient of determination value was (0.25) and the calculated-f value was (10.65) at (p>0.05) significance level. these results approved the existence of a significant relationship between governance dimensions and normative commitment. the beta values were (0.126, 0.372, and 0.081) denoted that these three dimensions (compliance with the corporate governance code, top management and control environment), respectively, have a significant impact on the normative commitment. in turn, the beta values for the dimension of transparency and disclosure was (beta = -0.108), and for the dimension of the rights of shareholders and stakeholders was (0.103) all at (p>0.05) level of significance. based on these results, alternative hypotheses of the first three dimensions should be accepted. this is clearly meant that these dimensions have a statistically significant impact on normative commitment. for the other two dimensions (transparency and disclosure and the rights of shareholders and stakeholders) the null hypotheses are accepted as there were no statistically significant impacts for both. then based on the third and the first hypotheses testing results, it could be concluded that the two dimensions of transparency and disclosure and the rights of shareholders and stakeholders, need extra attentions and efforts to improve the governance of jordanian companies. table 10. hypothesis test (h3): dependent variable: normative commitment. variables (independent) r r2 f sig beta t-statistic sig gc 0.517 0.267 10.657 0.000 0.126 1.440 0.012 tm 0.372 3.556 0.001 ce 0.081 1.775 0.040 td -0.108 -1.111 0.268 rs 0.103 1.157 0.249 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 131 table 11. the three-step model (h4). step variables unstandardized coefficient b std. error beta t sig. 1. cg >>>> ocom 0.775 0.329 0.536 7.780 0.000 2. cg >>>> ocul 0.789 0.089 0.606 9.337 0.000 3. cg and ocul >>>> ocom 0.548 0.114 0.411 4.822 0.000 0.211 0.087 0.206 2.414 0.017 notes: cg = corporate governance, ocom = organizational commitment, ocul = organizational culture the second hypothesis (h4) addressed the impact of the mediator variable (organizational culture) on the relationship between corporate governance and organizational commitment. a mediation analysis usually, assumes that the independent variable affects the mediator variable, which in turn impacts the dependent variable (baron and kenny, 1986, p1173). to perform the testing of this hypothesis, three steps of single mediator model was used (baron and kenny, 1986; mackinnon, 2008, p49). table 11, displayed the results of these three steps; i. corporate governance affects the organizational commitment, ii. corporate governance affects organizational culture (as a mediator variable), and iii. both corporate governance and organizational culture affect the organizational commitment. the testing results indicated that the impact of corporate governance (independent variable) was (0.411) and for organizational culture (mediator variable) was (0.206) at significant level (p<0.05) on the organizational commitment. consequently, it can be concluded that there is a positive impact of organizational culture on the relationship between corporate governance and organizational commitment. to support this conclusion, the researchers used the sobel test, where its results are presented in table 12. these results found that the value of the sobel statistic was (2.262) at the significance level of (p<0.05), which confirmed the organizational culture positive impact on this relationship. table 12. sobel test. input test statistic std. error pvalue a 0.775 sobel test 2.56174176 0.09408638 0.01041487 b 0.211 aroian test 2.51433205 0.09586045 0.01192581 sa 0.329 goodman test 2.61193859 0.0922782 0.00900304 sb 0.087 notes: * as the p-value for sobel test > 0.05 there is no impact or effect for the mediating variable on the relation between the independent variable and the dependent variable. najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 132 6. discussion this study sought to determine the impact of the five dimensions of corporate governance on organizational commitment. the statistical analysis results of this study have confirmed a positive effect of some dimensions (compliance with the corporate governance code, top management, and control environment) on organizational commitment. other dimensions either have negative effects or insignificant effects (transparency and disclosure, rights of shareholders and stakeholders). the findings of several previous studies have confirmed a positive effect of corporate governance on organizational commitment (purwanto, 2015) and other important indicators of organizational performance such as market value (durnev and kim, 2005; klapper and love, 2004).), stock market prices (durnev and kim, 2005), and financial performance (mwangi, 2013), improving corporate reputation (argenti and druck, 2004). corporate governance dimensions are actually correlated with control environment and stakeholder’s relationship, which enhances company’s accountability and consolidates its legitimacy and reputation in their societies (zadek, 2006). the study also has highlighted the important role of a company’s control environment, while the relationship with stakeholders still needs more efforts to plainly explore its nature to find the relevant method for enhancing its role in developing corporate governance. in relation to top management and the structure of management, it was found that the results of the current study were consistent with some of the outcomes of shanikat and abbadi's study (2011) on the governance of jordanian companies. by contrast, the results of this study concerning transparency and disclosure and the relationship with shareholders and stakeholders were not compatible with that of shanikat and abbadi's study (2011), which emphasized the important role of disclosure and transparency. transparency and accountability have a positive impact on organizational commitment (purwanto, 2015). conversely, a lack of transparency can have an unfavorable effect. the results of the study confirmed the need for senior management to strengthen the principles and standards of transparency in administrative policies and practices in jordanian companies in order to address this negative impact on the organizational commitment of employees. corporate governance that maintains the protection of the rights of stakeholders, including workers, would lead to improve employee satisfaction and increase their organizational commitment. however, the results of the study affirmed an insignificant impact of stakeholders and shareholders' rights on organizational commitment. an objective interpretation of these results could be related to inequitable factors between shareholders and other stakeholders. these results also reflect the failure of the top management (the ceo and the board of directors) in jordanian companies in the balanced and fair treatment of the rights of all stakeholders. according to mrabure, and iyoha’s study (2020), effective boards of directors in addressing shareholder interests prove effective in securing the interests of the rest of the company's stakeholders. there is a need to change the policy of negative competition based on shareholders versus stakeholders (smith, 2003) to a policy of positive cooperation based on shareholders with najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 133 stakeholders. strong organizational commitment can be found in japanese companies because of the lifelong employment policy (kamesaka, 2019). it might be true to state that corporate governance would be appropriate with the implementation of that type of recruitment policy which would maintain employee’s commitment and loyalty to their companies. in short, these study outcomes have emphasized the need for improvement of transparency and disclosure policy by jordanian companies to increase the positive impact of governance on employee’s organizational commitment. 7. implications and limitations corporate governance still needs to be rooted in the development of a concept through studies that help understand its patterns, values, and mechanisms in government institutions and business companies. this study sought to determine the impact of corporate governance on organizational commitment in jordanian companies, and governance, as an increasingly important topic, can affect important areas such as social responsibility, corporate citizenship, and business ethics or on organizational performance standards such as reputation, and market share in jordanian companies. therefore, these topics could represent areas for future studies. also, digital governance is still a new field that requires new research initiatives. in terms of the limitations, this study focused on determining the impact of the five dimensions of corporate governance in jordanian companies, and future empirical studies could be based on comparison with other companies from other countries. according to the results of this study, the two important dimensions (transparency and disclosure, the rights of shareholders and stakeholders) need more depth in case studies or comparisons between jordanian companies in different sectors. references adebayo, m. bakare, i. a. o. yusuf, b. & ishmael, o. (2014). good corporate governance and organisational performance: an empirical analysis, international journal of humanities and social science, vol. 4, no. 7(1), 170-178. agrawal, a. & chadha, s. (2005). corporate governance and accounting scandals, the journal of law and economics, vol 48, no. 2, 371-406. https://doi.org/10.1086/430808 alam, a. (2011). evaluation of allen and meyer's organizational commitment scale: a crosscultural application in pakistan, journal of education and vocational research, vol. 1, no. 3, pp. 80-86. https://doi.org/10.22610/jevr.v1i3.13 allen, n.j. & meyer, j.p. (1990): the measurement and antecedents of affective, continuance and normative commitment to the company, journal of occupational psychology, vol. 63 no. 1, pp. 1-18. https://doi.org/10.1111/j.2044-8325.1990.tb00506.x allen, n.j., & meyer, j.p. (1996). affective, continuance, and normative commitment to the organization: an examination of construct validity. journal of vocational behavior, 49, 252276. https://doi.org/10.1006/jvbe.1996.0043 altug, s. g. (2010). business cycles: fact, fallacy, and fantasy, new jersey: world scientific publishing co. https://doi.org/10.1142/6934 https://doi.org/10.1086/430808 https://doi.org/10.22610/jevr.v1i3.13 https://doi.org/10.1111/j.2044-8325.1990.tb00506.x https://doi.org/10.1006/jvbe.1996.0043 https://doi.org/10.1142/6934 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 134 aluchna, m. (2009). applying corporate governance in europe, in: aras, g. crowther, d. and aras, g. (eds). global perspectives on corporate governance and csr, pp 47-72, surrey: gower publishing limited. andreas r. (2010), collaborative governance, corporate governance, vol. 10, no. 4, 500-511. https://doi.org/10.1108/14720701011069713 baer, m. (2013). international corporations as actors in global governance, cologne: springer fachmedien wiesbaden. https://doi.org/10.1007/978-3-658-00406-4 barnes, j. jackson, d. w. utt, m. d. & kumar, a. (2006). the role of culture strength in shaping sales force outcomes," journal of personal setting and sales management, vol 26, n0, 3, 255-269. https://doi.org/10.2753/pss0885-3134260301 barns, s. (2016). mine your data: open data, digital strategies and entrepreneurial governance by code, urban geography, april, routledge, 1-18. https://doi.org/10.1080 /02723638.2016.1139876 basco, s. (2018). housing bubbles: origins and consequences, cham: palgrave macmillan. https://doi.org/10.1007/978-3-030-00587-0 bellingham, r. (2003), ethical leadership, amherst: hrd press, inc. bertelli, a. m. (2012). the political economy of public sector governance, new york: cambridge university press. berthelot, s. morris, t. & morrill, c. (2010). corporate governance rating and financial performance: a canadian study, corporate governance, vol 10, no, 5, 635-646. https://doi.org/10.1108/14720701011085599 cameron, k & quinn, r. (2006). diagnosing and changing organizational culture: based on the competing values framework. san francisco: jossey-bass. campbell, s. m. r., (2013): the applicability of commitment models in a unionized professional workplace, international journal of organizational analysis, vol. 21 iss 4 pp. 488 – 503. https://doi.org/10.1108/ijoa-06-2012-0591 cho, vincent & huang, xu, (2012): professional commitment, organizational commitment, and the intention to leave for professional advancement, information technology & people, vol. 25 iss 1 pp. 31 54. https://doi.org/10.1108/09593841211204335 companies control department (2009). jordanian corporate governance code, available at: https://ecgi.global (retrieved jan 3, 2020). daley, g. a. levy, d. g. kaganoff, t. augustine, c. a. benjamin, r. bikson, t. k. gates, s. m. moini, j. s. (2003). a strategic governance review for multi-organizational systems of education, training, and professional development, rand. de villiers, c. & dimes, r. (2021). determinants, mechanisms and consequences of corporate governance reporting: a research framework, journal of management and governance,vol 25, 7-26. https://doi.org/10.1007/s10997-020-09530-0 dignam, a. & lowry, j. (2006), corporate governance 1: corporate governance and corporate theory'', company law, core text series, new york: oxford university press. dimsdale, n. & prevezer, m. (1994), capital markets and corporate governance, oxford: clarendon press. https://doi.org/10.1093/acprof:oso/9780198287889.001.0001 dobbin, f. & jung, j. (2010). the misapplication of mr. michael jensen: how agency theory brought down the economy and why it might again, research in the sociology of organization, vol 30b, 29-64. https://doi.org/10.1108/s0733-558x(2010)000030b006 drumaux, a. & joyce, p. (2018). strategic management for public governance in europe, london: palgrave macmillan. https://doi.org/10.1057/978-1-137-54764-4 du plessis, j. j. hargovan, a. & bagaric, m. (2011). principles of contemporary corporate governance, cambridge: cambridge university press. durnev, artyom, and e. han kim. 2005. ''to steal or not to steal: firm attributes, legal environment, and valuation, 60, journal of finance, 1461-93. https://doi.org/10.1111/j.15406261.2005.00767.x eldomiaty, t. t. and choi. c. j. (2006). corporate governance and strategic transparency: east asia in the international business systems, corporate governance, vol 6, no. 3, 281-295. https://doi.org/10.1108/14720700610671882 fahy, m. r. j. and weiner, a. (2004). beyond governance, west sussex: john wiley & sons ltd, finkelstein, l.s. (1995), ''what is global governance?, global governance, vol. 1 no. 3, 367-72. https://doi.org/10.1163/19426720-001-03-90000007 gilson, r. j. (2018). from corporate law to corporate governance in: gordon, j. n. and ringe, w-g (eds). the oxford handbook of corporate law and governance, pp 1-30, oxford: oxford university press. https://doi.org/10.1108/14720701011069713 https://doi.org/10.1007/978-3-658-00406-4 https://doi.org/10.2753/pss0885-3134260301 https://doi.org/10.1080/02723638.2016.1139876 https://doi.org/10.1080/02723638.2016.1139876 https://doi.org/10.1007/978-3-030-00587-0 https://doi.org/10.1108/14720701011085599 https://doi.org/10.1108/ijoa-06-2012-0591 https://doi.org/10.1108/09593841211204335 https://ecgi.global/ https://doi.org/10.1007/s10997-020-09530-0 https://doi.org/10.1093/acprof:oso/9780198287889.001.0001 https://doi.org/10.1108/s0733-558x(2010)000030b006 https://doi.org/10.1057/978-1-137-54764-4 https://doi.org/10.1111/j.1540-6261.2005.00767.x https://doi.org/10.1111/j.1540-6261.2005.00767.x https://doi.org/10.1108/14720700610671882 https://doi.org/10.1163/19426720-001-03-90000007 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 135 grembergen, w. v. (2004). strategies for information technology governance, london: idea group publishing. https://doi.org/10.4018/978-1-59140-140-7 hackett, r. d., bycio, p., & hausdorf, p. a. (1994). further assessments of meyer and allen's (1991) three-component model of organizational commitment. journal of applied psychology, 79, 15-23. https://doi.org/10.1037/0021-9010.79.1.15 hair, j. f. page, m. and brunsveld, n. (2020). essentials of business research methods, new york: routledge. https://doi.org/10.4324/9780429203374 handley-schachler, m. juleff, l. and paton, c. (2007). corporate governance in the financial services sector. corporate governance, vol. 7, no. 5, 623-634. https://doi.org/10.1108/ 14720700710827202 haspeslagh, p. (2010), corporate governance and the current crisis, corporate governance, vol 10, no 4, 375-377. https://doi.org/10.1108/14720701011069614 hemphill, t. (2010). the creative capitalism: corporate governance model: how radical an approach to modern capitalism?, international journal of law and management, vol. 52 no. 2, 110-123. https://doi.org/10.1108/17542431011029415 heneman, h. g & judge, t. a. (2009), staffing organizations, new york: mcgraw hill international edition. heracleous, l. 2001. impact of corporate governance on organizational performance. mckinsey quarterly, vol 2 (20), 20-23. hertie school of governance and oecd (2016). the governance report 2016, oxford: oxford university press. hogler, r. and gross, m. a. & byrne, z. s. (2009). making governance work in academia, international journal of conflict management, vol 20, no. 1, 31-45. https://doi.org/10.1108/10444060910931594 hoopes, j. (2003). false prophets, cambridge: perseus publishing. iqbal, a. (2010). employee turnover: causes, consequences and retention strategies in the saudi organizations, the business review, cambridge, vol. 16, no. 2, 275-281. jaros, s. (2007). meyer and allen model of organizational commitment: measurement issues, the icfai journal of organizational behavior, vol. vi, no. 4, 7-25. jensen, m. (1993). the modern industrial revolution: exit and failure of internal control systems, journal of finance, vol 48, 931-880. https://doi.org/10.1111/j.1540-6261.1993.tb04022.x kamesaka, a. (2019). governance, risk and financial impact of mega disasters: lessons from japan, singapore: springer nature singapore pte ltd. https://doi.org/10.1007/978-981-139005-0 keohane, r.o. & nye, j.s. (2000), introduction'', in nye, j.s. and donahue, j.d. (eds), governance in a globalizing world, pp. 1-41, washington, dc, brookings institution press, khanifar, h.; hajlou , m. h., abdolhosseini b.; ataei, f. & soltani h. (2013): factors affecting the organizational commitment of employees and customer satisfaction, journal of basic and applied scientific research, 2 (11), pp. 11180-11185. khiari, w. karaa, a. & omri, a. (2007). corporate governance efficiency: an indexing approach using the stochastic frontier analysis, corporate governance, vol. 7, no, 2, 148-161. https://doi.org/10.1108/14720700710739796 klapper, l.f. & love, i. (2024), 'corporate governance, investor protection and performance in emerging markets'', journal of corporate finance, no. 10, 203-10. lakshana, a. & wijekoonb, w. (2012). corporate governance and corporate failure, procedia economics and finance, vol 2, 191 198. https://doi.org/10.1016/s2212-5671(12)00079-2 li, j. & harrison, j. r. (2008), corporate governance and national culture: a multi-country study, corporate governance, vol. 8, no. 5 607-621. https://doi.org/10.1108/14720700810913278 liao, s-h, retno, w. r. hu c, d-c & tasman, l.(2017). the mediating effect of psychological contract in the relationships between paternalistic leadership and turnover intention for foreign workers in taiwan, asia pacific management review, 22, 80-87. https://doi.org/10.1016/j.apmrv.2016.08.003 lin, d. kuo, h-c, and wang, l.h, (2013). chief executive compensation: an empirical study of fat cat ceos,the international journal of business and finance research, vol 7, no. 2, 2742. mcnutt, p. a. & batho, c. a. (2005)..code of ethics and employee governance, international journal of social economics. vol. 32 no. 8, 656-666. https://doi.org/10.1108/ 03068290510608183 https://doi.org/10.4018/978-1-59140-140-7 https://doi.org/10.1037/0021-9010.79.1.15 https://doi.org/10.4324/9780429203374 https://doi.org/10.1108/14720700710827202 https://doi.org/10.1108/14720700710827202 https://doi.org/10.1108/14720701011069614 https://doi.org/10.1108/17542431011029415 https://doi.org/10.1108/10444060910931594 https://doi.org/10.1111/j.1540-6261.1993.tb04022.x https://doi.org/10.1007/978-981-13-9005-0 https://doi.org/10.1007/978-981-13-9005-0 https://doi.org/10.1108/14720700710739796 https://doi.org/10.1016/s2212-5671(12)00079-2 https://doi.org/10.1108/14720700810913278 https://doi.org/10.1016/j.apmrv.2016.08.003 https://doi.org/10.1108/03068290510608183 https://doi.org/10.1108/03068290510608183 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 136 meyer, j. p., allen, n. j. & smith, c. a. (1993). commitment to organizations and occupation: extensions and test of a three-component conceptualization. journal of applied psychology, 78, 538-551. https://doi.org/10.1037/0021-9010.78.4.538 monks, r. a. g. & minow, n. (2004). corporate governance, malden: blackwell publishing ltd. mosadeghrad a. m., ferlie, e., & rosenberg, d (2008) "a study of the relationship between job satisfaction, organizational commitment and turnover intention among hospital employees", health services management research, 21, pp. 211227. https://doi.org/10.1258/hsmr.2007.007015 mostovicz, e. i. kakabadse, k. n. and kakabadse, a. (2011), corporate governance, corporate governance, vol. 11, no. 5, 613-626. https://doi.org/10.1108/14720701111177019 mrabure, k. o. & iyoha, a. a. (2020)..corporate governance and protection of stakeholders rights and interests, beijing law review, vol 11, 292-308. https://doi.org/10.4236/blr.2020.111020 muneeza, a. & hassan, r. (2014).shari'ah corporate governance: the need for a special governance code, corporate governance, vol. 14 no. 1,120-129. https://doi.org/10.1108/cg02-2011-0015 oecd (1999), principles of corporate governance, organization for economic co-operation and development, available at: http:thecorporatelibrary.com/docs/oecd/oecdprinciples.html (retrieved jan 15, 2020). oecd (2004). the oecd principles of corporate governance, paris: oecd publications service. https://doi.org/10.22201/fca.24488410e.2005.562 oecd (2019), oecd corporate governance fact book 2019, available at: http://www.oecd.org/corporate/corporate-governance-factbook.htm ortega-parra, a. & sastre-castillo, m. a (2013): impact of perceived corporate culture on organizational commitment, management decision, vol. 51 iss 5 pp. 1071-1083. https://doi.org/10.1108/md-08-2012-0599 osborne, s. (2010). the new public governance? emerging perspectives on the theory and practice of public governance. london: routledge. https://doi.org/10.4324/9780203861684 pattberg, p. (2005). the institutionalization of private governance: how business and nonprofit organizations agree on transnational rules, governance: an international journal of policy, administration, and institutions, vol. 18, no. 4, 589-610. https://doi.org/10.1111/j.14680491.2005.00293.x pavel, k. stanislav,t. petr, p. & pavel, p. (2012).corporate governance against recommendations: the cases of the strong executive and the strong ownership, journal of competitiveness, vol. 4, issue 3, 46-57. https://doi.org/10.7441/joc.2012.03.04 pozner, j.e, greve, h.r, & palmer, d. (2010). organisations gone wild: the causes, processes and consequences of organisational misconduct. the academy of management, vol.4, no. 1, 53-107. https://doi.org/10.5465/19416521003654186 purwanto, s.k. (2015). the effect of organizational governance on the performance and commitment of the lecturers, public policy and administration research, vol.5, no.1, 35-42. raipa, a. (2010). methodological aspects of the democratization of public governance, public policy and administration, 32: 7 -18. rampersad, h. & hussain, s. (2014). authentic governance, switzerland: springer international publishing. https://doi.org/10.1007/978-3-319-03113-2 rasche, a. (2010), collaborative governance, corporate governance, vol. 10, no. 4, 500-511. https://doi.org/10.1108/14720701011069713 rosenau, j.n. (1992), ''governance, order, and change in world politics'', in rosenau, j.n. and czempiel, e.-o. (eds), governance without government: order and change in world politics, pp 1-29, cambridge: cambridge university press. https://doi.org/10.1017/ cbo9780511521775.003 ruhanen, l. scott, n. ritchie, b. & tkaczynski, a. (2010), governance: a review and synthesis of the literature, tourism review, vol. 65 iss 4, 4-16. https://doi.org/10.1108/16605371011093836 sekaran, u. & bougie, r. (2016). research methods for business, west sussex: john wiley & sons ltd. https://doi.org/10.1037/0021-9010.78.4.538 https://doi.org/10.1258/hsmr.2007.007015 https://doi.org/10.1108/14720701111177019 https://doi.org/10.4236/blr.2020.111020 https://doi.org/10.1108/cg-02-2011-0015 https://doi.org/10.1108/cg-02-2011-0015 https://doi.org/10.22201/fca.24488410e.2005.562 http://www.oecd.org/corporate/corporate-governance-factbook.htm https://doi.org/10.1108/md-08-2012-0599 https://doi.org/10.1108/md-08-2012-0599 https://doi.org/10.4324/9780203861684 https://doi.org/10.1111/j.1468-0491.2005.00293.x https://doi.org/10.1111/j.1468-0491.2005.00293.x https://doi.org/10.7441/joc.2012.03.04 https://doi.org/10.5465/19416521003654186 https://doi.org/10.1007/978-3-319-03113-2 https://doi.org/10.1108/14720701011069713 https://doi.org/10.1017/cbo9780511521775.003 https://doi.org/10.1017/cbo9780511521775.003 https://doi.org/10.1108/16605371011093836 najm a. najm et al. / european journal of government and economics 11(1), june 2022, 113-137 137 shanikat, m.& abbadi, s. s. (2011)..assessment of corporate governance in jordan: an empirical study, australasian accounting, business and finance journal, volume 5, issue 3, 93-106. shleifer, a. & vishny, r. (1997) a survey of corporate governance." the journal of finance, vol 52, iss 2, 737-783. https://doi.org/10.1111/j.1540-6261.1997.tb04820.x singh, a. & gupta, b., (2015): job involvement, organizational commitment, professional commitment, and team commitment", benchmarking: an international journal, vol. 22, iss 6, 1192-1211. https://doi.org/10.1108/bij-01-2014-0007 singh, a. & gupta, b., (2015). job involvement, organizational commitment, professional commitment, and team commitment, benchmarking: an international journal, vol. 22 iss 6, 1192-1211. https://doi.org/10.1108/bij-01-2014-0007 smith, d. (1996). increasing employee productivity, job satisfaction, and organizational commitment, hospital health service administration, 41, 160-174. smith, h. j. (2003). the shareholders vs. stakeholders debate. mit sloan management review, vol 44, 85-90. solinger, o. n.; van olffen, w. & roe, r. a. (2007). beyond the three-component model of organizational commitment, maastricht research school of economics of technology and organizations, 1-43. https://doi.org/10.1037/0021-9010.93.1.70 stringham, e. p. (2015). private governance: creating order in economic and social life, new york: oxford university press. https://doi.org/10.1093/acprof:oso/9780199365166.001.0001 talamo, g. (2011). corporate governance and capital flows, corporate governance, vol 11, no. 3, 228-243. https://doi.org/10.1108/14720701111138661 torfing, j. & triantafillou, p. (2016). enhancing public innovation by transforming public governance, new york: cambridge university press. https://doi.org/10.1017/cbo9781316105337 weber. r. h. grosz, m. & weber, r. (2009). shaping internet governance: regulatory challenges, heidelberg: springer. https://doi.org/10.1007/978-3-642-04620-9 wiewiora, a. trigunarsyah, b. murphy, g. & coffey, v. (2013). organizational culture and willingness to share knowledge: a competing values perspective in australian context, international journal of project management 31, 1163-1174. https://doi.org/10.1016/j.ijproman.2012.12.014 zadek, s. (2008). global collaborative governance: there is no alternative', corporate governance, vol. 8 no. 4, pp. 374-88. https://doi.org/10.1108/14720700810899121 zoheir ezziane, z. & al shamisi, a. (2017). the impact of it governance compliance on enhancing organizational performance in: haes, s. d. and grembergen, w. v. (eds). strategic it governance and alignment in business settings, pp 209-236, hershey: igi global. https://doi.org/10.4018/978-1-5225-0861-8.ch008 https://doi.org/10.1111/j.1540-6261.1997.tb04820.x https://doi.org/10.1108/bij-01-2014-0007 https://doi.org/10.1108/bij-01-2014-0007 https://doi.org/10.1037/0021-9010.93.1.70 https://doi.org/10.1093/acprof:oso/9780199365166.001.0001 https://doi.org/10.1108/14720701111138661 https://doi.org/10.1017/cbo9781316105337 https://doi.org/10.1007/978-3-642-04620-9 https://doi.org/10.1016/j.ijproman.2012.12.014 https://doi.org/10.1108/14720700810899121 https://doi.org/10.4018/978-1-5225-0861-8.ch008 1. introduction 2. corporate governance and organizational commitment 3. method 4. validity and reliability 5. results 6. discussion 7. implications and limitations references european journal of government and economics 11(2), december 2022, 141-166 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 determinants of bilateral current account balance between the eurozone and the united states gerassimos bertsatos a, * a university of western macedonia, laboratory of applied economics, department of economics. kastoria campus, fourka area 52100, kastoria – greece. * corresponding author at: decon00001@uowm.gr abstract. longand short-run current account balance (cab) determinants of the nineteen eurozone (ez) member states vis-a-vis the united states (us) are examined. particularly, the competitiveness of the ez vs the us, the relationship between the current account deficit and the budget deficit (twin deficit), and other factors determining the current account balance are studied. quarterly data was used in a sample of the nineteen ez member states with the us as a trading partner over the period 2008 2018. it is found that the cab in the long run has a positive relationship with the real interest rate, real exchange rate, gross domestic product (gdp) per capita, and exchange rate volatility, but a negative relationship with the fiscal balance. in the short run, it is notable that only the real exchange rate affects the current account balance. finally, policy implications are discussed regarding the determinants of the current account. keywords. current account determinants, economic policy, pmg method, eurozone, usa jel codes. f30, f32, f40, f41 doi. https://doi.org/10.17979/ejge.2022.11.2.8925 1. introduction recognizing the current account determinants is a matter of great significance in exercising economic policy, as marked by a plethora of papers in the relevant literature. this work examines the current account determinants in the bilateral trade relations between the eurozone (ez) countries and the united states (us). the economic/trading relationship between the us and the ez in terms of trade volume is the greatest in the world and has grown more complicated over the years, encompassing an expanding quantity and variety of trade and financial operations that weave the economies into an increasingly interdependent partnership. the ez-us trade relations are affecting the gross domestic product (gdp) of both the us and the ez economies and, therefore, are in the spotlight of policymakers on both sides. apart from the bilateral trade relations, the ez member states and the us are leading members of the world trade organization (wto), the international monetary fund (imf), and the organization for economic cooperation and development (oecd). they both play crucial roles in establishing and implementing the goals of these institutions. the us and the ez are the largest economies in the world in terms of the gdp and volume of bilateral trade. their combined population in 2020 approached 800 million people, generating a https://creativecommons.org/licenses/by-nc/4.0/ mailto:decon00001@uowm.gr https://doi.org/10.17979/ejge.2022.11.2.8925 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 142 gdp that accounted for more than 40% of the global gdp. the combined ez-us world trade accounts for roughly 47% of total global trade, eurostat (2021). furthermore, they both have nearly identical levels of economic development and are among the most advanced in the world. they have the world's wealthiest and most educated populations. with a few exceptions, the us, and the ez are significant producers of innovative technology and services. as a result, most of the trade between the us and the eu is intra-industry (stöllinger 2020); that is, trade in similar commodities. from the above short description of the importance of the ez-us trade relations, it is evident that studying their determinants provides useful tools for exercising trade policy. identifying the determinants of the bilateral current account, in both the short and long run, provides the essential tools for forecasting real exchange rate policy, evaluating the effects of regime action on the open macro-economy, and determining the sustainability of current policies and the wealth of the two economies. in the present work, the bilateral relationship between the ez and the us and the determinants which affect their cab are examined. this work is novel because bilateral current account studies are scarce in the literature. there are numerous studies on the determinants of the trade balance of a country, vis a vis the rest of the world and the bilateral trade balance between individual countries. however, there are very few studies of the determinants of the bilateral cab mainly due to data unavailability in the trade in services and the other balances that form the current account, apart from the trade balance; some rare examples of bilateral current account studies are dettmann et al. (2012) and iqbal et al. (2017). disaggregate bilateral data for all balances formulating cab was made available only recently by a combination of sources (eurostat (2020), the oecd (2020), and the european central bank (ecb) (2020)). using quarterly panel data analysis, the effect of factors recognized in the literature as the primary determinants of the cab is examined in the trade between the 19 ez member states and the us from 2008 to 2018. these factors are real exchange rate, governmental fiscal balance, real interest rate, per capita gdp, and exchange rate volatility. the model was estimated using a panel autoregressive distributed lags (ardl) method, the pooled mean group (pmg). the article is arranged as follows: section 2 examines the literature on current account determinants; section 3 describes the empirical methodology and the data used, section 4 provides the results and section 5 summarizes the findings and lists policy implications. 2. literature review there is a growing literature on current account balance determinants since the 80s. some empirical studies examine the determinants of the current account to determine the extent of the current account that may be considered ‘normal’ for a country based on a set of structural and macroeconomic attributes. variables such as gdp per capita, demographics, fiscal balance, and initial net foreign assets are examples of economic essentials. the primary research on shortgerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 143 term current account variations is based on the concept that current account functions as a buffer against transient income shocks, smoothing consumption, and maximizing welfare. ghosh and ostry (1995), glick and rogoff (1995), kraay and ventura (2000), and nason and rogers (2006), completed the most significant of this early research in this field. the existence of various approaches with diverging estimates and variable selection to comprehend which aspects have a suitable role for determining cab led the topic to a multidimensional foundation. in theory, in equilibrium, the cab ought to be zero; however, when it comes to the real world, this is doubtful. in the case where a surplus or a deficit occurs in the current account, this signals something regarding the situation where the economy is in, both on its own and in comparison with the trading country. the imbalances in the current account, in the framework of income convergence, were studied by herrman and winker (2008) for the emerging economies in europe and asia. the authors stated that the progression of the financial market and financial integration were important factors in determining the current account balance. in the same notion, a more advanced financial market and financial integration might result in bigger deficits and lower surpluses. this scenario is possible because, as the convergence process progresses, countries with developed financial markets and integration may borrow more freely from abroad, resulting in higher domestic consumption and savings. yet, several financial integration and development indicators are insufficient to fully explain the diverse models of current account and real convergence in europe and asia’s growing countries. in the spirit of empirical application, the methodology used in the current work falls on a class currently comprised of many analyses that employ well-established econometric methodologies that examine the relationship between a wide range of macroeconomic factors and the cab, such as debelle and faruquee (1996); blanchard and giavazzi (2002); chinn and prasad (2003); herrmann and jochem (2005); ca'zorzi et al. (2009); nieminen (2015); das (2016); yoshida and zhai (2020). the difference between this study to the above literature is that the empirical studies above examine the current account determinants of a country with respect to the rest of the world or a group of countries and not bilaterally, while in the present study, the determinants of the ca are examined in the bilateral trade of the two largest economies in the world, the ez and the us. this work comprises a selection of some key determinants that affect the cab in the bilateral relationship between the eurozone and the united states. the first determinant identified in the literature as a key factor for the current account flows is the real exchange rate (rer). the net export component of the current account is commonly shown as a function of competitiveness, rer, and certain other external factors. fluctuations in the currency rate influence the balance of the current account. a currency depreciation, for example, is anticipated to rise the current account and lower the deficit. previously, if the exchange rate falls, the foreign price of this country's net exports will fall. this phenomenon will give the impression that the nation is more competitive and subsequently, an upsurge in the number of exports will occur. depreciation causes an upsurge in the current account deficit, which increases the value of gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 144 exports since export demand is relatively elastic. (kandil, 2009; purwono et al., 2018). a depreciation in the exchange rate, alternatively, will raise the cost of purchasing imported products. therefore, there will be less demand for imported goods, which will assist to minimize the current account deficit. hence, as the theory suggests, a depreciation in the exchange rate increases cab and vice versa1. calderon et al. (2007) conducted empirical research regarding the current account in emerging countries. the authors identified a negative association between the current deficit, a decline in terms of trade, and an appreciation of the rer. furthermore, prat et al. (2010), while assessing the current account balance for emerging countries as well, came across that the budget balance has a significant impact on the current account, and that this is enhanced by an increase in net foreign assets. falk (2008) concluded that a depreciation of the real effective exchange rate indicates that the trade balance is enchanted; in nations with a large positive net foreign investment holding and a negative trade balance, the trade balance is less responsive to fluctuations in the real effective exchange rate. the second determinant of the current account included in the model is the budget deficit. a budget deficit or fiscal deficit emerges as soon as a nation's expenditure surpasses its revenues. almost every year, the united states has been running fiscal deficits for decades past. naturally, a fiscal deficit does not seem like a pleasant situation for governments. but then again, arguments from the economists who support the keynesian view propose that deficits are not always damaging to an economy. nonetheless, deficit spending can be a favorite instrument for kickstarting a non-productive economy. in the literature, the link between the budget deficit and the current account deficit is one of the most controversial issues. the conclusions drawn from this relationship guide policymakers in regulating the preferred policy and the economic policy to be implemented. there are two approaches to analyzing the link between budget deficit and current account deficit: 1) the keynesian view, in which the budget deficit influences the deficit of the current account. in further depth, a causal link exists between the budget deficit and the current account deficit. if this relationship is positive, the twin deficit hypothesis can be used to explain it. 2) the ricardian equivalence, which asserts that the two deficits are uncorrelated. under the rational expectations hypothesis, an upsurge in public expenditure funded by issuing bonds induces a rise in savings to pay for future tax increases to repay the current increase in borrowing. in osoro’s et al. (2014) study, an investigation into the twin deficit hypothesis has been done, as well as a statistical analysis of the link between the budget and current account balances, as well as major macroeconomic variables for the republic of kenya for 196 years. the budget deficit and the current account deficit have a positive and substantial connection, according to empirical findings. the cointegrating slope coefficient's sign indicates a positive link between interest rates 1 this holds under the bickerdike-robinson-metzler condition of trade balance improvements after a currency depreciation. import-export demand and supply elasticities, as well as initial volume of trade, are said to affect changes in the foreign currency value of the trade balance. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 145 and the current account deficit, as well as a negative relationship between gdp and the money supply. this indicates that, in the long run, the current account deficit, combined with increases in the budget deficit, gdp, and interest rates, will increase the money supply. in their research, aloryito et al. (2016) inspected the twin deficit hypothesis considering countries in sub-saharan africa countries (ssa). the authors discovered that the deficit appeared to spread over the past decade in contrast with the positive output development. in this work data from 41 countries were examined for the era between 2000 and 2012, with the use of the generalized method of moments (gmm) estimation method, the foremost discovers specify that government deficits tend to enhance the current account and vice versa, in doing so the null hypothesis was rejected in favor of the twin hypothesis. a further determinant of the cab is exchange rate volatility (erv). it affects actual inflation as well as prospects about imminent price volatility (baharumshah, 2001). movements in the exchange rate tend to have an impact on the domestic prices of imported goods and services directly. erv, on the contrary, can have a similar impact on a country's cab through the impact it has on foreign trade. the unexpected movements in the exchange rate are known as exchange rate volatility. erv is the source of exchange rate risk and has repercussions for the volume of trade balance and, as a result, on the current account. increased exchange-rate volatility leads to higher costs for risk-averse traders and reduces foreign trade. if exchange rate fluctuations become unexpected, it generates uncertainty about potential earnings and, as a result, lowers the advantages of international trade. according to (de grauwe, 1988), the impact of exchange-rate uncertainty on exports should be proportional to the degree of risk aversion. that is, if exporters are risk averse, an increase in erv enhances the expected marginal utility of export revenue, motivating them to increase exports. using panel data analysis and the fuzzy approach2, nuroglu and kunst (2012) studied the influence of erv on the trade balance. using the gravity model of international economics, the authors examined the bilateral trade between the eu-15 nations. the estimated coefficient was proved to be negative, indicating that erv has a negative effect on bilateral trade flows. generally, the literature on the association between erv and cab is lacking. the research on the current account is limited, but at the same time, it is vast concerning the link between erv and trade balance. a recent study by purwono et al. (2018) analyzed the impact of erv on the current account deficit. the authors used data from 2005 to 2011, and for the empirical part, they used a simultaneous model of indonesia's current account deficit. depreciation, according to the simulation results, increases the surplus to the current account deficit. other than oil and gas, the decrease in export manufactured goods was higher than the increase in imports. furthermore, in literature, the relationship between interest rates and the current account argues that a rise in real interest rate reduces investment and increases savings, and as a result, the cab is improved. but the findings do not conclude that such a relationship exists. the 2 the method is particularly suited for ill-defined systems in which there is substantial ambiguity regarding the type and range of important input variables as well as the model’s underlying connections. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 146 intertemporal model by obstfeld and rogoff (1995) inspects the bond between the current account and interest rates; they discovered that fluctuations occurring in the current account are related negatively to interest rates if transaction costs are present, under the assumption of perfect capital mobility. bergin and sheffrin (2000) and bernhardsen (2000) found a positive slope coefficient between the current account and interest rates, indicating that an increase in the real interest rate caused a surge in the cab. additionally, anoruo and elike (2008) employed data from india, korea, thailand, and the philippines to survey the asymmetric bond between interest rates and current accounts, with the use of cointegration analysis and nonlinear unit root test. the outcomes of their study showed that positive shocks in the interest rates have a positive impact on the fluctuations of the current account for korea, the philippines, and india. but then again, interest rates for the case of thailand display a negative slope coefficient for the current account. the final, variable included in the model is the per capita income. the latter is not a determinant of the current account directly but is a key determinant of the trade balance, which constitutes the larger sub-balance of the current account. thus, the theory predicts that as gdp per capita increases, the purchasing power of the domestic agents increases as well. as a result, the domestic agents are now capable of purchasing domestic and imported goods more easily, which deteriorates the trade balance through the increase in the value of imports. falk (2008) conducted a study that focused on data from 32 developing and developed economies from 1990 to 2007. in his analysis, the author used linear mixed models and fixed effects in a panel data background to demonstrate that the trading partners' real foreign gdp per capita positively affects the balance of trade. the analysis also indicated that real domestic gdp per capita had a negative influence on the balance of trade. iyke and ho (2017) tested the outcome of fluctuations in the real exchange rate towards ghana’s trade balance by handling quarterly data for the era between 1986 and 2016 with the use of linear and non-linear specifications methods. this research found proof of an asymmetric influence of the exchange rate on the trade balance. at last, this research found the support of a j-curve effect and later disclosed that the per capita gdp and the per capita gdp of the partner country play a major roll-on ghana’s trade balance. 3. methodology and data 3.1 model description the model developed for examining the cab determinants in the bilateral relationship between the ez and the us employs variables recognized in the literature as the main determinants of the cab. these are the real exchange rate (rer), the budget deficit (bud), the real interest rate (rir), gdp per capita in ppps (gdp), and exchange rate volatility (erv). since our data is quarterly and seasonally unadjusted, three dummy variables for capturing seasonality (s1, s2, s3) were included. finally, another dummy variable was included demonstrating the year of the gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 147 entrance of each country in the eurozone (entry), this variable was included since each country was obligated to change its national currency to enter the eurozone and this might have a structural break effect on its current account panel data analysis has been employed for the estimation of the model 𝐶𝐶𝐶𝐶𝐵𝐵𝑖𝑖,𝑡𝑡 = 𝑓𝑓(𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−𝑠𝑠,𝐵𝐵𝐵𝐵𝐷𝐷𝑖𝑖,𝑡𝑡−𝑢𝑢,𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−𝑣𝑣 ,𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖,𝑡𝑡−𝑤𝑤,𝑅𝑅𝑅𝑅𝑉𝑉𝑖𝑖,𝑡𝑡−𝑥𝑥), [1] where i denotes country i and t denote time (quarterly data have been used for the period 2008:q1-2018:q4); cab is the current account balance of country i at time t as the ratio of us to eurozone; and the subscripts s, u, v, w, and x are the optimum time-lags for the regressors. rer is a measure of the eurozone's economic competitiveness in comparison to the united states. bud is defined as each eurozone country's budget deficit/surplus expressed as a percentage of gdp. rir stands for real interest rate, and it is computed as the nominal interest rate minus inflation of the us over that of the ez. gdp per capita is a measure of the purchasing power of domestic agents in constant prices and purchasing power parities (ppps) and it is constructed as the ratio of the us per capita gdp over that of the ez. furthermore, the erv variable measures the volatility of the exchange rate. it is calculated using a standard deviation of the moving average of the logarithm of the real exchange rate as a measure of time-varying exchange rate volatility3. 3.2 data description and variable specification table 1 provides a detailed list of the data used, frequency, source, and variable construction. the dataset includes the nineteen eurozone countries with the united states as a trading partner, and the cab data were obtained on a quarterly frequency to meet the time-frequency of the regressors. data for the cab involving the ez member states with the us as a trading partner was available from the oecd and eurostat databases from 2008 until 2018 on a quarterly frequency (oecd 2020, eurostat 2020). real exchange rate statistics were computed by quarterly nominal exchange rates, available from the ecb (european central bank 2020), and the consumer price index (cpi) of the 19 ez countries with the us as a trading partner from the eurostat database from the first quarter of 2008 until the last quarter of 2018, (eurostat 2020). gdp per capita in constant 2015 prices and ppps was available from the oecd database as well (oecd 2020) for both the ez and the us. real interest rate (rir) data was constructed by quarterly nominal interest rates (oecd 2020) and the cpi indexes also from the oecd database (oecd 2020). budget deficit (bud) data were available from the eurostat database for the 19 eurozone countries (eurostat 2020). finally, erv was calculated using the logarithmic moving average formula and the data for rer. regarding the us variables, the (oecd, 2020) and the (eurostat, 2020) databases have been used to draw our data on a quarterly frequency. the us 3 section 3.2 and table 1 provide a detailed analysis for the erv variable construction. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 148 variables are used in the model to create the ratio of us to ez in the estimation, except for the budget deficit variable. as the nominal exchange rate rises, the euro currency appreciates against the us dollar, reducing the competitiveness of eurozone countries. as a result, an increase in rer is expected to increase agents' real disposable income for goods and services, improving competitiveness and eventually improving the current account. the formula used to create the rer variable can be found in table 1 denoted by equation [2]. 𝜕𝜕𝐶𝐶𝐶𝐶𝐵𝐵 𝜕𝜕𝑅𝑅𝑅𝑅𝑅𝑅 > 0, 𝑠𝑠𝑠𝑠𝑠𝑠𝑐𝑐𝑐𝑐 𝜕𝜕𝜕𝜕 𝜕𝜕𝑅𝑅𝑅𝑅𝑅𝑅 > 0, 𝜕𝜕𝜕𝜕 𝜕𝜕𝑅𝑅𝑅𝑅𝑅𝑅 , < 0 in this part we should mention that 4 out of the 19 ez member states joined the ez in the period analyzed in this paper; estonia joined in 2011, latvia in 2014, lithuania in 2015, and slovakia in 2009. each country was obligated to change its national currency to enter the eurozone. before joining, the exchange rates for these counties were different from those of the ez. data regarding this issue is available from the ecb database through the nominal exchange rate. table 1. data sources and construction of the variables. data description frequency source variable nominal exchange rate quarterly; end of the period ecb (european central bank) (2020) reri,t: real exchange rate in 2015 prices reri,t= 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡 𝑈𝑈𝑈𝑈 𝐶𝐶𝐶𝐶𝐶𝐶𝑖𝑖,𝑡𝑡 𝐸𝐸𝐸𝐸 𝑐𝑐𝑖𝑖,𝑡𝑡 [2] 𝑐𝑐𝑖𝑖,𝑡𝑡: nominal exchange rate consumer price index, ez quarterly; base year 2015 eurostat (2020) consumer price index, us quarterly; base year 2015 eurostat (2020) consumer price index, ez quarterly; base year 2015 eurostat (2020) riri,t: real interest rate 𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡𝑈𝑈𝑈𝑈= 𝑠𝑠𝑖𝑖,𝑡𝑡-𝐶𝐶𝑃𝑃𝑅𝑅𝑖𝑖,𝑡𝑡𝑈𝑈𝑈𝑈 𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡𝐸𝐸𝐸𝐸= 𝑠𝑠𝑖𝑖,𝑡𝑡-𝐶𝐶𝑃𝑃𝑅𝑅𝑖𝑖,𝑡𝑡𝐸𝐸𝐸𝐸 𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡 = 𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡𝑈𝑈𝑈𝑈 𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡𝐸𝐸𝐸𝐸 𝑠𝑠𝑖𝑖,𝑡𝑡: nominal interest rate consumer price index, us quarterly; base year 2015 eurostat (2020) nominal interest rate quarterly; threemonth money market rate ecb (european central bank) (2020) budget deficit of the ez countries quarterly; percent of gdp eurostat (2020) budi,t: budget deficit of the ez countries budi,t is used without manipulation gross domestic product, ez quarterly; constant prices (base year 2015), ppps oecd (2020) gdpi,t: gross domestic product in constant prices and ppps 𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖,𝑡𝑡 = 𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖,𝑡𝑡𝑈𝑈𝑈𝑈 𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖 ,𝑡𝑡𝐸𝐸𝐸𝐸 gross domestic product, us quarterly; constant prices (base year 2015), ppps oecd (2020) exchange rate volatility quarterly; moving average author’s calculations ervi,t: exchange rate volatility 𝑉𝑉𝑖𝑖,𝑡𝑡+𝑚𝑚 = � 1 𝑚𝑚 �(𝑅𝑅𝑡𝑡+𝑖𝑖−1 − 𝑅𝑅𝑡𝑡+𝑖𝑖−2)2 𝑚𝑚 𝑖𝑖=1 � 1 2� [3] v: volatility of the exchange rate r: logarithm of the exchange rate m: number of periods gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 149 relating to the cab definition, the real interest rate (rir) is anticipated to have a positive relationship with the cab. the expected positive sign is explained by theory since an increase in real interest rates decreases investment while increasing savings, resulting in an improvement in the cab due to the difference between savings and investment. therefore, 𝑅𝑅𝑅𝑅𝑅𝑅 = 𝑅𝑅𝐶𝐶𝑅𝑅𝑈𝑈𝑈𝑈 𝑅𝑅𝐶𝐶𝑅𝑅𝐸𝐸𝐸𝐸 , 𝜕𝜕𝐶𝐶𝜕𝜕𝜕𝜕 𝜕𝜕𝑅𝑅𝐶𝐶𝑅𝑅 > 0. the budget deficit (bud) of the 19 eurozone member states as a percent of gdp has been used. a decrease in the value of the cab variable means an increase in the current account deficit. consequently, decreases in the cab variable are anticipated to be positively related to an increase in the budget deficit: 𝜕𝜕𝐶𝐶𝜕𝜕𝜕𝜕 𝜕𝜕𝜕𝜕𝑈𝑈𝜕𝜕 < 0. as noted in the literature review section, domestic agents' disposable income has a positive effect on the demand for goods and services. the model incorporates the agents' gdp per capita in constant prices and purchasing power parities (ppps) as a degree of the domestic agents' income. the estimated coefficient is anticipated to have a positive sign since agents’ income influences the demand for goods and services positively, i.e., a rise in the per capita gdp in the ez raises eurozone imports (m) from the us while a rise in the us per capita gdp has a positive effect on exports to the us (x). gdp = gdpus gdpez , ∂x ∂gdpus > 0, ∂m ∂gdpez > 0 therefore, ∂ca𝜕𝜕 ∂gdp > 0. in this work, an attempt was made to fill the gap in the literature on the relationship between cab and erv by employing an erv measure proposed by serenis and tsounis (2012, 2015). since erv is not directly observable, there is no clear, correct, or wrong way to measure it. nonetheless, several researchers over the years attempted to devise various methods of measuring it. in our approach, we employed a moving average formula of the logarithm of the exchange rate, the formula can be found in table 1 signified by equation (3). v is the volatility of the exchange rate, r stands for the logarithm of the real exchange rate, and m is the number of periods, in our case, 4 quarters have been used as a lag to observe the low and high peak values. this is an ad-hoc hypothesis based on adaptive expectations (see serenis and tsounis 2012, 2015). generally, erv negatively affects the cab because it raises uncertainty about import and export prices of goods and services when agents are risk averse. however, it may have a positive effect on trade flows if agents are risk-loving (agiomirgianakis et al. 2017). assumptions regarding the sign of the coefficient cannot be drawn ex-ante, this will be empirically estimated. if the erv coefficient displays a positive sign, this indicates that the eurozone countries’ exporters are more risk lovers than those of the united states and the opposite is if the sign of erv is negative. therefore, either 𝜕𝜕𝐶𝐶𝜕𝜕𝜕𝜕 𝜕𝜕𝐸𝐸𝑅𝑅𝜕𝜕 > 0 or 𝜕𝜕𝐶𝐶𝜕𝜕𝜕𝜕 𝜕𝜕𝐸𝐸𝑅𝑅𝜕𝜕 < 0. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 150 3.3 methodology a panel data cointegration analysis was performed to investigate the long-run relationship between the cab and its determinants. cointegration analysis examines the presence of a cointegrated mixture of the series to evaluate whether there is a statistically significant relationship among the variables. if the order of integration of this combination is low, it implies an equilibrium relationship among the initial series, which is known as cointegration. when applied to nonstationary time series, it is essential to employ cointegration analysis rather than standard linear regression techniques, as the latter would yield spurious results. an empirical model that investigates both the shortand long-run relationship between the cab and its determinants in the bilateral relationship between the us and ez member states was created. this is especially crucial when the econometric model is used to make policyimplementation inferences with time lags. both shortand long-term effects were assessed on the cab and its determinants, making use of a dataset comprised of the 19 eurozone member states with the us as a trading partner, rather than averaging the data for each state. the pmg method was used, which can be thought of as a panel error correction (ec) model, in which shortand long-term effects are estimated mutually using an ardl model (pesaran and shin, 1999) where the short-run effects are permitted to differ among cross-sections with common longrun coefficients. country heterogeneity is especially important in short-run relations, but long-run relations among cab are expected to be more homogenous among nations in the long run. the pmg technique also produces reliable estimates of the parameters in the long-run relation among stationary and integrated variables. when both i(0) and i(1) variables are incorporated in this manner, the model can be estimated, whereas other techniques need only i(0) or i(1) variables. however, the pmg approach still demands the regressors to be entirely exogenous. this is ensured if the dynamic specification of the model is sufficiently improved such that the regressors are strictly exogenous. increasing the number of regressors randomly, on the other hand, reduces the degrees of freedom. the residuals must also be serially uncorrelated. likewise, it is essential to ensure that the variables are not i(2) since the pmg technique would provide spurious results. thus, before advancing with model estimation, the order of integration of the regressors must be tested. this issue has been validated by applying the im, pesaran, and shin (ips) unit root test for panel data, as well as the levin, lin, and chu (llc) unit root test. the first step for the pmg method is to conduct panel unit root tests on the dependent and independent variables of interest. time-series unit root tests have given rise to panel unit root tests. this arose since the combining of the asymptotic properties of the time-series dimension t and the cross-sectional dimension n. im et al. (2003) developed the ips unit root test to capture dataset heterogeneity under the alternative hypothesis. the null hypothesis assumes that all series are non-stationary, whereas the alternative assumes that a portion of the series is stationary. table 2 shows the results of the ips panel unit root test. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 151 table 2. im, pesaran & shin unit root test. variables i (0) i (1) cab -5.508 -28.217* rer -2.958* -16.817* bud -10.419* -32.075* gdp 2.071 -12.817* erv -2.958* -16.817 rir -15.575* -29.912* source. author’s estimation / the * designates significance at least at the 5% level. ips test was performed utilizing the 5% significant level. h0 was rejected at the 5% statistically significant level for all variables except per capita gdp, which was discovered to be non-stationary in all panels. as a result, the variables cab, rer, bud, rir, and erv are i(0), whereas gdp is i(1). levin et al. (2002) created the second unit root test employed, which proposes hypotheses for assessing stationarity in panel data. the llc test demonstrates that each time series has a unit root under the null hypothesis, but each time series is stationary under the alternative hypothesis. the llc, like other unit root tests, assumes that each cross-separate section's operations are self-contained. from table 3, h0 was rejected at the 5% statistically significant level for cab, bud, and rir but not for rer, gdp, and erv, which, for all panels, were discovered to be non-stationary at their level. as a result, the variables cab, bud, and rir are i(0), whereas gdp, rer, and erv are i(1). table 3. levin, lin & chu unit root results. variables i (0) i (1) cab -4.907 -24.034* rer -0.935 -20.610* bud -7.700* -26.088* gdp -1.034 -6.762* erv -0.935 -20.610* rir -15.362* -28.865* source. author’s estimation / the * designates significance at least at the 5% level. llc test was performed using the 5% significant level. as can be seen from tables 2 and 3, the model contains both i(0) and i(1) regressors but not i(2), indicating that the variables are either stationary at their level or their first difference and that the pmg method can be applied. the sensitivity of the cointegrated variables to any deviation from the long-run equilibrium is a key feature. the pmg method is used on an error correction model to estimate the speed of adjustment to the long-run relationship while allowing for unrestricted cross-section heterogeneity in adjustment dynamics and fixed effects. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 152 following perasan et al. (1999), the pmg-restricted version of (1) is estimated on pooled data as 𝛥𝛥𝐶𝐶𝐶𝐶𝐵𝐵𝑖𝑖,𝑡𝑡 = 𝜑𝜑𝑖𝑖�𝐶𝐶𝐶𝐶𝐵𝐵𝑖𝑖,𝑡𝑡−1 − ∑ 𝜗𝜗𝑘𝑘,𝑖𝑖𝐺𝐺𝑘𝑘,𝑖𝑖,𝑡𝑡 𝜇𝜇 𝑘𝑘=1 � + ∑ 𝜆𝜆𝑖𝑖𝑖𝑖𝛥𝛥𝐶𝐶𝐶𝐶𝐵𝐵𝑖𝑖 ,𝑡𝑡−𝑖𝑖 + ∑ ∙𝜇𝜇 𝜅𝜅=1 ∑ 𝛽𝛽𝑘𝑘,𝑖𝑖,𝑡𝑡−𝑖𝑖𝐺𝐺𝑘𝑘,𝑖𝑖,𝑡𝑡 +𝑞𝑞−1 𝑖𝑖=0 𝑝𝑝−1 𝑖𝑖=1 ∑ 𝛾𝛾𝑧𝑧𝑠𝑠𝑐𝑐𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑧𝑧,𝑡𝑡 + 𝛿𝛿𝑐𝑐𝑠𝑠𝛿𝛿𝛿𝛿𝑦𝑦𝑡𝑡 + 𝑣𝑣𝑖𝑖 + 𝜀𝜀𝑖𝑖,𝑡𝑡3 𝑧𝑧=1 [4] where i = 1, …,19 indicates countries and t = 1, ..., 44 and indicates time; δ is the operator of the first difference; cabi,t denotes the cab of eurozone country i vis-à-vis the usa at time t; μ=6 and indicates the number of explanatory variables; g=( rer, erv, bud, rir, gdp) is the vector containing the regressors. also, in (4) the variable season represents three dummy variables accounting for seasonality and the variable entry represents a dummy variable for the entrant year of each country to the eurozone. note that the dataset is formed by panel data where each of the variables contained in the g vector has 19 time series, one for each country and, each time series has 44 time periods. the parameter φi represents the error-correcting speed of adjustment based on the long-run relationship, φi is significant since it indicates if the variables are cointegrated and is anticipated to be statistically significant and negative under the hypothesis that the variables exhibit a return to long-run equilibrium. furthermore, the estimated coefficients for 𝜗𝜗k, indicate the long-run relationship between the variables, whereas the βk,i the determinant variables' short-run coefficients, νi indicates the country-specific fixed-effect, ε is a time-varying disturbance term, μ=6 is the number of the determinants and q, and p is the number of lags. the following steps provide a brief explanation of the pmg method. it is necessary to identify the ardl order of the model denoted by [1] first. also, the value of q for each regressor must be calculated. thus, for each country, equation [1] was re-estimated, and the lagged ardl order was established using the schwarz bayesian criterion (sbc). the lag order of the ardl model for each cross-section is determined using a maximum of 6 lags in equation [1]. at that point, the most common lag order across cross-sections for each variable has been used, yielding the final form of equation (1) for estimation: 𝐶𝐶𝐶𝐶𝐵𝐵𝑠𝑠, 𝛿𝛿 = 𝜑𝜑𝑖𝑖� 𝐶𝐶𝐶𝐶𝐵𝐵𝑖𝑖,𝑡𝑡−1 − ∑ 𝜗𝜗𝑘𝑘,𝑖𝑖𝐺𝐺𝑘𝑘,𝑖𝑖,𝑡𝑡 𝜇𝜇 𝑘𝑘=1 � + 𝜆𝜆𝑖𝑖𝛥𝛥𝐶𝐶𝐶𝐶𝐵𝐵𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽1,𝑖𝑖,0𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡 + 𝛽𝛽1,𝑖𝑖,1𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽1,𝑖𝑖,2𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−2 + 𝛽𝛽1,𝑖𝑖,3𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−3 + 𝛽𝛽1,𝑖𝑖,4𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−4 + 𝛽𝛽2,𝑖𝑖,0𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡 + 𝛽𝛽2,𝑖𝑖,1𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽2,𝑖𝑖,2𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−2 + 𝛽𝛽2,𝑖𝑖,3𝛥𝛥𝑅𝑅𝑅𝑅𝑅𝑅𝑖𝑖,𝑡𝑡−3 + 𝛽𝛽3,𝑖𝑖,0𝛥𝛥𝐵𝐵𝐵𝐵𝐷𝐷𝑖𝑖 ,𝑡𝑡 + 𝛽𝛽3,𝑖𝑖,1𝛥𝛥𝐵𝐵𝐵𝐵𝐷𝐷𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽3,𝑖𝑖,2𝛥𝛥𝐵𝐵𝐵𝐵𝐷𝐷𝑖𝑖 ,𝑡𝑡−2 + 𝛽𝛽3,𝑖𝑖,3𝛥𝛥𝐵𝐵𝐵𝐵𝐷𝐷𝑖𝑖 ,𝑡𝑡−3 + 𝛽𝛽4,𝑖𝑖,0𝛥𝛥𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖,𝑡𝑡 + 𝛽𝛽4,𝑖𝑖,1𝛥𝛥𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽4,𝑖𝑖,2𝛥𝛥𝐺𝐺𝐷𝐷𝑃𝑃𝑖𝑖,𝑡𝑡−2 + 𝛽𝛽5,𝑖𝑖,0𝛥𝛥𝑅𝑅𝑅𝑅𝑉𝑉𝑖𝑖,𝑡𝑡 + 𝛽𝛽5,𝑖𝑖,1𝛥𝛥𝑅𝑅𝑅𝑅𝑉𝑉𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽5,𝑖𝑖,2𝛥𝛥𝑅𝑅𝑅𝑅𝑉𝑉𝑖𝑖,𝑡𝑡−2 + 𝛽𝛽5,𝑖𝑖,3𝛥𝛥𝑅𝑅𝑅𝑅𝑉𝑉𝑖𝑖,𝑡𝑡−3 + 𝛾𝛾1 𝑆𝑆1 + 𝛾𝛾2𝑆𝑆2 + 𝛾𝛾3𝑆𝑆3 + 𝛿𝛿𝑐𝑐𝑠𝑠𝛿𝛿𝛿𝛿𝑦𝑦𝑡𝑡 + 𝜈𝜈𝑖𝑖+𝜀𝜀𝑖𝑖,𝑡𝑡 [5] second, the maximum likelihood was used to estimate the long-run coefficients 𝜗𝜗𝑘𝑘,𝑖𝑖s across cross-sections. the final step involves estimating the short-run coefficients 𝜆𝜆𝑖𝑖s and 𝛽𝛽𝑘𝑘,𝑖𝑖,𝑖𝑖s, the speed of adjustment φi the country-specific intercepts vi, and the country-specific error variances gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 153 on a country-by-country basis, again employing the maximum likelihood method and the estimations of the long-run coefficients attained earlier. the pmg has been tested for the following specifications: the model's dynamic stability has been investigated; for our model to be dynamically stable, the coefficient of the error correction term must be negative and not lower than -2, i.e., inside the unit circle. its values were -0.417 and are statistically significant from zero at a significance level of less than 1%. as a result, the condition for dynamic stability is met. another requirement is to see if a cointegrated relationship exists, i.e., a statistically significant long-run relationship among the cab and its determinants. the coefficient of the error correction term φi must be negative and statistically significant to indicate cointegration. the percentage change in any disequilibrium between the dependent and the independent variables that are corrected during one period (in our case, one quarter) is represented by the value of this coefficient. its value indicates the rate at which the economy is adjusting to the long-run equilibrium. regarding our situation, the value of φi equals -0.417, indicating that the variables are cointegrated and that 41.7 percent of any disequilibrium between the dependent and the regressors is corrected within one quarter. in the end, the long-run elasticities must be identical across cross-sections for the pmg estimator to be applied. this pooling across cross-sections gives consistent and efficient estimates when the applied are correct, i.e., the long-run coefficients are the same across crosssections. the pmg estimations will be inconsistent if the slope parameter in the original model is heterogeneous. a hausman test is used to assess the homogeneity hypothesis. this hausman test is founded on a comparison of the mean group (mg) and the pmg estimators. the hausman test statistic was 3.91, with a 0.56 level of statistical significance (p). consequently, the null hypothesis that the difference in coefficients is not systematic cannot be rejected, and the model's slope parameters across cross-sections are proved to be homogenous. 4. results the dynamic specification of the model was found using the methods outlined above (see section 3.1 and eq. [5]) and is: ardl (1,5,4,4,3,4); each number in order represents the distributed lags for the variables cab, rer, rir, bud, gdp, and erv, respectively. table 4 presents the long and short-run coefficients of the cab determinants of the eurozone members vis-à-vis the united states. the values of the long-run coefficients in table 4 above illustrate the effect of the explanatory on the dependent variable. the long-run coefficients proved to be statistically significant at levels less than 1%. all of them are shown to have the anticipated signs: the us per capita income ratio over the ez per capita income influences the cab positively since a rise in the gdp per capita in the ez raises eurozone imports from the united states while an upsurge in the gdp per capita of the us has a positive impact on exports to the united states (0.033). nevertheless, the longrun coefficient value is smaller than one, indicating that when the us per capita income increases gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 154 by one ppp unit over the ez per capita income, the cab improves by 0.033. moreover, the increase in competitiveness stemming from a real depreciation of the real exchange rate (rer) (2.518) positively affects the balance of the current account. also, the real interest rate (rir) variable has the anticipated sign and positively affects cab (0.486) in the long run; an upsurge in real interest rates reduces expenditure (investment & consumption) by increasing savings, and as a result, current account deficit decreases or current account surplus increases. additionally, the variable representing the budget deficit (bud) negatively affects the cab (-0.348) since an increase in budget deficit decreases the cab variable through an increase in government expenditures. finally, the effect of erv on the cab is positive (7.551) for the eurozone countries, indicating that erv affects the actions of the domestic agents i.e., exporters in the eurozone are more risk-loving than importers and they see erv as an opportunity for speculated profit. by examining the short-run coefficients, the gdp variable does not affect the cab in any of the time lags, meaning that domestic agents have no intention to spend on goods and services among quarters until they gather a surplus on their income. at the fifth time lag, the short-run coefficients for the real exchange rate (rer) become statistically significant, indicating that the price difference between ez member states and the us has an impact on the cab after 5 quarters. moreover, the real interest rate (rir) in the short run does not affect the cab since it cannot reduce investment and increase savings and eventually cannot lead to a cab improvement. the variable of budget deficit (bud) also has no impact on the balance of the current account in the short run, indicating that the results adopt the ricardian equivalence hypothesis in the short run, in which the current account deficit and the budget deficit are uncorrelated. at the same time, this is a contradiction according to the theory because classical economists have studied the long-run equilibrium and not the short-run. besides, the erv does not affect the cab, thus inference of whether the exports are risk lovers or risk averters cannot be drawn. this can also indicate that the exporters/importers are not showing their true risk intentions in the short run. to assure the robustness of the results, robustness analysis has been performed and it can be found in the appendix. by changing the dynamic specification of the model in all cases,4 the signs and the level of statistical significance of the long-run coefficients for all regressors were the same as in the original model. on the other hand, the short-run coefficients proved to be nonstatistically significant (along with the hausman test) in all cases except for the specification, ardl (1,5,4,4,3,3), where the estimated coefficient for real exchange rate is statistically significant in the fifth time lag, as it is in the original model. furthermore, the value of coefficients does not deviate significantly from those shown in table 4 during the robustness tests. ultimately, for reasons of completeness, a ramsey rest test has been performed, accounting for omitted variable bias and functional misspecification. in the reset test, the null hypothesis that omitted variable bias and misspecifications exist was rejected (table 4 and the appendix). 4 the model originally used for the estimation is an ardl(1,5,4,4,3,4) and its results are presented in table 4. in the appendix robustness checks can be found were the specification of the ardl model has changed to ardl(1,4,4,4,3,4), ardl(1,5,4,4,2,4), and ardl(1,5,4,4,3,3). gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 155 table 4. long-run and short-run coefficients of the current account balance determinants. series coefficient st. error p-value long-run coefficients rer 2.518*** 0.899 0.000 rir 0.486*** 0.035 0.000 bud -0.348*** 0.004 0.000 gdp 0.033*** 0.001 0.000 erv 7.551*** 0.405 0.000 hausman test 3.91 0.562 error correction coef. (φ) -0.417*** 0.161 0.010 short-run coefficients δcabt-1 -0.222** 0.100 0.028 δrert 0.787 1.860 0.672 δrert-1 -1.195 2.166 0.581 δrert-2 -1.067 2.220 0.628 δrert-3 -0.758 2.411 0.753 δrert-4 -0.199 1.566 0.899 δrert-5 2.459* 1.226 0.045 δrirt 0.008 0.100 0.979 δrirt-1 -0.008 0.169 0.600 δrirt-2 0.104 0.192 0.588 δrirt-3 -0.014 0.162 0.370 δrirt-4 -0.013 0.145 0.365 δbudt -0.051 0.515 0.320 δbudt-1 -0.039 0.291 0.179 δbudt-2 -0.574 0.038 0.140 δbudt-3 -0.036 0.031 0.248 δbudt-4 -0.017 0.027 0.520 δgdpt 1.673 4.023 0.677 δgdpt-1 5.505 3.445 0.110 δgdpt-2 0.940 2.155 0.662 δgdpt-3 -1.218 2.289 0.595 δervt -9.740 9.251 0.292 δervt-1 -1.555 12.010 0.897 δervt-2 11.256 10.54 0.286 δervt-3 9.567 13.838 0.489 δervt-4 -1.095 6.946 0.875 s1 0.436 0.253 0.085 s2 0.349 0.355 0.325 s3 0.483** 0.221 0.029 intercept -0.586*** 0.203 0.004 entry 0.281 0.201 0.302 model misspecification test ramsey reset test f-statistic 0.329 dynamic specification ardl (1,5,4,4,3,4) estimation method pooled mean group (pmg) adjusting for country-fixed effects and seasonality number of countries 19 time period 2008: q1-2018: q4 (44 periods) observations 836 source. author’s estimation / ***, **, *: denote statistical significance at 1%, 5% and 10%, correspondingly. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 156 5. conclusion and policy implications this work studies the current account balance of the 19 eurozone member states in the bilateral relationship between them and the united states. the countries studied are considered to be the most prominent trading partners in the world. quarterly bilateral current account data have been used to provide an effective governmental policy implication. the novelty of this work lies in the fact that bilateral current account studies are scarce in literature and, to the best of our knowledge, do not exist for the current account balance of the ez member states vis-a-vis the us. the determinants of the current account balance in this bilateral relationship are tested over the period 2008 -2018 using quarterly data (2008:q1-2018:q4). it is found that the real bilateral exchange rate, the bilateral real interest rate, the budget deficit of the eurozone countries as a percent of gdp, the income of the domestic agents, and the exchange rate volatility are determinants of the bilateral current account balance of the ez vis-à-vis the us in the long run. moreover, since our data are seasonally unadjusted, three seasonality variables and a dummy variable maned entry, which represents the year of the entrance of each eurozone country have been included. in terms of empirical approach, our study is based on the theory of cointegrated panel data and error correction techniques for cointegrated variables utilizing the pmg method to cointegration. empirical findings indicate that all the long-run coefficients proved to be statistically significant meaning that the current account exhibits a cointegrated relationship along with its determinants and they can be used for policy implementation. in detail, the current account has a positive longrun relationship with the real exchange rate, the real interest rate, the exchange rate volatility, the gdp per capita, and a negative relationship with the budget deficit. then again, in the short run, only the real exchange rate is seen to be statistically significant, suggesting that the difference in prices can be used as a valid instrument for exercising economic policy for short periods. our conclusion on the current account balance suggests some beneficial policy implications regarding both counties. the policymakers in both the eurozone and the united states can evaluate the impact of the determinants of their policy implementation, i.e., by choosing the right determinants, they can prevent the current account from running long-lasting deficits. the pmg method used in this paper allows us to draw inferences on policy implications by evaluating the long-run and short-run coefficients. in the long run, all the determinants can be used as instruments for economic policy. more specifically, the value of the estimated long-run coefficients indicates that the current account balance is very sensitive to rer. ez policymakers may use this result for boosting growth: rer is a measure of ez competitiveness and depends on nominal er and the ratio of the price levels of ez and the us. nominal exchange rates, in a flexible er regime, in the long run, are determined by the fundaments of the two economies. on the other hand, price levels can be affected by economic policy. by keeping the level of prices in the ez at a lower level than that in the us i.e. by adopting anti-inflationary policies ez competitiveness increases and that will boost cab as, according to the estimated coefficients, a unit increase in competitiveness will improve cab by approximately 2.5 units, ceteris paribus. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 157 additionally, the estimated coefficient of the real interest rate variable shows that in the case where ez policymakers increase interest rates to bring down inflation, this action will make the price of exports relatively competitive, and as a result, exports will increase. even though an increase in interest rates by the ez leads to an appreciation of the euro currency as demand for the euro increases compared to the us dollar, the price of exports also increases resulting in a relatively small value of the estimated long-run coefficient (0.48). moreover, the decision to increase government expenditure by policymakers, most of the time, results in budget deficits. government expenditures are increased in recession periods since they are subject to countercyclical policy. as a result, the ez should increase borrowing for the us, which will have a negative effect on the current account. therefore, a budget deficit can lead to a current account deficit, in the long run, often known as a twin deficit. gdp is a tool used by policymakers to help assess the economy’s well-being and to make informed decisions or examine whether the economy is experiencing a recession. gdp can also be used as an indicator of trade surplus or deficit. finally, if policymakers do not fully anticipate exchange rate fluctuations, an increase in exchange rate volatility, which increases risk, may lead risk-averse agents to curtail import or export activity and reallocate production to the domestic market. this circumstance creates uncertainty in the market, which has a negative effect on the trade balance and eventually on the current account. thus, unexpected movements in the exchange rate should be monitored and regulated on a regular basis by policymakers. in the short run, the empirical results indicated that only real exchange rates could be used as a valid instrument for exercising economic policy. in the short run, the exchange rates are determined by forces of supply and demand. ez policymakers can regulate supply and demand through investments, loans, exchanges in the foreign market, and consumer decisions that affect the value of their economy’s currency to the us. note that the methodology employed in this work may be of use to assess other economic policies as well. references agiomirgianakis, g., serenis, d., & tsounis, n. (2017). effective timing of tourism policy: the case of singapore. economic modelling, 60(1), 29-38. https://doi.org/10.1016/j.econmod.2016.09.001 anoruo e., & elike. u. (2008). asymmetric dynamics in current account interest rate nexus: evidence from asian countries. investment management and financial innovations, 5(4). aloryito, g. k., senadza, b., & nketiah-amponsah, e. (2016). testing the twin deficits hypothesis: effect of fiscal balance on current account balance-a panel analysis of subsaharan africa. modern economy 7(09), 945. https://doi.org/10.4236/me.2016.79097 arize, a.c. (1994). cointegration test of a long-run relation between the real effective exchange rate and the trade balance. international economic journal, 8(3), 1-9. https://doi.org/10.1080/10168739400000001 https://doi.org/10.1016/j.econmod.2016.09.001 https://doi.org/10.4236/me.2016.79097 https://doi.org/10.1080/10168739400000001 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 158 baharumshah, a. z. (2001). the effect of exchange rate on bilateral trade balance: new evidence from malaysia and thailand. asian economic journal, 15(3), 291-312. https://doi.org/10.1111/1467-8381.00135 bernhardsen, t. (2000). the relationship between interest rate differentials and macroeconomic variables: a panel data study for european countries. journal of international money and finance, 19(2), 289-308. https://doi.org/10.1016/s02615606(00)00002-4 blanchard, o., & giavazzi f. (2002). current account deficits in the euro area: the end of the feldstein-horioka puzzle? brookings papers on economic activity 2002(2), 147-209. https://doi.org/10.1353/eca.2003.0001 bussière, m., fratzscher, m., & müller, g.j. (2004). current account dynamics in oecd and eu acceding countries an intertemporal approach. ecb working paper no. 311. https://doi.org/10.2139/ssrn.515074 calderón, c., chong, a., & loayza, n. (1999). determinants of current account deficits in developing countries. working papers central bank of chile 51, central bank of chile. calderón, c., chong, a., & zanforlin, l. (2007). current account deficits in africa: stylized facts and basic determinants. economic development and cultural change, 56(1), 191-221. https://doi.org/10.1086/520557 catão, l., & falcetti, e. (2002). determinants of argentina's external trade. journal of applied economics 5(1), 19-57. https://doi.org/10.1080/15140326.2002.12040570 ca' zorzi, m., dieppe, a., & chudik, a. (2009). current account benchmarks for central and eastern europe: a desperate search? ecb working paper no. 995. https://doi.org/10.2139/ssrn.1316215 chinn, m. d., & prasad e.s. (2003). medium-term determinants of current accounts in industrial and developing countries: an empirical exploration. journal of international economics, 59(1), 47-76. https://doi.org/10.1016/s0022-1996(02)00089-2 das, d.k. (2016). determinants of current account imbalance in the global economy: a dynamic panel analysis. journal of economic structures, 5(1), 8. https://doi.org/10.1186/s40008-0160039-6 debelle, g., & faruqee h. (1996). what determines the current account? a cross-sectional and panel approach. imf working paper no. 58. https://doi.org/10.5089/9781451966701.001 de grauwe, p. (1988). exchange rate variability and the slowdown in growth of international trade. imf econ rev, 35, 63-84. https://doi.org/10.2307/3867277 dettmann, g., möbert j., & weistroffer c. (2012). bilateral current account rebalancing in the emu. intereconomics, 47(4), 257-64. https://doi.org/10.1007/s10272-012-0426-7 european central bank (ecb) (2020). ecb reference exchange rate, us dollar/euro, 2:15 pm (c.e.t.). https://doi.org/10.14612/oldani_2_2012 eurostat (2020). harmonised index of consumer prices monthly data, luxemburg: eurostat publications. https://doi.org/10.1111/1467-8381.00135 https://doi.org/10.1016/s0261-5606(00)00002-4 https://doi.org/10.1016/s0261-5606(00)00002-4 https://doi.org/10.1353/eca.2003.0001 https://doi.org/10.2139/ssrn.515074 https://doi.org/10.1086/520557 https://doi.org/10.2139/ssrn.1316215 https://doi.org/10.1186/s40008-016-0039-6 https://doi.org/10.1186/s40008-016-0039-6 https://doi.org/10.2307/3867277 https://doi.org/10.1007/s10272-012-0426-7 https://doi.org/10.14612/oldani_2_2012 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 159 eurostat (2020). quarterly non-financial accounts for general government, luxemburg: eurostat publications. eurostat (2020). personal transfers and compensation of employees, luxemburg: eurostat publications. eurostat (2021). countries and regions: united states, european commission: dg trade, https://ec.europa.eu/trade/policy/countries-and-regions/countries/united-states/ (accessed 3rd january, 2022) ghosh, a., & ostry. j. d. (1995). the current account in developing countries: a perspective from the consumption-smoothing approach. the world bank economic review, 9(2), 30533. https://doi.org/10.1093/wber/9.2.305 glick, r., & rogoff k. (1995). global versus country-specific productivity shocks and the current account. journal of monetary economics, 35(1), 159-92. https://doi.org/10.1016/03043932(94)01181-9 greenwood, j. (1983). expectations, the exchange rate, and the current account. journal of monetary economics, 12(4), 543-69. https://doi.org/10.1016/0304-3932(83)90038-7 gruber, j. w., & kamin s.b. (2007). explaining the global pattern of current account imbalances. journal of international money and finance, 26(4), 500-522. https://doi.org/10.1016/j.jimonfin.2007.03.003 herrman, s. & winker, a. (2008). real convergence, financial markets, and the current account: emerging europe versus emerging asia, ecb working paper no. 88. https://doi.org/10.2139/ssrn.1084928 herrmann, s., & jochem, a. (2005). determinants of current account developments in the central and east european eu member states consequences for the enlargement of the euro area. discussion paper series 1: economic studies, 32, deutsche bundesbank. https://doi.org/10.2139/ssrn.2785215 huchet-bourdon, & m., korinek. j. (2011). to what extent do exchange rates and their volatility affect trade? oecd trade policy papers, no. 119. oecd publishing, paris. im, k.s., pesaran, m.h., & shin. y. (2003). testing for unit roots in heterogeneous panels. journal of econometrics, 115(1), 53-74. https://doi.org/10.1016/s0304-4076(03)00092-7 iqbal, b. a., rahman, m. n., & turay, a. (2017). indo-us bilateral fdi and current account balance: developing causal relationship. academic journal of interdisciplinary studies, 6(1), 129. https://doi.org/10.5901/ajis.2017.v6n1p129 iyke b.n., & ho s-y. (2017). the real exchange rate, the ghanaian trade balance, and the jcurve. journal of african business, 18(3): 380-392. https://doi.org/10.1080/15228916.2017.1315706 kalou, s., & paleologou s.m. (2012). the twin deficits hypothesis: revisiting an emu country. journal of policy modeling, 34(2): 230-41. https://doi.org/10.1016/j.jpolmod.2011.06.002 kandil, m. (2009). exchange rate fluctuations and the balance of payments: channels of interaction in developing and developed countries. journal of economic integration, 24(1): 151-174. https://doi.org/10.11130/jei.2009.24.1.151 https://ec.europa.eu/trade/policy/countries-and-regions/countries/united-states/ https://doi.org/10.1093/wber/9.2.305 https://doi.org/10.1016/0304-3932(94)01181-9 https://doi.org/10.1016/0304-3932(94)01181-9 https://doi.org/10.1016/0304-3932(83)90038-7 https://doi.org/10.1016/j.jimonfin.2007.03.003 https://doi.org/10.2139/ssrn.1084928 https://doi.org/10.2139/ssrn.2785215 https://doi.org/10.1016/s0304-4076(03)00092-7 https://doi.org/10.5901/ajis.2017.v6n1p129 https://doi.org/10.1080/15228916.2017.1315706 https://doi.org/10.11130/jei.2009.24.1.151 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 160 kraay, a., & ventura, j., (2000). current accounts in debtor and creditor countries. the quarterly journal of economics, 115(4), 1137-66. https://doi.org/10.1162/003355300555033 falk, m. (2008). determinants of the trade balance in industrialized countries, countries. wifostudies, wifo, number 34230, june. lane, p. r., & milesi-ferretti. g.m. (2002). external wealth, the trade balance, and the real exchange rate. european economic review, 46(6), 1049-71. https://doi.org/10.1016/s00142921(02)00160-5 lars e. o. s., & razin, a. (1983). the terms of trade and the current account: the harbergerlaursen-metzler effect. journal of political economy, 91(1), 97-125. https://doi.org/10.1086/261130 leiderman, l., & razin, a. (1991). determinants of external imbalances: the role of taxes, government spending, and productivity. journal of the japanese and international economies, 5(4), 421-50. https://doi.org/10.3386/w3738 levin, a., lin, c-f., & chu c-s.j. (2002). unit root tests in panel data: asymptotic and finitesample properties. journal of econometrics, 108(1): 1-24. https://doi.org/10.1016/s03044076(01)00098-7 medina l., prat j., & thomas a. h. (2010). current account balance estimates for emerging market economies. imf working papers no. 43: 1-22. https://doi.org/10.5089/9781451962994.001 mendoza, e. (1991). real business cycles in a small open economy." american economic review 81(4): 797-818. mendoza, e. (1995). the terms of trade, the real exchange rate, and economic fluctuations. international economic review, 36(1): 101-37. https://doi.org/10.2307/2527429 milesi-ferrett, g. m., & razin, a. (1996). sustainability of persistent current account deficits, no. 5467, nber working papers, national bureau of economic research, inc. https://doi.org/10.3386/w5467 nieminen, m. (2015). trade imbalances within the euro area and with respect to the rest of the world. economic modelling, 48: 306-314. https://doi.org/10.1016/j.econmod.2014.10.012 nuroglu, e., & fomonikunst, r.m. (2012). the effects of exchange rate volatility on international trade flows: evidence from panel data analysis and fuzzy approach. zbornik radova ekonomskog fakulteta u rijeci: časopis za ekonomsku teoriju i praksu, 30(1): 9-31. obstfeld, m., & rogoff k. (1995). the intertemporal approach to the current account. in grossman, g.m., rogoff, k. (eds.) handbook of international economics, vol. 3, northholland, 1731-1799. https://doi.org/10.1016/s1573-4404(05)80014-0 organisation for economic co-operation and development (oecd) (2020). trade in services by partner economy, paris: oecd publications. organisation for economic co-operation and development (oecd) (2020). inflation (cpi) (indicator). paris: oecd publications. https://doi.org/10.1787/eee82e6e-en organisation for economic co-operation and development (oecd) (2020). short-term interest rates (indicator). paris: oecd publications. https://doi.org/10.1787/2cc37d77-en https://doi.org/10.1016/s0014-2921(02)00160-5 https://doi.org/10.1016/s0014-2921(02)00160-5 https://doi.org/10.1086/261130 https://doi.org/10.3386/w3738 https://doi.org/10.1016/s0304-4076(01)00098-7 https://doi.org/10.1016/s0304-4076(01)00098-7 https://doi.org/10.2307/2527429 https://doi.org/10.3386/w5467 https://doi.org/10.1016/j.econmod.2014.10.012 https://doi.org/10.1016/s1573-4404(05)80014-0 https://doi.org/10.1787/eee82e6e-en https://doi.org/10.1787/2cc37d77-en gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 161 organisation for economic co-operation and development (oecd) (2020). gross domestic product expenditure approach, paris: oecd publications. organisation for economic co-operation and development (oecd) (2020). balance of payments balance on goods, paris: oecd publications. omoniyi, o. s., olasunkanmi, o.i., & babatunde o.a. (2012). empirical analysis of twins' deficits in nigeria." ijmbs. 2(3): 38-41. osoro, k. o., gor, s.o., & mbithi m.l. (2014). the twin deficit and the macroeconomic variables in kenya. international journal for innovation education and research, 2(9), 64-84. https://doi.org/10.31686/ijier.vol2.iss9.239 pesaran, m. h., shin, y., & smith. r.p. (1999). pooled mean group estimation of dynamic heterogeneous panels. journal of the american statistical association, 94(446), 621-634. https://doi.org/10.1080/01621459.1999.10474156 purwono, r., mucha, k., & mubin, m. k. (2018). the dynamics of indonesia's current account deficit: analysis of the impact of exchange rate volatility. the journal of asian finance, economics and business, 5(2), 25-33. https://doi.org/10.13106/jafeb.2018.vol5.no2.25 ratha, a. (2012). twin deficits or distant cousins? evidence from india. south asia economic journal, 13(1), 51-68. https://doi.org/10.1177/139156141101300103 rubaszek m., kolasa m., & taglioni, d. (2010). firms in the great global recession: the role of foreign ownership and financial dependence. emerging markets review, 11(4), 341-357. https://doi.org/10.1016/j.ememar.2010.06.001 serenis, d., & tsounis, n., 2012. a new approach for measuring volatility of the exchange rate." procedia economics and finance, 1, 374-82. https://doi.org/10.1016/s2212-5671(12)000433 serenis, d., & tsounis, n. (2015). the effects of exchange rate volatility on sectoral exports. evidence from sweden, the u.k. and germany. international journal of computational economics and econometrics, 5(1): 71-107. https://doi.org/10.1504/ijcee.2015.066204 stockman, a. (1987). the equilibrium approach to exchange rates". federal reserve bank of richmond economic review, march-april, 12-31. stöllinger, r. (2020). tradability of output and the current account in europe. international economics and economic policy, 17(1), 167-218. https://doi.org/10.1007/s10368-019-00449y tornell, a., & lane, p.r., 1998. are windfalls a curse? a non-representative agent model of the current account. journal of international economics, 44(1): 83-112. https://doi.org/10.1016/s0022-1996(97)00016-0 vieira, f. v., & macdonald. r. (2020). the role of exchange rate for current account: a panel data analysis. economia 21(1): 57-72. https://doi.org/10.1016/j.econ.2020.05.002 waliullah, m., kakar, r., & khan, w. (2010). the determinants of pakistan's trade balance: an ardl cointegration approach. the lahore journal of economics, 15 (1): 1-26. https://doi.org/10.35536/lje.2010.v15.i1.a1 https://doi.org/10.31686/ijier.vol2.iss9.239 https://doi.org/10.1080/01621459.1999.10474156 https://doi.org/10.1177/139156141101300103 https://doi.org/10.1016/j.ememar.2010.06.001 https://doi.org/10.1016/s2212-5671(12)00043-3 https://doi.org/10.1016/s2212-5671(12)00043-3 https://doi.org/10.1007/s10368-019-00449-y https://doi.org/10.1007/s10368-019-00449-y https://doi.org/10.1016/s0022-1996(97)00016-0 https://doi.org/10.1016/j.econ.2020.05.002 https://doi.org/10.35536/lje.2010.v15.i1.a1 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 162 yoshida y., & zhai w. (2021). revisiting the glick-rogoff current account model: an application to the current accounts of brics countries. in: dufrénot g., matsuki t. (eds) recent econometric techniques for macroeconomic and financial data. dynamic modeling and econometrics in economics and finance, vol 27. https://doi.org/10.1007/978-3-030-542528_10 https://doi.org/10.1007/978-3-030-54252-8_10 https://doi.org/10.1007/978-3-030-54252-8_10 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 163 appendix the appendix contains the robustness check for the panel ardl model. robustness checks are necessary to improve the credibility of the results presented. by changing the dynamic specification of the model, statistical significance in the long run and the expected results were obtained. along with the robustness check, a ramsey reset was employed, accounting for omitted variable bias and misspecification in the model. the results can be found in the tables below. gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 164 table 5. long-run and short-run coefficients of the current account balance determinants, ardl (1,4,4,4,3,4). series coefficient st. error p-value long-run coefficients rer 3.060*** 0.149 0.000 rir 0.050*** 0.005 0.000 bud -0.059*** 0.005 0.000 gdp 0.004*** 0.001 0.002 erv 9.251*** 0.772 0.000 hausman test 2.35 0.306 error correction coef. (φ) -0.222*** 0.094 0.019 short-run coefficients δcabt-1 -0.324*** 0.094 0.001 δrert 2.123 2.170 0.328 δrert-1 -2.484 1.875 0.185 δrert-2 -0.683 1.990 0.731 δrert-3 -1.689 2.436 0.488 δrert-4 -0.018 1.402 0.990 δrirt -0.002 0.007 0.715 δrirt-1 0.001 0.019 0.931 δrirt-2 0.002 0.016 0.875 δrirt-3 -0.013 0.016 0.403 δrirt-4 0.003 0.009 0.732 δbudt -0.032 0.042 0.446 δbudt-1 -0.008 0.018 0.631 δbudt-2 -0.031 0.023 0.188 δbudt-3 -0.012 0.027 0.633 δbudt-4 -0.016 0.023 0.479 δgdpt 1.455 2.026 0.473 δgdpt-1 3.444 2.495 0.168 δgdpt-2 -1.137 2.994 0.704 δgdpt-3 0.598 2.242 0.790 δervt 0.493 7.850 0.950 δervt-1 -14.640 10.679 0.170 δervt-2 10.668 9.103 0.241 δervt-3 5.762 13.189 0.662 δervt-4 -3.686 10.697 0.730 s1 0.250 0.184 0.175 s2 0.325 0.265 0.221 s3 0.358 0.178 0.045 intercept -0.613*** 0.193 0.001 entry 0.138 0.155 0.373 model specification test ramsey reset test f-statistic 0.296 dynamic specification ardl (1,4,4,4,3,4) estimation method pooled mean group (pmg) adjusting for country-fixed effects and seasonality number of countries 19 time period 2008: q1-2018: q4 (44 periods) observations 836 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 165 table 6. long-run and short-run coefficients of the current account balance determinants, ardl (1,5,4,4,2,4). series coefficient st. error p-value long-run coefficients rer 2.531*** 0.075 0.000 rir 0.050*** 0.003 0.000 bud -0.033*** 0.004 0.000 gdp 0.002*** 0.0003 0.000 erv 7.626*** 0.359 0.000 hausman test 6.011 0.761 error correction coef. (φ) -0.236** 0.119 0.048 short-run coefficients δcabt-1 -0.207** 0.108 0.046 δrert 0.426 1.818 0.815 δrert-1 -2.114 2.552 0.408 δrert-2 -0.094 1.910 0.960 δrert-3 -1.586 1.969 0.420 δrert-4 -0.498 1.567 0.750 δrert-5 1.508 0.991 0.128 δrirt 0.008 0.011 0.471 δrirt-1 -0.008 0.018 0.640 δrirt-2 0.005 0.018 0.764 δrirt-3 0.004 0.012 0.718 δrirt-4 -0.024 0.026 0.358 δbudt -0.041 0.052 0.436 δbudt-1 -0.017 0.023 0.453 δbudt-2 -0.056 0.039 0.153 δbudt-3 -0.033 0.028 0.232 δbudt-4 -0.024 0.027 0.366 δgdpt 0.235 4.431 0.958 δgdpt-1 4.119 4.372 0.346 δgdpt-2 1.248 2.337 0.593 δervt -10.406 9.073 0.251 δervt-1 -11.451 12.431 0.357 δervt-2 13.174 11.920 0.269 δervt-3 -1.614 10.104 0.873 δervt-4 1.724 8.800 0.845 s1 0.445 0.247 0.072 s2 0.398 0.376 0.290 s3 0.353 0.269 0.189 intercept -0.582** 0.233 0.013 entry 0.112 0.102 0.273 model specification test ramsey reset test f-statistic 0.148 dynamic specification ardl (1,5,4,4,2,4) estimation method pooled mean group (pmg) adjusting for country-fixed effects and seasonality number of countries 19 time period 2008: q1-2018: q4 (44 periods) observations 836 gerassimos bertsatos / european journal of government and economics 11(2), december 2022, 141-166 166 table 7. long-run and short-run coefficients of the current account balance determinants, ardl (1,5,4,4,3,3). series coefficient st. error p-value long-run coefficients rer 2.033*** 0.098 0.000 rir 0.042*** 0.003 0.000 bud -0.011** 0.006 0.031 gdp 0.002*** 0.0006 0.000 erv 5.824*** 0.454 0.000 hausman test 1.790 0.597 error correction coef. (φ) -0.268*** 0.126 0.034 short-run coefficients δcabt-1 -0.213** 0.107 0.048 δrert 0.382 1.850 0.836 δrert-1 -1.658 2.160 0.443 δrert-2 -0.456 2.160 0.833 δrert-3 -0.874 2.287 0.702 δrert-4 -0.526 1.401 0.707 δrert-5 2.465** 1.128 0.029 δrirt 0.001 0.011 0.926 δrirt-1 -0.004 0.017 0.780 δrirt-2 0.011 0.018 0.520 δrirt-3 -0.011 0.014 0.432 δrirt-4 -0.008 0.014 0.541 δbudt -0.052 0.050 0.299 δbudt-1 -0.033 0.029 0.245 δbudt-2 -0.052 0.036 0.157 δbudt-3 -0.036 0.031 0.240 δbudt-4 -0.023 0.026 0.364 δgdpt -0.326 3.771 0.931 δgdpt-1 4.906 3.236 0.130 δgdpt-2 2.360 2.048 0.249 δgdpt-3 -1.544 2.335 0.508 δervt -10.518 7.381 0.154 δervt-1 -3.627 11.346 0.749 δervt-2 10.374 9.804 0.290 δervt-3 8.432 12.461 0.499 s1 0.508** 0.241 0.035 s2 0.342 0.334 0.305 s3 0.422 0.235 0.072 intercept -0.500** 0.195 0.011 entry 0.212 0.200 0.291 model specification test ramsey reset test f-statistic 0.862 dynamic specification ardl (1,5,4,4,3,3 estimation method pooled mean group (pmg) adjusting for country-fixed effects and seasonality number of countries 19 time period 2008: q1-2018: q4 (44 periods) observations 836 abstract 1. introduction 2. literature review 3. methodology and data 4. results 5. conclusion and policy implications references appendix european journal of government and economics 10(1), june 2021, 30-45 european journal of government and economics issn: 2254-7088 natural resource abundance and broad-based financial development nexus in asean countries: accounting for globalization and human capital solomon prince nathaniel a* a department of economics, university of lagos, akoka, nigeria * corresponding author at: nathaniel_solomon21@yahoo.com abstract. sustainable resource consumption is important for the development of the financial system. besides, an advanced financial system eases the transfer of revenues from production activities and export to productive investments. the influence of natural resource (nr) abundance on financial development (fd) is still an ongoing debate with conflicting results. however, this study applies a novel proxy for fd, which measures the efficiency, accessibility, and depth of the financial market and institutions. therefore, the current study is a maiden attempt to explore the nexus between fd and nr abundance amidst globalization, human capital, and economic growth in asean economies. reliable panel econometric techniques, including second-generation unit root tests, westerlund (2007) cointegration tests, and the augmented mean group (amg) estimator are employed on the data for the period 1990-2017. the preliminary tests affirm the existence of cross-sectional dependence, unit root, and cointegrating relationship among the variables. the findings from the study reveal that nr abundance reduces fd, while globalization, human capital and economic growth add to fd. a feedback causality exists between nr abundance and fd. thus, this study argues that more investment to the manufacturing sector will ease the attainment of efficiency in financial sector accessibility and benefits from nr abundance. keywords. natural resource abundance; globalization; financial development; human capital; asean; amg. jel codes. e44; n57; p48; q32; q33. doi. https://doi.org/10.17979/ejge.2021.10.1.7202 1. introduction financial development (fd) is one of the strategies used by the world bank to ensure poverty reduction and sustainable growth in developing countries. loans and credits are the essential instruments adopted by the world bank to actualize these objectives (gokmenoglu and rustamov, 2019). the world bank directs these credits/loans to improve the private and financial sector, human and rural development, and public sector governance (world bank, 2017). there is evidence in the literature that most resource-rich/dependent countries are barely financially developed (frenkel 2012). more so, studies have revealed an inverse relationship between fd and natural resource (nr) abundance of resource-rich economies, but there is still no consensus, hence necessitating further probe to the nexus (nathaniel et al. 2020c; gylfason 2001; auty 2003). https://doi.org/10.17979/ejge.2021.10.1.7202 solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 31 in recent times, resource-rich regions (like africa, asia, and latin america) have witnessed minimal economic growth in relation to other regions with seemingly fewer nr (khan et al. 2020). the link between lagging economic performance and nr abundance compared to economies with relatively few resources is termed “resource curse hypothesis” popularized by (auty, 1993). the resource curse phenomenon has dominated recent policy and development discourse. also, it remained an influential area of research for practitioners and economists, mostly from developing countries, after the 1980s. the intention was to challenge/refute the conventional idea of considering nr as a blessing (xu et al. 2016; apergis and payne 2014 among others). the association of southeast asian nations (asean) founded in 1967, headquartered in jakarta, malaysia, is a regional grouping of ten (10) countries (the philippines, laos dpr, cambodia, brunei, malaysia, singapore, myanmar, thailand, vietnam, and indonesia) for security, economic, and political cooperation. the region had an average growth rate of 5.5% in 2018 (imf, 2019). the 21st century was tagged ‘the asian century’ as economic growth was shouldered by asia countries, and particularly countries in the asean bloc (nathaniel 2021; nathaniel & khan 2020). asean is a resource-rich region. the regions nr include bauxite, iron, petroleum, landmass, copper, energy (natural gas, oil, and coal), fertile land, nickel, tin, freshwater, and timber, among others. asean accounts for about 82% and 56% of the world’s total production of natural rubber and tin, respectively. with a growth rate of 5.2% and a collective gdp of 2.6 trillion usd, asean has assumed the status of an important economic bloc (nasir et al., 2019). the primary objective of this study is to examine the effect of nr abundance, globalization, and human capital on fd in asean. this study is super useful for asean countries considering the fact that it is a resource-rich region and has witnessed significant growth over the years. again, the region is opened to trade, which makes it globalized, but little or nothing is known as regards the influence of nr abundance, globalization, and human capital on fd in asean. this was the motivation for this study. thus, an adequate knowledge of the relationship between nr abundance and fd is required for better policy coordination and economic expansion. hence, any impact of nr abundance on fd is likely to ease the pace of economic growth in asean. in addition, exploring the relationship between fd and nr abundance will provide new insights for decision-makers in asean to utilize nr more as a blessing other than a curse. a sound financial system is vital for the efficient utilization and sustainable use of nr and the stimulation of economic growth (nathaniel et al., 2020e). fd is required to attain higher economic growth (redmond and nasir 2020; sun et al. 2020; nathaniel et al. 2020b; nawaz et al. 2019). a developed financial sector facilitates the transfer of funds, stimulate savings, create investment opportunities, encourage corporate control, drives innovation, and promotes risk management (murshed et al. 2020b; gokmenoglu and rustamov, 2019). asean is considered an economically dynamic region with laudable policies directed towards economic openness, heralding trade liberalization and globalization. considerable globalization and natural resource consumption, together with persistent economic growth, necessitated the issue of fd. globalization is an international phenomenon that economically and socially impacts solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 32 human lives, in terms of poverty reduction, and the financial well-being of the economy (ahmed et al. 2020a). evidence pertaining to the influence of globalization on the climate, economic growth, and inequality has been discussed in the literature, but the exact effect of globalization on fd has not been figured out (murshed et al., 2020a). globalization is becoming more and more important in asean, which becomes a powerful promoter of its financial sector. generally speaking, globalization can decrease the taxes and tariffs, and bring openness to trade and fd, which can boost economic growth (ulucak & khan, 2020). fd can decline the supporting costs of increasing financing networks, which could trigger enterprises to make more investment in buying new equipment (umar et al., 2020). this study adds to the literature in the following ways: (i) this is a maiden attempt to examine the impact of nr abundance on fd in asean (ii) unlike previous studies, this study is the first to adopt the recently developed broad-based financial development index of the imf to examine the fd-nr abundance nexus. this index is superior to those used by earlier studies in that, it considers the complex multidimensional nature of fd. it summarizes how developed financial institutions and financial markets are; in terms of access, depth, and efficiency. in addition, this study considered a more comprehensive human capital indicator which provides adjusted estimated returns to education for each country and covers labour market information. with these, the policy relevance of this study is assured. (iii) panel data have lots of issues (like serial correlation, heteroskedasticity, cross-sectional dependence (cd), heterogeneity, etc.) which could lead to inefficient and biased outcomes if ignored. the core of these issues is cd and heterogeneity (dogan et al. 2020). this study, unlike previous studies, adequately deal with these issues to obtain robust estimates via advanced econometrics techniques. the augmented mean group (amg) estimator, driscoll-kraay (dk), and the prais-winsten regression (otherwise known as panel-corrected standard errors (pcse) approach), have been applied to deal with these issues. the study is arranged as follows: section 2 presents the literature review which encompasses the theoretical framework and empirical review. section 3 addresses the methodology. results are presented and discussed in section 4. finally, section 5 concludes. 2. literature review 2.1. theoretical framework theoretically, the channels through which nr abundance adversely affects fd has been highlighted by beck (2002) who argued that nr sectors draw investment and skills away from financial sectors, concomitantly declining demand saving rates. in a similar gesture, rajan and zingales (2003) postulate the interest group theory of fd, which argued that existing powerful firms are always against fd by using their market power as an instrument to circumvent competition. in the presence of human capital, globalization, and economic growth, this study explores the nexus between fd and nr in asean. fd is determined by nr, human capital, and solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 33 globalization which, in turn, affects economic growth (yu et al. 2020; nawaz et al., 2019). a sound financial system distributes nr wealth for lucrative investment projects that can promote growth (shahbaz et al., 2018). nr could serve as an engine of growth and development, rather than the only driver of growth (badeeb et al., 2017). human capital enhances financial growth through the efficient utilization of nr (tiba and frikha, 2019). human capital adds to the effective use of nr as well as the growth and stability of the financial system (zaidi et al., 2019). an educated investor, as opposed to uneducated and unskilled people, can efficiently utilize financial resources (hatemi-j and shamsuddin, 2016). 2.2. empirical review according to the classical resource abundance studies, nr is a blessing for the host countries. however, theories like the “dutch disease” and “resource curse,” have provided evidence that these resources (nr) could impede the economic growth of the countries concerned. available studies affirmed that the blessings associated with nr abundance could transform to a curse amidst rent-seeking dependency, weak institutional management, low literacy rate, dutch disease, and poor human capital development, among others (nathaniel et al. 2020a; ahmed et al. 2020a; shahbaz et al., 2018a; dwumfour and ntow-gyamfi, 2018). besides, this “curse” is evidence in resource abundance economies. khan et al. (2020) explored the impact of nr, technological innovations, and human capital on fd in china from 1987–2017. their findings confirmed the negative impact of nr on fd in china. also, trade openness, technological innovations, and human capital exact a positive impact on fd. the authors argued for the development of human capital and technological innovations to ensure the sustainable use of nr to enhance fd. dwumfour and ntow-gyamfi (2018) discovered that nr has an unclear impact on fd in thirty-eight african countries. they attributed this to a weak institutional framework in the region. institutions, especially political institution, could minimize the nr curse in developing economies by promoting sustainable resource use in resource-rich countries (bhattacharyya and hodler 2014; mehlum et al., 2006; humphreys et al. 2007). guan et al. (2020) applied the fmols technique to investigate the nr-fd nexus in china while controlling for globalization, economic growth, and human capital. analogous to the study of khan et al. (2020), they discovered that nr is inimical to financial sector development, while globalization, economic growth, and human capital promote fd in china. further findings revealed a one-way causality, in the long run, from nr, economic growth, and human capital to fd. this further corroborates the findings of asif et al. (2020) for china. nevertheless, the majority of the recent studies affirmed that nr deteriorates fd in developing countries (asif et al., 2020). there is also a growing literature on how nr promotes fd in highincome countries (shahbaz et al., 2018a,b). gokmenoglu and rustamov (2019) examined the effect of nr on fd for selected countries, including turkmenistan, from 1992 to 2017. their findings confirmed that nr abundance plays a fascinating role in promoting fd in azerbaijan, kazakhstan, turkmenistan, and russia. zaidi et al. (2019) investigated the effects of nr, globalization, and human capital on fd in 31 oecd countries from 1990 to 2016. they discovered solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 34 that economic growth, nr, globalization, capital formation, and human capital exact a positive impact on fd. studies like (bravo-ortega and de gregorio 2005; behbudi et al. 2010; marchand and weber 2015; sibel et al. 2015; rickman et al. 2017; khan et al. 2020) have shown the positive impact of human capital in promoting fd. recent studies have also linked globalization and human capital to environmental degradation (ahmed et al. 2021a,b; ahmed et al. 2019; ahmed et al. 2020a,b,c; ahmed and wang 2019) now, the reviewed studies point to inconsistent results as regards the nexus between nr and fd. again, the majority of the studies are for a single country case, and there is/are no single study(s) on the asean economies which should be an attractive case study considering the region’s resource abundance, unprecedented growth, openness to trade, and budding financial sector. therefore, there is a dire need to examine the nr-fd nexus for the asean economies, including factors like human capital, globalization, and economic growth. 3. methodology 3.1. model construct and data source from the above analogy, a framework has been developed to investigate the impact of nr abundance on fd in asean by introducing human capital, economic growth, and globalization. the functional form of the model is given as: 𝐹𝐹𝐹𝐹𝑡𝑡 = 𝑓𝑓(𝑁𝑁𝑁𝑁𝑡𝑡, 𝐻𝐻𝐻𝐻𝑡𝑡 , 𝐺𝐺𝐺𝐺𝑡𝑡 , 𝐺𝐺𝑁𝑁𝑡𝑡) (1) where 𝐹𝐹𝐹𝐹𝑡𝑡, 𝑁𝑁𝑁𝑁𝑡𝑡, 𝐻𝐻𝐻𝐻𝑡𝑡, 𝐺𝐺𝐺𝐺𝑡𝑡, and 𝐺𝐺𝑁𝑁𝑡𝑡 represent financial development, natural resource, human capital, globalization, and economic growth, respectively. we linearized and transformed eq. (1) by taking the natural logarithm of the variables in line with the recent studies of khan et al. (2020) and sun et al. (2020) since log-linear models give reliable empirical results in elasticities (meo et al., 2020a,b). 𝑙𝑙𝑙𝑙𝑓𝑓𝑙𝑙𝑖𝑖,𝑡𝑡 = 𝜉𝜉0 + 𝜉𝜉1ln (𝑙𝑙𝑛𝑛)𝑖𝑖𝑡𝑡 + 𝜉𝜉2ln (ℎ𝑐𝑐)𝑖𝑖𝑡𝑡 + 𝜉𝜉3ln (𝑔𝑔𝑔𝑔)𝑖𝑖𝑡𝑡 + 𝜉𝜉4ln (𝑔𝑔𝑛𝑛)𝑖𝑖𝑡𝑡 + 𝜇𝜇𝑖𝑖𝑡𝑡 (2) where 𝑖𝑖 = 1,2, 3, …𝑁𝑁 for individual countries. 𝑡𝑡 = 1,2, 3, …𝑇𝑇 for time. detailed information on the variables are presented in table 1. the data for the study spans 1990-2017 for eight (8) asean countries, a decision constraint by data availability. table 1. description of variables. s/n variables measurement source supporting studies 1. natural resource natural resource rent (% of gdp) wdi (2019) sun et al. (2020) 2. gdp per capita in constant 2010 usd wdi (2019) nathaniel et al. (2020a) 3. globalization overall kof index kof (2017) guan et al. (2020) 4. financial development broad‐based index of financial depth imf (2016) kassouri & altıntaş (2020) access and efficiency 5. human capital human capital index penn world table khan et al. (2020) sources: author’s compilation. solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 35 3.2. econometric procedure 3.2.1. cross-sectional dependence and unit root tests it is always necessary to examine whether individual units in the panel are independent or not, as this will help to overturn biased estimates (omojolaibi and nathaniel 2020; adedoyin et al. 2020; adeleye et al. 2020). international treaties, trade agreements, and spillover effect are the possible causes of cd (chudik et al., 2016); and the asean countries have signed more than three hundred and fifty agreements after its inception, and are also signatories to various international agreements. therefore, it is possible that the cross-sections are not independent. this study applies the pesaran (2004) tests to investigate the dependence/independence of the crosssections. the tests equation is given as: 𝐻𝐻𝐹𝐹 = � 2𝑇𝑇 𝑁𝑁(𝑁𝑁−1) �∑ ∑ 𝜌𝜌𝑖𝑖𝜌𝜌𝑁𝑁−1 𝑗𝑗=𝑖𝑖+1 𝑁𝑁−1 𝑖𝑖=0 � 𝑁𝑁 (0,1), where n and t are the cross-sections and time horizon, respectively. ρij stands for cross-sections correlation of error between i and j. this test is suitable for this study due to the nature of our panel, given that 𝑇𝑇(time dimension) > 𝑁𝑁 (cross-section). in a situation where cd exist, secondgeneration tests are preferred. as such, the cadf (cross-sectionally augmented adf) and cips (cross-sectionally augmented ips) of pesaran (2007) are applied in this study. the cadf test equation is given as: ∆𝑦𝑦𝑖𝑖𝑡𝑡 = ∆𝜑𝜑𝑖𝑖𝑡𝑡 + 𝛽𝛽𝑖𝑖𝑥𝑥𝑖𝑖𝑡𝑡−1 + 𝜌𝜌𝑖𝑖𝑇𝑇 + ∑ 𝜃𝜃𝑖𝑖𝑗𝑗𝑛𝑛 𝑗𝑗=1 ∆𝑥𝑥𝑖𝑖,𝑡𝑡−𝑗𝑗 + 𝜀𝜀𝑖𝑖𝑡𝑡 , where 𝜑𝜑𝑖𝑖𝑡𝑡 , 𝑥𝑥𝑖𝑖𝑡𝑡 ,∆, 𝑇𝑇 and 𝜀𝜀𝑖𝑖𝑡𝑡 represent the intercept, study variables, difference operator, time span, and disturbance term respectively. 𝜌𝜌𝑖𝑖 is the proxy of the unobservable common factor, which pesaran (2007) introduced to eliminate cd emanating from common shocks that might affect all the units. previous studies have applied these tests (cips and cadf) amidst cd in the dataset (murshed 2020; saint akadırı et al. 2020). 3.2.2. cointegration and parameter estimation tests for long-run relationship are important for non-stationary variables, especially when variables are integrated at the same order, let’s say i(1). the study preferred the westerlund (2007) cointegration test because it has a greater explanatory power, and also robust even in the presence of cd and nuisance from endogeneity. the test constructs four statistics; the group mean statistics, 𝐺𝐺𝜏𝜏 = 1 𝑁𝑁 ∑ 𝛼𝛼𝚤𝚤� 𝑆𝑆𝑆𝑆(𝛼𝛼𝚤𝚤)� 𝑁𝑁 𝑖𝑖=1 and 𝐺𝐺𝛼𝛼 = 1 𝑁𝑁 ∑ 𝑇𝑇 𝛼𝛼𝚤𝚤� 𝛼𝛼𝚤𝚤�(1) 𝑁𝑁 𝑖𝑖=1 , and the panel mean tests, 𝑃𝑃𝜏𝜏 = 𝛼𝛼𝚤𝚤� 𝑆𝑆𝑆𝑆(𝛼𝛼𝚤𝚤)� and 𝑃𝑃𝛼𝛼 = 𝑇𝑇𝛼𝛼�. the former examines cointegration of the whole panel, while the latter explores the existence of cointegration in at least one of the units. since cointgration does not suggest long-run impact (li et al. 2020), the amg estimator, solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 36 popularized by bond & eberhardt (2013), is applied for parameter estimation, while the dk and pcse were adopted to confirm the consistency of the amg results. the amg estimator involves a two-step procedure: amg stage 1: ∆𝑦𝑦𝑖𝑖𝑡𝑡 = 𝛼𝛼𝑖𝑖 + 𝑔𝑔𝑖𝑖∆𝑥𝑥𝑖𝑖𝑡𝑡 + 𝑐𝑐𝑖𝑖𝑓𝑓𝑡𝑡 + ∑ 𝑙𝑙𝑡𝑡𝑇𝑇 𝑡𝑡=2 ∆𝐹𝐹𝑡𝑡 + 𝑒𝑒𝑖𝑖𝑡𝑡 amg stage 2: 𝑔𝑔�𝐴𝐴𝐴𝐴𝐴𝐴 = 𝑁𝑁−1 ∑ 𝑔𝑔�𝑖𝑖𝑁𝑁 𝑖𝑖=1 𝑥𝑥𝑖𝑖𝑡𝑡 and 𝑦𝑦𝑖𝑖𝑡𝑡 are the observables. 𝑓𝑓𝑡𝑡 represents the unobserved common factor. the countryspecific estimates of coefficients, the amg estimator, and the time dummies are respectively 𝑔𝑔𝑡𝑡, 𝑔𝑔�𝐴𝐴𝐴𝐴𝐴𝐴 , and 𝑙𝑙𝑡𝑡. this test was preferred because it suits the nature of our panel (𝑇𝑇 > 𝑁𝑁), it adequately addresses the two core panel data issues (cd and heterogeneity) and shows country-wise results which could inform the alignment of policies to suit countries peculiarities. the amg is also appropriate for nonstationary data typical of our panel. 4. results and discussion this section presents the trend of the variables, descriptive statistic and correlation, cd test, unit root, cointegration, and the parameter estimation tests (amg, pcse, and dk). from figure 1, malaysia, thailand and singapore are the most financially developed countries in asean, while laos dpr is the least developed. as shown in figure 2, all the countries are getting increasingly globalized. however, singapore is the most globalized, while lao dpr is the least globalized. in figure 3, nr rent is higher in brunei compared to the other asean countries. trend of the variables figure 1. fd in asean. figure 2. globalization in asean. 0 .2 .4 .6 .8 fi na nc ia l d ev el op m en t 1990 2000 2010 2020 year brunei thailand indonesia malaysia singapore philippines vietnam lao dpr 20 40 60 80 10 0 g lo ba liz at io n 1990 2000 2010 2020 year brunei thailand indonesia malaysia singapore philippines vietnam lao dpr solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 37 figure 3. nr abundance in asean. figure 4. human capital in asean. figure 5. economic growth in asean. singapore has witnessed more economic expansion (in terms of gdp growth) and human capital development in relation to the remaining countries in asean as shown in figure 4 and figure 5 respectively. table 2. descriptive statistic and correlation. fd nr gb gr hc mean 0.390 8.064 59.25 1163 2.384 max. 0.799 38.37 85.34 5674 3.947 mini. 0.001 0.000 24.00 433.2 1.512 std. d 0.190 8.081 14.73 1.547 0.416 correlation fd 1 nr -0.367 1 gb 0.329 -0.284 1 gr 0.409 0.263 0.422 1 hc 0.170 -0.007 0.384 0.634 1 source: author’s computation. table 2 reports the properties and correlation of the variables. economic growth has the highest average while fd has the least. these findings reveal that economic growth has been increasing faster than fd in asean countries. fd and nr have a minimum value of 0.001 and 0.000, respectively. all the variables are positively associated with fd, except nr. economic growth and human capital show a positive correlation with globalization, while globalization is negatively associated with nr. 0 10 20 30 40 n at ur al r es ou rc es a bu nd an ce 1990 2000 2010 2020 year brunei thailand indonesia malaysia singapore philippines vietnam lao dpr 1. 5 2 2. 5 3 3. 5 4 h um an c ap ita l 1990 2000 2010 2020 year brunei thailand indonesia malaysia singapore philippines vietnam lao dpr 0 20 00 0 40 00 0 60 00 0 ec on om ic g ro w th 1990 2000 2010 2020 year brunei thailand indonesia malaysia singapore philippines vietnam lao dpr solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 38 table 3. cross-sectional dependence result. variables breusch-pagan lm pesaran scaled lm pesaran cd lnef 621.3251a 21.56473a 15.57385a lngr 345.3425a 45.67483a 45.67464a lngr 153.4738a 43.57382a 23.68937a lnhc 214.4636a 98.56785a 44.64445a lnnr 327.3524a 67.78294a 34.76589a source: author’s computation. note: ‘a’ represents significance at 1% level. table 3 confirms the existence of cd across the three tests. table 4 revealed that the variables are non-stationary at i(0), but i(1). therefore, with i(1) variables, cointegration is a possibility. the westerlund (2007) test affirms cointegration as ga, pt, and pa are significant. table 5 reports the amg, dk, and pcse results. the focus is on the amg results. the dk and pcse are applied to check the robustness of the amg results. from the findings, nr abundance reduces fd. this is consistent with the findings of sun et al. (2020) and guan et al. (2020) for seven emerging economies and china, respectively. table 4. unit root and cointegration results. variables level first difference cips cadf cips cadf fd (log) -2.621 10.62 -5.387a 21.21a nr (log) -2.779 11.32 -5.440a 25.12a hc (log) -1.528 11.82 -1.911b 23.56a gb (log) -1.782 10.67 -4.314a 28.12a gr (log) westerlund (2007) -0.733 gt -3.016a 11.92 ga 8.782 -2.624a pt -7.474a 20.43a pa -9.783b note: a and b represent statistical significance at 1% and 10% levels, respectively. source: author’s computation. this suggests a trade-off between both variables. this finding could be attributed to the fact that asean countries are still emerging, with weak institutional quality, and difficulty in efficiently managing the available nr. as such, the resource curse phenomenon, as it relates to fd, still exist. another plausible reason is the increase in nr exports which decline the number of investments channelled to the manufacturing sector in asean. more so, crowding-out investment from the manufacturing and industrial sector, as well as, inadequate infrastructure could be the other potential reasons for a decrease in fd in asean. further findings showed that human capital and globalization promote fd, though the influence of the former is not significant. human capital is vital for fd. it helps in efficient utilization of nr, promotes financial inclusion and financial literacy, which contribute to fd. besides, education is necessary for fd. an uneducated human capital is unaware of the mechanism of the financial sector, hence it may contribute little or nothing to its development. more human capital triggers productivity and skill-building economies, which is associated with more opportunities in the financial sector. however, the findings confirmed that solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 39 human capital is not yet at a desirable level to efficiently contribute to fd in the asean bloc. table 5. amg, dk, and pcse results. variables amg driscoll/kraay pcse nr (log) -0.010 (-3.91) a -0.001 (-9.11) a -0.019 (-4.20) a hc (log) 1.246 (0.21) 0.080 (1.31) 0.209 (1.56) gb (log) 0.123 (3.45) a 0.538 (10.5) a 1.877 (14.8) a gr (log) 0.327 (2.96)a 0.009 (6.88)a 0.011 (0.49) _cons. -2.219 (-3.35) a -1.786 (-7.60) a -8.389 (-20.8)a source: author’s computation. note: ‘a’ represents significance at 1% level. the z/t-values are in parenthesis. therefore, human capital development should be top on the agenda of policymakers in asean, especially when enacting policies that relate to financial sector development. on the other hand, globalization opens up an economy. it allows for the importation of advanced technologies which could boost manufacturing, promote gainful trade, enhance productivity, and ensure fd. this is in line with the findings of zaidi et al. (2019) and guan et al. (2020). just like globalization, economic growth adds to fd in asean, consistent with the findings of (nawaz et al. 2019; sun et al. 2020). this suggests that economic expansion in asean have created jobs and resulted in increased wage (purchasing power), leading to higher investment and consumption which promotes financial services, and hence increase fd. table 6. country-specific amg results. countries lnnr lnhc lngb lngr brunei -43.1 (-1.96)b 34.1 (4.43)a -1.31 (-0.96) 1.19 (0.49) thailand -0.11 (-1.31) 0.62 (5.88)a 0.40 (3.09)a -0.80 (-1.56) indonesia 0.05 (0.90) -3.09 (-3.91)a 0.69 (1.68)c 0.07 (0.27) malaysia -0.00 (-0.13) -0.34 (-1.13) 0.75 (8.23)a 0.02 (2.07)b singapore philippines vietnam lao dpr -0.13 (-3.04)a -0.11 (-3.30)a 0.00 (2.08)b 0.10 (1.65)c -0.34 (-1.13) -9.84 (-1.46) 2.85 (0.61) -22.1 (-2.81)a 0.10 (7.23)a 1.11 (1.79)c -0.27 (-4.21)a 0.32 (0.68) 1.21 (3.89)a 0.04 (3.34)a 1.87 (3.34)a 1.84 (1.93)b source: author’s computation. note: ‘a’, ‘b’, and ‘c’ represent significance at 1%, 5%, and 10% levels respectively. the z/t-values are in parenthesis. the dk and pcse results affirmed the outcome of the amg estimator. therefore, similar explanation applies. in table 6, nr improves fd only in vietnam, laos dpr, and indonesia. the influence of human capital is mixed, but it promotes fd in brunei, thailand, and vietnam. globalization harms fd in brunei and vietnam, while economic growth is not compatible with fd in thailand. solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 40 table 7. dumitrescu & hurlin (dh) (2012) results. null hypothesis w-stat. zbar-stat. probability decision lnnr → lnfd lnfd → lnnr 10.01 9.376 9.987 7.908 0.000 0.000 bidirectional causality lnhc → lnfd lnfd → lnhc 7.650 2.988 5.783 1.345 0.003 0.299 unidirectional causality lngb → lnfd lnfd → lngb 8.910 5.382 7.557 4.185 0.000 0.001 bidirectional causality lngr → lnfd lnfd → lngr 5.780 4.799 2.962 2.278 0.003 0.023 bidirectional causality lnnr → lngr lngr → lnnr 6.789 2.560 4.129 1.519 0.000 0.212 unidirectional causality lngb → lngr lngr → lngb 1.078 1.346 0.347 0.667 0.565 0.765 no causality source: author’s computation. note: ‘→’ shows the direction of causality. the results in table 7 suggests a feedback causality between nr and fd, globalization and fd, and between economic growth and fd. this further confirms the link between fd and nr, and why nr and economic growth policies should consider/be connected with policies that relate to financial sector development, vice versa. 5. conclusion and policy implications this study examined the nexus between fd and nr for eight (8) asean countries from 1990 2017. the preliminary estimation procedures involve the investigation of possible cd among the cross-sections. the three cd tests exposed the presence of cd which necessitated the adoption of robust second-generation unit root and cointegration tests to ameliorate the possible adverse effects associated with cd. besides, the existence of a long-run relationship among the variables informed the choice of the amg estimator. the causal direction between variables is investigated by using the dh panel causality method. the amg results showed that human capital, globalization, and economic growth promote fd. on the flipside, nr abundance reduces fd, which confirms the resource curse hypothesis for asean economies. the countrywise results are mixed. however, economic growth harms fd only in thailand. these findings necessitate relevant policy directions. in line with these findings, this study argues that more investment in the industrial and manufacturing sectors will help asean economies attain efficiency in financial sector accessibility, efficiency, depth, and enhance the benefits from their nr abundance. the most serious issue is to manage nr to support fd; nr can be transformed into value-added exports to enhance export revenues and promote fd. therefore, revisiting nr abundance utilization for a more productive output is recommended. since human capital and economic growth trigger fd, policies regarding sustainable growth should be maintained and the formation of human capital should be focused. more so, there is a dire need to develop human capital in asean economies. an investment in human capital will enhance the available labour market with skills to direct the benefits from trade to the financial sector. it will also promote financial inclusion, and help in the efficient utilization of nr for fd. the estimates of the causality test indicate a feedback causality solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 41 between economic growth and fd, globalization and fd, and between nr and fd. these further confirm the link between nr, globalization, and fd in the asean bloc. thus, policies to reduce customs duties, taxes, and trade restrictions, and boosting foreign investment would be the right option in the asean region as globalization increases fd. however, trade and foreign investment in environmentally friendly technology should be encouraged to reap the economic and environmental benefits of globalization. in conclusion, future investigation of the interaction between nr and fd in different countries/regions may lead to more generalizable results, as financial policies and nr abundance vary from country to country/region to region. future studies may need to incorporate the moderating role of technical innovations and governance with nr to re-examine fd using other advanced measures. references adedoyin, f. f., nathaniel, s., adeleye, n. (2020). an investigation into the anthropogenic nexus among consumption of energy, tourism, and economic growth: do economic policy uncertainties matter? environmental science and pollution research, 1-13. https://doi.org/10.1007/s11356-020-10638-x adeleye, b. n., adedoyin, f., nathaniel, s. (2020). the criticality of ict-trade nexus on economic and inclusive growth. information technology for development, 1-21. https://doi.org/10.1080/02681102.2020.1840323 ahmed, z., wang, z. (2019). investigating the impact of human capital on the ecological footprint in india: an empirical analysis. environmental science and pollution research, 26(26), 26782-26796. https://doi.org/10.1007/s11356-019-05911-7 ahmed, z., asghar, m. m., malik, m. n., nawaz, k. (2020c). moving towards a sustainable environment: the dynamic linkage between natural resources, human capital, urbanization, economic growth, and ecological footprint in china. resources policy, 67, 101677. https://doi.org/10.1016/j.resourpol.2020.101677 ahmed, z., cary, m., le, h. p. (2021a). accounting asymmetries in the long-run nexus between globalization and environmental sustainability in the united states: an aggregated and disaggregated investigation. environmental impact assessment review, 86, 106511. https://doi.org/10.1016/j.eiar.2020.106511 ahmed, z., nathaniel, s. p., shahbaz, m. (2020a). the criticality of information and communication technology and human capital in environmental sustainability: evidence from latin american and caribbean countries. journal of cleaner production, 125529. https://doi.org/10.1016/j.eiar.2020.106511 ahmed, z., wang, z., mahmood, f., hafeez, m., ali, n. (2019). does globalization increase the ecological footprint? empirical evidence from malaysia. environmental science and pollution research, 26(18), 18565-18582. https://doi.org/10.1007/s11356-019-05224-9 ahmed, z., zafar, m. w., ali, s. (2020b). linking urbanization, human capital, and the ecological footprint in g7 countries: an empirical analysis. sustain. cities soc, 55, 102064. https://doi.org/10.1016/j.scs.2020.102064 ahmed, z., zhang, b., cary, m. (2021b). linking economic globalization, economic growth, financial development, and ecological footprint: evidence from symmetric and asymmetric ardl. ecological indicators, 121, 107060. https://doi.org/10.1016/j.ecolind.2020.107060 apergis, n., payne, j. e. (2014). the oil curse, institutional quality, and growth in mena countries: evidence from time-varying cointegration. energy economics, 46, 1-9. https://doi.org/10.1016/j.eneco.2014.08.026 https://doi.org/10.1007/s11356-020-10638-x https://doi.org/10.1080/02681102.2020.1840323 https://doi.org/10.1007/s11356-019-05911-7 https://doi.org/10.1016/j.resourpol.2020.101677 https://doi.org/10.1016/j.eiar.2020.106511 https://doi.org/10.1016/j.eiar.2020.106511 https://doi.org/10.1007/s11356-019-05224-9 https://doi.org/10.1016/j.scs.2020.102064 https://doi.org/10.1016/j.ecolind.2020.107060 https://doi.org/10.1016/j.eneco.2014.08.026 solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 42 asif, m., khan, k. b., anser, m. k., nassani, a. a., abro, m. m. q., zaman, k. (2020). dynamic interaction between financial development and natural resources: evaluating the ‘resource curse’hypothesis. resources policy, 65, 101566. https://doi.org/10.1016/j.resourpol.2019 .101566 auty, r., (1993). sustaining development in mineral economies: the resource curse thesis. routledge, london. badeeb, r. a., lean, h. h., clark, j. (2017). the evolution of the natural resource curse thesis: a critical literature survey. resources policy, 51, 123-134. https://doi.org/10.1016/ j.resourpol.2016.10.015 beck, t. (2002). financial development and international trade: is there a link? journal of international economics, 57(1), 107-131. https://doi.org/10.1016/s0022-1996(01)00131-3 behbudi, d., mamipour, s., karami, a. (2010). natural resource abundance, human capital and economic growth in the petroleum exporting countries. journal of economic development, 35(3), 81. https://doi.org/10.35866/caujed.2010.35.3.004 bhattacharyya, s., hodler, r. (2014). do natural resource revenues hinder financial development? the role of political institutions. world development, 57, 101-113. https://doi.org/10.1016/j.worlddev.2013.12.003 bond, s., eberhardt, m. (2013). accounting for unobserved heterogeneity in panel time series models. university of oxford. bravo-ortega, c., & de gregorio, j. (2005). the relative richness of the poor? natural resources, human capital, and economic growth. the world bank. https://doi.org/10.1596/1813-94503484 chudik, a., mohaddes, k., pesaran, m. h., raissi, m. (2016). long-run effects in large heterogeneous panel data models with cross-sectionally correlated errors. emerald group publishing limited. https://doi.org/10.1108/s0731-905320160000036013 dogan, e., ulucak, r., kocak, e., isik, c. (2020). the use of ecological footprint in estimating the environmental kuznets curve hypothesis for bricst by considering cross-section dependence and heterogeneity. the science of the total environment, 723, 138063. https://doi.org/10.1016/j.scitotenv.2020.138063 dumitrescu, e. i., hurlin, c. (2012). testing for granger non-causality in heterogeneous panels. economic modelling, 29(4), 1450-1460. https://doi.org/10.1016/j.econmod.2012.02.014 dwumfour, r. a., ntow-gyamfi, m. (2018). natural resources, financial development and institutional quality in africa: is there a resource curse? resources policy, 59, 411-426. https://doi.org/10.1016/j.resourpol.2018.08.012 gokmenoglu, k. k., rustamov, b. (2019). examining the world bank group lending and natural resource abundance induced financial development in kart countries. resources policy, 63, 101433. https://doi.org/10.1016/j.resourpol.2019.101433 guan, j., kirikkaleli, d., bibi, a., zhang, w. (2020). natural resources rents nexus with financial development in the presence of globalization: is the “resource curse” exist or myth? resources policy, 66, 101641. https://doi.org/10.1016/j.resourpol.2020.101641 gylfason, t. (2001). natural resources, education, and economic development. european economic review, 45(4-6), 847-859. https://doi.org/10.1016/s0014-2921(01)00127-1 hatemi-j, a., & shamsuddin, m. (2016). the causal interaction between financial development and human development in bangladesh. applied economics letters, 23(14), 995-998. https://doi.org/10.1080/13504851.2015.1128066 humphreys, m., sachs, j. d., stiglitz, j. e., soros, g., humphreys, m. (2007). escaping the resource curse. columbia university press. international monetary fund (imf) (2019). world economic outlook database, october 2019. retrieved 11 october 2019. international monetary fund (imf) (2016). introducing a new broad‐based index of financial development, imf working paper wp/16/5. https://data.imf.org/?sk=388dfa60-1d26-4adeb505-a05a558d9a42 kassouri, y., altıntaş, h. (2020). human well-being versus ecological footprint in mena countries: a trade-off? journal of environmental management, 263, 110405. https://doi.org/10.1016/j.jenvman.2020.110405 https://doi.org/10.1016/j.resourpol.2019.101566 https://doi.org/10.1016/j.resourpol.2019.101566 https://doi.org/10.1016/j.resourpol.2016.10.015 https://doi.org/10.1016/j.resourpol.2016.10.015 https://doi.org/10.1016/s0022-1996(01)00131-3 https://doi.org/10.35866/caujed.2010.35.3.004 https://doi.org/10.1016/j.worlddev.2013.12.003 https://doi.org/10.1596/1813-9450-3484 https://doi.org/10.1596/1813-9450-3484 https://doi.org/10.1108/s0731-905320160000036013 https://doi.org/10.1016/j.scitotenv.2020.138063 https://doi.org/10.1016/j.econmod.2012.02.014 https://doi.org/10.1016/j.resourpol.2018.08.012 https://doi.org/10.1016/j.resourpol.2019.101433 https://doi.org/10.1016/j.resourpol.2020.101641 https://doi.org/10.1016/s0014-2921(01)00127-1 https://doi.org/10.1080/13504851.2015.1128066 https://www.imf.org/external/pubs/ft/weo/2019/02/weodata/index.aspx https://data.imf.org/?sk=388dfa60-1d26-4ade-b505-a05a558d9a42 https://data.imf.org/?sk=388dfa60-1d26-4ade-b505-a05a558d9a42 https://doi.org/10.1016/j.jenvman.2020.110405 solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 43 khan, z., hussain, m., shahbaz, m., yang, s., jiao, z. (2020). natural resource abundance, technological innovation, and human capital nexus with financial development: a case study of china. resources policy, 65, 101585. https://doi.org/10.1016/j.resourpol.2020.101585 li, z. z., li, r. y. m., malik, m. y., murshed, m., khan, z., umar, m. (2020). determinants of carbon emission in china: how good is green investment? sustainable production and consumption, 27, 392-401. https://doi.org/10.1016/j.spc.2020.11.008 marchand, j. t., weber, j. (2015). the labor market and school finance effects of the texas shale boom on teacher quality and student achievement (no. 2015-15). university of alberta, department of economics. mehlum, h., moene, k., torvik, r. (2006). cursed by resources or institutions? world economy, 29(8), 1117-1131. https://doi.org/10.1111/j.1467-9701.2006.00808.x meo, m. s., nathaniel, s. p., khan, m. m., nisar, q. a., fatima, t. (2020b). does temperature contribute to environment degradation? pakistani experience based on nonlinear bounds testing approach. global business review, 0972150920916653. https://doi.org/10.1177 /0972150920916653 meo, m., nathaniel, s., shaikh, g., kumar, a. (2020a). energy consumption, institutional quality and tourist arrival in pakistan: is the nexus (a) symmetric amidst structural breaks? journal of public affairs, e2213. https://doi.org/10.1002/pa.2213 murshed, m. (2020). an empirical analysis of the non-linear impacts of ict-trade openness on renewable energy transition, energy efficiency, clean cooking fuel access and environmental sustainability in south asia. environmental science and pollution research, 27(29), 3625436281. https://doi.org/10.1007/s11356-020-09497-3 murshed, m., abbass, k., rashid, s. (2020a). modelling renewable energy adoption across south asian economies: empirical evidence from bangladesh, india, pakistan and sri lanka. international journal of finance & economics. https://doi.org/10.1002/ijfe.2073 murshed, m., nurmakhanova, m., elheddad, m., ahmed, r. (2020b). value addition in the services sector and its heterogeneous impacts on co 2 emissions: revisiting the ekc hypothesis for the opec using panel spatial estimation techniques. environmental science and pollution research, 27(31), 38951-38973. https://doi.org/10.1007/s11356-020-09593-4 nasir, m. a., huynh, t. l. d., tram, h. t. x. (2019). role of financial development, economic growth & foreign direct investment in driving climate change: a case of emerging asean. journal of environmental management, 242, 131-141. https://doi.org/10.1016/j.jenvman .2019.03.112 nathaniel, s. p. (2021). environmental degradation in asean: assessing the criticality of natural resources abundance, economic growth and human capital. environmental science and pollution research, 1-13. https://doi.org/10.1007/s11356-020-12034-x nathaniel, s. p., nwulu, n., bekun, f. (2020a). natural resource, globalization, urbanization, human capital, and environmental degradation in latin american and caribbean countries. environmental science and pollution research, 1-15. nathaniel, s. p., omojolaibi, j. a., ezeh, c. j. (2020b). does stock market-based financial development promotes economic growth in emerging markets?: new evidence from nigeria. serbian journal of management, 15(1), 45-54. https://doi.org/10.5937/sjm15-17704 nathaniel, s. p., yalçiner, k., bekun, f. (2020c). assessing the environmental sustainability corridor: linking natural resources, renewable energy, human capital, and ecological footprint in brics. resources policy, 1-13. https://doi.org/10.1016/j.resourpol.2020.101924 nathaniel, s., khan, s. a. r. (2020). the nexus between urbanization, renewable energy, trade, and ecological footprint in asean countries. journal of cleaner production, 122709. https://doi.org/10.1016/j.jclepro.2020.122709 nathaniel, s., aguegboh, e., iheonu, c., sharma, g., shah, m. (2020d). energy consumption, fdi, and urbanization linkage in coastal mediterranean countries: re-assessing the pollution haven hypothesis. environmental science and pollution research, 27(28), 35474-35487. https://doi.org/10.1007/s11356-020-09521-6 nathaniel, s. p., omojolaibi, j. a., ezeh, c. j. (2020e). does stock market-based financial development promotes economic growth in emerging markets? new evidence from nigeria. serbian journal of management, 15(1), 45-54. https://doi.org/10.5937/sjm15-17704 https://doi.org/10.1016/j.resourpol.2020.101585 https://doi.org/10.1016/j.spc.2020.11.008 https://doi.org/10.1111/j.1467-9701.2006.00808.x https://doi.org/10.1177/0972150920916653 https://doi.org/10.1177/0972150920916653 https://doi.org/10.1002/pa.2213 https://doi.org/10.1007/s11356-020-09497-3 https://doi.org/10.1002/ijfe.2073 https://doi.org/10.1007/s11356-020-09593-4 https://doi.org/10.1016/j.jenvman.2019.03.112 https://doi.org/10.1016/j.jenvman.2019.03.112 https://doi.org/10.1007/s11356-020-12034-x https://doi.org/10.5937/sjm15-17704 https://doi.org/10.1016/j.resourpol.2020.101924 https://doi.org/10.1016/j.jclepro.2020.122709 https://doi.org/10.1007/s11356-020-09521-6 https://doi.org/10.5937/sjm15-17704 solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 44 nawaz, k., lahiani, a., roubaud, d. (2019). natural resources as blessings and finance-growth nexus: a bootstrap ardl approach in an emerging economy. resources policy, 60, 277-287. https://doi.org/10.1016/j.resourpol.2019.01.007 omojolaibi, j., nathaniel, s., p. (2020). assessing the potency of environmental regulation in maintaining environmental sustainability in mena countries: an advanced panel data estimation. journal of public affairs, e2526. https://doi.org/10.1002/pa.2526 pesaran, m. h. (2004). general diagnostic tests for cross-section dependence in panels. fac. econ. https://doi.org/doi.org/10.17863/cam.5113 pesaran, m. h. (2007). a simple panel unit root test in the presence of cross-section dependence. j. appl. econom. 22, 265–312. https://doi.org/10.1002/jae.951 rajan, r. g., zingales, l. (2003). the great reversals: the politics of financial development in the twentieth century. journal of financial economics, 69(1), 5-50. https://doi.org/10.1016/s0304405x(03)00125-9 redmond, t., nasir, m. a. (2020). role of natural resource abundance, international trade and financial development in the economic development of selected countries. resources policy, 66, 101591. https://doi.org/10.1016/j.resourpol.2020.101591 rickman, d. s., wang, h., winters, j. v. (2017). is shale development drilling holes in the human capital pipeline? energy economics, 62, 283-290. https://doi.org/10.1016/j.eneco .2016.12.013 saint akadırı, s., alola, a. a., usman, o. (2021). energy mix outlook and the ekc hypothesis in brics countries: a perspective of economic freedom vs. economic growth. environmental science and pollution research, 1-5. https://doi.org/10.1007/s11356-020-11964-w shahbaz, m., naeem, m., ahad, m., tahir, i. (2018a). is natural resource abundance a stimulus for financial development in the usa? resources policy, 55, 223-232. https://doi.org/10.1016/j.resourpol.2017.12.006 shahbaz, m., nasir, m. a., roubaud, d. (2018b). environmental degradation in france: the effects of fdi, financial development, and energy innovations. energy economics, 74, 843-857. https://doi.org/10.1016/j.eneco.2018.07.020 sibel, b. e., kadir, y. e., ercan, d. (2015). local financial development and capital accumulations: evidence from turkey. panoeconomicus, 62(3), 339-360. https://doi.org/10.2298/ pan1503339e sun, y., ak, a., serener, b., xiong, d. (2020). natural resource abundance and financial development: a case study of emerging seven (e− 7) economies. resources policy, 67, 101660. https://doi.org/10.1016/j.resourpol.2020.101660 tiba, s., frikha, m. (2019). the controversy of the resource curse and the environment in the sdgs background: the african context. resources policy, 62, 437-452. https://doi.org/ 10.1016/j.resourpol.2019.04.010 ulucak, r., khan, s. u. d. (2020). does information and communication technology affect co2 mitigation under the pathway of sustainable development during the mode of globalization? sustainable development. https://doi.org/10.1002/sd.2041 umar, m., ji, x., kirikkaleli, d., shahbaz, m., zhou, x. (2020). environmental cost of natural resources utilization and economic growth: can china shift some burden through globalization for sustainable development? sustainable development, 28(6), 1678-1688. https://doi.org/10.1002/sd.2116 westerlund, j. (2007). error correction based panel cointegration tests. oxford bulletin of economics and statistics, 69, 709-748. https://doi.org/10.1111/j.1468-0084.2007.00477.x world bank (2017). world bank lending: fiscal 2017. the world bank group. retrieved from the world bank. website: http://pubdocs.worldbank.org/. world development indicator (wdi) (2019). world bank development indicators database (online) available at https://data.worldbank.org/ accessed date 24.10.2019. xu, x., xu, x., chen, q., che, y. (2016). the research on generalized regional “resource curse” in china's new normal stage. resources policy, 49, 12-19. https://doi.org/10.1016 /j.resourpol.2016.04.002 yu, a., jia, z., zhang, w., deng, k., herrera, f. (2020). a dynamic credit index system for tsmes in china using the delphi and analytic hierarchy process (ahp) methods. sustainability, 12(5), 1715. https://doi.org/10.3390/su12051715 https://doi.org/10.1016/j.resourpol.2019.01.007 https://doi.org/10.1002/pa.2526 https://doi.org/doi.org/10.17863/cam.5113 https://doi.org/10.1002/jae.951 https://doi.org/10.1016/s0304-405x(03)00125-9 https://doi.org/10.1016/s0304-405x(03)00125-9 https://doi.org/10.1016/j.resourpol.2020.101591 https://doi.org/10.1016/j.eneco.2016.12.013 https://doi.org/10.1016/j.eneco.2016.12.013 https://doi.org/10.1007/s11356-020-11964-w https://doi.org/10.1016/j.resourpol.2017.12.006 https://doi.org/10.1016/j.eneco.2018.07.020 https://doi.org/10.2298/%0bpan1503339e https://doi.org/10.2298/%0bpan1503339e https://doi.org/10.1016/j.resourpol.2020.101660 https://doi.org/10.1016/j.resourpol.2019.04.010 https://doi.org/10.1016/j.resourpol.2019.04.010 https://doi.org/10.1002/sd.2041 https://doi.org/10.1002/sd.2116 https://doi.org/10.1111/j.1468-0084.2007.00477.x http://pubdocs.worldbank.org/ https://doi.org/10.1016/j.resourpol.2016.04.002 https://doi.org/10.1016/j.resourpol.2016.04.002 https://doi.org/10.3390/su12051715 solomon prince nathaniel / european journal of government and economics 10(1), june 2021, 30-45 45 zaidi, s. a. h., wei, z., gedikli, a., zafar, m. w., hou, f., iftikhar, y. (2019). the impact of globalization, natural resources abundance, and human capital on financial development: evidence from thirty-one oecd countries. resources policy, 64, 101476. https://doi.org/ 10.1016/j.resourpol.2019.101476 https://doi.org/10.1016/j.resourpol.2019.101476 https://doi.org/10.1016/j.resourpol.2019.101476 ©the authors 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 206-224 https://doi.org/10.17979/ejge.2024.13.2.10173 submitted: dec 14, 2023 accepted: may 14, 2024 published: dec 3, 2024 article unlocking regional value chains: fintech’s role in the western balkans' integration into global markets gentjan shaqiri,1 jolta kacani1, *, gazmend qorraj2 1 university of tirana, albania 2 jean monnet chair, university of prishtina, kosovo *correspondence: joltakacani@gmail.com abstract. digitalization and automation are restructuring the geographical landscape of operations and labor, significantly altering operations in global value chains (gvcs). digital technologies are expected to boost resilience, enhance transparency, and foster sustainability in gvcs. in this context, this paper looks at how the application of fintech solutions fosters the creation of regional value chains (rvcs), increasing attractiveness and generating new opportunities for enterprises in the western balkan region to participate in gvcs to fill the existing gap for granular analysis at the industry level, the paper examines if enterprises in the western balkans that apply of fintech solutions in financial management have better opportunities to create rvcs than the ones that resist fintech implementation. using a benchmarking instrument on fintech solutions implemented on 370 enterprises, this paper tries to identify industries with the highest potential to create rvcs and higher prospects for integration in gvcs. our findings indicate that fintech solutions and applications can be one of the main driving forces in transforming and upgrading rvcs in western balkan countries. keywords: fintech; global value chains; financial ratio analysis; benchmarking; western balkans jel classification: c39 ; c51 ; h26 1. introduction global value chains (gvss) have become increasingly fragmented and dispersed, with a growing number of longer and wider-ranging networks of smalland medium-sized (sme) upstream and downstream enterprises across major international markets. according to kano et al. (2020) and kacani and shaqiri (2023), gvcs comprise a series of stages involved in the production of a product or service that is ultimately sold to consumers. each stage contributes to the creation of value, and at least two stages are produced in distinct countries. gvcs create global trade routes that connect local producers from emerging economies to global marketplaces. they provide a link between suppliers of the raw materials, the producer, and the final consumer. in the dynamic environment of international markets, technological evolution is one of the main factors influencing how gvcs evolve (banga, 2020). new technologies are making it easier to reconfigure gvcs. within the technological evolution, the european union (eu) has recognized digitalization as a significant catalyst for https://creativecommons.org/licenses/by-nc/4.0/ unlocking regional value chains economic and social transformation, albeit the process has been characterized as uneven. digitalization stimulates the upgrading of gvcs as it intensifies research and development expenditure, maximizes resource allocation, encourages technological innovation, empowers digital transformation, augments efficiency, and promotes goodwill and reputation (sun et al., 2023). in gvcs, digitalization reduces multilateral resistance to trade, expands export market access, increases the export market, and helps exports grow marginally (elia et al., 2021). over the past decade, digitalization has introduced several innovations in gvcs, including fintech solutions. in a nutshell, fintech is a digital record of financial transactions about trade that can be traced in real-time and distributed among participants who possess access rights. fintech permits actors in gvcs to administer their databases, bringing together all the pertinent data into one digital document (gleibner et al., 2019). fintech solutions can be used to preserve, track, supervise, and exchange both physical and digital assets in a way that is both effective and transparent. by implementing this approach, the technology can serve as a trust mechanism that can be easily integrated and utilized, thereby enabling other emerging technologies to attain greater scale. fintech solutions can also be used to improve the traceability of goods throughout the production process (arrfelt et al., 2018). numerous research studies have revealed that the application of fintech solutions exhibits numerous advantages, including the enhancement of openness, confidence, and safety, while simultaneously lowering expenses, resulting in enhanced effectiveness. (javorcik, 2020). to continue, global trade can be encouraged due to the availability of smart contracts through fintech solutions. smart contracts typically refer to software applications that store rules and actions between parties. fintech solutions provide an opportunity for enterprises operating in gvcs to modify their size and configuration, thereby facilitating firms to reduce transaction costs (kano et al., 2020). with big data, cloud technology, and powerful algorithms, gvc relationships and work organization have changed. (le et al., 2021). moreover, the application of fintech solutions is expected to boost resilience, enhance transparency, and foster sustainability in regional markets and regional value chains. in line with the global dynamics of gvcs, the new european union (eu) regional innovation agenda (euria) for the western balkan countries 1aims to create regional value chains (rvcs) within the western balkan identifying industries that can easily participate in gvcs attracting new investments. according to pasquali et al. (2020), the concept of regional value chains (rvcs) refers to production and consumption networks that are not globally coordinated, but rather within a single global geographic or administrative zone. rvcs help local economies participating in rvcs to develop political, social, and strategic relationships. this helps the region become more prosperous and grow sustainably. participation in rvcs can create income and work opportunities, thus reducing poverty. it can also reduce uneven development and intra-regional inequalities (kacani & shaqiri, 2023). active participation in gvcs brings new perspectives for developing countries and in this case for western balkan ones, as they are exposed to new technologies, innovations, and automation (kacani & shaqiri, 2023). given the new landscape in gvcs, this paper examines if enterprises in the western balkans that apply fintech solutions in financial management have better opportunities to create rvcs than the ones that resist fintech implementation. the study highlights that enterprises 1 western balkan countries include albania, bosnia herzegovina, kosovo, montenegro, north macedonia, serbia. 208 gentjan shaquiri et al. applying fintech solutions in their financial operations have a higher potential to create regional value chains. these enterprises are considered by european ones as better candidates for additional investment and better integration in gvcs (antonucci et al., 2021). this research contribution starts with a theoretical background followed by a thorough explanation of the benchmarking methodology. the paper continues by presenting the main results and industry trends along with conclusions and policy recommendations both at an industry and regional level. 2. theoretical background the rise of globalization has made gvcs even more intricate, involving numerous parties requiring a lot of coordination. this makes it more expensive to run these global networks and move goods and services between countries (avom et al., 2021). the primary advantage of fintech solutions is the ability to facilitate the introduction of novel business models. in the digital age, enterprises need to make decisions in real-time to stay competitive, by introducing new products more quickly, cheaply, and with better quality (juliet et al., 2022). implementation of fintech solutions has the potential to disrupt global trade by facilitating the exchange of real-time data and reducing the expenses associated with delayed information, administrative trade frictions, and logistical obstacles like order delays (li et al, 2019). in the literature, the application of fintech solutions stimulates regionalization that is expected to reduce the length of gvcs, concentrating their geographical distribution and promoting exports of high-value goods from regional markets into gvcs (pananond et al, 2020). fintech solutions affect gvcs through cost reduction and improvement in communication efficiency. in particular, the implementation of paperless trade practices such as electronic invoices, electronic signatures, pre-clearance of shipments, single windowing, and the utilization of blockchain ledger technology in the supply chain for tax or border compliance processes, has reduced the days and costs needed for international transactions (meltzer, 2019). additionally, fintech solutions eliminate information asymmetry, enhancing the value added of exports and promoting the upgrading of rvcs into gvcs (feroz et al., 2021). the significance of fintech solutions is valid for both small and large multinational enterprises, as they offer significant advantages in the areas of sales, payment, finance, and logistics. digitization helps large enterprises transport their goods faster across borders and helps small enterprises overcome problems when exporting (thun et al., 2021). fintech solutions empower smes to transform their business models by interacting with multi-layered digital platforms and by enhancing existing alliances (helmerich et al., 2021). by working together, a shared digital platform among smes can also be seen as a global virtual value chain characterized by technology and intangible production inputs. smes can use fintech platforms to manage their cash flows in a revolutionary way (kan et al., 2022). artificial intelligence and transactional data can be used by smes to understand transaction histories and make accurate forecasts of their future cash flows. smes can make more informed decisions by using this data-driven approach to optimize their working capital and mitigate their risk of insolvency. for example, fintech platforms offer new pathways for sme suppliers to access funds (lee & falahat, 2019). by digitizing assets and utilizing unlocking regional value chains technology, small-scale suppliers can improve liquidity and reduce their financing costs (liu et al., 2021). by enhancing the resilience of smes, these platforms can aid in enhancing the resilience of the entire value chain. the digital payment landscape has changed thanks to fintech companies like paypal, payneer, apple pay, and google pay (qi & ren, 2021). nowadays, smes can digitalize their assets through the utilization of fintech platforms. in addition, the emergence of fintech platforms provides substantial incentives for the procurement and distribution of goods produced in gvcs. the availability of cloud-based information, data analysis software applications, and related information "dashboards" enhances the visibility of the value chain to enable real-time tracking and rapid allocation of resources to meet customer demands. at the same time, these aspects also help deliver service and provide customer care (niemand et al., 2021). to continue, the western balkan region has recently witnessed a surge in the utilization of fintech services. in the region, the online banking system was used by 8% of account holders com, compared to 58% in the eu. e-commerce is still not widely used in the region, but it is more common than in other countries that recently joined the eu (gruber, 2019). this trend is also identified in demirgüç-kunt et al. (2022), with specific indicators presented in table 1. table 1. data on the western balkans country used a mobile phone or the internet to check account balance (% age 15+) use a mobile phone or the internet to make payments, buy things, or send or receive money using a financial institution account (% age 15+) store money using a financial institution or a mobile money account (% age 15+) has access to the internet (% age 15+) made or received a digital payment (% age 15+) used a mobile phone or the internet to buy something online (% age 15+) used a mobile phone or the internet to pay bills (% age 15+) albania 13% 8% 25% 78% 35% 17% 5% bosnia and herzegovina 35% 18% 48% 92% 67% 31% 13% kosovo 14% 10% 31% 93% 48% 29% 10% montenegro 39% 23% n.a n.a 60% 13% 7% north macedonia 37% 30% 43% 88% 74% 31% 28% serbia 41% 28% 48% 84% 87% 33% 27% source: demirgüç-kunt et al. (2022) the western balkan countries are well-known for their troubled past and unstable politics. however, they have been making significant progress in embracing information and communication technologies (ict) for the past decade (nguyen et al., 2020). their growing ict sector is trying to get attention from investors and global tech enterprises. the western balkan countries are well210 gentjan shaquiri et al. positioned for a promising future because of a surge in the region's ict industry, where well-paying and high-quality employment opportunities have been scarce in the past. it might be what the region needs to catch up with other eu member states in the area (santo-jaen et al., 2023). there has been a steady increase in the importance of rvcs over the last few years. eliminating trade obstacles and boosting cross-border commerce are expected to streamline business collaboration between enterprises in the region. this, in turn, will allow enterprises to pool their resources, skills, and abilities, and to create more innovative and competitive goods and services. the more enterprises in the region work together, the more foreign investments there will be in the region (mccormick and somaya, 2020). this will make the regional market in the western balkan more stable and easier to integrate into gvcs, creating additional prospects for large enterprises as well as smes and start-ups. to continue, the literature argues that developing countries including the western balkan ones have made their integration into gvcs faster by creating rvcsthat align with the common goals of gvcs (radosavljević, 2023). rvcs support developing countries to build better political, social, and strategic relationships with each other. this helps the region grow more economically and sustainably (kacani et al., 2022). in particular, the significance of rvcs was amplified during the covid-19 pandemic. there are many reasons why rvcs have started to appear. first, geographic location helps regions cluster. the clustering of adjacent regions promotes the interaction among nations, resulting in expedited delivery times and enhanced adoption and implementation of innovations in several industries (wielgos et al., 2021). the lengthy transition into the free-market economy and the slow progress in key structural reforms have hampered the region's competitiveness despite the low salaries paid by its citizens compared to those in european nations (pournader et al, 2020). to enhance its competitiveness, the economic transformation strategy is currently focusing on altering the business model and adoption of ict, including fintech solutions that will allow enterprises in the western balkan region to absorb new developments in several relevant sectors that are present in international markets like the eu or the united states (us) (goldfarb & tucker, 2019). a few initiatives are currently in place in the region of north macedonia and montenegro that involve multiple stakeholders. according to the international communication union2, since june 2019, the innovation office in north macedonia has been fully operational. increasing internal capacity to handle fintech issues, spreading awareness about the benefits and dangers of the fintech sector, and assisting market players with regulatory clarity are some of the objectives of the office. the goal is to issue legally non-binding opinions regarding inquiries from fresh market players and innovative products and services. the office will cover several areas, including digital identity, open banking, tokenization of financial instruments, crowdfunding, crypto assets, and possibilities for developing a central bank digital currency. its primary objective is to develop a fintech strategy and action plan to help develop fintech activities. the innovation office will establish a detailed overview of the fintech sector and investigate the various business strategies in greater detail. the office will examine the existing legal and regulatory framework to identify where the impediments originate. 2 https://www.itu.int/dms_pub/itu-d/opb/inno/d-inno-profile.northmacedonia-2023-pdf-e.pdf https://www.itu.int/dms_pub/itu-d/opb/inno/d-inno-profile.northmacedonia-2023-pdf-e.pdf unlocking regional value chains in addition, the first and, to date, only regulatory sandbox 3in the western balkans was established at the securities commission of montenegro in september 2019. the securities commission has established comprehensive regulations that govern the scheme's objectives, criteria, and application procedure. the sandbox initiative aims to provide a controlled regulatory environment for potential new market entrants and established fintech firms to evaluate financial innovation, particularly in instances where the applicable regulation is inappropriate or unclear. the european commission will monitor and evaluate the risks associated with new products or services, as well as their compatibility with existing rules, while they're in the sandbox. to be deemed acceptable, enterprises must present innovative financial solutions that provide distinct advantages to consumers, such as more practical, safer, or cheaper financial services. in return, individuals will derive numerous advantages from participating in the scheme, including the receipt of guidance from relevant authorities, a more lenient approach towards penalties, and a tolerant interpretation of existing regulations based on the principle of proportionality (oecd, 2023). 3. methodology and case study overview financial ratios are widely used by researchers to evaluate the operational performance of enterprises. in the literature, financial ratios are regarded as reliable indicators for evaluating sustainability in both enterprises and industries. in addition, financial ratios are used to assess not only the current financial well-being of enterprises but also their upcoming financial prospects, such as expansion into new markets or intensifying their integration into gvcs. they are estimated to investigate how the previous financial performance of an enterprise compares to enterprises operating in the same industry. laitinen (2018) argues that financial ratios serve two primary functions. firstly, they are used to forecast financial outcomes, such as returns, earnings, and late or non-payments. secondly, financial ratios are considered credible indicators to compare across different industries, a process commonly referred to as benchmarking analysis (arrfelt et al., 2018). furthermore, financial ratios are solid indicators of the financial sustainability of enterprises in the existing literature. to answer our main research questions on whether enterprises implementing fintech solutions have a higher possibility to create rvcs than the ones that do not consider fintech in their financial operations, this paper looks at the financial indicators related to the use of fintech solutions in 370 enterprises that operate in western balkan countries. the financial data was taken from a large survey addressed to the management of these enterprises. relevant data was retrieved for the five years spanning from 2018 to 2022 inclusive. the enterprise selection criteria introduced in table 2 targeted enterprises that use fintech solutions and operate in more than one market. 3 regulatory sandboxes are regulatory instruments that enable enterprises to evaluate and experiment with novel and innovative products, services, or enterprises under the supervision of a regulatory body for a limited duration. the purpose of regulatory sandboxes is to foster business learning, i.e. the creation and validation of innovations in a real-world setting, and to support regulatory learning, i.e. the formulation of experimental legal frameworks to guide and support businesses in their innovation endeavors. https://www.europarl.europa.eu /regdata/etudes/brie/2022/733544/eprs_bri(2022)733544_en.pdf https://www.europarl.europa.eu/regdata/etudes/brie/2022/733544/eprs_bri(2022)733544_en.pdf https://www.europarl.europa.eu/regdata/etudes/brie/2022/733544/eprs_bri(2022)733544_en.pdf 212 gentjan shaquiri et al. the survey design: the survey was designed to obtain relevant information on the financial performance of enterprises. the questionnaire has three main parts: the first part collects general information about the enterprise, the second one aims to retrieve financial data on the return of fintech investments, and the third part targets fintech operations. the survey was conducted online following a prior introduction of the research objectives and a brief explanation of the methodology. after completing the survey, the authors conducted a focus group to identify the challenges, opportunities, necessary policy implications, and future trends in the use of fintech. the qualitative method enabled us to gather more elaborative responses so to generate meaningful insights into this study. population and sample: the population is based on a purposeful sample targeting enterprises operating in the western balkan region that met the pre-selection criteria introduced in table 2. in addition to the selection criteria, the enterprises were selected according to the gics 2023 industry classification used in the study. to continue, participants in the focus group were selected randomly and included participants from all western balkan countries and industries included in the study. a set of closed and open questions were prepared for participants. implementation procedure: data collection was a transparent and simple process as respondents are managers of enterprises and, therefore, aware of the importance of the research process and the benchmarking instrument. ethical issues were considered, such as respondents’ anonymity and confidentiality of the enterprise information. table 2: selection criteria applied for enterprises operating in western balkan countries. no. selection criteria occurrence in the sample size 1. turnover of over 5 million/eur within the last three years. • 13% of enterprises in the sample have a turnover of over 10 million /eur. • 29% of enterprises have a turnover between 7-9 million/eur. • 58% of enterprises in the sample have a turnover between 5 to 7 million/eur. 2. over 100+ employees • 11% of enterprises have over 250 employees. • 14 % of enterprises have between 200-250 employees. • 32% of enterprises have between 150-200 employees. • 43% of enterprises have between 100-150 employees. 3. provision of services in at least two international markets like the european union, the united states, etc. • 22% of enterprises provide services to the regional, european, and united states markets. • 67% of enterprises provide services to the regional and the european market. • 11% of enterprises provide services only to the regional market. 4. the number of �intech solutions implemented within the last 5 years. • 12% of enterprises have over 5 fintech investments. • 36% of enterprises have made between 3 to 5 fintech investments. • 52% of enterprises have made up to 3 fintech investments. source: prepared by authors unlocking regional value chains table 3: enterprise classification according to gics, 2023 no. industry subindustry number of enterprises subindustry level total number of enterprises 1. consumer discretionary distribution & retail automotive retail 21 248 apparel retail 64 computer & electronics retail 37 home improvement retail 18 home furnishing retail 26 food distributors 49 food retail 33 2. customer services hotels, resorts & cruise lines 59 122 leisure facilities 13 restaurants 35 education services 17 source: prepared by authors enterprises in the sample are categorized into industries and subindustries to obtain a more granular analysis for a specific industry. this way, it will be able to identify the industry trends based on the enterprise application of fintech solutions. their activity is identified based on responses from the management obtained through the large-scale survey. the final categorization of the operational activity is presented in table 3 based on the global industry classification standard (gics) 4 developed by msci and standard and poor in two main categories: consumer discretionary distribution and retail and customer services. initially, for every enterprise, each financial ratio is calculated for five years. the formulae applied are presented in table 4. enterprises in the sample are categorized into industries and subindustries to obtain a more granular analysis for a specific industry. this way, it will be able to identify the industry trends based on the enterprise application of fintech solutions. their activity is identified based on responses from the management obtained through the large-scale survey. the final categorization of the operational activity is presented in table 3 based on the global industry classification standard (gics) 5 developed by msci and standard and poor in two main categories: consumer discretionary distribution and retail and customer services. initially, for every enterprise, each financial ratio is calculated for five years. the formulae applied are presented in table 4. 4 please refer to the official website for the latest classification of enterprises introduced in july 2023 https://www.msci.com/our-solutions/indexes/gics 5 please refer to the official website for the latest classification of enterprises introduced in july 2023 https://www.msci.com/our-solutions/indexes/gics https://www.msci.com/our-solutions/indexes/gics https://www.msci.com/our-solutions/indexes/gics 214 gentjan shaquiri et al. table 4: financial indicators used to evaluate the degree of fintech applications in enterprises operating in the western balkan region. no. financial indicator formula explanation 1. asset turnover from fintech assets • asset turnover = total sales generated from fintech assets / average total assets asset turnover ratio measures the value of sales or revenues generated relative to the value of its assets. 2. investment turnover from fintech investments • revenues generated from fintech solutions/ (stockholders' equity + debt) the investment turnover indicates how effectively an enterprise uses its resources to generate revenues. 3. account receivables turnover (days) • account receivables turnover= net annual credit sales generated from fintech solutions/ average accounts receivables • account receivables turnover/365 days the accounts receivables turnover ratio measures the number of times an enterprise collects its average accounts receivable. it measures the efficiency of an enterprise in collecting outstanding balances from clients and managing its line of credit process. 4. account payables turnover (days) • account payable turnover= supplier purchases made through fintech solutions/average accounts payable • account receivables turnover/365 days the accounts payable turnover ratio is used to quantify the rate at which an enterprise pays off its suppliers. it indicates how many times an enterprise pays off its accounts payable during a given period. 5. customer lifetime value (5 last years) • total net profit generated from customers using enterprise fintech applications / total net profit generated from all enterprise customers (in %). • the difference is calculated between two periods to indicate the growth rate (in %). this indicator helps to identify the percentage of profit generated from customers who use fintech applications within the last 5 years. at the same time, it is an indicator that is used to estimate the growth rate of the indicator within the last five years. 6. online payments made by customers • (number of online payments made from customers through fintech applications / total number of payments made by customers (in %). • the difference is calculated between two periods to indicate the growth rate (in %). indicates the percentage of payments customers made through fintech applications, within the last five years. also, the growth in the use of fintech applications. 7. online payments made to suppliers • number of online payments made to suppliers through fintech applications / total number of payments made to suppliers (in %). • the difference is calculated between two periods to indicate the growth rate (in %). indicates the percentage of payments made to suppliers through fintech applications, within the last five years. also, the growth in the use of fintech applications 8. online inventory management • time used to manage inventory through fintech applications – time used to manage time through non-fintech applications. indicates time reduction achieved through the implementation of fintech applications in inventory management. the reduction in time is converted into quicker order delivery to end customers. 9. electronic invoicing • number of electronic invoices issued through fintech applications/total number of invoices (in %). • the difference is calculated between two periods to indicate the growth rate (in %). indicates the percentage of invoices issued both to customers and suppliers through fintech applications in the daily operations of the enterprise within the last five years. also, the growth in the use of fintech applications. source: prepared by authors unlocking regional value chains after obtaining the results of financial ratios for each enterprise, they are classified according to the range introduced in table 5, indicating whether there is (i) a low potential for the creation of rvcs, (ii) a middle potential for the creation of rvcs, and (iii) a high potential for the creation of rvcs. these indicators serve as the basis for estimating industry values and applying a benchmark analysis, thus identifying industry trends for the creation of rvcs based on enterprise fintech implementation in financial operations. table 5: benchmarking indicators for fintech applications fintech applications in enterprises operating in the western balkan region. component financial ratio low potential for regional value chain middle potential for regional value chain high potential for regional value chain return on fintech investment asset turnover from fintech assets <1 1–1.5 >1.5 investment turnover from fintech investments <0.5 0.5–1 >1 account receivables turnover (days) >10 days 5-10 days <5 days account payables turnover (days) >10 days 5-10 days <5 days customer lifetime value (last 5 years) <2,000 eur 2,000–4,000 eur >4,000 eur fintech operations online payments made by customers <30% 30%–50% >50% online payments made to suppliers <40% 40%-60% >60% online inventory management <50% 50%-80% >80% electronic invoicing <50% 50%-80% >80% source: prepared by authors after the calculation of financial ratios, benchmarking is the primary instrument used to evaluate the potential for the creation of rvgs. the notion of benchmarking pertains to a systematic and continuous evaluation of the performance of an enterprise about those of established market leaders, both locally, regionally, and globally. benchmarking helps enterprises retrieve valuable information to improve their performance. it has been a successful instrument for integration in gvcs because benchmarking can be simplified into learning from the best. the industry benchmarks are calculated using several methods. the two main methods are “the average of ratios” and “the ratio of averages”. the data collected led us to adopt the ratio of averages as a more accurate representation of the relatively small number of enterprises included in the sample and to determine the percentile in which enterprises in the sample are placed, according to equation 1 in which y r, t represents the industry and subindustry, while a and b represent ratio components. this approach ensures that outliers do not significantly affect the result. the results are presented in subindustry segments to mitigate the bias toward extensive categorization of enterprises at the industry level. similar methodologies have been applied in studies in the literature (kacani et al., 2022; kasasbeh (2021). the evaluation indicators of the 216 gentjan shaquiri et al. benchmarking instrument are based on fintech solutions used in the financial management of enterprises participating in the survey. indicators include, among others, a reduction in the lead time to perform a financial transaction, the number of automated financial management processes (payments, customer invoicing, payrolls, etc.), customer lifetime value generated through fintech solutions, return on fintech investments, return on fintech assets, etc. (𝑡𝑡) = ∑ 𝐴𝐴𝑖𝑖(𝑡𝑡)𝑡𝑡 𝑖𝑖 ∑ 𝐵𝐵𝑖𝑖(𝑡𝑡)𝑡𝑡 𝑖𝑖 = 𝐸𝐸[𝐴𝐴𝑖𝑖(𝑡𝑡)] 𝐸𝐸[𝐵𝐵𝑖𝑖(𝑡𝑡)] [1] the ratio of averages formula was then implemented to measure benchmarking indicators for the subindustry/industry level and judge whether each enterprise included in the sample size performs below or above the estimated industry level. this comparison allows for a better judgment of where each enterprise is positioned. for example, as introduced in table 6, enterprise a has most financial ratios falling into the middle potential for the creation of regional value chains according to the criteria introduced in table 4. as such, the final placement for enterprise a has a middle potential for the creation of regional value chains. a similar analysis is applied to enterprises b and c. in case an enterprise has the same number of financial ratios for all categories presented in table 5, then the final decision is based on the financial ratios with more dominant values (kiselakova et al., 2018). table 6: example of the category an enterprise falls regarding the creation of regional value using fintech applications. scenario low potential for regional value (no. of financial ratios). middle potential for regional value chain (no. of financial ratios). high potential for regional value chain (no. of financial ratios). final individual categorization of enterprises enterprise a 3 5 1 middle enterprise b 6 2 1 low enterprise c 2 3 5 high enterprise d 3 3 3 ratios with the highest value will prevail. source: prepared by authors 4. main results our findings indicate that most enterprises demonstrate a medium to low potential in creating regional value chains resulting from fintech applications. with of enterprises in the sample have generally demonstrated a medium to low outsourcing potential. the most affected subindustries are automotive retail (52%) and home improvement retail (61%). these subindustries demonstrate lengthy account receivables over 20 days and have less than 30% of customer payments made through fintech technologies. the lengthy payment cycle results from the type of goods (automobiles, home furniture, electric appliances, etc.) these industries sell to customers goods, which are mostly considered an investment rather than an expense. as such, enterprises in these subindustries prefer to receive bank transfers rather than online payments. however, for the provision of services like unlocking regional value chains repairs, both customers and enterprises prefer online payments. differently, in the european value chain, fintech applications are spreading rapidly in the automotive industry, such as the new partnership of mercedez-benz financial services with autogravity applications. this partnership will allow potential buyers to select their car by identifying all nearby dealerships. users can apply for online financing by having access to mercedez-benz lending purchasing options. in addition, in the home retail industry, there is an increase in the buy now, pay later (bnpl) services that are reshaping retail. these platforms allow consumers to make purchases and spread the cost over installment payments, often with low or no interest rates. bnpl services are attractive to consumers who look for flexibility and affordability while making purchases and are reluctant to use traditional credit cards and debt (kacani et al., 2022). despite these services having gained much popularity in the european value chain, there are several limitations to implementing such novelties in the western balkan countries. legislative constraints such as limited lump sum amounts of e-payments without declaring the origin of funds amplify the administrative burden for customers to make large purchases such as automobiles or furniture. in addition, the weak supervision of financial institutions, the high risk of payment default, and the dubious source of funds in emerging economies like those in the western balkans, restrain head offices in the automotive and home retail from allowing the use of bnpl or online financing (yue, 2020). on the other hand, subindustries that demonstrate the highest potential for the creation of regional value chains are hotels, resorts, and cruise lines (42%), apparel retail (38%), and food retail (36%). these are also the industries with the highest level of e-commerce transactions in the region, which boomed, especially during the covid-19 pandemic. enterprises in the apparel and food retail re-evaluated their business operating model, acquiring new technological skill sets and making additional fintech investments, especially in blockchain technologies to meet the ever-increasing demand for online sales. as a growing number of individuals made online purchases, it presented novel prospects for emerging enterprises to expand their digital presence and broaden their reach in a broader e-commerce landscape, including operations and services in several countries in the western balkans. fintech innovations are crucial for e-payments and e-commerce, with a few successful examples in the region, such as the albanian gjirafa and serbian fishingbooker.com. nevertheless, during a discussion with the management of these enterprises, it was suggested that a multitude of obstacles hindering the advancement of fintech in e-commerce exist. first, e-commerce associations need a special license in bosnia and herzegovina, serbia, and montenegro. this prevents european firms with a license abroad from attempting to enter the country. moreover, there are a limited number of payment gateway providers, which results in elevated expenses, and transaction fees for payment processing are a significant expense. a further obstacle to progress is that online sellers may encounter difficulties if, for instance, their business falls under the category of deposittaking, as it does in albania. ultimately, a significant number of customers are required to make payments in cash upon their arrival due to a lack of utilization of electronic signatures, thereby hindering the provision of a favorable user experience (odorović et al., 2020). electronic transactions in these subindustries reached over 70%, spanning additional markets like the european one. enterprises come up with unique ways to communicate and find new ways to improve the customer experience. https://www.autogravity.com/ 218 gentjan shaquiri et al. for example, in the hospitality industry, fintech is an efficient way to achieve customer loyalty. using fintech will help enterprises operating in this industry to gain and improve customer loyalty. from a sales perspective, one way is by implementing convertible loyalty points as an online payment option, this is more likely to make customers return and repurchase enterprise services due to their economic benefits. in the apparel retail e-commerce payment solutions, digitized receipts are provided to return clothing items, saving tons of paper used for physical receipts. in supermarkets and other food stores, innovative point-of-sale solutions are driving fintech innovation (yue, 2022). retailers are responding to customers' demand for cashless shopping by partnering with new payment platforms, such as cloud-based solutions, as well as contactless payment providers. while in enterprise management, fintech supports enterprises using blockchain, where brands can digitize, track, and trace the entire lifecycle of a product item. with blockchain, enterprises can create an immutable record of all steps in the value chain, capture specific data points, such as sustainability certifications and claims, and provide open access to this data publicly (truby, 2022). to implement fintech solutions, enterprises had to overcome numerous difficulties, including securing adequate funding for the digital and physical infrastructure, providing continuous training of their employee, and keeping track and updates on the latest fintech developments. the major challenge was to educate customers on using fintech services, particularly age groups that are not well acquainted with using technological innovations, including fintech. multiple benefits arise by embracing fintech, including improved customer experience, faster delivery times, easier product tracking, and paperless financial management (nugraha et al., 2022). table 7. benchmarking results for enterprises using fintech applications and operating in the western balkan region. no. industry subindustry low potential for regional value chain (%) middle potential for regional value chain (%) high potential for regional value chain (%) 1. consumer discretionary distribution & retail automotive retail 52% 37% 11% apparel retail 29% 33% 38% computer & electronics retail 28% 58% 14% home improvement retail 58% 29% 13% home furnishing retail 54% 28% 16% food distributors 25% 43% 32% food retail 29% 35% 36% 2. customer services hotels, resorts & cruise lines 27% 31% 42% leisure facilities 37% 49% 14% restaurants 53% 38% 9% education services 31% 43% 26% source: prepared by authors unlocking regional value chains these enterprises succeeded in using fintech solutions to manage their supply chain entirely online, including online monitoring of inventory with order tracking applications, online payments, and electronic invoicing for suppliers and customers, reducing by at least three days the time of receivables and payables. these subindustries are also characterized by a high asset turnover from fintech applications with a ratio of over 3, a high investment turnover with values over 2, and a customer lifetime value exceeding 5,000 euros. most of the enterprises in these subindustries have already established product and service delivery by opening branches or by establishing partnerships with similar service and product providers in regional countries. to continue, enterprises with middle potential for establishing regional value chains through the implementation of fintech solutions are the majority in the sample size. enterprises within this category are providers of both goods and services. they have taken several steps in incorporating fintech solutions into their operations; however, this is most dominant in their interaction with suppliers rather than with customers. as such, these enterprises have an average of six days of accounts payable, with online payments to suppliers reaching almost 60%. in these enterprises, fintech solutions are made mostly as a necessity to interact with international suppliers rather than a decision made by the management to generate a forward-looking initiative for further integration into the regional market. their reluctance is observed mostly toward customers with minor transactions occurring through fintech applications. in the interviews with the focus group, several obstacles were identified in the application of fintech solutions. data security has emerged as a priority issue in the digital community, whether it's mobile banking or payment applications. traditional financial institutions like banks believe they're capable of safeguarding client data. but things are not as easy as we might think when it comes to digital security. credit card and bank account numbers, addresses, and responses to security inquiries are some of the personal and financial information that fintech applications possess. as such, data breaching is a major risk and highly damaging to users (ishamuddin et al., 2023). this makes security a top priority for fintech enterprises and users of fintech solutions. cybersecurity services like pen testing can help reduce the risk in the digital environment. for instance, enterprises are using intrusion detection systems to test the degree of data security. given the increased threats, it is imperative to contemplate a cybersecurity strategy to safeguard fintech operations within an enterprise. although these strategies might be costly and require experienced professionals, enterprises within the western balkan at early stages may opt to become users rather than providers of fintech applications that have a two-factor authorization, biometric authentication, data encryption, and real-time alerts and notifications. such fintech applications are widely spread across different segments of gvcs, facilitating both integration and upgrading of regional value chains into gvcs (truby et al., 2022). overall, enterprises operate in a regional environment with relatively tight ties with each other. this is manifested especially in electronic invoicing and online payment transfers made to suppliers and the presence of fintech applications. notwithstanding the category in which enterprises belong, all of them have made at least two significant investments toward fintech implementation, indicating an unexploited potential for strengthening regional value chains that will lead to further integration into global ones. our findings indicate that several enterprises can outperform the industry by exhibiting a much higher potential for the creation of sustainable regional value chains (adeleye et al., 2021). 220 gentjan shaquiri et al. 5. conclusions and policy recommendations our research shows that enterprises in the western balkan region have a moderate to weak potential for the establishment of regional value chains, given the slow progress made in the implementation of fintech solutions. enterprises and policymakers need to foster a business environment driven by a higher degree of fintech transactions to encourage a higher degree of regional interconnection, leading to a higher integration into gvcs (egwuonwu et al., 2022). based on the findings of this paper, the creation of regional value chains encouraged through the application of fintech solutions has long-term benefits for western balkan countries in improving operational financial transparency and attracting additional investment from foreign enterprises, especially the european ones. to take full advantage of the eu market, institutional actions need to initiate and implement pre-emptive policies to improve financial infrastructure in the private sector and strengthen regional investment competitiveness (abou-foul et al., 2021). these policies can be implemented in the framework of global shifts occurring in several gvcs, accelerating integration for enterprises that are more inclined to form regional value chains. in addition, the implementation of joint initiatives and policies to enable the creation of sustainable regional value chains in specific industries brings additional benefits from participation in the eu market (caputo et al., 2021). this way, enterprises in the western balkan region are exposed to new fintech technologies, innovations, and automation. this exposure improves competitiveness and facilitates compliance with international standards in the provision of goods and services, resulting in higher customer satisfaction and compliance with international standards for the provision of goods and services (dünhaupt and herr,2021). fintech innovations are increasingly based on blockchain technology and artificial intelligence. blockchain technology has the potential to disrupt traditional value chains by enabling more secure, transparent, and efficient transactions and value exchange. on the other hand, in the coming years, it is expected that ai-enabled systems will speed up cumbersome customs procedures, remove roadblocks, and make it easier for goods to cross borders. moreover, demand forecasting and visibility platforms powered by ai will be developed to help companies optimize inventory levels and provide real-time visibility in various gvcs. such optimization will enable enterprises to ai tools to facilitate environmental, social, and governance (esg) reporting in various industries (longbin et al., 2021). namely, ai-enabled analytics can fill disclosure gaps and facilitate sustainable investing. enterprises in the western balkans can follow a gradual implementation approach of such technologies, in their operations. in doing so, they can follow in the footsteps of main counterparties well integrated in gvcs, such as customers, suppliers, or enterprise headquarters, embracing best practices, thus avoiding challenges and failures previously experienced by them. in parallel, the development of continuous training programs for employees to use fintech innovations is mandatory to ensure a smooth transition from traditional operation systems into the new ones based on blockchain and ai. this way, enterprises can collaborate effectively with third-party experts and align expert recommendations with the objectives of implementing fintech in the operations of the enterprise. by doing so, enterprises in the western balkan region will find it easier to become integrated into gvcs (guo et al., 2022). https://payspacemagazine.com/articles/how-ai-can-facilitate-sustainable-investing/ https://payspacemagazine.com/articles/how-ai-can-facilitate-sustainable-investing/ unlocking regional value chains with the fast-paced developments in information and communication technologies, fintech solutions and applications can be one of the main driving forces behind the transformation and upgrading of regional value chains among western balkan countries. enterprises in the western balkan region need to transform their operating business model oriented toward digital technologies such as mobile internet, cloud computing, big data, internet of things, blockchain, and artificial intelligence, to promote the upgrading of several industries in the gvcs (kacani & shaqiri, 2023). by working together in sustainable regional value chains aiming at making the application of fintech solutions more common enterprises will be able to increase their competitiveness toward foreign investment. the increase in the use of fintech solutions results in more efficient enterprises in the western balkan, thus reducing transaction costs, improving service delivery, exporting to international markets, and upgrading their participation in gvcs. even though several western balkan countries work together on fintech projects through the european integration process, there is still room for a regional group to work together on fintechrelated regulations. small national markets, fragmented regulatory frameworks, and the inability to pass licenses to other countries are some of the most common remarks from industry participants. the coordination of her efforts would confer an additional advantage by potentially enabling fintech’s to offer their services beyond their domestic markets. moreover, the protection of consumers who utilize financial services is rarely adapted to digital communication channels. the recently adopted law on the protection of financial service consumers in distance contracts in serbia may serve as a valuable model for other countries in the region. several reforms can be jointly implemented, which include, among others: • establish a point of contact or innovation hub within each regulatory body to address fintech-related questions about licensing and clarification. • raise awareness of fintech opportunities, and facilitate discussion forums, conferences, or roundtables with relevant industry and consumer stakeholders. • establish a regulatory sandbox or innovation center in every country of the region within the financial oversight bodies to allow fintech enterprises and systems to flourish in a controlled testing environment. • join forces among regulators from different authorities in each country in the western balkans for capacity-building and training to establish a common understanding of fintech and promote cross-border interaction and engagement in this field. • collect information about how customers use bank and non-bank financial services and important signs of how the fintech market is changing. this will help industry and regulators talk and understand each other better. • revise the payment system regulations are needed to ensure and facilitate the availability of payment systems to non-bank financial service providers. • assess the highly challenging prospect of establishing a regional regulatory sandbox, to enable innovative regional fintech solutions to attain crucial economies of scale across borders. to fully exploit these prospects, regulators, providers, and governments in the western balkans must unite and implement the requisite reforms. adopting modernization and encouraging cooperation, 222 gentjan shaquiri et al. the economies of the western balkans can broaden the range of goods and services offered to end consumers, ultimately positioning the region as an active and spirited contender in the european financial landscape, thus creating a sustainable regional value chain that could be easily fully incorporated into gvcs. references abou-foul, m., ruiz-alba, j.l. & soares, a. (2021). the impact of digitalization and servitization on the financial performance of a firm: an empirical analysis. production planning and control, 32 (12), 975-989. https://doi.org/10.1080/09537287.2020.1780508 adeleye, b.n., adedoyin, f. & nathaniel, s. (2021). the criticality of ict trade nexus on economic and inclusive growth. information technology for development, 27 (2), 293-313. https://doi.org/10.1080/02681102.2020.1840323 antonucci, y.l., fortune, a. & kirchmer, m. (2021). an examination of associations between business process management capabilities and the benefits of digitalization: all capabilities are not equal. business processes management journal, 27 (1) 124-144. arrfelt, m., mannor, m., nahrgang, j.d. & christensen, a.l. (2018). all risk-taking is not the same: examining the competing effects of firm risk-taking with meta-analysis. review of managerial science, 12, 621-60. https://doi.org/10.1007/s11846-016-0225-9 avom, d., dadegnon, a.k. & igue, c.b. (2021). does digitalization promote net job creation? empirical evidence from waemu countries. telecommunications policy, 45 (8), 1-17. https://doi.org/10.1016/j.telpol.2021.102215 banga, k. (2020). digital technologies and product upgrading in global value chains: empirical evidence from indian manufacturing firms. the european journal of development research, 34, 77-102. https://doi.org/10.1057/s41287-020-00357-x caputo, a., pizzi, s., pellegrini, m. m. & dabic, m. (2021). digitalization and business models: where are we going? a science map of the field. journal of business research, 123, 489-501. https://doi.org/10.1016/j.jbusres.2020.09.053 demirgüç-kunt, a., klapper, l., singer, d., & ansar, s. (2022). the global findex database 2021: financial inclusion, digital payments, and resilience in the age of covid-19. washington, dc: world bank. dünhaupt, p. & herr, h. (2021). global value chains-a ladder for development? international review of applied economics, 35 (4), 456-474. https://doi.org/10.1080/02692171.2020.1815665 egwuonwu, a., mordi, c., egwuonwu, a. & uadiale, o. (2022). the influence of blockchains and internet of things on global value chain. strategic change, 31 (1), 45-55. https://doi.org/10.1002/jsc.2484 elia, s., giuffrida, m., mariani, m.m. & bresciani, s. (2021). resources and digital export: an rbv perspective on the role of digital technologies and capabilities in cross-border e-commerce. journal of business research, 132, 158-169. https://doi.org/10.1016/j.jbusres.2021.04.010 feroz, a. k., zo, h. & chiravuri, a. (2021). digital transformation and environmental sustainability. sustainability, 13 (3), 1530-1551. https://doi.org/10.3390/su13031530 gleibner, w., günther, t. & walkshäusl, c. (2022). financial sustainability: measurement and empirical evidence. journal of business economics, 92, 467-516. https://doi.org/10.1007/s11573-022-01081-0 goldfarb, a. & tucker, c. (2019). digital economics. journal of economic literature, 57, 3-43. https://doi.org/10.1257/jel.20171452 gruber, h. (2019). proposals for a digital industrial policy for europe. telecommunication policy, 43 (2).116-127. https://doi.org/10.1016/j.telpol.2018.06.003 guo.l, chen, j., li,s., li,y. & lu,j. (2022). a blockchain and iot-based lightweight framework for enabling information transparency in supply chain finance. digital communication and networks, 8(4), 576-587. https://doi.org/10.1016/j.dcan.2022.03.020 helmerich, n., raj-reichert, g. & zajak, s. (2021). exercising associational and networked power through the use of digital technology by workers in global value chains. competition and change, 25 (2), 42-166. https://doi.org/10.1177/1024529420903289 international telecommunication union. (2023). digital innovation profile: north macedonia. itu publications, geneva, switzerland. https://www.itu.int/dms_pub/itu-d/opb/inno/d-innoprofile.northmacedonia-2023-pdf-e.pdf https://doi.org/10.1080/09537287.2020.1780508 https://doi.org/10.1080/02681102.2020.1840323 https://doi.org/10.1007/s11846-016-0225-9 https://doi.org/10.1016/j.telpol.2021.102215 https://doi.org/10.1057/s41287-020-00357-x https://doi.org/10.1016/j.jbusres.2020.09.053 https://doi.org/10.1080/02692171.2020.1815665 https://doi.org/10.1002/jsc.2484 https://doi.org/10.1016/j.jbusres.2021.04.010 https://doi.org/10.3390/su13031530 https://doi.org/10.1007/s11573-022-01081-0 https://doi.org/10.1257/jel.20171452 https://doi.org/10.1016/j.telpol.2018.06.003 https://doi.org/10.1016/j.dcan.2022.03.020 https://doi.org/10.1177/1024529420903289 https://www.itu.int/dms_pub/itu-d/opb/inno/d-inno-profile.northmacedonia-2023-pdf-e.pdf https://www.itu.int/dms_pub/itu-d/opb/inno/d-inno-profile.northmacedonia-2023-pdf-e.pdf unlocking regional value chains ishamuddin, m., vaicondam, y., agha, j., burkhanov, a.u. & binti yusof, s. h. (2023). cybersecrity challenges and solutions in the fintech mobile app ecosystem. international journal of interactive mobile technologies, 17 (22), 100-116. https://doi.org/10.3991/ijim.v17i22.45261 javorcik, b. (2020). reshaping of global supply chains will take place, but it will not happen fast. journal of chinese economic and business studies, 18 (4), 321-325. https://doi.org/10.1080/14765284.2020.1855051 juliet orji, i., ojadi, f. & kalu okwara, u. (2022). the nexus between e-commerce adoption in a health pandemic and firm performance: the role of pandemic response strategies. journal of business research, 145,616-635. https://doi.org/10.1016/j.jbusres.2022.03.034 kacani, j. & shaqiri, g. (2023). emerging ict clusters in the western balkans: a convergence model based on financial statement analysis. regional science policy and practice, 1, 1-16. https://doi.org/10.1111/rsp3.12687 kacani, j., mukli, l. & hysa, e. (2022). a framework for shortvs. long-term risk indicators for outsourcing potential for enterprises participating in global value chains: evidence from western balkan countries. journal of risk and financial management, 5 (9), 401-415. https://doi.org/10.3390/jrfm15090401 kan, d., lyu, l., huang, w. & yao, w. (2022). digital economy and the upgrading of the global value chain of china's service industry. journal of theoretical and applied electronic commerce research, 17, 1279-1296. https://doi.org/10.3390/jtaer17040065 kano, l., tsang, e. w. & yeung, h. w. c. (2020). global value chains: a review of the multi-disciplinary literature. journal of international business studies, 51 (4), 577-622. https://doi.org/10.1057/s41267-020-00304-2 kasasbeh, f.i. (2021). impact of financing decisions ratios on firm accounting-based performance: evidence from jordan listed companies. future business journal, 7 (1), 15-16. https://doi.org/10.1186/s43093-021-00061-0 kiselakova, d., sofrankova,b., cabinova,v. & soltesova, j. (2018). analysis of enterprise performance and competitiveness to streamline managerial decisions. polish journal of management studies, 17 (2), 101-11. https://doi.org/10.17512/pjms.2018.17.2.09 laitinen, e.k. (2018). financial reporting: long-term change of financial ratios. american journal of industrial and business management, 8 (9), 1893-927. https://doi.org/10.4236/ajibm.2018.89128 le, h.t., hoang, d.p., doan, t.n., pham, c.h. & to, t.t. (2021). global economic sanctions, global value chains and institutional quality: empirical evidence from cross-country data. journal of international trade and economic development, 31 (3), 427-449. https://doi.org/10.1080/09638199.2021.1983634 lee, y. & falahat, m. (2019). the impact of digitalization and resources on gaining competitive advantage in international markets: mediating role of marketing, innovation and learning capabilities. technology innovation management review, 9 (11), 26-39. li, f., frederick, s. & gereffi, g. (2019). e-commerce and industrial upgrading in the chinese apparel value chain. journal of contemporary asia, 49 (1), 24-53. https://doi.org/10.1080/00472336.2018.1481220 liu, w., zhang, j., wei, s. & wang, d. (2021). factors influencing organisational efficiency in a smartlogistics ecological chain under e-commerce platform leadership. international journal of logistics research and applications, 24 (4), 364-391. https://doi.org/10.1080/13675567.2020.1758643 longbing c., yang, q. & yu, p.s. (2021). data science and ai in fintech: an overview. international journal of data science and analytics, 12:81-99. https://doi.org/10.1007/s41060-021-00278-w lukacova, m., korecko, j., jencova, s. & juskova, m. (2020). analysis of selected indicators of tax competition and tax harmonization in the eu. entrepreneurship and sustainability, 8 (1), 123-37. https://doi.org/10.9770/jesi.2020.8.1(8) mccormick, m. & somaya, d. (2020). born globals from emerging economies: reconciling early exporting with theories of internationalization. global strategy journal, 10 (2), 251-281. https://doi.org/10.1002/gsj.1368 meltzer, j.p. (2019). governing digital trade. world trade review, 18 (1), 23-48. https://doi.org/10.1017/s1474745618000502 nguyen, c.p., schinckus, c. & su, t.d. (2020). the drivers of economic complexity: international evidence from financial development and patents. international economics, 164, 140-150. https://doi.org/10.1016/j.inteco.2020.09.004 niemand, t., rigtering, j.p.c., kallmünzer, a., krauch, s. & maalaoui, a. (2021). digitalization in the financial industry: a contingency approach of entrepreneurial orientation and strategic vision on digitalization. european management journal, 39 (3): 317-326. https://doi.org/10.1016/j.emj.2020.04.008 https://doi.org/10.3991/ijim.v17i22.45261 https://doi.org/10.1080/14765284.2020.1855051 https://doi.org/10.1016/j.jbusres.2022.03.034 https://doi.org/10.1111/rsp3.12687 https://doi.org/10.3390/jrfm15090401 https://doi.org/10.3390/jtaer17040065 https://doi.org/10.1057/s41267-020-00304-2 https://doi.org/10.1186/s43093-021-00061-0 https://doi.org/10.17512/pjms.2018.17.2.09 https://doi.org/10.4236/ajibm.2018.89128 https://doi.org/10.1080/09638199.2021.1983634 https://doi.org/10.1080/00472336.2018.1481220 https://doi.org/10.1080/13675567.2020.1758643 https://doi.org/10.1007/s41060-021-00278-w https://doi.org/10.9770/jesi.2020.8.1(8) https://doi.org/10.1002/gsj.1368 https://doi.org/10.1017/s1474745618000502 https://doi.org/10.1016/j.inteco.2020.09.004 https://doi.org/10.1016/j.emj.2020.04.008 224 gentjan shaquiri et al. nugraha, d.p., setiawan, b., nathan, r.j. & fekete-farkas, m. (2022). fintech adoption drivers for innovation for smes in indonesia. journal of open innovation: technology, market and complexity, 8(4), 208. https://doi.org/10.3390/joitmc8040208 odorović, a., mckain, g., garvey, k., schizas, e., zhang, b.z., rowan, ph. & ziegler, t. (2020). fintech innovation in the western balkans: policy and regulatory implications and potential interventions. avalaible at ssrn: https://doi.org/10.2139/ssrn.3619214 oecd (2023). regulatory sandboxes in artificial intelligence. oecd digital economy papers, no. 356, oecd publishing, paris. pananond, p., gereffi, g. & pedersen, t. (2020). an integrative typology of global strategy and global value chains: the management and organization of cross-border activities. global strategy journal, 10 (3), 421-443. https://doi.org/10.1002/gsj.1388 pasquali, g., godfrey, s. & nadvi, k. (2021). understanding regional value chains through the interaction of public and private governance: insights from southern africa's apparel sector. journal of international business policy, 4, 368-389. https://doi.org/10.1057/s42214-020-00071-9 pournader, m., shi, y., seuring, s. & lenny koh, s.c. (2020). blockchain applications in supply chains, transport and logistics: a systematic review of the literature. international journal of production research, 58 (7), 2063-2081. https://doi.org/10.1080/00207543.2019.1650976 qi, j.y., & ren, y.d. (2021). the impact of digital economy penetration on global value chain position: transnational empirical research based on industrial heterogeneity. journal of international trade, 9, 105-121. https://doi.org/10.1007/s00146-022-01524-w radosavljević, k. (2023). opportunities and challenges in sustainability digital transformation and risk mitigation in emerging insurance markets: a comparative analysis between china and serbia. opportunities and challenges in sustainability, 2, 104-115. santos-jaén, j.m., gimeno-arias, f., gómez, a.l. & palacios-manzano, m. (2023). the business digitalization process in smes from the implementation of e-commerce: an empirical analysis. journal of theoretical and applied electronic commerce research, 18 (4), 1700-1720. https://doi.org/10.3390/jtaer18040086 sun, y., li, s., & wang, r. (2023). fintech: from budding to explosion-an overview of the current state of research. review of managerial science, 17, 715-755. https://doi.org/10.1007/s11846-021-00513-5 thun, s., bakås, o. & storholmen, t.c.b. (2021). development and implementation processes of digitalization in engineer-to-order manufacturing: enablers and barriers. ai and society, 37 (2), 725-743. https://doi.org/10.1007/s00146-021-01174-4 truby, j., dahdal, a., caudevilla, o. (2022). global blockchain-based trade finance solutions: analysis of governance models and impact on local laws in six jurisdictions. global journal of comparative law, 11(2), 167-196. https://doi.org/10.1163/2211906x-11020001 wielgos, d.m., homburg, c. & kuehnl, c. (2021). digital business capability: its impact on firm and customer performance. journal of the academy of marketing science, 49, 762-789. https://doi.org/10.1007/s11747-021-00771-5 yue, w. (2022). foreign direct investment and the innovation performance of local enterprises. humanities and social science communication, 9(1), 252. https://doi.org/10.3390/joitmc8040208 https://doi.org/10.2139/ssrn.3619214 https://doi.org/10.1002/gsj.1388 https://doi.org/10.1057/s42214-020-00071-9 https://doi.org/10.1080/00207543.2019.1650976 https://doi.org/10.1007/s00146-022-01524-w https://doi.org/10.3390/jtaer18040086 https://doi.org/10.1007/s11846-021-00513-5 https://doi.org/10.1007/s00146-021-01174-4 https://doi.org/10.1163/2211906x-11020001 https://doi.org/10.1007/s11747-021-00771-5 1. introduction 2. theoretical background 3. methodology and case study overview 4. main results 5. conclusions and policy recommendations references © theauthor(s)2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 1 (2024), pages 29-47 https://doi.org/10.17979/ejge.2024.13.1.9953 submitted: october 2, 2023 accepted: february 15, 2024 published: june 6, 2024 article exploring citizen participation in fiscal control: insights from manizales, colombia carlos david cardona-arenas,1,* johary andrea campuzano 2 1 universidad de manizales, colombia 2 universidad autónoma de manizales, colombia *correspondence: carloscardona@umanizales.edu.co abstract. the objective of this study is to identify the fundamental determinants of citizen participation within the framework of the "let's take control" initiative, implemented by the general comptroller's office of manizales municipality during 2020-2021, with a specific focus on fostering public participation in fiscal control. utilizing a two-stage mixed-method methodology, the research encompasses surveys administered to citizen observers, municipal officials, and the general population, followed by a descriptive exploratory analysis. subsequent to data collection, a series of statistical analyses is conducted, including chi-square and anova tests, along with ordinary least squares linear regression. additionally, a generalized linear model (glm) is estimated. the findings unequivocally demonstrate a positive impact on citizen participation facilitated by the "let's take control" strategy. furthermore, positive correlations emerge between participation frequency and variables such as invitations to participation exercises, age groups of participants, and their socio-economic strata within events organized by the general comptroller's office of manizales municipality-colombia. keywords: public strategy; citizen participation; fiscal control; government jel classification: h11; h83; d72; d83 1. introduction citizen participation is a fundamental right that allows citizens to become involved in social affairs and contribute to the decision-making process. in colombia, the political constitution1 establishes mechanisms for citizen participation in the exercise of their sovereignty, which include voting, plebiscite, referendum, popular consultation, open town hall meetings, legislative initiative, and revocation of mandate. these mechanisms of citizen participation are essential for democracy and have been discussed, constructed, and used in many democracies in response to the crisis of representation. fiscal control is a fundamental mechanism for ensuring transparency in public management and the protection of the state's assets. in colombia, this control is exercised through the general 1 see constitución política de colombia (2023). https://creativecommons.org/licenses/by-nc/4.0/ 30 cardona-arenas and campuzano comptroller's office of the republic and territorial comptroller's offices2, as well as through social control, in which the community plays a fundamental role. law 850 of 2003 regulates citizen observer committees as mechanisms of democratic representation that allow citizens and community organizations to monitor public management at all levels and areas where public resources are used. the intersection of citizen control and fiscal control is participatory fiscal control, which seeks the care of public goods and resources. according to eberhardt (2015), in the set of western democracies, mechanisms of citizen participation such as the popular initiative, the public consultation, the public hearing, and the participatory budget have been incorporated in response to the loss of credibility of both the rulers elected through the popular vote and regarding the functioning of traditional representative bodies. lerner, j. (2017) and hong and cho (2018) suggest that the implementation of participatory budgeting can contribute to improving equity in resource allocation, albeit it may be less effective than bureaucratic decisionmaking in achieving public objectives. the law 850 of 2003 regulates citizen oversight as democratic representation mechanisms that allow citizens and community organizations to monitor public management at all levels and areas where public resources are used. the intersection of citizen oversight and fiscal control is participatory fiscal control, which seeks to ensure the care of public assets and resources. according to diaz et al. (2003), in colombia, the fiscal control model defined by the constitution of 1991 has been applied, in which departments and municipalities have autonomy for the organization and selection of their respective comptrollers. however, from the author’s perspective, instead of improving the fiscal control system, politicking and political patronage have deteriorated its quality. the study conducted by galais, navarro, and fontcuberta (2013), emphasizes that the quality of participatory processes or experiences is often a multidimensional phenomenon. it is observed that the levels of this quality vary among dimensions and that these dimensions are differently related to various aspects of the context in which they take place. this study aims to analyze the role of municipal comptroller's offices in colombia and demonstrate the importance of linking citizen participation and information and communication technologies (ict) to achieve greater transparency and effectiveness in the fiscal management of public resources. citizen participation in fiscal control is essential to guarantee the protection of public assets and permanent monitoring of public resources, and ict can be a valuable tool to promote participation and transparency in public management. this article analyzes the results of the "let's take control" strategy implemented by the general comptroller's office of the municipality of manizales during the period 2020-2021, in terms of citizen participation in fiscal control. to achieve this objective, a mixed methodology was used, consisting of two stages. in the first stage, a survey was applied to citizen watchdogs, officials of the central administration of the municipality, and citizens in general, followed by a descriptive exploratory analysis. in the second stage, the collected information was consolidated, and a statistical analysis was performed that included chi-square and anova tests, as well as the estimation of a generalized linear model (glm). 2 see contraloría general de la republica (2023). exploring citizen participation in fiscal control the results indicated a positive influence of citizen participation in fiscal control through the "let's take control" strategy. additionally, evidence was found that being invited to citizen participation exercises, age range, and socioeconomic stratum of participants are significantly related to the frequency of participation in events called by the general comptroller's office of the municipality of manizales. in colombia, fiscal control is essential to ensure transparency in public management and the protection of the state's heritage. citizen participation in fiscal control is a manifestation of the principle of participatory democracy, and citizen participation mechanisms have been discussed, built, and used in many democracies in response to civic disaffection. in conclusion, it is imperative to underscore that this study offers indispensable insights into the specific context of manizales, with a keen focus on the dynamics of citizen participation and fiscal control. such a targeted approach will undoubtedly enhance the depth and contextual understanding for of readers vested in these participatory processes. for policy-makers, this study stands as a valuable resource, weaving vital connections between theory and regulations concerning citizen involvement within the realm of public administration. boasting a considerable social significance, this study tackles critical issues such as institutional trust and the battle against social exclusion. not only does its pinpoint challenges and institutional barriers surrounding citizen participation, but it also puts forth tangible solutions and recommendations, with the implementation of participatory budgeting standing out prominently. these proposals hold the potential to reshape the current landscape and nurture a positive societal transformation, ultimately reshaping the citizen's role in public governance and concurrently igniting change from a governmental vantage point. it is important to note that recent developments regarding the determinants of citizen participation focus on: i. factors explaining citizens' participation in budgetary processes; or ii. the effects resulting from patterns of participation. this study primarily focuses on the first perspective, following a conceptual framework very similar to that established by ríos, benito, and bastida (2017) and manes-rossi et al. (2023). both studies examine the driving or determining factors of citizen participation. in our case, these driving factors are framed within the "let's take control" strategy. additionally, we analyze the relationships between these driving or determining factors, which include the spaces provided and the calls made by the comptroller's office to encourage citizen participation, as well as the socio-economic characteristics of the population. the main findings of the research provide valuable contributions to the theory of citizen participation. these include the influence of educational level and socioeconomic stratum on participation, the relationship between geographic location and willingness to participate, civic motivation for reporting corruption, the effectiveness of technology-based participation strategies, the need to strengthen participation as an integral part of public management, and the importance of understanding who participates and why. collectively, these findings enrich our understanding of how to foster more effective and meaningful citizen participation in public decision-making. this research article is structured as follows: in the first section, a literature review is presented, highlighting the importance of citizen participation in public management and fiscal control as the articulating analytical category, followed by the methodology section, emphasizing the methods, variables, data, and instruments applied. the results and discussion section are presented next, followed by the conclusions of the study in the final section. 32 cardona-arenas and campuzano 2. literature review the importance of the concept of citizen participation has aroused significant interest in both political and academic circles, and this interest has its roots in different historical moments and geographical contexts. in contemporary europe and north america, this interest has been largely driven by the pursuit of the welfare state3. in contrast, in latin america, the focus has been largely on the discontent generated by exclusion and the lack of representation in political decision-making. this critical approach is evident in the words of hengstenberg (1999), who criticize political parties for their limited capacity to channel citizen demands. therefore, the relevance of citizen participation in public administration has been extensively researched in academic literature. arnstein's (1969) "ladder" of citizen participation, ranging from manipulation to citizen control, provides a framework illustrating different levels of involvement in decision-making in public management. in a similar context, the debate presented by crosby, kelly, and schaefer (1986) offers an intriguing perspective on forms of participation, introducing the concept of "maximum feasible participation." in their studies, a call for citizen participation is made in the united states, with an emphasis on the fact that most participatory programs focus on intervention at the local administrative level, neglecting broader political agendas, such as those at the national level. an important study for the objectives of the present research is the research conducted by aguirre (2021). in this case study, the implementation process of participatory budgeting in the district of villa el salvador (peru) between 2000 and early 2002 is examined. this participatory initiative in villa el salvador serves as an exercise in constructing inclusive environments. the study particularly delves into the experience and service outcomes that the residents of villa el salvador had regarding the participatory budget for the years 2000-2002, highlighting the necessary interaction with organizations and public officials. seven valuable lessons are extracted for those interested in promoting citizen participation, including the coordination of participation and accountability, the use of citizen recognition as an incentive, attention to implementation processes, the creation of conditions for organizational learning, the institutionalization of changes, the minimization of competitive processes, and the promotion of commitment from elected officials. if we consider the possibility that dynamics similar to a principal-agent relationship may arise in the relationship between government and citizens in terms of information availability, we can also recognize the existence of power relations between both parties. these power relations are constructed in the social structure, as castells (2013) points out through spatiotemporal formations that operate at both local and global levels. in this context, communication networks acquire immense social power, especially those with a high degree of interconnectivity due to recent technological innovations. 3 in the literature review, we have sought to recognize the diverse traditions of citizen participation worldwide, with a particular emphasis on northern and southern countries. overall, empirical research in this field barely considers such diversity, especially in europe and north america. the reason behind this is that in these regions, participatory tradition has a more entrenched history and has developed very identity-centric approaches. exploring citizen participation in fiscal control according to rowe and frewer (2004), assessing the benefits of participation is challenging, given that there are relatively few cases where the effectiveness of participatory exercises has been systematically examined. this situation contrasts with the abundance of studies that approach the subject subjectively. however, there is a common denominator among these studies, which is to highlight the potential of citizen participation as an empowerment tool, in addition to its contribution to improving the efficiency and effectiveness of government management in public administration. in this line, gaventa (2006) emphasizes the potential of participation as a catalyst for empowerment and social transformation, requiring the identification of areas of influence and innovative forms of civic participation. in the context of participatory budgeting in european nations, sintomer, herzberg, and röcke (2012) emphasize the crucial role of citizen engagement, as it enhances transparency, accountability, and the legitimacy of democratic processes. furthermore, baiocchi and ganuza (2014) argue that participation should go beyond the mere efficiency of services, aiming to empower individuals. ramos and reyes (2005) maintain that citizen participation, through civil organizations and society in general, is primarily characterized in by local entities, within the context, for example, of mexican federal states. alternatively, ríos, benito, and bastida (2017) focus on identifying the socio-economic, institutional, and political factors that promote public participation in the central government budgetary process within a context that encompasses 93 countries. the results obtained indicate that internet penetration, population diversity, government financial situation, and budget transparency are key determinants of the opportunities for public participation in this process. on the other hand, faúndez and arancibia (2022) present an innovative model of citizen participation institutionalized in public bodies. using the analytic hierarchy process methodology, they examine critical factors for a successful implementation of participation. this model addresses fundamental criteria such as leadership, normative scope, team commitment, and management culture, offering a comprehensive approach to promote effective citizen engagement. to further enrich the literature review on citizen participation, governance, and public management, prominent sources such as draude, börzel, and risse (2018), as well as bovens, goodin, and schillemans (2021), are recommended. these sources delve into the interactive dynamics between citizen participation and effective governance. delving deeper into the literature review, we highlight the key contributions of these studies. gonçalves (2014) conducts a comprehensive analysis of the effects of participatory budgeting on municipal expenditures and infant mortality in brazil, emphasizing the fundamental importance of citizen participation in the budgetary decision-making process. on the other hand, schneider and busse (2019) provide a meticulous review of empirical findings related to the implementation of participatory budgeting in germany, underscoring its relevance and adaptability in a european context. montambeault (2016) explores the various pathways followed by participatory citizens in the context of participatory budgeting in brazil, investigating how this practice significantly contributes to the formation of an active citizenship. furthermore, coleman and cardoso sampaio (2017) conduct thorough research into the sustainability of this democratic innovation through the study of electronic participatory budgets in belo horizonte, identifying both the challenges it faces and the opportunities offered by electronic participation. together, these research studies provide a comprehensive understanding of the impacts and prospects of participatory budgeting in different contexts. 34 cardona-arenas and campuzano now, these studies connect with the findings of garcía (2009) regarding participatory fiscal control in colombia. the collaboration between government entities and citizens, facilitated by the general comptroller's office, strengthens oversight mechanisms. despite identifying possible fiscal responsibilities through citizen complaints integrated into audits, comprehensive citizen supervision of public management is still in its early stages. in turn, participatory fiscal control promotes a more effective use of public resources and leads to greater community trust in the entities responsible for fiscal control in the country (garcía mejía, 2015). therefore, it can be accepted that fiscal control and citizen participation are integral components of effective public management. citizen participation contributes to uncovering corruption, while arnstein's theory suggests that increasing citizen participation in fiscal control processes can be beneficial. the proposal for a national fiscal control system, in line with the ideas of ochoa and charris (2003), seeks to promote fair supervision. the evolution of citizen participation throughout history is crucial for shaping constitutional and fiscal oversight processes. in this regard, chavez and niño (2018) delve into participatory fiscal control in montería, revealing various forms of citizen participation. this case study aims to analyze the historical journey of citizen participation, as well as address the importance of the legacy of citizen participation from past generations. on the other hand, montero (2021), as a local precedent, relates fiscal control to the principles of the social and solidarity economy, establishing a conceptual framework between these categories of analysis. this study specifically examines participatory fiscal control in manizales, based on a comprehensive review of colombia's territorial fiscal control system. sánchez-torres el al (2006) categorizes problems to propose improvements to territorial fiscal control, emphasizing a strong trend towards the development of activities related to the promotion and implementation of citizen participation mechanisms to address citizen demands. however, as highlighted by goméz and pinzón (2012), there are significant gaps in territorial comptrollerships in terms of strategies and even results regarding citizen participation exercises. they also explore the relevance of complementary tools in colombian participatory fiscal control. these contributions align with contemporary public administration and participatory practices. cardenas et al. (2013) raises relevant questions about the effectiveness of the general comptroller's participatory fiscal control strategies. garcía mejía (2015) emphasizes the importance of knowledge in the effectiveness of participatory fiscal control, emphasizing perspectives in the search for effective governance in the city of manizales. based on the promotion of civic education aimed at fostering direct participation in fiscal control, the proposal by liendo and de las nieves (2021) presents a theoretical-practical approach that underscores the crucial importance of citizen training in the context of public management control, specifically within the comptroller's office of sucre, state of aragua, venezuela. this article addresses the imperative need to involve citizens in the auditing process, focusing specifically on the execution of works and the provision of public services within their own communities. the fundamental purpose of this research is to raise awareness among comptroller's office officials regarding the relevance of citizen participation in fiscal control of public management. additionally, the aim is to motivate the general public to actively participate and develop the necessary technical skills to effectively contribute to fiscal control conducted by comptroller's offices. this scientific exploring citizen participation in fiscal control article seeks to provide a solid theoretical and practical framework supporting the importance of citizen education in the realm of fiscal control, thereby promoting the construction of a more engaged and empowered society in the oversight and improvement of public management. the studies reviewed up to this point in this state-of-the-art aim to emphasize the central role of citizen participation in public administration. it is concluded that it is essential to promote democracy, empowerment, and transformative change in local social structures, which require greater representation in control and oversight bodies. furthermore, it goes beyond the mere importance of citizen participation in physically supervising government efficiency, focusing on promoting an engaged and active citizenry in transparent and inclusive government exercises. 3. methods this research employs a mixed-method design, utilizing a cross-sectional approach and short-term longitudinal analysis for monitoring the participation indicators of the 'take control' strategy, covering the strategy period from q1 2020 to q4 2021, aligned with legislative act 04 of 2019 and the municipal administration's timeline from 2020 to 2023. it is worth noting that the data used for model estimation follows a cross-sectional approach. the study is empirically analytical with an explanatory scope, aiming to identify institutional determinants influencing citizen participation in social and fiscal control. the examination revolves around management indicators from the contraloría de manizales. methodologically, the study outlines the initiation and execution of the "let's take control and monitor the proper use of public resources" strategy during 2020-2021. this is based on quarterly reports from the office of citizen participation to the auditor general of the republic, forming the longitudinal aspect of the study. information analysis begins with a descriptive and analytical assessment using primary data collection. an instrument validated by experts characterizes key stakeholders in the "let's take control" strategy. these stakeholders are central to analyzing the determinants of citizen participation within this public management approach. subsequently, variables' behavior is examined, considering both the literature and the strategy itself. notably, this methodological design addresses initial research information needs and strives to minimize bias while upholding anonymity and interpretational rigor principles. the majority of literature on the outcomes of participatory fiscal control has primarily taken a descriptive-analytical approach. this study stands out as one of the few that makes an empirical effort to analyze the results of participatory fiscal control. however, within the conceptual framework of citizen participation, empirical studies are more common. notably, the work by theocharis and van deth (2018) employs various statistical inference methods, such as logistic regressions, to analyze different mechanisms of political participation. it is crucial to note that theocharis and van deth's (2018) work underscores the limited presence of instruments for measuring newer forms of participation in large-scale surveys. these surveys consistently identify and measure political participation solely through established repertoires. consequently, the present study aims to address this gap by designing and implementing 36 cardona-arenas and campuzano an expert-validated instrument. this instrument seeks to collect maximum objective information from a previously unstudied sample, enabling the discovery of valuable latent information regarding the relationship between the described variables (variables, data, and information sources). 3.1 variables, data, and information sources the study focuses on officials from the general comptroller's office of the municipality of manizales, as well as citizen oversight groups and the general public participating in the activities of the comptroller's office. as pimienta, gulfo, and suarez (2021) indicate, the construction of a social rule of law is grounded in practical citizen experiences. therefore, these actors contribute a significant perspective based on their knowledge and experiences to carry out the present research. obtaining opinions regarding the subjects of control is of utmost importance, as they are the ones audited and for whom the reports are intended. in accordance with the research question, it is important to highlight that, for the purposes of this study, effects on differentiated groups are not considered, and the explained variance for such groups is not decomposed. this is because, from both the study's perspective and that of the oversight body, citizens, officials, and employees are considered subjects of control, forming part of a broader category of individuals under study. due to the population's diversity and asymmetry, a convenience sampling approach is chosen. this decision is supported by limited citizen tracing and the study's alignment with the "let's take control" strategy. to mitigate selection bias, systematic sampling is used, involving regular interval selection from the population list. randomization is introduced during participant selection, resulting in a sample of 48 individuals, including citizen overseers and control subjects. ethical standards are upheld, with informed consent obtained and data protection regulations followed. in accordance with the research question, it is emphasized that, for the purposes of this study, effects on differentiated groups are not considered, and the explained variance for such groups is not decomposed. this is because, both from the study's perspective and that of the oversight body, citizens, officials, and employees are regarded as subjects of control, forming part of a broader category of individuals under study. the data include discrete quantitative information, quarterly aggregated, and coded polytomous qualitative data using ordinal likert scales. the study covers the period from q1 2020 to q4 2021, aligned with legislative act 04 of 2019 and the municipal administration's timeline from 2020 to 2023. methods involve descriptive analysis, anova estimations, and generalized linear models (glm). data sources encompass primary survey data, secondary data from official websites, and comptroller's office's facebook platforms, such as the integrated system of audits-mission sia and the citizen attention system. exploring citizen participation in fiscal control 3.2 anova estimation anova models are highly valuable for this study due to the presence of categorical predictor variables with distinct classification levels. each subject was assessed using a primary data collection tool, and the target response variable in the initial model is "the number of times they have participated in citizen participation exercises organized by the comptroller's office of manizales city." this response variable is numerical, and the primary goal is to examine the mean across different levels or categories of explanatory variables. beyond comparing various groups, there's also a focus on numerically quantifying these differences through such estimation. given the nature of the observations, employing this type of model is crucial for achieving the research objective. in essence, anova models are indispensable for comprehensively analyzing the collected data and deriving robust conclusions from this study. the general representation is presented below: y_ij=μ+β_i g_j+ε_ij [1] μ is a constant, and β_i for i=1,.., g, are independently distributed random variables across their different levels or categories. for this model, it is verified that e[y_ij ]= μ and the variance of y_ij denoted by σ_y^2=σ_β^2+σ^2 where σ_β^2,σ^2 are components of the variance. this is why the model is referred to as the variance components model. it's worth noting that the main difference from regression models is that in this case, we do not model the entire observed response, but rather the observed mean of the set of observations for each level of the factor. 3.3 generalized linear model (glm) estimation in this model, it is assumed that the dependent variable is linearly related to the set of predictor variables. this model is particularly useful as it is an extension of linear models that allow for the use of non-normal distributions. the model fit is achieved using the maximum likelihood method. the single equation linear model is defined as: y_i=x_i β+u_i, [2] where y_i and u_i are scalars and x_i and β are column vectors, for this model, it is verified that: e[u_i ]= 0, the residuals of the model are independent from each other. the relationship between the dependent variable and the independent variables is linear in the predictor space: y_i=β_0+β_1 x_1i+β_2 x_2i+..+β_k x_ki+u_i. where y_i is the dependent variable, x_(1 ),x_(2 ),x_(k )are the independent variables, β0, β1, β2, ..., βk are the coefficients of the model, and u_i is the random error. the variance of the dependent variable is constant at all values of the independent variables (homoscedasticity): var(y) = σ^2, where σ^2 is a constant that represents the variance of the dependent variable. the errors of the dependent variable are independent and identically distributed: u_i ~ i.i.d. n (0, σ^2), i.i.d. means "independent and identically distributed. and the link 38 cardona-arenas and campuzano function of the model is determined by: μ =e[y]=β_1 x_1+⋯+β_k x_k, according to müller (2004) the resulting estimator β has an asymptotic normal distribution. for further reading on glm we refer to dunn, p. k., and smyth, g. k. (2018). 3.4 citizen participation in the municipality of manizales the citizen participation office of the general comptroller's office of the municipality of manizales carried out training and citizen control activities with citizen watchdogs, officials and employees of control subjects, student auditors, and the general public between 2020 and 2021, following the schedule presented in table 1. table 1. schedule of citizen participation activities of the manizales comptroller's office. management carried out by the citizen participation office, implementing the "let's take control" strategy. 2020 2021 promotion of social control and participatory fiscal control to raise awareness among young people in educational institutions in manizales. training in crimes against the public administration information about the processes carried out by the comptroller's office through political control debates. introduction to audit processes information on the processes carried out by the office of the comptroller through political control debates. contractual analysis in audited processes fiscal control and citizen participation mechanisms. consulting for public and private entities mechanisms of citizen participation first departmental meeting of student comptrollers and deputy comptrollers observe sia platform general structure of the special jurisdiction for peace common errors in fiscal management. principles and elements of an audit with an intosai approach integrated planning and management model (mipg) social control and transparency law appropriation of state contracting processes by student governments. work table with ombudsmen and student comptrollers analysis of institutional performance index results. verification of public discourse and misinformation general aspects of the audit process. design of strategies to continue the student controllers’ program public procurement launch of the project monitoring mental health policy in educational institutions general information on the application of decree 806 of 2020 for the legal defense of public entities. control route, space for dialogue, training for the citizens of vereda la cabaña the role of the comptroller general of the republic in citizen participation reform to the general disciplinary code and its implications. exploring citizen participation in fiscal control management carried out by the citizen participation office, implementing the "let's take control" strategy. 2020 2021 transparency and access to information social control in the public procurement process network of citizen oversight information as a mechanism for citizen oversight from the media linking in the week of citizen participation biocultural manizales, process of social control and citizen participation function of the comptroller environmental social control from public entities procedure for supervision and interventions citizen participation and social control meeting with municipal representatives: fiscal control and the most common errors in fiscal management. products strategic alliance with the uam role of the agr in fiscal control public accounting regime effective social control in times of covid19 cultural event "pact for participation constitutional mechanisms in social control. promotion of social control with young people of saint pius x virtual updating day practical aspects for the formulation of projects. general aspects of the manizales comptroller's office with young people from the cyprus institute objectives of sustainable development and control of legality of manifest emergencies dialogue with citizens on the management of the manizales comptroller's office 4. results and discussion to assess the extent of knowledge and involvement in participatory fiscal control, a survey was conducted (see appendix 1). this instrument was administered to citizen monitors, officials from control entities, and the general public, resulting in 48 completed surveys. the findings from the surveys are depicted in figure 1. notably, 25% of the 48 respondents fall within the 25 to 34 age group. this aligns with the education-related query, where 29% of participants possess higher education degrees at the undergraduate level. conversely, 21% of respondents are aged between 18 and 24, with a predominance of primary education qualifications. in figure 2, it is apparent that 40% of respondents belong to socioeconomic stratum four (4) in the city of manizales, with 33% residing in the atardeceres commune. these outcomes underscore the significance of citizen participation in shaping public policies. as escobar (2017) emphasizes, the advancements in state modernization across latin america underscore the imperative of involving civil society as a pivotal player in this trajectory. 40 cardona-arenas and campuzano figure 1. age range and level of education. figure 2. commune of residence and socioeconomic stratum. throughout this study, reference is made to the ideas put forth by arnstein (1969), and this work contributes to the field of knowledge by presenting a more comprehensive model of partnership compared to the one proposed by the aforementioned author. this marks a new starting point for the emerging dialogue on the interaction between local government and community groups, aimed at the next generation of planners and scholars. this contribution prompts a profound reflection on the need to more dynamically promote initiatives that foster equitable citizen participation, advocating for the establishment of long-term "partnerships" with local community groups. for participation to become an instrument of development, empowerment, and social equity, it is crucial to be meaningful and authentic, involving all stakeholders and differentiating but synchronizing their roles across various educational realms. this extends beyond the classroom, encompassing both educational policy and formal and informal education. torres (2001) emphasizes that education is not confined to schooling, and the necessary learnings for life, work, participation, and full citizenship cannot be limited to a specific life stage. the learning process commences at birth and extends throughout life, transcending school institutions to encompass a wide array of institutions, modalities, relationships, and practices. thus, education, the educational community, and educational policy prove to be much broader entities than their school-based counterparts, revealing a more holistic and dynamic landscape. consequently, this is how the budget, as a mechanism of political decision-making according to nebot (2001), serves as the document that accurately reflects the economic and social priorities of the government. this instrument translates governmental policies and commitments into decisions that determine how resources will be obtained and where they will be spent. therefore, it is the concrete tool through which governments implement action plans and programs aimed at addressing the needs of citizens, defining their quality of life. consequently, it becomes the means by which society can exercise control over the government's management of public resources. 6% 21% 25% 19% 10% 19% 21% 13% 10% 25% 29% 2% 0% 10% 20% 30% 40% under 18 years between 18 and 24… between 25 and 34… between 35 and 44… between 45 and 54… more than 54 years elementary school technique technological undergraduate postgraduate none 33% 6% 13% 0% 10% 6% 2% 8% 0% 15% 6% 4% 8% 29% 40% 15% 4% 0% 20% 40% 60% atardeceres ciudadela norte cumanday ecoturístico cerro… estación fuente macarena palogrande san josé tesorito universitaria stratum 1 stratum 2 stratum 3 stratum 4 stratum 5 stratum 6 exploring citizen participation in fiscal control figure 3: times (a) and frequency (b) of participation. figure 4: participation in the control strategy, and channels of participation. within this context, participatory budgeting is described as a direct-democracy approach to decisionmaking in resource allocation (calabrese, williams & gupta, 2020). in this regard, calabrese, williams & gupta (2020) highlight that participatory budgeting can be an interesting policy device, where fiscal allocation may result in increased funding for projects, albeit at lower amounts. based on this analysis, the present research suggests not only advancing participatory control but also considering mechanisms for participatory budget allocation. while the latter is not the main focus of this work, it represents a promising direction for future research. figure 3 shows the number of times and the frequency with which the subjects have participated in the activities convened by the comptroller’s office. it is highlighted that 21% have participated more than once in total, but at the same time, 56% have never participated. similarly, in terms of frequency, the response ‘never’ also predominates with 33%, followed by ‘at least once a week’ with 23%, and ‘at least once a year’ with 15%. this response raises uncertainties, especially considering the efforts this entity has made to involve citizens in state decisions. traditionally, the municipality has been perceived from a purely technical-administrative standpoint, sidelining considerations of social underpinnings and political dynamics. similarly, figure four shows the means by which people have participated in the ‘take control’ strategy of the general comptroller’s office of the municipality of manizales. it is highlighted that facebook live stands out with 38%. figure 6 provides valuable insights into the extent of familiarity respondents from manizales have with control agencies nationwide. impressively, over 60% of participants demonstrated a comprehensive understanding of all control agencies, showcasing a notable level of awareness on this matter. turning to figure 7, which delves into the motivations behind reporting instances of corrupt practices, a significant 69% of respondents viewed it as a civic duty to expose such misconduct to the appropriate authorities. 56% 4% 4% 4% 2% 4% 4% 21% 0% 20% 40% 60% never 1 time 2 times 3 times 4 times 5 times 6 times more than… times 23% 10% 2% 8% 8% 15% 33% 0% 20% 40% once a week twice a month once a month every 2 months once a semester once a year never frecuency 38% 4% 4% 52% 2% 0% 20% 40% 60% facebook live youtube face to face neither all 42 cardona-arenas and campuzano . figure 6. reasons that motivate the allegations of corruption. figure 7. perception of the work carried out by the general comptroller's office of the municipality of manizales. this alignment of viewpoints in figure 7 finds support in cardona et al (2020) emphasize the relationship between institutions and economic development from the perspective of sen’s (2000) approaches. in this case, to achieve true results of economic freedom, it is important to enhance political freedoms and guarantees of transparency. thus, the ‘take control’ strategy contributes to the achievement of these objectives in the context of the municipality of the city of manizales. in this section, the statistical analyses of the information collected through the perception and knowledge survey on participatory fiscal control in the city of manizales are shown. first, descriptive statistics of the variables are presented, followed by statistical tests of the relationship between the implementation of the let's take control strategy and monitoring of the proper use of public resources with the determined citizen participation variables, and finally, the estimated models are presented. anova test between variables for the association test between variables, the basis is the anova analysis. according to the study by dagnino (2014), this anova test4, which stands for analysis of variance, was developed by ronald aylmer fisher (1890-1962) and allows measuring the statistical significance of the differences 4 it’s important to note that anova assumes that the data follow a normal distribution and that the samples are independent. additionally, anova is based on comparing the variance between groups with the variance within groups. if the variability between groups is much greater than the variability within groups, it is likely that a significant association exists between the independent variable and the dependent variable. 23% 8% 69% 0% 10% 20% 30% 40% 50% 60% 70% 80% 0% 10% 21% 46% 23% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 50% 30 cardona-arenas and campuzano between the means of the groups determined in the dependent variable "number of times they have participated in activities called by the general comptroller's office of the municipality of manizales" for the dichotomous and polychotomous qualitative variables that have been constructed from the applied information collection instrument. the equations corresponding to the estimation of anova 1 and anova 2 models presented in table 2 are for anova 1: 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇_𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑝𝑝𝑇𝑇𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑝𝑝𝑖𝑖 = 𝑝𝑝 + 𝑇𝑇𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑠𝑠𝑇𝑇𝑖𝑖 + 𝑇𝑇𝑠𝑠𝑇𝑇𝑇𝑇𝑠𝑠𝑠𝑠𝑇𝑇𝑝𝑝𝑖𝑖 + 𝑠𝑠𝑖𝑖 . and, for anova 2: 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇_𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑝𝑝𝑇𝑇𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑝𝑝𝑖𝑖 = 𝑝𝑝 + comptroller′s office 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑝𝑝𝑇𝑇𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑠𝑠𝑠𝑠𝑖𝑖 + 𝐿𝐿𝑇𝑇𝑝𝑝´𝑇𝑇 𝑝𝑝𝑝𝑝𝑡𝑡𝑇𝑇 𝑝𝑝𝑠𝑠𝑠𝑠𝑝𝑝𝑝𝑝𝑠𝑠𝑐𝑐 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑝𝑝𝑇𝑇𝑝𝑝𝑝𝑝𝑝𝑝𝑇𝑇𝑠𝑠𝑠𝑠𝑖𝑖 + 𝑠𝑠𝑖𝑖. tables 2 and 3 present the estimated coefficients and statistical results. table 2. anova 1 number of obs: 48 r-squared: 0.5999 root mse: 501.159 adj r-squared:0.5414 partial ss df ms f prob>f model 1544.2205*** 6 257.37009 10.25 0.0000 dep: times participated in citizen participation activities socioeconomic stratum 338.90565** 5 67.781129 2.70 0.0338 have you been summoned to the events held at the comptroller general of the municipality of manizales 905.16805*** 1 905.16805 36.04 0.0000 residual 1029.7586 41 25.116064 total 2573.9792 47 54.765514 source: own calculations. data collection instrument. note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. above table 2 there is a summary of the underlying regression. the model was fit on 48 observations, and the root mean squared error (root mse) is 501. the r2 for the model is 0.5999, and the adjusted r2 is 0.5414. the first line of the table summarizes the model. the sum of squares (partial ss) for the model is 1544.22 with 6 degrees of freedom (df). the corresponding f statistic is 10.25 and has a significance level of 0.000. thus, the model appears to be significant at the 0.0001% level and a confidence level of 99%. table 3 includes a summary of the underlying regression. the model was fit on 47 observations, and the root mean squared error (root mse) is 4.40. the r2 for the model is 0.6591, and the adjusted r2 is 0.6176. the first line of the table summarizes the model. the sum of squares (partial ss) for the model is 1535.31 with 5 degrees of freedom (df). the corresponding f statistic is 15.86 and has a significance level of 0.000. thus, the model appears to be significant at the 0.0001% level and a confidence level of 99%. exploring citizen participation in fiscal control table 3. anova 2 number of obs: 47 r-squared: 0.6591 root mse: 4.40056 adj r squared: 0.6176 model partial ss df ms f prob>f 1535.3133*** 5 307.06266 15.86 0.000 dep: times participated in citizen participation activities have you participated in the citizen events organized by the general comptroller's office of the municipality of manizales? 142.78901*** 1 142.78901 7.37 0.009 have you participated in the "take control" strategy of the general comptroller's office of the municipality of manizales? 98.807531** 1 98.807531 5.10 0.029 what is your perception of the work carried out by the general comptroller's office of the municipality of manizales? 103.21083 3 34.403609 1.78 0.166 residual 793.9633 41 19.364959 total 2573.9792 46 50.636448 source: own calculations. data collection instrument. note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. appendix 2 at the end of the document presents the correlation matrix. specifically, this matrix reveals that the number of times individuals have participated in citizen participation events scheduled by the controller's office correlates positively and significantly, with a confidence level of 99%, with age range. conversely, it correlates negatively and significantly, with the same confidence level, with participation frequency, knowledge about citizen participation processes, participation in social networks, knowledge about the "take control" strategy, and socioeconomic stratum, with a confidence level of 90%. these findings suggest that participating a certain number of times may lead to a decrease in interest in participating. furthermore, contrary to intuition, better socioeconomic conditions and greater awareness of citizen participation mechanisms negatively correlate with participation frequency. this highlights intriguing cultural aspects that warrant deeper investigation. it is worth noting that participation in the "take control" strategy negatively and significantly correlates, with a confidence level of 99%, with age range and educational level. however, it shows a positive and significant correlation, at the same confidence level, with participation in social networks, knowledge about citizen participation, and socioeconomic stratum. this may indicate a positive relationship with the process of embracing citizen participation mechanisms through the public strategy "take control." it is interesting to observe, from a gender perspective, that considering gender (male or female) bears no relevance to the set of variables under analysis. this implies that, at least in manizales, the gender variable is not significant in relation to citizen participation processes. it's worth noting that certain counterintuitive relationships will be the subject of analysis in future research. for the time being, in estimating the glm model, statistically robust and most 32 cardona-arenas and campuzano relevant relationships have been considered to fulfill the objective set forth in this research article. the outcomes of the generalized linear model (glm) analysis (table 4) indicate a decline in citizen participation prompted by the general comptroller's office in manizales. the coefficients in the glm reflect how independent variables relate to the dependent variable, factoring in other predictors within the model. these coefficients reveal the size and direction of the impact that independent variables exert on the dependent variable—here, the citizen participation spurred by the general comptroller's office in the municipality of manizales, while accounting for other variables in the model. it's crucial to recognize that the coefficient's interpretation hinges on the type of variable being modeled. for categorical variables, the coefficient signifies the difference in the dependent variable between the reference category and the modeled category. specifically, a negative coefficient in a glm with a categorical independent variable indicates that the response variable decreases on average when transitioning from the reference category to the corresponding category, keeping other variables constant. hence, in our context of modeling citizen participation prompted by the general comptroller's office in the municipality of manizales based on educational level, socioeconomic stratum, attendance at the office's events, and awareness of participatory fiscal control, the findings reveal that at higher socioeconomic strata and educational levels, the average frequency of citizen participation called by the comptroller's office decreases by -10.01 and -8.9, respectively. intriguingly, contrary to expectations, higher cultural backgrounds and socioeconomic statuses do not promote citizen participation. this could be attributed to time limitations imposed by work commitments for individuals with advanced education and greater socioeconomic standing. according to the results of the glm model, a robustness test has been conducted, which involves estimating a linear regression model with heteroskedasticity-robust errors to decompose the effect of being in each socioeconomic stratum on the frequency of participation in citizen events for fiscal control. thus, regarding the model predicting the frequency of participation in citizen participation activities for fiscal control based on socioeconomic stratum, the results suggest that as the socioeconomic stratum increases, the negative coefficient becomes more pronounced and statistically significant. this finding is observed with a 95% confidence level for strata 2, 3, and 5, and a 99% confidence level for strata 1, 4, and 6. notably, the negative coefficient reaches its maximum absolute value in socioeconomic stratum five. this reflects how as socioeconomic status rises, participation in citizen fiscal control activities tends to decrease. in summary, the results support an inverse relationship between socioeconomic stratum and participation in citizen fiscal control activities. as individuals belong to higher socioeconomic strata, they are less likely to actively participate in these activities (see the estimation results in table 5). exploring citizen participation in fiscal control table 4. generalized linear model – glm2 generalized linear models no. of obs: 48 residual df: 42 optimization: ml iteration 0: log likelihood: -143.59486 scale parameter: 26.5432 deviance: 1114.814219 (1/df) deviance: 26.5432 pearson 1114.814219 (1/df) pearson: 26.5432 log likelihood: -143.5948585 aic: 6.233119 bic: 952.2238 coef. oim std. err. z p>i z i [95% conf. interval] dep: times participated in citizen participation activities have you been summoned to the events -7.611135*** 2.452402 -3.10 0.002 -12.41776 -2.804514 education level -8.903197** 3.913082 -2.28 0.028 -16.80583 -1.000568 stratum socioeconomic -10.01808** 3.828666 -2.62 0.012 -17.75023 -2.285937 do you know what is participatory fiscal control -13.34252*** 4.111532 -3.25 0.002 -21.64593 -5.039117 _cons 20*** 3.543731 5.64 0.000 12.84329 27.15671 source: own calculations. data collection instrument. note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. table 5. least squares linear regression dep var: how many times have you participated in citizen participation activities indep var: socioeconomic stratum ss df ms number of obs: 48 model 1544.2205 6 257.370088 f (6,41) 10.25 residual 1029.75864 41 25.1160643 prob>f 0.0000 total 2573.97917 47 54.7655142 r-squares: 0.5999 adj r-squares: 0.5414 socioeconomic stratum coef. std. err. t p>i t i [95% conf. interval] stratum 1 -1288706*** 4.488201 -2.87 0.006 -21.95117 -3.822957 stratum 2 -8.903197** 3.913082 -2.28 0.028 -16.80583 -1.000568 stratum 3 -10.01808** 3.828666 -2.62 0.012 -17.75023 -2.285937 stratum 4 -13.34252*** 4.111532 -3.25 0.002 -21.64593 -5.039117 stratum 5 -13.42471** 5.069212 -2.65 0.011 -23.66219 -3.187228 stratum 6 -9.150583*** 1.524263 -6.00 0.000 -12.22889 -6.072271 _cons 20*** 3.543731 5.64 0.000 12.84329 27.15671 source: own calculations. data collection instrument. note: *, **, and *** indicate statistical significance at the 10%, 5%, and 1% levels, respectively. 34 cardona-arenas and campuzano the pearson chi-square test was applied, a non-parametric test that measures the discrepancy between an observed distribution and a theoretical one, indicating the degree of dependence between two variables. finally, is established the following the hypothesis: h1: the "let's take control" strategy developed by the general comptroller's office of the municipality of manizales, utilizing participation mechanisms based on information and communication technologies (icts), has positively influenced the outcomes of citizen participation that engages in social control over fiscal management during the period 2020 2021. null h1: the "let's take control" strategy developed by the general comptroller's office of the municipality of manizales, utilizing participation mechanisms based on information and communication technologies (icts), has negatively influenced the outcomes of citizen participation that engages in social control over fiscal management during the period 2020 2021. the cramer's v test is also applied, which is calculated from the pearson chi-square test and the sample size. it is a normalized form of the chi-square statistic that considers the size of the contingency table. the interpretation of the cramer's v coefficient is similar to that of the correlation coefficient. a value close to zero indicates that there is no association between the two categorical variables, while a value close to 1 indicates a strong association between the two categorical variables. see table 6. interpretation of effect size, below. table 6. interpretation of effect size effect size (es) interpretation es ≤0.2 the result is weak. although the result is statistically significant, the fields are only weakly associated.". 0.2 < es ≤ 0.6 the result is moderate. the fields are moderately associated es > 0.6 the result is strong. the fields are strongly associated source: ibm (2023) according to the result of the cramér's v coefficient, which is equal to 0.54 (see table 7), a moderate association can be inferred between the place of residence and participation in civic events organized by the municipal comptroller's office of manizales. this finding indicates that both variables are significantly linked in the city of manizales, suggesting that geographical location influences citizens' willingness to participate in the initiatives of the municipal comptroller's office. for more detailed information, it is recommended to refer to table 7. exploring citizen participation in fiscal control table 7. chi-square test for 'municipality of residence and participation in the citizen events of the municipal comptroller's office of manizales. 𝑥𝑥2 test value number of observations 47 𝑥𝑥2 pearson 14,14 average 0,078 cramér´s v: 0,54 to verify and confirm the hypothesis of this study, we conducted a chi-square test (x^2) between two variables: the neighborhood of residence and participation in citizen events organized by the general comptroller's office of the municipality of manizales. the null hypothesis (h0) stated that there is no association between the neighborhood of residence and participation in these events (see table 7). the results of the analysis yielded an x^2 value of 14.14 with a corresponding probability (pr) of 0.078. as the chosen significance level was α = 0.5, and since the obtained p-value is lower than this significance level, we reject the null hypothesis. this leads us to conclude that there is indeed a relationship between a citizen's neighborhood of residence and their participation in the events organized by the general comptroller's office of the municipality of manizales. based on the information collected through the survey, we find that a citizen's neighborhood of residence significantly influences their participation in the activities organized by the control entity. an analysis was carried out to ascertain the extent of association between the variables of educational level and participation in citizen events organized by the general comptroller's office of the municipality of manizales. this perspective was gleaned from inputs provided by citizen oversight groups, officials responsible for the comptroller's control subjects, and the general public. the null hypothesis posited that no association exists between these variables. however, the findings have shown that this null hypothesis was rejected with a significance level of 1%. instead, the alternative hypothesis was accepted, indicating a connection between the educational level of citizens in manizales and their active involvement in the activities orchestrated by the comptroller's office. it is worth noting that while there is an association between the mentioned variables, education level, and participation in events organized by the comptroller's office, this relationship is inverse, as can be further confirmed in the results of the model in table 4. building upon the aforementioned outcomes, the general comptroller's office of the municipality of manizales possesses the requisite tools to empower citizens in participating effectively and efficiently in the mission of ensuring proficient and timely participatory fiscal oversight over the utilization and surveillance of public resources managed by territorial entities. as stipulated by law 134 of 1994, "administrative participation as a right of individuals. participation in administrative management shall be exercised by private individuals and civil organizations in accordance with the terms of the constitution, and those established by law which develop the final clause of article 103 of the political constitution and establish the regulatory procedures required for this purpose, the requirements to be met, the definition of the decisions and matters subject to participation, as well as their exceptions and the entities in which these procedures shall operate" (p. 26). 36 cardona-arenas and campuzano to effectively and efficiently manage public resources, it becomes imperative to engage citizens as watchdogs and stimulate their participation in territorial fiscal control. this not only optimizes the utilization of public resources but also rebuilds citizens' trust and reinforces the fulfillment of the fundamental purposes of the state. therefore, citizen participation is necessary to enhance budget transparency (ríos, benito, and bastida, 2017) in the context of the municipality of manizales, the challenges of insufficient citizen participation and the weakening of community organizations stand as significant roadblocks to achieving transformation and progress within the city. while new roles founded on values like coexistence and respect for diversity are advocated for, the responsibility often falls upon leaders who exploit the situation for their personal gain. these findings align with a study conducted in the municipality of monteria, as noted by chávez and niño (2018). a potential remedy for restoring citizens' trust and fostering democracy and participation could involve the infusion of fresh paradigms into public administration via participatory budgeting within the municipality of manizales. such an approach could facilitate the planning of city projects that address prevailing challenges and enable the inclusion of communities in all matters pertaining to public and political affairs. alternatively, and from a prospective standpoint, hong and cho (2018) suggest that the implementation of participatory budgeting can contribute to improving equity in resource allocation, albeit it may be less effective than bureaucratic decision-making in achieving public objectives. the authors propose the 'social pressure hypothesis' as an explanation, contending that the influence of social pressure during discussions in public forums can lead participating citizens to make decisions geared towards resource redistribution. the manizales general comptroller's office has instituted strategies aimed at facilitating the participation of all citizens, as evidenced by the roster of activities carried out in 2020 and 2021. nonetheless, there appears to be a sense of apathy among citizens towards engaging with the initiatives emanating from the oversight body. the lack of commitment on the part of citizens to actively drive change in their surroundings represents one of the primary hurdles to effective participatory fiscal control. often, citizens demand changes and outcomes from territorial entities without exerting any concerted effort themselves. to cultivate a more robust and direct involvement of communities in matters of mutual interest, the proposition of a public administration tool tailored to the context of manizales is warranted. through such a mechanism, heightened participation of city residents in the planning and execution of city projects with tangible societal benefits could be achieved. in conclusion, our research underscores the significance of carefully considering who participates, in what context, and for whose benefit. these considerations become even more crucial when considering how changes in preferences, customs, and traditions can impact citizens' willingness to engage, whether directly or indirectly. furthermore, it is imperative to acknowledge that the " let's take control" strategy goes beyond traditional methods of participation. the digital age, particularly social media networks, has proven to be an effective tool in fostering interaction between public administration and citizens. this approach aligns with the predictions of coleman and blumler (2009), who foresaw the internet's potential to enhance public communication and enrich democracy. achieving such a vision, including exploring citizen participation in fiscal control the implementation of the let's take control" sstrategy, requires imaginative and proactive policy formulation. dvir, liu and vedlitz (2023) highlights the significant role of citizen participation in government policies, whose effects vary depending on three distinct modes of citizen participation: policy endorsement, co-production, and co-investment. in the case of the let's take control" strategy, the mode aligns with policy endorsement. however, this aspect is a deeper field of analysis that deserves closer attention in future research. ultimately, our findings highlight the need for informed and active participation in the public sphere, enriching democracy and shaping a more engaged and participatory future. 5. conclusions current research in the field of citizen fiscal participation primarily focuses on the factors that explain citizens' involvement in budgetary processes. this study primarily emphasizes the first perspective. this study was conducted to verify the hypothesis of whether there is a relationship between the district where a citizen resides and their participation in the civic events of the municipality of manizales' office of the comptroller general. the results obtained through the chi-square test indicate that the null hypothesis is rejected, which means that there is a relationship between both variables. it is concluded that the district where a citizen resides conditions their participation in the activities of the comptroller's office in manizales. the findings suggest that participating a certain number of times may lead to a decrease in interest in participating. furthermore, contrary to intuition, better socioeconomic conditions and greater awareness of citizen participation mechanisms negatively correlate with participation frequency. this highlights intriguing cultural aspects that warrant deeper investigation. additionally, the degree of association between the level of education and citizen participation in the events of the municipality of manizales' office of the comptroller general was evaluated. the results indicate that the null hypothesis is rejected and the alternative hypothesis is accepted, demonstrating that the level of education influences citizen participation in the activities of the comptroller's office. given these findings, it is crucial to invest in awareness and sensitization regarding citizen participation strategies and initiatives. the aim is to promote the benefits and positive externalities that stem from them, such as increased civic empowerment, greater transparency in public administration, and consequently, a better understanding of the workings of the public sector. these aspects contribute to an enhancement of institutional quality. depending on the knowledge of public servants and the general public, the effectiveness of citizen participation tools in participatory fiscal control will vary. thus, it is necessary to delve deeper into the characteristics of each tool, including their scope, limitations, and the appropriate time to use them. citizen participation mechanisms are more than just a legal tool; they are society's best allies in facing inconsistencies and irregularities in the proper use of public resources, which are ultimately executed by different public entities, as well as some private sector entities that manage state resources. there are several factors that influence the lack of citizen participation, including leadership and citizens' knowledge, which enable them to exercise social control over participatory fiscal 38 cardona-arenas and campuzano control. one of the factors that prevent citizens from developing social control is corruption, which is associated with the poor functioning of planning and control systems. community participation and control lead to efficient results in the audit processes developed by the municipality of manizales' office of the comptroller general, activated and executed by community expressions of dissatisfaction with the management and use of public resources, which must ultimately seek common benefits that positively impact the entire municipality. additionally, these resources must be executed efficiently and timely. the results obtained from the application of the instrument indicate that the citizen participation office of the comptroller's office in manizales has shown commendable management and effectiveness in its activities during the 2020-2021 period. the office demonstrated its commitment by promptly receiving and addressing complaints within the legal timeframes established by law 1755 of 2015. furthermore, they delivered audit reports that provided responses to the requests made by the city of manizales. these achievements highlight the importance of strengthening the regulatory framework governing participatory fiscal control. it is crucial for the government to prioritize and enhance this tool to ensure that each of the mission objectives is accomplished in collaboration with the administrations of territorial entities, always working hand in hand with the community. by reinforcing the regulatory framework and emphasizing participatory fiscal control, the government can foster transparency, accountability, and citizen participation. this approach will contribute to more effective public administration and ultimately lead to better outcomes for the city of manizales and its residents. references aguirre, f.j. (2021). participación ciudadana en la gestión municipal para crear entornos incluyentes: caso del presupuesto participativo en perú, villa el salvador 2000-2002 y años recientes. revista erua, 13(26), 12. https://rua.uv.mx/index.php/rua/article/viewfile/136/111 arnstein, s.r. (1969). a ladder of citizen participation. journal of the american institute of planners, 35(4), 216-224. https://doi.org/10.1080/01944366908977225 baiocchi, g., & ganuza, e. (2014). participatory budgeting as if emancipation mattered. politics and society, 42(1), 81-110. https://doi.org/10.1177/0032329213512978 bovens, m., goodin, r.e., & schillemans, t. (2014). the oxford handbook public accountability. oxford handbooks. https://repository.essex.ac.uk/11939/ calabrese, t., williams, d., & gupta, a. (2020). does participatory budgeting alter public spending? evidence from new york city. administration and society, 52(9), 1382-1409. https://doi.org/10.1177/0095399720912548 cárdenas corredor, l.a., saleh alba, r.a., & sierra cuervo, e.a. (2013). auditorias articuladas, una herramienta de control fiscal participativo. retrieved from: https://hdl.handle.net/10901/6982 cardona-arenas, c.d., osorio-barreto, d., & sotelo-zemanate, d.y. (2020). desempeño de las subregiones del departamento del cauca y su contribución al desarrollo territorial (2015-2017). revista facultad de ciencias económicas: investigación y reflexión, 28(2), 161-179. https://doi.org/10.18359/rfce.4645 castells, m. (2013). comunicación y poder. siglo xxi editores méxico. chavez ricardo, l.g., & niño rodríguez, l.f. (2018). control fiscal participativo en el municipio de montería. retrieved from https://repository.unad.edu.co/handle/10596/21018 https://rua.uv.mx/index.php/rua/article/viewfile/136/111 https://doi.org/10.1080/01944366908977225 https://doi.org/10.1177/0032329213512978 https://repository.essex.ac.uk/11939/ https://doi.org/10.1177/0095399720912548 https://hdl.handle.net/10901/6982 https://doi.org/10.18359/rfce.4645 https://repository.unad.edu.co/handle/10596/21018 exploring citizen participation in fiscal control coleman, s., & blumler, j.g. (2009). the internet and democratic citizenship: theory, practice, and policy. cambridge university press. https://cir.nii.ac.jp/crid/1361137045896774912 coleman, s., & cardoso sampaio, r. (2017). sustaining a democratic innovation: a study of three eparticipatory budgets in belo horizonte. information, communication and society, 20(5), 754769. https://doi.org/10.1080/1369118x.2016.1203971 constitución política de colombia. (2023). artículo 2. retrieved from http://www.constitucioncolombia.com/titulo-1/capitulo-0/articulo-2 contraloría general de la republica. (2023). participación ciudadana. estrategias en el control fiscal participativo. crosby, n., kelly, j.m., & schaefer, p. (1986). citizens panels: a new approach to citizen participation. public administration review, 170-178. https://doi.org/10.2307/976169 dagnino, j. (2014). análisis de varianza. revista chilena de anestesia, 43(43), 306-310. díaz, h.o., rebellón, s.c., cifuentes, j.c.a., solano, j.a., mora, j.j., correa, g.a.s., ... (2003). propuesta de un modelo de control fiscal para el estado colombiano: el sistema de control fiscal nacional. foro «hacia un control fiscal orientado a resultados, 7, 13. http://www.scielo.org.co/pdf/eg/v19n89/v19n89a01.pdf draude, a., börzel, t.a., & risse, t. (2018). the oxford handbook of governance and limited statehood (online ed.). oxford academic. retrieved april 5, 2018. https://doi.org/10.1093/oxfordhb/9780198797203.001.0001 dunn, p.k., & smyth, g.k. (2018). generalized linear models with examples in r (vol. 53). new york: springer. dvir, r., liu, x., & vedlitz, a. (2023). exploring public participation modes in government: the case of infrastructure policies. public management review, 1-22. eberhardt, m.l. (2015). democracias representativas en crisis. democracia participativa y mecanismos de participación ciudadana como opción. araucaria. revista iberoamericana de filosofía, política y humanidades, 17(33). https://doi.org/10.12795/araucaria.2015.i33.04 faúndez, r., arancibia, s., & delamaza, g. (2022). criterios determinantes de la institucionalización de la participación ciudadana en organismos públicos: una propuesta de modelo. gestión y política pública, 31(2), 35-61. https://doi.org/10.60583/gypp.v31i2.1256 galais, c., navarro, c.j., & fontcuberta, p. (2013). la calidad de los procesos participativos locales: indicadores y factores explicativos contextuales. el caso de andalucía. revista española de ciencia política, (32), 65-87. garcía mejía, m.y. (2015). el control fiscal participativo: una estrategia para la eficaz administración territorial en colombia. retrieved from http://hdl.handle.net/10654/6407 gaventa, j. (2006). finding the spaces for change: a power analysis. ids bulletin, 37(6), 23-33. https://doi.org/10.1111/j.1759-5436.2006.tb00320.x gómez garcía, d.i., & carrero sáenz, o.r. (2009). revisión del concepto sobre control fiscal participativo en la contraloría general de la república.retrieved from https://ciencia.lasalle.edu.co/contaduria_publica/497/ gómez, e.c., & pinzón maldonado, h. (2012). responsabilidad fiscal en colombia. prolegómenos. derechos y valores, 15(29), 173-188. http://www.redalyc.org/articulo.oa?id=87625419011 gonçalves, s. (2014). the effects of participatory budgeting on municipal expenditures and infant mortality in brazil. world development, 53, 94–110. https://doi.org/10.1016/j.worlddev.2013.01.009 hengstenberg, p. (1999). sociedad civil en américa latina: representación de intereses y gobernabilidad. editorial nueva sociedad. hong, s. (2015). citizen participation in budgeting: a trade-off between knowledge and inclusiveness? public administration review, 75(4), 572-582. https://doi.org/10.1111/puar.12377 hong, s., and cho, b. s. (2018). citizen participation and the redistribution of public goods. public administration, 96(3), 481-496. https://doi.org/10.1111/padm.12521 lerner, j. (2017). conclusion: time for participatory budgeting to grow up. new political science, 39(1), 156-160. https://doi.org/10.1080/07393148.2017.1278860 liendo, c., and de las nieves, y. (2021). formación ciudadana para la participación directa en el control fiscal. luz 20(4), 133-141. manes-rossi, f., brusca, i., orelli, r.l., lorson, p.c., & haustein, e. (2023). features and drivers of citizen participation: insights from participatory budgeting in three european cities. public management review, 25(2), 201-223. https://doi.org/10.1080/14719037.2021.1963821 montambeault, f. (2016). participatory citizenship in the making? the multiple citizenship trajectories of participatory budgeting participants in brazil. journal of civil society, 12(3), 282-298. https://doi.org/10.1080/17448689.2016.1213508 montero, r. (2021). la fiscalidad participativa y la economía social y solidaria. revista de derecho fiscal, 19, 89-108. universidad externado de colombia. https://doi.org/10.18601/16926722.n19.04 https://cir.nii.ac.jp/crid/1361137045896774912 https://doi.org/10.1080/1369118x.2016.1203971 http://www.constitucioncolombia.com/titulo-1/capitulo-0/articulo-2 https://doi.org/10.2307/976169 https://doi.org/10.1093/oxfordhb/9780198797203.001.0001 https://doi.org/10.12795/araucaria.2015.i33.04 http://hdl.handle.net/10654/6407 https://doi.org/10.1111/j.1759-5436.2006.tb00320.x https://ciencia.lasalle.edu.co/contaduria_publica/497/ http://www.redalyc.org/articulo.oa?id=87625419011 https://doi.org/10.1016/j.worlddev.2013.01.009 https://doi.org/10.1111/puar.12377 https://doi.org/10.1111/padm.12521 https://doi.org/10.1080/07393148.2017.1278860 https://doi.org/10.1080/14719037.2021.1963821 https://doi.org/10.1080/17448689.2016.1213508 https://doi.org/10.18601/16926722.n19.04 40 cardona-arenas and campuzano nebot, c.p. (2001). gobiernos locales: participación ciudadana en el proceso presupuestario. gestión y análisis de políticas públicas, 161-172. https://doi.org/10.24965/gapp.vi22.286 ochoa díaz, h., & charris rebellón, s. (2003). propuesta de un modelo de control fiscal para el estado colombiano: el sistema de control fiscal nacional. estudios gerenciales, 19(89), 13-42. https://www.icesi.edu.co/revistas/index.php/estudios_gerenciales/article/view/121 pimienta, j.l.r., gulfo, y.e.c., & suarez, a.d. (2021). participación ciudadana en el sistema de seguridad social en salud en colombia. revista de ciencias sociales (ve), 27(1), 230-240. https://www.redalyc.org/articulo.oa?id=28065533018 ramos, j.m., & reyes, m. (2015). gobiernos locales y participación ciudadana: hacia un enfoque de gestión estratégica asociada. espiral estudios sobre estado y sociedad, 12(34). recuperado a partir de http://espiral.cucsh.udg.mx/index.php/eees/article/view/1601 ríos, a.m., benito, b., & bastida, f. (2017). factors explaining public participation in the central government budget process. australian journal of public administration, 76(1), 48-64. https://doi.org/10.1111/1467-8500.12197 rowe, g., & frewer, l.j. (2004). evaluating public-participation exercises: a research agenda. science, technology, and human values, 29(4), 512-556. https://doi.org/10.1177/0162243903259197 sánchez-torres, c.a., lópez obregón, c., peña gonzález, e., & aldana arévalo, d. (2006). control fiscal territorial diagnóstico y perspectivas. estudios socio jurídicos, 8 , 44 http://www.redalyc.org/articulo.oa?id=73380204 schneider, s. h., & busse, s. (2019). participatory budgeting in germany–a review of empirical findings. international journal of public administration, 42(3), 259-273. https://doi.org/10.1080/01900692.2018.1426601 sen, a. (2000). desarrollo y libertad. barcelona: planeta. sintomer, y., herzberg, c., röcke, a., & allegretti, g. (2012). transnational models of citizen participation: the case of participatory budgeting. journal of public deliberation, 8(2), article-9. https://hdl.handle.net/10316/41034 theocharis, y., & van deth, j.w. (2018). the continuous expansion of citizen participation: a new taxonomy. european political science review, 10(1), 139-163. https://doi.org/10.1017/s1755773916000230 torres, r.m. (2001). participación ciudadana y educación. una mirada amplia y, 20, 24-25 segunda reunión de ministros de educación del consejo interamericano para el desarrollo integral -cidi (punta del este, uruguay, 24-25 septiembre, 2001) https://doi.org/10.24965/gapp.vi22.286 https://www.icesi.edu.co/revistas/index.php/estudios_gerenciales/article/view/121 https://www.redalyc.org/articulo.oa?id=28065533018 https://doi.org/10.1111/1467-8500.12197 http://www.redalyc.org/articulo.oa?id=73380204 https://doi.org/10.1080/01900692.2018.1426601 https://hdl.handle.net/10316/41034 https://doi.org/10.1017/s1755773916000230 exploring citizen participation in fiscal control appendix 1. primary information collection instrument title: unraveling the factors driving citizen participation in fiscal control: lessons from the city of manizales – colombia this survey is part of a research project with the aim of understanding and identifying the conceptions that citizens of manizales have regarding participatory fiscal control. the information you provide will be confidential, and only authorized individuals will have the right to handle it. 1. survey age range: _ a) under 18 years old. b) between 18 and 24 years old. c) between 25 and 34 years. d) between 35 and 44 years. e) between 45 and 54 years. f) more than 54. education level: a) basic primary. b) basic secondary. c) technique. d) technological. e) undergraduate. f) graduate. g) none. commune where you reside: a) atardeceres. b) ciudadela del norte c) cumanday. d) ecotourstico cerro de oro. e) la estación. f) la fuente g) la macarena. h) palo grande. i) san josé. j) tesorito k) universitaria 42 cardona-arenas and campuzano socioeconomic stratum: a) stratum 1. b) stratum 2. c) stratum 3. d) stratum 4. e) stratum 5. f) stratum 6. mark with an (x) as appropriate: sex: female _____ male ________ other ________ have you been summoned to the events held at the comptroller general of the municipality of manizales? have you participated in the citizen events of the comptroller general of the municipality of manizales? how many times have you participated in citizen participation activities convened by the comptroller general of the municipality of manizales? number of times you have participated: ____________________________________________ how often do you participate in citizen participation activities convened by the general comptroller of the manizales municipality? a) 1 time per week. b) 2 times a month. c) 1 time a month. d) every 2 months. e) 1 time every semester. f) 1 time a year. are you familiar with the "let's take control" strategy of the comptroller general of the municipality of manizales? yes ___ no___ have you participated in the "let's take control" strategy of the comptroller general of the municipality of manizales? yes ___ no___ exploring citizen participation in fiscal control by what means have you participated in the "let's take control" strategy of the comptroller general of the municipality of manizales? a) facebook live. b) youtube c) face-to-face do you know what is participatory fiscal control is? yes ___ no___ do you know the mechanisms for citizen participation? yes ___ no___ which is it? a) vote. b) plebiscite. c) referendum d) referendum. e) town meeting. f) legislative initiative g) revocation of the mandate. do you know what are the control or regulatory bodies are? yes ___ no___ which is it? a) comptroller general of the republic. b) territorial controllers. c) office of the attorney general of the nation. d) ombudsman. e) district and municipal ombudsman. f) general audit office of the republic. do you know, what are citizen oversights? yes ___ no___ have you participated in any citizen oversight? yes ___ no___ which of the following reasons do you consider to be the main one for not reporting acts of corruption? a) for fear of reprisals against him. b) because they are indifferent to the situation. c) because they receive benefits from such behaviors. d) because they don't believe in control entities. 19. which of the following reasons motivate the denunciations of acts of corruption? 44 cardona-arenas and campuzano a) economic benefits. b) recognition before society. c) citizen duty. we respectfully request that you answer the following question by marking an x, based on the rating scale: 20. what is your perception of the work carried out by the comptroller general of the municipality of manizales? 1: poor 2: regular 3: acceptable 4: good 5: excellent descriptive statistics figure a.1. sex figure a.2. question 6. ¿ have you been summoned to the events held at the general comptroller's office of the municipality of manizales? 46% 54% 40% 45% 50% 55% female male 42% 58% 0% 20% 40% 60% 80% yes no exploring citizen participation in fiscal control figure a.3. question 7¿have you participated in the citizens´ events of the general comptroller's office of the municipality of manizales? figure a.4 question 10. ¿do you know the strategy implemented by the general comptroller's office of the municipality of manizales? figure a.5: question 13. ¿do you know what participation in fiscal control? figure a.6: question 16. ¿do you know what citizen-veedurias are? 40% 60% 0% 20% 40% 60% 80% yes no 44% 56% 0% 20% 40% 60% yes no 56% 44% 0% 20% 40% 60% yes no 63% 38% 0% 20% 40% 60% 80% yes no 46 cardona-arenas and campuzano figure a.7. question 17. ¿have you participated in any citizen´s oversight? figure a.8. question 18. ¿which of the following reasons do you consider to be the main one for not reporting acts of corruption? 10% 90% 0% 20% 40% 60% 80% 100% yes no 40% 29% 10% 21% 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% for fear of reprisals in his against because they are indifferent inthe face of the situation because they receive benefits from said behaviors because they don´t believe in control entities © theauthor(s)2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license appendix 2. correlation matrix variable ag e ra ng e ed uc at io n_ le ve l co m m un e st ra tu m se x su m m on ed pa rt ic ip at ed ti m es p ar ti ci pa te d pa rt ic ip at io n_ fr eq ue nc y le t’s ta ke c on tr ol pa rt ic ip at io n so ci al m ed ia pa rt ic ip at io n kn ow le dg e ab ou t ci ti ze n pa rt ic ip at io n k no w le dg e ab ou t m ec ha ni sm k no w le dg e ab ou t ci ti ze n ov er si gh ts ti m es c it iz en ov er si gh ts d en un ci at io ns co rr up ti on m ot iv at io n co m pt ro lle r ge ne ra l f un ct io n pe rc ep ti on k no w le dg e ab ou t age range 1 education_level 0.260** 1 commune -0.267** -0.245* 1 stratum 0.160 0.12 -0.184 1 sex -0.078 -0.115 0.056 0.001 1 summoned -0.399*** -0.485*** 0.090 0.036 -0.099 1 participated -0.419*** -0.471*** 0.107 -0.068 -0.146 0.871*** 1 times participated 0.406*** 0.365*** -0.155 -0.261* 0.088 -0.684*** -0.751*** 1 participation_ frequency -0.315** -0.477*** 0.246* -0.016 0.010 0.518*** 0.545*** -0.516*** 1 let’s take control participation -0.379*** -0.474*** 0.129 0.119 -0.157 0.788*** 0.826*** -0.762*** 0.586*** 1 social_media_participation -0.136 -0.323** -0.161 0.211 -0.071 0.620*** 0.620*** -0.581*** 0.476*** 0.768*** 1 knowledge about citizen participation -0.200 -0.520*** -0.002 0.045 -0.272* 0.740*** 0.708*** -0.554*** 0.359** 0.646*** 0.606*** 1 knowledge about mechanism -0.036 -0.030 0.040 0.026 0.186 0.100 0.006 -0.004 0.127 -0.006 0.037 0.003 1 knowledge about citizen oversights -0.400*** -0.169 0.229 0.076 -0.306** 0.124 0.148 -0.262* 0.126 0.389*** 0.291** 0.140 -0.181 1 times citizen oversights 0.085 -0.283** -0.003 0.075 0.091 0.419** 0.438*** -0.382** 0.352** 0.335** 0.116 0.320** 0.194 -0.252* 1 denunciations corruption motivation -0.055 0.021 -0.320** -0.270* -0.152 -0.044 -0.120 0.230 -0.074 -0.172 -0.120 0.018 0.085 -0.197 -0.064 1 comptroller general function perception 0.249* 0.023 -0.236 0.287** 0.117 -0.298** -0.340*** 0.212 -0.128 -0.225 -0.136 -0.271* 0.043 0.076 -0.084 0.004 1 knowledge about let's take control" strategy -0.434*** -0.553*** 0.168 0.022 -0.221 -0.788*** 0.832*** -671*** 0.492*** 0.840*** 0.634*** 0.687*** 0.019 0.466*** 0.261* -0.083 -0.259* 1 https://creativecommons.org/licenses/by-nc/4.0/ 1. introduction 2. literature review 3. methods 4. results and discussion 5. conclusions references © theauthor(s)2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 1 (2024), pages 5-28 https://doi.org/10.17979/ejge.2024.13.1.9773 submitted: june 13, 2023 accepted: march 5, 2024 published: june 6, 2024 article public-private partnerships caught in a trust-trap and the potential role of information: the burden of divergent rationales moritz kappler,1,* rahel schomaker,2, 3 michael bauer 4 1 speyer university, germany 2 cuas villach, austria 3 german research institute for public administration, germany 4 european university institute, italy *correspondence: kappler@uni-speyer.de abstract. trust is known to be beneficial for cooperation, though remains underdeveloped in public-private partnerships (ppps), which are caught in a trust trap due to the cooperative regime’s internal heterogeneity. accordingly, ppp’s inherent trust relationship is disentangled, and the burden of divergent actor rationales is introduced. we hypothesize that potential ppps find themselves ex ante in a situation that requires certain repair mechanisms before the ‘usual’ venues of trust development can become effective. as such, exogenous information is theorized as quasi-substitute for relational experience that may support the development of a ‘leap of faith’. theoretical considerations are tested by a vignette experiment with representatives of both the public and the private sector. within and between-group analyses find strong support for the burden of divergent rationales and stable levels of (role-based) trust. however, the effect of the information treatment on trust levels is limited. keywords: organizational trust; public-private partnerships; experiment; trust repair; trust development jel classification: c91; d91 1. introduction although trust is seen as ‘vital for the maintenance of cooperation in society’ (zucker, 1986, p. 56), it remains constantly underdeveloped in public-private partnerships (ppp) (schomaker & bauer, 2020). as such, a lack of trust challenges the operation of ppps or even inhibits their creation. in general, it contributes to ppps’ underwhelming performance (eurodad, 2018). the consequences are wide-reaching, since it de facto deprives public service provision of benefitting from bundled strengths of the public and the private partners and, thereby, severely undermines major public sector reforms of the past two decades that made ppps a top priority. whereas the positive relationship between trust and cooperation is subject to a plethora of studies across the disciplines, it is low levels of trust and the subsequent problems that is flying under https://creativecommons.org/licenses/by-nc/4.0/ 6 kappler, schomaker and bauer the radar of ppp-related trust literature. this paper contributes, first, theoretically, by adding to a differentiated understanding of the enactment of trust and, second, by adding experimental evidence related to trust gaps. this paper’s major proposition put forward is similar to nooteboom’s (2002) notion of pathological mistrust that interaction necessary for developing trust does not take place because of trust levels being too low to reasonably engage in interaction. disaggregating ppp’s trust relationship, this paper argues that ppps find themselves in a trust trap constituted by severe actor heterogeneity and divergent rationales. as such, it is suggested to approach the trust relationship in ppps ex ante as one that shows similar characteristics to one that experienced trust violation, and to consider exogenous information as a potential quasi-substitute for (missing) relational experience necessary for trust creation. empirical evidence found strongly supports ppps’ burden of divergent rationales and the corresponding perception of trust violation ex ante. trust levels are found to be very stable, with exogenous information substituting for relational experience but with limited impact. this paper proceeds as follows. first, ppps’ constituting elements and the trust-related interplay between them are analyzed, allowing for grasping (negative) dynamics, tensions, and potential breaking points within. on this basis, the nexus of heterogeneous actors, trust development and information-sharing within ppps is discussed and hypotheses are derived. second, an experimental design puts the impact of exogenous information on and the stability of trust levels under scrutiny by conducting a vignette experiment with representatives of both the public and the private sector. primary data is then put under in-depth statistical scrutiny. fourth, empirical findings are discussed and embedded in the body of extant literature. 2. disentangling trust in ppp and the role of information to cooperate always means to rely on and to be at the mercy of someone who is not oneself, hence, someone whose actions cannot be fully controlled. it therefore inevitably comes along with a certain degree of ‘irreducible social vulnerability and uncertainty’ (möllering, 2006, p. 115). as such, cooperation necessarily requires actors – independent of the actor being an individual or an organisation (kroeger, 2012) to take a certain risk that the action of the other may be harmful (dietz, 2011). to overcome the state of a paralyzed and reclusive society, it is trust understood as the ‘psychological state comprising the intention to accept vulnerability based upon positive expectations of the intentions or behavior of another’ (rousseau et al., 1998, p. 393) that enables cooperation and secures its smooth proceeding (hardin, 2002; luhmann, 1988).1 generally, the benefit of trust for partnerships has been recognized widely and informed a plethora of empirical studies across the disciplines (barber, 1983; bidault & castello, 2009; fukuyama, 1995; kappler & schomaker, 2023; lascaux, 2005; lewis & weigert, 1985; luhmann, 1979; möllering, 2006; oecd, 2011; putnam, 1993, 2000; sztompka, 1999; zucker, 1986) and also in 1 indeed, levels of trust being too high are acknowledged as a potential ‘dark side’ of trust (anderson & jap, 2005; möllering & sydow, 2019; nooteboom, 2002). since the context of ppp struggles with very low levels of trust, the danger of high trust levels is not discussed in this paper. public-private partnerships caught in a trust-trap and the potential role of information the specific case of ppps (brogaard, 2019; edelenbos & klijn, 2007; klijn et al., 2010; schomaker & bauer, 2020; warsen et al., 2018). put simply, large parts of the extant literature on the cooperationtrust nexus – independent of the (economic or political) yardstick used consider trust as an explanatory variable and agree on a positive relationship between trust levels and both procedural and outcome indicators. approaching ppp and trust between the partners generically, edelenbos and klijn (2007) find trust being the single most important condition for a successful ppp. for example, edelenbos and klijn (2007), klijn et al. (2010), and schomaker and bauer (2020) scrutinize transaction costs occurring in ppps as a result of uncertainty and low levels of trust. they refer to trust as a substitute for control carving out the relationship between higher levels of trust, corresponding positive expectations about the trustor’s action, a lower demand for additional safeguards when investing resources in a joint endeavor and consequently lower transaction costs. amongst others, warsen et al. (2018) include procedural indicators and take good cooperation (e.g. measures on smooth resolution of conflicts and the ability to overcome deadlocks) in a process into account as a part of ppp performance sui generis. while there is consensus about the fruitful and dynamic relationship between trust and cooperation, the stagnant and even declining trajectory of ppps is hardly targeted by extant literature. this is a serious void since low levels in the extreme, the absence2 of trust can reasonably be seen as a severe obstacle for smooth cooperation. some few theoretical and empirical exceptions that work on ppp-inherent trust problems, e.g. schomaker and bauer (2020), demonstrate that the instrument of ppp, despite extensive promotion in the past two decades, is largely neglected by public decision-makers with about two-thirds not considering ppp a valid alternative for public provision. confirming trust in ppp as generally problematic, they compare ppp with public-only provision and find that ppps are prone to lower trust levels and higher transaction costs that severely hamper cooperation between public and private actors. along similar veins, even ex post ‘failures’ of ppps, for example, in the form of costand time over-runs or the service’s remunicipalization seem more often to be due to lack of trust than to factual performance problems (hall, 2012). large-scale empirical work, such as the special reports by the european court of auditors (2018) and eurodad (2018) confirm ppp inherent problems of delays, cost increases, underuse and general ineffectiveness. disaggregating trust in ppps in its basic terms, every trust relationship can be summarized as a trustor (x) who trusts a trustee (y) relating to a specific issue (z) (hardin, 2002; nooteboom, 2002; six & verhoest, 2016). whereas ppps elements are well known (a private partner, a public partner, a contract, and the objective of service provision), digging deeper into the interplay of such exposes several trust-related peculiarities. starting with x, both actors in charge (the private and the public partner), first, need to freely and actively opt for collaboration and, second, determine operational issues throughout the lifespan 2 we consciously do not refer to concepts of distrust or mistrust. 8 kappler, schomaker and bauer of a ppp (luhmann, 1988). thus, x can be either the private or the public partner, each of which deciding and acting independently. accordingly, in order to conduct ppps (i.e. the actual enactment of trust), both partners need to accept vulnerability and have positive expectations of the other’s intentions and, in general, the joint endeavor. substantial differences in the public and the private sector lead to different incentive systems, distinct decision-making, varying strategies and issues at stake; here subsumed as divergent rationales. affirmatively, empirical work on stakeholder opposition within ppps finds strong evidence for the gap between expectations on both the process and the outcome of the project between the distinct actors driving internal tensions and strong resistance (de schepper et al., 2014; el-gohary et al., 2006; zhang, 2005). as such, the incentive structures and basis motives of the public and private sectors vary significantly and may even be contradictory. basically, the former’s rationale is to improve public well-being, while the latter’s raison d’être is to maximize corporate profits. whereas the former is largely accountable to the public, the latter will report to, e.g., an executive board and shareholders. considering – in a slightly oversimplified manner the divergent claims between the public that aims for a ‘decent life’ and the shareholder that aims for higher returns on investment, the perceptions of ‘desirable behaviors or end states’ (edwards & cable, 2009, p. 654) and of general appropriateness differ significantly between the actors. as such, a ppp understood as a very heterogeneous cooperative regime with an inherent dissent in sense-making, it is most likely prone to not share neither a joint vision (lee et al., 2018), nor a common belief system, nor a reasonable degree of value congruence amongst each other (arnold et al., 2012; bouckaert, 2011; parmentier & vervaeke, 2011; van gestel & de poorter, 2019) all of which instrumentally secure collaborators from one another’s deviant and undesired behavior. on the contrary, using a sociological frame, approaching the two sectors as two independent socially constructed systems of norms, values, and beliefs, the definition of what is legitimate can vary significantly across the systems (goodin et al., 2011; kappler, 2022; scott, 2010). translating to trust in ppps, an actor within a joint endeavor who is deciding and acting differently than oneself would do, put the entire project at a rather unstable grounding. own objectives are in continuous peril and need to be defended rather than jointly advanced. accordingly, bridging to rousseau’s basic definition of trust, relying on someone else (or on a joint endeavor with the other) that has a divergent rationale, vulnerability is increased and positive expectations about both the other’s intentions and the joint endeavor’s appropriate conduct are at best difficult to develop. as such, in addition to expected generally low levels of trust, expectations and intentions of distinct collaborators most likely are not congruent and, therefore, trust levels can thoroughly vary within one trust relationship depending on the specific trustor (hardin, 2002; nooteboom, 2002). h1: trust levels into ppps vary between the two trustors. that especially holds up in the case of an asymmetric z – as is the case in ppps – meaning that the actual objective of the partnership belongs to the specific working field of only one of the actors. in fact, the major objective and actual reason for establishing ppps across the world is to deliver improved services – vis-à-vis public-only service provision and better value for money aiming at the collective wellbeing and the ‘general interest’ (ppp knowledge lab, 2021; australian public-private partnerships caught in a trust-trap and the potential role of information government, 2016). along these lines, ppiaf (2012) justifies the existence of ppps exclusively as a means to potentially better ‘achieve the government’s objectives’ (p. 1) than public-only service provision. considering ppps basic alignment with the public sector, one may be tempted to see disadvantages for the private partner due to the alleged mismatch of the actual objective. however, the opposite may be the case. the decision of a private partner to join a ppp is based on relative clear cost-benefit calculations, informed by future gains, risks, and potential for economies of scale and scope (boivaird, 2004). oversimplified, if the expected return on investment of a ppp is lower than could be realized elsewhere, the private partner will most likely refrain from cooperating with a public partner. regarding the public partner, the decision on ppps is more complex. each ppp puts the public actor into a tension field characterized by the urge to ‘harness the market’ and the inherent loss of political control. farazmand (1999, p. 152) puts it as ‘with sweeping privatization of public enterprises and other major governmental functions, the capacity and ability of governments in public management are seriously diminished’. as such, to safeguard public accountability and democratic control and, thus, to cope with the actual public mandate3, an intrinsic preference is to keep specific tasks and responsibilities within the public sector rather than to privatize. furthermore, a failure in public service provision, would cause very negative societal effects so that the public actor has ‘no right to be wrong’ (rittel & webber, 1973, p. 160). this raises the stakes and making it more difficult to rely on someone else. accordingly, considering a risk-related perspective, the concept of divergent rationales is further distinguished leading to different trust levels hypothesized as follows: h2: the public partner trusts less in ppps than the private partner. alongside the inner heterogeneity due to divergent rationales, ppp’s organisation through a contractual structure makes the underlying contract a controversial trust-related issue. as such, contracts inform both the capability to compensate for divergent rationales and the perception of the y of the basic trust relationship. on the one hand, contracts provide a frame and a certain order relating to obligations, responsibility-, taskand, in particular, risk-sharing arrangements over a given period of time (mostly between one and thirty years) (schomaker & bauer, 2020). along the same lines, it can involve several safeguards for both the public and the private partner securing each other from the other’s actual rationale. as such, indeed, the underlying contractual arrangements4 attempt to moderate discrepancies. on the other hand, typical ppp problems are relatively wicked inasmuch as they are ‘complex, unpredictable, open-ended, or intractable’ (head & alford, 2015) and – to meet public accountability require a high degree of responsiveness to changing public opinion and, therefore, limit the scope of pre-set rules. accordingly, contracts do not rule out divergent rationales completely, keeping the ppp a tension field. however, the contractual architecture changes the appearance significantly and makes a ppp different to the sum of its two constituent partners. a ppp makes two independent entities part 3 the public mandate needs to be considered, since roughly half of citizens are concerned about private enterprises being involved in public services (james et al., 2016). 4 for an overview on varying ppp designs, e.g., grimsey & lewis (2005), teicher et al. (2006); bovaird (2004). 10 kappler, schomaker and bauer of a joint endeavor that forms an entity on its own (de schepper et al., 2014). accordingly, if one of the partners builds an expectation relating to the advancement of specific means and ends within a ppp, then assessing the entire ppp’s ‘intention’ – considering the interplay of the two partners and the contractual arrangement may be a more solid grounding than to exclusively assess the partner in an isolated manner. these considerations and the dynamics that urge out of contractual safeguards strongly inform the perception of the y. along these lines, this work partly dissents from the extant literature that perceives trust in ppps as an inter-organisational matter and, accordingly, the y as either the public or the private partner (edelenbos & klijn, 2007; klijn et al., 2010; warsen et al., 2018). oversimplified, hitherto, the likelihood to engage in ppps is explained by trust in the potential partner exclusively, excluding the specific contractual arrangements and interplay dynamics. however, this work proposes an analytical shift that, in fact, overlaps (potentially in large parts) with the literature on inter-organisational trust, though it departs from it by considering the complexity of ppps as an, indeed, very heterogeneous and dynamic entity in itself. as such, including the self, and allowing for internal dynamics within the ppp, this work proposes to understand the trust relationship within ppps as supra-organisational matter, as depicted in figure 1. the y, accordingly, is understood as the supra-organisational composite of various organisations with an independent ‘legislation’ that is valid for all members and that requires the single partners to convey certain agency and sovereignty to the ppp. for example, in extreme terms, even if one partner does not trust the other one, there may be good reason to trust the entity of the ppp if the contractual design is advantageous. figure 1. approaches on trust in ppps (own illustration). figure 2. ppps’ basic trust relationship (own illustration). public-private partnerships caught in a trust-trap and the potential role of information translating trust disaggregation discussed throughout this chapter into the basic trust relationship of ppps, figure 2 depicts the multifaceted nature of the trustor (x), the compound (supraorganisational) characteristic of the trustee (y), and ppp’s overarching and potentially conflicting objective(s) (z) all of which find together in a context characterized by divergent rationales. trust development in ppps as prior discussion elaborates, trust allows and supports cooperation. it is ppp’s specific composition and the inherent diverging rationales, though, that depicts a rather infertile ground for trust to emerge and develop. accordingly, the question of how trust develops generally and in a ppp in particular is of further interest. along these lines, this work borrows from trust-building and trust-repair literature that provides several revealing starting points. recalling rousseau et al. (1998), trust is developed when positive expectations are met by the counterpart’s action. accordingly, large parts of the corresponding literature (dietz, 2011; korsgaard, 2018; lewicki & bunker, 1996; schoorman et al., 2007; six & skinner, 2010) approach trust development through a positive feedback loop, respectively through a virtuous cycle of increasing trust and cooperation resulting in a positive experience that again positively shapes future expectations. whereas one can reasonably assume that higher levels of trust are reached – if at all – slower in a heterogeneous partnership such as ppps, it may be the starting point of this trust-creating process that explains the largest parts of the stagnant trust relationship at hand. in general, all cooperation necessarily starts at some point in time when trust is not yet developed. to begin cooperation despite absence of trust, amongst others, simpson (2012, p. 551) argues that to a certain degree trust is taken-for-granted with a general propensity to trust since ‘in living socially, people must rely on others to act cooperatively’. along the same line, stern (2016, p. 29) emphasizes the unconditional existence of trust as ‘trust is not of our own making but is something given with the nature…’ and, therewith, confirms the notion of trust as a moral duty in social life (hollis, 1998). accordingly, in order to overcome the initial obstacle of lacking experience and knowledge, möllering (2006) states that in all cooperative action there is always a ‘leap of faith’ involved. relating to ppps, however, one may argue that the public and the private partner do not start their relationship within ppps at a usual starting ground in which a leap of faith helps to start the trust-developing process. this work suggests that, despite not having interacted before, built upon the well-known divergent rationales discussed above, the public and the private partner have a certain ex ante (negative) role-based trust level that puts trust levels well below ones in the case of not knowing each other (kramer, 2010). in line with kramer (1999), such a form of role-based trust is not built upon experience but is depersonalized and informed by the expectation that the potential partner plays a particular (here partly contradictory) role informed by, for example, sectorappropriate belief systems and notions of appropriateness. as such, ppp’s initial state of trust contains certain characteristics of a violated trust relationship, even though none of the partners has actively transgressed, respectively disappointed the other, nor has the ppp itself shown any problem. accordingly, even without prior trust failure, divergent rationales lead to a trust gap resulting from the expected sector-appropriate ‘action of 12 kappler, schomaker and bauer organisational agents that threaten the legitimacy of the ppp and that has the potential to harm the well-being of one or more of the organisation’s stakeholders’ (gillespie & dietz 2009, p. 128). this perceived violation ex ante lowers expectation to a level that makes partners refrain from cooperating with the unequal partner in ppps. this widened gap between the partners can be understood as the burden of divergent rationales. it hampers – if not prohibits trust development in ppp to start right at the outset, inhibiting the enactment of trust in terms of the initialization and implementation of ppps. to the worse, as discussed above, the (public) partner who needs to initiate ppps and, thus, who needs to overcome the initial obstacle is the one who is supposed to have even lower trust levels in ppp. along these lines, the rule of ppp’s public initiation5 reinforces trust-related problems in ppp. ppps ‘not even being considered as an alternative’ (schomaker & bauer 2020, p. 4) by the vast majority of public officials, remarkably indicate that divergent rationales can deepen a trust gap until it turns into a de facto trust trap. in such a trap, similar to nooteboom’s notion of pathological mistrust (2002), necessary interaction for developing trust does not take place because trust levels are too low to engage in interaction, or as nooteboom (2002, p. 207) puts it, the absence of trust ‘will keep one from entering into relations, which robs one of the opportunity of favorable experience’. along these lines, to allow for initial trust development in ppps, trust needs to be established by improving expectations up to a level that enables the partners to make themselves vulnerable and to grant the necessary ‘leap of faith’. organisational trust repair literature offers certain potential venues to improve expectations such as transparency measures and intensified regulation that, however, face severe limitations in the case at hand (bachmann et al., 2015; denize & young, 2007). recalling the virtuous cycles introduced above leads to (ever) increasing positive experiences culminating in an accumulation of knowledge into the other actor ‘so that the other behavior is anticipatable’ (lewicki & bunker 1996, p. 121). conforming with, amongst others, van de walle (2017) and rousseau et al. (1998, p. 399), the basic elements within each round of interaction building knowledge, hence, is the ‘information available to the trustor from within the relationship’. however, since interaction (here cooperation in a joint endeavor) is not taking place in the discussed trust trap, it may be (positive) exogenously driven information – from outside of the actual relationship about the other and the ppp that could increase initial trust levels. accordingly, this work suggests to compensate for a lack of relational information by considering exogenous information as external quasi-experience, substituting for organic experience. along these lines, focusing on external information as a driver of trust, this work is well in line with, for example, kaplan and haenlein (2009), who investigate information provision as a means of public marketing, and, amongst others, van der meer (2017), kumlin and haugsgjerd (2017), radin (2006), van de walle (2017), and van ryzin (2011) that work on political trust and the role of open data in general and, in particular, publicly available performance indicators. along these lines, exogenous information comes with the advantage that it can provide quasi-experience about the entire ppp (and not merely about the potential partner). basically, exogenous information is assumed to inform expectations relating to the ppp and, thus, also the willingness to accept vulnerability within such. accordingly, this work hypothesizes that: 5 privately initiated projects / unsolicited proposals are a very rare exception and describe the situation in which the private actor reaches out to the government with a project proposal (ppiaf, 2012). public-private partnerships caught in a trust-trap and the potential role of information h3: exogenous information impacts trust levels. analogous to the quality of a personal experience, the quality of the substitute can be expected to have a similar effect. in very brief, that means that positive information substitutes for positive personal experience, both leading to improved expectations, i.e. an increased trust level; and vice versa. accordingly, the information’s quality (in terms of indicating either positive or negative content about ppps) is hypothesized to impact trust as follows: h4: positive / negative information increases / decreases trust levels. taking h4 one step further, allowing effect sizes to differ between negative and positive information, one can – making information a sensitive issue assume negative information to cause a larger delta in trust levels (negative) than positive information (here a positive delta) does. as such, thorbecke (as cited in six, 2005, p. 5) gets to the heart of potential change in trust levels by speaking metaphorically stating that ‘trust comes on foot, but leaves on horseback’, therewith emphasizing the gradual and general time-consuming development of trust in small steps and the consequences of violation as occurring in a rather ‘catastrophic manner’. according to, amongst others (ross & lacroix, 1996), decisive for such different effects may be the more vivid perception of violation visà-vis the mostly taken-for-granted experience of good cooperation. following, this work hypothesizes that: h5: negative information weighs more heavily than positive one does. one step further, better understanding effect sizes, beyond the specific kind of intervention, it is the trust levels themselves that need to be considered well. as such, trust levels can be supposed to vary in stability over time and, thus, over interventions. put it simply, information will be of larger impact when existing trust levels are rather loose; respectively, the impact will be smaller when trust levels are firm. following this line of thought, trust levels’ inherent stability and the effect size of information are assumed to correlate negatively. as such, it is of major importance to understand, generally, how and, particularly, on which pillars existing trust in ppps is built upon, since specific paths of trust development (and violation) may lead to either firm or loose trust levels. borrowing from extant literature, one may subsume the corresponding discussion as the question of if (perceived) trust violation took place either ‘strategically’ motivated by integrity issues or ‘accidentally’ caused by a lack of competence or even mere misfortune. in very brief, on the one hand, one may be positive about trust repair (i.e. expecting low stability measures), since – in this case perceived trust violation is most likely attributed to the divergent rationales in the sectors of origin and not to the specific partner’s deliberation and strategic deception within ppps. accordingly, as grover et al. (2014) indicate, such unintentional violations can be recovered more easily than intentional ones. on the other hand, as discussed above, it may be exactly these divergent rationales indicating different perceptions of appropriateness and sound principles regarding legitimate means and ends, leading perceived violation to be interpreted as ‘integrity-based’. in comparison to ‘competence-based’ violations that can be a sheer mistake, integrity-based ones are more difficult to 14 kappler, schomaker and bauer repair due to general doubts about the moral values of the other, which will most likely keep informing future behavior within ppps (grover et al., 2014). along the same lines, differentiating causes for violation based on their expected permanence, in line with tomlinson and mayer (2009), integrity-based violation, when compared with competence-based one, is supposed to remain constant and, thus, exposes a constant threat. hence, building upon kramer (1999), such integritybased trust levels that anticipate the expected (undesired) role of the other in a ppp in the context of divergent rationales, one may assume the (low) expectation into ppp of both the public and private partners to be relatively firm, making them difficult to change. as such, it is hypothesized that: h6: initial (pre-test) trust level matters. ppp’s disaggregated trust relationship caught in a trust trap and the potential role of exogenous information are put under empirical scrutiny in the subsequent chapter. the theoretically derived hypotheses are put to empirical contestation using an experimental design and quantitative analyses. 3. results in order to test the hypotheses raised, an experimental vignette setting is applied that is based on an (external) information-providing intervention. this methodology comes with several crucial advantages, in particular it allows to identify causal mechanisms based on systematic and balanced treatment variation and seems to be an appropriate and innovative research method for the purpose at hand (anderson & edwards, 2014; bouwman & grimmelikhuijsen, 2016; james et al., 2016; margetts, 2011; walker et al., 2017). experimental setting this study applies a campus-based classroom experiment at two german universities in which the classroom serves as a controlled ‘laboratory” and allows for high levels of control. the study group is composed of management students from a private business school as well as students of ‘public administration’ at a school of administrative sciences; the overall n equals 80 subjects.6 as such, the group composition involves (future) decision-makers from the public and the private sector and, therefore, represents the target population fairly well. all students stem from higher semesters and come up with a fair amount of study and working experience and can, therefore, be reasonably assumed to be specifically socialized in their corresponding sectors. since this experiment aims at decision-makers in both the private and the public sectors rather than at citizens, this experimental setting does not face the challenges of low external validity when drawing inferences from students 6 the power analysis used to determine the sample size or number of observations required to detect an effect of a reasonable size, provided us with a value 0,78 and thus is close enough to the 0,8 value that is usually taken as a benchmark for such experiments (suresh & chandrashekara, 2012). public-private partnerships caught in a trust-trap and the potential role of information to citizens. on the contrary, the homogeneous groups of public administration and business students proxy well the relatively narrow elite group of public and private decision-makers and allow sound comparison between experimental groups (morton & williams, 2010). generally, the applied experiment combines a withinand a between-group logic and, therefore, allows to capture changes in trust levels based on, e.g. sectoral belonging and over ‘time’ (before and after the intervention). along these lines, to capture the enactment of trust i) the respective trust levels are queried before and after the intervention and ii) the independent variable (information) is manipulated to scrutinize the potential effect on the dependent variable, (trust levels). the type of frame this work applies is an ‘equivalency’ or ‘valence’ frame, as it casts true information about ppps, just in either a positive or negative light (druckman, 2011). these logically equivalent statements concerning ppps are expected to make the participants to choose different options (rabin, 1998, p. 36). the texts (see appendix) for the intervention were written by the study authors specifically for that purpose and follow the logic of ppp’s supra-organisational character. the first text for group a (‘neutral intervention’) serves as control and merely explains why researchers use questionnaires and surveys and is used merely as a ‘distractor’ to bridge the time gap for the control group, thereby avoiding a time bias, ensuring that the time lag exposure of all groups is the same, approximately three minutes. the second text (group b, ‘negative intervention’) provides information emphasizing potential risks and challenges related to the new cooperation regimes between the public and private sector. the third text (group c, ‘positive intervention’) provides information about the positive effects and potential benefits related to the new cooperation arrangements. in order to improve validity of the experimental design and to secure a proper understanding of manipulations, the experiment was cognitively pre-tested with representatives of the targeted groups. accordingly, due to this relatively high degree of internal consistency, one may not expect significant disruptive factors. nonetheless, to correct for potential structural differences in individual trustor-related drivers of trust (arnold et al., 2012; parmentier & vervaeke, 2011) such as age and gender, the experimental setting and the subsequent quantitative analyses control thoroughly for such. the experiment was conducted by a professor of the respective faculty that is known to the students, but not the one that was currently teaching the class. this approach was chosen to control for the potential bias that may arise if an actively teaching professor conducts the experiment, but to ensure that the source of information overall is trusted by the participants. after providing basic information about the fact that a scientific experiment will be conducted, and the related question of whether the individual was willing to participate7, as well as a short neutral introduction of the fact that in many countries the public sector includes private agents in the service delivery, questionnaires were randomly assigned and double-blind distributed amongst the study group participants. 7 being fully aware of the experiment’s purpose, the handling of data, and its voluntary nature, all participants affirmed the desire to participate in advance of the experiment and, therewith, provided informed consent. 16 kappler, schomaker and bauer figure 3. experimental design (own illustration). as indicated in figure 3, the questionnaires contained a set of pre-test questions that were the same in all groups, one randomly distributed vignette text a, b or c, and a set of identical posttest questions for all groups. all questions (of which not all are used for this paper) and texts are fully shown in the appendix. relating to the measurement of trust levels, well in line with, e.g., dietz (2011), and dietz and den hartog (2006), this work emphasizes the processual character of trust relationships. accordingly, it considers ‘trust stages’ that begin with the assessment of the other’s trustworthiness, then leads to the decision to either trust or not, and results in the trust-informed action, i.e. the enactment of trust here to opt for a ppp. the trust items developed and used in this experiment assess trust in terms of the ‘transition category’, that is the decision to either use ppps or not, i.e. to accept or refuse vulnerability in the sense of enactment of trust. by querying trust levels into ppps in general – and not exclusively into the potential partner trust levels are queried according to the supra-organisational nature of ppps.8 quantitative analyses relating to differences between groups, both h1 and h2 as discussed above target the hypothesized basic trust problem in ppps that is the internal actor heterogeneity. it is hypothesized that divergent rationales translate into i) different trust levels based on sectoral belonging and ii) relatively lower levels of trust of the public partner, both culminating in a burden of divergent rationales. to empirically approach both differences in trust levels in general and, in particular, the specific kind of differences based on belonging to either the private or the public sector tests for group differences are run. since it is the empirical baseline (pre-experimental) condition of ppp’s trust relationship and the inherent divergent rationales that is of prime importance, pre-test trust levels are scrutinized using the nonparametric mann-whitney u test to test for differences of the ordinal dependent variable pre-test trust level between the private and the public sector. as assessed by visual inspection of population pyramids (indicating frequency of different trust levels isolated for both 8 the appendix provides the detailed questions and the texts used for framing. public-private partnerships caught in a trust-trap and the potential role of information public and private sector representatives), distributions of trust levels for public and private sector respondents were similar. accordingly, median trust levels could be applied for further scrutiny. in general, evidence was found supporting h1 and h2, since the median trust level score was statistically significantly lower (i.e. also different) for public sector respondents (3.5) than for private sector respondents (4.0), u=604, z=-2.032, p=.042.9 in addition, cautiously confirming sectoral belonging as partly explaining trust in ppps, scrutinizing group difference of post-test trust levels based on sectoral belonging, excluding different interventions for now, confirms both trust differences within ppps and relatively lower values for the public partner (with a median of 3.0 for public sector respondents and 3.5 for private ones, though becoming insignificant, u=609, z=-1.887, p=.059.10 summarized in figure 4, evidence found (stronger in the case of pre-test trust levels) supports the notion of ppp’s internal heterogeneity and the low(er) public trust levels. along these lines, both h1 and h2 can be accepted. to overcome the empirically shown burden of divergent rationales, the role of information as quasi-experience has been tested accordingly. in a first approach, a wilcoxon signed-rank test was applied to test for differences in trust levels before and after the intervention. as such, it tested the impact of the intervention in general. 80 sectoral representatives were recruited to understand potential changes in trust levels based on information treatment as measured with a 5-point liker scale. as test statistics indicate, information treatment elicited a statistically significant median increase in post-test trust levels compared to ex-ante trust levels, z=-2.67, p=.008. accordingly, not distinguishing the type of intervention so far, this analysis suggests that interventions matter to a certain degree, and, thus, exogenous information impacting trust levels leading to a preliminary acceptation of h3. however, since preand post-test trust items are, indeed, querying the same concept, but with non-identical wording, one needs to cautiously interpret findings and contest such by further analyses, as is done subsequently. figure 4. boxplots trust levels based on sectoral belongings. 9 same pattern can be observed when comparing mean ranks instead of median scores with public sector representatives (mean rank =35.28) scoring significantly lower than public sector representatives (mean rank=44.77). 10 again, comparison of mean ranks results in similar patterns with public sector representatives (mean rank=35.42) scoring significantly lower than public sector representatives (mean rank=44.66). 18 kappler, schomaker and bauer table 1. hierarchical multiple ordinal logistical regressions (glm). (i) (ii) (iii) variables exp(b) exp(b) exp(b) pre-test trust levels 11 13.091*** 14.016*** 13.319*** neg. intervention (reference is neutral intervention) 0.559 0.582 pos. intervention (reference is neutral intervention) 1.153 1.185 private sector belonging (reference is public sector) 1.527 age 1.025 1.015 1.037 gender (reference is female) 1.582 1.542 1.516 note. n = 80; dependent variable: post-test trust levels. significance level: *** 1%, ** 5%, * 10% to further elaborate on, amongst others, stability effects, intervention effects on (post-test) trust levels and expand analyses by the inclusion of other (control) variables and interaction effects, the following part of the statistical analyses conducts various multivariate ordinal logistic regression in a hierarchical fashion, as summarized in table 1. in line with h6, to test for stability of trust, model (i) was run to determine the effect of the pre-test trust levels that are assumed to be rather rolebased, on the post-trust levels. an increase in pre-test trust levels was associated with an increase in the odds of showing high post-test trust levels, with an odds ratio of 13.319 (95% ci, 5.710 to 30.013), χ2(1) = 36.916, p < .005. as such, trust levels in ppps can be assumed to be a very stable trait providing strong empirical support for h6 that can be well accepted. noteworthy, both control variables (age and gender) do not show significant differences in odds ratios in this estimation, and, anticipating, continue to be insignificant across all models tested.12 contesting findings relating to h3 made above and further expanding analyses to consider h4 (including different interventions), model (ii) allows to test for specific intervention effects on post-test trust levels. negative and positive interventions are introduced and compared with the control and reference group ‘neutral intervention’. as such, beyond the findings made through the wilcoxon signed-rank test above, this approach allows to fully benefit from the experimental setting and enables us to grasp causality between the specific kind of intervention and the corresponding change in the dependent variable. the odds of someone who received a positive intervention having 11 generally, the various robustness checks made throughout quantitative analyses are not included for the sake of better readability. noteworthy, however, all regressions have been likewise conducted considering the ordinal nature of the ex-ante trust levels, fully supporting findings made above. 12 while, particularly, the non-effect of gender may be slightly controversial with extant literature on, for example, citizen trust, it seems reasonable in this case surveying ‘elite’ groups indicating a high degree of homogeneity within groups, therewith, strengthening the theorized firm joint belief system within sectors. public-private partnerships caught in a trust-trap and the potential role of information higher trust levels was 1.153 (95% ci, 0.306 to 4.345) times that for someone who received a neutral intervention, χ2(1) = 0.044, p = .833, indicating both a very small and insignificant impact of positive information treatment on trust levels. showing a slightly stronger impact, though remaining insignificant, the odds of someone who received a negative intervention having higher trust levels was 0.559 (95% ci, 0.158 to 1.977) times that for someone who received a neutral intervention χ2(1) = 0.814, p = .367. as such, both interventions, when compared with the control group, show small impacts with the negative intervention slightly outweighing the positive one. accordingly, the relationship between the specific (positive / negative) intervention and the direction of change in trust levels as hypothesized in h4 finds cautious support from within the sample but cannot be accepted due to insignificance. the same holds true for h5, inasmuch as negative information has a slightly stronger negative impact than positive information, which has a positive impact within the sample but cannot be accepted in the context of this specific study due to insignificance. translated to practical meaning, different information treatments have – if at all a minor and, on average, slightly stronger impact for negative information. ex-ante trust levels keep on weighing in strongly inasmuch as an increase in ex-ante trust levels was associated with an increase in the odds of showing high ex-post trust levels, with an odds ratio of 14.016 (95% ci, 5.959 to 32.963), χ2(1) = 36.612, p < .005. accordingly, considering both pre-test trust levels and the different interventions, information treatments hardly contribute to explaining post-test trust levels, which are explained in large parts by the corresponding (assumed role-based) pre-test trust levels. model (iii) was run to determine and double-check the impact of sectoral belonging on the post-test trust levels. as table 1 indicates, considering multiple independent variables and largely affirming findings from above, the effect of sectoral belonging to trust levels remains stable relating to the positive sign when referring to the private sector, though it becomes insignificant. 4. discussion h1 and h2 find strong empirical support that suggests generally very low trust levels and that the public partner has substantively lower trust in ppps when compared with the private one. this makes the case for divergent rationales both identifiable and decisive. distinct incentive and risk structures and different perceptions of vulnerability and uncertainty seem to play an important role when assessing trust in ppps. that feeds back to theory building and practical handling of trust in ppps through various venues. it sheds light on the difficulty of trust development in heterogenous cooperative regimes and, indeed, makes ppps a least likely case for trust to emerge and to develop. next to the recognition of trust being problematic in the case of ppps (hall, 2012; eurodad, 2018; schomaker & bauer, 2020), this paper contributes an explanation for low trust levels that result in, generally, stagnation and (under-) performance of ppps. it makes ppps an organization in which own interests need to be defended against competing ones rather than to be jointly advanced. relating to the low(er) levels of trust of, particularly, the public partner with a public mandate and ‘no right to be wrong’ (rittel & webber, 1973, 160), it seems at best difficult to rely on someone within a joint endeavor who have contradictory motives and corporate objectives. in conjunction with ppp’s rule of public initiation, trust problems impede the creation of ppps and most 20 kappler, schomaker and bauer likely further strengthen the concept of the corresponding trust trap. the potential role of information as quasi-substitute for relational experience in trust development is rather sobering and h3, h4, and h5 cannot be accepted. indeed, interventions have the expected impact that validates the hypothesized cause-effect relationship though exclusively within the sample. it may, however, suggest that such information treatment will not suffice to make ppps a considerable alternative for both public and private actors. limited effects of the information treatment may be due to the perceived trust violation ex-ante being integrity-based rather than competence-based. as found by grover et al. (2014), kim et al. (2013), and dirks et al. (2011), such violation is likely to be perceived as strategic and will most likely be repeated. hence, corresponding ongoing negative expectations may translate into long-term attitudes and, consequently, stable trust levels despite of positive information. this line of thought is strongly confirmed by empirical evidence found within the context of h6 that hypothesizes trust in ppps being firm. by far the strongest explanatory variables to explain post-test trust levels was the trust level that was queried before the treatment. limited effects of the information treatment, however, may also be explained by the specifics of the treatment. first, the information treatment was relatively technical and descriptive and may fall short of resembling personal experience that is made through interaction (***). following these lines, more personal, emotional and salient information could have a stronger impact. second, information was provided once. particularly, in the context of relatively stable trust levels, one could also raise doubts on the impact of a one-off interaction-based experience. following the virtuous cycle in which interaction and trust develop mutually over time and various frequent interactions, one may also think of information as if delivered more frequently to have stronger impact through the functioning of a similar cycle (korsgaard, 2018; six & skinner, 2010). along these lines, rather than one-off information as applied in this paper, one may provide information more frequently or on demand similar to accountability and transparency measures which are found to have stronger effects on trust levels (kumlin & haugsgjerd, 2017; van de walle, 2017). third, against the backdrop of divergent rationales, it may also be more effective to provide more tailor-made information that better suits the idiosyncratic incentiveand belief-system of the different partners. in particular, it could help if information addresses specific reservations and ways to circumvent partner-related risks rather than more generic information as provided in the empirical study at hand. with a closer look at h5, the effect of negative information being stronger than the one of positive information is revealing and confirms literature by large. while it proposes a very slow and incremental generation of trust, trust destruction seems to be a rather disruptive endeavor (six, 2005). following basic trust psychology, it may be, in particular, the case of ppps’ divergent rationales in which negative information can have a relatively large effect since it is ‘confirmatory’ inasmuch as it meets yet existing negative reservations that again can serve as fertile ground for further decreasing trust. public-private partnerships caught in a trust-trap and the potential role of information 5. conclusions this paper sheds light on (low) trust as an explanatory variable for ppp-related problems and advances the field of trust in heterogeneous partnerships by introducing the supra-organisational structure of ppp, the burden of divergent rationales, and exogenous information as potential (quasi) substitute for relational experience. this paper explained ppps as being caught in a trust trap in which the necessary interaction for trust development does not take place because trust levels are too low to interact in the first place. the experimental setting applied served well to empirically scrutinize different trust levels within ppps based on the different rationales in either the public or the private sector. making the rule of public initiation a particular problem, public sector representatives showed significantly lower trust levels in ppps than their private counterparts do. accordingly, the proposed existence of divergent rationales within ppps, indeed, develops into a de facto burden when approaching trust development. as a potential means to develop a sufficient trust level that allows for interaction to occur, exogenous information was hypothesized to act as a quasi-substitute for relational experience. empirical evidence strongly supported the concept of divergent rationales but found a relatively low impact of information treatments. explaining partially the lack of effect of information on trust in ppps, trust levels were very stable and, thus, very difficult to change. that again strongly backs both the concept of divergent rationales and the perceived-trust violation being integrity-based informed by divergent rationales. as such, this paper contributes to better understanding of the nexus between ppps and (low) trust, providing a rather sobering image for trust prospects though. however, it is this point where this paper’s limitations and suggestions for future research concur. information can vary in presentation, frequency of provision, and content. accordingly, findings made are valid exclusively for the kind of information (written, moderate, general role-based information, one-time) given in this experiment and – if at all – give an idea on information effects in general. given the stability of trust levels, it may be important for information to better mimic personal relational experience in order to attain a larger effect, similar to the one of relational experience. tailor-made information for each of the trustors, considering divergent rationales, potentially making it more personal, more emotional, and more salient may increase the effect of information on trust levels. accordingly, findings made here would highly benefit from further contestation by the inclusion of different forms of information. 22 kappler, schomaker and bauer references anderson, d.m., & edwards, b.c. (2015). unfulfilled promise: laboratory experiments in public management research. public management review 17(10), pp. 1518-1542. anderson, e., & jap, s.d. (2005). mitsloan management review. retrieved from the dark side of close arnold, c., sapir, e.v., & zapryanova, g. (2012). trust in the institutions of the european union: a crosscountry. european integration online papers, 16. australian government. (2016). www.infrastructure.gov.au. retrieved from: https://www.infrastructure.gov.au/sites/default/files/migrated/infrastructure/ngpd/files/nat ional-ppp-policy-framework-oct-2015.pdf bachmann, r., gillespie, n., & priem, r. (2015). repairing trust in organizations and institutions: toward a conceptual framework. organization studies, 1123-1142. https://doi.org/10.1177/01708406 15599334 barber, b. (1983). the logic and limits of trust. new brunswick: rutgers university press. bidault, f., & castello, a. (2009). trust and creativity: understandingthe role of trust in creativity-oriented joint developments. r&d management39, 3, pp. 259-270. https://doi.org/10.1111/j.14679310.2009.00557.x bouckaert, g. (2011). "reforming for performance and trust: some reflections." core speech at the nispacee conference, varna, bulgaria, may 19-22. bouwman, r., & grimmelikhuijsen, s. (2016). experimental public administration from 1992 to 2014: a systematic literature review and ways forward. international journal of public sector management 29(2), pp. 110-131. bovaird, t. (2004). public-private partnerships: from contested concepts to prevalent practice. international review of administrative sciences 70(2), pp. 199-215. https://doi.org/10.1177/ 0020852304044250 brogaard, l. (2019). business value in public-private partnerships:the positive impact of trust and taskrelevantcompetencies on business outcomes in ppps. international public management journal, 22:4, pp. 617-642. https://doi.org/10.1080/10967494.2018.1457107 de schepper, s., dooms, m., & haezendonck, e. (2014). stakeholder dynamics and responsibilities in public-private partnerships: a mixed experience. international journal of project management, pp. 1210-1222. https://doi.org/10.1016/j.ijproman.2014.01.006 denize, s., & young, l. (2007). concerning trust and information. industrial marketing management (36), pp. 968-982. https://doi.org/10.1016/j.indmarman.2007.06.004 dietz, g. (2011). going back to the source: why do people trust each other? journal of trust research, 1, 215-222. https://doi.org/10.1080/21515581.2011.603514 dietz, g., & den hartog, d.n. (2006). measuring trust inside organisations. personnel review 35(5), s. 557588. dirks, k.t., kim, p.h., ferrin, d.l., & cooper, c.d. (2011). understanding the effects of substantive responses on trust following a transgression. organizational behavior and human decision processes, 114(2), pp. 87-103. https://doi.org/10.1016/j.obhdp.2010.10.003 druckman, j.n. (2011). what's it all about?: framing in political science. in g. keren, perspectives on framing. new york: psychology press. edelenbos, j., & klijn, e.h. (2007). trust in complex decision-making networks: a theoretical and empirical exploration. administration and society, pp. 25-50. https://doi.org/10.1177/ 0095399706294460 edwards, j.r., & cable, d.m. (2009). the value of value congruence. journal of applied psychology, 654677. https://doi.org/10.1037/a0014891 el-gohary, n., osman, h., & el-diraby, t.e. (2006). stakeholder management for public private partnerships. international journal of project management 24, pp. 595-604. https://doi.org/10.1016/j.ijproman.2006.07.009 eurodad. (2018). history repppeated: how public private partnerships are failing. retrieved from https://d3n8a8pro7vhmx.cloudfront.net/eurodad/pages/508/attachments/original/1590679 608/how_public_private_partnerships_are_failing.pdf?1590679608 european court of auditors. (2018). public private partnerships in the eu: widespread shortcomings and limited benefits. retrieved from https://www.eca.europa.eu/lists/ecadocuments/sr18_09/ sr_ppp_en.pdf farazmand, a. (1999). globalization, privatization, and the future of modern governance: a critical assessment. public finance & management, 2(1), pp. 151-186. fukuyama, f. (1995). trust: the social virtues and the creation of prosperity. new york: free press. gillespie, n., & dietz, g. (2009). trust repair after an organization-level failure. the academy of management review, pp. 127-145. https://doi.org/10.5465/amr.2009.35713319 https://www.infrastructure.gov.au/sites/default/files/migrated/infrastructure/ngpd/files/national-ppp-policy-framework-oct-2015.pdf https://www.infrastructure.gov.au/sites/default/files/migrated/infrastructure/ngpd/files/national-ppp-policy-framework-oct-2015.pdf https://doi.org/10.1177/0170840615599334 https://doi.org/10.1177/0170840615599334 https://doi.org/10.1111/j.1467-9310.2009.00557.x https://doi.org/10.1111/j.1467-9310.2009.00557.x https://doi.org/10.1177/0020852304044250 https://doi.org/10.1177/0020852304044250 https://doi.org/10.1080/10967494.2018.1457107 https://doi.org/10.1016/j.ijproman.2014.01.006 https://doi.org/10.1016/j.indmarman.2007.06.004 https://doi.org/10.1080/21515581.2011.603514 https://doi.org/10.1016/j.obhdp.2010.10.003 https://doi.org/10.1177/0095399706294460 https://doi.org/10.1177/0095399706294460 https://doi.org/10.1037/a0014891 https://doi.org/10.1016/j.ijproman.2006.07.009 https://d3n8a8pro7vhmx.cloudfront.net/eurodad/pages/508/attachments/original/1590679608/how_public_private_partnerships_are_failing.pdf?1590679608 https://d3n8a8pro7vhmx.cloudfront.net/eurodad/pages/508/attachments/original/1590679608/how_public_private_partnerships_are_failing.pdf?1590679608 https://www.eca.europa.eu/lists/ecadocuments/sr18_09/sr_ppp_en.pdf https://www.eca.europa.eu/lists/ecadocuments/sr18_09/sr_ppp_en.pdf https://doi.org/10.5465/amr.2009.35713319 public-private partnerships caught in a trust-trap and the potential role of information goodin, r., march, j., & olsen, j. (2011). the logic of appropriateness. in the oxford handbook of political science. https://doi.org/10.1093/oxfordhb/9780199604456.001.0001 grimsey, d., & lewis, m.k. (2005). are public private partnerships value for money?: evaluating alternative approaches and comparing academic and practitioner views. accounting forum 29(4), pp. 345378. https://doi.org/10.1016/j.accfor.2005.01.001 grover, s. l., hasel, m.c., manville, c., & serrano-archimi, c. (2014). follower reactions to leader trust violations: a grounded theory of violation types, likelihood of recovery, and recovery process. european management journal 32(5), pp. 689-702. hall, d. (2012). re-municipalising municipal services in europe. a report commissioned by epsu to public services international research unit (psiru). hardin, r. (2002). trust and trustworthiness. russell sage foundation. head, b.w., & alford, j. (2015). wicked problems: implications for public policy and management. administration & society vol. 47(6), pp. 711-739. https://doi.org/10.1177/0095399713481601 hollis, m. (1998). trust within reason. cambridge: cambridge universtity press. https://doi.org/ 10.1017/cbo9780511612244 james, o., jilke, s., petersen, c., & van de walle , s. (2016). citizens' blame of politicians for public service failure: experimental evidence about blame reduction through delegation and contracting. public administration review, pp. 83-93. https://doi.org/10.1111/puar.12471 kaplan, a.m., & haenlein, m. (2009). the increasing importance of public marketing:explanations, applications and limits of marketingwithin public administration. european management journal, 197-212. https://doi.org/10.1016/j.emj.2008.10.003 kappler, m. (2022). western multinational corporations in latin america: conflating capitalisms and institutional dynamics of inter-systemic actor exchange. cham: springer nature. https://doi.org/10.1007/978-3-030-93805-5 kappler, m., & schomaker, r.m. (2023). vertrauen in governance und regulierung. verwaltung & management 29(6), pp. 243-253. kim, p.h., ferrin , d. l., cooper, c. d., & dirks, k.t. (2013). repairing trust with individuals vs. groups. organizational behavior and human decision processes, 120(1), pp. 1-14. https://doi.org/ 10.1016/j.obhdp.2012.08.004 klijn, e.h., edelenbos, j., & steijn, b. (2010). trust in governance networks; its impact and outcomes. administration and society, pp. 193-221. https://doi.org/10.1177/0095399710362716 korsgaard, m.a. (2018). reciprocal trust: a self-reinforcing dynamic process. in r. h. searle, a. m. nienaber, & s. b. sitkin, the routledge companion to trust (pp. 14-18). london: routledge. https://doi.org/10.4324/9781315745572-3 kramer, r.m. (1999). trust and distrust in organizations: emerging perspectives, enduring questions. annual reviews in psychology, 50, 569-598. https://doi.org/10.1146/annurev.psych.50.1.569 kramer, r.m. (2010). collective trust within organizations: conceptual foundations and empirical insights. corporate reputation review 13, pp. 82-97. https://doi.org/10.1057/crr.2010.9 kroeger, f. (2012). trusting organizations: the institutionalization of trust in interorganizational relationships. organization, pp. 743-763. https://doi.org/10.1177/1350508411420900 kumlin, s., & haugsgjerd, a. (2017). the welfare state and political trust: bringing performance back in. in s. zmerli, & t.w. van der meer, handbook on political trust (pp. 285-301). cheltenham: edward elgar. https://doi.org/10.4337/9781782545118.00029 lascaux, a. (2005). trust and transaction costs. in j. finch, & m. orillard, complexity and the economy: implications for economic policy (pp. 151-171). cheltenham: edward elgar. https://doi.org/ 10.4337/9781845428044.00018 lee, m., han, x., quising, p. f., & villaruel, m. l. (2018). hazard analysis on public-private partnership projects in developing asia. manila: adb economics working paper series. https://doi.org/ 10.22617/wps189453-2 lewicki, r.j., & bunker, b.b. (1996). developing and maintaining trust in work relationships. in r. m. kramer, & t. r. tyler, the new face of government: how public managers are forging a new approach to governance (pp. 114-139). boca raton: auerbach publications. https://doi.org/ 10.4135/9781452243610.n7 lewis, d.j., & weigert, a. (1985). trust as a social reality. social forces, 63 (4), 967-985. https://doi.org/10.2307/2578601 luhmann, n. (1979). trust and power. chicester: john wiley and sons. luhmann, n. (1988). familiarity, confidence, trust. problems and alternatives. in d. gambetta, trust. making and breaking cooperative relations (pp. 94-107). oxford: blackwell. margetts, h. (2011). experiments for public management research. public management review 13(2), pp. 189-208. möllering, g. (2006). trust: reason, routine, reflexivity. amsterdam: elsevier. https://doi.org/10.1093/oxfordhb/9780199604456.001.0001 https://doi.org/10.1016/j.accfor.2005.01.001 https://doi.org/10.1177/0095399713481601 https://doi.org/10.1017/cbo9780511612244 https://doi.org/10.1017/cbo9780511612244 https://doi.org/10.1111/puar.12471 https://doi.org/10.1016/j.emj.2008.10.003 https://doi.org/10.1007/978-3-030-93805-5 https://doi.org/10.1016/j.obhdp.2012.08.004 https://doi.org/10.1016/j.obhdp.2012.08.004 https://doi.org/10.1177/0095399710362716 https://doi.org/10.4324/9781315745572-3 https://doi.org/10.1146/annurev.psych.50.1.569 https://doi.org/10.1057/crr.2010.9 https://doi.org/10.1177/1350508411420900 https://doi.org/10.4337/9781782545118.00029 https://doi.org/10.4337/9781845428044.00018 https://doi.org/10.4337/9781845428044.00018 https://doi.org/10.22617/wps189453-2 https://doi.org/10.22617/wps189453-2 https://doi.org/10.4135/9781452243610.n7 https://doi.org/10.4135/9781452243610.n7 https://doi.org/10.2307/2578601 24 kappler, schomaker and bauer möllering, g., & sydow, j. (2019). trust trap? self-reinforcing processes in the constitution of interorganizational trust. in m. sasaki, trust in contemporary society. leiden: brill. https://doi.org/10.1163/9789004390430_009 morton, r.b., & williams, k.c. (2010). experimental political science and the study of causality: from nature to the lab. new york: cmbridge university press. nooteboom, b. (2002). trust: forms, foundations, functions, failures and figures. cheltenham, uk: edward elgar. https://doi.org/10.4337/9781781950883 oecd (2011). the call for innovative and open government: an overview of national initiatives. paris: oecd publishing. https://doi.org/10.1787/9789264107052-en parmentier, s., & vervaeke, g. (2011). in criminal justice we trust? a decade of public opinion research in belgium. european journal of criminology, 8(4), pp. 286-302. https://doi.org/10.1177/ 1477370811411460 ppiaf (2012). ppp basics and principles of a ppp framework. retrieved from: https://www.ppiaf.org/documents/3118 ppp knowledge lab. (2021). https://pppknowledgelab.org. retrieved from https://pppknowledgelab.org /guide/sections/22-ppp-program-objectives putnam, r.d. (1993). making democracy work: civic traditions in modern italy. princeton: princeton university press. https://doi.org/10.1515/9781400820740 putnam, r.d. (2000). bowling alone: the collapse and revival of american community. new york: touchstone. https://doi.org/10.1145/358916.361990 rabin, m. (1998). psychology and economics. journal of economic literature 36(1), 11-46. radin, b. (2006). challenging the performance movement: accountability, complexity, and democratic values. washington, dc: georgetown univ. press. rittel, h.w., & webber, m.m. (1973). dilemmas in a general theory of planning. policy sciences 4, pp. 155169. https://doi.org/10.1007/bf01405730 ross, w., & lacroix, j. (1996). multiple meanings of trust in negotiation theory and research: a literature review and integrative model. international journal of conflict management, 7, 4, pp. 314-360. https://doi.org/10.1108/eb022786 rousseau, d.m., sitkin, s.b., burt, r.s., & camerer, c. (1998). not so different after all: a cross-discipline view of trust. academy of management review 23(3), pp. 393-404. https://doi.org/10.5465 /amr.1998.926617 schomaker, r.m., & bauer, c. (2020). trust and transaction costs in public-private partnerships-theoretical reflections and empirical findings. public money & management, 1-7. https://doi.org/10.1080/09540962.2020.1801882 schoormann, f.d., mayer, r.c., & davis, j.h. (2007). an integrative model of organizational trust: past, present, and future. academy of management review, pp. 344-354. https://doi.org/10.5465/ amr.2007.24348410 scott, r.w. (2010). reflections: the past and future of research on institutions and institutional change. journal of change management 10:1, 5-21. https://doi.org/10.1080/14697010903549408 simpson, t.w. (2012). what is trust? pacific philosophical quarterly, 93(4), 550-569. https://doi.org/10.1111/j.1468-0114.2012.01438.x six, f. (2005). the trouble with trust the dynamics of interpersonal trust building. cheltenham: edward elgar. https://doi.org/10.4337/9781845426873 six, f., & skinner, d. (2010). managing trust and trouble in interpersonal work relationships: evidence from two dutch. the international journal of human resource management, 21(1), s. 109-124. https://doi.org/10.1080/09585190903466913 six, f., & verhoest, f. (2016). trust in regulatory regimes: scoping the field. in f. six, & k. verhoest, trust in regulatory regimes (pp. 1-36). cheltenham: edward elgar. https://doi.org/10.4337/ 9781785365577.00005 stern, r. (2016). løgstrup on the priority of trust. in p. faulkner, & t. w. simpson, the philosophy of trust (pp. 272-294). oxford: oxford university press. https://doi.org/10.1093/acprof:oso/ 9780198732549.003.0016 suresh, k.p., & chandrashekara, s. (2012). sample size estimation and power analysis for clinical research studies. journal of human reproductive sciences 5(1), pp. 7-13. sztompka, p. (1999). trust as a sociological theory. cambridge : cambridge university press. teicher, j., alam, q., & van gramberg, b. (2006). managing trust and relationships in ppps: some australianexperiences. international review of administrative sciences, pp. 85-100. https://doi.org/10.1177/0020852306061624 tomlinson, e.c., & mryer, r.c. (2009). the role of causal attribution dimensions in trust repair. academy of management reviewvol. 34, no. 1. https://doi.org/10.1163/9789004390430_009 https://doi.org/10.4337/9781781950883 https://doi.org/10.1787/9789264107052-en https://doi.org/10.1177/1477370811411460 https://doi.org/10.1177/1477370811411460 https://www.ppiaf.org/documents/3118 https://pppknowledgelab.org/guide/sections/22-ppp-program-objectives https://pppknowledgelab.org/guide/sections/22-ppp-program-objectives https://doi.org/10.1515/9781400820740 https://doi.org/10.1145/358916.361990 https://doi.org/10.1007/bf01405730 https://doi.org/10.1108/eb022786 https://doi.org/10.5465/amr.1998.926617 https://doi.org/10.5465/amr.1998.926617 https://doi.org/10.1080/09540962.2020.1801882 https://doi.org/10.5465/amr.2007.24348410 https://doi.org/10.5465/amr.2007.24348410 https://doi.org/10.1080/14697010903549408 https://doi.org/10.1111/j.1468-0114.2012.01438.x https://doi.org/10.4337/9781845426873 https://doi.org/10.1080/09585190903466913 https://doi.org/10.4337/9781785365577.00005 https://doi.org/10.4337/9781785365577.00005 https://doi.org/10.1093/acprof:oso/9780198732549.003.0016 https://doi.org/10.1093/acprof:oso/9780198732549.003.0016 https://doi.org/10.1177/0020852306061624 public-private partnerships caught in a trust-trap and the potential role of information van de walle, s. (2017). trust in public administration and public services. in european commission, trust at risk: implications for eu policies and institutions (pp. 118-128). luxembourg: publications office of the european union. van der meer, t.w. (2017). democratic input, macroeconomic output. in s. zmerli, & t. w. van der meer, handbook on political trust (pp. 270-284). cheltenham: edward elgar. https://doi.org/ 10.4337/9781782545118.00028 van gestel, r., & de poorter, j. (2019). in the court we trust: cooperation, coordination and collaboration. 2019: cambridge university press. https://doi.org/10.1017/9781108686556 van ryzin, g.g. (2011). outcomes, process, and trust of civil servants. journal of public administration research and theory, pp. 745-760. https://doi.org/10.1093/jopart/muq092 walker, r.m., james, o., & brewer, g.a. (2017). replication, experiments and knowledge in public management research. public management review, pp. 1221-1234. warsen, r., nederhand, j., klijn, e. h., grotenberg, s., & koppenjan, j. (2018). what makes public-private partnerships work? survey research into the outcomes and the quality of cooperation in ppps. public management review, 20:8, 1165-1185. https://doi.org/10.1080/14719037.2018 .1428415 zhang, x. (2005). critical success factors for public-private partnerships in infrastructure development. journal of construction engineering and management. https://doi.org/10.1061/(asce)07339364(2005)131:1(3) zucker, l.g. (1986). production of trust: institutional sources of economic structure, 1840 1920. research in organizational behavior, 8, pp. 53-111. https://doi.org/10.4337/9781782545118.00028 https://doi.org/10.4337/9781782545118.00028 https://doi.org/10.1017/9781108686556 https://doi.org/10.1093/jopart/muq092 https://doi.org/10.1080/14719037.2018.1428415 https://doi.org/10.1080/14719037.2018.1428415 https://doi.org/10.1061/(asce)0733-9364(2005)131:1(3) https://doi.org/10.1061/(asce)0733-9364(2005)131:1(3) 26 kappler, schomaker and bauer appendix pre-test questions statistical information: age ________________ gender m f course of study: general management other ______________ public administration? yes no state ________________________________________________ pre-test questions regarding the provision of public services we are interested in your opinion regarding the provision of public services and kindly ask you to answer the following questions. please mark the answer that most closely matches your opinion. the scientific evaluation of the questionnaires will be anonymous; the results will only be used for research regarding public administration and will not be used for other purposes. 1) the public sector (municipal, regional, or state level) in germany and the european union is efficient and effective in the provision of infrastructure such as water/sewerage, construction of schools, or transportation infrastructure. strongly agree agree neutral disagree strongly disagree 2) you may be familiar with the concept of ‘public-private partnerships (ppp)’ – contractual agreements between the public sector and private enterprises to provide infrastructure and services in areas such as water or transportation. the aim of such cooperation is riskand work-sharing. the private partner is responsible for providing the service efficiently and provides capital, while the public partner takes over responsibility for meeting overall welfare goals. the inclusion of private enterprises in these arrangements in germany or the eu is very desirable desirable neutral not desirable completely undesirable public-private partnerships caught in a trust-trap and the potential role of information treatment groups group a economists and researchers in public administration deal with questions that are of general political interest as well as relevant to the current practice of administration. the information gathered from this research may contribute to more efficient and effective service provision by the administration in the long run. this applies to germany and the european union level. in this context surveys such as this one help, among other things, to collect information about the attitude of individuals towards politics and administrative actions. this information, qualitative or quantitative, not only brings new insights to research, but also forms the basis for political consulting. thank you for your participation! group b although the number of ppps is increasing, their theoretical benefit is quite controversial, even the public authorities are not completely convinced about the usefulness of ppps as an alternative procurement method. not all ppp-concepts are fully developed, resulting in time overruns, subsequent cost increases, or even the cancellation of ppps. thus, many decision makers, committees, and civil servants tend to execute their planning and procurement under their own direction due to inexperienced administrative procedures. in most areas there is an uncertainty about lawful and administrative procedures. a unified nationwide ppp law that sets rules for assuming risk and specifies the problem handling ppps is still missing so far. group c many advantages are expected in connection to ppps: first, an anticipatory effect as ppp investments can be helpful to the indebted public sector. this method is easier compared to the normal, often tedious budget planning for its own raising of credit. in addition, ppp-projects result in a time gain with their faster, on schedule, and in budget completion as well as an increase in efficiency through the pure economic management of the private sector in the maintenance and operation of systems. ppps are true to the principle that the government and the private sector should concentrate on their strengths and core competences, resulting in optimization of the service for the users and increases in the overall efficiency of the performance. the private companies have long-term experience in the optimal design of specific types of projects, in particular in structural engineering, and therefore they can better estimate the existing project and operational risk. thus, through ppps a more economically efficient provision of services can be achieved for the citizens’ benefit. post-test questions in the following you will find further questions on service provision. we kindly ask you to answer them, marking the answer that best reflects your opinion. for question 3, more than one answer can be marked. 28 kappler, schomaker and bauer 1) with regard to the importance of services for the society and regarding to the potential benefits and risks related to ppps, private enterprises should be included more often/whenever possible yes no 2) please mark your level of agreement ‘ppps should be used’ strongly agree agree neutral disagree strongly disagree 1. introduction 2. disentangling trust in ppp and the role of information 3. results 4. discussion references © the author(s) 2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 1 (2024), pages 103-119 https://doi.org/10.17979/ejge.2024.13.1.9902 submitted: august 26, 2023 accepted: april 24, 2023 published: june 6, 2024 article democracy, governance, and environmental policy effectiveness: a cross-country analysis with sustainable governance indicators sevi dokuzoğlu,1,* arif eser güzel 2 1 hatay mustafa kemal university, department of public finance, turkey. 2 hatay mustafa kemal university, department of economics, turkey. *correspondence: dokuzoglusevi@gmail.com abstract. the effectiveness of environmental policies has been extensively studied, often focusing solely on environmental degradation indicators. however, successful environmental policies have broader societal impacts. this study addresses this limitation by utilizing the sustainable governance indicators (sgis) for a more comprehensive measurement of environmental policy effectiveness. analyzing 41 countries with data from 2014-2021, a fixed-effects model was employed to assess the impact of democratic and good governance practices on environmental policy success. the findings reveal that democracy and government quality are crucial determinants of effective environmental policies, highlighting the need for improved institutional structures. the robustness of these results is confirmed using alternative democracy and governance indicators. additionally, the study identifies education, urbanization, and population density as significant factors influencing environmental policy performance. keywords: environmental policy effectiveness; democracy; sustainable governance indicators; education; urbanization. jel classification: o13; q58; p18; i29; d73 1. introduction production, employment, food security, and environmental sustainability are at risk due to degraded ecosystems, rising biodiversity losses, and climate catastrophe. these consequences were intensified even further by the covid-19 outbreak (fao, 2022a). the efficiency of production factors decreases as a result of climate change, which also limits access to factor utilization. it has a major negative economic impact, particularly in climate-sensitive industries, including forestry, agriculture, tourism, and fisheries (unctad, 2021). given that the tourism industry contributes 6.1% of the global gdp and the agricultural sector 4% (wtcc, 2022; fao, 2022b), it is crucial to preserve these industries from environmental deterioration. if adequate policy measures are not put in place, the economic and social costs of environmental deterioration may become much worse than covid-19 in the future. in addition, environmental problems that feed off one another, such as climate change, biodiversity loss, air pollution, and water pollution, may render people more susceptible to https://creativecommons.org/licenses/by-nc/4.0/ 104 dokuzoğlu and güzel epidemics in the future (oecd, 2020). however, even legally binding policy targets set out in the convention on biological diversity and the paris agreement, which include temperature limits and greenhouse gas emission reductions, have still not been achieved (ekardt, bärenwaldt & heyl, 2022; ekardt et al. 2023). although the environmental kuznets curve characterizes the relationship between aggregate income and environmental degradation by the level of welfare, increased welfare may also lead to further degradation due to the rebound effect (stern, 2020). to manage environmental quality, governments can use a variety of policy measures, including taxes, subsidies, rationing, prohibition, technical standards, public production, and moral persuasion (oates & baumol, 1975). performance on behalf of institutions and governments determines how effective these policy instruments are. in the presence of institutional and political flaws, both a government failure and a market failure may occur (hepburn, 2010). however, political, institutional, and demographic variables do influence the nature and course of this connection. for instance, according to deacon and norman (2006), the link between pollution and income varies by nation. the effects of environmental deterioration brought on by economic growth will lessen as environmental regulations and institutions become more effective. effective measures, according to panayotou (1997), make the environmental kuznets curve flatter and lower the environmental costs of growth. a positive linear link between environmental degradation and economic growth is predicted in the absence of appropriate policies (nicolli et al. 2012). additionally, some elements, such as climate conditions that have an impact on environmental quality and natural resources, are out of the political process's control (pellegrini & gerlagh, 2006). therefore, it is important to be interested in how well government programs are working. in this paper, we show how democracy and the quality of governments affect the success of environmental policies. the study's use of distinctive environmental performance indicators sets it apart from previous research in the field. environmental policy stringency index and environmental performance index have been employed as performance indicators of environmental policies in several research works (damania, fredriksson & list, 2003; pellegrini & gerlagh, 2009; mavragani, nikolaou & tsagarakis, 2016). studies have also linked environmental outcomes, such as greenhouse gas emissions or ecological footprint, to the success of policies (panayotou, 1997; torras & boyce, 1998; easty & porter, 2005; ward, 2008). instead, we make use of the bertelsmann stiftung's sustainable governance indicators. the data set already covers widely referenced global environmental agreements like kyoto as well as policy targets like co2 emissions, waste generation, footprints, etc. we might say that the index is more thorough than others in this regard. the remainder of the essay is structured as follows: the literature on environmental policy effectiveness and its institutional determinants is reviewed in section 2. the model, data, and approach are described in section 3. the empirical findings and discussion in section 4 are followed by the conclusion in section 5. democracy, governance, and environmental policy effectiveness 2. literature review the economic analysis of environmental policies is based on the idea that economic activities create externalities by damaging the environment. the market mechanism fails to internalize such externalities because of the free-rider problem and transaction costs, therefore public intervention is required (pigou, 1932; coase, 1960). for a long time, the effectiveness of environmental policies has been discussed using a pigouvian framework. according to pigou (1932), the government should impose taxes on economic activities to equal the marginal damage of pollution. however, the pigouvian tax is not sufficient for environmental quality and is usually not implemented correctly (fullerton, hong and metcalf, 2000). in addition to the pigouvian approach, socio-economic, political, and institutional factors have recently been widely discussed among economists (carraro & metcalf, 2000; esty & porter, 2005). environmental protection is considered as a public good (kirchgässner & schneider, 2003). the provision of public goods is closely related to the level of democracy. in autocracies, the state budget is allocated through transfers to politically influential groups. however, in democratic regimes, a wide range of people need to be satisfied. therefore, governments spend on public goods where economies of scale occur. according to deacon (2009), in democracies, the level of public policies on environmental protection (e.g. clean water, sanitation, pollution control) is greater than in non-democratic ones. in democratic systems, voters who want to protect the environment and are against the waste of natural resources, pressure the government to be more accountable. freedom of expression further increases this pressure. democracy has often been associated with the free flow of information about environmental degradation and the ability of citizens to resist it (pellegrini and gerlagh, 2006). governments in democratic states participate more in international collaborations and agreements on environmental protection than non-democratic states (congleton, 1992). as the level of democracy increases, people are more concerned with the burden on future generations (fiorino, 2011). quality of governance improves the performance of governments and makes policy outcomes effective. good governance better responds to voter demands (fiorino, 2011). in the political economy framework, environmental policies are important in that they could represent many government decision-making processes. because environmental policies are usually determined by economic and political self-interest rather than public interest (aidt, 1998; fredriksson & svensson, 2002). corruption causes environmental policies to differ in their legal form and practice. when bureaucrats are corrupt, non-compliance with policy rules is not punished. thus, the stringency of government policies is reduced (damania, 2002). however, public support for environmental policies increases when the citizens believe that government is effective and fair (huber, wicki & bernauer, 2020). thus, voluntary compliance with environmental policies increases. harring (2014) concluded that corruption reduces the effectiveness of environmental policies because of the unwillingness to comply with policy decisions. in addition, the effect of a rise in the demand for environmental policy also depends on corruption (damania, fredriksson & list, 2003). according to damania (2002), environmental regulations can lead to more corruption in countries that already have corruption problems due to environmental taxation. governments differ from each other in the effectiveness of public policies (tanzi, 1998; afonso, shuknecht & tanzi, 2005; afonso, 106 dokuzoğlu and güzel shuknecht & tanzi, 2010; adam, delis & kammas, 2014). according to afonso, shuknecht & tanzi (2005), rajkumar & swaroop (2008), hwang & akdede (2011), as well as traditional indicators (such as gdp growth, inflation, unemployment, etc.) education, rule of law, bureaucracy, corruption and demography are important determinants on the overall performance of the public sector. it has been discussed for a long time which of the policy instruments for environmental protection is effective. however, a consensus has not been reached on the factors that determine the effectiveness of environmental policies (steinebach, 2022). most of the studies focus on certain policy tools or environmental outcomes, while several others emphasize environmental policy performance. morley (2012) examined policy performance through the effectiveness of environmental taxes using a dynamic panel for the eu countries. nerudova and solilova (2016) compared the relative effectiveness of environmental taxes and public expenditure on environmental protection using the vector error correction model. there is extensive literature on which the environmental policy goal is to reduce greenhouse gas emissions and energy use (panayotou, 1997; torras & boyce, 1998; easty & porter, 2005; wawrzyniak & doryń, 2020; acheampong, dzator & savage, 2021). ward (2008) also revealed the effect of democracy based on ecological footprint as a measure of policy success. nicolli, mazzanti, and iafolla (2012) analyzed the socio-economic and political factors that determine the effectiveness of environmental policies through waste dynamics for the eu15 in the sure model. according to the study, population density has a positive impact on waste generation and a negative impact on landfilled waste, depending on whether economies of scale arise. on the other hand, neumayer, gates and gleditsch (2002) argued that environmental outcomes such as the amount of emission and soil erosion may occur depending on the type of energy used and climate conditions. both are hard for institutional and political actors to control. this is why, in many studies, the effect of institutional factors such as democracy on emissions is not robust. instead of environmental outcomes, neumayer, gates and gleditsch (2002) used environmental commitments, including international environmental agreements, memberships in environmental organizations, the number of protected area statuses, etc. according to the study, the impact of democracy on environmental commitments is positive, significant, and robust. bättig and bernauer (2009) discussed policy effectiveness by using an index consisting of climate change policy and emissions in a cross-section of data from 185 countries. while the impact of democracy on climate change policy is positive, its impact on environmental outcomes is uncertain. damania, fredriksson and list (2003) examined the effectiveness of policies in the context of environmental policy stringency using panel data for developed and developing countries by developing a political economy model. the study indicates that corruption reduces the stringency of policies. pellegrini and gerlagh (2006) also used the environmental policy stringency index to point out the effects of socio-economic and institutional factors. the stringency increases at high levels of democracy and decreases at high levels of corruption. kelleher, kim and chang (2009), mukherjee and chakraborty (2013), mavragani, nikolaou and tsagarakis (2016), chang and hao (2017) used the environmental performance index (epi) developed by yale university to examine the effects of institutions and government quality. control of corruption improves the environmental performance in kelleher, kim, and chang (2009); democracy, control of corruption and education are positively related to the epi in mukherjee and democracy, governance, and environmental policy effectiveness chakraborty (2013); quality of government and control of corruption have a positive effect on the epi in mavragani, nikolaou and tsagarakis (2016); government quality causes better environmental performance in chang and hao (2017). adam and tsarsitalidou (2019) evaluated the environmental policy effectiveness of 39 countries with the dea method. the study indicates that the effectiveness of environmental policies is affected by economic, demographic, and political factors. according to the regression results, population density and corruption have no effect on policy effectiveness, while urbanization has a positive effect. the effect of democracy varies depending on the income level. the first gap identified in the literature is that there is no consensus on the effects of democracy and governance indicators on the effectiveness of environmental policies. while some studies have revealed a strong positive relationship, some have concluded that this relationship is not certain, while others have concluded that it differs depending on the income level of the countries. it is seen that the main difference in the empirical literature is among the methods of measuring effectiveness, which means the degree of achieving certain policy goals (ekardt, 2022). in many studies, the effectiveness of environmental policies has been directly measured by indicators of environmental degradation. however, it is debatable whether co2 emission or ecological footprint alone can be an indicator of environmental policy success. reducing co2 emissions is part of environmental policy objectives, but broader indicators should be used to measure this success. for this purpose, some studies have used the environmental policy stringency index published by oecd as an indicator of environmental policy effectiveness. this index considers the strength of marketbased, non-market instruments and government support when measuring how strictly environmental policy instruments are implemented. however, strict implementation of environmental policies does not guarantee that these policies will be effective in reducing environmental degradation. there is also no indication of the outputs of environmental policies in the calculation of the index. in several studies, the environmental performance index published by yale university and developed by wolf et al. (2022) is used as an indicator of environmental policy effectiveness. although the sub-components of this index are published in the form of time series, it is not appropriate to assemble them in the form of time series or panel data. this is because the dataset and methodology vary between different versions of epi (wolf et al., 2022). sustainable governance indicators offer a broader measure than indicators of environmental degradation such as co2 emissions. they are also appropriate for comparison across countries in the context of environmental policy effectiveness. in addition to the empirical studies in the literature, this study investigates the role of democracy and government quality in the performance of environmental policies utilizing the sustainable governance indicators dataset published by bertelsmann stiftung (2023). 3. model, data, and methodology as a measure of the effectiveness of environmental policies, the environmental policy performance index of the sustainable governance indicators published by bertelsmann stiftung (schiller, hellmann & paulini, 2022) was used. this indicator is more comprehensive than environmental 108 dokuzoğlu and güzel degradation measures such as co2 emission and ecological footprint, which are frequently used in the literature. the indicator has two main sub-components. these are environmental policy and global environmental protection. 50% of the environmental policy sub-component includes expert assessments of environmental policy effectiveness. the remaining 50% covers energy productivity, co2 emission, particulate matter, biocapacity, waste generation, material recycling, biodiversity, renewable energy, and material footprint indicators. 50% of the global environmental protection subcomponent also includes expert assessments. 25% includes participation in multilateral environmental agreements and 25% success in meeting the kyoto targets. in the study, sgi robust democracy and good governance indicators published by bertelsmann stiftung were used as proxies of democracy and government quality (dem and gov in equations). the composition of the robust democracy indicator consists of electoral processes, access to information, civil rights and political liberties, and the rule of law sub-components. good governance indicator considers government quality in the context of accountability and executive capacity (schiller, hellmann & paulini, 2022). real gdp per capita and its square, global energy price index, education, urbanization, and population density variables were used as control variables. in the literature, these variables have been associated with the success of environmental policies. there is extensive literature that focuses on the relationship between national income and environmental quality based on the environmental kuznets curve (grossman and krueger, 1995; holtz-eakin and selden, 1995; cole, rayner, and bates, 2001; stern and common, 2001). according to the hypothesis, environmental degradation is temporary, and the problem can be solved with sufficient growth rate and technological progress, although the threshold level of national income that improves environmental conditions is not fully clear (bimonte, 2002). the energy price index is also used as a control variable. it is expected that as non-renewable energy prices increase, the tendency towards alternative energy sources will increase. in this respect, high energy prices can contribute to the reduction of co2 emissions. (li, fang & he, 2020). there are important demographic factors that affect the success of environmental policies. the first is education. educated people have a high perception of environmental problems. they demand policies that increase environmental quality and are more participatory and organized. rivera-batiz (2002) empirically confirmed that as the level of education increases, democratic participation will increase, and individuals will be more demanding of effective policies. in addition, since the compliance of educated people with environmental regulations is high, the effectiveness of the policies increases (bimonte, 2002; farzin & bond, 2006). on the other hand, having a majority of educated people may not always guarantee a smaller ecological footprint since it is highly correlated with economic development (heyl & ekardt, 2022). an increase in urban population may lead to higher levels of energy consumption and, therefore, higher emissions. however, a few studies addressed the possible positive effects on environmental policy effectiveness. urbanization reduces the cost of transportation and transactions required for people to come together and organize. this may lead to an increase in environmental activism. rising activism increases politicians' awareness of environmental degradation (farzin & bond, 2006). urbanization also creates economies of scale, for example in the provision of sanitation facilities to reduce environmental pollution (torras & boyce, 1998). population density is also associated with economies of scale. as population density democracy, governance, and environmental policy effectiveness increases, economies of scale can occur through urbanization (mazzanti & zoboli, 2009). economies of scale offset the pollution-increasing effect of population density. the social costs of environmental degradation are high in densely populated areas (nicolli, mazzanti & iafolla, 2012). therefore, the government is more sensitive to environmental problems in such areas. the models established following the related literature are as follows. eppit=β0+β1gdpit+β2gdpit 2 +β3epit+β4eduit+β5urbit+β6pdit+β7demit+𝜇𝜇it [1] eppit=𝜃𝜃0+𝜃𝜃1gdpit+𝜃𝜃2gdpit 2 +𝜃𝜃3epit+𝜃𝜃4eduit+𝜃𝜃5urbit+𝜃𝜃6pdit+𝜃𝜃7govit+εit [2] in equations 1 and 2, epp refers to the environmental policy performance index. democracy and good governance indicators are dem and gov respectively. since these variables are highly correlated, models were set up separately. gdp is real gdp per capita, ep is energy prices, edu is expected years of schooling as a proxy of education, urb is urban population, and pd is population density. variable definitions and sources are presented in table 1. table 1. variable definitions and sources variable definition source epp environmental policy performance index bertelsmann stiftung (2023) gdp real gdp per capita (constant 2015, us $) world bank (2023a) ep energy price index (2016=100). it is the average of global oil, coal, and natural gas price indices. imf (2023) edu expected years of schooling for children. undp (2023) urb urban population (% of the total population) world bank (2023a) pd population density (people per sq. km of land area) world bank (2023a) dem quality of democracy bertelsmann stiftung (2023) gov good governance bertelsmann stiftung (2023) the data set includes data from 41 countries. the availability of data sets was considered as the selection criteria. table 2 presents a list of the countries. table 2. countries included in the sample australia czechia iceland malta slovakia united states austria denmark ireland mexico slovenia belgium estonia israil netherlands south korea bulgaria finland italy new zealand spain canada france japan norway sweden chile germany latvia poland switzerland croatia greece lithuania portugal türkiye cyprus hungary luxemburg romania united kingdom 110 dokuzoğlu and güzel table 3. descriptive statistics epp gdp ep edu urb pd dem gov mean 5.958 35278.66 133.16 16.860 76.878 168.06 7.131 6.589 median 5.894 31372.5 123.30 16.418 79.577 104.74 7.30 6.592 max 8.796 108351 212.32 23.089 98.079 1610.4 9.292 8.925 min 3.091 6796.69 91.70 13.897 53.557 3.056 2.692 3.607 sd 1.086 22195.13 37.757 1.767 12.150 245.74 1.355 1.112 skewness 0.290 1.071 1.059 0.732 -0.340 3.864 -0.731 0.050 kurtosis 2.970 4.049 3.099 3.413 2.143 20.55 3.436 2.715 n 287 287 287 287 287 287 287 287 the time dimension of the panel covers the period 2014-2020. descriptive statistics for the data set are presented in table 3. all variables are turned into logarithmic form. since the data set has a low time dimension, a choice was made between fixed effects and random effects estimators. according to the hausman test results, the null of random effects was rejected in all estimations. in addition, to see the sensitivity of the results, estimates were made for the two sub-components of the environmental policy performance index. this is also important in terms of determining whether the effects on environmental policy performance occur on energy and environmental indicators or the adaptation process to global environmental policies. 4. empirical results the fixed effects estimation results for models 1 and 2 are presented in table 4. according to the results, the effects of democracy and governance are positive and significant at 1%. a 1% improvement in democracy leads to a 0.38% increase in environmental policy performance. the effect of a 1% improvement in good governance is 0.47%. there is an inverted u relationship between gdp per capita and environmental policy performance. this result is the opposite of that predicted by the environmental kuznets curve hypothesis. for this hypothesis to be valid, the opposite of the relationship (u-shaped) between gdp and environmental degradation would have to be seen in environmental policy performance. estimation results show that demographic variables such as education, urbanization, and population density have a positive effect on environmental policy. in model 1, a 1% increase in expected years of schooling leads to a 0.51% increase in policy performance, while in model 2, this effect is 0.34%. the effects of urbanization and population density are also positive and statistically significant for both models. in model 1, the effect of a 1% increase in urban population is 1.49%, while it is 1.14% in model 2. a 1% increase in population density leads to about 0.50% increase in environmental policy performance. the effect of the energy price index is insignificant in both models. democracy, governance, and environmental policy effectiveness table 4. fixed effects estimation results 𝑬𝑬𝑬𝑬𝑬𝑬 gdp 2.6737*** (2.79) 1.6384* (1.77) gdp2 -0.1402*** (-2.96) -0.0939** (-2.04) ep 0.0004 (0.03) 0.0002 (0.02) edu 0.5109*** (2.89) 0.3458* (1.93) urb 1.4919*** (2.64) 1.1396** (2.01) pd 0.5554*** (3.35) 0.5040*** (3.03) dem 0.3806*** (7.06) gov 0.4709*** (6.70) constant -21.934*** (-4.03) -14.121*** (-2.75) f 12.30*** 11.50 𝑹𝑹𝒂𝒂𝒂𝒂𝒂𝒂𝟐𝟐 0.933 0.932 hausman 36.21*** 32.51*** nxt 287 287 ***, **, and * indicate the rejection of null hypothesis at 1%, 5%, and 10% respectively. t statistics are given in parentheses. table 5 presents the estimation results obtained for the environmental sub-component of environmental policy performance. the effects of democracy and good governance variables are positive for both models and are significant at 1%. a 1% improvement in democratic institutions leads to an increase of 0.32% in the environment sub-component. the effect of a 1% increase in good governance is 0.33%. when the control variables are examined, it is seen that the kuznets curve hypothesis is still invalid. while the effect of the increase in real gdp is positive, this effect turns negative after the threshold value. on the other hand, the signs of demographic factors are positive and statistically significant. education, urbanization, and population density have positive effects on the environmental sub-component of the environmental policy performance index. energy prices do not have a statistically significant effect in this model either. the results regarding the impact of democracy and good governance in achieving global environmental policy goals are shown in table 6. according to the results, the effect of democracy is positive and significant at the 1% level. a 1% increase in the democracy index improves compliance with global environmental policy targets by 0.46%. the effect of governance quality is similarly positive and significant at the 1% level. a 1% increase in this indicator creates a 0.66% increase in the dependent variable. the effect of gdp again shows that the ekc hypothesis is invalid. considering the impact of demographic indicators, it is seen that the effects of education, urbanization and population density indicators are positive and significant. while education and urbanization are positive and significant at the 1% level, population density is significant at the level of 5% in the first model and 10% in the second model. the impact of the energy price index is insignificant. 112 dokuzoğlu and güzel table 5. fixed effects estimation results for 𝐸𝐸𝐸𝐸𝐸𝐸𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑬𝑬𝑬𝑬𝑬𝑬𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆𝒆 gdp 2.4790*** (2.66) 1.4008* (1.80) gdp2 -0.1266*** (-2.75) -0.0775** (-2.02) ep -0.0170 (-1.45) -0.0174 (-1.41) edu 0.6332*** (3.68) 0.5088*** (3.23) urb 2.1832*** (3.97) 1.8721*** (2.78) pd 0.3924** (2.43) 0.3266* (1.72) dem 0.3244*** (6.48) gov 0.3319*** (3.53) constant -23.816*** (-4.49) -15.943*** (-3.60) f 13.19*** 11.95*** 𝑹𝑹𝒂𝒂𝒂𝒂𝒂𝒂𝟐𝟐 0.92 0.92 hausman 48.42*** 43.29*** nxt 287 287 ***, **, and * indicate the rejection of null hypothesis at 1%, 5%, and 10% respectively. t statistics are given in parentheses. table 6. fixed effects estimation results for 𝐸𝐸𝐸𝐸𝐸𝐸𝑔𝑔𝑔𝑔𝑒𝑒𝑔𝑔𝑔𝑔𝑔𝑔 𝑬𝑬𝑬𝑬𝑬𝑬𝒈𝒈𝒈𝒈𝒆𝒆𝒈𝒈𝒂𝒂𝒈𝒈 gdp 3.4852*** (2.49) 2.5083* (1.82) gdp2 -0.1840*** (-2.62) -0.1417** (-2.04) ep 0.1645 (0.94) 0.0162 (0.95) edu 0.4907** (2.32) 0.2714 (1.16) urb 0.7149 (1.10) 0.3076 (0.43) pd 0.7413*** (3.21) 0.7092*** (2.98) dem 0.4553*** (4.48) gov 0.6563*** (4.32) constant -23.219*** (-3.16) -15.498** (-2.24) f 7.57*** 9.37*** 𝑹𝑹𝒂𝒂𝒂𝒂𝒂𝒂𝟐𝟐 0.90 0.90 hausman 18.71*** 17.69** nxt 287 287 ***, **, and * indicate the rejection of null hypothesis at 1%, 5%, and 10% respectively. t statistics are given in parentheses. democracy, governance, and environmental policy effectiveness different indices are estimated to test whether the results on democracy and governance quality are robust. the political rights index published by freedom house is used as an indicator of the quality of democracy. the political rights index measures the quality of the electoral process, political pluralism and participation, and the functioning of the government (freedom house, 2023). another index used as an indicator of democracy is the voice and accountability index of worldwide governance indicators published by the world bank. this index measures the extent to which citizens can participate in elections, freedom of association, freedom of expression and freedom of the media (world bank, 2023b). higher levels of indexes indicate higher levels of democracy. two different indicators were used as measures of governance quality. the first is the arithmetic average of government effectiveness, rule of law, and control of corruption indices of the world bank's world governance indicators. these indices, which took values between -2.5 and +2.5, were normalized between 0-1 and their natural logarithms were taken. the second governance indicator is the arithmetic average of the corruption, bureaucracy quality, and law and order indices obtained from the international country risk guide database published by the prs group (2023). the index value was normalized between 0-1 and used in logarithmic form. robustness check results are shown in table 7. the effect of democracy indicators is positive and significant at the 1% level. similar results were obtained for the governance quality indices. the effect of the icrg index is significant at the 5% level, while the wgi index is significant at the 1% level. as a result, it is seen that the quality of democracy and governance is strictly robust in terms of environmental policy success. while the effect of gdp per capita confirms the results in the opposite direction of the ekc hypothesis in the estimations made with democracy indices, it becomes insignificant in estimations made with governance quality indices. therefore, it is seen that the inverted-u-shaped relationship between income level and the effectiveness of environmental policies is not robust. estimates of control variables show that the positive effect of education on the effectiveness of environmental policies is robust. the coefficient of education for all models is positive and significant at the 1% level. the effect of population density is also seen to be positive and significant in all models. this result indicates that economies of scale are strongly effective in the success of environmental policies. while the effect of urbanization is positive and significant in the estimations made with the democracy indices and the wgi governance index, it is statistically insignificant in the estimation made only with the icrg index. there is no significant relationship between energy prices and the success of environmental policies in all models estimated. 114 dokuzoğlu and güzel table 7. robustness check 𝑬𝑬𝑬𝑬𝑬𝑬 gdp 1.9733** (2.06) 1.9288* (1.95) 0.6350 (0.64) 0.6884 (0.70) gdp2 -0.1090** (-2.30) -0.1053** (-2.16) -0.0444 (-0.90) -0.0459 (-0.94) ep 0.1033 (0.83) 0.0057 (0.45) 0.0042 (0.31) -0.0033 (-0.26) edu 0.5247*** (2.90) 0.4821*** (2.61) 0.5637*** (2.83) 0.5716*** (2.91) urb 1.8338*** (3.13) 1.3299** (2.26) 0.7834 (1.25) 1.0350* (1.70) pd 0.6952*** (3.96) 0.5976*** (3.38) 0.4802** (2.56) 0.5488*** (2.87) dem𝒇𝒇𝒇𝒇 0.4555*** (6.14) dem𝒘𝒘𝒈𝒈𝒆𝒆 0.7414*** (5.17) gov𝒆𝒆𝒊𝒊𝒆𝒆𝒈𝒈 0.3259** (2.12) gov𝒘𝒘𝒈𝒈𝒆𝒆 0.4550*** (2.69) constant -21.1146*** (-3.75) -16.3993*** (-2.96) -7.0672 (-1.30) -8.7751 (-1.61) f 10.36*** 8.58*** 5.01*** 5.45*** 𝑹𝑹𝒂𝒂𝒂𝒂𝒂𝒂𝟐𝟐 0.93 0.93 0.92 0.92 ***, **, and * indicate the rejection of null hypothesis at 1%, 5%, and 10% respectively. t statistics are given in parentheses. 5. discussion the study demonstrates that democracy and government quality indicators have a positive effect on environmental policy performance. besides, the findings are not sensitive to alternative measures and, therefore, are robust. the study has results similar to those of other studies in the existing literature in many respects. to summarize, democracy improves environmental policy effectiveness as in neumayer, gates and gleditsch (2002), pellegrini and gerlagh (2006), deacon (2009), bättig and bernauer (2009), mukherjee and chakraborty (2013). democracy is often associated with sustainable development, public accountability and participation opportunities (baker & jehlička, 1998; scholte, 2002). in democracies, the standards for environmental protection are quite high. environmental law is effectively enforced with the rule of law based on democracy (bosselmann, 2009). democracy also enables civil society on environmental issues. as the level of democracy increases, governments care about citizens' environmental concerns more and allow them to act together. (dalton and rohrschneider, 2002). participation and empowerment of civil society are essential for sustainability (munslow and ekoko, 1995). as public pressures increase, governments follow policies more effectively in more democratic regimes. the findings regarding democracy differ from the findings of adam and tsarsitalidou (2019). the authors concluded that democracy negatively affects the effectiveness of environmental policies. they explained these findings, which democracy, governance, and environmental policy effectiveness were incompatible with the literature, with differences in income levels between countries and showed that an interaction term consisting of per capita income level and democracy variables turned the relationship positive. our findings suggest that democracy has a robust positive effect. although the non-robust results of adam and tsarsitalidou (2019) that higher corruption is associated with higher policy effectiveness, our study confirms that the quality of government increases the success of environmental policies similar to kelleher, ki and chang (2009), harring (2014), mavragani, nikolaou and tsagarakis (2016), and chang and hao (2017). government quality ensures political freedom and effective provision of public goods such as environmental protection (la porta et al., 1999). government quality prevents environmental intervention caused by market failure from turning into government failure, such as regulatory capture (becker, 1983; helm, 2010). policy performance also suffers as unaccountable governments lack public support (anderson and tverdova, 2003). therefore, environmental policies cannot be expected to be effective if compliance with environmental measures is low. the finding on the positive effect of education on the effectiveness of environmental policies is in line with bimonte (2002), farzin and bond (2006), and kelleher, kim and chang (2009). accordingly, we can claim that education is related to compliance with environmental rules, more participation in the policy-making process and more environmental awareness. furthermore, education enables environmental activist groups to have more supporters, so it is easier to create public pressure to influence the government to protect the environment (dalton & rohrschneider, 2002). in the study, urbanization increases policy effectiveness, confirming the economies of scale in line with mazzanti and zoboli (2009), adam and tsarsitalidou (2019). the proposition that population density creates economies of scale, which means more recycling through urbanization, is also supported empirically (mazzanti & zoboli, 2009). in addition, urbanization makes it easier for individuals with common environmental concerns to come together and take action (dalton & rohrschneider, 2002). limitations an important limitation of the study is related to the data structure. although sustainable governance indicators provide multidimensional measurement, the time dimension of the data set is relatively short. democracy and government quality, which are the main elements of the analysis, change in the relatively longer term. this leads to a methodological limitation. according to hill et al. (2020), the basis of the fixed effects model is changing data characteristics over time. since the time dimension of our study is relatively short in the context of institutional change, the information it provides for analysis is limited. these constraints may be reviewed in future studies utilizing panels with longer time periods. 6. conclusions the factors affecting the success of environmental policies are a frequently discussed topic in the literature. at this point, the number of studies investigating the role of democracy and government effectiveness is limited. in addition, studies investigating the effectiveness of environmental policies 116 dokuzoğlu and güzel in the literature mostly focused on a single environmental degradation indicator. however, the effectiveness of environmental policies is a broader issue. a successfully implemented environmental policy is expected to lead to improvements in many indicators of environmental degradation. therefore, more comprehensive measurements are needed. in this study, the role of democracy and government quality in environmental policy performance was investigated with the sustainable governance indicators dataset. empirical results show that both democracy and good governance are positively associated with environmental policy performance, and the relationship is robust under alternative measurements. according to the empirical results, democracy and the quality of the government have a significant role in the success of environmental policy. implementing and enforcing environmental policies and ensuring that they are based on community needs and objectives require democratic governance systems and effective government institutions. a more effective democracy will enable individuals and organizations with high environmental awareness in society to express their demands to the government effectively. they can also create political pressure to use public resources to solve environmental problems. based on the findings, strengthening democratic institutions is recommended. key stakeholders in the development of democracy are parliaments and governments. parliaments can contribute to the more effective implementation of environmental policies by creating an institutional structure that will ensure better representation of citizens and promote transparency and accountability. governments should respond to the democratic demands of voters regarding environmental concerns. they should also improve the quality of governance. thus, more effective use of public resources in the fight against environmental degradation and climate change can be achieved. moreover, the rule of law should be established, bureaucratic efficiency should be increased, and corruption should be reduced. these measures reduce the private interest-based actions of politicians and bureaucrats and enable public interest policies, such as reducing environmental degradation, to come to the fore more. environmental policies combined with good governance can also be implemented more restrictively. in this context, country-specific studies can be conducted to make more specific policy recommendations. future studies can examine individual environmental policies of countries and conduct comparative analysis following improvements in measurement methods. references acheampong, a.o., dzator, j., & savage, d.a. (2021). renewable energy, co2 emissions and economic growth in sub-saharan africa: does institutional quality matter?. journal of policy modeling, 43(5), 1070-1093. https://doi.org/10.1016/j.jpolmod.2021.03.011 adam, a., & tsarsitalidou, s. (2019). environmental policy efficiency: measurement and determinants. economics of governance, 20, 1-22. https://doi.org/10.1007/s10101-018-0219-y adam, a., delis, m.d., & kammas, p. (2014). fiscal decentralization and public sector efficiency: evidence from oecd countries. economics of governance, 15, 17-49. https://doi.org/10.1007/s10101013-0131-4 afonso, a., schuknecht, l., & tanzi, v. (2005). public sector efficiency: an international comparison. public choice, 123(3-4), 321-347. https://doi.org/10.1007/s11127-005-7165-2 afonso, a., schuknecht, l., & tanzi, v. (2010). public sector efficiency: evidence for new eu member states and emerging markets. applied economics, 42(17), 2147-2164. https://doi.org/10.1080/ 00036840701765460 https://doi.org/10.1016/j.jpolmod.2021.03.011 https://doi.org/10.1007/s10101-018-0219-y https://doi.org/10.1007/s10101-013-0131-4 https://doi.org/10.1007/s10101-013-0131-4 https://doi.org/10.1007/s11127-005-7165-2 https://doi.org/10.1080/00036840701765460 https://doi.org/10.1080/00036840701765460 democracy, governance, and environmental policy effectiveness aidt, t.s. (1998). political internalization of economic externalities and environmental policy. journal of public economics, 69(1), 1-16. https://doi.org/10.1016/s0047-2727(98)00006-1 anderson, c.j., & tverdova, y.v. (2003). corruption, political allegiances, and attitudes toward government in contemporary democracies. american journal of political science, 47(1), 91-109. https://doi.org/10.1111/1540-5907.00007 baker, s., & jehlička, p. (1998). dilemmas of transition: the environment, democracy and economic reform in east central europe-an introduction. environmental politics, 7(1), 1-26. https://doi.org/10.1080/09644019808414370 bättig, m.b., & bernauer, t. (2009). national institutions and global public goods: are democracies more cooperative in climate change policy?. international organization, 63(2), 281-308. https://doi.org/10.1017/s0020818309090092 becker, g.s. (1983). a theory of competition among pressure groups for political influence. the quarterly journal of economics, 98(3), 371-400. https://doi.org/10.2307/1886017 bertelsmann stiftung (2023). sustainable governance indicators. https://www.sgi-network.org/2022/ data. bimonte, s. (2002). information access, income distribution, and the environmental kuznets curve. ecological economics, 41(1), 145-156. https://doi.org/10.1016/s0921-8009(02)00022-8 bosselman, k. (2009). earth democracy: institutionalizing sustainability and ecological integrity. k. bosselmann & l. westra (eds), democracy, ecological integrity and international law, (91-114). newcastle: cambridge scholars publishing. carraro, c., & metcalf, g.e. (2000). behavioral and distributional effects of environmental policy introduction (no. 7648). nber. https://doi.org/10.3386/w7648 chang, c.p., & hao, y. (2017). environmental performance, corruption and economic growth: global evidence using a new data set. applied economics, 49(5), 498-514. https://doi.org/10.1080/ 00036846.2016.1200186 coase, r. (1960). the problem of social cost. the journal of law & economics, 3, 1-44. https://doi.org/10.1086/466560 cole, m.a., rayner, a.j., & bates, j.m. (1997). the environmental kuznets curve: an empirical analysis. environment and development economics, 2(4), 401-416. https://doi.org/10.1017/ s1355770x97000211 congleton, r.d. (1992). political institutions and pollution control. the review of economics and statistics, 74(3), 412-421. https://doi.org/10.2307/2109485 dalton, r.j., & rohrschneider, r. (2002). political action and the political context: a multi-level model of environmental activism. d. fuchs, e. roller & b.webe1s (eds), bürger und demokratie in ost und west, (pp. 333-350). springer. https://doi.org/10.1007/978-3-322-89596-7_18 damania, r. (2002). environmental controls with corrupt bureaucrats. environment and development economics, 7(3), 407-427. https://doi.org/10.1017/s1355770x02000256 damania, r., fredriksson, p.g., & list, j.a. (2003). trade liberalization, corruption, and environmental policy formation: theory and evidence. journal of environmental economics and management, 46(3), 490-512. https://doi.org/10.1016/s0095-0696(03)00025-1 deacon, r. t. & norman, c.s. (2006). does the environmental kuznets curve describe how individual countries behave?. land economics, 82(2), 291-315. https://doi.org/10.3368/le.82.2.291 deacon, r.t. (2009). public good provision under dictatorship and democracy. public choice, 139, 241262. https://doi.org/10.1007/s11127-008-9391-x easty, d.c., & porter, m.e. (2005). national environmental performance: an empirical analysis of policy results and determinants. environment and development economics, 10(4), 391-434. https://doi.org/10.1017/s1355770x05002275 ekardt, f. (2022). economic evaluation, cost-benefit analysis, economic ethics. springer: cham, switzerland. https://doi.org/10.1007/978-3-030-99284-2 ekardt, f., bärenwaldt, m. & heyl, k. (2022). the paris target, human rights, and ipcc weaknesses: legal arguments in favour of smaller carbon budgets. environments, 9(112). https://doi.org/10.3390/environments9090112 ekardt, f., günther, p., hagemann, k., garske, b., heyl, k. & weyland, r. (2023). legally binding and ambitious biodiversity protection under the cbd, the global biodiversity framework, and human rights law. environmental sciences europe, 35(80). https://doi.org/10.1186/s12302-02300786-5 fao (2022a). world food and agriculture statistical yearbook. fao-food & agriculture organization of the united nation, rome. fao (2022b). the state of world fisheries and aquaculture 2022. towards blue transformation. faofood & agriculture organization of the united nation, rome. farzin, y.h., & bond, c.a. (2006). democracy and environmental quality. journal of development economics, 81(1), 213-235. https://doi.org/10.1016/j.jdeveco.2005.04.003 https://doi.org/10.1016/s0047-2727(98)00006-1 https://doi.org/10.1111/1540-5907.00007 https://doi.org/10.1080/09644019808414370 https://doi.org/10.1017/s0020818309090092 https://doi.org/10.2307/1886017 https://www.sgi-network.org/2022/data https://www.sgi-network.org/2022/data https://doi.org/10.1016/s0921-8009(02)00022-8 https://doi.org/10.3386/w7648 https://doi.org/10.1080/00036846.2016.1200186 https://doi.org/10.1080/00036846.2016.1200186 https://doi.org/10.1086/466560 https://doi.org/10.1017/s1355770x97000211 https://doi.org/10.1017/s1355770x97000211 https://doi.org/10.2307/2109485 https://doi.org/10.1007/978-3-322-89596-7_18 https://doi.org/10.1017/s1355770x02000256 https://doi.org/10.1016/s0095-0696(03)00025-1 https://doi.org/10.3368/le.82.2.291 https://doi.org/10.1007/s11127-008-9391-x https://doi.org/10.1017/s1355770x05002275 https://doi.org/10.1007/978-3-030-99284-2 https://doi.org/10.3390/environments9090112 https://doi.org/10.1186/s12302-023-00786-5 https://doi.org/10.1186/s12302-023-00786-5 https://doi.org/10.1016/j.jdeveco.2005.04.003 118 dokuzoğlu and güzel fiorino, d.j. (2011). explaining national environmental performance: approaches, evidence, and implications. policy sciences, 44, 367-389. https://doi.org/10.1007/s11077-011-9140-8 fredriksson, p.g., & svensson, j. (2003). political instability, corruption and policy formation: the case of environmental policy. journal of public economics, 87(7-8), 1383-1405. https://doi.org/10.1016/s0047-2727(02)00036-1 freedom house (2023). freedom in the world. https://freedomhouse.org/report/freedom-world. fullerton, d., hong, i., & metcalf, g.e. (2001). a tax on output of the polluting industry is not a tax on pollution: the importance of hitting the target. in behavioral and distributional effects of environmental policy (pp. 13-44). university of chicago press. https://doi.org/10.7208/ chicago/9780226094809.003.0002 grossman, g.m., & krueger, a.b. (1995). economic growth and the environment. the quarterly journal of economics, 110(2), 353-377. https://doi.org/10.2307/2118443 harring, n. (2014). corruption, inequalities, and the perceived effectiveness of economic proenvironmental policy instruments: a european cross-national study. environmental science & policy, 39, 119-128. https://doi.org/10.1016/j.envsci.2013.08.011 helm, d. (2010). government failure, rent-seeking, and capture: the design of climate change policy. oxford review of economic policy, 26(2), 182-196. https://doi.org/10.1093/oxrep/grq006 hepburn, c. (2010). environmental policy, government, and the market. oxford review of economic policy, 26(2), 117-136. https://doi.org/10.1093/oxrep/grq016 heyl, k., & ekardt, f. (2022). barriers and methodology in transitioning to sustainability: analysing web news comments concerning animal-based diets. journal of cleaner production, 330, 1-13. https://doi.org/10.1016/j.jclepro.2021.129857 hill, t.d., davis, a.p., roos, j.m., & french, m.t. (2020). limitations of fixed-effects models for panel data. sociological perspectives, 63(3), 357-369. https://doi.org/10.1177/0731121419863785 holtz-eakin, d., & selden, t.m. (1995). stoking the fires? co2 emissions and economic growth. journal of public economics, 57(1), 85-101. https://doi.org/10.1016/0047-2727(94)01449-x huber, r.a., wicki, m.l., & bernauer, t. (2020). public support for environmental policy depends on beliefs concerning effectiveness, intrusiveness, and fairness. environmental politics, 29(4), 649-673. https://doi.org/10.1080/09644016.2019.1629171 hwang, j., & akdede, s. h. (2011). the influence of governance on public sector efficiency: a cross-country analysis. the social science journal, 48(4), 735-738. https://doi.org/10.1016/j.soscij.2011 .04.002 imf (2023). primary commodity prices. https://www.imf.org/en/research/commodity-prices. kelleher, d., kim, g.s., & chang, y. j. (2009, june). do differences in political institutions explain differences in environmental policy performance across countries? in appam-kdi international conference on environmental policy and teaching methods, seoul. kirchgässner, g., & schneider, f. (2003). on the political economy of environmental policy. public choice, 115(3-4), 369-396. https://doi.org/10.1023/a:1024289627887 la porta, r., lopez-de-silanes, f., shleifer, a. & vishny, r. (1999). the quality of government. the journal of law, economics, and organization, 15(1), 222-279. https://doi.org/10.1093/jleo/15.1.222 li, k., fang, l., & he, l. (2020). the impact of energy price on co2 emissions in china: a spatial econometric analysis. science of the total environment, 706, 135942. https://doi.org/10.1016/j.scitotenv .2019.135942 mavragani, a., nikolaou, i.e., & tsagarakis, k.p. (2016). open economy, institutional quality, and environmental performance: a macroeconomic approach. sustainability, 8(7), 601. https://doi.org/10.3390/su8070601 mazzanti, m., & zoboli, r. (2009). municipal waste kuznets curves: evidence on socio-economic drivers and policy effectiveness from the eu. environmental and resource economics, 44, 203-230. https://doi.org/10.1007/s10640-009-9280-x morley, b. (2012). empirical evidence on the effectiveness of environmental taxes. applied economics letters, 19(18), 1817-1820. https://doi.org/10.1080/13504851.2011.650324 mukherjee, s., & chakraborty, d. (2013). is environmental sustainability influenced by socioeconomic and sociopolitical factors? cross-country empirical evidence. sustainable development, 21(6), 353371. https://doi.org/10.1002/sd.502 munslow, b., & ekoko, f.e. (1995). is democracy necessary for sustainable development?. democratization, 2(2), 158-178. https://doi.org/10.1080/13510349508403434 nerudova, d., & solilova, v. (2016). the efficiency of environmental policy: empirical evidence based on the application of vec model. engineering economics, 27(5), 527-537. https://doi.org/ 10.5755/j01.ee.27.5.13394 neumayer, e., gates, s. & gleditsch, n. p. (2002). environmental commitment, democracy and inequality: a background paper to the world development report 2003 (no. 27847). the world bank. https://doi.org/10.1007/s11077-011-9140-8 https://doi.org/10.1016/s0047-2727(02)00036-1 https://doi.org/10.7208/chicago/9780226094809.003.0002 https://doi.org/10.7208/chicago/9780226094809.003.0002 https://doi.org/10.2307/2118443 https://doi.org/10.1016/j.envsci.2013.08.011 https://doi.org/10.1093/oxrep/grq006 https://doi.org/10.1093/oxrep/grq016 https://doi.org/10.1016/j.jclepro.2021.129857 https://doi.org/10.1177/0731121419863785 https://doi.org/10.1016/0047-2727(94)01449-x https://doi.org/10.1080/09644016.2019.1629171 https://doi.org/10.1016/j.soscij.2011.04.002 https://doi.org/10.1016/j.soscij.2011.04.002 https://www.imf.org/en/research/commodity-prices https://doi.org/10.1023/a:1024289627887 https://doi.org/10.1093/jleo/15.1.222 https://doi.org/10.1016/j.scitotenv.2019.135942 https://doi.org/10.1016/j.scitotenv.2019.135942 https://doi.org/10.3390/su8070601 https://doi.org/10.1007/s10640-009-9280-x https://doi.org/10.1080/13504851.2011.650324 https://doi.org/10.1002/sd.502 https://doi.org/10.1080/13510349508403434 https://doi.org/10.5755/j01.ee.27.5.13394 https://doi.org/10.5755/j01.ee.27.5.13394 democracy, governance, and environmental policy effectiveness nicolli, f., mazzanti, m. & iafolla, v. (2012). waste dynamics, country heterogeneity and european environmental policy effectiveness. journal of environmental policy & planning, 14(4), 371-393. https://doi.org/10.1080/1523908x.2012.719694 oates, w., & baumol, w. (1975). the instruments for environmental policy. e.s. mills (ed), economic analysis of environmental problems, (pp. 95 132). cambridge: nber. oecd (2020). building back better: a sustainable, resilient recovery after covid-19. oecd publishing. panayotou, t. (1997). demystifying the environmental kuznets curve: turning a black box into a policy tool. environment and development economics, 2(4), 465 484. https://doi.org/10.1017 /s1355770x97000259 pellegrini, l., & gerlagh, r. (2006). corruption, democracy, and environmental policy: an empirical contribution to the debate. the journal of environment & development, 15(3), 332-354. https://doi.org/10.1177/1070496506290960 pigou, a.c. (1932). the economics of welfare. london: macmillan. rajkumar, a.s., & swaroop, v. (2008). public spending and outcomes: does governance matter?. journal of development economics, 86(1), 96-111. https://doi.org/10.1016/j.jdeveco.2007.08.003 rivera-batiz, f.l. (2002). democracy, governance, and economic growth: theory and evidence. review of development economics, 6(2), 225-247. https://doi.org/10.1111/1467-9361.00151 schiller, c., hellmann, t., & paulini, p. (2022). sustainable governance indicators 2022. bertelsmann stiftung, https://www.sgi-network.org/2022/publications. scholte, j.a. (2002). civil society and democracy in global governance. global governance, 8(3), 281-304. https://doi.org/10.1163/19426720-00803004 steinebach, y. (2022). instrument choice, implementation structures, and the effectiveness of environmental policies: a cross-national analysis. regulation & governance, 16(1), 225-242. https://doi.org/10.1111/rego.12297 stern, d.i. & common, m.s. (2001). is there an environmental kuznets curve for sulfur?. journal of environmental economics and management, 41(2), 162-178. https://doi.org/10.1006/jeem.2000.1132 stern, d.i. (2020). how large is the economy-wide rebound effect?. energy policy, 147, 1-7. https://doi.org/10.1016/j.enpol.2020.111870 tanzi, v. (1998). government role and the efficiency of policy instruments. p.b. sørensen (ed), public finance in a changing world, (pp.51-69). london: palgrave macmillan. https://doi.org/10.1007/ 978-1-349-14336-8_3 the prs group (2023). international country risk guide. https://www.prsgroup.com/explore-ourproducts/icrg/. torras, m., & boyce, j.k. (1998). income, inequality, and pollution: a reassessment of the environmental kuznets curve. ecological economics, 25(2), 147-160. https://doi.org/10.1016/s09218009(97)00177-8 unctad (2021). trade and environment review 2021. the united nations conference on trade and development, new york. undp (2023). human development index. https://hdr.undp.org/data-center/documentation-anddownloads. ward, h. (2008). liberal democracy and sustainability. environmental politics, 17(3), 386-409. https://doi.org/10.1080/09644010802055626 wawrzyniak, d., & doryń, w. (2020). does the quality of institutions modify the economic growth-carbon dioxide emissions nexus? evidence from a group of emerging and developing countries. economic research-ekonomska istraživanja, 33(1), 124-144. https://doi.org/10.1080/1331677x.2019 .1708770 wolf, m.j., emerson, j.w., esty, d.c., sherbinin, a.d., & wendling, z.a. (2022). 2022 environmental performance index (epi) results. new haven, ct: yale center for environmental law & policy. world bank (2023a). world development indicators. https://databank.worldbank.org/source/worlddevelopment-indicators. world bank (2023b). worldwide governance indicators. https://databank.worldbank.org/source/ worldwide-governance-indicators. wtcc (2022). travel & tourismeconomic impact august 2022. world travel tourism council, london. https://doi.org/10.1080/1523908x.2012.719694 https://doi.org/10.1017/s1355770x97000259 https://doi.org/10.1017/s1355770x97000259 https://doi.org/10.1177/1070496506290960 https://doi.org/10.1016/j.jdeveco.2007.08.003 https://doi.org/10.1111/1467-9361.00151 https://www.sgi-network.org/2022/publications https://doi.org/10.1163/19426720-00803004 https://doi.org/10.1111/rego.12297 https://doi.org/10.1006/jeem.2000.1132 https://doi.org/10.1016/j.enpol.2020.111870 https://doi.org/10.1007/978-1-349-14336-8_3 https://doi.org/10.1007/978-1-349-14336-8_3 https://www.prsgroup.com/explore-our-products/icrg/ https://www.prsgroup.com/explore-our-products/icrg/ https://doi.org/10.1016/s0921-8009(97)00177-8 https://doi.org/10.1016/s0921-8009(97)00177-8 https://hdr.undp.org/data-center/documentation-and-downloads https://hdr.undp.org/data-center/documentation-and-downloads https://doi.org/10.1080/09644010802055626 https://doi.org/10.1080/1331677x.2019.1708770 https://doi.org/10.1080/1331677x.2019.1708770 https://databank.worldbank.org/source/world-development-indicators https://databank.worldbank.org/source/world-development-indicators https://databank.worldbank.org/source/worldwide-governance-indicators https://databank.worldbank.org/source/worldwide-governance-indicators 1. introduction 2. literature review 3. model, data, and methodology 4. empirical results 5. discussion 6. conclusions references ©the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 1 (2025), pages 55-70 https://doi.org/10.17979/ejge.2025.14.1.11356 submitted: oct 10, 2024 accepted: apr 25, 2024 published: jun 20, 2025 article the influence of economic factors on sustainable energy consumption: a comparative approach between the oecd and western balkan countries atdhetar gara,1 argjira bilalli,1,* artenisa beka 1 1 faculty of business and economics, south east european university, tetovo, north macedonia *correspondence: ab31379@seeu.edu.mk abstract. this study investigates the influence of economic factors—such as energy use, foreign direct investment, inflation, gdp per capita, trade, population growth, and unemployment—on sustainable energy consumption (sec) and renewable energy consumption (rec) across 38 oecd and 6 western balkan countries from 2010 to 2022. adopting a comparative approach, the research sheds light on how varying economic contexts shape energy sustainability outcomes in developed and emerging economies. using panel data from the world bank and the global economy database, we estimate regression models (ols, fixed effects, random effects) and employ the generalized method of moments (gmm). we accounted for potential endogeneity with the use of gmm and verified it with arellano-bond and sargan tests. the analysis found that gdp per capita and the openness of a country to trade were positively connected to sec. also, inflation and unemployment can serve as disincentives for the use of renewable energy. the findings provide insights for effective energy sustainable development thinking, especially given the economic development and goals for sustainable energy outcomes. keywords: sustainable energy consumption; economic growth; inflation; foreign direct investment jel classification: b22 ; q4 ; q43 1. introduction due to the effects of climate change and depleting petroleum reserves, sustainable energy systems must be used worldwide. while there has been a historical transition between the main energy sources, these changes have often lasted more than a century and were prompted by advances in technology, labor costs that were prohibitive, and resource shortages (solomon & krishna, 2011). in addition, national and international seminars, and conferences adequately address and explore the significance of renewable energy for environmental preservation, sustainable development, and diversification of the energy supply. the authors consider energy sustainable if its net effects on the biosphere do not significantly degrade its capabilities for supporting existing species in their current abundance and diversity, this is a very conservative definition, one requiring that such energy be less https://creativecommons.org/licenses/by-nc/4.0/ 56 gara, bilalli and beka harmful to the environment than most forms in use now. global climate change threatens people's health, economic prospects, and basic food and water sources. additionally, people could have false beliefs about the properties of various energy sources and how they affect the environment. a wide range of changes in household energy behavior is needed to realize a sustainable energy transition. the scope of subjects covered by sustainable energy is remarkably extensive. therefore, in addition to energy and environmental technology, sustainable energy must also take into account the political, social, and economic aspects of human lifestyles (tester et al., 2012). sustainable energy solutions mean better efficiency, better cost-effectiveness, better resources used, better design and analyses, better energy security, and a better environment (dincer & zamfirescu, 2011). it's critical to comprehend the variables influencing the acceptance of energy system modifications and policies. therefore, considering the role of sustainable energy in all spectrums of life especially in the economy, this study aims to investigate its effects with a rigorous methodological approach. accept the variables selected namely, energy use per capita (eup), foreign direct investment (fdi), inflation (inf), gdp per capita (gdpc), trade (tr), population growth (pog), unemployment (une) and sustainable energy consumption (sec). future research papers could be extended by including some other economic factors, such as the efficiency of the institution's performance, taking into account the fact that the mechanism of the institutions is considered fundamentally in terms of the policies in energy and their application, as well as additional groups of counties in order to measure the influence of such economic factors on sustainable energy consumption as a permanent source of life. the contribution of this study to the literature is multifaceted. first, it provides a clear comparison of the impact of economic factors on sustainable energy consumption between two groups of countries: the western balkans and oecd countries. this comparison is made by focusing on economic factors, development contexts, and policy implications. while existing research on sustainable energy consumption exists, there is limited literature directly comparing oecd countries (which are generally more economically developed) with the western balkan countries (which are transitioning economies). these groups have distinct economic structures, energy consumption patterns, and sustainability goals, making the comparative approach unique. furthermore, many studies on sustainable energy consumption primarily focus on environmental or technological factors, often overlooking the crucial role of economic variables. by focusing specifically on economic factors, this study highlights how national economic conditions influence energy use patterns and sustainability. the study also addresses the gap in understanding how a country's development stage affects its energy consumption choices and sustainable practices. the economic context of the western balkans, with its infrastructure challenges and economic transition, contrasts with the more stable, industrialized economies of the oecd. by examining both groups, the study offers valuable insights into how policies might be tailored for developed versus developing countries, helping to shape more targeted energy policies that consider economic realities. furthermore, it provides guidelines for policymakers to create more effective sec-related policies. the sections of this paper are arranged as follows: except for the first section, which unfolds the introduction part, the second one provides reviews of the literature, and the third section contains the research methodology and data of the panel evidence extracted from reliable sources for western balkan and oecd countries. further, section four provides the results generated using an appropriate methodology and discussion arising from the relevant topic. section five provides the the influence of economic factors on sustainable energy consumption discussion and findings and the last section provides the conclusion given by the authors. additionally, the main argument for this study is that only a few studies have examined the influence of economic factors on sustainable energy consumption. moreover, this study is characterized by using the comparative approach between two groups of counties, the methodology applied, the time period, and the variables included in the sample. consequently, there is a gap in the literature that this study aims to fill. 2. literature review since the middle of the 20th century, the growth of industry and population has resulted in a massive increase in the need for energy. as a result, countries worldwide are now prioritizing the development of new and renewable energy sources. many empirical studies, conducted by various authors, and categorized by various countries and periods, examine the influence of economic factors on sustainable energy consumption. renewable energy consumption on economic growth can be either positive, negative, or not significant. renewable energy consumption should contribute to economic growth (yildirim et al., 2012). renewable energy consumption has positive impacts on economic growth (alper et al., 2016; shahbaz et al., 2020; lin & moubarak, 2014; inglesi-lotz, 2016). the positive effect of renewable energy consumption on economic growth for oecd countries was also examined by wang & wang (2020). however, this positive relationship varies as the threshold values change, indicating a nonlinear role in increasing the use of renewable energy to promote economic development for these countries. the significant impact of renewable energy consumption on economic growth in the balkans and black sea is recognized by koçak & şarkgüneşi (2017). in germany, a 1% increase in renewable energy consumption boosts german economic growth by 0.2194% (rafindadi & ozturk, 2017). chen et al. (2020) found that renewable energy consumption has no significant effect on economic growth in developed countries, but it has a positive and significant effect on economic growth in oecd countries. ocal & aslan(2013), for the region of turkey, examined the negative impact of renewable energy consumption on economic growth, applying empirical tests from the ardl approach. however, it is worth noting that developed countries have passed the phase of increasing the consumption of renewable energy compared to less developed countries, which are still in the initiation phase. however, moving from non-renewable energy to sustainable sources requires substantial investments, eu countries should pay closer attention to investment in research and development in order to sustain the plan for long-term advancement in sustainable power sources for feasible energy and economic development. (adedoyin et al., 2020). sadiq et al. (2023), using the autoregressive distributed lag (ardl) indicated that economic growth, fdi, inflation, and population growth were positively associated with sec in china. by applying a panel ardl estimations (qamruzzaman & jianguo, 2020) confirmed that in the long run, financial development, trade openness, and foreign capital flow positively influence renewable energy consumption. the analysis included three subgroups of countries namely, for low-income countries, middle-income countries, and upper-middle-income panels. akpanke et al. (2023) found out that gdp in both short-run and long-run does not significantly impact re while, fdi and public https://www.sciencedirect.com/topics/earth-and-planetary-sciences/balkans 58 gara, bilalli and beka sector credit, have a significant positive effect on re use in the long run, moreover, inflation rate and broad money have a significant negative long-run and significant positive short-run effect on re use also emphasizing that poor-financial resource countries must encourage fdi inflows to promote re development. (khandker et al., 2018) examined the relationship between fdi and renewable energy consumption in the context of bangladesh applying johansen’s cointegration test confirming that variables are cointegrated in the long run and the granger causality test reveals that there is a bidirectional causality between respective variables also, through vector error correction model (vecm), we found no causality between the variables in the short run. many studies have discussed that foreign direct investments (fdi) potentially increase the production and dissemination of renewable energy since foreign firms have different experiences regarding environmental awareness in the host country. pao et al. (2011) emphasize that fdi can help reduce co2 emissions by promoting the adoption of cleaner technologies, particularly in the context of renewable energy. apergis and payne (2010) highlight the impact of fdi on renewable energy consumption and discuss how foreign investments can encourage energy efficiency and cleaner energy solutions in the u.s., also, this study suggests that fdi might not always contribute positively to renewable energy consumption, especially if it leads to investments in energy-intensive sectors. it highlights how fdi can sometimes be directed toward industries that increase reliance on non-renewable energy. authir stern (2004) discusses the link between trade liberalization, industrial development, and energy use. it argues that trade can sometimes contribute to higher energy consumption if environmental policies are not sufficiently strong. ozturk and acaravci (2010) analyze the relationship between energy consumption, economic growth, and inflation in turkey. it suggests that inflationary pressures can have mixed effects on energy consumption depending on the stage of economic development and the type of energy sources in use. stern (2011) discusses how economic factors such as inflation, trade, and fdi interact with energy consumption. while there is an assumption that these factors generally promote economic growth, the relationship with sustainable energy consumption is more complex, and in some cases, may be negative or insignificant, particularly when the economic growth model is energy-intensive. after the wide discussion of different works of literature, we set our hypothesis relevant to our research as follows: h1: economic factors have a positive impact on sustainable energy consumption h2: there are significant differences in terms of energy consumption between bp and oecd countries. 3. methodology the impact of economic factors on sustainable energy consumption is measured using panel data encompassing 6 western balkan countries and 38 oecd countries spanning the period from 2010 to 2022, totaling 13 years. methodologically, various panel data models are employed, including pooled ols, ols robust, fixed effects (fe), and random effects (re). to determine the most reliable model between fixed and random effects, the hausman test is applied for comparative purposes. ensuring objective, dependable, and efficient estimations of this coefficient requires the use of an appropriate the influence of economic factors on sustainable energy consumption estimation method. also, the methodological approach is applied showing a comparison between the respective countries. furthermore, the analysis utilizes the system gmm estimator, developed by arellano and bover in 1995, which combines a regression in differences. this particular estimator aims to mitigate inconsistencies and inefficiencies observed in alternative panel estimators, as established by arellano and bond in 1991, blundell and bond in 1998, and blundell and bond in 2000. its validity hinges on the instruments' appropriateness and the assumption that the differenced error terms lack second-order serial correlation—both prerequisites for the consistency of the gmm estimator. to validate these assumptions, the sargan test of overidentifying limitations has already been conducted, assessing the overall reliability of the instruments by examining the sample analog of the moment conditions applied in the estimation technique. 3.1 model specification we specify a log-linear model for the study to estimate the responses of economic factors on sustainable energy consumption. the study uses a macroeconomic series that consists of yearly observations between 2010-2022. the specification of the models treating the variables chosen to highlight this impact takes the form as follows: sec𝑖𝑖𝑖𝑖 = 𝑐𝑐 + 𝛽𝛽1 (sec𝑖𝑖𝑖𝑖−1) + 𝛽𝛽2 (eup) + 𝛽𝛽3 (fdi) + 𝛽𝛽4 (inf) + 𝛽𝛽5 (gdpc) + 𝛽𝛽6 (tr) + 𝛽𝛽7 (pog) + 𝛽𝛽8 (une) + 𝑢𝑢𝑖𝑖𝑖𝑖 where sec is the dependable variable, which in this case is sustainable energy consumption, i = 1….44 (countries), t = 2010…2022 (years); c is constant; the explanatory variables include: 𝑦𝑦𝑖𝑖𝑖𝑖−1, which is the first lagged of the dependent variable, eup (energy use per capita); fdi (foreign direct investment); inf (inflation); gdpc (growth domestic product per capita); tr (trade); pog (population growth); une (unemployment) and uit is the error term. table 1. definition of variables. variable abbreviations unit source sustainable energy consumption sec million kwh the global economy database energy use per capita eup kwh world bank database foreign direct investment fdi % of gdp world bank database inflation inf % world bank database gdp per capita gdpc us $ world bank database trade tr % of gdp world bank database population growth pog % world bank database unemployment une % world bank database 3.2 data’ our panel contains annual data, obtained from the world bank (wbd) and the global economy database. the dataset covers a balanced panel of 6 western balkan countries (kosovo, albania, north macedonia, bosnia and herzegovina, montenegro, and serbia) and 38 oecd countries (colombia, turkey, mexico, costa rica, chile, poland, hungary, greece, latvia, slovak republic, lithuania, 60 gara, bilalli and beka portugal, czechia, estonia, slovenia, spain, korea, rep., italy, japan, france, united kingdom, new zealand, germany, israel, belgium, canada, austria, finland, netherlands, australia, sweden, denmark, iceland, the united states, norway, switzerland, ireland, and luxembourg), over the period 2010-2022. the dependent variable of energy consumption lies at the heart of the challenge of meeting sustainable consumption of energy resources. the influence of independent variables in this situation is of great depth. for example, increased foreign direct investment (fdi) can bring innovative and efficient technologies for energy production and use. high fdi can encourage the use of renewable resources and improved energy consumption practices. inflation also has an impact on energy consumption. when inflation rises, there is often an increase in energy prices, which can lead to a decrease in total consumption. on the other hand, an increase in gross domestic product (gdp) per capita is often accompanied by an overall increase in energy consumption, because economic growth often needs more energy. trade also has an impact on energy consumption. improvements in trade and further international linkages may lead to overall increases in energy needs for transport, generation, and distribution. at the same time, population growth and rising unemployment can create new challenges, affecting the needs and the way energy is used in different communities and countries. 4. empirical findings and discussion . this chapter presents the empirical results. first, descriptive statistics for all 44 nations are presented. next, a comparison between the results of the descriptive statistics in the western balkans (6) and the oecd countries (38) is provided. in the context of the data analysis for the 44 countries of the study (table 2), some important dynamics can be observed in relation to energy and economic aspects. on average, sustainable energy consumption reaches 27089.6 million kilowatt hours, an important result that indicates the global use of sustainable energy resources. in this context, the use of energy per capita, which is about 3579.26 kwh, brings to attention the need for the use of energy at individual levels and the efficiency of its use. beyond the aspect of energy use, the data show significant links between economic aspects and energy consumption. foreign direct investments (fdi) constitute about 3.84% of the gross domestic product (gdp), reflecting the confidence and commitment of foreign investors in these economies. at the same time, the unemployment average of 9.35% shows the challenges in the labor market and the opportunities for the use of energy in the context of the economic development of these countries. the influence of economic factors on sustainable energy consumption table 2. descriptive statistics for all sampled countries. variable obs mean std. dev. min max sec 572 27089.60 60780.43 115 468931 eup 572 3579.26 2765.35 432 18432 fdi 572 3.84 22.46 -391.43 234.46 inf 572 2.68 4.22 -1.73 72.3 gdpc 529 34092.02 25456.67 3009.52 133590.15 tr 572 100.89 57.30 23.38 388.12 pog 572 .423 .82 -2.25 2.67 une 570 9.35 6.42 2.02 35.41 source: author’s calculation after presenting the descriptive statistics for all the countries together, table 3 shows the results separated for the countries of the western balkans and the oecd countries. based on the results presented in this table it can be seen that the oecd countries have much higher average sustainable energy consumption, reaching 30897 million kilowatt hours, compared to only 2976 for the western balkan countries. this shows a large difference in the use of sustainable energy resources between these two groups of countries. another fundamental difference is in energy use per capita. the average in the countries of the western balkans is 1488.2 kwh, while in the oecd countries, it is much higher, about 3909.43 kwh per individual. this shows that, on average, individuals in oecd countries use more energy than those in the western balkans. regarding foreign direct investment (fdi) as a percentage of gdp, the western balkan countries have a higher average, about 6.3%, compared to the average of 3.451% in oecd countries. this shows that, as a percentage of gdp, the economies of the western balkans are more open to foreign investments compared to oecd countries. in terms of unemployment, the countries of the western balkans have a much higher average, about 20.269%, while in oecd countries, it is lower, about 7.629%. this big difference indicates a more difficult situation in the labor market for the countries of the western balkans compared to the oecd countries. these differences in levels of energy consumption, individual energy use, foreign investment, and unemployment highlight the differences in economic development and use of energy resources between the two groups of countries. table 3. comparison of results in western balkan countries and oecd countries. variable western balkans oecd obs. mean std.dev obs. mean std.dev sec 78 2976 2362.42 494 30897 64586 eup 78 1488.2 660.683 494 3909.43 2826.29 fdi 78 6.3 3.694 494 3.451 24.107 inf 78 2.757 3.459 494 2.678 4.341 gdpc 74 5655.9 1443.74 455 38716.8 24496.9 tr 78 97.165 19.675 494 101.478 61.156 pog 78 -0.403 0.636 494 0.554 0.78 une 78 20.269 7.044 492 7.629 4.251 source: author’s calculation 62 gara, bilalli and beka as we analyze the connections between sustainable energy consumption and other variables (table 4), we can recognize a broad panorama of the relationships between them. a weak, positive correlation between sustainable energy consumption and per capita energy use indicates that always with a low impact, individual levels of energy use can be related to global sustainable energy consumption. however, the direct impact of individual use on this consumption appears to be weak. changes in foreign direct investment (fdi) do not show a significant impact on sustainable energy consumption, as shown by the weak negative correlation between these two variables. this weak link may present a paradox, suggesting that, despite the increase in foreign investment, their contribution to the use of sustainable energy in these countries is limited. on the other hand, the weak negative correlation between sustainable energy consumption and inflation levels suggests that the impact of rising inflation on sustainable energy use is minimal. a similar trend appears between levels of sustainable energy consumption and gdp per capita, indicating that individual wealth growth has a consistent relationship with sustainable energy consumption. also, it appears that a higher percentage of trade in gdp is associated with a lower level of sustainable energy consumption, as shown by the negative correlation expressed between them. this may reflect a lower use of energy resources through trade in goods and services, marking an interesting dynamic between sustainable energy use and trade activity. table 4. correlation analysis. variables (1) (2) (3) (4) (5) (6) (7) (8) (1) sec 1.000 (2) eup 0.140 1.000 (3) fdi -0.068 -0.028 1.000 (4) inf -0.049 -0.034 -0.003 1.000 (5) gdpc 0.190 0.514 0.076 -0.221 1.000 (6) tr -0.298 0.063 0.291 -0.105 0.322 1.000 (7) pog 0.009 0.319 0.040 0.101 0.504 0.001 1.000 (8) une -0.147 -0.374 0.014 0.011 -0.502 -0.079 -0.425 1.000 source: author’s calculation table 5 reports the results of the gmm model in the western balkans and oecd countries. the findings demonstrate the thorough specification of all estimated dynamic panel models. the sargan test for identifying constraints (obtained from the findings of the second phase) is acknowledged as a reliable tool. it supports the claim that there is no relationship between the instrument variables and the residuals. in the first order, the arellano-bond test of ar (1) and ar (2) is rejected; in the second order, it is approved. none of the two presented models suffers from multicollinearity and heteroskedasticity, where these two problems were tested by the vif test and the breusch pagan test. based on the results of the gmm model presented in table 4, the impact of sustainable energy use has a positive impact on the first difference of lag in the countries of the western balkans (b=0.256) and oecd (0.796). where both coefficients are statistically significant at the 1% level. foreign direct investments have a positive impact on the use of sustainable energy consumption, in the countries of the western balkans with coefficient b=0.035 and oecd b=0.067, where the coefficients are statistically significant at the level of 5% and 1%. foreign direct investment (fdi) positively impacts sustainable energy consumption by financing renewable energy projects, the influence of economic factors on sustainable energy consumption introducing advanced energy-efficient technologies, and improving environmental standards. fdi also facilitates the transfer of green technologies, creates green jobs, and encourages collaboration between the private sector and governments to implement sustainable energy policies. it can lead to positive spillover effects, where local businesses adopt sustainable practices, accelerating the transition to cleaner energy sources and contributing to a more sustainable energy future. the standard of living also has a positive and statistically significant impact, where in the countries of the western balkans the coefficient is b=0.124 (statistically significant at the 5% level), while in the oecd countries, the impact is positive b=0.156 (statistically significant at the level of 1%). higher living standards often correlate with greater access to resources and technologies that promote energy efficiency and renewable energy use. as individuals' income levels rise, they are more likely to invest in energy-efficient appliances, sustainable housing, and green technologies. wealthier nations or populations tend to have more robust infrastructure and policies that support the transition to clean energy, such as incentives for renewable energy adoption and investments in energy-saving technologies. this leads to an overall increase in sustainable energy consumption as the standard of living improves. further, the last variable with a positive and significant impact on the increase in the use of sustainable energy consumption is trade, where in the countries of the western balkans (b=0.016) it is statistically significant at the level of 5%, while in the oecd countries s b=0.028, statistically significant at the 1% level. while countries engage in international trade, they often import and adopt energy-efficient technologies and renewable energy systems that might not be locally available. trade stimulates competition and innovation, encouraging countries to adopt more sustainable practices to stay competitive in the global market. trade agreements also incentivize the development of green industries, supporting the transition to cleaner energy sources and ultimately leading to higher sustainable energy consumption. on the other hand, population growth has a negative impact on the use of sustainable energy consumption both in the countries of the western balkans (b=-0.348) and in the oecd countries (b=-0.036), where both coefficients show a statistically significant impact at the 5% level. population growth has a negative impact on sustainable energy consumption because as the population increases, so does the overall demand for energy. this means higher energy consumption, which, if not paired with the adoption of renewable energy sources, results in greater reliance on nonrenewable and environmentally harmful energy sources like coal and oil. rapid population growth can also strain existing infrastructure, making it more difficult to implement energy-efficient technologies or renewable energy systems. in many developing regions, the focus may be more on meeting basic energy needs rather than prioritizing sustainability, which can hinder efforts to promote clean energy consumption. 64 gara, bilalli and beka table 5. results of the gmm model for western balkan countries and oecd countries. variable western balkans oecd coef std. err p-value coef std. err p-value lag of sec 0.256*** 0.035 0.000 0.796*** 0.032 0.000 eup -0.001 0.002 0.367 -0.006 0.002 0.309 fdi 0.035** 0.012 0.021 0.067 0.002*** 0.000 inf -0.018 0.03 0.544 -0.013 0.013 0.992 gdpc 0.124** 0.06 0.017 0.156*** 0.003 0.000 tr 0.016** 0.007 0.008 0.028*** 0.003 0.002 pog -0.348** 0.016 0.023 -0.036** 0.011 0.027 une 0.047 0.039 0.22 -0.024* 0.014 0.092 constant 5.118 9.216 0.579 2.065 0.527 0.000 mean dependent var 7.75 9.15 number of obs 68 417 sd dependent var 0.659 1.591 chi-square 72351.22 831489.16 arellano – bond test for ar (1) 0.032 0.037 arellano – bond test for ar (2) 0.94 0.098 sargan 0.454 0.752 vif 2.67 1.51 hettest 0.1526 0.4521 source: author’s calculation. notes: * p<0.1, ** p<0.05, *** p<0.01 the sargan test, with p-values of 0.454 for the western balkans and 0.752 for the oecd countries, indicates that the instruments used in the gmm model are valid. this test assesses whether the instruments are exogenous, i.e., independent of model errors, and the null hypothesis is that the instruments are valid. since both p-values are significantly above the usual threshold of 0.05, there is no statistical evidence to reject the null hypothesis, suggesting that the selected instruments (such as variable lags or other factors) are not correlated with the error terms and are appropriate to address endogeneity in the model. this strengthens the reliability of the gmm results for both groups of countries, supporting the interpretation of the effects of economic variables on sustainable energy consumption (sec). variance inflation factor (vif) with values of 2.67 for the western balkans and 1.51 for the oecd countries indicates the level of multicollinearity between the independent variables in the model. a vif below 5 suggests that multicollinearity is not a serious problem for both groups of countries, with the oecd showing an even lower level (1.51), close to the total absence of strong correlation between the variables. this means that the regression coefficients are reliable and are not affected by excessive overlap between economic factors such as fdi, gdpc, and tr, supporting the stability of the gmm results. the heteroskedasticity test (hettest) with p-values of 0.1526 for the western balkans and 0.4521 for the oecd tests whether the variance of the errors is constant. since both p-values are above 0.05, there is no statistical evidence to reject the null hypothesis of homoskedasticity, meaning the influence of economic factors on sustainable energy consumption the errors are relatively uniform for both groups. this strengthens the reliability of the model, as methods like gmm assume consistent errors to provide accurate estimates. 5. discussion comparing the outcomes of our study with findings from various authors provides a comprehensive perspective on the intricate relationship between economic factors and sustainable energy consumption. firstly, our investigation aligns with numerous previous studies indicating that renewable energy consumption positively impacts economic growth. this assertion is supported by works from yildirim et al. (2012), alper et al. (2016), shahbaz et al. (2020), and others, emphasizing the constructive link between renewable energy use and economic development. similarly, our results portray positive impacts of sustainable energy consumption on economic growth in both western balkans and oecd countries, echoing the assertions made by wang & wang (2020) and koçak and şarkgüneşi (2017) regarding the significant influence of renewable energy on economic advancement. however, contrasting views also exist in the literature. for instance, chen et al. (2020) found no significant effect of renewable energy consumption on economic growth in developed countries, while ocal and aslan (2013) indicated a negative impact in turkey. this divergence highlights the complexity of these relationships and the potential variability across different economies and stages of development. in terms of foreign direct investment (fdi), our study concurs with insights suggesting a positive impact on sustainable energy consumption, as noted by akpanke et al. (2023) and khandker et al. (2018). the observed positive effects of fdi on sustainable energy use align with the notion that foreign investments could contribute to the development and adoption of renewable energy sources due to the varied experiences and environmental awareness brought by foreign firms. moreover, our findings regarding trade and its negative correlation with sustainable energy consumption complement the suggestions made by qamruzzaman and jianguo (2020) regarding the potential for lower energy resource utilization through trade activities. this indicates an interesting dynamic that merits further exploration concerning the interplay between trade patterns and sustainable energy consumption. additionally, the negative impact of population growth on sustainable energy consumption, consistent in both the western balkans and oecd countries, echoes the findings by sadiq et al. (2023), underscoring the challenges posed by population dynamics in achieving sustainable energy goals. based on the extensive analysis and empirical findings presented, we test the hypotheses stated at the outset of the study. h1: economic factors have a positive impact on sustainable energy consumption the empirical evidence from the analysis supports this hypothesis to a certain extent. variables such as foreign direct investment, gdp per capita, and trade exhibit positive and statistically significant impacts on sustainable energy consumption in both western balkan countries and oecd nations. 66 gara, bilalli and beka these findings suggest that economic indicators indeed play a role in influencing and potentially promoting sustainable energy consumption within these economies. h2: there are significant differences in terms of energy consumption between bp and oecd countries the results strongly support this hypothesis. the comparison between western balkan countries and oecd countries reveals substantial disparities in sustainable energy consumption and energy use per capita. oecd countries generally demonstrate higher levels of both measures compared to western balkan countries. additionally, factors such as foreign direct investment and unemployment display notable variations between these groups, indicating distinct economic landscapes and likely differing approaches to energy consumption and development strategies. considering the statistical significance of the coefficients obtained from the gmm model across both groups of countries, the empirical evidence confirms the influence of various economic factors on sustainable energy consumption. moreover, the contrasting impacts of population growth on energy consumption, negative in both groups, further emphasize the intricate relationship between demographic factors and sustainable energy usage. the results obtained from the comprehensive analysis provide robust support for the hypotheses, underlining the influence of economic factors on sustainable energy consumption and substantiating the significant disparities in energy consumption between western balkan and oecd countries. the advantage of comparing the oecd countries and western balkan countries lies in the distinct economic, developmental, and energy contexts of these two groups. oecd countries are typically more economically developed, with established infrastructures, advanced energy policies, and a stronger focus on sustainability. western balkan countries are in transition, facing challenges such as developing energy-efficient infrastructures, shifting from fossil fuels to renewable sources, and managing economic growth while ensuring environmental sustainability. by comparing these two groups, the research highlights how economic factors such as foreign direct investment, trade, and inflation affect sustainable energy consumption differently in developed versus transitioning economies. this comparison provides valuable insights into the specific needs, barriers, and opportunities faced by each group, helping to identify tailored policy recommendations and strategies for promoting sustainable energy consumption in both contexts. findings can reveal how development stages, energy consumption patterns, and economic structures influence the effectiveness of energy policies and the transition to renewable energy sources. 6. conclusions the objective of this paper was to empirically examine the influence of economic factors on sustainable energy consumption for western balkan and oecd countries using econometric methods. results were generated by applying stata software as a data processing tool, despite the opposition of the comparative approach between the two respective groups of countries (western balkan and oecd). in terms of the methodology, the gmm model, assuming that the differenced error terms do not display the second-order serial correlation and the validity of the instruments, reveals that variables in the sample have a significant relationship. moreover, to verify these presumptions, the influence of economic factors on sustainable energy consumption we apply the sargan test of overidentifying limitations, which examines the sample analog of the moment conditions utilized in the estimating technique to assess the overall validity of the instruments. based on the results of the gmm model the impact of sustainable energy use has a positive impact on the first difference of lag in the countries of the western balkans and oecd where both coefficients are statistically significant at the 1% level. foreign direct investments have a positive impact on the use of sustainable energy consumption in the countries of the western balkans, based on the coefficients b = 0.035 and oecd b = 0.067, with both coefficients statistically significant at the 5% and 1% levels. additionally, the standard of living also has a positive and statistically significant impact, where in the countries of the western balkans the coefficient is b=0.124 (p<0.05), while in the oecd countries, the impact is positive b=0.156 (p<0.01). trade have positive and statistically significant impacts on sustainable energy consumption in western balkan and oecd countries, where in the countries of the western balkans (b=0.016) it is statistically significant at the level of 5%, while in the oecd countries b=0.028, statistically significant at the 1% level. however, population growth has a negative impact on the use of sustainable energy consumption in both groups of countries. energy is recognized as a fundamental necessity for both social and economic development, and by extension, for sustainable development. furthermore, energy policy plays a crucial role in shaping energy systems, particularly in promoting the development of renewable or sustainable energy. therefore, it is challenging for businesses specializing in re to enter and operate in the energy market. (cai & aoyama, 2018). this paper highlights the significance of policy implications and institutional efficiency in enhancing sustainable energy consumption. the dynamics of the energy market are often disrupted due to factors such as the lack of administrative authority in decentralized systems, which limits the creation of an effective framework for renewable energy regulation and implementation. it is important to recognize a significant distinction between developing and developed countries in the context of renewable energy and energy efficiency. while some developing countries are making earnest efforts to explore and promote renewable energy technologies, they are also striving to enhance energy efficiency by seeking collaboration with more advanced, developed countries. these partnerships aim to leverage expertise and resources to accelerate the adoption of sustainable energy solutions. the results from the gmm model highlight several key economic factors that positively impact sustainable energy consumption in both the western balkans and oecd countries, underscoring the importance of targeted government policies to enhance these dynamics. for example, the positive influence of foreign direct investment (fdi) on sustainable energy consumption suggests that governments can attract fdi by creating favorable policies such as tax incentives, subsidies, and streamlined regulatory frameworks that encourage private sector investment in renewable energy technologies. the statistically significant impact of the standard of living on sustainable energy use implies that governments should focus on improving living standards, particularly through policies that promote energy efficiency and access to clean energy. additionally, the positive correlation between trade and sustainable energy consumption emphasizes the importance of trade policies that foster international cooperation in energy markets, supporting the exchange of renewable energy technologies and resources. 68 gara, bilalli and beka negative impact of population growth on sustainable energy use points to the need for governments to implement policies that address the increased demand for energy through innovations like energy efficiency measures, smart grid technologies, and the expansion of renewable energy infrastructure. in both regions, energy policy plays a crucial role in shaping these outcomes, and governments should develop long-term strategies that incorporate these economic factors. this could include setting renewable energy targets, encouraging energy diversification, and fostering an institutional environment conducive to the sustainable energy transition. by aligning government policies with these economic drivers, governments can more effectively promote sustainable energy consumption and move toward a more sustainable future. references adedoyin, f. f., bekun, f. v., & alola, a. a. (2020). growth impact of transition from non-renewable to renewable energy in the eu: the role of research and development expenditure. renewable energy, 159, 1139–1145. https://doi.org/10.1016/j.renene.2020.06.015 akpanke, t. a., deka, a., ozdeser, h., & seraj, m. (2023). does foreign direct investment promote renewable energy use? an insight from west african countries. renewable energy focus, 44, 124–131. https://doi.org/10.1016/j.ref.2022.11.007 alper, a., & oguz, o. (2016). the role of renewable energy consumption in economic growth: evidence from asymmetric causality. renewable and sustainable energy reviews, 60, 953–959. https://doi.org/10.1016/j.rser.2016.01.123 apergis, n., & payne, j. e. (2010). energy consumption and growth in south america: evidence from a panel error correction model. energy economics, 32(6), 1421-1426. arellano, m., & bover, o. (1995). another look at the instrumental variable estimation of error-components models. journal of econometrics, 68(1), 29–51. https://doi.org/10.1016/0304-4076(94)01642-d arellano, m., & bond, s. (1991). some tests of specification for panel data: monte carlo evidence and an application to employment equations. the review of economic studies, 58(2), 277–297. https://doi.org/10.2307/2297968 blundell, r., & bond, s. (1998). initial conditions and moment restrictions in dynamic panel data models. journal of econometrics, 87(1), 115–143. https://doi.org/10.1016/s0304-4076(98)00009-8 blundell, r., & bond, s. (2000). gmm estimation with persistent panel data: an application to production functions. econometric reviews, 19(3), 321–340. https://doi.org/10.1080/07474930008800475 cai, y., & aoyama, y. (2018). fragmented authorities, institutional misalignments, and challenges to renewable energy transition: a case study of wind power curtailment in china. energy research & social science, 41, 71–79. https://doi.org/10.1016/j.erss.2018.04.021 chen, c., pinar, m., & stengos, t. (2020). renewable energy consumption and economic growth nexus: evidence from a threshold model. energy policy, 139, 111295. https://doi.org/10.1016/j.enpol.2020.111295 dincer, i., & zamfirescu, c. (2011). sustainable energy systems and applications. new york, ny: springer science & business media. https://doi.org/10.1007/978-0-387-95861-3 inglesi-lotz, r. (2016). the impact of renewable energy consumption to economic growth: a panel data application. energy economics, 53, 58–63. https://doi.org/10.1016/j.eneco.2015.01.003 khandker, l. l., amin, s. b., & khan, f. (2018). renewable energy consumption and foreign direct investment: reports from bangladesh. journal of accounting, 8(3), 72–87. koçak, e., & şarkgüneşi, a. (2017). the renewable energy and economic growth nexus in black sea and balkan countries. energy policy, 100, 51–57. https://doi.org/10.1016/j.enpol.2016.10.007 lin, b., & moubarak, m. (2014). renewable energy consumption-economic growth nexus for china. renewable and sustainable energy reviews, 40, 111–117. https://doi.org/10.1016/j.rser.2014.07.128 ocal, o., & aslan, a. (2013). renewable energy consumption-economic growth nexus in turkey. renewable and sustainable energy reviews, 28, 494–499. https://doi.org/10.1016/j.rser.2013.08.036 https://doi.org/10.1016/j.renene.2020.06.015 https://doi.org/10.1016/j.ref.2022.11.007 https://doi.org/10.1016/j.rser.2016.01.123 https://doi.org/10.1016/0304-4076(94)01642-d https://doi.org/10.2307/2297968 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1080/07474930008800475 https://doi.org/10.1016/j.erss.2018.04.021 https://doi.org/10.1016/j.enpol.2020.111295 https://doi.org/10.1016/j.eneco.2015.01.003 https://doi.org/10.1016/j.enpol.2016.10.007 https://doi.org/10.1016/j.rser.2014.07.128 https://doi.org/10.1016/j.rser.2013.08.036 the influence of economic factors on sustainable energy consumption ozturk, i., & acaravci, a. (2010). co₂ emissions, energy consumption and economic growth in turkey. renewable and sustainable energy reviews, 14(9), 3220–3225. https://doi.org/10.1016/j.rser.2010.07.005 pao, h. t., yu, h. c., & yang, y. h. (2011). modeling the co₂ emissions, energy use, and economic growth in russia. energy, 36(8), 5094–5100. https://doi.org/10.1016/j.energy.2011.06.004 qamruzzaman, m., & jianguo, w. (2020). the asymmetric relationship between financial development, trade openness, foreign capital flows, and renewable energy consumption: fresh evidence from panel nardl investigation. renewable energy, 159, 827–842. https://doi.org/10.1016/j.renene.2020.06.069 rafindadi, a. a., & ozturk, i. (2017). impacts of renewable energy consumption on the german economic growth: evidence from combined cointegration test. renewable and sustainable energy reviews, 75, 1130–1141. https://doi.org/10.1016/j.rser.2016.11.093 sadiq, m., ou, j. p., duong, k. d., van, l., & bui, t. x. (2023). the influence of economic factors on the sustainable energy consumption: evidence from china. economic research–ekonomska istraživanja, 36(1), 1751–1773. https://doi.org/10.1080/1331677x.2022.2093244 shahbaz, m., raghutla, c., chittedi, k. r., jiao, z., & vo, x. v. (2020). the effect of renewable energy consumption on economic growth: evidence from the renewable energy country attractive index. energy, 207, 118162. https://doi.org/10.1016/j.energy.2020.118162 solomon, b. d., & krishna, k. (2011). the coming sustainable energy transition: history, strategies, and outlook. energy policy, 39(11), 7422–7431. https://doi.org/10.1016/j.enpol.2011.09.009 stern, d. (2004). the rise and fall of the environmental kuznets curve. world development, 32(8), 1419– 1439. https://doi.org/10.1016/j.worlddev.2004.03.004 stern, d. i. (2011). the role of energy in economic growth. annals of the new york academy of sciences, 1219(1), 26–51. https://doi.org/10.1111/j.1749-6632.2010.05921.x tester, j. w., drake, e. m., driscoll, m. j., golay, m. w., & peters, w. a. (2012). sustainable energy: choosing among options. cambridge, ma: mit press. wang, q., & wang, l. (2020). renewable energy consumption and economic growth in oecd countries: a nonlinear panel data analysis. energy, 207, 118200. https://doi.org/10.1016/j.energy.2020.118200 yildirim, e., saraç, ş., & aslan, a. (2012). energy consumption and economic growth in the usa: evidence from renewable energy. renewable and sustainable energy reviews, 16(9), 6770–6774. https://doi.org/10.1016/j.rser.2012.09.004 https://doi.org/10.1016/j.rser.2010.07.005 https://doi.org/10.1016/j.energy.2011.06.004 https://doi.org/10.1016/j.renene.2020.06.069 https://doi.org/10.1016/j.rser.2016.11.093 https://doi.org/10.1080/1331677x.2022.2093244 https://doi.org/10.1016/j.energy.2020.118162 https://doi.org/10.1016/j.enpol.2011.09.009 https://doi.org/10.1016/j.worlddev.2004.03.004 https://doi.org/10.1111/j.1749-6632.2010.05921.x https://doi.org/10.1016/j.energy.2020.118200 https://doi.org/10.1016/j.rser.2012.09.004 70 gara, bilalli and beka appendix a: granger causality test results western balkans oecd f-stat (p-value) f-stat (p-value) fdi → sec 3.45 (0.034)* 4.12 (0.019)* sec → fdi 1.23 (0.298) 0.89 (0.415) gdpc → sec 2.87 (0.045)* 5.67 (0.004)** sec → gdpc 0.95 (0.387) 1.45 (0.236) tr → sec 3.12 (0.029)* 3.89 (0.023)* sec → tr 1.67 (0.192) 1.02 (0.367) pog → sec 2.54 (0.067) 2.98 (0.055) sec → pog 0.78 (0.456) 0.65 (0.523) the granger test indicates the possible direction of causality between economic variables and sustainable energy consumption (sec) for both groups of countries. in the western balkans, fdi (p = 0.034), gdp per capita (gdpc, p = 0.045), and trade (tr, p = 0.029) “granger-cause” sec at the 5% significance level, suggesting that these variables can predict changes in sec. however, there is no evidence of reverse causality (sec → fdi, gdpc, or tr), as the p-values are high (p > 0.05), indicating that sec does not affect these economic factors. for population growth (pog), the result is borderline significant (p = 0.067), leaving open the possibility of a weak negative impact, consistent with your findings. for oecd countries, the results are similar but stronger: fdi (p = 0.019), gdpc (p = 0.004), and trade (p = 0.023) “granger-cause” sec with statistical significance, at the 5% and 1% levels, respectively, reinforcing the idea that these economic factors are drivers of sustainable energy consumption. as in the western balkans, there is no evidence of reverse causality from sec to these variables (p > 0.05). pog also shows a marginal effect (p = 0.055), suggesting a possible but not statistically significant negative impact. 1. introduction 2. literature review 3. methodology 4. empirical findings and discussion 5. discussion 6. conclusions references appendix a: the role of automatic stabilizers in business cycle: the case of indonesia european journal of government and economics 10(2), december 2021, 167-184 european journal of government and economics issn: 2254-7088 the role of automatic stabilizers in business cycle: the case of indonesia haryo kuncoro a* a faculty of economics, state university of jakarta, indonesia * corresponding author at: har_kun@feunj.ac.id abstract. the use of large fiscal stimulus packages to dampen the impact of covid-19 recently has raised concerns about the effectiveness of the discretionary fiscal policy. this paper aims at analysing the feasibility of automatic fiscal stabilisers to mitigate economic fluctuations in the case of indonesia. using the imf standard model for quarterly data over the period of 2001(1) to 2019(4), we find that the role of automatic fiscal stabilisers is getting greater both in revenue and spending. this implies that the automatic fiscal stabilisers are feasible as the main fiscal policy instrument for economic stability goals in the future. however, given the existing circumstances, indonesia has to reform economic, regulatory, and institutional ecosystems in adopting the automatic fiscal stabilisers. keywords. automatic stabilisers; business cycle; discretionary fiscal policy; government spending; taxes. jel codes. e32; e62; h61; h62. doi. https://doi.org/10.17979/ejge.2021.10.2.7366 1. introduction following the coronavirus disease (covid-19) outbreak around the world in late 2019, fiscal policy has received much attention. many advanced countries adopt the fiscal stimulus packages relative to monetary policy to survive the adverse impacts emerging from covid-19. for example, as of october 2020, member countries of g-20 announced that the fiscal stimulus packages ranged from 7 percent of gdp in china to 13 percent of gdp in the us, and more than 21 percent of gdp in japan (szmigiera, 2020). while in developed countries, the role of automatic fiscal stabilisers is very important, it is likely less prominent in developing countries (debrun and kapoor, 2010). to mitigate the negative impacts of the covid-19 pandemy, emerging market countries rely on the fiscal stimulus packages only. the main cause is the revenue to gdp and expenditure to gdp ratios are far smaller than that in advanced countries. on the revenue side, the tax base is smaller so that the share of income-elastic taxes is smaller. on the expenditure side, there are few automatic stabilisers in developing countries (cornia, 2006). for academic circumstances, the above phenomena are interesting. the main question here is whether fiscal stimulus packages are really effective to stabilise macroeconomic conditions. despite the lag length to affect, such discretionary fiscal policies potentially lead to procyclical. on the one hand, the discretionary fiscal policies which originally are subject to stabilisation can mailto:har_kun@feunj.ac.id https://doi.org/10.17979/ejge.2021.10.2.7366 haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 168 destabilise (gnip, 2011). hence, there is considerable debate in the literature about the efficacy of discretionary fiscal policy in developing countries (see: doraisami, 2013). on the other hand, the automatic stabilisers -programs that automatically scale up in recessions and drawdown during booms to stabilise the economy -play a critical role in fighting every recession (leiserson, 2020). for policymakers, understanding the size and the role of automatic stabilisers is crucial. while automatic stabilisers are a fairly established concept in the fiscal policy literature, there is still no consensus about their actual nature and their effectiveness (veld, larch, and vandeweyer, 2012). this leaves no room for discretionary fiscal policy and highlights the importance of knowing whether automatic stabilisers alone can deliver sufficient stabilisation. reconsidering issues arising from discretionary fiscal policies will also help the authority in emerging markets to avoid output fluctuation better by using a passive fiscal policy (vera, 2016). indonesia is not an exception. since the government officially announced covid-19 for the first time on march 2, 2020, indonesia launched the first two fiscal packages amounting to idr 33.2 trillion (0.2 percent of gdp), the government announced an additional package of idr 405 trillion (2.6 percent of gdp) on march 31, 2020. they were further expanded to idr 677.2 trillion (4.2 percent of gdp) on june 4, 2020, as part of a national economic recovery program. the national economic recovery program has been continuously refined and currently stands at idr 695.2 trillion (unddr, 2020). at present, indonesia only has automatic stabilisers in its taxation through progressive income tax. the country’s corporate income tax, however, is no longer progressive from 2020 with a single rate of 22 percent. on the spending side, the country does not have automatic stabilisers yet, as typically the case in developing economies. the inadequate fiscal space, relatively high population, and low share of public spending as a share of gdp need more time to implement in indonesia at least within the medium term (trihartanto, 2019). the local governments are still heavily reliant on central government fiscal transfer that covers around two-thirds of their budget. their current aggregate own-source revenue merely stands at 2.4 percent of gdp. as such, this time seems to be not the perfect time yet for indonesia to provide such benefits or the time being, indonesia’s fiscal stabilisers still rely on discretionary fiscal measures due to its still limited automatic stabilisers. accordingly, countries with weak automatic stabilisers have enacted larger fiscal stimulus programs (dolls, fuest, and peichl, 2012). regardless of the absence of automatic fiscal stabilisers, indonesia also has fiscal rules discipline (most notably capping fiscal deficit to 3 percent of gdp and debt ratio to 60 percent of gdp adopted from maastricht pact since 2004). it is unallowed to rely heavily on fiscal expansion during the bad times, where unemployment benefits and other social transfers work as automatic stabilisers. however, such a fiscal discipline still effectively proves to make indonesia’s fiscal sound so far. this paper enriches the literature on fiscal policy in developing countries with a focus on indonesia. we measure the automatic fiscal stabiliser and relate it to the discretionary fiscal https://equitablegrowth.org/people/greg-leiserson/ haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 169 policy. then, we attempt to estimate their role and assess their feasibility in managing the economic impact of covid-19. the lessons learnt from indonesia would be useful for other emerging countries to achieve price stabilisation as well as to promote economic growth. the article is organised as follows. in the next section, we briefly present the literature and previous empirical researches both in developed countries and developing countries, including indonesia. the third section describes the dataset and empirical techniques used. the results of the empirical tests are presented in the results and discussion section. in the end, we summarise arguments and key findings in the conclusion section. 2. literature review basically, a country's fiscal policy has three major economic functions, namely allocation, distribution and stabilisation (musgrave and musgrave, 1989). the allocation function is carried out through the state expenditures side, while the distribution and stabilisation functions are mostly executed through taxes and expenditures. in principle, the fiscal policy will transfer resources from the public to the government and then redistribute them to the public with certain considerations (arestis, 1985). at the most basic technical level, the fiscal policy can be categorised into three groups, namely automatic fiscal stabilisation, systematic fiscal discretion, and non-systematic fiscal discretion (see: surjaningsih, utari, and trisnanto, 2012; kuncoro, 2017). while automatic fiscal stabilisation policies and systematic fiscal discretion are aimed at dampening business cycle fluctuations, the non-systematic fiscal discretion is related to changes in government revenues and/or expenditures due to apart from the changes in the economic cycle (fatas and mihov, 2003). empirical studies regarding the three types of fiscal policy, primarily automatic fiscal stabilisers, provide diverging results. on the revenue side, mabbett and schelkle (2007) assessed the stabilisation effect of the tax system. they found that various tax-benefit instruments vary across countries in the eu. however, suescún (2007) finds that the degree of smoothing provided by the automatic revenue stabilisers -described by various properties of the tax system -is negligible in latin america. looking at the type of tax revenue, buettner and fuest (2010) analysed the effectiveness of the corporate income tax as an automatic fiscal stabiliser. by employing a unique firm-level data set of german manufacturers, they found that the stabilisation effect varies over the business cycle and tends to increase during cyclical downturns. in contrast, sen (2013), using granger causality test, shows that in the case of turkey, personal income tax is the most effective tax in stabilising business cycle fluctuations. the effectiveness of personal income tax as an automatic stabiliser is not because of the progressivity of tax, but due to the sensitivity of employment and average wages to gdp fluctuations (krajewsky and pilat, 2017). besides income tax, value-added tax (vat) could be an instrument of automatic fiscal stabiliser. the vat would be more progressive than the payroll tax it replaces because there is no ‘taxable maximum’ on a vat that slashes the effective rate paid by high earners (ritz, 2020). haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 170 a vat is also neutral to the treatment of capital and labour, which means that employers will be based on economic benefits for their investment decisions, rather than solely on favourable tax treatment (nunns and rosenberg, 2016). hence, the vat affects indirectly the employed labour’ income. by comparing us to the eu, dolls, fuest, and peichl (2012) found that, in us, tax-benefit systems absorb a greater proportion of income variation generated by income shock than the unemployment shock. the difference is largely explained by the coverage and generosity of unemployment benefits. automatic stabilisers in the case of an unemployment shock are basically replacement rates for a transition from employment to non-work. rather than work incentives, they reflect how much the tax-benefit system absorbs market income losses due to becoming unemployed or exiting the labour market altogether. as a result, the tax-benefit system reduced inequality of net incomes, and so helped offset the inequality‐increasing impact of growing disparities in gross market incomes (paulus and tasseva, 2020). on the expenditure side, government size -measured by the gdp ratio of government spending -plays the role of an automatic stabiliser in developing countries such as latin america, subject to weak smoothing effect (suescún, 2007). in terms of expenditure type, darby and melitz (2008) argue that ageand health-related social expenditures react to the cycle in a stabilising manner. in contrast, granado, gupta, and hajdenberg (2013) concluded that social spending in developed countries is acyclical, implying the ineffectiveness of social spending as automatic fiscal stabilisers. in the us, the stabilisers affect welfare significantly through the provision of social insurance (mckay and reis, 2016). the macroeconomic literature on automatic fiscal stabilisation tends to focus on taxes and dismiss the relevance of government expenditure except for unemployment compensation. for instance, salgado et al. (2014) analysed the distribution of replacement rates when simulating the unemployment shock in six eu countries due to the great recession. they distinguish between shortand long-term unemployment, and their findings confirmed that higher replacement rates in the short term could lead to serious challenges for minimum income schemes with the consequences of the crisis in the longer term. moreover, poghosyan, senhadji, and cottarelli (2016) found that fiscal transfers smooth regional shocks in three large federation countries: the u.s., canada, and australia. they find that fiscal transfers offset 4-11 percent of idiosyncratic shocks (risk-sharing) and 13-24 percent of permanent shocks (redistribution). this fiscal insurance largely operates through automatic stabilisers embedded in a central budget, primarily through federal taxes and transfers to individuals, rather than transfers from the central government to state budgets. the analysis of the role of automatic fiscal stabilisation in the various forms of social protection in reducing economic shocks has been discussed by estep, ajilore, and madowitz (2019). caldara et al. (2020) also concluded that various social protection programs in the us are effective in reducing the impact of an economic recession. however, the adoption of those various social protection programs for the case of developing countries such as indonesia needs further adjustments due to differences in terms of economic, regulatory, and institutional ecosystems. haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 171 hence, there is a need for further research on this issue, especially from indonesian case. whereas, indonesian fiscal policy tends to be more a-cyclical or even procyclical. the a-cyclicality of fiscal policy leads to the presumption of the role of automatic fiscal stabilisers, while the procyclicality of fiscal policy further leads to the role of fiscal discretion, due to the fact that scholars found that there was no counter-cyclicality in fiscal policy (baldacci, 2009; akitoby et al., 2004). in terms of discretionary fiscal policy, simorangkir and adamanti (2010) evaluated the effectiveness of fiscal stimulus in accordance with the global financial crisis. using a financial computable general equilibrium approach, they revealed the combination of fiscal expansion without monetary policy expansion and monetary expansion without fiscal expansion. they infer that the combination of fiscal and monetary expansion has a large multiplier effect, which increases aggregate demand. on the other hand, surjaningsih, utari, and trisnanto (2012) indicated the absence of discretionary fiscal policy made by the government of indonesia. their study also concluded that short-term adjustment suggests that an increase in government spending has a positive effect on output, while a tax increase has a negative effect. there is a greater influence of government spending on output compared to taxation policies in the short term. therefore, government spending is more effective to stimulate economic growth, especially in times of recession, compared to taxation policies. furthermore, hur, mallick, and park, (2014) found similar results. the fiscal stimulus programs have contributed substantially to developing asia’s countries (including indonesia) faster and stronger than expected recovery from the global financial crisis. basri and raharja (2011) suggest improving the quality of spending in controlling fiscal deficits. to contribute greatly in economic stabilisation, fiscal space can be maintained by converting unproductive spending into productive spending. abdurohman and resosudarmo (2017) investigated the practical behaviour of fiscal policy in indonesia in response to economic cycles. they showed that fiscal policy in indonesia tends to be procyclical. recently, resosudarmo et al. (2020) discovered that fiscal stimulus packages during the global financial crisis had a positive impact on aggregate demand and on poverty prevention, principally via stimulating private consumption. corporate income tax cuts have the largest economic impact in the long run, and cash transfers are the most useful policy tool for alleviating poverty. however, a fiscal stimulus package could have an uneven spatial distributional effect on output across regions, particularly in the short term. according to prior literature, indonesia tends to support discretionary fiscal policy instead of automatic fiscal stabilisers to overcome economic turbulence. in addition, most of the studies above are done in various circumstances. therefore, there is no general conclusion, which requires further investigations. the procyclicality of fiscal policy in indonesia opens room to implement automatic stabilisers alongside fiscal discretion. in particular, designing better automatic stabilisers is one of the most promising routes for better macroeconomic policy (blanchard, dell'ariccia, and mauro, 2010). haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 172 3. analytical model and data the purpose of this section is to develop an analytical framework within which this can be clearly stated as a set of formal propositions. the notion of government budget constraint proposes that the total government spending (exp) should be sufficiently financed by the total domestic revenues (rev). if rev is inadequate to cover exp, the available financing option is debt, resulting in interest payment (irp) in the next period. the overall fiscal balance (i.e. deficits or surpluses) is the difference between rev and exp: 𝑂𝑂𝑂𝑂 = 𝑅𝑅𝑅𝑅𝑅𝑅 − 𝐸𝐸𝐸𝐸𝐸𝐸 [1] when irp is excluded from the total government expenditure, we get the primary fiscal balance (pb): 𝑃𝑃𝑂𝑂 = 𝑅𝑅𝑅𝑅𝑅𝑅 − (𝐸𝐸𝐸𝐸𝐸𝐸 − 𝐼𝐼𝑅𝑅𝑃𝑃). [2a] then, 𝑂𝑂𝑂𝑂 = 𝑃𝑃𝑂𝑂 − 𝐼𝐼𝑅𝑅𝑃𝑃 [2b] looking solely at changes in the fiscal balance can thus be misleading: these movements may give an impression of expansionary (or contractionary) discretionary policy actions, even though the changes are driven by cyclical factors. this is why cyclical adjustment is applied, to filter the impact of cyclical movements on fiscal variables and assess the ‘underlying’ fiscal stance. following the oecd methodology (giorno et al., 1995), the primary fiscal balance may be decomposed into the cyclically adjusted primary balance (capb) and the cyclical primary balance (cpb): 𝑃𝑃𝑂𝑂 = 𝐶𝐶𝐶𝐶𝑃𝑃𝑂𝑂 + 𝐶𝐶𝑃𝑃𝑂𝑂 [3] the cpb is the part of the primary balance that automatically reacts to the cycle. interest payments are often kept separate because of their movements, while ‘automatic’ in the sense of not generally reflecting discretionary fiscal policy actions. they may not be necessarily correlated with cyclical output changes. from [2b] and [3], changes in the ob can be decomposed into: (i) the automatic response of fiscal variables to changes in output; (ii) the response of fiscal variables to changes in discretionary policy; and (iii) changes in interest payments, as follows: ∆𝑂𝑂𝑂𝑂 = ∆𝐶𝐶𝐶𝐶𝑃𝑃𝑂𝑂 + ∆𝐶𝐶𝑃𝑃𝑂𝑂 − ∆𝐼𝐼𝑅𝑅𝑃𝑃 [4] where δ is the difference between two consecutive years, t and t+1 (or the difference relative to a reference year). haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 173 eventually, the automatic stabilisers (as) are defined as the change in the cyclical primary balance: 𝐶𝐶𝐴𝐴 = ∆𝐶𝐶𝑃𝑃𝑂𝑂 = ∆𝑂𝑂𝑂𝑂 − ∆𝐶𝐶𝐶𝐶𝑃𝑃𝑂𝑂 + ∆𝐼𝐼𝑅𝑅𝑃𝑃 [5] the change in the cyclically adjusted primary balance can be derived from cyclically adjusted revenue and expenditure. in particular, the cyclically adjusted component of revenue revca is defined as 𝑅𝑅𝑅𝑅𝑅𝑅𝐶𝐶𝐶𝐶 = 𝑅𝑅 �𝑌𝑌 𝑝𝑝 𝑌𝑌 � 𝜀𝜀𝑅𝑅 [6] where yp is potential output (that is, the maximum output compatible, at any given time, with the absence of unexpected inflation), y is actual output, and εr is the elasticity of revenue with respect to the output gap. similarly, the cyclically adjusted component of expenditure expca is defined as 𝐸𝐸𝐸𝐸𝐸𝐸𝐶𝐶𝐶𝐶 = 𝐸𝐸 �𝑌𝑌 𝑝𝑝 𝑌𝑌 � 𝜀𝜀𝐸𝐸 [7] where εe is the elasticity of expenditure with respect to the output gap. 𝑔𝑔𝑔𝑔𝐸𝐸 = �𝑌𝑌−𝑌𝑌 𝑝𝑝 𝑌𝑌𝑝𝑝 � [8] subtracting [6] and [7], we obtain the cyclically adjusted primary balance 𝐶𝐶𝐶𝐶𝑃𝑃𝑂𝑂 = 𝑅𝑅𝑅𝑅𝑅𝑅 �𝑌𝑌 𝑝𝑝 𝑌𝑌 � 𝜀𝜀𝑅𝑅 − 𝐸𝐸𝐸𝐸𝐸𝐸 �𝑌𝑌 𝑝𝑝 𝑌𝑌 � 𝜀𝜀𝐸𝐸 [9] the cyclically adjusted primary balance is often measured in relation to potential output − the “natural” scaling variable since cyclically adjusted balances measure what the fiscal balance would have been if the output had been at its potential level − as follows: 𝑐𝑐𝑔𝑔𝐸𝐸𝑐𝑐 = 𝑅𝑅𝑅𝑅𝑅𝑅𝐶𝐶𝐶𝐶 𝑌𝑌𝑝𝑝 − 𝐸𝐸𝐸𝐸𝐸𝐸𝐶𝐶𝐶𝐶 𝑌𝑌𝑝𝑝 = 𝑅𝑅𝑅𝑅𝑅𝑅 𝑌𝑌 �𝑌𝑌 𝑝𝑝 𝑌𝑌 � 𝜀𝜀𝑅𝑅−1 − 𝐸𝐸𝐸𝐸𝐸𝐸 𝑌𝑌 �𝑌𝑌 𝑝𝑝 𝑌𝑌 � 𝜀𝜀𝐸𝐸−1 [10a] where gap is the output gap and small letters r and g denote ratios of revenue and expenditure to gdp while small letter for capb denotes the ratio of cyclically adjusted primary balance to potential gdp 𝑐𝑐𝑔𝑔𝐸𝐸𝑐𝑐 = 𝑟𝑟(1 + 𝑔𝑔𝑔𝑔𝐸𝐸)−(𝜀𝜀𝑅𝑅−1) − 𝑅𝑅(1 + 𝑔𝑔𝑔𝑔𝐸𝐸)−(𝜀𝜀𝐸𝐸−1) [10b] haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 174 this expression captures the “structural” primary balance i.e. primary balance not affected by cyclical fluctuations. changes in the capb are often used to estimate the size/cost of discretionary policy. 𝑐𝑐𝐸𝐸𝑐𝑐 = 𝐶𝐶𝐶𝐶𝐶𝐶 𝑌𝑌 = 𝐶𝐶𝐶𝐶 𝑌𝑌 − 𝐶𝐶𝐶𝐶𝐶𝐶𝐶𝐶 𝑌𝑌𝑝𝑝 = 𝐸𝐸𝑐𝑐 − 𝑐𝑐𝑔𝑔𝐸𝐸𝑐𝑐 [11a] 𝑐𝑐𝐸𝐸𝑐𝑐 = 𝑟𝑟(𝜀𝜀𝑅𝑅 − 1)𝑔𝑔𝑔𝑔𝐸𝐸 − 𝑅𝑅(𝜀𝜀𝐸𝐸 − 1)𝑔𝑔𝑔𝑔𝐸𝐸 [11b] where pb and cpb are the primary balance and the cyclical primary balance in percent of actual gdp. the contribution from automatic stabilisers to changes in the overall balance is then given by the change in [11] between two periods. 𝑔𝑔𝑎𝑎 = 𝐶𝐶𝐴𝐴 𝑌𝑌 = ∆𝑐𝑐𝐸𝐸𝑐𝑐 = 𝑐𝑐𝐸𝐸𝑐𝑐𝑡𝑡 − 𝑐𝑐𝐸𝐸𝑐𝑐𝑡𝑡−1 [12] to identify potential output, we adopt hodrick-prescott (hp) filtering method. this method is widely used among macroeconomists to obtain a smooth estimate of the long-term trend component of a series. the method was first used in a working paper (circulated in the early 1980’s and published in 1997) by hodrick and prescott (1997) to analyse postwar us business cycles. technically, the hp filter is a two-sided linear filter that computes the smoothed series τ of y by minimising the variance y of around τt, subject to a penalty that constrains the second difference of τ. the hp filter then chooses s to minimise: ∑ (𝑦𝑦𝑡𝑡 − 𝜏𝜏𝑡𝑡)2𝑇𝑇 1 + 𝜆𝜆∑ [(𝜏𝜏𝑡𝑡+1 − 𝜏𝜏𝑡𝑡) − (𝜏𝜏𝑡𝑡 − 𝜏𝜏𝑡𝑡−1)]2𝑇𝑇−1 2 [13] the penalty parameter λ controls the smoothness of the series τ. the larger the λ, the smoother the τ. as λ = ∞, τ approaches a linear trend. the default value of λ in eviews is set to be 1,600 for quarterly data. for this empirical study, we define rev ∈ {it, vat, others} and exp ∈ {cex, rex, irp}. it is income tax, vat is value added tax, and others is other government revenues. cex is central government expenditures, rex is government transfers to lower-layer governments, and irp is interest payments, respectively. the government revenue is divided into two grand categories, taxes and non-taxes received excluded grants. the term ‘government expenditure’ used in this study is central government general consumption or recurrent expenditure realisation (mostly allocated onto wage/salary and goods/services purchase) and capital expenditure. inclusively, we also assess the spending of transfer to regions. deficits are the difference between government spending and government revenues. the fiscal data are taken from the ministry of finance. the selected key macroeconomic variable is gdp. the gdp is used as the main factor for the government to set the state budget projection for the next year. the gdp data is available on a haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 175 quarter-basis. those variables are presented at the 2010 constant price. price levels are derived from the gdp in current price divided by gdp in constant prices (2010=1). the deflator index is also used to convert all variables into real values. the sample periods chosen for this study extend from 2001(1) to 2019(4). the total observation is 76 sample points. the year 2001 is set as the starting observation related to the implementation of fiscal decentralisation. all of the data are taken from the central bank of indonesia and the central board of statistics. most of the results are calculated in the econometric program eviews 9. price levels are derived from the gdp in current price divided by gdp in constant prices (2010=1). the deflator index is also used to convert all variables into real values. 4. results and discussion table 1 provides descriptive statistics covering mean, median, and extreme (maximum and minimum) values for each variable of interest. the average shares of income tax and other revenue are not far from each other. nevertheless, other revenue is more volatile compared to income tax as presented by the higher standard deviation as well as the range of two extreme values. the share of valueadded tax to total revenue is relatively low but stable indicated by the lowest standard of deviation. the highest kurtosis and the negative sign of skewness suggest that value-added tax has a big potential to be the main automatic fiscal stabiliser tool. central government spending absorbs almost half of total expenditure. this is followed by transfer to regions which takes 37 percent. the rest of total expenditure is allocated to interest payment. special attention should be paid more to the interest payment since this outlay tends to remarkably increase in recent years. while most of the central government expenditure is mandatory spending, the tight fiscal space (represented by an increase in primary balance deficits) is primarily due to the increase in interest payment. table 1. descriptive statistics. revenue expenditure ob ratio it vat others cex rex irp mean 0.3794 0.2418 0.3787 0.4959 0.3676 0.1366 -0.2695 median 0.3856 0.2499 0.3582 0.4939 0.3716 0.1169 -0.2328 maximum 0.5206 0.2999 0.6121 0.6824 0.5356 0.3600 0.5744 minimum 0.1883 0.1142 0.2347 0.2533 0.1962 0.0609 -1.0479 std. dev. 0.0679 0.0350 0.0906 0.1048 0.0813 0.0665 0.0351 skewness -0.3207 -0.9323 0.5231 -0.2041 0.0753 1.6285 -0.3571 kurtosis 2.6405 4.2651 2.4385 2.2779 2.4025 5.3280 2.6592 haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 176 figure 1. output gap and overall balance ratio. figure 1 presents the dynamics of the output gap and overall balance deficit ratio. output gap was low in the early 2000s, associated with the diminishing impacts of the 1997/1998 asian monetary crisis. output gap tended to be high, approximately 4 percent, in accordance with the 2009 global financial crisis. figure 1 shows that fiscal policy in indonesia during the sample observation periods is typically a-cyclical, as found by akitoby et al. (2004) and baldacci (2009). this figure will be confirmed by the statistical result. the overall balance budget ranges from (deficit) -1.05 to (surplus) 0.6 percent of gdp. the lowest deficit ratio took place in 2012 when the commodity boom began. the highest overall deficit ratio occurred in the mid-2010s when the commodity boom ended. it seems that the government revenue is strongly dependent on some external factors. therefore, implementing pro-growth, pro-job, and pro-poor fiscal measures is likely to require an increase in the size of the government revenue. the above results implicitly offer some fundamental implications. the government can use the output gap of plus/minus 5 percent as the preliminary reference for carrying out the active fiscal policy. the active fiscal policy might be conducted through either fiscal discretion or automatic fiscal stabilisers. each of these policies will be described further in the next section. according to sen (2013), granger causality can be explored to test the ability of a particular revenue to be an automatic fiscal stabiliser. table 2 reports the standard granger causality test results for the three types of revenues. since the granger causality test is very sensitive to the selection of lag lengths, the lag lengths are determined by akaike information criteria (aic). standard granger causality test results show that there is a bi-directional causality in all three revenue categories. it means that all revenues prospectively could be an automatic fiscal stabiliser instrument in the future. -.06 -.04 -.02 .00 .02 .04 .06 -1.2 -0.8 -0.4 0.0 0.4 0.8 1.2 2002 2004 2006 2008 2010 2012 2014 2016 2018 gap obr haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 177 table 2. causality test of revenue and expenditure to gdp. null hypothesis: obs f-stat prob. d(log(y)) does not granger cause d(log(it)) 73 10.1578 0.0001 d(log(it)) does not granger cause d(log(y)) 8.1725 0.0007 d(log(y)) does not granger cause d(log(vat)) 73 24.7563 0.0000 d(log(vat)) does not granger cause d(log(y)) 19.5834 0.0000 d(log(y)) does not granger cause d(log(others)) 73 56.5161 0.0000 d(log(others)) does not granger cause d(log(y)) 44.0929 0.0000 d(log(y)) does not granger cause d(log(cex)) 73 59.6569 0.0000 d(log(cex)) does not granger cause d(log(y)) 46.3288 0.0000 d(log(y)) does not granger cause d(log(rex)) 73 15.8953 0.0000 d(log(rex)) does not granger cause d(log(y)) 17.5873 0.0000 d(log(y)) does not granger cause d(log(irp)) 73 16.2580 0.0000 d(log(irp)) does not granger cause d(log(y)) 14.8332 0.0000 similar results are found in the context of spending. there is a bi-directional granger causality between gdp and all of the three types of expenditure. the larger spending exerted by the government, the higher the gdp. the higher gdp requires the government to spend more to facilitate economic activities. they imply further that inducing the government spending can act as an automatic fiscal stabiliser, particularly in the time of recessions since granger causality test runs from government spending to gdp, which are considered in this study. the first estimation of equations (6) and (7) using the log-linear model, unfortunately, did not yield fiscal variables elasticity with respect to output gap as expected by the results of previous studies. the nlls (nonlinear least squares) model approach also produces similar results. the elasticity of the fiscal variable with respect to output gap even gives a negative sign and insignificant coefficient. for this reason, equations (6) and (7) are modified by assuming the elasticity of the fiscal variable to actual output is equal to unity. within this assumption, the elasticity of the fiscal variable is calculated with respect to potential output, instead of the output gap. table 3 provides elasticity of fiscal variables with respect to potential output. the parameter estimates indicate that value-added tax (0.98) is the most effective tax in stabilising business cycle fluctuations. the second most effective tax is the income tax (0.94). in line with the prior researches, we expect that income tax should be the most effective tax and has a positive effect on the business cycle fluctuations. it seems that in the case of indonesia, income tax (consisting of corporate and personal) could act as an automatic fiscal stabiliser mostly due not to the progressive tax rates, but because of the sensitivity of employment to gdp fluctuations. this confirms krajewsky and pilat (2017). haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 178 table 3. potential output elasticity. revenue elasticity expenditure elasticity it 0.9393 cex 0.9065 vat 0.9809 rex 1.1907 others 0.2769 irp 0.1546 total 0.6695 total 0.8457 on the spending side, transfer to regions is the most effective expenditure (1.2) in stabilising the business cycle fluctuations. the second most effective spending is the central government expenditures (0.91). those results make sense. since 2001, fiscal responsibilities between central and local governments have been more clearly defined. the fiscal decentralisation and regional autonomy turned into an additional source of central government expenditure to fulfil the vertical and horizontal fiscal imbalances. these imply that various compensations and social protection programs can be well distributed either through central government spending or transfer to regions. figure 2 explains further the magnitude of automatic fiscal stabilisers and fiscal discretion policy which are derived from table 3 and figure 1. the automatic fiscal stabilisers appeared to be very high in 2005 in relation to the changes in budgetary regulation. after that, the size of automatic fiscal stabilisers is relatively stable, which rises and falls within the plus/minus 0.4 percent of the overall balance budget deficit. the same pattern is found in systematic fiscal discretion. figure 2. automatic fiscal stabilisers (as) and discretionary fiscal policy (dp). -0,006 -0,004 -0,002 0 0,002 0,004 0,006 0,008 0,01 20 01 q 1 20 02 q 1 20 03 q 1 20 04 q 1 20 05 q 1 20 06 q 1 20 07 q 1 20 08 q 1 20 09 q 1 20 10 q 1 20 11 q 1 20 12 q 1 20 13 q 1 20 14 q 1 20 15 q 1 20 16 q 1 20 17 q 1 20 18 q 1 20 19 q 1 as dp haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 179 the magnitude of systematic fiscal discretion was notably very high in 2005 and during the 2009 global financial crisis. the release of the new taxation law in 2008 has triggered the size of systematic fiscal discretion largely. overall, the size of automatic fiscal stabilisers is greater than that of systematic fiscal discretion. however, the government consistently uses both automatic fiscal stabilisers and systematic fiscal discretion to manage macroeconomic performance through the state budget deficit. when we break it down into two sub-periods of observation, the conclusion does not change. table 4 summarises the automatic fiscal stabilisers contribution of each revenue side and expenditure side derived from equation (12). the contribution of automatic fiscal stabilisers from the revenue side reached 0.21, with automatic stabilisers of income tax having the largest contribution (0.11). meanwhile, the automatic fiscal stabilisers contribution from the spending side reached 0.31. the contribution of automatic fiscal stabilisers in terms of central government and transfer to regions expenditures are almost the same, for about 0.17. in general, the contribution of automatic fiscal stabilisers is getting bigger. this conclusion is in contrast to boone and buti (2019), who obtained empirical evidence that automatic fiscal stabilisers in developed countries are economically weak. the increase in contribution indicates that automatic fiscal stabilisers have a big potential to be the main fiscal weapon for economic stability goals. this finding is in accordance with the imf's suggestion (baunsgaard and symansky, 2009; spilimbergo et al., 2010). based on those results, we conclude that automatic fiscal stabilisers are potentially implemented to fight the current economic fluctuations induced by covid-19. table 4 contribution of automatic fiscal stabilisers. revenue expenditure period total it vat others total cex rex irp 2001-09 0.1178 0.0560 0.0372 0.0217 0.1580 0.0939 0.0694 0.0039 2010-19 0.2154 0.1195 0.0826 0.0305 0.3118 0.1737 0.1657 0.0059 2001-19 0.1950 0.1063 0.0731 0.0287 0.2797 0.1571 0.1456 0.0054 note. processed from table 1 and 3. the following sections focus more on automatic fiscal stabilisers. the comparison of the effectiveness between automatic fiscal stabilisers and fiscal discretion policy is summarised in table 5. the automatic fiscal stabilisers are negatively related to the primary balance budget deficit and the overall balance budget deficit ratios. since both the primary balance and the overall balance are deficits (have minus signs), the correlation should be understood in the opposite direction. this is consistent with the theoretical prediction that automatic stabilisers will increase the deficit during the downturns, and, vice versa, they narrow the budget deficit during the upswings (trihartanto, 2019). meanwhile, automatic fiscal stabilisers have a negative correlation with the output gap. this means that the automatic stabilisers component can be properly operated if the potential output is greater than the actual output. haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 180 the discretionary fiscal component, as found in the automatic fiscal stabilisers, is positively correlated to the primary balance budget deficit and the overall balance budget deficit, and negatively to the output gaps. these support the above analysis that the government uses both automatic fiscal stabilisers and systematic fiscal discretion to manage macroeconomic performance. however, since the correlation of automatic fiscal stabilisers to gdp is greater than that of discretionary fiscal components, we find that automatic fiscal stabilisers are more effective to overcome recessions. furthermore, the primary balance budget deficit and the overall balance budget deficit are associated with output gaps in the same direction. since they represent the fiscal stance, this study discovers that the indonesian fiscal policy is characterised by procyclical or even a-cyclical. the procyclicality of fiscal policy tends to induce deficit bias which demands non-systematic discretionary fiscal policy instead of automatic fiscal stabilisers or systematic discretionary fiscal policy. eventually, the relatively low stabilisation component of fiscal policy is claimed to be the main cause of fiscal stimulus policies to be large and expensive (dolls, fuest, and peichl, 2012). table 5 correlation matrix, 2001-2019. as dp pbr obr gap as 1.0000 0.7553 -0.4431 -0.5183 -0.6159 dp 0.7553 1.0000 -0.3032 -0.3356 -0.8305 pbr -0.4431 -0.3032 1.0000 0.8637 0.2916 obr -0.5183 -0.3356 0.8637 1.0000 0.2821 gap -0.6159 -0.8305 0.2916 0.2821 1.0000 note. as = automatic fiscal stabilizer obr = overall balance to gdp ratio dp = discretionary fiscal policy gap = output gap between actual pbr = primary balance to gdp ratio to potential output 5. conclusions the use of large fiscal stimulus packages to dampen the impact of covid-19 recently has raised concerns about the effectiveness of the discretionary fiscal policy. while the role of automatic fiscal stabilisers is very important in developed countries, it is likely less prominent in developing countries. automatic fiscal stabilisers – which can be described as a built-in response of public finance by scaling up and drawing down automatically to economic fluctuations – play a critical role in fighting every recession. this paper aims at analysing the feasibility of automatic fiscal stabilisers to mitigate economic fluctuations in the case of indonesia. using the imf standard model for quarterly data over the period of 2001(1) to 2019(4), we find that the role of automatic fiscal stabilisers is getting greater haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 181 both in the revenues and spendings. compared to systematic fiscal discretion, the automatic fiscal stabiliser is more effective to overcome business cycles. in other words, the automatic fiscal stabiliser is feasible as the main fiscal policy instrument for economic stability goals in the future, including the current economic fluctuations induced by covid-19. eventually, this research suggests that indonesia’s government should immediately install automatic fiscal stabilisers in its long-term grand fiscal policy. however, given the existing circumstances, indonesia should reform its economic, regulatory, and institutional ecosystems in adopting completely automatic fiscal stabilisers. without initiating the use of automatic fiscal stabilisers permanently, indonesia will always tend to rely on the non-systematic discretionary fiscal policy in the form of fiscal stimulus in anticipating any unexpected recessions. references abdurohman, and resosudarmo, b.p. (2017). the behavior of fiscal policy in indonesia in response to economic cycles, singapore economic review 62, 2: 377-401. https://doi.org/10.1142/s0217590816500041 akitoby, b., clements, b., gupta, s., and inchauste, g. (2006). public spending, voracity, and wagner’s law in developing countries. european journal of political economy 22, 4: 908–24. https://doi.org/10.1016/j.ejpoleco.2005.12.001 arestis, p., (1985), is there any crowding-out of private expenditure by fiscal actions? in arestis, p. and skouras, t. (edt.), post keynesian economic theory, sussex, wheatsheaf books: 99-123. baldacci, e., (2009). neither sailing against the wind, nor going with the flow: cyclicality of fiscal policy in indonesia, imf country report no. 09/231. basri, m.c., and rahardja, s. (2011). mild crisis, half-hearted fiscal stimulus: indonesia during the gfc. in ito, t. and parulian, f. (edt.). assessment on the impact of stimulus, fiscal transparency, and fiscal risk, eria research project report 2010-01: 169-211. baunsgaard, t., and symansky, s.a. (2009), automatic fiscal stabilisers, imf staff position note no. 2009/23, september. https://doi.org/10.5089/9781455290567.004 blanchard, o., dell’ariccia, g., and mauro, p. (2010), rethinking macroeconomic policy, journal of money, credit, and banking 42, 1: 199–215. https://doi.org/10.1111/j.15384616.2010.00334.x buettner, t., and fuest, c. (2010), the role of the corporate income tax as an automatic stabiliser, international tax public finance 17: 686–98. https://doi.org/10.1007/s10797-010-9155-7 caldara, a., maravalle, a., rawdanowicz, l., and chico, a.s. (2020), strengthening automatic stabilisers could help combat the next downturn, voxeu, 23 march 2020. cornia, g., (2006), (edt.), pro-poor macroeconomics: potential and limitations, new york: palgrave macmilan. darby, j., and melitz, j. (2008). social spending and automatic stabilisers in the oecd, economic policy 23, 56: 715-56. https://doi.org/10.1111/j.1468-0327.2008.00210.x https://doi.org/10.1111/j.1538-4616.2010.00334.x https://doi.org/10.1111/j.1538-4616.2010.00334.x haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 182 debrun x. and kapoor r. (2010), fiscal policy and macroeconomic stability: automatic stabilizers work, always and everywhere. imf working paper wp/10/111 may. dolls m., fuest, c., and peichl, a. (2011), automatic stabilisers and economic crisis: us vs. europe, journal of public economics 96, 3–4, april: 279-94. https://doi.org/10.1016/ j.jpubeco.2011.11.001 doraisami, a., (2013). fiscal policy challenges in developing countries: the indonesian experience in responding to the global financial crisis, journal of southeast asian economies 30, 3: 322–33. https://doi.org/10.1355/ae30-3g estep, s., ajilore, o., and madowitz, m. (2019), the importance of automatic stabilisers in the next recession, center for american progress, washington. fatás, a., and mihov, i. (2003), the case for restricting fiscal policy discretion, quarterly journal of economics 118, 4: 1419-47. https://doi.org/10.1162/003355303322552838. giorno c., richardson, p., roseveare, d., and van den noord, p. (1995), potential output, output gaps, and structural budget balances, oecd economics department working papers no. 152. gnip, a.g., (2011), discretionary measures and automatic stabilisers in the croatian fiscal policy, economic research ekonomska istraživanja 24, 3: 45-74. https://doi.org/10.1080/ 1331677x.2011.11517467 granado, j.a., gupta, s., and hajdenberg, a. (2013), is social spending procyclical?, world development 42, february: 16-27. https://doi.org/10.1016/j.worlddev.2012.07.003 hodrick, r.j. and prescott, e.c. (1997). post-war u.s. business cycles: an empirical investigation, journal of money, credit, and banking 29: 1-16. hur, s., mallick, s. and park, d. (2014), fiscal policy and crowding out in developing asia, environment and planning c: politics and space 32, 6, december: 1117–32. https://doi.org/10.1068/c12185r. krajewsky, p., and piłat, k. (2017), does a progressive pit stabilize the economy? a comparison of progressive and flat taxes, comparative economic research 20,1: 21-33. https://doi.org/10.1515/cer-2017-0002 kuncoro, h., (2017), fiscal policy and stock market returns volatility: the case of indonesia, international journal of economic policy in emerging economies 10, 2: 153–70. https://doi.org/10.1504/ijepee.2017.085286 leiserson g. (2020), the coronavirus recession highlights the importance of automatic stabilisers. washington center for equitable growth, june 30. mabbett d., and schelkle, w. (2007), bringing macroeconomics back into the political economy of reform: the lisbon agenda and the 'fiscal philosophy' of emu, journal of common market studies 45, 3: 81-103. https://doi.org/10.1111/j.1468-5965.2007.00704.x mckay, a., and reis, r. (2016), the role of automatic stabilisers in the u.s. business cycle, econometrica 84, 1: 141-94. https://doi.org/10.3982/ecta11574 musgrave, r.a., and musgrave, p.b. (1989), public finance in theory and practice, 5th edition, singapore: mcgraw-hill company. https://doi.org/10.1016/j.jpubeco.2011.11.001 https://doi.org/10.1016/j.jpubeco.2011.11.001 https://doi.org/10.1355/ae30-3g https://doi.org/10.1162/003355303322552838 https://doi.org/10.1080/1331677x.2011.11517467 https://doi.org/10.1080/1331677x.2011.11517467 https://doi.org/10.1515/cer-2017-0002 https://doi.org/10.1504/ijepee.2017.085286 https://equitablegrowth.org/people/greg-leiserson/ https://doi.org/10.1111/j.1468-5965.2007.00704.x haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 183 nunns, j.r., and rosenberg, j. (2016), a federal consumption tax as replacement for the employer payroll tax, tax policy center. 22. july 6. paulus, a., and tasseva, i.v. (2020), europe through the crisis: discretionary policy changes and automatic stabilisers, oxford bulletin of economics and statistics 82, 4: 864-88. https://doi.org/10.1111/obes.12354 poghosyan, t., senhadji, a., and cottarelli, c. (2016), the role of fiscal transfers in smoothing regional shocks: evidence from existing federations, imf working paper wp/16/141, june. https://doi.org/10.5089/9781498379601.001 resosudarmo, b.p., abdurohman, yusuf, a.a., and hartono, d. (2020), spatial impacts of fiscal stimulus policies during the 2009 global financial crisis in indonesia. asia-pacific journal of regional sciences 5: 305-26. https://doi.org/10.1007/s41685-020-00176-2 ritz, b., (2020), emergency economics: fighting a recession in 2020 and beyond, the progressive policy institute march. 17, 2020, retreived on februari 20, 2021. https://www.progressivepolicy.org/wp-content/uploads/2020/03/ppiemergencyeconomics.pdf salgado, m.f., figari, f., sutherland, h., and tumino, a. (2014), welfare compensation for unemployment in the great recession, review of income and wealth 60, s1: s177-s204. https://doi.org/10.1111/roiw.12035 şen, h., (2013), the role of taxes as an automatic stabiliser: evidence from turkey”, economic analysis & policy 43, 3, december: 303-13. https://doi.org/10.1016/s0313-5926(13)50033-6 simorangkir, i., and adamanti, j. (2010), the role of fiscal stimulus and monetary easing in indonesian economy during global financial crisis: financial computable general equilibrium approach, bulletin of monetary economics and banking 13, 2: 169-92. https://doi.org/10.21098/bemp.v13i2.389 spilimbergo, a., symansky, s.a., and cottarelli, c., blanchard, o.j. (2009). fiscal policy for the crisis, imf staff position notes 2008/001, international monetary fund. szmigiera, m. (2020), value of covid-19 stimulus packages in the g20 countries as of october 2020, retreived on february 20, 2021, from https://www.statista.com/ suescún, r., (2007), the size and effectiveness of automatic fiscal stabilisers in latin america, policy research working papers wps4244. https://doi.org/10.1596/1813-9450-4244 surjaningsih, n., utari, g.a.d. and trisnanto, b. (2012), the impact of fiscal policy on the output and inflation, bulletin of monetary economics and banking 14, 4: 367-96. https://doi.org/10.21098/bemp.v14i4.409 trihartanto, a.y., (2019), a fiscal instrument: the role of automatic stabilisers, the jakarta post, march 22: 6. undrr, (2020), disaster risk reduction in the republic of indonesia: status report 2020, united nations office for disaster risk reduction, regional office for asia and the pacific, bangkok. veld, j., larch, m., and vandeweyer, m., (2012), automatic fiscal stabilisers: what they are and what they do, european commission economic papers no, 452, april. https://doi.org/ 10.2139/ssrn.2125796 https://www.progressivepolicy.org/wp-content/uploads/2020/03/ppi-emergencyeconomics.pdf https://www.progressivepolicy.org/wp-content/uploads/2020/03/ppi-emergencyeconomics.pdf https://www.statista.com/ https://doi.org/10.2139/ssrn.2125796 https://doi.org/10.2139/ssrn.2125796 haryo kuncoro / european journal of government and economics 10(2), december 2021, 167-184 184 vera, l.v., (2016), fluctuation paths and controversial issues of fiscal policy: where do these leave us?, romanian journal of fiscal policy 7, 1: 24-43. the role of automatic stabilizers in business cycle:the case of indonesia 1. introduction 2. literature review 3. analytical model and data 4. results and discussion 5. conclusions references © the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 2 (2025), pages 219-234 https://doi.org/10.17979/ejge.2025.14.2.11723 submitted: feb 6, 2025 accepted: sep 8, 2025 published: dec 17, 2025 article how government spending matters to fdi inflows in developing countries? thi bich thuy dao 1, * 1 vnu university of economics and business, hanoi, vietnam *correspondence: thuydaokt@vnu.edu.vn abstract. this study explores the effect of government spending on fdi inflows in developing country groups by different level of development. the united nations’ country classification by income level is used to classify country level of development. the study covers a sample of 100 developing countries worldwide over the 2002-2022 period. applying different estimation techniques, the findings reveal a declining relationship between government spending and fdi inflows along the rise in country level of development. the low-income country group has the largest positive effect of government spending on fdi inflows, the effect decreases in the lower-middle income, the upper-middle income, and finally it turns to be negative in the high-income country group. policy implications are derived that at lower level of development, larger size of government spending may attract inward fdi but as country level of development rises, fdi inflows favor smaller size of government spending. keywords: fdi inflows; government spending; developing countries jel classification: f21; h50; o57 1. introduction foreign direct investment (fdi) is widely recognized as an engine for economic growth and development of developing countries. the positive growth effect of fdi in developing countries is evident and confirmed in many empirical studies (sahu, 2021). fdi supplements the host country’s capital stock, raises the skill level of local workforce and contributes to technology transfer (abbes et al., 2015). thanks to the high reputation of fdi contribution to economic growth, many developing countries wish to lure fdi and therefore well-understanding of fdi determinants is crucial for policymakers. great attention has been focused on the determinants of fdi including macroeconomic stability, market size, growth prospect, human capital, labor costs, infrastructure, level of technology, endowment of natural resources, institutional environment (islam and beloucif, 2024). market-seeking fdi with the aim of serving foreign market is attracted to a country with large market size due to a high expected demand and prospect for efficient utilization of resources and exploitation of economies of scale (scaperlanda and mauer, 1969). foreign firms are seeking for https://creativecommons.org/licenses/by-nc/4.0/ 220 dao countries with large market sizes as well as high purchasing power of their people. growth in per capita gdp results in a larger market size and higher peoples’ living standard which offers more opportunities for profitable investments and thus a prosperous business expansion for foreign firms in the host country. according to onyeiwu and shrestha (2004), high-growth economies are typically expected to implement stable and credible macroeconomic policies which attract fdi. human capital plays an important role in the attraction of fdi, especially for efficiencyseeking fdi (karimi et al., 2013). higher level of human capital makes physical capital more productive (lucas, 1988) and therefore fdi should be attracted to countries with high levels of human capital. besides, the advantageous source of multinational corporations is the possession of state-of-the-art technology and modern techniques of production and this requires the availability of skilled workforce in the host countries (abbas et al., 2021). fdi represents a long-term investment in the host country. long-term investment decisions work well under the stability of a country’s currency and market-determined prices. a stable currency preserves its value and firms are more confident to make long-term plans and more able to achieve the efficiency of allocation of resources which lowers the costs of doing business. inflation distorts the value of money, causes uncertainty which makes long-term planning difficult and therefore discourages investment (reece and sam, 2012). in this respect, monetary policies to ensure the stable and reliable of currency should have a positive effect on the influx of fdi. the idea that tax burden influences location decisions is initiated in the work of tiebout (1956) on models of location. in choosing an investment location, owners of mobile factors of production consider the costs of taxation. reductions in taxes provide incentives for foreign firms to expand as they see more profitable for their investments in the host country. as well argued by rădulescu and druica (2014), tax reductions create two positive effects on fdi. the direct effect of tax cuts is to increase business profits while the indirect effect works through the purchasing decisions of households. a decrease in vat or duty taxes raises the consumption purchasing power and a lower personal income tax increases disposable income which encourages more households’ consumption and therefore higher domestic demand for foreign firms’ products stimulates the growth of fdi. fdi is lured by the attractiveness of the investment environment which in turns is shaped by institutional quality. improvement in institutional quality such as a more stable political environment reduces the risk of property loss, destruction, or expropriation and thus encourage fdi (chandra and handoyo, 2020). foreign investors are more confident to undertake entrepreneurial activity when their properties are protected by secure property rights (tag, 2021) and the law and order system functions efficiently (tag and degirmen, 2022). relaxation on regulations containing rules, procedures and requirements imposed on business would directly affect the ease of entry and exit, doing business and profit capture (saha et al., 2022). recent interest emphasized on the role of government spending in the attraction of inward fdi in developing countries (othman et al., 2018). the impact of government spending on fdi can be traced back to the argument raised by aschauer (1989) who claimed that government spending has two opposing effects – complementary and hindrance effects on private investment. theoretically, government spending which is directed to productive investment such as provision for infrastructure and human development would raise the economy’s resource productivity which in how government spending matters to fdi inflows in developing countries? turns increase returns on private investment and therefore encourage private investment activities. on the other hand, the hindrance effect works when government spending crowds out private investment. higher government spending pushes pressure upward on interest rates which makes it more costly for private firms to borrow to make investment. furthermore, increases in government spending may force the government to increase taxes leading to higher costs of production for firms and make business adventures less profitable. higher government spending, if not accompanied with higher taxes, would result in more accumulation of public debt. since private sector may consider public debt as an indicator of economic uncertainty (ahlborn and schweickert, 2018), rising public debt discourages private investment activities. empirical studies supported for both effects of government spending on private investment. the positive effect of government spending on private investment is evident in studies by narayan (2004), akinlo and oyeleke (2018), ramirez (2000), and nguyen (2023). in contrast, many other studies found that government spending crowds-out private investment, among them are furceri and sousa (2011), cavallo and daude (2011), şen and kaya (2014), kim and nguyen (2020) and carvelli (2023). in another line of argument, researchers claimed that the impact of government spending on private investment depends on the type and composition of government spending. in a study for 14 oecd countries for the 1979-1988 period, argimon et al. (1997) showed that government investment asserts a positive effect while government consumption has a negative effect on private investment. the same conclusion is reached in a study for greece by mamatzakis (2001). for the case of canada during the 19612000 period, wang (2005) found that government spending on education and health crowds-in whereas government spending on capital and infrastructure crowd-out private investment. rahman et al. (2015) examined the impact of government spending on private investment in pakistan in the period from 1974 to 2010 and concluded that while government spending on transport and communication, health and agriculture has a positive effect on private investment, the effect of debt servicing and community servicing expenditures on private investment is negative. there are few studies attempted to examine the impact of government spending on fdi inflows and yet the results are mixed. in a study for selected countries in middle east and north africa from 2000 to 2016, anwar (2017) claimed that government spending has a negative effect on fdi inflows. however, a study done by othman et al. (2018a) for the case of asean-5, china and india in the 1982-2016 period found a positive effect of government spending on inward fdi. the same conclusion is reached that government spending promotes fdi inflows in the study by othman et al. (2018b) for a group of 24 developing countries in the period from 1982 to 2014. at the compositional level of government spending, a study by shah and iqbal (2016) for pakistan in the 1972-2013 period found a positive long run relationship between government expenditure on health, education and development and fdi while the relationship between defense expenditure and fdi is negative. azolibe et al. (2020) showed that government spending on road, transportation, defense, and health infrastructure has a positive relationship with fdi in nigeria. there is an ample room for studies discovering how government spending affects fdi inflows in developing countries. developing countries possess with different levels of development. in our argument, the effect of government spending on fdi inflows is subject to the role of government in the promotion for the economy’s resource productivity. when the government plays a major role in providing resources for infrastructure and human development, an increase in 222 dao government spending is much needed to modernize the economy’s infrastructure and enhance human capital formation leading to higher economy’s resource productivity. however, if this role is shifted to the private sector then an increase in government spending dampens private investment and therefore slows down the growth rate of the economy’s resource productivity. the involvement of government in the provision of resources for infrastructure and human development seems to depend on country level of development. in countries at lower level of development, a large share of provision for infrastructure and human development relies on public investment. as country level of development increases, the private sector gradually takes over the public sector in undertaking these activities. since the government has less role in the contribution to the growth of the economy’s resource productivity, any increase in government spending would only exaggerate the negative effect on private investment. this leads us to propose the following hypotheses: hypothesis 1: the effect of government spending on fdi inflows in developing countries changes along country level of development. hypothesis 2: the effect of government spending on fdi inflows in developing countries is positive at first and then turns to be negative as country level of development rises. to test for the hypotheses, section 2 presents model specification, data and methodology. results are provided in section 3 and finally section 4 is concluding remarks. 2. model specification, data and methodology 2.1 model specification the model is developed that specifies government spending as a determinant of fdi inflows. some key control variables are included into the model as per the existing literature including market growth (islam and beloucif, 2024), human capital (abbas et al., 2021), sound money (dia and ondoa, 2023), tax burden (esteller-more et al., 2020) and institutional quality (emako et al., 2022). the regression equation for the model is written as 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖 ,𝑡𝑡 = 𝑐𝑐 + 𝛽𝛽1𝐺𝐺𝐺𝐺𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖 ,𝑡𝑡 + 𝛽𝛽2𝐺𝐺𝑃𝑃𝐺𝐺𝑡𝑡,𝑖𝑖 + 𝛽𝛽3𝐻𝐻𝐺𝐺𝐻𝐻𝑖𝑖 ,𝑡𝑡 + 𝛽𝛽4𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖 ,𝑡𝑡 + 𝛽𝛽5𝑇𝑇𝐻𝐻𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝛽𝛽6𝐹𝐹𝑆𝑆𝑆𝑆𝑇𝑇𝑖𝑖,𝑡𝑡 +𝛽𝛽7𝐸𝐸𝑇𝑇𝐺𝐺𝑖𝑖 ,𝑡𝑡 + 𝑒𝑒𝑖𝑖,𝑡𝑡 (1) where country and time (measured in year) are denoted by subscript i and t, respectively. fdi is fdi inflows measured as net inflows of foreign direct investment (% of gdp). grgdpc is per capita gdp growth measured as the annual growth rate of gdp per capita. pop is population growth measured as the annual growth rate of population. data on fdi inflows, per capita gdp how government spending matters to fdi inflows in developing countries? growth and population growth are taken from world development indicators (world bank)1. hca is human capital level measured as the natural logarithm of mean years of schooling. human development index (undp)2 provides data on mean years of schooling. smn is sound money which is measured by the index of monetary freedom taken from the index of economic freedom (heritage foundation)3. the index has the value ranging from 0 to 100 and a higher value means a more commitment of the government to conduct monetary policies to ensure the stable and reliable of currency. tax is reduction in tax burden which is measured by the index of tax burden taken from the index of economic freedom (heritage foundation). the index has the value ranging from 0 to 100 and a higher value means lower burden of taxes. inst is institutional quality which is proxied by governance index constructed by kaufmann et al. (1999). the index comprises of six governance indicators including government effectiveness, regulatory quality, rule of law, control of corruption, political stability and absence of violence, and voices and accountability. the index has the value ranging from -2.5 to 2.5 and a higher value means better institutional quality. following hayatt (2019) and others, institutional quality is computed by taking the average of the six governance indicators. data on governance index is taken from world governance indicators (world bank)4. exp is government spending measured as total expenditure of general government (% of gdp). data on total expenditure of general government is taken from world economic outlook database (international monetary fund)5. 2.2 data the objective of this study is to investigate the relationship between government spending and fdi inflows in developing countries by level of development. we adopt the united nations’ country classification by income level to classify country level of development. using the list of countries by income level provided in the world economic situation and prospects 2019 (united nations)6, the sample of 100 developing countries worldwide is divided into four groups by income level including low income (19 countries), lower-middle income (37 countries), uppermiddle income (30 countries) and high income (14 countries). the period of study is chosen from 2002 to 2022 which is mainly dictated by the availability of data for all variables included in the model. descriptive statistics are provided in table 1. all data are recorded in yearly basis. the 1 world development indicators (world bank): https://databank.worldbank.org/source/world-developmentindicators 2 human development index (undp): https://hdr.undp.org/data-center/documentation-and-downloads 3 index of economic freedom (heritage foundation): https://www.heritage.org/index/pages/all-countryscores 4 world governance indicators (world bank): https://databank.worldbank.org/source/worldwidegovernance-indicators 5 world economic outlook database (international monetary fund): https://www.imf.org/en/publications/weo/weo-database/2025/april/select-country-group 6 world economic situation and prospects 2019 (united nations): https://www.un.org/development/desa/dpad/wp-content/uploads/sites/45/wesp2019_book-web.pdf 224 dao table shows that the high income country group has highest mean value for fdi inflows, and followed by the low income, upper-middle income and finally lower-middle income country group. the lower and upper-middle income country groups experienced higher per capita gdp growth than the low and high income country groups. population growth in the lower-middle income and upper-middle income country groups is relatively lower than in the low and high income country groups. the upper-middle and high income country groups have higher mean value for sound money than the low and lower-middle income country groups. there are a clear pattern for human capital, reduction in tax burden and institutional quality as their mean values increase along the rise in income level. noticeably, government spending tends to increase steadily from the low to the lower-middle, the upper-middle and the high income country group. table 1. descriptive statistics high income country group upper-middle income country group max min std. dev. mean max min std. dev. mean fdi inflow 58.518 -6.967 9.270 6.478 27.760 -8.401 3.707 3.775 per capita gdp growth 14.710 -17.145 4.515 1.388 35.833 -34.203 5.321 2.014 population growth 19.360 -4.170 3.190 2.159 11.794 -2.880 1.237 1.503 human capital 13.443 5.977 1.717 10.007 11.607 3.474 1.516 8.237 sound money 93.000 62.300 6.073 79.314 94.300 33.400 8.538 74.764 reduction in tax burden 100.000 52.700 13.636 86.655 98.800 40.900 8.652 78.612 institutional quality 1.623 -0.488 0.536 0.596 0.870 -1.318 0.487 -0.205 government spending 62.503 9.015 8.894 28.729 52.123 10.793 8.178 27.035 lower-middle income country group low income country group max min std. dev. mean max min std. dev. mean fdi inflow 43.912 -37.172 5.425 3.561 103.330 -4.846 10.292 5.309 per capita gdp growth 16.404 -21.567 3.777 2.631 27.831 -36.777 4.753 1.734 population growth 4.156 -0.402 0.849 1.806 5.663 -0.077 0.616 2.838 human capital 11.250 1.380 1.951 5.897 7.206 0.977 1.442 3.292 sound money 90.400 38.100 7.163 73.483 90.300 46.700 7.715 73.822 reduction in tax burden 91.400 48.200 8.743 77.220 91.100 44.600 9.775 73.392 institutional quality 0.589 -1.753 0.442 -0.563 0.042 -1.726 0.410 -0.849 government spending 66.442 6.641 9.067 23.290 42.187 4.173 6.748 19.175 correlation matrices for all countries as a whole group and four individual country groups are provided in tables 2. table 2.1 to 2.4 show that there are low correlations among independent variables. in these tables, the variance inflation factor (vif) value for each variable is less than 2 implying no prevalence of multicollinearity problem. in table 2.5, institutional quality variable is highly correlated with government spending variable. institutional quality how government spending matters to fdi inflows in developing countries? variable has vif value of 2.54, so to avoid the multicollinearity problem, this variable is excluded from the regression equation for the high income country group. table 2.1 correlation matrix. all countries as a whole group. grgdpc pop hca tax smn inst exp vif grgdpc 1.0000 1.10 pop -0.2090 1.0000 1.19 hca -0.0471 -0.2678 1.0000 1.63 tax -0.0004 0.0920 0.3039 1.0000 1.16 smn -0.0329 -0.0180 0.0419 -0.0073 1.0000 1.18 inst 0.0340 -0.1903 0.5128 0.2176 0.3330 1.0000 1.67 exp -0.1274 -0.0694 0.3271 0.0729 0.0059 0.3220 1.0000 1.20 table 2.2 correlation matrix. low income country group. grgdpc pop hca tax smn inst exp vif grgdpc 1.0000 1.03 pop 0.0025 1.0000 1.16 hca -0.0354 -0.2442 1.0000 1.28 tax 0.0694 -0.2314 0.3034 1.0000 1.45 smn -0.0378 0.1689 -0.3112 -0.1415 1.0000 1.19 inst 0.1190 0.0430 -0.0518 0.3831 0.1830 1.0000 1.31 exp -0.0447 0.0494 -0.0763 0.2398 0.0107 0.2259 1.0000 1.09 table 2.3 correlation matrix. lower-middle income country group. grgdpc pop hca tax smn inst exp vif grgdpc 1.0000 1.11 pop -0.1865 1.0000 1.20 hca -0.0496 -0.2402 1.0000 1.18 tax 0.1175 -0.1628 0.1756 1.0000 1.18 smn -0.0782 -0.0784 -0.1503 -0.2869 1.0000 1.26 inst 0.0823 -0.2283 0.1323 -0.0481 0.2671 1.0000 1.37 exp -0.0564 -0.1511 0.2576 -0.0906 -0.0056 0.4244 1.0000 1.35 table 2.4 correlation matrix. upper-middle income country group. grgdpc pop hca tax smn inst exp vif grgdpc 1.0000 1.10 pop -0.1846 1.0000 1.14 hca -0.1310 -0.1272 1.0000 1.11 tax 0.0290 0.1035 0.0645 1.0000 1.11 smn -0.0318 0.1202 -0.0422 0.1785 1.0000 1.13 inst 0.0789 -0.1865 0.1877 -0.0077 0.1933 1.0000 1.24 exp -0.1537 0.0352 0.0867 -0.2041 -0.0432 0.2166 1.0000 1.18 table 2.5 correlation matrix. high income country group. grgdpc pop hca tax smn inst exp vif grgdpc 1.0000 1.30 pop -0.3638 1.0000 1.38 hca 0.0884 -0.3097 1.0000 1.49 tax -0.1592 0.3778 -0.4498 1.0000 1.66 smn 0.2418 -0.2327 0.3240 -0.2401 1.0000 1.34 inst 0.2040 -0.2631 0.4037 -0.3860 0.4337 1.0000 2.32 exp -0.2878 0.0909 -0.3031 0.0191 -0.3866 -0.6389 1.0000 2.18 the cointegration tests are carried whose results confirm a long run relationship between 226 dao fdi inflows and government spending. table 3. panel cointegration pedroni cointegration test decision phillips-perron augmented dickey fuller cointegration exists statistic -17.455 -16.877 p-value 0.000 0.000 kao cointegration test augmented dickey fuller cointegration exists statistic -9.042 p-value 0.000 westerlund cointegration test variance ratio cointegration exists statistic -5.899 p-value 0.000 2.3 estimation technique in the regression model, there is a high possibility that per capita gdp growth is endogenous variable since output growth is determined by growth of inputs and total factor productivity. we conduct the hausman test to compare ols model and gmm model in the presence of potential endogeneity. the null hypothesis of the test is no endogeneity. with the result of chi2(7) = 68.76, p-value = 0.0000, we reject the null hypothesis and conclude that endogeneity is present. to ensure the robustness of the result, two estimation techniques are used to address the endogeneity problem. first is the two-step system gmm model developed by arellano and bover (1995) and blundell and bond (1998). to estimate gmm, the lagged fdi variable is included into the regression model which makes the model be a type of dynamic. as claimed by roodman (2009), gmm model is effective for a sample of large n and small t. in this study, the sample has the number of identities (n = 100 countries) higher than the number of time (t = 21 years) and therefore satisfies the condition for an effective gmm estimation. besides, roodman (2009) argued that the number of instruments should not exceed the number of identities. to assess differences in the impact of government spending on fdi inflows between different country groups, three dummies variables are introduced. 𝐹𝐹1 = �1, lower middle income countries 0, others 𝐹𝐹2 = �1, upper middle income countries 0, others 𝐹𝐹3 = �1, high income countries 0, others how government spending matters to fdi inflows in developing countries? the regression equation for the gmm model is specified as 𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖 ,𝑡𝑡 = 𝛽𝛽1𝐹𝐹𝐹𝐹𝐹𝐹𝑖𝑖,𝑡𝑡−1 + 𝛽𝛽2𝐺𝐺𝐺𝐺𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖 ,𝑡𝑡 + 𝛽𝛽3𝐺𝐺𝑃𝑃𝐺𝐺𝑡𝑡,𝑖𝑖 + 𝛽𝛽4𝐻𝐻𝐺𝐺𝐻𝐻𝑖𝑖,𝑡𝑡 + 𝛽𝛽5𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 + 𝛽𝛽6𝑇𝑇𝐻𝐻𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝛽𝛽7𝐹𝐹𝑆𝑆𝑆𝑆𝑇𝑇𝑖𝑖 ,𝑡𝑡 +𝛽𝛽8𝐸𝐸𝑇𝑇𝐺𝐺𝑖𝑖,𝑡𝑡 + 𝛽𝛽9𝐹𝐹1 ∗ 𝐸𝐸𝑇𝑇𝐺𝐺𝑖𝑖,𝑡𝑡 + 𝛽𝛽10𝐹𝐹2 ∗ 𝐸𝐸𝑇𝑇𝐺𝐺𝑖𝑖,𝑡𝑡 + 𝛽𝛽11𝐹𝐹3 ∗ 𝐸𝐸𝑇𝑇𝐺𝐺𝑖𝑖,𝑡𝑡 + 𝑒𝑒𝑖𝑖,𝑡𝑡 (2) where fdi is lagged fdi inflows. the rest variables are defined as in equation (1). in the regression model, lagged fdi inflows and per capita gdp growth are treated as endogenous variables and the rest variables are exogenous. year dummies are included to account for year-specific effects. second, per capita gdp growth may be correlated with the error term in the model then to deal with endogeneity problem, following the suggestion by kumar and woo (2010) and cecchetti et al. (2011), we use two-year forward average of fdi inflows rather than annual fdi inflows. panizza and presbitero (2014) pointed out that the use of overlapping periods can cause autocorrelation in the model. feasible generalized least square (fgls) regression is an effective estimator to deal with autocorrelation problem and so is chosen. the regression equation for the fgls model is specified as 𝐹𝐹𝐹𝐹𝐹𝐹�����𝑖𝑖,(𝑡𝑡,𝑡𝑡+1) = 𝑐𝑐 + 𝛽𝛽1𝐺𝐺𝐺𝐺𝐺𝐺𝐹𝐹𝐺𝐺𝐺𝐺𝑖𝑖 ,𝑡𝑡 + 𝛽𝛽2𝐺𝐺𝑃𝑃𝐺𝐺𝑡𝑡,𝑖𝑖 + 𝛽𝛽3𝐻𝐻𝐺𝐺𝐻𝐻𝑖𝑖,𝑡𝑡 + 𝛽𝛽4𝑆𝑆𝑆𝑆𝑆𝑆𝑖𝑖,𝑡𝑡 + 𝛽𝛽5𝑇𝑇𝐻𝐻𝑇𝑇𝑖𝑖,𝑡𝑡 +𝛽𝛽6𝐹𝐹𝑆𝑆𝑆𝑆𝑇𝑇𝑖𝑖,𝑡𝑡 + 𝛽𝛽7𝐸𝐸𝑇𝑇𝐺𝐺𝑖𝑖,𝑡𝑡 + 𝑒𝑒𝑖𝑖,𝑡𝑡 (3) where 𝐹𝐹𝐹𝐹𝐹𝐹�����𝑖𝑖,(𝑡𝑡,𝑡𝑡+1) is two-year forward average of fdi inflows and the rest variables are the same as in equation (1). regression is conducted for each country group to examine the magnitude of the effect that government spending has on fdi inflows in each country group. year dummies are included to account for year-specific effects. 3. results table 4 presents regression results for the gmm model. as the table shows, the lagged fdi inflows is significantly positive which implies that the current period inflow of fdi depends positively on its last period. the success of foreign firms in the host country is a factor attracting further foreign investments. multinational firms are more likely attracted to countries that already have an accumulated sizable fdi. per capita gdp growth has a positive effect on fdi inflows. stronger growth of gdp per capita results in rising purchasing power of domestic residents which lead to a larger demand for goods including those produced by foreign firms. there is a positive relationship between population growth and fdi inflows. population growth not only results in a larger market size for the host countries but also provides a larger pool of workforce that keep labor costs at a competitive level. higher level of human capital is a determinant attracting inward fdi in developing countries. this is an evident that it is the quality of labor force, not cheap labor costs in the host country that becomes an increasingly important factor to attract fdi. a dominant competitive advantage of multinational firms over domestic firms is their possession of advanced technology. 228 dao however, foreign advanced technology is only implemented in the host countries if the domestic workers can handle it. foreign firms demand for high skilled workers and the more skilled workers are available the more chance for foreign firms to implement advanced technology to yield higher productivity. the relationship between sound money and the inflow of fdi is positive. sound money implies a higher commitment of the government to conduct monetary policies endeavors to fight inflation and maintain stable money. higher macroeconomic stability enables firms to make long term plans and induce them to engage in more investment ventures. reduction in tax burden contributes positively to the attraction of fdi inflows. institutional quality has a positive effect on fdi inflows indicating that improvement in institutional quality encourages inward fdi. table 4. determinants of fdi inflows in developing countries by income level. gmm model dependent variable: fdi inflows explanatory variables coef. std. err. fdi : lagged fdi inflows 0.585*** 0.007 grgdpc: per capita gdp growth 0.080*** 0.013 pop: population growth 0.084** 0.041 hca: human capital 0.750** 0.312 smn: sound money 0.027*** 0.008 tax: reduction in tax burden 0.018* 0.009 inst: institutional quality 0.902*** 0.190 exp: government spending 0.057*** 0.021 d1 (dummy – lower-middle income) * exp -0.050*** 0.019 d2 (dummy – upper-middle income) * exp -0.063*** 0.022 d3 (dummy – high income) * exp -0.078*** 0.024 year dummies yes ar(1) test (p-value) 0.003 ar(2) test (p-value) 0.317 hansen test (p-value) 0.236 number of instruments 70 observations 1940 note: *** p < 0.01, ** p < 0.05, * p < 0.1 the coefficient value for government spending in the low income country group is positive (β = 0.057, p-value < 0.01). in comparison to the low income country group, the magnitude of the government spending effect on fdi inflows is lower in the lower-middle income country group (by the difference of β = -0.050, p-value < 0.01) and even lower in the upper-middle income country group (by the difference of β = -0.063, p-value < 0.01), and finally it is lowest in the high income country group (by the difference of β = -0.078, p-value < 0.01). this implies that the size of the effect that government spending has on fdi inflows decreases as country level of development rises. regression results for the fgls model are presented in table 5. there is a clear pattern on the relationship between government spending and fdi inflows in country groups by income level. government spending has a largest positive effect on fdi inflows in the low income country group (β = 0.363, p-value < 0.01). the effect declines in the lower-middle income country group (β = 0.130, pvalue < 0.01), becomes lower in the upper-middle income country group (β = 0.035, p-value < 0.05) and finally it turns to be negative in the high income country group (β = -0.354, p-value < 0.01). these results again imply a declining effect of government spending on fdi inflows along the rise in country level of development. how government spending matters to fdi inflows in developing countries? table 5. determinants of fdi inflows in developing countries by income level. fgls model dependent variable: fdi inflows explanatory variables low income lower-middle income upper-middle income high income coef. (std. err.) coef. (std. err.) coef. (std. err.) coef. (std. err.) grgdpc: per capita gdp growth 0.134 (0.132) 0.240*** (0.054) 0.091*** (0.034) 0.043 (0.127) pop: population growth 1.687* (0.961) 0.831*** (0.210) 0.678*** (0.109) -0.402** (0.172) hca: human capital 3.730*** (1.188) 1.452** (0.500) 1.712** (0.737) 8.545** (3.642) smn: sound money -0.040 (0.068) 0.103*** (0.025) 0.101*** (0.016) 0.452*** (0.096) tax: reduction in tax burden -0.068 (0.060) 0.068*** (0.020) 0.061*** (0.015) 0.159*** (0.040) inst: institutional quality 0.944 (1.42) -0.021 (0.439) 0.054 (0.283) exp: government spending 0.363*** (0.081) 0.130*** (0.021) 0.035** (0.016) -0.354*** (0.058) year dummies yes yes yes yes observations 356 710 584 279 note: *** p < 0.01, ** p < 0.05, * p < 0.1 to interpret the declining pattern of the government spending and fdi inflows relationship, we reason that government spending has two opposing effects – complementary and hindrance effects on private investment, including fdi. the complementary effect encourages fdi inflows via the positive effect of productive public investment on the economy’s resource productivity. the hindrance effect discourages fdi inflows via the pressure of government spending on tax burden imposed on tax payers and the creation of economic uncertainty due to the possibility of higher public debt. governments involve in many economic activities in countries where level of development is low. public investment accounts for a large share of provision for infrastructure and human development. the majority of infrastructure development projects in roads, transportation, telecommunication, power and water supply and human development programs in education and health care are often undertaken and funded by the government. in these countries, governments take a major role in providing resources for infrastructure and human development which increase resource productivity in the economy. as country level of development rises, the share of government sector in economic activities declines. the provision of resource for infrastructure and human development may be gradually shifted from the government to private sector. in the structure of government spending, more funds are allocated to current government expenditures instead of public investment. therefore, the complementary effect of government spending decreases. in the low income country group, the complementary effect of government spending may be largest which dominates the hindrance effect leading to the net positive effect of government spending on fdi inflows. since the complementary effect decreases as country level of development rises, the size of the positive effect of government spending on fdi inflows declines in the lowermiddle income country group and keeps declining in the upper-middle income country group. finally, the complementary effect of government spending is outweighed by the hindrance effect in 230 dao the high income country group that causes a negative government spending effect on fdi inflows. 4. concluding remarks this study proposes hypotheses that the effect of government spending on fdi inflows in developing countries changes along country level of development and the effect is positive at first and then turns to be negative as country level of development rises. the hypotheses are tested with the sample of 100 developing countries worldwide grouped into low, lower-middle, upper-middle and high income countries in the 2002-2022 period. applying different estimation techniques including gmm and fgls estimators, the results show that the effect of government spending on fdi inflows declines as country level of development rises. government spending has a positive effect on fdi inflows in the low, lower-middle and upper-middle income country groups, though the magnitude of the impact decreases along the rise in country level of development. in these country groups, increases in the size of government spending play as a stimulus factor to fdi inflows. the effect turns to be negative in the high income country group indicating that sizable government spending acts as a barrier to the inflow of fdi. these findings imply that expansion of government spending is an effective tool to attract inward fdi in countries at low level of development, while in countries at high level of development governments should pursue contraction of their spending for the sake of fdi attraction. references abbas, a., moosa, i., & ramiah, v. (2021). the contribution of human capital to foreign direct investment inflows in developing countries. journal of intellectual capital, 23(1), 9–26. https://doi.org/10.1108/jic-12-2020-0388 abbes, s. m., mostéfa, b., seghir, g., & zakarya, g. y. (2015). causal interactions between fdi and economic growth: evidence from dynamic panel cointegration. procedia economics and finance, 23, 276– 290. https://doi.org/10.1016/s2212-5671(15)00541-9 ahlborn, m., & schweickert, r. (2018). public debt and economic growth: economic systems matter. international economics and economic policy, 15, 373–403. https://doi.org/10.1007/s10368017-0396-0 akinlo, t., & oyeleke, o. j. (2018). effects of government expenditure on private investment in nigerian economy (1980–2016). emerging economy studies, 4(2), 147–156. https://doi.org/10.1177/2394901518795067 anwar, s. m. (2017). political stability and attracting foreign direct investment: a comparative study of middle east and north african countries. science international (lahore), 29(3), 679–683. arellano, m., & bover, o. (1995). another look at instrumental variable estimation of error-components models. journal of econometrics, 68, 29–51. https://doi.org/10.1016/0304-4076(94)01642-d https://doi.org/10.1108/jic-12-2020-0388 https://doi.org/10.1016/s2212-5671(15)00541-9 https://doi.org/10.1007/s10368-017-0396-0 https://doi.org/10.1007/s10368-017-0396-0 https://doi.org/10.1177/2394901518795067 https://doi.org/10.1016/0304-4076(94)01642-d how government spending matters to fdi inflows in developing countries? argimon, i., gonzalez-paramo, j. m., & roldan, j. m. (1997). evidence of public spending crowding-out from a panel of oecd countries. applied economics, 29(8), 1001–1010. https://doi.org/10.1080/000368497326390 aschauer, d. a. (1989). is public expenditure productive? journal of monetary economics, 23, 177–200. https://doi.org/10.1016/0304-3932(89)90047-0 azolibe, c. b., okonkwo, j. j., & adigwe, p. k. (2020). government infrastructure expenditure and investment drive in an emerging market economy: evidence from nigeria. emerging economy studies, 6(1), 61–85. https://doi.org/10.1177/2394901520907722 blundell, r., & bond, s. r. (1998). initial conditions and moment restrictions in dynamic panel data models. journal of econometrics, 87, 115–144. https://doi.org/10.1016/s0304-4076(98)00009-8 carvelli, g. (2023). the long-run effects of government expenditure on private investments: a panel csardl approach. journal of economics and finance, 47(3), 620–645. https://doi.org/10.1007/s12197-023-09617-y cavallo, e., & daude, c. (2011). public investment in developing countries: a blessing or a curse? journal of comparative economics, 39, 65–81. https://doi.org/10.1016/j.jce.2010.10.001 cecchetti, s. g., mohanty, m., & zampolli, f. (2011). achieving growth amid fiscal imbalances: the real effects of debt. in economic symposium conference proceedings (vol. 352, no. 8, pp. 145–196). chandra, t. a., & handoyo, r. d. (2020). determinants of foreign direct investment in 31 asian countries for the 2002–2017 period. contemporary economics, 14(4), 566–581. dia, i., & ondoa, h. a. (2023). does economic freedom improve fdi inflows in sub-saharan africa? journal of economic integration, 38(3), 383–410. https://doi.org/10.11130/jei.2023.38.3.383 emako, e., nuru, s., & menza, m. (2022). determinants of foreign direct investments inflows into developing countries. transnational corporations review, 1–14. https://doi.org/10.1016/j.tncr.2023.100006 esteller-moré, a., rizzo, l., & secomandi, r. (2020). the heterogeneous impact of taxation on fdi: a note on djankov et al. (2010). economics letters, 186, 108775. https://doi.org/10.1016/j.econlet.2019.108775 furceri, d., & sousa, r. m. (2011). the impact of government spending on the private sector: crowding-out versus crowding-in effects. kyklos, 64(4), 516–533. https://doi.org/10.1111/j.14676435.2011.00518.x hayat, a. (2019). foreign direct investments, institutional quality, and economic growth. the journal of international trade & economic development, 28(5), 561–579. https://doi.org/10.1080/09638199.2018.1564064 islam, m. s., & beloucif, a. (2024). determinants of foreign direct investment: a systematic review of the empirical studies. foreign trade review, 59(2), 309–337. https://doi.org/10.1177/00157325231158846 karimi, m. s., yusop, z., hook, l. s., & chin, l. (2013). effect of human capital on foreign direct investment inflows. journal of economic research, 18(1), 79–179. https://doi.org/10.17256/jer.2013.18.1.004 kaufmann, d., kraay, a., & zoido-lobatón, p. (1999). aggregating governance indicators (vol. 2195). world bank publications. https://doi.org/10.2139/ssrn.188548 kim, t., & nguyen, q. h. (2020). the effect of public spending on private investment. review of finance, 24(2), 415–451. https://doi.org/10.1093/rof/rfz003 kumar, m., & woo, j. (2010). public debt and growth. imf working paper, wp/10/174. https://doi.org/10.5089/9781455201853.001 https://doi.org/10.1080/000368497326390 https://doi.org/10.1016/0304-3932(89)90047-0 https://doi.org/10.1177/2394901520907722 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1007/s12197-023-09617-y https://doi.org/10.1016/j.jce.2010.10.001 https://doi.org/10.11130/jei.2023.38.3.383 https://doi.org/10.1016/j.tncr.2023.100006 https://doi.org/10.1016/j.econlet.2019.108775 https://doi.org/10.1111/j.1467-6435.2011.00518.x https://doi.org/10.1111/j.1467-6435.2011.00518.x https://doi.org/10.1080/09638199.2018.1564064 https://doi.org/10.1177/00157325231158846 https://doi.org/10.17256/jer.2013.18.1.004 https://doi.org/10.2139/ssrn.188548 https://doi.org/10.1093/rof/rfz003 https://doi.org/10.5089/9781455201853.001 232 dao lucas, r. e., jr. (1988). on the mechanics of economic development. journal of monetary economics, 22(1), 3–42. https://doi.org/10.1016/0304-3932(88)90168-7 mamatzakis, e. c. (2001). public spending and private investment: evidence from greece. international economic journal, 15(4), 33–46. narayan, p. k. (2004). do public investments crowd out private investments? fresh evidence from fiji. journal of policy modeling, 26, 747–753. https://doi.org/10.1016/j.jpolmod.2004.06.002 nguyen, v. b. (2023). the relationship between public expenditure and private investment in developed and developing economies: policy implications based on the difference. hacienda pública española / review of public economics, 244(1), 37–55. https://doi.org/10.7866/hpe-rpe.23.1.2 onyeiwu, s., & shrestha, h. (2004). determinants of foreign direct investment in africa. journal of developing societies, 20(1–2), 89–106. https://doi.org/10.1177/0169796x04048305 othman, n., andaman, g., yusop, z., & ismail, m. m. (2018b). impact of public expenditures on fdi inflows into developing countries. pertanika journal of social sciences & humanities, 26(2). othman, n., yusop, z., andaman, g., & ismail, m. m. (2018a). impact of government spending on fdi inflows: the case of asean-5, china and india. international journal of business and society, 19(2), 1–6. panizza, u., & presbitero, a. f. (2014). public debt and economic growth: is there a causal effect? journal of macroeconomics, 41, 21–41. https://doi.org/10.1016/j.jmacro.2014.03.009 rădulescu, m., & druică, e. (2014). the impact of fiscal policy on foreign direct investments. empiric evidence from romania. economic research–ekonomska istraživanja, 27(1), 86–106. https://doi.org/10.1080/1331677x.2014.947133 rahman, m., ullah, i., & jebran, k. (2015). effects of government expenditure on private investment: evidence from pakistan. journal of basic and applied scientific research, 5(2), 14–23. ramirez, m. d. (2000). the impact of public investment on private investment spending in latin america: 1980–95. atlantic economic journal, 28(2), 210–225. https://doi.org/10.1007/bf02298362 reece, c., & sam, a. g. (2012). impact of pension privatization on foreign direct investment. world development, 40(2), 291–302. https://doi.org/10.1016/j.worlddev.2011.06.003 roodman, d. (2009). how to do xtabond2: an introduction to difference and system gmm in stata. the stata journal, 9(1), 86–136. https://doi.org/10.1177/1536867x0900900106 saha, s., sadekin, m. n., & saha, s. k. (2022). effects of institutional quality on foreign direct investment inflow in lower-middle income countries. heliyon, 8(10), e10828. https://doi.org/10.1016/j.heliyon.2022.e10828 sahu, j. p. (2021). does inflow of foreign direct investment stimulate economic growth? evidence from developing countries. transnational corporations review, 13(4), 376–393. https://doi.org/10.1080/19186444.2020.1833603 scaperlanda, a. e., & mauer, l. j. (1969). the determinants of u.s. direct investment in the eec. the american economic review, 59(4), 558–568. şen, h., & kaya, a. (2014). crowding-out or crowding-in? analyzing the effects of government spending on private investment in turkey. panoeconomicus, 61(6), 631–651. https://doi.org/10.2298/pan1406631s shah, n., & iqbal, y. (2016). impact of government expenditure on the attraction of fdi in pakistan. international journal of economics and empirical research, 4(4), 212–219. https://doi.org/10.1016/0304-3932(88)90168-7 https://doi.org/10.1016/j.jpolmod.2004.06.002 https://doi.org/10.7866/hpe-rpe.23.1.2 https://doi.org/10.1177/0169796x04048305 https://doi.org/10.1016/j.jmacro.2014.03.009 https://doi.org/10.1080/1331677x.2014.947133 https://doi.org/10.1007/bf02298362 https://doi.org/10.1016/j.worlddev.2011.06.003 https://doi.org/10.1177/1536867x0900900106 https://doi.org/10.1016/j.heliyon.2022.e10828 https://doi.org/10.1080/19186444.2020.1833603 https://doi.org/10.2298/pan1406631s how government spending matters to fdi inflows in developing countries? tag, m. n. (2021). judicial institutions of property rights protection and foreign direct investment inflows. international review of law and economics, 65, 105975. https://doi.org/10.1016/j.irle.2020.105975 tag, m. n., & degirmen, s. (2022). economic freedom and foreign direct investment: are they related? economic analysis and policy, 73, 737–752. https://doi.org/10.1016/j.eap.2021.12.020 tiebout, c. (1956). a pure theory of local expenditures. journal of political economy, 64, 416–424. https://doi.org/10.1086/257839 wang, b. (2005). effects of government expenditure on private investment: canadian empirical evidence. empirical economics, 30(2), 493–504. https://doi.org/10.1007/s00181-005-0245-9 https://doi.org/10.1016/j.irle.2020.105975 https://doi.org/10.1016/j.eap.2021.12.020 https://doi.org/10.1086/257839 https://doi.org/10.1007/s00181-005-0245-9 234 dao appendix list of developing countries in the study developing countries (100 countries) low income countries: burkina faso, burundi, central african republic, chad, congo (d.r.), ethiopia, gambia, guinea, guineabissau, liberia, madagascar, malawi, mali, mozambique, niger, rwanda, sierra leone, togo, uganda. lower-middle income countries: angola, bangladesh, benin, bhutan, bolivia, cabo verde, cambodia, cameroon, comoros, congo (republic), côte d'ivoire, egypt, el salvador, ghana, haiti, honduras, india, indonesia, kenya, lao, lesotho, mauritania, mongolia, morocco, myanmar, nepal, nicaragua, nigeria, pakistan, philippines, sao tome and principe, senegal, sri lanka, tanzania, tunisia, viet nam, zambia. upper middle income countries: algeria, argentina, belize, botswana, brazil, china, colombia, costa rica, dominican republic, ecuador, equatorial guinea, fiji, gabon, guatemala, iran, jamaica, jordan, lebanon, maldives, malaysia, mauritius, mexico, namibia, panama, paraguay, peru, south africa, suriname, thailand, turkey. high income countries: bahrain, barbados, brunei darussalam, chile, israel, korea (republic), kuwait, oman, qatar, saudi arabia, singapore, trinidad and tobago, united arab emirates, uruguay. 1. introduction 2. model specification, data and methodology 3. results 4. concluding remarks references ©the authors 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 165-188 https://doi.org/10.17979/ejge.2024.13.2.10225 submitted: dec 29, 2023 accepted: may 23, 2024 published: dec 3, 2024 article investor sentiment and euro area etfs: an empirical analysis of consumer behaviour pavlo dziuba,1,* maryna hrysenko,1 kyryl shtogrin1 1 taras shevchenko national university of kyiv, ukraine *correspondence: pavlo_dziuba@ukr.net abstract. the goal of this article is to study the relationship between investor sentiment and euro area exchange traded funds (etfs), providing insights into the dynamics of market behaviour and the impact of consumer behaviour on investment patterns within the eurozone. this paper has employed panel data analysis, correlation analysis, regression analysis, and the granger causality test to investigate the relationship between etfs and the consumer confidence index (cci). results suggest that cci significantly negatively affects etf pricing. while a weak positive correlation and regression relationship between ishares msci eurozone etf prices and cci 19 euro area markets have been found, the granger causality test did not provide evidence of a causal relationship between the two variables. other factors, such as financial stability, government regulations, and market-specific factors, may also affect etf prices. the study concludes that, while cci may be important, it should not be the only factor considered. keywords: markets; euro area; confidence index; behavioural finance; european exchange traded fund. jel classification: e22; g14 1. introduction the exchange traded fund (etf) is a popular investment vehicle (open-end investment fund traded on the stock exchange) with the main purpose of serving as a proxy for a reference index (benchmark), and its performance is usually measured through a comparison with the underline benchmark’s performance (henriques et al., 2022). etf includes all securities (usually stocks or bonds) that make up an index. however, there are a few etfs that track indices using derivatives. the main benefit of etfs is the opportunity to get exposure to a basket of different securities, such as stocks, bonds, and cryptocurrencies, without having to deal with information costs, transaction costs, or regulatory differences. marta and riva (2022) explain the incredible success of etfs in the investment industry through their advantages relating to diversification, transparency, tax efficiency, management fees, and liquidity. in this paper, the authors specifically investigate only stock etfs for several reasons. etfs are a convenient way to assess the overall performance of a particular market since they represent a https://creativecommons.org/licenses/by-nc/4.0/ 166 pavlo dziuba et al. basket of different securities. limiting the choice of etfs to specific markets enables better isolation of consumer confidence effects. having in mind that the etfs in question are market-specific, i.e., etfs are composed of companies’ stocks within a certain market, they are more sensitive to consumer confidence within the country's market. finally, stock etfs are the most popular kind of etf, as they are highly available to different types of investors all around the globe and are publicly traded on the main platforms. exchange-traded funds have already transformed from a quite unfamiliar investment instrument at the beginning of 2000 to a traditional structured security with trillions of dollars in value twenty years later. according to data from etfgi (2023), the global etf market has $9.6 trillion worth of assets. it has been constantly growing since 2000 (except for 2018 and 2022), with an annual growth rate of ca. 17% over the last 13 years. in 2017, the rate of growth hit a record of 38% per year. however, in 2021, the total value of the global etf market has dropped by 10%, from $10 trillion to $9 trillion. in 2023, the global market has witnessed a growth of 6.5%. it is essential to note that the usa accounts for the dominant share of the global etf market. as of 2023, the us market has the largest share (ca. 70%) in global etfs, accounting for $6.765 trillion, while the european etf market accounts only for $1.454 trillion (ca. 15%). more detailed information on the etf values’ dynamics for different regions can be found in appendix a. the second independent variable in this analysis is the consumer confidence index (cci), which is a generally available and widely used index that tracks consumers' expectations of the economy and finance. the version of the cci applied in the paper has been developed by the organisation for economic cooperation and development (oecd). the oecd (2023) indicates that the cci is a tool used to forecast the future trends of households’ consumption and saving habits. it is based on households’ responses related to their financial situation, economic sentiment, unemployment, and savings potential. if the index value is over 100, it means that consumers have a positive outlook towards the future economy, which may lead to increased spending on major purchases in the coming year. contrary to that, if the value is below 100, consumers may be pessimistic about the economy, resulting in a higher tendency to save and a lower willingness to consume. zurowska (2022) suggests that the increasing attractiveness of etfs for investors can be explained through instant (i.e., intraday) liquidity as well as a convenient tool for hedging and speculative trading. among other benefits of etf investing, cumming and monteiro (2023) argue that an etf is a liquid and diversified investment that could be tailored to an investment strategy. henriques et al. (2022) ground the growing interest in etfs on the opportunities for better portfolio diversification and the convenience of trading individual stocks. however, the specified benefits of etfs require further research since the global financial markets transform and face new challenges. despite all the opportunities, etfs may have their dark side. as singh (2022) puts forward, cryptocurrency etfs might be fraught with misleading, opaque structures and eccentric risk and performance reporting frameworks. li and zhu (2022) point out that one of the main etf drawbacks is the transmission mechanism of non-fundamental demand shocks to the index components. brown et al. (2021) also emphasise the significant impact of non-fundamental shocks on etfs. moreover, due to the spillover effects, underlying stocks may also face increased volatility. since the authors of the paper investigate the relationship between etfs and investor investor sentiment and euro area etfs sentiment, it is necessary to outline the basics of the investor sentiment concept. according to the effective market hypothesis and random walk theory (malkiel & fama, 1970; fama, 1965), employed by the traditional approach to studying finance, there are no opportunities for investors to use all publicly available information to discover securities with excess returns. although the traditional approach has failed to explain variations in stock returns from the fundamental factors’ positions. the same pertains to etfs’ pricing specifics (hong et al., 2022). on the other hand, from a behavioural finance perspective, investors’ opinions in the form of sentiment can impact asset returns. therefore, investor sentiment is one of the main research topics in behavioural finance, and there is a large strand of literature covering the investors’ sentiment impact on global markets (akçay, 2022; loang, 2022; vuong & suzuki, 2022). in simple terms, investor sentiment is the anticipation of securities prices by an investor. investor sentiment can also be presented in the form of factors that impact investors’ decisions and drive stock prices away from their fundamental values (piccoli & chaudhury, 2018; loang, 2022; liepert, 2024). however, the concept of investor sentiment is quite complex and elusive in nature, and it is possible to outline two main reasons for understanding investor sentiment. firstly, sentiment can drive prices away from underlying fundamentals, making it important for investment managers to monitor their portfolios for changes in sentiment (ahmed, 2020). investor sentiment can cause deviations in securities’ prices from their intrinsic values and promote ineffective pricing mechanisms in the market (de long et al., 1990). vuong and suzuki (2022) emphasise that many scientists have questioned and investigated how investor sentiment impacts securities returns, with some researchers arguing that investor sentiment has a significant impact on stock returns while others consider this relationship to be negligible. moreover, investor sentiment may cause significant spillover effects that can quickly spread throughout the market, impacting investor risk aversion and portfolio selection regardless of fundamental value (ahmed, 2020). therefore, investment sentiment presents a challenge not only for academics and practitioners but has become an important indicator for policymakers to follow trends that might disrupt the normal course of financial market development. examining the association between investor sentiment, as shown by the cci and the european etf price is the primary goal of the research. the principal aim of this study is to examine if consumer confidence levels in the euro area markets influence market-specific etf pricing. the paper investigated two hypotheses. the main hypothesis suggests a significant relationship exists between consumer confidence levels in the euro area markets and the pricing of ea market-specific etfs. the sub-hypothesis focuses specifically on the ishares msci eurozone etf and the impact of cci 19 euro area markets (ea19) on its pricing. 2. literature review the literature review section analyses three sets of papers that discuss various topics related to this research. the first set of papers explores the impact of consumer confidence on stock valuation. these papers can be divided into two groups based on their arguments. the first group presents arguments supporting the idea that consumer confidence affects stock prices. for instance, hsu et al. 168 pavlo dziuba et al. (2011) conducted a study investigating investor sentiment developments in stock markets. the researchers found a two-way causal relationship between consumer confidence and the stock market, indicating that changes in stock returns impact changes in consumer confidence, and vice versa. the authors suggest that stock returns can act as leading indicators of future economic development, influencing investors’ behaviour. they also propose that consumer confidence is a useful tool for predicting stock returns, especially for small-cap stocks. in contrast to that, the second group of papers, which focuses on the relationship between etfs and investor sentiment, provides opposing arguments, suggesting that there is no significant positive correlation between consumer confidence and stocks. bremmer (2008) examined the relationship between consumer confidence and stock prices and found that while stock prices impact consumer confidence in the short run, there is no long-run relationship between the two. the author also identified that unexpected changes in consumer confidence have a direct relationship with stock prices, while expected changes have no effect. these findings deviate from previous studies and complement the efficient market hypothesis (malkiel & fama, 1970). unlike the first set, this set presents quite consistent opinions, suggesting that consumer confidence has an impact on etf performance. ben-david et al. (2023) investigated exchange-traded funds and their performance over time. the scientists found that basic, inclusive etfs are associated with low-cost diversification, while specialised etfs that follow popular investment trends often include overvalued stocks and, therefore, perform poorly. the authors suggest that the underperformance of specialised etfs is caused by providers targeting investors’ overoptimistic beliefs by launching etfs after the peak of excitement around popular investment lines. the conclusion is that investor sentiment may be positively correlated with specialised etfs’ pricing. tseng and lee (2016) explored the features of the etf market in asia and examined the relationship between investor sentiment, etf liquidity, and each other. the researchers employed the generalised autoregressive conditional heteroscedastic (garch) model to analyse etfs from different asian markets. the authors found that investor sentiment has a significant impact on etf liquidity. notably, trading volume and investor sentiment were identified as significant factors for etf liquidity. this finding emphasises the importance of considering investor sentiment when making investment decisions and adjusting investment portfolios using etfs. several other studies in this set also support the idea that investor sentiment, as measured by consumer confidence, influences etf performance. ma et al. (2018) examined the relationship between market sentiment and the price deviations of asian etfs from their underlying fundamentals. the researchers confirmed that investor overreaction may cause etf prices to deviate from their intrinsic value, with sentiment playing a significant role. bahadar et al. (2019) demonstrated the significance of the behavioural approach in understanding the peculiarities of etf trading. the authors found strong herding behaviour among investors in the leveraged exchangetraded funds (letfs) market, particularly during bear markets and the global financial crisis. the third set of literature focuses on the etf markets in europe. yiannaki (2015) concludes that european etfs closely track their benchmarks without significant deviation. feder-sempach and miziołek (2023) support this conclusion and demonstrate that tracking errors in european etfs are generally low. however, zawadzki (2020) presents contrasting arguments, analysing the performance of ishares etfs in relation to their specific benchmarks. the author finds that etfs do investor sentiment and euro area etfs not consistently follow the performance of their benchmarks, and the level of discrepancies varies across different regions and markets. 3. materials and methods in this study, the authors have performed several methods of data analysis to investigate the relationship between the etfs and the cci in the euro area. the first method employed in the paper is panel data analysis, a widely used approach in empirical research to investigate the impact of independent variables over time. panel data regression models are useful for examining correlations between variables in situations where the data contains both a cross-sectional and a time-series component. by taking into consideration the temporal dynamics within each market, this method enables the researchers to account for potential variability across the various markets. the panel data model includes 7 euro area developed markets such as austria, belgium, france, germany, italy, the netherlands, and spain. the analysis of this study covers the time period from march 1st, 2015, to march 1st, 2023. this time period has been chosen to capture the recent trends in the economic development of the markets in question. it allows for the model to examine the relationship between the etfs and cci over a dynamic time frame that may provide new insights into significant trends in the data. this panel model uses data on specific etfs’ prices for each market. the prices of the following etfs have been used in the model: ishares atx ucits etf (de) (ex01) for austria, lyxor ucits bel 20 tr (bel) for belgium, ishares core dax® ucits etf (de) eur (acc) (gdaxiex) for germany, accion ibex 35 cotizado armonizado fi (bbvai) for spain, lyxor ucits cac 40 (dr) deur (cac) for france, lyxor ucits ftse mib (etfmib) for italy, ishares aex ucits (iaex) for the netherlands. these models offer thorough historical pricing data for their exchange-traded funds, which helped the researchers gather a large dataset encompassing the markets of interest in the euro area. the independent variable in the model is the cci value for each of the seven markets that have been extracted from the oecd database. the oecd is a well-established international organisation that collects and publishes a wide range of economic and social indicators, including the cci, which is a widely recognised measure of consumer sentiment. the study guarantees the accuracy and legitimacy of the data utilised in the analysis by utilising these reputable data sources. the values for etf price dynamics and cci dynamics for each market can be found in appendix b and appendix c. after conducting the panel data analysis, the authors of this study have decided to look at the hypothesis from another perspective and investigate the relationship between the ishares msci eurozone etf and the cci ea19, employing additional methods such as correlation analysis, regression analysis, and the granger causality test. time series plots for both ishares msci eurozone etf prices and values of cci ea19 can be found in appendix d and appendix e. correlation analysis has been used to explore the linear association and measure the strength and direction of the relationship between the two variables. this type of analysis is useful for understanding the extent to which the etf and cci are related and identifying any potential patterns in their relationship. regression analysis has been performed to estimate the impact of cci on the ishares msci eurozone etf. regression analysis helps to better understand the extent to which the cci might affect the 170 pavlo dziuba et al. ishares msci eurozone etf. finally, granger causality tests have been performed to investigate the direction of causality between the etf and the cci. granger causality tests are common means of data investigation that help determine whether the cci causes changes in the ishares etf or vice versa. the combination of econometric methods in question should help explore the nature of the relationship between consumer confidence and etf pricing in the markets of the euro area in a comprehensive and robust manner. it is necessary to note that the methods employed in this paper have several limitations. the authors only investigate the relationship between cci and etf in the euro area markets, limiting the paper’s findings only to one region. also, the researchers only consider stock etfs, ignoring other types of etfs such as bond, commodity, or currency etfs, which may behave differently in terms of consumer confidence. in addition, a single independent variable is employed, the consumer confidence index, to explain the pricing of etfs, while other factors that may affect etf pricing, such as interest rates, macroeconomic indicators, or political events, are omitted. finally, the time period may also limit the generalisability of the findings, as it only covers a specific period, not considering trends or cyclical patterns for the longer time period. 4. results 4.1 comprehensive analysis of the relationship between investor sentiment and european etf pricing the study has employed a random effects panel regression with etf prices as the dependent variable and cci values as the independent variable. this sample comprises seven markets (austria, belgium, france, germany, italy, the netherlands, and spain) and 678 observations. the following scatterplot demonstrates the relationship between the etf price and the cci value for each market (table 1, figure 1). table 1. panel data analysis results. dependent variable: etf_price model 1 model 2 cci_value -0.491*** (0.18) -0.49*** (0.18) constant 97.027*** (22.343) observations 678 678 r2 0.011 0.011 adjusted r2 0.001 0.009 f statistic 7.451*** (df=1; 670) 7.452*** note: *p<0.1; **p<0.05; ***p<0.01. source: compiled by the authors. investor sentiment and euro area etfs figure 1. the etfs prices and cci values for 7 countries. source: compiled by the authors. the results of the test demonstrate that consumer confidence has a significant negative effect on etf pricing. model 1 and model 2 show that for every one-unit increase in the cci value, the etf price decreases by approximately 0.491 units. this result may have important implications for investors who wish to trade etfs. a higher level of consumer confidence might suggest that the market is overpriced, and a market correction may occur soon. as a result, investors may think about decreasing their exposure to etf holdings in their investment portfolios under such circumstances. the standard errors are the same for both models, and the p-values show the coefficients’ statistical significance level. the constant term is also statistically significant in model 2, indicating that there are other factors besides cci that affect etf prices. the r-squared values are low, indicating that levels of consumer confidence explain only a small portion of the variation in etf prices. this leads to the conclusion that there should be other factors that may influence etf prices, such as financial stability, government regulations, political instability, or other market developments. it should be noted that these factors may be market-specific. in general, these results imply that consumer confidence might be an important factor to consider when analysing euro area etfs’ pricing, but it should not be the only factor considered. for example, investors should be aware of political and economic developments in each market that may affect etf prices. by considering both the market’s consumer confidence and market-specific factors, investors can make more informed decisions and potentially improve their investment performance (tkachenko et al., 2023; shubalyi, 2023). the impact of investor emotion on investment choices is one possible reason for this inverse relationship. a low level of consumer confidence may be an indication of greater economic uncertainty and household despair. investors may adopt a more risk-averse strategy as a result of this mindset, and they may be more likely to shift their holdings away from riskier assets like equities 172 pavlo dziuba et al. and etfs and towards safer alternatives. the peculiarities of exchange-traded funds may also exacerbate the detrimental impact of customer confidence on etf pricing. etfs are frequently utilised by investors as a practical and affordable means of obtaining broad market exposure. the demand for etfs that track certain sectors or the overall market may decline as consumer confidence declines, indicating a more pessimistic shift in investor sentiment and consequent pressure on etf prices. it is noteworthy that the findings of the panel data analysis show that the cci only partially explains the volatility in etf prices, indicating that other factors are probably more essential in determining how well etfs perform. the pricing of etfs in the euro area markets may be influenced by regulatory changes, industry-specific dynamics, macroeconomic conditions, and other investorrelated variables. the study proceeds by examining the link between the ishares msci eurozone etf and the cci ea19. it further employs various techniques, including correlation analysis, regression analysis, and the granger causality test, to investigate this relationship (table 2). table 2. correlation analysis matrix etf_price cci_value etf_price 1.0000000 0.2232225 cci_value 0.2232225 1.0000000 source: compiled by the authors. a correlation coefficient of 0.223 indicates a weak positive correlation between the ishares msci eurozone etf prices and the cci. it implies that there is a tendency for the two variables to increase or decrease together, but the strength of this relationship is relatively weak. this weak correlation may be attributed to other economic and/or financial causes that may affect this specific etf price or to the possibility of reverse causality. although the positive correlation indicates that slight gains in the price of the eurozone etf may be linked to improvements in overall consumer sentiment, the relationship’s limited strength emphasises the significance of a more comprehensive approach to etf dynamics analysis. further research is required to investigate these factors and their potential impact on etf investments. in general, changes in consumer confidence levels could influence investment decisions related to ishares etfs to some extent, but it is important to consider other factors as well to make effective investment decisions (figure 2). investor sentiment and euro area etfs figure 2. correlation between cci ea19 and ishares msci eurozone etf. source: compiled by the authors. the conclusions from the correlation study are supported by the scatterplot in figure 2, which shows the linear relationship between the two variables graphically. there is observable dispersion and a lack of tight clustering around the fitted line, which highlights the relatively weak correlation further in the scatter of data points surrounding the regression line. there are several potential reasons for the observed weak correlation. in the context of the european etf market, the cci could not be a comprehensive enough indicator of investor mood. the cci is intended to represent household spending intentions and the overall economic outlook, but it might not accurately represent the nuanced behavioural patterns and complicated motivations of the wide range of investors taking part in the eurozone etf market. etf trading is frequently heavily impacted by institutional investors, such as asset managers and pension funds, whose choices on what to buy may be influenced by more than just market mood. the member nations of the euro area have differing degrees of political stability, economic development, and financial market maturity. the cci data’s aggregation to the eurozone level might ignore significant country-specific variations in how investor sentiment influences the etf price. another possible explanation for the weak correlation could be the averaging out of these processes at the national level. the possible temporal lags and indirect impacts in the relationship between consumer confidence and etf performance must also be taken into account. etf price changes may not reflect shifts in consumer mood right away because it can take time for investors to modify their trading plans and portfolios. by quantifying the possible effects of changes in the cci ea19 on the ishares msci eurozone etf’s price in this subsection, the authors aim to provide additional insights into the dynamics of the relationship between investor mood and this important european exchange-traded fund. regression analysis enables a more comprehensive examination of data, including the strength and direction of the relationship and accounting for other potential factors that may have an impact. regression 174 pavlo dziuba et al. analysis may provide more insight into the factors that affect etf prices and guide more effective investment decisions. the results of the test are exhibited in table 3 and figure 3. table 3. regression analysis matrix residuals min 1q median 3q max -8.5953 -3.2488 -0.8432 2.8852 10.1847 coefficients estimate std. error t value pr(>|t|) (intercept) -21.163 27.3437 -0.774 0.441 cci_ea19_value 0.6091 0.2729 2.232 0.028* note: residual standard error: 4.746 on 95 degrees of freedom; multiple r-squared: 0.04983, adjusted rsquared: 0.03983; f-statistic: 4.982 on 1 and 95 df, *p-value: 0.02796. source: compiled by the authors. figure 3. regression between cci ea19 and ishares msci eurozone etf. source: compiled by the authors. the intercept term in the model represents the expected value of the dependent variable (ishares msci eurozone etf prices) when the independent variable (cci) is equal to zero, which is not a realistic scenario, and therefore this result does not have any practical significance. the coefficient of the independent variable (cci_ea19_value) suggests that for each unit increase in the cci, the ishares msci eurozone etf prices are expected, ceteris paribus, to increase by 0.6091 units. the pvalue associated with the independent variable is less than the significance level of 0.05, indicating that there is a significant relationship between the two variables. even though the intercept coefficient is statistically significant, it is relatively small and implies a weak positive relationship between the two variables. in addition, the adjusted r-squared value demonstrates that only about 3.98% of the variation in ishares msci eurozone etf prices can be explained by changes in the investor sentiment and euro area etfs consumer confidence index. this relatively weak explanatory power of the cci may be attributed to the complex and multifaceted nature of the factors influencing etf pricing in the euro area. in general, while the results of the regression analysis provide some insight into the relationship between the cci and ishares msci eurozone etf prices, they do not paint a clear picture of all the factors that might affect etf prices. therefore, it is necessary to explore other factors that might impact the prices of the ishares msci eurozone etf in addition to the cci for more comprehensive guidance for investment decisions related to euro area etfs. the granger causality test results demonstrate no evidence of granger causality between the ishares msci eurozone etf prices (value.x) and the cci (value.y). in the first model, which tests whether lagged values of both variables can predict changes in value.x, the p-value is greater than 0.05, which means one cannot reject the null hypothesis that lagged values of value.y do not grangercause value.x (table 4). table 4. granger causality test results model res. df df f pr(>f) value.x does granger cause value.y 94 1 0.337 0.563 value.y does granger cause value.x 94 1 0.001 0.97 source: compiled by the authors. similarly, in the second model, which tests whether lagged values of both variables can predict changes in value.y, the p-value is greater than 0.05, which means one cannot reject the null hypothesis that lagged values of value.x do not granger-cause value.y. therefore, based on the granger causality test results, it may not be concluded that changes in the euro area markets’ consumer confidence influence the ishares msci eurozone etf prices. granger causality’s absence indicates that a complex network of interrelated factors is likely influencing the relationship between these variables rather than a straightforward, unidirectional one. in addition, it is important to note that the results of the granger causality test complement the results of the previous tests and, therefore, help come to quite convincing conclusions on the hypothesis. 4.2 the impact of the external factors and behavioural patterns of investors in the euro area as was previously indicated, various external factors can have a substantial impact on the relationship between consumer confidence and etf pricing in the euro region. establishing a stable and consistent relationship between investor sentiment and etf performance might be difficult due to the potential breaks or shifts in the dynamics caused by them. for example, consumer confidence can be significantly impacted by global events. increased geopolitical tensions, unstable political environments, or significant international conflicts might reduce investor and household optimism, which can lower the consumer confidence index (kudrina & ivchenko, 2023). equity-based etfs may see a decline as a result of this shift in attitude, which may also cause a flight to safety and heightened risk aversion. 176 pavlo dziuba et al. in a similar way, the relationship between the cci and etf performance in the euro region can also be influenced by broader global economic trends. prolonged periods of robust economic expansion and growth in large economies, like the us or china, can positively influence consumer confidence and investor attitudes within the eurozone. the demand for etfs based in the eurozone may rise as a result, which might reinforce the favourable relationship between etf prices and the cci. nevertheless, even when the core fundamentals of the member companies of european etfs stay reasonably consistent, global economic downturns or recessions have the potential to undermine consumer confidence and adversely affect the pricing of these products. monetary policy changes also have a significant impact. demand for euro area etfs, for instance, may increase if expansionary monetary policies like interest rate reductions or quantitative easing are implemented. these measures can also increase consumer confidence. at the same time, the implementation of contractionary measures like the rise in interest rates may cause investors to become more riskapprehensive, which in turn may cause consumer confidence to weaken and etf prices to fall. even though the data point to a narrow direct correlation between the cci and european etf performance, it is still vital to take into account the more nuanced and indirect ways that investor emotion may affect both market dynamics and investment choices. according to behavioural finance theory, social circumstances, emotions, and cognitive biases can all have an impact on an investor’s decision-making process, and they do not always act in a perfectly rational way. changes in consumer confidence within the euro region may set off specific behavioural patterns that have an effect on etf trading and price. for instance, times of high consumer confidence may encourage retail investors to engage in more speculative and risky activity. they may also be more likely to allocate capital to riskier assets, such as specialised or sector-specific etfs. this may increase demand and raise costs for specific etfs, irrespective of their underlying principles. on the other hand, if investor confidence wanes, it could result in a widespread flight to safety and additional selling pressure on etfs, even in cases where the underlying companies’ long-term prospects remain unchanged. furthermore, shifts in consumer opinion may also have an impact on institutional investors’ behavioural habits. even though the cci is not the only factor influencing their choices, these bigger players might modify the allocations and holdings of etfs they own depending on how they see the state of the market. such behavioural modifications may have repercussions throughout the european etf market, adding to the intricate dynamics shown in the findings. 4.3 comparison of the relationship of the cci and etfs in europe to those in the us and asia there are several intriguing parallels and distinctions between the relationship between consumer confidence and etf pricing in the euro region and other significant financial markets, like the us and asia. a substantial amount of research has examined the relationship between exchange-traded fund performance and investor sentiment in the united states, with consumer confidence indexes frequently serving as proxies. in comparison to the findings from the euro area described in this research, studies conducted in the us have generally revealed a more significant association between investor sentiment and euro area etfs these characteristics. for instance, chen et al. (2017) looked at how investor sentiment affected the performance of national etfs in the us market. higher consumer mood levels were linked to better etf performance, according to the authors, especially for etfs that track industries that are vulnerable to shifts in household spending and the state of the economy. this is in line with the behavioural finance viewpoint, which contends that rising demand and inflows into equity-based investment products can be caused by a positive investor attitude. studies conducted in asian markets have also demonstrated the importance of investor emotion in influencing the dynamics of exchange-traded funds. tseng and lee (2016) showed that trading volume and other sentiment proxies, which measure investor sentiment, had a significant effect on etf liquidity in their examination of the asian etf market. this implies that, in the asian setting, exchange-traded fund demand patterns and pricing dynamics may alter in response to shifts in investor mood and expectations. on the other hand, the euro area analysis is finding of a relatively weak correlation between consumer confidence and etf price may point to a more intricate and varied set of factors affecting the european etf market. as the article discusses, the impact of investor mood alone on etf performance in the eurozone is probably not as great as that of macroeconomic conditions, regulatory changes, and other market-specific developments. these inconsistent results across several regional markets highlight how crucial it is to take into account the distinctive traits and structural variations of every financial ecosystem when evaluating the factors that influence etf pricing. the variable degree and nature of the relationship between consumer confidence and etf performance can be attributed to various factors, including the age of the etf market, the makeup and diversity of the investor base, and the general economic and political situation (lin, 2015). the results of this study provide valuable insights into the relationship between investor sentiment, specifically measured as cci, and the european etf market. the analysis employed panel data techniques and other methods to investigate the relationship between etfs and the cci. the results indicate that consumer confidence has a significant negative impact on etf pricing, suggesting that as consumer confidence decreases, etf prices tend to decline. however, it is important to note that the influence of consumer confidence on etf pricing explains only a small portion of the overall variation. this implies that while consumer confidence plays a role in shaping etf prices, it is not the sole determinant of their movement. the study reveals a weak positive correlation and regression relationship between ishares msci eurozone etf prices and the cci ea19. this suggests that there is a certain degree of association between the two variables, indicating that as consumer confidence improves, there is a tendency for eurozone etf prices to increase. although the analysis did not find evidence of a causal relationship between the cci and etf prices, the granger causality test did not provide significant results. this reveals that there are other factors at play that impact etf prices beyond the influence of consumer confidence. 5. discussion the reviewed studies examine this relationship from a number of angles, such as the impact of macroeconomic variables, the dynamics within particular market sectors and geographical areas, 178 pavlo dziuba et al. and the feelings of foreign, domestic, and global investors. these results emphasise how complex the relationship between sentiment and etf return is and how crucial it is to take a variety of factors into account when examining the pricing and performance of etfs. in the following discussion, the authors will summarise the most important takeaways from the literature, point out areas where research has converged and diverged, and go over the practical ramifications for traders and investors who want to use sentiment analysis in their etf investment strategy. the paper by yakubovskiy et al. (2020) supports the idea that the us financial markets have a significant impact on the global financial environment. for instance, the researchers argue that outward investments from the usa are more profitable for us investors than the profitability of inward investments in the usa for foreign investors. the paper by dziuba et al. (2022) discusses the relationship between levels of international diversification and the risk/return ratio. the authors conclude that a higher level of international diversification in a specific market does not imply a higher return on investment. the scientists argue that the most effective risk-return ratio can be found in markets with an average level of international diversification. therefore, there is more evidence in support of the idea that the international nature of etfs may imply a low correlation between etfs and local impact factors, such as the consumer confidence of local investors. zadoia et al. (2022) investigate the topic of market convergence within the eu environment. the researchers argue that, in some cases, several eu markets fall behind their european partners in terms of economic development. for instance, the authors found significant discrepancies between the euro’s real purchasing power and that of eu countries. therefore, aggregate consumer confidence across eu markets may lead to misleading results. chen et al. (2017) provide valuable insights into the relationship between investor sentiments, economic freedom, and the returns of country-specific exchange-traded funds. the author utilises unbalanced panel data from 27 ishares msci country-specific etfs and employs quantile regression to examine the influence of global, foreign, and us investor sentiments on etf returns. it is highlighted in the findings that investor sentiments, as well as the etf expense ratio, significantly determine the returns of the etfs traded in the us markets. the quantile regression approach reveals nuanced relationships between investor sentiments and etf returns across different quantiles. high-return etfs show positive sensitivity to changes in global sentiment, such as market turnover, vix (a measure of market volatility), and the us federal funds rate. furthermore, the author uncovers interesting findings regarding the effects of market volatility and foreign inflation on etf returns. the results indicate a reversal in the relationship, where returns from lower quantiles show a negative association with vix and foreign inflation, while higher quantiles exhibit a positive relation. this suggests that the impact of market volatility and foreign inflation on etf returns varies across different levels of returns. the scientist also highlights that not all components of economic freedom have equal effects on etf returns. the specific components of economic freedom that are examined in this study likely have varying impacts on the returns of the etfs studied, underscoring the importance of considering specific aspects of economic freedom when analysing etf performance. yang and chi (2023) provide additional insights on this topic, with a focus on the chinese market. the researchers examined the impact of investor sentiment on the volatility of etfs in china. the shanghai stock exchange composite index and the investor confidence index, two sentiment investor sentiment and euro area etfs measures, were used by the authors to determine the strong association between investor sentiment and etf volatility. in particular, the findings imply that while negative sentiment is linked to increased volatility, a positive mood among investors is linked to lower volatility in etf prices. these results align with the behavioural finance viewpoint presented in the current work, which holds that asset prices, including those of etfs, can be driven by investor mood away from their intrinsic values. optimistic investors might be more likely to purchase etfs, which would raise demand and possibly reduce volatility. meanwhile, etfs may face selling pressure during times of pessimism and risk aversion, which would increase the volatility of their prices. this supports the findings of the current paper by emphasising how crucial it is to take investor mood into account when analysing the dynamics of etf performance in the european context as well as in other international financial markets. investors in etfs and other market participants can use this data to guide their investing strategies and practices related to risk management. the scientists naeem et al. (2022) examine the relationship between investor sentiment and the returns of exchange-traded funds in different sectors of the european market. in the theoretical part of the work, the authors provide a comprehensive understanding of market sentiment, including its measures and their applications. in the practical part of the work, the relationship between sentiment and returns of sector etfs is examined using the analysis of the impulse response function (irf). the combined irf is constructed for the european sentiment indicator (esi), which serves as a proxy for general european sentiment and the returns of sector etfs. furthermore, the irf is employed for two specialised sentiment proxies – the financial services indicator and the consumer indicator – and the corresponding etf returns. the results of the analysis are reviewed, and recommendations are proposed for the successful use of sentiment indicators when trading specific sector stocks. this suggests that the work not only aims to investigate the relationship between sentiment and etf returns but also offers practical guidance on utilising sentiment indicators in the context of sector-based trading. by examining the relationship between investor sentiment and etf returns in different sectors of the european market, this work contributes to the understanding of market dynamics and the potential predictive power of sentiment indicators. the findings can be valuable for investors and traders who seek to incorporate sentiment analysis into their investment strategies, particularly when focusing on sector-specific etfs. goel and dash (2022) focus on the impact of macroeconomic variables and investor sentiment on etf returns. through a comprehensive analysis of panel data, the authors explore the effects of variables such as gdp growth, inflation rate, and interest rates on the relationship between investor sentiment and etf returns. the researchers suggest in the findings that the influence of investor sentiment on etf returns varies depending on the prevailing macroeconomic conditions. specifically, during periods of high gdp growth and low inflation rates, investor sentiment exhibits a stronger positive effect on etf returns. conversely, in times of economic downturn and higher inflation, the impact of investor sentiment on etf returns is diminished. this analysis provides valuable insights for investors, highlighting the importance of considering macroeconomic factors when interpreting the relationship between sentiment and etf performance. cullen (2023) delves into the role of investor sentiment in sector-specific etfs. by examining the performance of etfs across different sectors, the author uncovers the effects of sentiment on sector-based etf returns. the findings reveal that the relationship between investor sentiment and 180 pavlo dziuba et al. etf performance varies significantly across sectors. sectors that are more sensitive to consumer spending, such as retail and consumer goods, demonstrate a stronger positive correlation between sentiment and etf returns. conversely, sectors influenced by macroeconomic factors, such as manufacturing and industrial sectors, exhibit a weaker or even inverse relationship between sentiment and etf returns. these results highlight the importance of considering sector-specific dynamics when analysing the impact of investor sentiment on etf performance. fang et al. (2021) adopt a cross-country perspective to analyse the effects of investor sentiment on international etf returns. by examining a broad range of etfs across different countries and regions, the authors investigate how local and global investor sentiment interact to shape etf performance. the findings suggest that the impact of investor sentiment on etf returns is influenced by both local and global sentiment factors. in regions where local sentiment aligns with global sentiment, the effect on etf returns is amplified. conversely, in regions where local sentiment diverges from global sentiment, the impact on etf returns is attenuated or even reversed. this analysis highlights the significance of considering both local and global sentiment factors when assessing the relationship between investor sentiment and international etf returns. overall, the researchers underscore the complex and multifaceted nature of the relationship between investor sentiment and etf returns. these findings offer valuable insights for investors and traders, enabling them to make more informed decisions by considering specific factors and conditions that shape the impact of investor sentiment on etf performance. by leveraging such knowledge, market participants can enhance their investment strategies and navigate the dynamic landscape of etfs more effectively. the authors highlight the importance of considering other factors when analysing etf pricing. factors such as financial stability, government regulations, and marketspecific factors may also exert an influence on etf prices. these additional factors may explain the remaining variation in etf pricing that is not accounted for by consumer confidence alone. therefore, while consumer confidence is an important aspect to consider, it should not be the sole focus when assessing etf pricing dynamics. it is worth noting that this study has its limitations. the analysis relies on panel data and other methods, which have inherent limitations and may not capture the full complexity of the relationship between investor sentiment and etf pricing. additionally, the study focuses on the european etf market and may not be fully generalisable to other regions or asset classes. 6. conclusions this paper is devoted to the analysis of the relationship between investor sentiment in the form of a cci and the european etf market. the results of the panel data analysis demonstrate that consumer confidence has a significant negative effect on etf pricing, with a beta of -0.491 and p<0.01. the rsquared values are low (0.011), indicating that levels of consumer confidence explain only a small portion of the variation in etf prices. this suggests that other factors may influence etf prices. having regard to the previous studies, such factors may include financial stability, government regulations, political instability, or other market developments. the results of the panel analysis do not provide evidence in support of this hypothesis. investor sentiment and euro area etfs the correlation coefficient of 0.223 indicates a weak positive correlation between the two variables, so the two variables may increase or decrease together, but the strength of this relationship is relatively weak. the regression analysis results show a weak positive relationship between the two variables, with each unit increase in the cci leading to an expected increase of 0.6091 units in etf prices. the p-value of the independent variable is less than 0.05, indicating a significant relationship. the adjusted r-squared value of 0.03983 suggests that only 3.98% of the variation in etf prices can be explained by changes in the cci. the granger causality test results indicate that there is no evidence of granger causality between the two variables. this means that lagged values of the cci do not predict changes in the ishares msci eurozone etf prices, and lagged values of the etf prices do not predict changes in the consumer confidence index. the p-values obtained from the test were greater than 0.05, which suggests that the null hypothesis cannot be rejected. while analysing euro area etfs, consumer confidence may be a significant consideration, but it shouldn’t be the only one. aside from the variables listed above, investors and financial analysts should also look into other aspects of the market, including inflation, interest rates, government laws, financial stability, and market dynamics. investors might potentially enhance their investing performance in the european etf market by adopting a more thorough approach and accounting for a wider range of variables. it is imperative for policymakers to acknowledge the intricate correlation between investor attitude and etf pricing. additionally, they should contemplate the potential impact of alterations in consumer confidence and other macroeconomic elements on the stability and growth of the european etf market. though the main hypothesis has not been supported by the panel analysis results, the weak positive relationship found between ishares msci eurozone etf prices and the cci suggests that consumer confidence may still be an important factor to consider when analysing euro area etfs’ pricing, but it should not be the only factor. the granger causality test also did not provide evidence of a causal relationship between the two variables. therefore, it is important to investigate other factors that may affect etf prices, such as financial stability, government regulations, interest rates, inflation rates, or foreign exchange rates. references ahmed, b. (2020). understanding the impact of investor sentiment on the price formation process: a review of the conduct of american stock markets. the journal of economic asymmetries, 22, e00172. https://doi.org/10.1016/j.jeca.2020.e00172 akçay, s. (2022). investor sentiment and oil prices in the united states: evidence from a time-varying causality test. energy research letters, 3(2). https://doi.org/10.46557/001c.32633 bahadar, s., mahmood, h., & zaman, r. (2019). the herds of bulls and bears in leveraged etf market. journal of behavioral finance, 20(4), 408-423. https://doi.org/10.1080/15427560.2019.1553177 ben-david, i., franzoni, f., kim, b., & moussawi, r. (2023). competition for attention in the etf space. the review of financial studies, 36(3), 987-1042. https://doi.org/10.1093/rfs/hhac048 bremmer, d. (2008). consumer confidence and stock prices. paper presented at the 72nd annual meeting of the midwest economics association “beliefs, preferences and stock prices,” chicago, march 16. retrieved from https://www.rosehulman.edu/~bremmer/professional/bremmer_midwest2008.pdf brown, d. c., davies, s. w., & ringgenberg, m. c. (2021). etf arbitrage, non-fundamental demand, and https://doi.org/10.1016/j.jeca.2020.e00172 https://doi.org/10.46557/001c.32633 https://doi.org/10.1080/15427560.2019.1553177 https://doi.org/10.1093/rfs/hhac048 https://www.rose-hulman.edu/%7ebremmer/professional/bremmer_midwest2008.pdf https://www.rose-hulman.edu/%7ebremmer/professional/bremmer_midwest2008.pdf 182 pavlo dziuba et al. return predictability. review of finance, 25(4), 937-972. https://doi.org/10.1093/rof/rfaa027 chen, m.-p., lee, c. c., & hsu, y. c. (2017). investor sentiment and country exchange traded funds: does economic freedom matter? the north american journal of economics and finance, 42, 285-299. https://doi.org/10.1016/j.najef.2017.07.015 cullen, j. (2023). exchange-traded funds, capital market efficiency and systemic risk. in i. chiu & i. macneil (eds.), research handbook on global capital markets law (pp. 318-333). cheltenham: edward elgar publishing. https://doi.org/10.4337/9781800379305.00030 cumming, d. j., & monteiro, p. (2023). hedge fund investment in etfs. frankfurt: center for financial studies. https://dx.doi.org/10.2139/ssrn.4058765 de long, j. b., shleifer, a., summers, l. h., & waldmann, r. j. (1990). noise trader risk in financial markets. journal of political economy, 98(4), 703-738. retrieved from https://www.jstor.org/stable/2937765 dziuba, p., glukhova, d., & shtogrin, k. (2022). risk, return and international portfolio diversification: kmeans clustering data. baltic journal of economic studies, 8(3), 53-64. https://doi.org/10.30525/2256-0742/2022-8-3-53-64 etfgi. (2023). etfgi reports the global etfs industry gathered us$19.96 billion of net inflows in february 2023. retrieved from https://etfgi.com/news/press-releases/2023/03/etfgi-reportsglobal-etfs-industry-gathered-us1996-billion-net-inflows fama, e. f. (1965). random walks in stock market prices. financial analysts journal, 21(5), 55-59. https://doi.org/10.2469/faj.v21.n5.55 fang, h., chung, c.-p., lu, y.-c., lee, y.-h., & wang, w.-h. (2021). the impacts of investors’ sentiments on stock returns using fintech approaches. international review of financial analysis, 77, 101858. https://doi.org/10.1016/j.irfa.2021.101858 feder-sempach, e., & miziołek, t. (2023). how precisely european equity etfs mirror their flagship benchmarks? evidence from funds replicating performance of euro stoxx 50 index. journal of asset management, 24, 121-135. https://doi.org/10.1057/s41260-022-00287-9 goel, g., & dash, s. r. (2022). investor sentiment and government policy interventions: evidence from covid-19 spread. journal of financial economic policy, 14(2), 242-267. https://doi.org/10.1108/jfep-02-2021-0038 henriques, c. o., neves, m. e., castelão, l., & nguyen, d. k. (2022). assessing the performance of exchange traded funds in the energy sector: a hybrid dea multiobjective linear programming approach. annals of operations research, 313(1), 341-366. https://doi.org/10.1007/s10479-021-04323-6 hong, c. y., li, f. w., luo, j., & subrahmanyam, a. (2022). financial intermediaries and comovement in market efficiency: the case of etfs. https://dx.doi.org/10.2139/ssrn.4062962 hsu, c.-c., lin, h.-y., wu, j.-y., & commerce development research institute. (2011). consumer confidence and stock markets: the panel causality evidence. international journal of economics and finance, 3(6), 91-98. https://doi.org/10.5539/ijef.v3n6p91 investing.com. (2024). world etfs statistics. retrieved from https://www.investing.com/etfs/world-etfs kudrina, o., & ivchenko, yu. (2023). organisational and economic mechanism for ensuring the innovation and investment potential of an enterprise in the context оf post-conflict transformation. economic bulletin of cherkasy state technological university, 23(2), 24-30. https://doi.org/10.24025/23064420.69.2023.288376 li, f. w., & zhu, q. (2022). short selling etfs. the review of asset pricing studies, 12(4), 960-998. https://doi.org/10.1093/rapstu/raac005 liepert, m. (2024). a meta-analysis on private equity technology-driven value creation. scientific bulletin of mukachevo state university. series “economics,” 11(1), 29-39. https://doi.org/10.52566/msuecon1.2024.29 lin, j.-c. (2015). premiums on exchange-traded funds as a sentiment indicator: evidence from taiwan. international review of management & business research, 4(1), 40-58. loang, o. k. (2022). overreaction, investor sentiment and market sentiment of covid-19. vision: the journal of business perspective. https://doi.org/10.1177/09722629221087386 ma, j. z., ho, k.-c., yang, l., & chu, c.-c. (2018). market sentiment and investor overreaction: evidence from new york listed asian country exchange traded funds. emerging markets finance and trade, 54(11), 2455-2471. https://doi.org/10.1080/1540496x.2018.1464907 malkiel, b. g., & fama, e. f. (1970). efficient capital markets: a review of theory and empirical work. the journal of finance, 25(2), 383-417. https://doi.org/10.1111/j.1540-6261.1970.tb00518.x marta, t., & riva, f. (2022). do etfs increase the comovements of their underlying assets? evidence from a switch in etf replication technique. https://dx.doi.org/10.2139/ssrn.4079302 naeem, m. a., peng, z., bouri, e., shahzad, s. j. h., & karim, s. (2022). examining the asymmetries between equity and commodity etfs during covid-19. resources policy, 79, 103048. https://doi.org/10.1016/j.resourpol.2022.103048 oecd. (2023). consumer confidence index (cci). retrieved from https://doi.org/10.1093/rof/rfaa027 https://doi.org/10.1016/j.najef.2017.07.015 https://doi.org/10.4337/9781800379305.00030 https://dx.doi.org/10.2139/ssrn.4058765 https://www.jstor.org/stable/2937765 https://doi.org/10.30525/2256-0742/2022-8-3-53-64 https://etfgi.com/news/press-releases/2023/03/etfgi-reports-global-etfs-industry-gathered-us1996-billion-net-inflows https://etfgi.com/news/press-releases/2023/03/etfgi-reports-global-etfs-industry-gathered-us1996-billion-net-inflows https://doi.org/10.2469/faj.v21.n5.55 https://doi.org/10.1016/j.irfa.2021.101858 https://doi.org/10.1057/s41260-022-00287-9 https://doi.org/10.1108/jfep-02-2021-0038 https://doi.org/10.1007/s10479-021-04323-6 https://dx.doi.org/10.2139/ssrn.4062962 https://doi.org/10.5539/ijef.v3n6p91 https://www.investing.com/etfs/world-etfs https://doi.org/10.24025/2306-4420.69.2023.288376 https://doi.org/10.24025/2306-4420.69.2023.288376 https://doi.org/10.1093/rapstu/raac005 https://doi.org/10.52566/msu-econ1.2024.29 https://doi.org/10.52566/msu-econ1.2024.29 https://doi.org/10.1177/09722629221087386 https://doi.org/10.1080/1540496x.2018.1464907 https://doi.org/10.1111/j.1540-6261.1970.tb00518.x https://dx.doi.org/10.2139/ssrn.4079302 https://doi.org/10.1016/j.resourpol.2022.103048 investor sentiment and euro area etfs https://data.oecd.org/leadind/consumer-confidence-index-cci.htm#indicator-chart piccoli, p., & chaudhury, m. (2018). overreaction to extreme market events and investor sentiment. applied economics letters, 25(2), 115-118. https://doi.org/10.1080/13504851.2017.1302052 shubalyi, o. (2023). analysis of the behavior of european investors during the war in ukraine. economic forum, 1(4), 50-55. https://doi.org/10.36910/6775-2308-8559-2023-4-6 singh, p. (2022). is the financial market ready for cryptocurrency etfs? – a critical evaluation. the journal of risk finance, 23(4), 456-460. https://doi.org/10.1108/jrf-08-2022-241 tkachenko, a., levchenko, n., pozhuieva, t., sevastyanov, r., & levchenko, s. (2023). modified assessment methodology esg competitiveness of enterprises to a new generation of investors. iop conference series: earth and environmental science, 1254(1), 012126. https://doi.org/10.1088/17551315/1254/1/012126 tseng, y.-c., & lee, w.-c. (2016). investor sentiment and etf liquidity – evidence from asia markets. advances in management & applied economics, 6(1), 89-111. vuong, n. b., & suzuki, y. (2022). the moderating effect of market-specific factors on the return predictability of investor sentiment. sage open. https://doi.org/10.1177/21582440221114322 yakubovskiy, s., dominese, g., rodionova, t., & tsviakh, a. (2020). comparative analysis of the return on foreign investments of the united states, germany and japan. journal global policy and governance, 9(2), 17-27. https://doi.org/10.14666/2194-7759-9-2-002 yang, c., & chi, j. (2023). investor sentiment and volatility of exchange-traded funds: evidence from china. international journal of finance and economics, 28(1), 668-680. https://doi.org/10.1002/ijfe.2443 yiannaki, s. m. (2015). etfs performance europe – a good start or not? procedia economics and finance, 30, 955-966. https://doi.org/10.1016/s2212-5671(15)01346-5 zadoia, a. o., mahdich, a. s., & zadoia, o. a. (2022). macroeconomic criteria of currency and financial convergence: guidelines for ukraine. academy review, 57(2), 7-23. https://doi.org/10.32342/2074-5354-2022-2-57-1 zawadzki, k. (2020). the performance of etfs on developed and emerging markets with consideration of regional diversity. quantitative finance and economics, 4(3), 515-525. https://doi.org/10.3934/qfe.2020024 zurowska, e. (2022). etf primary market structure and its efficiency. philadelphia: university of pennsylvania. https://dx.doi.org/10.2139/ssrn.4143952 https://data.oecd.org/leadind/consumer-confidence-index-cci.htm#indicator-chart https://doi.org/10.1080/13504851.2017.1302052 https://doi.org/10.36910/6775-2308-8559-2023-4-6 https://doi.org/10.1108/jrf-08-2022-241 https://doi.org/10.1088/1755-1315/1254/1/012126 https://doi.org/10.1088/1755-1315/1254/1/012126 https://doi.org/10.1177/21582440221114322 https://doi.org/10.14666/2194-7759-9-2-002 https://doi.org/10.1002/ijfe.2443 https://doi.org/10.1016/s2212-5671(15)01346-5 https://doi.org/10.32342/2074-5354-2022-2-57-1 https://doi.org/10.3934/qfe.2020024 https://dx.doi.org/10.2139/ssrn.4143952 184 pavlo dziuba et al. appendix a aum’s values dynamics for three markets (usa market, european market, and other markets) for the period from 2010 to 2023. the worth of these aums in trillions 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 other 0.139 0.149 0.211 0.247 0.287 0.322 0.390 0.628 0.647 0.900 1.226 1.441 1.317 1.396 europe 0.281 0.268 0.331 0.395 0.438 0.488 0.542 0.762 0.726 0.974 1.194 1.506 1.333 1.454 us 0.893 0.938 1.212 1.612 1.918 2.060 2.465 3.300 3.310 4.320 5.316 7.074 6.378 6.765 0 2 4 6 8 10 12 us europe other investor sentiment and euro area etfs appendix b the country-specific etfs’ price dynamics for seven countries for the period from march 1st, 2015, to march 1st, 2023, denominated in eur note: the prices of the following etfs have been used in the model such as ishares atx ucits etf (de) (ex01) for austria, lyxor ucits bel 20 tr (bel) for belgium, ishares core dax® ucits etf (de) eur (acc) (gdaxiex) for germany, accion ibex 35 cotizado armonizado fi (bbvai) for spain, lyxor ucits cac 40 (dr) d-eur (cac) for france, lyxor ucits ftse mib (etfmib) for italy, ishares aex ucits (iaex) for the netherlands. source: compiled by the authors based on investing.com (2023). 186 pavlo dziuba et al. appendix c the country-specific cci value dynamics for seven countries for the period from march 1st, 2015, to march 1st, 2023. source: compiled by the authors based on oecd (2023). investor sentiment and euro area etfs appendix d time series plot of ishares msci eurozone etf prices for the period from march 1st, 2015, to march 1st, 2023, denominated in usd. source: compiled by the authors based on investing.com (2023). 188 pavlo dziuba et al. appendix e time series plot of aggregate consumer confidence index values for 19 euro area markets for the period from march 1st, 2015, to march 1st, 2023. source: compiled by the authors based on oecd (2023). 1. introduction 2. literature review 3. materials and methods 4. results 5. discussion 6. conclusions references appendix a appendix b appendix c appendix d appendix e © the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 2 (2025), pages 256-270 https://doi.org/10.17979/ejge.2025.14.2.12671 submitted: oct 23, 2025 accepted: nov 23, 2025 published: dec 17, 2025 article public policies and sustainable mobility in spain: empirical analysis of household expenditure on bicycles sánchez-gabarre, mary elena 1, * 1 universidade da coruña, department of economics, spain *correspondence: mary.e.sanchezg@udc.es abstract. this study investigates the socioeconomic determinants of household expenditure on sports bicycles in spain, providing novel insights into the affordability and equity dimensions of sustainable mobility policies. whilst cycling promotion has emerged as a central pillar of urban and environmental strategies, the economic factors influencing household-level bicycle adoption remain under-researched, particularly within southern european contexts. employing nationally representative microdata and a two-part modelling approach, this analysis explores how income, household characteristics and regional variations shape cycling-related expenditure patterns. the findings demonstrate that bicycle expenditure has become progressively concentrated among higher-income households, suggesting an evolution towards cycling being perceived as a discretionary purchase rather than an accessible transport solution. persistent disparities are observed across age groups, family structures and geographical regions, highlighting unequal access to the benefits of cycling. this research challenges prevailing assumptions regarding the inherent affordability of cycling and proposes targeted policy interventions to mitigate financial and structural barriers. by connecting macro-level sustainability objectives with micro-level consumption behaviour, this study contributes a conceptual framework for developing inclusive mobility policies that balance environmental ambitions with social equity considerations. the analysis offers timely evidence for policymakers seeking to align spain's cycling strategy with principles of distributive justice and broader sustainable development goals. keywords: cycling policy; economic policy; household expenditure; sustainable mobility jel classification: 1. introduction the promotion of sustainable lifestyles, in line with the paris agreement’s climate targets, is critical to limiting global warming to well below 2°c, with an aspirational goal of 1.5°c (ipcc, 2018). sustainable mobility plays a central role in this effort due to its capacity to reduce transport-related carbon emissions (creutzig et al., 2015). low-emission transport modes, enhanced public transport, and the adoption of electric vehicles are frequently cited as key tools for climate mitigation (sims et al., 2014; ipcc, 2022). within this framework, cycling has gained prominence as a cost-effective, https://creativecommons.org/licenses/by-nc/4.0/ public policies and sustainable mobility in spain health-promoting, and low-carbon mode of urban transport (pucher & buehler, 2017). beyond its environmental contribution, cycling also offers social and economic benefits. in sport economics, bicycles are recognised as both recreational and functional consumer goods. according to the sports habits survey (2022), 54.3% of the spanish population own a bicycle, a proportion rising to 66.4% among individuals who have engaged in sport during the past year. this data underscores the role of bicycles as key components of physically active lifestyles. public policy interventions, such as infrastructure investment, purchase subsidies, and bike-sharing systems, have fostered the growth of cycling in many urban areas (gössling & choi, 2015). empirical evidence from european cities shows that cycling infrastructure not only increases ridership but also mitigates congestion and emissions (fishman, 2015). in spain, cycling has shifted from a leisure activity to a mainstream mobility option, supported by increasing environmental awareness and improved infrastructure (observatorio de la bicicleta en españa, 2022; muñoz et al., 2013). higher cycling rates have been linked to public health gains through increased physical activity and improved air quality (de nazelle et al., 2011). in recent years, national and regional authorities have intensified their efforts to promote cycling through dedicated strategies and funding (ministry for ecological transition and the demographic challenge, 2023). nevertheless, the implementation of these policies remains uneven across administrative levels, and limited evidence exists on their economic impact, particularly regarding household-level bicycle spending. this paper contributes to filling this gap by empirically analysing the determinants of household bicycle expenditure in spain, with particular attention to its income elasticity. the findings aim to shed light on the economic barriers and facilitators affecting bicycle acquisition and use. since bicycle ownership is a prerequisite for both utilitarian and recreational cycling, understanding these patterns is critical for policy design in the areas of transport, health, and sustainability. furthermore, it should be noted that the empirical strategy integrates economic (becker, 1964; mcfadden, 1974), psychosocial (ajzen, 1991), and environmental-political (bronfenbrenner, 1979) frameworks, offering a multidimensional approach to analysing consumer behaviour in the cycling sector. 2. literature review cycling is widely recognised as a key component of sustainable mobility, particularly in its potential to reduce greenhouse gas emissions and support public health objectives (creutzig et al., 2015; sims et al., 2014). however, despite the extensive literature on infrastructure investment, modal choice, and behavioural determinants of cycling, relatively few studies have focused on household expenditure on bicycles, especially within european economies. understanding household spending behaviour is crucial for evaluating the distributive effects of cycling policies and for designing interventions that are both efficient and equitable. from an economic perspective, household expenditure on bicycles can be conceptualised as an investment in health and human capital. according to becker (1964), individuals allocate resources to improve their long-term productivity and well-being. in this framework, bicycles 258 sánchez-gabarre represent durable goods that facilitate regular physical activity, thereby contributing to improved health outcomes and potentially reducing future healthcare costs. this justifies the inclusion of bicycle-related spending within broader analyses of health-oriented consumption and the effectiveness of preventive public policy. transport economics offers a complementary framework. based on the random utility model developed by mcfadden (1974), households choose among transport modes by maximising expected utility, which depends on costs, time, convenience, and preferences. bicycle purchases reflect the relative utility of cycling compared to other modes, and this utility is shaped by contextual variables such as infrastructure availability, fuel prices, or perceived safety. policy instruments that alter these parameters, such as cycle lanes, intermodality schemes, or car disincentives, can directly affect both the probability and magnitude of bicycle-related expenditure. in addition, psychosocial factors influence consumption decisions. ajzen’s (1991) theory of planned behaviour identifies attitudes, subjective norms, and perceived behavioural control as predictors of intentional behaviour. households are more likely to invest in bicycles when they associate cycling with health, environmental responsibility, or social desirability, and when they feel capable of cycling safely in their environment. this implies that public policy must go beyond cost reduction and also address informational, cultural, and social barriers through awareness campaigns and communitybased programmes. the ecological systems theory of bronfenbrenner (1979) provides a useful framework for analysing the interaction between individual decisions and the institutional and environmental contexts in which they occur. from this perspective, spending on bicycles is not only determined by personal attributes such as income or education, but also by variables such as the built environment, political support and social norms. public interventions, such as subsidies, regulation, or investment in infrastructure, influence household behaviour through direct and indirect channels. finally, the diffusion of innovations framework (rogers, 1962) explains how new behaviours and consumption patterns spread across populations. in the case of cycling, symbolic associations with modernity, sustainability, or well-being may accelerate adoption, particularly among early adopters in urban areas. public authorities can leverage this dynamic through framing and communication strategies that present bicycles as aspirational goods aligned with contemporary lifestyle values. the limited empirical literature on bicycle expenditure confirms the relevance of these theoretical insights. thibaut et al. (2016), analysing belgian survey data, show that spending is influenced not only by demographics but also by cycling frequency, type of use, and online engagement. their concept of cycling capital, as a composite of identity, competence, and habit, demonstrates that behavioural and cultural factors are critical determinants of expenditure. environmental and spatial factors are also empirically significant. zhang et al. (2024) indicates that urban design, infrastructure quality, and perceived safety are strong predictors of cycling frequency and related expenditure. furthermore, access to shared bicycle systems can produce spillover effects on household transport behaviour and leisure-related consumption. despite these contributions, evidence for the spanish context remains scarce. most existing studies are situated in countries with consolidated cycling cultures and relatively homogeneous infrastructure. moreover, few papers analyse household-level bicycle expenditure using microdata, public policies and sustainable mobility in spain or explore its elasticity with respect to income and other structural variables. this gap is particularly relevant in light of recent policy efforts to promote active mobility in spain, which require evidencebased evaluation of their economic and behavioural impacts. this paper addresses these gaps by estimating a two-part model of household bicycle expenditure using spanish microdata. by linking expenditure decisions to sociodemographic, geographic, and behavioural variables, by explicitly estimating income elasticity, the study contributes to a better understanding of the economic drivers of active mobility and provides empirical evidence to support the design of equitable and effective public policy. 3. data and variables the data used in this analysis come from the household budget surveys (encuestas de presupuestos familiares, epf) conducted by the spanish national statistics institute (instituto nacional de estadística, ine). these surveys collect detailed information on private household expenditures, enabling the analysis of consumption trends across various goods and services. to identify goods and services specifically related to bicycles, we employed the european classification of individual consumption by purpose (ecoicop) at its most disaggregated level. in particular, bicycle-related expenditures were grouped following the same classification criterion used in the spanish yearbook of sports statistics, which associates household spending on sports with code 07130. a methodological clarification is warranted regarding the scope of our expenditure variable. the ecoicop code 07130 specifically captures expenditure on 'sports bicycles'. while this classification may encompass some bicycles used for utilitarian purposes, it primarily reflects recreational cycling consumption. this distinction is crucial for interpreting our findings, as the socioeconomic determinants and income elasticity of sports bicycle expenditure may differ substantively from those of strictly utilitarian bicycles, which are often lower-cost purchases for daily transport needs. furthermore, it is important to note that this category refers exclusively to conventional non-motorised bicycles. electric bicycles (e-bikes) are classified separately under a different expenditure code together with other motorised two-wheeled vehicles. this focused approach ensures conceptual consistency with active mobility while acknowledging that future research could benefit from analysing e-bike expenditure separately as statistical classifications evolve. for the empirical analysis, this study focuses on two reference years: 2019 and 2023. the selection of these years is guided by analytical and contextual considerations. on the one hand, 2019 represents the last full year prior to the covid-19 pandemic, capturing household expenditure patterns under pre-pandemic mobility conditions. on the other hand, 2023 reflects a post-pandemic scenario, shaped by a new policy and behavioural landscape in which sustainable transport modes, including cycling, have gained renewed prominence. comparing these two years allows us to explore structural changes in expenditure behaviour potentially driven by the long-term effects of the pandemic on urban mobility and health-oriented consumption. additionally, this choice avoids the inclusion of transitory distortions in household consumption data that may have occurred during the peak pandemic years (2020-2021), which were 260 sánchez-gabarre marked by exceptional restrictions on mobility and disruptions in retail markets. the exclusion of 2020 and 2021 is further justified by the methodological note provided by the ine, which indicates significant irregularities in consumption data during lockdown periods (ine, 2021). moreover, studies using the epf show that the 2020 lockdown produced heterogeneous and unexpected shifts in expenditure distribution and household savings behaviour (jiménez-martín et al., 2024), and analyses of the pandemic’s impact on consumption underscore that the weakening of demand was particularly pronounced for durable goods and did not fully revert in 2021 (banco de españa, 2025). finally, labour-market and consumption linkage studies reveal that the reestablishment of the prepandemic patterns of consumption was still incomplete in 2021 (alvargonzález et al., 2022). focusing on 2019 and 2023 thus ensures a more stable and representative comparison, while also enhancing the clarity of the empirical strategy without introducing excessive temporal variability into the model. to analyse and identify the determinants of household expenditure on bicycles in spain, total household expenditure is used as a proxy for household income, following established practice in consumption studies where self-reported income data may be unreliable or incomplete (deaton, 2019). this proxy enables the estimation of income elasticity of demand, a key objective of the analysis. a set of control variables captures relevant sociodemographic and geographic factors influencing bicycle-related spending. regional variables include all spanish autonomous communities and municipality size, categorised into four groups based on population (<20,000; 20,000–50,000; 50,000–100,000; >100,000 residents). autonomous communities are treated individually to capture region-specific differences in infrastructure, urban planning, and environmental awareness, while the municipality size grouping balances granularity with adequate sample sizes, ensuring robust spatial analysis. these controls account for heterogeneity in cycling infrastructure, policy promotion, and environmental conditions across regions, which can affect household cycling behaviour and expenditure (handy et al., 2014; pucher & buehler, 2008). household type reflects differences in consumption preferences and priorities, distinguishing between single-person households and those with children, for example. larger or more complex households may have distinct transport and leisure needs, potentially increasing the likelihood of bicycle ownership or use. similarly, the number of household members helps capture economies of scale or complementary consumption patterns within families (attanasio & weber, 2010). age is included to control for life-cycle effects, as consumption and engagement in physical or recreational activities vary across age groups (wagner & hanna, 1983). younger individuals may be more likely to use bicycles for commuting or sport, whereas older individuals may exhibit lower demand or different motivations. gender and marital status are standard demographic variables capturing behavioural and social differences in consumption. prior research has identified gender differences in sports consumption and transport preferences, which can influence spending on bicycles (thibaut et al., 2016; götschi et al., 2015). also, educational attainment serves as a proxy for human capital and is positively correlated with health awareness, environmental consciousness, and participation in sustainable transport public policies and sustainable mobility in spain modes (scheerder et al., 2011). higher education levels tend to be associated with a greater valuation of cycling as a healthy and eco-friendly activity (pucher & buehler, 2010) as well as stronger proenvironmental attitudes (diamantopoulos et al., 2003). finally, employment status of the household head reflects both economic capacity and time availability, two key components in the household production of leisure and mobility. following becker’s (1965) model, those with more stable or flexible employment may have more time and resources to allocate to cycling-related activities and purchases. employment status also influences travel behaviour and leisure time allocation, affecting transport mode choices and expenditure patterns (gatersleben & uzzell, 2007; sener et al., 2009). all expenditure variables have been standardised to ensure consistency across the spatial and temporal units of the household budget survey used. the combination of these variables provides a rich framework for capturing the multifaceted nature of household decision-making and identifying the key factors driving bicycle-related expenditure in spanish households. 4. methodology this study adopts a two-part modelling approach to empirically examine the determinants of household expenditure on bicycles in spain and to estimate its responsiveness to income. the choice of this methodological framework is motivated by the nature of the dependent variable (annual household expenditure on bicycles) which exhibits a large proportion of zero observations. these zero values correspond to households that did not report any bicycle-related spending during the reference period, either because they did not purchase a bicycle or because such purchases occurred outside the one-year time frame considered by the survey. given this distributional feature, conventional linear models such as ordinary least squares (ols) are inadequate, as they do not account for the underlying decision process that leads to zero expenditure. applying such models in this context would likely result in biased and inconsistent parameter estimates. as noted by mullahy (1998), separating the participation and consumption decisions provides consistent and interpretable estimates in health and consumption econometrics. to address this issue, the analysis employs a two-part model, which separately captures the extensive and intensive margins of household behaviour regarding bicycle consumption. the twopart model applied here is particularly suited for data with a large mass at zero, a common feature in consumption studies involving durable goods (duan et al., 1983). in the first stage, the model estimates the probability that a household incurs any expenditure on bicycles using a probit specification. this captures the extensive margin, reflecting the binary decision of whether or not to spend. in the second stage, the model estimates the level of expenditure conditional on positive spending, using a log-log specification applied to the sub-sample of households that reported non-zero expenditure. this functional form is particularly suitable for estimating income elasticity, as the coefficient on total household expenditure (used as a proxy for permanent income) can be directly interpreted as an elasticity. specifically, it reflects the percentage change in bicycle-related spending associated with a 1% change in household income. the value of this elasticity coefficient is central to understanding the consumption 262 sánchez-gabarre characteristics of bicycles. an elasticity greater than one indicates that bicycles are treated as luxury goods, implying that expenditure increases more than proportionally with income. an elasticity between zero and one suggests that bicycles are considered necessity goods, with expenditure rising less than proportionally. while theoretically possible, a negative elasticity, indicating that bicycles are inferior goods, is unlikely, given the generally positive associations of cycling with health, sustainability, and well-being. to ensure that the estimated effect of income is not confounded by other factors, the model includes a set of control variables that capture relevant sociodemographic and geographic characteristics. these include the household’s region of residence, the size of the municipality, the composition of the household, and individual attributes of the household head, such as age, gender, marital status, educational attainment, and employment status. the empirical framework is grounded in the theory of household production (becker, 1965), which conceptualises consumption not merely as the acquisition of market goods but as the production of utility-yielding activities using time and purchased inputs. within this framework, bicycles are understood as both durable goods and productive inputs for the generation of leisure, physical activity, and mobility, which are outcomes that households value and seek to maximise. the data used in this study come from the spanish household budget survey (encuesta de presupuestos familiares, epf). the survey provides rich microdata on household expenditure patterns and sociodemographic characteristics, allowing for a detailed and representative analysis. to ensure population-level representativeness, the estimations apply the sampling weights provided by the national statistics institute. robust standard errors are used throughout to account for potential heteroscedasticity in the second-stage regression. by jointly modelling the likelihood of bicycle expenditure and the conditional amount spent, and by explicitly estimating income elasticity, this methodological approach provides a comprehensive and policy-relevant perspective on household behaviour. it offers insights not only into the structure of demand for bicycles in spain, but also into broader questions related to sustainable mobility, consumer affordability, and equity in access to low-emission transport alternatives. 5. results and discussion the results of the two-part model estimations are presented in tables 1 and 2 and provide a comprehensive account of the determinants of household expenditure on bicycles in spain for the years 2019 and 2023. the findings confirm the pivotal role of income in shaping both the likelihood of bicycle-related spending and the conditional amount spent, with income elasticity figures suggesting substantial responsiveness. with regard to the determinants of spending decisions, table 1 shows those derived from the two-part model estimation. as might be expected, income has a direct and significant impact on spanish households' spending on bicycles. specifically, total household expenditure, a proxy for permanent income, is positively and significantly associated with both the probability of spending and the amount spent. this effect is robust across both years and underlines the importance of public policies and sustainable mobility in spain economic capacity in enabling bicycle-related purchases. however, beyond income, several key household characteristics play substantial roles. firstly, age of household head is negatively associated with the probability of spending in both years. older household heads are less likely to spend on bicycles, reflecting typical life-cycle consumption patterns and declining physical activity with age. interestingly, in 2023, age is positively associated with the amount spent, suggesting that among those who do spend, older individuals may purchase higher-end or electric bicycles, potentially for recreational or health purposes. secondly, about the household composition, we can see families with children under 16 show a positive association with bicycle expenditure, though this relationship does not reach conventional statistical significance in the 2023 probit model (p value=0.576). this supports the notion that bicycles are often purchased for children's mobility or recreational use. however, the amount spent by these households tends to be lower, which may reflect more basic or child-specific bicycle purchases rather than investment in expensive models. thirdly, households with children aged 0-15 show a significantly higher probability of spending, though often with lower conditional amounts. this suggests bicycle acquisition in these households is common but cost-constrained. the presence of household members aged 25-34 or 3564 is negatively associated with the amount spent, which could be due to competing financial priorities or substitution with other transport modes. regarding employment status, shows variable effects. unemployed individuals are less likely to report any expenditure, which is consistent with income limitations. retired individuals show higher spending in 2019 but significantly lower conditional spending in 2023. homeworkers and students exhibit inconsistent results, although students in 2019 report significantly lower conditional expenditure. also, on the one hand, education level has limited predictive power. university-educated individuals are more likely to spend in 2019, but this relationship fades in 2023. and, on the other hand, gender and marital status are generally not statistically significant, suggesting that cycling expenditure may no longer be strongly gendered or tied to family structure, at least in the aggregate. finally, the analysis reveals notable geographical variation. in both years, regions such as murcia, catalonia, and the basque country show higher probabilities of expenditure, which likely reflects better cycling infrastructure, more favourable policies, and more deeply rooted cycling cultures. the region of navarre shows strong participation effects in 2019, which dissipate in 2023. in contrast, madrid and valencia show an increase in participation and conditional spending in 2023, possibly reflecting recent investments in urban cycling networks. the significant negative coefficient for la rioja in 2023 (conditional spending) may reflect unique regional dynamics, potentially including a combination of sample size limitations that amplify statistical volatility alongside structural factors such as distinctive local market conditions or particularly pronounced population ageing trends in the region. in contrast, the size of the municipality does not appear to be a significant factor. this suggests that regional policies and infrastructure provision may be more important than mere urban density or population scale. 264 sánchez-gabarre table 1. two-part model estimation: log-log model. 2019 2023 probit reg. mco probit reg. mco coef. p>|z| coef. p>|z| coef. p>|z| coef. p>|z| c -5.13 0.000 -0.196 0.874 -6.692 0.000 -3.746 0.010 tot expenditure 0.374 0.000 0.654 0.000 0.491 0.000 0.875 0.000 region andalucía 0.248 0.009 -0.962 0.003 0.447 0.001 -0.033 0.925 aragón 0.019 0.870 -0.072 0.836 0.248 0.096 -0.636 0.114 asturias -0.019 0.882 0.412 0.174 0.344 0.055 -0.207 0.551 baleares -0.126 0.340 0.467 0.160 0.190 0.265 -0.656 0.082 canarias 0.033 0.771 0.370 0.207 0.332 0.018 0.243 0.571 cantabria 0.052 0.592 0.041 0.859 0.236 0.093 0.086 0.825 castilla y león -0.062 0.581 -0.004 0.986 0.440 0.002 0.293 0.396 c.–la mancha -0.084 0.325 0.117 0.569 0.374 0.001 0.152 0.627 cataluña 0.065 0.436 0.126 0.552 0.450 0.000 0.084 0.794 c. valenciana 0.112 0.271 0.089 0.722 0.343 0.014 -0.049 0.900 extremadura 0.029 0.769 -0.346 0.208 0.340 0.005 0.066 0.825 galicia -0.107 0.268 0.221 0.312 0.075 0.542 -0.070 0.830 madrid 0.069 0.548 -0.236 0.256 0.369 0.005 0.278 0.416 murcia 0.263 0.013 -0.162 0.577 0.657 0.000 -0.438 0.226 navarra 0.201 0.009 -0.608 0.011 0.485 0.000 0.021 0.944 país vasco 0.114 0.345 -0.085 0.759 0.602 0.000 0.132 0.707 rioja, la -0.191 0.522 0.162 0.602 -0.294 0.475 -4.175 0.000 ceuta 0.902 0.002 -0.062 0.918 0.321 0.247 -1.064 0.133 residents 10m<res<20m -0.039 0.624 0.192 0.349 -0.029 0.767 -0.388 0.103 20m<res<50m -0.055 0.437 -0.013 0.946 0.485 0.542 0.019 0.928 50m<res<100m -0.120 0.136 0.065 0.733 -0.002 0.985 -0.090 0.674 >100m res -0.090 0.141 -0.061 0.698 -0.039 0.605 -0.019 0.922 families couple without children 0.128 0.241 0.139 0.621 -0.094 0.439 0.075 0.795 couple with children < 16 years old 0.328 0.004 -0.380 0.137 0.780 0.576 -0.476 0.078 other family nuclei 0.269 0.048 -0.123 0.718 -0.144 0.329 -0.367 0.294 other households 0.414 0.008 -0.073 0.851 -0.098 0.636 -0.157 0.708 age -0.013 0.000 -0.011 0.203 -0.014 0.000 0.027 0.005 no. family members by age group 0 4 0.223 0.000 -0.423 0.000 0.123 0.057 -0.331 0.013 5 15 0.161 0.000 -0.118 0.200 0.169 0.000 -0.201 0.013 16 24 -0.133 0.026 0.118 0.438 -0.024 0.678 -0.106 0.476 25 34 -0.080 0.216 -0.278 0.117 0.052 0.530 -0.323 0.002 35 64 -0.025 0.705 -0.347 0.054 0.087 0.263 -0.195 0.190 65 – 84 -0.055 0.471 -0.657 0.020 0.004 0.973 -0.269 0.226 85 or more -0.301 0.040 1.085 0.007 -0.212 0.221 0.196 0.424 gender female -0.076 0.143 0.039 0.743 -0.062 0.298 -0.187 0.164 marital status couple -0.042 0.494 0.122 0.415 -0.076 0.284 -0.070 0.698 educational level intermediate studies 0.103 0.050 0.054 0.721 -0.093 0.190 0.221 0.223 university 0.133 0.025 0.110 0.489 -0.056 0.470 0.162 0.366 employment status unemployed -0.204 0.067 -0.308 0.188 0.079 0.599 -0.146 0.554 retired -0.055 0.577 0.556 0.074 -0.005 0.968 -0.773 0.052 student 0.114 0.808 -1.671 0.000 n/a n/a n/a n/a homeworker -0.725 0.001 -1.231 0.200 -0.390 0.189 0.276 0.520 inactive -0.135 0.558 0.880 0.000 -0.179 0.364 0.744 0.001 note. n/a: insufficient data for a reliable estimate. public policies and sustainable mobility in spain table 2 shows the income elasticity values derived from the estimate for spanish households' expenditure on bicycles. concretely, the income elasticity of bicycle expenditure, decomposed into participation elasticity (extensive margin), conditional elasticity (intensive margin), and total elasticity. the results show a clear shift in the nature of bicycle consumption between 2019 and 2023. table 2. results of the estimation of bicycle expenditure models: elasticities. mod. probit + log-log elast. partic. elast. cond. elast. total 2019 0.9039 0.6541 1.5580 2023 1.2709 0.8752 2.1461 the income elasticity estimates derived from the two-part model offer meaningful insight into the nature of household expenditure on sport-oriented bicycles in spain. in both 2019 and 2023, total income elasticity exceeds one (rising from 1.56 to 2.15) indicating that this category of expenditure is treated as a luxury good in economic terms. this result stems from increased responsiveness at both the extensive margin (probability of expenditure) and the intensive margin (level of spending among those who purchase). the growing elasticity suggests that, over time, bicycles have become more prominent in household budgets as discretionary recreational goods, potentially reflecting shifts in lifestyle preferences, growing health awareness, or post-pandemic changes in leisure consumption. however, the strong income dependence also raises distributional concerns. if higher bicycle expenditure is increasingly concentrated among higher-income households, this could reinforce inequalities in access to the physical, social, and environmental benefits associated with cycling. the increase in participation elasticity (from 0.90 to 1.27) suggests that the decision to initiate bicycle expenditure has become substantially more income-sensitive than the decision regarding how much to spend. this phenomenon could be explained by the post-pandemic 'normalization' of cycling as a mainstream recreational activity among higher-income groups, for whom initial adoption represents a discretionary lifestyle choice. concurrently, lower-income households may face persistent financial barriers to market entry, even for basic models, while those who do participate show more stable spending patterns relative to their income. the high-income elasticity estimates must be interpreted within the context of our 'sports bicycle' focus. it is plausible that expenditure on purely utilitarian bicycles would exhibit lower income sensitivity, as these represent essential transport goods for daily mobility rather than discretionary recreational purchases. this distinction suggests that policy interventions may need differentiation: while redistributive fiscal measures could address affordability across both segments, promotional campaigns might need tailoring, emphasizing practical utility for lowerincome groups versus recreational benefits for higher-income adopters. 266 sánchez-gabarre 6. conclusions this study has examined the determinants of household expenditure on sport bicycles in spain, using a two-part model applied to microdata from the spanish household budget survey for the years 2019 and 2023. the analysis focused on capturing both the decision to engage in bicycle-related spending and the amount spent, providing insights into the sociodemographic, geographic, and economic variables that influence cycling-related consumption. importantly, the findings have direct implications for the design of public policies aimed at fostering sustainable and inclusive mobility. as a main result, we can indicate that income consistently plays a significant role in explaining household spending on sports bicycles. both the probability of spending and the conditional amount spent rise with income, and this effect has strengthened over time. the estimated income elasticities suggest that sport bicycles are increasingly treated as luxury goods. this pattern indicates that, while income growth can stimulate demand for cycling-related goods, it also contributes to widening disparities in access to bicycles, particularly for low-income households. in this context, public policy must not assume that promoting cycling infrastructure alone will ensure equitable adoption. as banister (2008) points out, infrastructure provision is necessary but insufficient: effective transport equity requires complementary interventions that target behavioural, cultural, and financial barriers. moreover, the growing income dependence of bicycle expenditure raises distributional concerns. if cycling becomes increasingly income-sensitive, as the elasticity results suggest, public authorities risk reinforcing socio-economic inequalities in access to active mobility options (lucas, 2012). to counter this trend, a comprehensive policy approach should integrate three complementary strands of intervention. first, fiscal measures must directly address affordability barriers through means-tested purchase subsidies, vat reductions on non-motorised bicycles, and support for affordable second-hand markets. second, infrastructural policies need to ensure equitable access by developing safe, age-friendly cycling networks (including dedicated routes for schools and families) and promoting coherent regional implementation to address spatial disparities. third, educational and behavioural initiatives are essential to foster cultural change through public awareness campaigns, cycling education programmes in schools, and participatory planning to overcome safety concerns and social norms. this integrated strategy finds further application when considering specific demographic patterns. household structure and composition substantially influence expenditure decisions, as families with children under the age of 16 are more likely to spend on bicycles, albeit often at lower price points. public policy should respond by combining the three policy pillars: supporting familyoriented cycling through infrastructure adapted for children, such as school cycling lanes; implementing fiscal measures like discounts for multi-bicycle purchases; and developing educational programmes that combine mobility and education objectives. similarly, the negative association between age and the likelihood of spending highlights the relevance of life-cycle effects. to promote lifelong cycling, strategies for older populations should include investment in age-friendly infrastructure, public rental schemes with adapted models, and educational partnerships with the health sector that encourage recreational cycling as part of active ageing policies. the analysis also identifies significant regional disparities in expenditure behaviour, which public policies and sustainable mobility in spain likely reflect uneven implementation of cycling policies, differences in urban planning, and cultural variations in transport preferences. regions such as cataluña, the país vasco, and murcia show consistently higher levels of bicycle-related expenditure, suggesting that coherent cycling policies can positively influence household consumption. these findings underscore the need for decentralised and place-based policy approaches that combine the three pillars within regional contexts. while the spanish national cycling strategy provides a common framework, its effectiveness depends on strong implementation at regional and municipal levels through coordinated fiscal incentives, targeted infrastructural investments, and context-specific educational programmes. a final consideration concerns the scope of our analysis, which focuses specifically on conventional bicycles. while the exclusion of e-bikes may omit part of the recent market expansion, it allows a clearer focus on conventional bicycles, which are most closely associated with active mobility and direct physical activity benefits. as statistical classifications evolve to better distinguish electric and non-electric bicycles, future research could revisit the analysis to capture the changing composition of cycling expenditure. this integrated strategy, combining financial support, inclusive infrastructure, and targeted communication, is crucial to prevent cycling from becoming stratified along socioeconomic lines (gössling & choi, 2015) and to align spain's mobility transition with principles of equity and environmental justice. household bicycle expenditure is not merely a function of preferences or income, but is shaped by a complex interplay of structural, territorial, and policy factors. if cycling is to play a meaningful role in spain's sustainable mobility transition, public policy must move beyond isolated interventions and embrace a holistic approach that simultaneously addresses affordability, accessibility, and social acceptance across all demographic groups and territories. references ajzen, i. (1991). the theory of planned behavior. organizational behavior and human decision processes, 50(2), 179–211. https://doi.org/10.1016/0749-5978(91)90020-t alvargonzález, p., gómez, m., martínez-carrascal, c., pidkuyko, m., & villanueva, e. (2022). analysis of labor flows and consumption in spain during covid-19 (banco de españa occasional paper no. 2202). banco de españa. https://doi.org/10.2139/ssrn.4060174 attanasio, o. p., & weber, g. (2010). consumption and saving: models of intertemporal allocation and their implications for public policy. journal of economic literature, 48(3), 693–751. https://doi.org/10.1257/jel.48.3.693 banco de españa. (2025). algunos factores explicativos de la debilidad del consumo de los hogares después de la pandemia. boletín económico, 25(1), 1–20. https://www.bde.es/f/webbe/ses/secciones/publicaciones/informesboletinesrevistas/boleti neconomico/25/t1/fich/be2501-art01.pdf becker, g. s. (1964). human capital: a theoretical and empirical analysis. columbia university press. becker, g. s. (1965). a theory of the allocation of time. the economic journal, 75(299), 493–517. https://doi.org/10.2307/2228949 https://doi.org/10.1016/0749-5978(91)90020-t https://doi.org/10.2139/ssrn.4060174 https://doi.org/10.1257/jel.48.3.693 https://www.bde.es/f/webbe/ses/secciones/publicaciones/informesboletinesrevistas/boletineconomico/25/t1/fich/be2501-art01.pdf https://www.bde.es/f/webbe/ses/secciones/publicaciones/informesboletinesrevistas/boletineconomico/25/t1/fich/be2501-art01.pdf https://doi.org/10.2307/2228949 268 sánchez-gabarre borràs, j., jiménez-martín, s., & labeaga, j. m. (2024). the effects of the 2020 lockdown on consumption and savings in spain (fedea documento de trabajo eee 2024-23). fedea. https://documentos.fedea.net/pubs/eee/2024/eee2024-23.pdf bronfenbrenner, u. (1979). the ecology of human development: experiments by nature and design. harvard university press. https://doi.org/10.4159/9780674028845 creutzig, f., jochem, p., edelenbosch, o., mattauch, l., van vuuren, d., mccollum, d., & minx, j. (2015). transport: a roadblock to climate change mitigation? science, 350(6263), 911–912. https://doi.org/10.1126/science.aac8033 deaton, a. (2019). the analysis of household surveys: a microeconometric approach to development policy. world bank publications. https://doi.org/10.1596/978-1-4648-1331-3 de nazelle, a., nieuwenhuijsen, m. j., antó, j. m., brauer, m., briggs, d., braun-fahrländer, c., … dill, j. (2011). improving health through policies that promote active travel: a review of evidence to support integrated health impact assessment. environment international, 37(4), 766–777. https://doi.org/10.1016/j.envint.2011.02.003 diamantopoulos, a., schlegelmilch, b. b., sinkovics, r. r., & bohlen, g. m. (2003). can socio-demographics still play a role in profiling green consumers? a review and empirical investigation. journal of business research, 56(6), 465–480. https://doi.org/10.1016/s0148-2963(01)00241-7 duan, n., manning, w. g., morris, c. n., & newhouse, j. p. (1983). a comparison of alternative models for the demand for medical care. journal of business & economic statistics, 1(2), 115–126. https://doi.org/10.2307/1391852 fishman, e. (2015). bikeshare: a review of recent literature. transport reviews, 36(1), 92–113. https://doi.org/10.1080/01441647.2015.1033036 gatersleben, b., & uzzell, d. (2007). affective appraisals of the daily commute: comparing perceptions of drivers, cyclists, walkers, and public transport users. environment and behavior, 39(3), 416–431. https://doi.org/10.1177/0013916506294032 götschi, t., garrard, j., & giles-corti, b. (2015). cycling as a part of daily life: a review of health perspectives. transport reviews, 36(1), 45–71. https://doi.org/10.1080/01441647.2015.1057877 gössling, s., & choi, a. s. (2015). transport transitions in copenhagen: comparing the cost of cars and bicycles. ecological economics, 113, 106–113. https://doi.org/10.1016/j.ecolecon.2015.03.006 handy, s., van wee, b., & kroesen, m. (2014). promoting cycling for transport: research needs and challenges. transport reviews, 34(1), 4–24. https://doi.org/10.1080/01441647.2013.860204 instituto nacional de estadística (ine). (2019). encuesta de presupuestos familiares (epf): microdatos 2019. https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736176806& menu=resultados&idp=1254735976608 instituto nacional de estadística (ine). (2020). household expenditure survey 2020: press release. https://www.ine.es/en/prensa/epf_2020_en.pdf instituto nacional de estadística. (2021). encuesta de presupuestos familiares 2020: nota metodológica sobre el efecto de la covid-19. https://www.ine.es/metodologia/t25/efecto_covid_epf_2020.pdf https://documentos.fedea.net/pubs/eee/2024/eee2024-23.pdf https://doi.org/10.4159/9780674028845 https://doi.org/10.1126/science.aac8033 https://doi.org/10.1596/978-1-4648-1331-3 https://doi.org/10.1016/j.envint.2011.02.003 https://doi.org/10.1016/s0148-2963(01)00241-7 https://doi.org/10.2307/1391852 https://doi.org/10.1080/01441647.2015.1033036 https://doi.org/10.1177/0013916506294032 https://doi.org/10.1080/01441647.2015.1057877 https://doi.org/10.1016/j.ecolecon.2015.03.006 https://doi.org/10.1080/01441647.2013.860204 https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736176806&menu=resultados&idp=1254735976608 https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736176806&menu=resultados&idp=1254735976608 https://www.ine.es/en/prensa/epf_2020_en.pdf https://www.ine.es/metodologia/t25/efecto_covid_epf_2020.pdf public policies and sustainable mobility in spain instituto nacional de estadística (ine). (2023). encuesta de presupuestos familiares (epf): microdatos 2023. https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736176806& menu=resultados&idp=1254735976608 instituto nacional de estadística (ine) & consejo superior de deportes (csd). (2022). encuesta de hábitos deportivos en españa 2022. https://www.educacionfpydeportes.gob.es/servicios-alciudadano/estadisticas/deportes/encuesta-habitos-deportivos-espana.html ipcc. (2018). global warming of 1.5°c. intergovernmental panel on climate change. https://www.ipcc.ch/sr15/ ipcc. (2022). climate change 2022: mitigation of climate change. intergovernmental panel on climate change. https://doi.org/10.1017/9781009157926 lucas, k. (2012). transport and social exclusion: where are we now? transport policy, 20, 105–113. https://doi.org/10.1016/j.tranpol.2012.01.013 martens, k. (2016). transport justice: designing fair transportation systems. routledge. https://doi.org/10.4324/9781315746852 mcfadden, d. (1974). conditional logit analysis of qualitative choice behavior. in p. zarembka (ed.), frontiers in econometrics (pp. 105–142). academic press. ministerio para la transición ecológica y el reto demográfico. (2023). national cycling strategy 2030. government of spain. https://www.miteco.gob.es/es.html mullahy, j. (1998). much ado about two: reconsidering retransformation and the two-part model in health econometrics. journal of health economics, 17(3), 247–281. https://doi.org/10.1016/s01676296(98)00030-7 muñoz, b., monzón, a., & lois, d. (2013). cycling habits and other psychological variables affecting commuting by bicycle in madrid, spain. transportation research record, 2382(1), 1–9. https://doi.org/10.3141/2382-01 observatorio de la bicicleta en españa. (2022). annual report on cycling in spain. https://www.redbici.org/wp-content/uploads/2022/12/observatorio-de-la-bicicleta2022-diciembre-2022.pdf pucher, j., & buehler, r. (2008). making cycling irresistible: lessons from the netherlands, denmark, and germany. transport reviews, 28(4), 495–528. https://doi.org/10.1080/01441640701806612 pucher, j., & buehler, r. (2010). cycling to sustainability in amsterdam. sustain, 21(1), 36–40. https://vtc.rutgers.edu/publication/cycling-to-sustainability-in-amsterdam/ pucher, j., & buehler, r. (2017). cycling towards a more sustainable transport future. transport reviews, 37(6), 689–694. https://doi.org/10.1080/01441647.2017.1340234 rogers, e. m. (1962). diffusion of innovations. free press. scheerder, j., vos, s., & taks, m. (2011). expenditures on sport apparel: creating consumer profiles through interval regression modelling. european sport management quarterly, 11(3), 251–274. https://doi.org/10.1080/16184742.2011.577931 sener, i. n., eluru, n., & bhat, c. r. (2009). who are bicyclists? why and how much are they bicycling? transportation research record, 2134(1), 63–72. https://doi.org/10.3141/2134-08 sims, r., et al. (2014). transport. in climate change 2014: mitigation of climate change (pp. 599–670). ipcc. https://doi.org/10.1017/cbo9781107415416.014 thibaut, e., vos, s., lagae, w., puyenbroeck, t., & scheerder, j. (2016). partaking in cycling, at what cost? determinants of cycling expenses. international journal of sport management and marketing, 16(3–6), 221–238. https://doi.org/10.1504/ijsmm.2016.077932 https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736176806&menu=resultados&idp=1254735976608 https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736176806&menu=resultados&idp=1254735976608 https://www.ipcc.ch/sr15/ https://doi.org/10.1017/9781009157926 https://doi.org/10.1016/j.tranpol.2012.01.013 https://doi.org/10.4324/9781315746852 https://www.miteco.gob.es/es.html https://doi.org/10.1016/s0167-6296(98)00030-7 https://doi.org/10.1016/s0167-6296(98)00030-7 https://doi.org/10.3141/2382-01 https://www.redbici.org/wp-content/uploads/2022/12/observatorio-de-la-bicicleta-2022-diciembre-2022.pdf https://www.redbici.org/wp-content/uploads/2022/12/observatorio-de-la-bicicleta-2022-diciembre-2022.pdf https://doi.org/10.1080/01441640701806612 https://vtc.rutgers.edu/publication/cycling-to-sustainability-in-amsterdam/ https://doi.org/10.1080/01441647.2017.1340234 https://doi.org/10.1080/16184742.2011.577931 https://doi.org/10.3141/2134-08 https://doi.org/10.1017/cbo9781107415416.014 https://doi.org/10.1504/ijsmm.2016.077932 270 sánchez-gabarre wagner, j., & hanna, s. (1983). the effectiveness of family life cycle variables in consumer expenditure research. journal of consumer research, 10(3), 281–291. https://doi.org/10.1086/208967 zhang, x., lim, e. s., & chen, m. (2024). promoting sustainable urban mobility: factors influencing e-bike adoption in henan province, china. sustainability, 16(22), article 10136. https://doi.org/10.3390/su162210136 https://doi.org/10.1086/208967 https://doi.org/10.3390/su162210136 1. introduction 2. literature review 3. data and variables 4. methodology 5. results and discussion 6. conclusions references the definitive vat system and its impact on tax collection european journal of government and economics 10(2), december 2021, 146-166 european journal of government and economics issn: 2254-7088 the definitive vat system and its impact on tax collection hana zídkováa *, kristýna balíkováa a prague university of economics and business, czechia * corresponding author at: hana.zidkova@vse.cz abstract. value added tax (vat) is a significant source of fiscal revenues in the eu. however, the vat treatment of cross-border supplies enables large-scale tax frauds, such as the missing trader intra-community (mtic), which takes each year billions of euros from member states' public budgets. in 2016 a definitive vat system was proposed by the european commission to respond to the shortcomings of the current temporary system. this new system should reduce the possibilities of mtic fraud for intra-community transactions through the collection of vat by the supplier in the same way as for domestic transactions. the tax collection by the supplier would impact the administrative costs of the financial authorities. this paper contributes to the discussion about the advantages and disadvantages of the newly suggested system. the analysis focuses on the study of the change in administrative costs and vat revenues for individual member states and across the eu. the results are that after implementing the definitive vat system, total administrative costs of the member states would increase at least by eur 107 million, whereas total vat revenues would rise by eur 40 billion. this indicates the overall positive impact of the definitive vat system for the eu. however, individual member states would not benefit equally. the net exporters, whose intra-community supplies exceed the intra-community acquisitions, would spend more than others for the collection of vat in connection with the international trade of goods. keywords. definitive vat system, international trade statistics, vat action plan, temporary vat system, mtic fraud jel codes. h21, h26, k34. doi. https://doi.org/10.17979/ejge.2021.10.2.7803 1. introduction the value added tax (vat) is one of the main sources for financing public spending in the eu member states. unfortunately, the collection of vat is affected by missing trader intra-community frauds (mtic frauds). according to poniatowski et al. (2020), the value of vat gap, which represents the volume of vat evasion, including mtic fraud, reached eur 137 billion in 2019 in the whole eu. fighting against vat fraud, therefore, belongs to major priorities of eu’s tax policy (european commission, 2010). in this connection, a draft amending the council directive on common system of vat (european council, 2006) was filed in 2017 (european commission, 2017). the purpose of the draft was the transformation of current vat treatment applicable for intra-community supplies of goods between businesses (intra-community b2b supplies) to a new mechanism, socalled definitive vat system. the current regime was implemented in 1993 and was meant to be temporary. as the main disadvantages of the current system are considered, both predisposition to tax evasion and high demands on tax compliance of businesses while trading on the intracommunity level. https://doi.org/10.17979/ejge.2021.10.2.7803 zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 147 the key difference between the current and the definitive system lies in replacement of two mirror “taxable supplies” requiring mirror reporting involved in cross-border transactions. these two “mirror supplies” are intracommunity supply and intracommunity acquisition. they will be replaced by only one transaction supply of goods within eu. this supply will be taxed by suppliers in their own countries by the vat rate of the destination country. the collection of vat on b2b cross-border supplies will be organized through so-called one stop shop. in practice, when doing business on the intra-community level, the invoice would include tax of the country of destination (where the goods are being delivered) while the supplier would pay the tax to the tax authority in his own country (european commission, 2017a). subsequently, local tax authorities would transfer collected vat to tax authorities in destination countries. the system would be working similarly to mini one stop shop that is already used for electronic, telecommunication and broadcast services provided to non-taxable persons and recently (since 1. july 2021) expanded to cross-border supplies of goods to non-taxable persons (b2c cross border supplies). according to european commission (2015), the drafted definitive system would lead to a reduction of compliance duties of businesses. the study also shows that the definitive system reduces the mtic frauds in the eu, which would result in higher vat revenues for public budgets. nonetheless, we believe that the benefit of the definitive system would not be equal in all eu member states. the value of savings would be affected by the volume of mtic frauds in each state. the higher the size of these frauds in the country is, the higher the amount of additional revenue should be expected. negative consequences of the definitive system in terms of increased tax administration costs will most likely arise due to the additional transfer of collected vat between financial authorities of different member states (lamensch, 2012). like the additional vat revenues, the change in the administrative costs would not be uniform in all engaged states. a relatively better position (decrease in administrative costs after implementing the definitive system) might be expected in member states with higher b2b intra-community acquisitions of goods than b2b intra-community supplies as they will stop collecting the vat from acquisitions and start collecting it from the intracommunity supplies. on the other hand, states with more b2b intra-community supplies than acquisitions would collect payments for other countries, and their administration costs would increase relatively more compared to the current system, where they levy vat on intra-community acquisitions and not intracommunity supplies. provided that there exists a positive relationship between additional vat revenues achieved by the definitive system and the change in administrative costs compared to the current situation, the implementation of the new system would be reasonable and thus acceptable for all member states. on the condition that additional tax revenue is higher than additional administration costs, the transition into a definitive system would be advantageous also to countries with a higher amount of intra-community acquisitions. this paper analyses impact of implementing the definitive vat system on eu member states through calculation of the change in administration costs, additional tax revenues and comparative analysis of these two factors. firstly, we examine which of the member states are zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 148 more supplyand acquisition-oriented (net exporters and importers). we only use the difference between b2b intracommunity supplies and acquisitions because we want to find out what the change in administrative costs will be. in the original system, the state had to collect vat from b2b cross-border acquisitions, and in the new system, it will collect vat from b2b intracommunity supplies. therefore, if the supplies are higher, the state’s administrative costs will increase. if, on the contrary, the acquisitions are higher, his administrative costs will decrease. the goal is to calculate the amount of the change in administration costs incurred by individual countries while collecting vat for other member states in the definitive system. secondly, the paper provides calculation of additional tax revenue for each member state and examines whether there is a correlation between the change in administration costs and additional vat revenues. 2. literature review the current system of vat is based on the invoice-credit method and fractional payments. in such system, the supplier collects tax from the transaction while the purchaser has a right to deduct the tax paid. therefore, the tax administrator collects the vat gradually from all vat payers inside the distribution chain (e.g. schenk et al., 2015; terra and kajus, 2015). the described system has been impacted by temporary rules for cross-border transactions applicable since 1993, when customs borders between member states were cancelled. under this temporary system, the invoice for a b2b cross-border transaction does not include vat (intracommunity supply is vat exempt) as the tax is fully levied by a state of consumption on the b2b intracommunity acquisition of goods. at the same time, to fulfil the purposes of vat and its neutrality, the state of origin enables the supplier to deduct vat incurred on purchases related to the b2b intracommunity supply of goods, i.e. vat exemption with credit is applicable on the b2b intracommunity supplies. the element of tax exemption likely increases vat fraud opportunities. fraudsters exploit the vat exemption for b2b cross-border transactions to conduct mtic frauds (european commission, 2016). mtic fraud mainly represents a situation when a taxpayer acquires goods from other member state vat free and supplies the goods locally afterwards. the fraud appears when the taxpayer (here called missing trader) vanishes, does not file the vat return, becomes no-contact. this missing trader does not pay to the financial authorities the tax on the local supply of goods despite selling the goods, including vat (fedeli and forte, 2011). the next member in the distribution chain – local purchaser – frequently does not know about the missing trader’s intent and therefore normally pays vat in the price to the supplier and then applies for vat deduction (ainsworth, 2009). the definitive vat system keeps the principle of fractional payments, which have the advantage of the so-called self-policing nature (pomeranz, 2015). the tax administrator is permitted to check whether the purchasers’ claim on input vat corresponds with vat declared zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 149 by the supplier (cnossen, 2010). each vat payer is motivated to get tax documents from the supplier to claim back vat on input. input vat can only be claimed through the vat return. in the definitive system, vat on theb2b cross-border acquisition of goods is not paid by the buyer anymore. the tax from the acquisition of goods and the right to deduct it are not cumulated with one and the same vat payer. therefore, a vat payer who becomes a missing trader that does not file the tax return, is not only avoiding paying the output tax from his local supply but also cannot claim back the input tax incurred in the price that he paid for the goods purchased from another member state. in other words, since vat definitive system replaces the tax exemption on cross-border transactions, the purchaser acquires from other member state taxed goods. therefore, if he/she disappears without payment of output vat on subsequent local supply, he/she must sacrifice input tax deduction (amand, 2014). this should lead to the reduction of mtic frauds in eu (european commission, 2015). however, there may still be tax evasion if a supplier from another member state does not pay the tax in his state (state of origin) and the buyer claims it back in his state (state of consumption). this may be the case, in particular, if information systems are not interconnected between member states. the main advantage the definitive vat system would be a significant reduction of mtic fraud across the eu. however, some authors also point out several disadvantages of the definitive system. for example, vat expert group (2016) criticizes administrative burdens of businesses and the possibility of making a mistake while selecting the correct vat rate or while assigning the transport in a chain transaction with goods. furthermore, the necessity of mutual payments between member states is another setback. zídková and šťastná (2019) also notice potential deterioration in businesses cash flow due to including vat in the price for cross-border supplies of goods and the obligation to remit this vat to the financial authorities often before receiving the payment from the customer. at the same time, the authors point out the loss of financial authorities' control over cross-border transactions should the intra-community sales list be canceled. de la feria (2018) criticizes insufficient harmonization of tax bases in the present system, and catarino and moraes e soares (2019) notice insufficient harmonization of tax rates. nevertheless, all these disadvantages may be outweighed by increased member states' tax revenues. european commission (2017a) estimates an increase in vat revenues across the eu of eur 40 billion per year due to the implementation of the definitive vat system. this estimate is based on the size of carousel frauds (eur 50 billion per year) and on the estimate of 20% margins on intra-community supplies. the definitive system would lead to a revenue loss in value of eur 10 billion instead of eur 50 billion because missing traders would not pay vat on output but also would not have a right to claim input vat. the tax revenue lost in such a case would only be the vat from their margin and not the vat calculated from the total value of the goods as in the current system1. however, mtic fraud is not the only form of vat evasion. there are also other types of fraudulent behavior, e.g., hiding taxable revenues, intentional use of lower rates, 1 in the current system, the intracommunity supply of goods is vat exempt, and the missing trader does not incur any input vat when purchasing the goods cross-border. therefore, non-filing the tax return doesn’t mean the loss of vat deduction. zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 150 or avoiding the vat registration (tait, 1988, keen and smith, 2006). the implementation of the definitive vat system would not solve these forms of vat evasion. it is obvious that there would not be an equal benefit of the definitive system in all member states because the additional tax revenues after its implementation would differ, inter alia, due to economy’s business structure and the volume of vat evasion caused by the mtic fraud in each country. therefore, in our paper, we try to estimate potential additional vat revenues for individual member states in addition to calculating their change of administrative costs. the change in their administrative costs depends on the trade balance of that state with other member states, as explained above. the increase in vat revenues depends on the volume of carousel fraud present in that member state. the result of the overall trade balance of the state with other member states and mtic fraud present in that state are not related in any way. however, they express the cost and benefit of the definitive system for each state. therefore, we will analyze the position of each state in these two respects, and we are interested in whether the costs and benefits are correlated or whether some member states incurred only costs whereas others only benefits. 3. data and methods this section shows data and explains methods used for the calculation of the definitive system’s impact on each member state. firstly, the data used for the estimation of additional vat revenue and the change of administration expenses are presented. secondly, the chapter describes correlation analysis applied to find a relationship between additional vat revenues and the change of administration costs. 3.1 trade balances and their adjustments for the purposes of our analysis, data from eurostat’s international trade in goods statistics (itgs) were utilized. itgs includes both intra-eu and outer-eu business transactions. the amounts of imported and exported goods per each state and year since 1988 are collected there (eurostat, 2020a). we used only the data on exports and imports within the eu. in the terminology of vat legislation, exports are called intra-community supplies, and imports are intra-community acquisitions. data of import and export should be adjusted for the purposes of our calculation. it is necessary to remove cross-border transactions, which would be exempted or not taxed in the vat definitive system. these transactions can be found in the council directive on common system of vat (european council, 2006) – further referred as vat directive. although the vat directive is not enforceable in member states directly, every country is obliged to implement principles and the content of this directive into its own legislation. therefore, we assume that transactions formulated in the vat directive as non-taxable are treated so in all zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 151 member states, and we removed them from the itgs. to be precise, we focused on the phrase “member states shall exempt” used by the directive. in such cases, the goods are exempted across the eu and there are no doubts about taxing them. oppositely, if the wording of the vat directive contains “member states may exempt”, we concluded not to remove the goods from the statistics. the reason is that the exemption of such goods is voluntary, and not all states apply that. based on article 132 of directive, activities in the public interest are exempted. these activities include inter alia postal services, hospital and medical care, services of dental technicians, services of young people’s education or specific culture services. further, article 136 also exempts the supply of goods utilized solely for all exempted activities in the article 132. in other words, member states are obliged to exempt, for instance, medical supplies. other exemptions are provided in article 148 of directive, which focuses on the exemption of international transportation. based on this article, the supply of goods for fueling, reparation, maintenance and provisioning of vessels used for selected purposes is exempted as the activities itself. the exemption is applicable by analogy for the supply of aircrafts, goods for its maintenance or their reparation. we utilized two types of nomenclature to distinguish between taxed and non-taxed supplies. firstly, the classification of products by activity (cpa) nomenclature divides supplies according to nace2 activity, so one might distinguish supplies of medical care products for health care from supplies of medical care for wholesale. secondly, the standard international trade classification (sitc) nomenclature was employed to search for supplies on the lower levels of codes. both types of nomenclature show the data on a monthly basis in eur currency. for the reason of clarity, we summarized outputs on an annual basis. 3.2 administration costs the difference in administration costs caused by the definitive vat system can be quantified by cost of collection indicator. the cost of tax collection is annually calculated and reported by oecd (2020) and represents the value of administration costs incurred by the financial authorities per one unit of tax revenue (not only vat but all taxes). the indicator is calculated as the ratio of annual costs of tax administration to net tax revenues collected during the respective year as follows. 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑐𝑐𝑇𝑇𝑠𝑠𝑇𝑇𝑠𝑠 𝑇𝑇𝑜𝑜 𝑇𝑇𝑇𝑇𝑡𝑡 𝑇𝑇𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑠𝑠𝑇𝑇𝑎𝑎𝑇𝑇𝑇𝑇𝑎𝑎𝑇𝑇𝑎𝑎 𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇𝑇 𝑇𝑇𝑇𝑇𝑡𝑡 𝑎𝑎𝑟𝑟𝑟𝑟𝑟𝑟𝑎𝑎𝑟𝑟𝑟𝑟𝑠𝑠−𝑉𝑉𝑉𝑉𝑇𝑇 𝑇𝑇𝑎𝑎 𝑎𝑎𝑎𝑎𝑖𝑖𝑇𝑇𝑎𝑎𝑇𝑇 [1] equation [1] includes values in national currencies. to improve the comparability of results, vat on imports is subtracted from total tax revenues in the denominator. further, oecd (2020) 2 nomenclature générale des activités économiques dans les communautés européennes zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 152 multiplies the fraction by 100. for the purposes of our analysis, multiplication by 100 is not done so that the result represents collection costs per one currency unit of revenue. collection costs of tax administration can be useful for prediction of additional administration costs incurred in the definitive vat system. each state’s additional costs can be calculated as the collection tax administration costs of each state multiplied by the expected amount of additional vat collected by this member state for another member states. we calculate the difference in the tax administration costs from the amount of intracommunity supplies exceeding intracommunity acquisitions, as explained earlier. the reason is that under the current system, each state collects the vat from the intra-community acquisitions and not from the intracommunity supplies but in the definitive system, it will be exactly the opposite. to quantify the change in administration costs, we multiply the balance of intra-community supplies and acquisitions of each state by the average vat rate in eu (used for simplification as explained in next section). thus, we get the amount of vat to be collected in addition to what was collected in the current system. these amounts are then multiplied by the tax administration collection costs of each state. even though such method of quantification is not perfectly accurate, it helps to demonstrate the effect of the definitive vat system’s implementation. the inaccuracy is caused by several factors. firstly, we build the calculation on data of the current system where no payments between tax administrators take place (except of vat collected from electronic, telecommunication and broadcast services). the process of payments might cause other additional costs that we are not able to predict. secondly, tax administration costs reported by oecd (2020) are calculated per unit of overall tax revenue and not only vat revenue. however, due to the lack of more precise data we use it for our calculation as an approximation of the costs of vat collection. furthermore, vat collection costs are also influenced by other factors specific to vat and different from other types of taxes, as for example the average length of the distribution chain in the economy or the use of a local reverse-charge mechanism. 3.3 vat rates on cross-border transactions to calculate additional costs incurred by definitive system, vat rate is highly important as it serves for computation of additional payments collected and distributed to other member states. in our analysis, we utilized the arithmetic average of standard vat rates applied in all member states in 2017 (european commission, 2020). rates from 2017 were used to correspond to other data in the analysis that are last available in that year (but as table 1 shows, they have not changed till 2019). zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 153 table 1. standard vat rates (source: european commission, 2020) 2017 2018 2019 belgium 21 21 21 bulgaria 20 20 20 czech republic 21 21 21 denmark 25 25 25 estonia 20 20 20 finland 24 24 24 france 20 20 20 croatia 25 25 25 ireland 23 23 23 italy 22 22 22 cyprus 19 19 19 lithuania 21 21 21 latvia 21 21 21 luxemburg 17 17 17 hungary 27 27 27 malta 18 18 18 germany 19 19 19 netherlands 21 21 21 poland 23 23 23 portugal 23 23 23 austria 20 20 20 romania 19 19 19 greece 24 24 24 slovakia 20 20 20 slovenia 22 22 22 united kingdom 20 20 20 spain 21 21 21 sweden 25 25 25 the average vat rate used for our analysis was calculated at 21.519 %. we realize that the use of the average basic rate is a great simplification, but in our opinion, it is sufficient for the preliminary calculation that we want to provide in this paper. for more accurate results in potential further research, we would suggest calculating the average effective vat rate individually for each state according to its trade structure and vat rates (both standard and reduced). 3.4 estimation of mtic fraud for the calculation of additional vat revenues, it is important to estimate the size of mtic fraud because additional vat revenues will be realized through the elimination or at least reduction of this type of vat evasion. the value of mtic fraud might be calculated from the vat gap. zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 154 vat gap means the difference between theoretical vat liability and real vat revenue (european commission, 2016a). one of the vat gap’s components is mtic fraud. nonetheless, there are also other components, as for example, bankruptcy of businesses or mistakes in tax calculation. it is estimated that mtic fraud represents approximately 36 % of the vat gap in eu (european commission, 2016a). the estimates are based only on data collected from a few countries. all member states were asked to provide data on the share of mtic fraud in their vat gap. however, only nine countries were able to provide data in required detail. according to european commission (2015), in some member states the mtic fraud is expected to be lower – approximately 20 %. a lower estimate also stems from the work of other authors (borselli, 2015 or poniatowski et al., 2016). some other authors provided a calculation of mtic fraud on certain goods or between individual countries (stiller and heinemann, 2019 or vaškovič et al., 2021) using the discrepancies in trade balances. furthermore, stiller and heinemann (2021), based on the analysis of reporting discrepancies in trade balances between member states estimated mtic fraud on electronic devices in the whole eu. however, these estimates are not available for all goods and all states. therefore, we used extrapolation and weighted average to estimate the overall share of mtic fraud on vat gap at 24 %. this percentage was further applied to vat gap in each member state estimated for 2017 in case (2019). the use of the average share of mtic fraud in the vat gap is a limitation that could only be overcome with more detailed data from individual member states. however, these data are not publicly available. 3.5 correlation analysis the relationship between the change in administration costs and the value of mtic frauds might be appraised by correlation analysis. from the analysis, we will find out whether the states that will incur relatively more tax administrative collection costs are also states with higher mtic fraud and thus higher additional revenues. this will help us to find out if member states will be motivated to implement a vat definitive system, because changing the vat treatment on cross-border supplies of goods will increase their administrative costs but will also reduce their mtic fraud and these costs and benefits are in balance. to determine the tightness of the dependence of those two variables, we chose spearman’s rank correlation coefficient, which can be expressed as 𝑟𝑟𝑠𝑠 = 1 − 6 ∑ 𝑎𝑎2𝑖𝑖 𝑛𝑛 𝑖𝑖=1 𝑎𝑎 (𝑎𝑎2−1) , 𝑑𝑑𝑎𝑎 = �𝑖𝑖𝑡𝑡 − 𝑖𝑖𝑦𝑦�, [2] where the variables represent 𝑛𝑛…𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑛𝑟𝑟 𝑜𝑜𝑜𝑜 𝑜𝑜𝑛𝑛𝑜𝑜𝑛𝑛𝑟𝑟𝑜𝑜𝑜𝑜𝑜𝑜𝑖𝑖𝑜𝑜𝑛𝑛𝑜𝑜 𝑖𝑖𝑡𝑡 … 𝑜𝑜𝑟𝑟𝑑𝑑𝑛𝑛𝑟𝑟 𝑜𝑜𝑜𝑜 𝑜𝑜ℎ𝑛𝑛 𝑆𝑆𝑜𝑜𝑜𝑜𝑜𝑜𝑛𝑛 𝑜𝑜𝑎𝑎𝑎𝑎𝑜𝑜𝑟𝑟𝑑𝑑𝑖𝑖𝑛𝑛𝑎𝑎 𝑜𝑜𝑜𝑜 𝑜𝑜ℎ𝑛𝑛 𝑎𝑎ℎ𝑜𝑜𝑛𝑛𝑎𝑎𝑛𝑛 𝑖𝑖𝑛𝑛 𝑜𝑜𝑑𝑑𝑛𝑛𝑖𝑖𝑛𝑛𝑖𝑖𝑜𝑜𝑜𝑜𝑟𝑟𝑜𝑜𝑜𝑜𝑖𝑖𝑜𝑜𝑛𝑛 𝑎𝑎𝑜𝑜𝑜𝑜𝑜𝑜𝑜𝑜 𝑖𝑖𝑦𝑦 … 𝑜𝑜𝑟𝑟𝑑𝑑𝑛𝑛𝑟𝑟 𝑜𝑜𝑜𝑜 𝑜𝑜ℎ𝑛𝑛 𝑆𝑆𝑜𝑜𝑜𝑜𝑜𝑜𝑛𝑛 𝑜𝑜𝑎𝑎𝑎𝑎𝑜𝑜𝑟𝑟𝑑𝑑𝑖𝑖𝑛𝑛𝑎𝑎 𝑜𝑜𝑜𝑜 𝑜𝑜ℎ𝑛𝑛 𝑜𝑜𝑜𝑜𝑣𝑣𝑛𝑛𝑛𝑛 𝑜𝑜𝑜𝑜 𝑀𝑀𝑀𝑀𝑀𝑀𝑀𝑀 𝑜𝑜𝑟𝑟𝑜𝑜𝑛𝑛𝑑𝑑𝑜𝑜 the spearman’s correlation coefficient is a non-parametric method that utilizes the order of variables’ values. it is based on the calculation of statistical dependence between the rankings of https://en.wikipedia.org/wiki/correlation_and_dependence https://en.wikipedia.org/wiki/ranking zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 155 two variables. it assesses how well the relationship between two variables can be described using a monotonic function. the great advantage of this method lies in its broad applicability. it might be used for all types of linear or non-linear correlation and furthermore, there is no need for normal data distribution. spearman’s correlation coefficient is mostly used for variables that are expected to be non-linear or without normal distribution (corder, 2014). for the evaluation of correlation’s statistical significance, we used hypotheses about the correlation coefficient. hypothesis 𝐻𝐻0 represents independence of between variables while alternative hypothesis 𝐻𝐻1 means dependence (hindls, 2007). 𝐻𝐻0 = 𝜌𝜌𝑦𝑦𝑦𝑦 = 0; 𝐻𝐻1 = 𝜌𝜌𝑦𝑦𝑦𝑦 ≠ 0 [3] further, we calculated test criterion t, which was afterwards compared to a critical region on selected significance level 𝑊𝑊𝑊𝑊. 𝑜𝑜 = 𝑎𝑎𝑠𝑠 �1− 𝑎𝑎𝑠𝑠2 √𝑛𝑛 − 2; 𝑊𝑊 𝑊𝑊 = {𝑜𝑜; |𝑜𝑜| ≥ 𝑜𝑜1−𝑊𝑊/2} [4] where 𝑜𝑜1 − 𝛼𝛼 2 represents quantile of t-test with 𝑛𝑛 − 2 degrees of freedom. we use a significance level α αof 5 % (i.e., 95% reliability), the critical value is given by the quantile 𝑜𝑜0,975 (25) = 2,060 4. findings and discussion this section provides our results and related discussion about the impact of the definitive vat system on member states. firstly, trade balances of each member state are calculated. secondly, administration costs’ change and additional vat revenues are estimated. thirdly, the results of correlation analysis between the change of administration costs and additional vat revenues are provided. 4.1 trade balances trade balances were calculated as the values of net supplies of each member state to other member states. the trade balances were adjusted by items deducted as non-taxable according to directive. these items were about 2% of the total intra-community supplies and acquisitions. table 2 contains “net supplies” (analogy of net export or trade surplus) calculated for each member state as the sum of intra-community acquisitions of goods from all other member states subtracted from the total of intracommunity supplies of goods to all other member states. in the last two columns, the countries are sorted by their net supplies in 2017 to get the overview of which state was the net biggest exporter (intra-community supplier) and which had the biggest imports (intra-community acquisitions) https://en.wikipedia.org/wiki/variable_(mathematics)#applied_statistics https://en.wikipedia.org/wiki/monotonic_function zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 156 table 2. net supplies after adjustments (eur million). country 2019 2018 2017 2017 countries (sorted by their trade surplus in 2017) austria 16,908 -16,525 -15,179 174,064 netherlands belgium 28,333 26,145 25,641 25,641 belgium bulgaria -1,401 -1,312 -1,834 21,888 ireland croatia -9,353 -8,340 -7,758 20,887 germany cyprus -3,560 -3,669 -3,469 18,510 czech republic czech republic 23,019 20,323 18,510 9,981 poland denmark -5,895 -7,106 -5,390 7,671 hungary estonia -2,317 -2,451 -2,390 3,474 slovenia finland -10,486 -10,373 -10,158 2,677 slovakia france -116,548 -113,908 -113,055 -1,568 malta germany -1,515 10,572 20,887 -1,834 bulgaria greece -10,424 -10,325 -10,401 -1,995 italy hungary 9,549 7,724 7,671 -2,390 estonia ireland 31,284 26,431 21,888 -3,469 cyprus italy 3,203 1,327 -1,995 -3,602 latvia latvia -3,881 -3,582 -3,602 -4,668 lithuania lithuania -4,642 -4,808 -4,668 -5,390 denmark luxembourg -6,040 -6,777 -6,029 -5,546 spain malta -1,619 -2,050 -1,568 -6,029 luxembourg netherlands 181,915 184,964 174,064 -7,758 croatia poland 19,155 12,426 9,981 -10,158 finland portugal -14,004 -14,086 -13,771 -10,401 greece romania -12,284 -10,857 -10,491 -10,491 romania slovakia 311 2,173 2,677 -13,771 portugal slovenia 4,411 4,276 3,474 -15,179 austria spain -3,258 -5,983 -5,546 -18,670 sweden sweden -17,502 -17,975 -18,670 -113,055 france note. source: eurostat (2020) and own calculation. a positive value of net supplies represents a higher number of intra-community supplies than intra-community acquisitions. in such a case, the member state would, after the implementation of the definitive system, collect vat from b2b intra-community supplies for other states, which will be higher than the vat collected from b2b intra-community acquisitions in the current system. therefore, states with the highest trade surplus as netherlands, belgium, ireland, germany and czech republic would incur more administrative costs caused by a definitive vat system compared to the administrative cost that they are spending now. the impact of a definitive system would not be equal for all member states, and countries with positive trade balances would have the worse administrative position. in the left part of table 2, trade balances are calculated for three consecutive years (2017, 2018, 2019) to ascertain the trend. most of the countries remain at either negative or positive values, except for italy and germany. as for germany, b2b intra-community supplies exceed zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 157 acquisitions in 2017 and 2018. in 2019, germany reached higher intra-community acquisitions than supplies. in the case of italy, there were higher intra-community acquisitions than supplies in 2017. for the next two years, intra-community supplies exceed acquisitions. from the last two columns, it can be observed that intra-community supplies predominate in the case of belgium, the czech republic, hungary, ireland, the netherlands, poland as well as slovakia and slovenia. these states are expected to have a worse administrative position in the definitive vat system since they would have a higher administrative burden caused by the additional collection of output vat on intra-community supplies for another member states. the greatest difference between intra-community supplies and acquisitions is in the netherlands. in the long term, the netherlands has higher intra-community supplies than acquisitions by approx. eur 181 billion. thus, it is expected that the netherlands would encounter the highest increase of administration costs in the definitive vat system. on the contrary, france would improve its position as it has on average about eur 116 billion surplus in intra-community acquisitions. 4.2 change in administration costs based on the results above, we analyze which countries would incur higher administration costs in the definitive vat system and which countries will, on the contrary, reduce their tax collection costs. table 3 shows the estimates of the change in administrative costs for all member states in 2017. the second column displays the change in vat collected by member states under the definitive system calculated as net supplies multiplied by the average vat rate. the third column shows tax collection costs per unit of revenue. unfortunately, the data about the cost of collection are not available for hungary, poland, and slovakia. to estimate the change in tax administration costs in those three member states, we calculated average tax collection costs amounting to 0.0078 based on the values for all other 24 states. the fourth column displays the final estimation of the change in tax administration costs caused by the implementation of the definitive vat system. the estimation is based on the change in the amount of administered vat and actual tax collection costs in each state. it is necessary to remark that slightly higher costs should be expected since the definitive system would also require new operations (such as forwarding of payments among tax administrators). the highest benefit of vat system’s transformation would experience france. france has the lowest net supplies in the eu connected to an above-average cost of collection (0.0090). generally, if the state with negative net supplies also has a high value of collection costs, it would benefit even more. the reason is that the tax, which is now collected not efficiently by such a state, would be collected by another member state. zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 158 table 3. change in administration costs. country vat administered for other member states (eur million) cost of collection change in administration costs (eur million) netherlands 37,456 0.0076 286 germany 4,495 0.0137 62 czech republic 3,983 0.0132 53 belgium 5,518 0.0092 51 ireland 4,710 0.0055 26 poland 2,148 0.0078 17 hungary 1,651 0.0078 13 slovenia 748 0.0068 5 slovakia 576 0.0078 4 malta -337 0.0042 -1 estonia -514 0.0033 -2 italy -429 0.0084 -4 bulgaria -395 0.0100 -4 denmark -1,160 0.0052 -6 latvia -775 0.0081 -6 lithuania -1,004 0.0070 -7 cyprus -746 0.0105 -8 spain -1,194 0.0069 -8 luxembourg -1,297 0.0079 -10 croatia -1,669 0.0072 -12 sweden -4,018 0.0035 -14 finland -2,186 0.0065 -14 greece -2,238 0.0064 -14 romania -2,258 0.0080 -18 austria -3,266 0.0084 -27 portugal -2,963 0.0113 -33 france -24,328 0.0090 -219 total 110 note. source: oecd (2020) and own calculation. highlighted countries in the table are expected to experience an increase in administrative costs after the implementation of the definitive vat system. the greatest rise of administration costs would occur in the netherlands, followed by germany, the czech republic and belgium. collection costs play a significant role in determining the change in administration costs. the high value of collection costs causes an increase in the final effect, as can be seen in table 3. this explains why the czech republic should expect more additional administration costs than belgium and ireland although their trade balances show that they will administer more vat for other countries. as the results are calculated for all member states, it is possible to estimate the final change in administration costs across the eu as eur 110 million (sum of the last column totalling the zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 159 changes in all member states). this calculation is, however, not precise as we expect higher additional costs due to new operations and services, as, for example, cross-border payments’ settlement between member states. table 3. vat gap and estimated value of mtic frauds. country vat gap (eur million) mtic frauds (eur million) italy 33,629 8,071 germany 25,016 6,004 france 12,030 2,887 greece 7,399 1,776 romania 6,413 1,539 poland 5,764 1,383 belgium 3,996 959 netherlands 2,744 659 austria 2,444 587 denmark 2,235 536 czech republic 2,082 500 ireland 1,938 465 portugal 1,929 463 hungary 1,893 454 spain 1,806 433 slovakia 1,791 430 finland 1,622 389 lithuania 1,119 269 sweden 654 157 bulgaria 625 150 croatia 459 110 latvia 385 92 slovenia 128 31 estonia 122 29 luxembourg 23 6 cyprus 11 3 malta 13 3 total 28,385 note. sources: case, 2019; european commission, 2015; own calculation) zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 160 4.3 additional vat revenues additional revenue in the definitive vat system would be caused by eliminating mtic frauds. thus, calculation of additional vat revenue requires evaluation of mtic frauds. table 4 displays vat gaps according to case (2019) for 2017 and our estimated value of mtic frauds. as explained earlier, mtic fraud is estimated based on commission’s study (european commission, 2015) as 24 % of each member state’s vat gap. results vary among all member states. based on our calculation, the highest savings are expected in italy (eur 8 billion). the lowest additional revenue would experience cyprus and malta (eur 3 million). total additional vat revenues in the eu would be eur 28 billion in 2017. 4.4 correlation analysis and overall impact on member states the change of administration costs and potential vat revenues that would result from the definitive vat system should be compared to find out whether the definitive system is suitable for individual members and the european union as such. from the point of view of individual states, their net proceeds from the new system can be calculated if we subtract the change in administrative costs from the additional vat revenues (which are equal to the member states’s size of mtic fraud). from the last two columns of table 5, it can be concluded that the new system is beneficial for all countries because the resulting proceeds of the definitive system are positive. this results from the fact that the mtic fraud (i.e., expected additional vat revenue) is much higher than the expected change of tax administration costs. best results are achieved by italy, germany and france, which get the highest proceeds from the new system. this is caused by the size and volume of mtic fraud that is eliminated by the definitive system. on the other side of the spectrum, there are small countries as malta, cyprus or luxemburg that will have the least benefit from the new system. furthermore, we want to assess the fairness of the implementation of the system for the eu as a whole because the costs and benefits of the system are not distributed evenly as explained above. to get an answer, the correlation analysis between the change of the tax administration costs and the estimates of mtic frauds, i.e., a potential increase in vat revenues, was performed. ranking of member states for the correlation analysis are shown in the left part of table 5 below. both variables (change in administrative costs and value of mtic frauds) were arranged from the highest to lowest value, and according to their position in a row, the variables 𝑖𝑖𝑡𝑡 and 𝑖𝑖𝑦𝑦 are the positions of the states based on their values of the change of tax administrative cost and mtic fraud. spearman’s coefficient is further calculated according to equation 2 as: 𝑟𝑟𝑠𝑠 = 1 − 6∗3,056 27 (272−1) = − 0.0672 [5] zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 161 since the value of spearman’s coefficient approaches zero, variables are independent. independence might be verified by test of hypothesis on correlation coefficient. if we substitute variables in equation 4, we get test criterion 𝑜𝑜 = − 0.0672 �1− (−0.0672)2 √27 − 2 = − 0.3365 [6] table 4. correlation analysis and net impact of the definitive system on member states country change in administrative costs (eur million) value of mtic frauds (eur million) 𝒊𝒊𝒙𝒙 𝒊𝒊𝒚𝒚 𝒅𝒅𝒊𝒊 (ix – iy) 𝒅𝒅𝟐𝟐𝒊𝒊 countries ranked according to net proceeds from the def. system net proceeds from the definitive system austria -27 587 25 9 16 256 malta 4 belgium 51 959 4 7 -3 9 cyprus 11 bulgaria -4 150 13 20 -7 49 luxembourg 16 croatia -12 110 20 21 -1 1 slovenia 26 cyprus -8 3 17 27 -10 100 estonia 31 czech republic 53 500 3 11 -8 64 latvia 98 denmark -6 536 14 10 4 16 croatia 122 estonia -2 29 11 24 -13 169 bulgaria 154 finland -14 389 22 17 5 25 sweden 171 france -219 2,887 27 3 24 576 lithuania 276 germany 62 6,004 2 2 0 0 netherlands 373 greece -14 1,776 23 4 19 361 finland 403 hungary 13 454 7 14 -7 49 slovakia 426 ireland 26 465 5 12 -7 49 ireland 439 italy -4 8,071 12 1 11 121 hungary 441 latvia -6 92 15 22 -7 49 spain 441 lithuania -7 269 16 18 -2 4 czech republic 447 luxembourg -10 6 19 25 -6 36 portugal 496 malta -1 3 10 26 -16 256 denmark 542 netherlands 286 659 1 8 -7 49 austria 614 poland 17 1,383 6 6 0 0 belgium 908 portugal -33 463 26 13 13 169 poland 1366 romania -18 1,539 24 5 19 361 romania 1557 slovakia 4 430 9 16 -7 49 greece 1790 slovenia 5 31 8 23 -15 225 france 3106 spain -8 433 18 15 3 9 germany 5942 sweden -14 157 21 19 2 4 italy 8075 note. source: case (2019), ey (2015), and own calculation. zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 162 for significance value of 5%, critical value is represented by quantile: 𝑜𝑜0.975 (25) = 2.060 [7] considering test criterion, |−0.3365 | < 2.060 , we cannot reject hypothesis 𝐻𝐻0 that the two variables are independent. we can therefore conclude that the introduction of the definitive system is good for the eu as a whole, but there is no balance (correlation) between the costs and benefits of the system for individual member states. some states will invest more in the new system than others. 5. conclusion definitive vat system primarily aims to reduce mtic fraud, but it should be noted that this system also causes additional burdens and financial costs, especially for member states, which are supply-oriented in the eu market. the reason is that vat from b2b cross-border supply would be collected in the country of origin and then forwarded to the country of consumption. member state from which the goods are delivered is responsible for the process of payment and forwarding the vat on b2b intra-community supplies. that would cause additional administration costs for the financial authorities of that state (state of origin). in the current vat system, member states collect vat on the b2b acquisition of goods from other member states, while in the new system, they will collect vat on the b2b supply of goods to other member states. therefore, administrative costs will increase in countries that are net exporters. based on our analysis, it was ascertained that a higher additional administration burden might arise in the netherlands, germany, the czech republic, belgium, ireland, poland, hungary, slovenia, and slovakia. these member states would experience an increase in tax administration costs because their intra-community supplies exceed intra-community acquisitions. according to our calculations, the total increase in tax administration costs in all member states together is expected to be approximately eur 110 million (see table 3). this value can be characterized as the lowest value possible generated by the definitive vat system because costs of new processes and operations (inter alia, settlement of payments between member states or control of refunds paid to businesses purchasing goods from other member states) are not involved in our analysis. additional vat revenues that could be generated thanks to the vat definitive system are equal to the value of eliminated mtic frauds. total additional vat revenues across the eu are estimated as eur 40 billion, according to european commission (2017a). this calculation was based on the estimate of the vat gap from another study of the commission (european commission, 2015). since then, the vat gap has decreased across the eu thanks to various measures fighting against vat fraud (for example, reverse charge applied by many member states on risk commodities). therefore, potential additional vat revenue would be nowadays zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 163 lower than in 2015, they would reach eur 28 billion in 2017 (see table 4) and would still be in billions of eur in the year 2021. additional revenues would certainly exceed the additional administrative costs that we calculated. it is clear from the net proceeds of the individual member states (in table 5) that not all countries would benefit to the same extent from the definitive system. italy, france and germany would gain the most. malta, cyprus and luxemburg would have the least advantage from the system in absolute terms due to their small volume of mtic fraud. if only these two factors are considered, the definitive vat system would be advantageous for the european union. nonetheless, benefits for member states are not balanced since there is no correlation between benefits (the additional vat revenues) and costs (additional administration costs) incurred by them. to implement the definitive vat system, a unanimous agreement of all member states is necessary. therefore, we would recommend some sort of compensation of additional administrative costs for the member states with a worse administrative position. we also find it necessary to emphasize that there are also other aspects of the definitive vat system that should be evaluated to be able to decide about its implementation. for instance, the change in cash-flow of businesses and their higher compliance costs on cross-border transactions could have a negative impact on cross-border trade in single market or increased compliance costs for business. acknowledgments the paper was prepared as one of the outputs of a research project of the faculty of finance and accounting at the prague university of economics and business “economic and institutional aspects of public finance”, registered by the internal grant agency of prague university of economics and business under the number f1/7/2019, and it was also funded by the institutional support ip 100040 at the faculty of finance and accounting at the prague university of economics and business. references ainsworth, r. (2009). the morphing of mtic fraud: vat fraud infects tradable co2 permits. boston university school of law working paper. 09-35. https://doi.org/10.2139/ssrn.1443279 amand c. (2014). taxation of intracommunity supplies of goods. international vat monitor 25(4). 188-196. retrieved from: https://www.ibfd.org/ibfd-products/journal-articles/internationalvat-monitor/collections/ivm/html/ivm_2014_04_e2_3.html. case. (2019). study and reports on the vat gap in the eu-28 member states: final report. retrieved from: https://ec.europa.eu/taxation_customs/sites/taxation/files/vat-gap-full-report2019_en.pdf https://doi.org/10.2139/ssrn.1443279 https://www.ibfd.org/ibfd-products/journal-articles/international-vat-monitor/collections/ivm/html/ivm_2014_04_e2_3.html https://www.ibfd.org/ibfd-products/journal-articles/international-vat-monitor/collections/ivm/html/ivm_2014_04_e2_3.html https://ec.europa.eu/taxation_customs/sites/taxation/files/vat-gap-full-report-2019_en.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/vat-gap-full-report-2019_en.pdf zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 164 catarino, j. r., & moraes e soares, r. (2019). restructuring the european vat tax system: advantages and disadvantages of the adoption of a single-rate model – a study based on the portuguese case. european journal of government and economics 8(2). 145-160. doi: https://doi.org/10.17979/ejge.2019.8.2.5478 cnossen, s. (2010). three vat studies. hague: central pauberau. retrieved from: https://cpb.nl/sites/default/files/publicaties/download/bijz90.pdf corder, w. g. (2014). nonparametric statistics: step-by-step approach (2nd ed.). hoboken, nj: john wileys & sons. de la feria, r. (2018). the definitive vat system: breaking with transition. ec tax review 27(3). 122-126. retrieved from: https://papers.ssrn.com/sol3/papers.cfm?abstract_id= 3485788 european commission (2010). green paper: on the future of vat. retrieved from: https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/common/con sultations/tax/future_vat/sec%282010%291455_en.pdf european commission (2015). implementing the ‘destination principle’ to intra-eu b2b supplies of goods, feasibility and economic evaluation study, final report. retrieved from: https://op.europa.eu/en/publication-detail/-/publication/f62d669b-503f-41b7-b3b3a77e87df81be/language-en european commission (2016). action plan on vat: communication from the commission to the european parliament, the council and the european economic and social committee. brussels. retrieved from: https://ec.europa.eu/taxation_customs/sites/taxation/files/com_ 2016_148_en.pdf european commission (2016a). report on vat gap estimations by fiscalis tax gap project group (fpg/041): the concept of tax gaps. brussels. retrieved from: https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/tgpg_report_en.pdf european commission (2017). council directive amending directive 2006/112/ec as regards harmonising and simplifying certain rules in the value added tax system and introducing the definitive system for the taxation of trade between member states. retrieved from: https://eurlex.europa.eu/legal-content/en/txt/html/?uri=celex:52017pc0569&from=cs european commission (2017a). impact assessment accompanying the document proposal for a council directive amending directive 2006/112/ec as regards harmonising and simplifying certain rules in the value added tax system and introducing the definitive system for the taxation of trade between member states. retrieved from: https://ec.europa.eu/ transparency/regdoc/rep/10102/2017/en/swd-2017-325-f1-en-main-part-1.pdf european commission (2020). vat rate in the eu as of 1 january 2020. retrieved from: https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/taxation/vat/ how_vat_works/rates/vat_rates_en.pdf european council (2006). council directive 2006/112/ec of 28 november 2006 on the common system of value added tax. retrieved from: https://eur-lex.europa.eu/legalcontent/en/txt/html/?uri=celex:32006l0112&from=cs https://doi.org/10.17979/ejge.2019.8.2.5478 https://cpb.nl/sites/default/files/publicaties/download/bijz90.pdf https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3485788 https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3485788 https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/common/consultations/tax/future_vat/sec%282010%291455_en.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/common/consultations/tax/future_vat/sec%282010%291455_en.pdf https://op.europa.eu/en/publication-detail/-/publication/f62d669b-503f-41b7-b3b3-a77e87df81be/language-en https://op.europa.eu/en/publication-detail/-/publication/f62d669b-503f-41b7-b3b3-a77e87df81be/language-en https://ec.europa.eu/taxation_customs/sites/taxation/files/com_2016_148_en.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/com_2016_148_en.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/docs/body/tgpg_report_en.pdf https://eur-lex.europa.eu/legal-content/en/txt/html/?uri=celex:52017pc0569&from=cs https://eur-lex.europa.eu/legal-content/en/txt/html/?uri=celex:52017pc0569&from=cs https://ec.europa.eu/transparency/regdoc/rep/10102/2017/en/swd-2017-325-f1-en-main-part-1.pdf https://ec.europa.eu/transparency/regdoc/rep/10102/2017/en/swd-2017-325-f1-en-main-part-1.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/taxation/vat/how_vat_works/rates/vat_rates_en.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/resources/documents/taxation/vat/how_vat_works/rates/vat_rates_en.pdf https://eur-lex.europa.eu/legal-content/en/txt/html/?uri=celex:32006l0112&from=cs https://eur-lex.europa.eu/legal-content/en/txt/html/?uri=celex:32006l0112&from=cs zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 165 eurostat (2020). international trade in goods aggregated data. retrieved from: https://ec.europa.eu/eurostat/cache/metadata/en/ext_go_agg_esms.htm fedeli, s. & forte, f. (2011). eu vat frauds. european journal of law and economics 31(2). 143-166. retrieved from: https://doi.org/10.1007/s10657-009-9135-5. hindls, r. (2007). statistika pro ekonomy (8th ed.). praha: professional publishing. keen, m. & smith, s. (2006). vat fraud and evasion: what do we know, and what can be done? national tax journal 59(4). 861-887. retrieved from: https://www.ntanet.org/ ntj/59/4/ntj-v59n04p861-87-vat-fraud-evasion-what.html lamensch, m. 2012. are ‘reverse charging’ and the ‘one-stop-scheme’ efficient ways to collect vat on digital supplies? world journal of vat/gst law 1(1). 1-20. retrieved from: https://core.ac.uk/download/pdf/80792379.pdf oecd (2019). international trade by commodity statistics. retrieved from https://www.oecdilibrary.org/trade/international-trade-by-commodity-statistics/volume-2019/issue-6_g2g9fb2aen oecd (2020). tax administration 2019: comparative information on oecd and other advanced and emerging economies. retrieved from: https://www.oecd.org/tax/administration/taxadministration-23077727.htm pomeranz, d. (2015). no taxation without information: deterrence and self-enforcement in the value added tax. american economic review 105(8). 2539-69. https://doi.org/ 10.1257/aer.20130393 poniatowski, grzegorz and neneman, jaroslaw and michalik, tomasz, vat non-compliance in poland under scrutiny (june 28, 2016). mbank – case seminar proceedings, no. 142/2016, retrieved from: https://ssrn.com/abstract=2813160 or http://dx.doi.org/10.2139/ssrn.2813160 poniatowski, g. et al. (2020). study and reports on the vat gap in the eu-28 member states: 2020 final report. warsaw: case. retrieved from: https://ec.europa.eu/taxation_customs /sites/taxation/files/vat-gap-full-report-2020_en.pdf schenk, a., thuronyi, v. and cui, w. 2015. value added tax: a comparative approach (2nd ed.). new york: cambridge university press. isbn: 978-1-107-04298-8. stiller, w., heinemann m. 2019. reverse-charge-mechanism an effective measure against vat fraud in germany and austria? betriebswirtschaftliche forschung und praxis 71(2):177-192 stiller, w., heinemann m. 2019. do more harm than good? the optional reverse charge mechanism against vat fraud. preprint on research gate. https://doi.org/10.13140/ rg.2.2.13802.44482 tait, a. a. (1988): value added tax: international practice and problems. washington, dc: international monetary fund. isbn:9781557750129. doi: http://dx.doi.org/ 10.5089/9781557750129.071 terra, b. and kajus, j. (2015). a guide to the european vat directives. hoorn: ibfd. isbn 97890-8722-308-3. https://ec.europa.eu/eurostat/cache/metadata/en/ext_go_agg_esms.htm https://doi.org/10.1007/s10657-009-9135-5 https://www.ntanet.org/ntj/59/4/ntj-v59n04p861-87-vat-fraud-evasion-what.html https://www.ntanet.org/ntj/59/4/ntj-v59n04p861-87-vat-fraud-evasion-what.html https://core.ac.uk/download/pdf/80792379.pdf https://www.oecd-ilibrary.org/trade/international-trade-by-commodity-statistics/volume-2019/issue-6_g2g9fb2a-en https://www.oecd-ilibrary.org/trade/international-trade-by-commodity-statistics/volume-2019/issue-6_g2g9fb2a-en https://www.oecd-ilibrary.org/trade/international-trade-by-commodity-statistics/volume-2019/issue-6_g2g9fb2a-en https://www.oecd.org/tax/administration/tax-administration-23077727.htm https://www.oecd.org/tax/administration/tax-administration-23077727.htm https://doi.org/10.1257/aer.20130393 https://doi.org/10.1257/aer.20130393 https://ssrn.com/abstract=2813160 https://dx.doi.org/10.2139/ssrn.2813160 https://ec.europa.eu/taxation_customs/sites/taxation/files/vat-gap-full-report-2020_en.pdf https://ec.europa.eu/taxation_customs/sites/taxation/files/vat-gap-full-report-2020_en.pdf https://doi.org/10.13140/rg.2.2.13802.44482 https://doi.org/10.13140/rg.2.2.13802.44482 http://dx.doi.org/10.5089/9781557750129.071 http://dx.doi.org/10.5089/9781557750129.071 zídková and balíková / european journal of government and economics 10(2), december 2021, 146-166 166 vat expert group (2016). definitive vat regime for intra-eu trade. first step. issues to be examined. retrieved from: https://circabc.europa.eu/sd/a/1f174f75-8a88-4f58-94b0c2d748371ab6/57%20-%20definitive%20regime%20for%20intra-eu%20trade%20%20first%20step%20-%20issues%20to%20be%20examined.pdf vaškovič, m. zídková, h. arltová, m. the volume of mtic fraud between poland and czechia in electronic devices trade: general method of carousel fraud estimation. international journal of economic policy in emerging economies. 2021. issn 1752-0452. url: https://www.inderscience.com/info/ingeneral/forthcoming.php?jcode=ijepee#96641 zídková, h., vrána, t. (2020). is vat administration system efficient? the case of the czech republic. central european public administration review 18(2), 121–134. doi: https://doi.org/10.17573/cepar.2020.2.06 zídková, h., šťastná, a. (2019). vat collection methods. acta universitatis agriculturae et silviculturae mendelianae brunensis 67(3). 883-895. https://doi.org/10.11118/actaun 201967030883 https://circabc.europa.eu/sd/a/1f174f75-8a88-4f58-94b0-c2d748371ab6/57%20-%20definitive%20regime%20for%20intra-eu%20trade%20-%20first%20step%20-%20issues%20to%20be%20examined.pdf https://circabc.europa.eu/sd/a/1f174f75-8a88-4f58-94b0-c2d748371ab6/57%20-%20definitive%20regime%20for%20intra-eu%20trade%20-%20first%20step%20-%20issues%20to%20be%20examined.pdf https://circabc.europa.eu/sd/a/1f174f75-8a88-4f58-94b0-c2d748371ab6/57%20-%20definitive%20regime%20for%20intra-eu%20trade%20-%20first%20step%20-%20issues%20to%20be%20examined.pdf https://www.inderscience.com/info/ingeneral/forthcoming.php?jcode=ijepee#96641 https://doi.org/10.17573/cepar.2020.2.06 https://doi.org/10.11118/actaun201967030883 https://doi.org/10.11118/actaun201967030883 the definitive vat system and its impact on tax collection 1. introduction 2. literature review 3. data and methods 4. findings and discussion 5. conclusion acknowledgments references © theauthor(s)2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 1 (2024), pages 48-66 https://doi.org/10.17979/ejge.2024.13.1.9918 submitted: september 10, 2023 accepted: december 6, 2023 published: june 6, 2024 article impact of government quality on post-covid subjective well-being in pakistan: the mediating role of financial stress muhammad hassan danish1,2, *, muhammad ashfaq3, salman azam joiya4 1 university of management and technology, lahore, pakistan 2 research fellow at department of economics, university of religions and denominations, qom, 3749113357, iran 3 iu international university of applied sciences, germany 4 lahore business school, the university of lahore, pakistan *correspondence: hassan.danish@umt.edu.pk abstract. covid-19 has badly affected the psychological well-being of people all over the globe, especially in developing countries, due to the loss of jobs, social capital and stress. the present study is also designed to analyze the impact of government quality on subjective well-being (swb) after the pandemic through the mediating role of financial stress. this study uses the data of 599 respondents in pakistan and applies mixed methodology in generalized structure equation model (gsem) structure with tobit and order logit. results reveal that people perceived more financial and economic stress and perceived poor satisfaction with the government’s role in fighting the pandemic. regression analyses confirmed that least economic stress, good perceived governance, and high level of income are negatively associated with financial stress and positively relate to happiness and life satisfaction. moreover, a significant relationship exists between the mediator (financial stress) and swb. our results suggest government bodies should encourage investment for business start-ups. this will not only reduce unemployment and economic and financial stress but also expand the size of the economy, which leads to achieving the wellbeing of people. keywords: covid-19; social capital; financial stress; post-covid stress; happiness; life satisfaction jel classification: h1; i1 1. introduction the covid-19 pandemic broke out in pakistan in february 2020. pakistan has recorded more than 1.5 million confirmed cases and over 30,000 deaths due to covid-191. the country has faced multiple waves of the pandemic, with varying severity. pakistan, like many other countries, has taken various measures to combat the spread of covid-19. these measures included the closure of schools and universities, the suspension of international and domestic travel, the imposition of lockdowns in certain areas, the requirement for face masks and social distancing and the ramping up of vaccination efforts. however, compliance with these measures has varied and the country has experienced 1 see, for instance: https://covid19.who.int/region/emro/country/pk https://creativecommons.org/licenses/by-nc/4.0/ https://covid19.who.int/region/emro/country/pk impact of government quality on post-covid subjective well-being multiple waves of the pandemic with varying severity. due to restrictions and safety measures to prevent severe health effects, people in pakistan have faced negative impacts on their psychological and mental health. during the lockdowns, people were restricted as regards going outside for recreation activities, they lost their social capital, many people lost their jobs and losses in businesses were also reported. these factors all have a negative impact on psychological well-being and increase financial and mental stress. the impact of covid-19 on psychological well-being is still notable in pakistan. before the pandemic, pakistan was in 65th position in happiness among 145 countries around the globe (world happiness report, 2018-19), but a recent report (whr-2022) shows that pakistan has significantly dropped down in the ranking of happiness to 122nd among 146 countries analysed. this is due to the bad economic situation, poor social capital as a result of social distancing during the pandemic, poor government quality, increased unemployment and financial stress. the prevailing situation in the country and the sudden decrease in the happiness level motivates us to identify the factors that cause financial stress and the lower level of well-being among the people of pakistan. there are various factors associated with the low level of wellbeing after the covid-19. the most common outcomes that impact the well-being of people include economic uncertainty, joblessness, physical activities, disruption in daily life routine, healthcare problems, and social distancing. in less developed countries, covid-19 has led to a health crisis, inducing an economic crisis and, consequently, an increase in the problems of people falling into low-income strata. residents from less developed countries are forced to live in vulnerable conditions due to increased economic stress, over-dependence on the informal sector, unavailability of mental healthcare services and the prevalence of mental health issues (baranov et al., 2020; maselko et al., 2018; muhammad gadit & mugford, 2007). economic stress has increased around the world, which also negatively influenced physical and mental health problems. it should be the foremost priority of policymakers to address such problems (bareket-bojmel et al., 2021). in addition to economic and financial stress, daily life routines and physical activities have badly affected people's mental health and well-being. there is no denying the fact that physical activity is directly related to well-being and mental health. research shows that exposure to nature positively increases happiness, reduces stress and enhances vitality (nguyen & brymer, 2018). it is necessary to be engaged in physical activity through multiple ‘lenses’. these lenses may be evaluated in the form of social and physical environments (sallis et al., 2015). the psychological factors related to an individual include factors related to the physical, social and emotional environment. in addition to these, recreational facilities are also important for the psychological framework of an individual (bauman et al., 2012; sallis et al., 2015). physical activity is strongly recommended by public health advocates to protect people from the negative impacts of covid-19 in the form of psychological and physical detriments (hammami et al., 2022). there is a clear side effect of covid-19 on the lives of people in the form of anger, confusion and stress (brooks et al., 2020). sustainable development agenda focuses on improving the well-being of people but the rank of pakistan has significantly decreased in happiness. this situation might be control by the government, but poor governance after the pandemic has a bad influence on the economic condition and psychological wellbeing of people. post covid conditions in pakistan were not controlled and still there are multiple factors, including a poor healthcare system and increasing mental health problems 50 danish, ashfaq and azam among people. additionally, economic crisis and stagflation have also become major issues in increasing financial stress and psychological disorders among people. examining these impacts motivates the present research to have a more thorough grasp of the pandemic's consequences and take preventative action to deal with the difficulties that people and society could encounter in the wake of it. therefore, the present study is designed to analyze the role of government quality on swb with the mediating effect of financial stress after the covid period. the present study is carried out by collecting data from 599 households in pakistan through the survey method. the present study is contributing to the literature in many ways. 1) this research adds to the literature by analyzing the mediating role of financial stress in the relationship between government quality and swb. 2) the present research uses a generalized structure equation model (gsem) due to ordinal outcome of dependent variables. 3) previous studies have been conducted during the pandemic but this study captures the post-covid impact on subjective wellbeing. the present study will assist policymakers in working on factors that can minimize financial stress and improve the well-being of people. the rest of the study is organized as follows: the second section highlights the recent literature on covid-19 related stress and wellbeing. section 3 highlights the methodology and data collection procedure. results are presented and interpreted in section 4, while the study is concluded in section 5. 2. literature review the literature on the effect of covid-19 on daily life and the psychological well-being of people is discussed vastly around the globe. previous studies have examined the impact of covid-related issues during the pandemic or lockdown period on well-being and financial stress (see, for instance, alfawaz et al., aslam et al., petrovic et al., rodrigues et al., 2021; baranov et al.). however, covidrelated issues are still not resolved, especially in developing countries, and no study has been found that has examined the post-covid issues on well-being, especially in the context of pakistan. as discussed earlier, covid-related issues have created a gap in people and job losses, economic crisis and high levels of inflation have increased the financial stress among people, which is also the cause of low levels of happiness and life satisfaction. therefore, there was a need to address these issues and highlight those factors that are responsible for financial and psychological stress. covid-19, financial position and swb covid-19 has significantly affected people's happiness and financial position. several indicators, including the gross national happiness index, are used to measure and compare happiness before and after the imposition of the lockdown (petrovič et al., 2021). people going through serious conditions reported their mental health to be much worse compared to before the first covid-19 lockdown (brown et al., 2021). economic well-being is not only affected by the loss of income but also because of degradation in the quality of life due to social distancing, leading to a decline in economic well-being as well. though there has been a major income loss during the pandemic time, the well-being of people with an undergraduate degree who were working in the non-health sectors impact of government quality on post-covid subjective well-being reported higher income loss as compared to others (tran et al., 2020). whereas a decline in the income levels of people during covid-19 has led to a decline in the life satisfaction of people as compared to those who had no income loss (baranov et al., 2022; cheng et al., 2020). psychological and mental impact of covid-19 an online survey administered to administrative staff, employees, and students of king saud university was conducted to measure the psychosocial and mental impact of covid-19, and it was found that upper-income quantile groups feel less stress than lower-income groups. as far as coping mechanisms for loneliness are concerned, strong familial bonding has proved to be crucial in maintaining mental health. this was particularly true for the female respondents (alfawaz et al., 2021). social distancing by parents has also significantly increased stress, depression and anxiety in norway during covid-19 (johnson et al., 2019). xi et al. (2020) have analyzed the longitudinal changes in individual responses throughout the pandemic. higher levels of psychological distress are found in well-educated respondents due to their awareness of the health consequences that covid19 could induce, but interestingly, they reported lower levels of distress in a follow-up survey (huang & zhao, 2020). traumatic stress, anxiety and well being there exists a positive correlation between traumatic stress and growth. a cross-sectional study by zhao et al. (2021) used online surveys to identify this relationship. they have found that females, low-educated people and middle-age groups are more vulnerable to stress, whereas exploring the impact of covid-19 on well-being, fear and anxiety is higher in adults (aslam et al., 2021). there is a significant but negative impact of anxiety and well-being and there is a direct relationship between economic crisis and global health crisis with life satisfaction. social distancing by parents has also significantly increased stress, depression and anxiety in norway during covid-19 (johnson et al., 2019). post pandemic stress disorder and stress women are found to have developed post-traumatic stress disorder, while middle-aged people worried about their children and ageing parents. government employees who were battling on the frontline against the pandemic faced distress (xi et al., 2020). the psychological impact of post covid-19 on hospital staff is notable, as the frequency of psychological distress tends to be higher among health service staff, according to an online cross-sectional survey administered to nurses, midwives, doctors and health staff in a large metropolitan health service located in melbourne, australia (holton et al., 2020). healthcare professionals are at a higher risk of psychological problems as one out of five healthcare professionals experienced mental health issues and the need for psychological support interventions is inevitable (ceri & cicek, 2021). 52 danish, ashfaq and azam lockdown and physical activity restriction there is an impact of lockdown restrictions on outdoor activities, self-isolation on physical activity and eating habits as these new conditions increased anxiety and depression, lowered average quality of life and provoked unhealthy behaviors as coping mechanisms, especially in young people as they are more vulnerable to sudden changes (dragun et al., 2020; lesser & nienhuis, 2020). students have faced more drastic changes and stress due to delayed graduation, job losses and a reduction in job offers, according to a cross-sectional study that targeted students from schools and medical colleges (dragun et al., 2020). another research (lu et al., 2020) conducted an online survey to collect data related to depression, fear and anxiety by using the hamilton depression scale (hamd) and found that psychological stress increased in the medical worforce during covid-19. saalwirth and leipold (2021) conducted a survey of 665 respondents in germany and found that social coping and covidrelated worries negatively affect well-being and sleep patterns in families. china has been more affected by the pandemic, where sleeping patterns and day-time physical activities have been negatively affected by covid-19 (wang et al., 2021). loneliness, mental health issues and disturbances li and wang (2020) related the frequency of feelings of loneliness during covid-19. they found that young people and women are more likely to develop a general psychiatric disorder (gpd) and loneliness. on the other hand, having a job and living with a partner are factors that are associated with reduced loneliness and general psychiatric disorders. whereas godinić and obrenovic (2020) concluded that there exists a positive correlation between job uncertainty and identity disturbance and a negative correlation between job uncertainty and psychological well-being, job uncertainty resulting from an economic recession can lead to helplessness, distress and paranoia, substance abuse and suicide. according to de kock et al. (2021), mental health issues have a positive correlation between job uncertainty and identity disturbance and a negative correlation between job uncertainty and psychological well-being. previous studies have either studied the impact of financial stress and households’ economic situation on wellbeing and psychological stress (alola et al., 2021; rodrigues et al., 2021 and baranov et al. 2022) or the impact of government quality on swb during the pandemic (alamsyah & zhu, 2022) or estimated the anxiety and mental stress during covid 19 (holton et al., 2020; lu et al., 2020; cohen-louck & levy, 2022). these studies have carried out research on the teachers, nurse staff, medical students or staff during the covid 19, but no study has been conducted on the postcovid effect. moreover, the methodology used in these studies is very simple. they use either simple ols or sem in mediation. but if the dependent variable is in an ordinal category, ols results become spurious. therefore, the present research is carried out to address all those issues which are prevailing after the covid 19 and uses gsem with order logit and tobit model. the results of this study will help policymakers to control the variables that have a negative impact on swb and provide policies to achieve sustainable development goals. impact of government quality on post-covid subjective well-being 3. material and methods 3.1 data the data of this study was collected from residents of pakistan who are above the age of 18 through a google survey link. data was collected between the period of september 2022 and 15 january 2023, and the google form was shared on social media like facebook, whatsapp, and linkedin and emailed to potential participants. due to financial constraints, authors cannot visit city to city, and people feel restrictions to meet unknown persons. therefore, it was difficult to collect data from personal meetings and interviews and the authors decided to use the social media network and personal contacts for survey. the final data we received for this study was comprised of 640 participants in the survey from all over pakistan, but after cleaning, only 599 responses are included for data analysis of this study due to missing information in the remaining data. our sample consists of 34.5% females and 65.5% male respondents to analyze the impact of covid-19 on the financial stress and well-being of individuals in pakistan. 3.2 measures the survey for this study is comprised of 8 parts: demographic, income and financial stress after covid-19, dietary habits, government support during/after covid-19, stress, social capital, economic stress and measures of well-being. demographic characteristics include gender, age, education, and employment status, but age is not included in the final model as these variables affect the robustness of the model and are not significant. income is measured on an ordinal scale while financial stress is measured on a five-point likert scale. social capital includes relationships with family, friends, neighbours and other people after covid-19. economic stress includes stress on child education, prices and the economic situation after covid-19. government support is used as a proxy of government effectiveness, which includes questions regarding social security, creating jobs, reducing inequality, educational needs and health care after covid-19. finally, well-being is measured on an ordinal scale from 0-10 in the context of happiness and life satisfaction. subjective well-being the present study uses happiness and life satisfaction as a measure of swb, which is the worldwide recommended measure (helliwell et al., 2021). both measures are ranked on an ordinal scale from 0-10, where ‘0’ represents “not at all happy or satisfied with life”, while ‘10’ represents “completely happy or satisfied with life”. financial stress financial stress is measured through pca (principal component analysis) on spss, which includes four items on a five-point likert scale from “never” (1) to “always” (5). these items include depression, irritation and a drain of emotions due to financial situation. the measure has strong internal consistency (α= 0.84), and the value of bartlett’s test is significant at a 1% level, which shows 54 danish, ashfaq and azam that this is a valid measure for the analysis. this measure is suggested by cardona-montoya et al. (2022). financial stress increases negative feelings and emotions, which badly affect mental and physical health. in pakistan, people feel irritated to meet people if they are unemployed, lose their emotions and many people attempt suicide due to continuous depression. therefore, this variable is added to the factor of analysis. covid-19-related stress individual stress is measured by six items, including anxiety, sleep disturbance and sleeping patterns, hopelessness and stress after covid-19. each item is asked on a dichotomous scale (1= ‘yes’ and 0 = ‘no’). to measure the final values, the sum of responses is used to estimate the stress variable for each individual. economic stress after the covid-19, the inflation rate has consistently increased in double digits in pakistan. due to this, educational institutions have also raised the cost of education, and the overall economic situation is becoming worse day by day. therefore, this variable can be a significant factor in analysing swb. households’ and economic-related stress is measured through three items, which include stress on child education, prices and overall economic situation in the country after covid19 on 5 5-point likert scale from “very much” (1) to “not at all” (5). this measure was also adopted by (cohen-louck & levy, 2022) and internal consistency was satisfactory (α = 0.653). the variable is constructed through principle component analysis (pca) on spss and the value of bartlett’s test is significant at a 1% level, which shows that this is a valid measure for the analysis. pca is used to standardize the values of variables for dimension reduction in the dataset containing multiple items of a variable with multiple responses or likert-scale. social capital respondents were asked about their relationships with their family, friends, neighbours and other people after covid-19 on a five-point likert scale from 1 (extremely good) to 5 (extremely bad) and a measure of social capital is developed through pca on spss. the measure of social capital also has strong internal consistency and validity (α = 0.82). zhao et al. (2022) have also used the measure of social capital to estimate its impact on happiness and mental health. government quality government support is used as a proxy of government quality after covid-19 as a political and social responsibility. respondents were asked about the performance of the government in the last 2 years regarding social security allowance, creating jobs, reducing inequality, providing educational needs and health care on a five-point likert scale from 1 (very poor) to 5 (very good). this measure is adapted from the world happiness report (2022) and is used according to the country’s situation. government effectiveness is also measured through pca on spss and possesses a strong internal consistency (α = 0.86). this measure has also been used in previous studies to assess government effectiveness or government quality (see, for example, kim & kim, 2012; danish & nawaz, 2022). impact of government quality on post-covid subjective well-being demographic and income demographic variables include gender (female = 0 and male = 1), education (up to primary = 1, up to matric = 2, up to graduation = 3 and master’s and above = 4), employment status (unemployed/retired/housewife = 1, part-time employed = 2, full time employed = 3 and selfemployed = 4). moreover, income is measured on 5 ranked scales (up to 25k = 1, 25-50k = 2, 50-75k = 3, 75-100k = 4 and more than 100k = 5). 3.3 empirical model and method we described data characteristics that covered frequency, percent, mean and standard deviation. the data obtained from the survey is analyzed using spss 22 and stata 15 software. data is described through frequency distribution, mean values and percentage distribution. the main objective of this study is to analyze the government support on swb (happiness and life satisfaction) through the mediating channel of financial stress after covid-19. in the first model, financial stress is regressed over employment, income, economic stress and government effectiveness, while in the second model, financial stress is used as a mediator to regress happiness and life satisfaction along with other factors, including gender, social capital, dietary habits, stress, education and living area (equation 1 & 2). 𝐹𝐹𝐹𝐹𝑖𝑖∗ = 𝛽𝛽0 + 𝛽𝛽1𝑖𝑖. 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖∗ + 𝛽𝛽2𝑖𝑖. 𝑖𝑖𝑒𝑒𝑖𝑖𝑒𝑒𝑒𝑒𝑒𝑒.𝑖𝑖∗+ 𝛽𝛽3𝑒𝑒𝑖𝑖𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖𝑖𝑖 𝑠𝑠𝑒𝑒𝑠𝑠𝑒𝑒𝑠𝑠𝑠𝑠𝑖𝑖∗ + 𝛽𝛽4𝑔𝑔𝑒𝑒𝑔𝑔𝑒𝑒𝑠𝑠𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑠𝑠𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖∗ + 𝜇𝜇𝑖𝑖 … … … (1) 𝐹𝐹𝑆𝑆𝑆𝑆𝑖𝑖∗ = 𝛾𝛾0 + 𝛾𝛾1𝑖𝑖.𝑔𝑔𝑒𝑒𝑒𝑒𝑔𝑔𝑒𝑒𝑠𝑠𝑖𝑖∗ + 𝛾𝛾2𝑖𝑖. 𝑎𝑎𝑠𝑠𝑒𝑒𝑎𝑎𝑖𝑖∗ + 𝛾𝛾3𝑖𝑖. 𝑒𝑒𝑔𝑔𝑒𝑒𝑖𝑖𝑎𝑎𝑒𝑒𝑖𝑖𝑒𝑒𝑒𝑒𝑖𝑖∗ + 𝛾𝛾4𝑠𝑠𝑒𝑒𝑖𝑖𝑖𝑖𝑎𝑎𝑒𝑒 𝑖𝑖𝑎𝑎𝑒𝑒𝑖𝑖𝑒𝑒𝑎𝑎𝑒𝑒𝑖𝑖∗ + 𝛾𝛾5 𝑔𝑔𝑖𝑖𝑒𝑒𝑒𝑒𝑖𝑖∗ + 𝛾𝛾6 𝑠𝑠𝑒𝑒𝑠𝑠𝑒𝑒𝑠𝑠𝑠𝑠𝑖𝑖∗ + 𝛾𝛾7 𝑔𝑔𝑖𝑖𝑒𝑒𝑒𝑒𝑖𝑖∗ + 𝛾𝛾8 𝑖𝑖. 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖∗ + 𝛾𝛾9𝑖𝑖. 𝑖𝑖𝑒𝑒𝑖𝑖𝑒𝑒𝑒𝑒𝑒𝑒.𝑖𝑖∗ + 𝛾𝛾10𝑒𝑒𝑖𝑖𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖𝑖𝑖 𝑠𝑠𝑒𝑒𝑠𝑠𝑒𝑒𝑠𝑠𝑠𝑠𝑖𝑖∗ + 𝛾𝛾11𝑔𝑔𝑒𝑒𝑔𝑔𝑒𝑒𝑠𝑠𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒 𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑒𝑖𝑖𝑒𝑒𝑖𝑖𝑔𝑔𝑒𝑒𝑒𝑒𝑒𝑒𝑠𝑠𝑠𝑠𝑖𝑖∗ + 𝜈𝜈𝑖𝑖 … … … (2) ‘fs’ denotes ‘financial stress’, which is computed through pca, and ‘swb’ is the measure of the individual's happiness and life satisfaction. to analyze the model in equation 1 and 2, a gsem (generalized structure equation model) is used with the mixed methodologies of tobit and order logistic regression according to the nature of dependent variables. tobit model is estimated continuous dependent variable when it is censored (tobin, 1958). error terms are normally distributed in the tobit model. in the second model, happiness and life satisfaction (ls) are measured on an ordinal scale from 0-10; therefore, order logistic regression is an appropriate methodology for such measures. the order logit model is specifically designed to estimate such a kind of dataset when the dependent variable is in order form or ranked and does not consist of any continuous scale (cameron & trivedi, 2005). a step-wise regression is applied for both the happiness and ls model with and without control variables. 56 danish, ashfaq and azam 4. results 4.1 descriptive analysis table 1 presents the demographic characteristics of the sample respondents in the study. 65.44% of respondents in our study are male, while the remaining 34.56% are female. most of the respondents are from urban areas (73.56%) and have completed more than 12 years of education (85.65%). 45.58% of respondents are full-time employed, while 19.20% are self-employed. more than half (53%) of respondents belong to upper-middle and upper-income groups. table 1. descriptive statistics variables freq. % of the sample gender male 392 65.44 female 207 34.56 area rural 159 26.54 urban 440 73.46 education level primary or below 16 2.67 up to “matriculation” 70 11.69 higher secondary 296 49.42 graduation and above 217 36.23 employment full-time 273 45.58 part-time 76 12.69 self-employed 115 19.2 unemployed/domestic work/retired 135 22.54 income level up to 25000 69 11.52 25001-50000 113 18.86 50001-75000 98 16.36 75001-100000 116 19.37 >100000 203 33.89 table 2 summarizes the frequency distributions of all factors of variables used in the regression analysis. after the pandemic very few respondents have reported that they have perceived a good relationship with their social contacts. 53% of respondents perceived that they have an extremely good or good relationship with their families and friends, while only 43-45% of respondents perceived a good relationship with relatives and neighbours, whereas this ratio is very low with other people (40%). most of the respondents (40%-50%) perceived that government-level support was average after covid-19 in providing social security allowance, educational needs and other matters. around 35%-60% of respondents perceived that government support was poor in providing basic needs after covid-19, especially in providing jobs and reducing inequality. most of the respondents (55%-80%) are under stress due to rising prices and the poor economic impact of government quality on post-covid subjective well-being situation after the pandemic. additionally, most of the respondents (more than 50%) reported a high level of stress, anxiety and sleep disturbance after the pandemic. the mean scores of happiness and ls are 6.41 and 6.24, respectively. 37% of the respondents reported happiness and ls level below 6, while 63% reported happiness and ls level from 6-10 (figure 1 and 2). in conclusion, covid-19 has badly affected people's social relationships. it has also increased the stress, whether it is economic or personal stress, due to anxiety, hopelessness and sleep disturbances. figure 1. distribution of happiness responses figure 2. distribution of life satisfaction responses 4% 2% 2% 5% 5% 18% 12%15% 14% 7% 16% 0 1 2 3 4 5 6 7 8 9 10 5% 2% 3% 6% 5% 19% 10%13% 14% 9% 14% 0 1 2 3 4 5 6 7 8 9 10 58 danish, ashfaq and azam table 2. frequencies distribution of responses related to social capital, government support, economic stress, stress, and dietary habits. social capital extremely good slightly good neutral slightly bad extremely bad how do you consider your relationship with your friends and family after the outbreak of covid-19? 26.9% 26.4% 39.6% 5.2% 2.0% how do you perceive the relationships with your relatives after the period of the covid-19 17.9% 28.0% 45.9% 6.5% 1.7% how do you perceive the relationships with your neighbours after the period of the covid-19 17.2% 26.5% 48.6% 5.8% 1.8% how do you perceive the relationships with other people after/after the period of the covid-19 12.5% 24.4% 52.4% 8.3% 2.3% government support very poor poor average good very good providing social security allowance 16.0% 20.7% 46.2% 13.5% 3.5% creating jobs 23.7% 32.4% 33.1% 8.5% 2.3% reducing the gap between rich and poor 34.4% 26.2% 29.2% 7.7% 2.5% providing educational needs 20.5% 24.2% 37.7% 12.2% 5.3% providing health care 13.2% 16.5% 36.2% 26.9% 7.2% economic stress very much somewhat neutral not much not at all to what extent do you feel a burden because of your children’s education after covid? 34.9% 22.9% 26.0% 7.2% 9.0% to what extent do you feel stress with the current economic situation of your country after covid? 53.9% 22.4% 18.4% 4.7% 0.7% to what extent do you feel stress with the rising level of prices in your country after covid? 68.6% 11.7% 14.4% 4.0% 1.3% stress yes no have you experienced any stress after/after covid? 72.5% 27.5% have you experienced any anxiety after/after covid? 62.8% 37.2% do you categorize the time before the start of covid as less stressful? 71.0% 29.0% have you experienced any sleep disturbance after covid? 47.7% 52.3% has your sleeping pattern changed since the covid? 57.4% 42.6% have you experienced any feelings of hopelessness after/after covid? 52.1% 47.9% dietary habits yes no have you experienced any loss of appetite after covid? 32.6% 67.4% have your eating habits changed after covid? 49.1% 50.9% have you engaged in stress eating after covid? 30.1% 69.9% has your weight increased after covid? 43.2% 56.8% impact of government quality on post-covid subjective well-being 4.2 gsem analysis the results of the baseline model and structural model are presented in table 3, where the financial stress (fs) is the mediator in the first step and is regressed with employment level, income, economic stress and perceived government role after the pandemic. while in the structural model, happiness and life satisfaction are regressed separately in each model with and without control variables to ensure the robustness of the model. the first column of table 3 denotes all the variables while the second column represents the regressors fs. finally, column three to 6 shows the regressors of happiness and life satisfaction (ls). for the fs model, coefficient values are provided in column 2, and odd ratios (or) are given for the model of happiness and ls, while standard errors are shown in parentheses. results of the study reveal that level of income is negatively related to financial stress. people with higher income (> 100000) are less likely to suffer stress with the financial position of the household by 0.5 standard deviation points than the lower income (<25000). previous studies also suggest that a higher level of income after covid is associated with a lower level of financial stress (cardona-montoya et al., 2022). economic-related stress has also affected households after the pandemic and is significantly related to financial stress among individuals. results of the present study show that respondents who perceived no stress with increasing prices and the economic situation have reported the least financial stress. this means that for people who have perceived stress regarding economic conditions after the pandemic, their financial stress significantly increases by 0.14 standard deviation (s.d) points. our results are in line with the previous study of alola et al. (2021), who find that daily economic uncertainty is associated with a high level of financial stress among households. moreover, part-time and full-time employed persons are less financially stressed than retirees and unemployed. additionally, for people who perceived that the government has supported basic needs after covid, their financial stress is likely to reduce by 0.07 s.d points but this relationship is significant at a 10% level. the government response in the pandemic reduces the consequences of economic risk and reduces the difficulties in financial support (chłoń-domińczak & holzer-żelażewska, 2021). after the outbreak of covid-19, the government of pakistan disbursed many social security allowances like direct transfer payments, health insurance and subsidies, which might have reduced the financial problems due to covering household expenses but, overall, the effect was very minor as government cannot fully support each household due to a low budget and weak financial position of the economy. 60 danish, ashfaq and azam table 3. structural equation model for the association between government quality and swb: mediating role of financial stress (methodology: gsem using tobit and order logit) (1) (2) (3) (4) (5) fs happiness happiness ls ls fs 0.51*** 0.61*** 0.50*** 0.57*** (0.04) (0.05) (0.04) (0.05) employment part-time -0.28** 1.46 1.49 1.39 1.36 (0.14) (0.38) (0.40) (0.36) (0.36) full-time -0.24** 1.26 1.27 1.22 1.12 (0.10) (0.24) (0.25) (0.23) (0.22) self-employed -0.15 0.97 1.02 1.04 1.06 (0.12) (0.22) (0.23) (0.23) (0.24) income 25-50 k -0.09 1.49 1.62* 1.98** 2.02** (0.15) (0.41) (0.46) (0.54) (0.56) 50-75k -0.32** 1.87** 1.84** 1.75** 1.64* (0.15) (0.54) (0.54) (0.50) (0.48) 75-100k -0.54*** 2.42*** 2.30*** 2.65*** 2.49*** (0.15) (0.67) (0.66) (0.74) (0.71) more than 100k -0.75*** 2.37*** 2.46*** 2.40*** 2.37*** (0.13) (0.62) (0.69) (0.63) (0.67) economic stress -0.11*** 1.10 1.10 1.01 1.02 (0.04) (0.09) (0.09) (0.08) (0.08) government quality -0.07* 1.22** 1.14 1.28*** 1.20** (0.04) (0.09) (0.09) (0.10) (0.09) gender: male 1.01 0.95 (0.16) (0.15) social capital 0.67*** 0.76*** (0.05) (0.06) dietary habits 0.50** 0.48*** (0.14) (0.13) covid-related stress 0.56** 0.71 (0.16) (0.20) education up to matriculation 0.85 2.47* (0.44) (1.28) higher secondary 1.04 1.97 (0.51) (0.95) graduation and above 0.90 2.36* (0.46) (1.18) area: urban 1.37* 0.91 (0.24) (0.16) constant 0.60*** (0.13) log likelihood -2064.8667 -2039.7902 -2083.3309 -2067.7581 observations 599 599 599 599 599 note: ***, **, and * indicate level of significance at 1%, 5% and 10% level. the results of the structural model are also presented in table 3, wherein the first step, happiness and ls are regressed with a mediator and its determinants, while in the second step, control variables are also included in the final model. odd ratios (or) are presented for each model by using the gsem. our results strongly reveal that financial stress has significantly reduced the level of happiness and ls after the pandemic by odds of 0.5 (p < 0.01). during the pandemic, the growing level of prices and unemployment in the country has increased the financial stress among families and individuals, which has led to a lower level of happiness and ls. rodrigues et al. (2021) also impact of government quality on post-covid subjective well-being studied the negative effect of financial stress on well-being in families in portugal after the outbreak of covid-19. level of income is positively related to both happiness and ls. individuals in the upperincome group are happier by odds of 2.37 and ls by odds of 2.40. the role of the government cannot be ignored after the pandemic outbreak in reducing financial and mental stress. moreover, a market economy cannot be restored only by market forces, but government support is also a fundamental factor in the recovery process (stiglitz, 2021). results of the present study reveal that for people who perceived good government support in healthcare and other benefits after covid-19, their happiness has raised by odds of 1.22 (or = 1.22; p< 0.05) and ls by odds of 1.28 (or = 1.28; p < 0.01). alamsyah and zhu (2022) reveal that government information quality led to quicker responses among citizens, which reduces stress and anxiety while being positively related to life satisfaction. bad social capital has also significantly reduced the level of happiness by odds of 0.33 (or = 0.67; p < 0.01) and ls by 0.24 (or = 0.76; p < 0.01). due to the long-term lockdown during the pandemic, people avoided meeting with friends, relatives, and neighbours, which badly affected social cohesion and relationships with each other. due to this adverse effect of the pandemic, mental stress is increased among people, which adversely affects happiness. previous studies suggest that stranger trust, family social capital and strong social cohesion is positively related with the level of happiness and psychological well-being (zhao et al., 2022 and sarmiento prieto et al., 2022). people whose dietary habits have changed after covid-19 and have taken to stress eating or perceived any hopelessness, their level of happiness also reduced by 0.50 odds (or = 0.50; p < 0.05) and ls by 0.52 odds (or = 0.48; p < 0.05). during the pandemic, many people started stress eating due to loneliness and frustration, which has decreased physical activity and increased obesity, thus resulting in anxiety and a lower level of emotional well-being (cecchetto et al., 2021). moreover, unhealthy lifestyles among people during the pandemic have also lowered the well-being of people and negatively affected mental health (hu et al., 2020). additionally, stress and anxiety after the pandemic have also negatively affected happiness (or = 0.56; p < 0.05) but it is not significantly related with ls. cohen-louck and levy (2022) also find the negative impact of stress and anxiety on happiness. werner et al. (2021) also evaluated the patterns of sleep and stress in germany during the pandemic and found that covid-19 related stress have negatively affected sleep quality. people living in urban areas are happier than rural areas, but this relationship is less robust (p< 0.1). additionally, employment and education are statistically not significant in the structural model, but they pertain to the positive effect in both models, especially full-time employees who are likely to be happier than part-time employees and the unemployed. finally, males reported more happiness than females, while ls among females was higher than males after the pandemic. however, statistically, this relationship is not significant (p > 0.1). 5. conclusions although covid-19 cases have declined over the last year, the negative impact on society, especially on psychological well-being is still not overcome. social connection among people is still not restored to how it was before covid-19. moreover, government effectiveness has also affected swb during 62 danish, ashfaq and azam the pandemic, especially in developing countries like pakistan. therefore, the present study is designed to examine the effect of perceived government quality on swb (happiness and ls) with the mediation of financial stress after the pandemic by using gsem with the tobit and ordered logit model. the study uses the final data of 599 respondents from all over pakistan. results of the study reveal that most of the respondents had declared a less stressful time before covid-19 and reported sleep disturbance, anxiety and stress after the pandemic. more than 60% of people reported that government quality was average or poor after the pandemic in terms of the economic situation and prices. the present study finds that a higher level of income and improved government quality lower financial stress and increased happiness after covid-19. additionally, lower economic stress is also negatively related to financial stress after the pandemic. results of the study also revealed that financial stress is a strong mediator between government quality and swb and high level of financial stress has significantly reduced the level of happiness and ls after the pandemic. finally, covidrelated stress, poor dietary habits, and poor social capital significantly reduced the level of happiness after the pandemic, while education and gender have no significant impact on this study. policy implications and limitations of the study the present study provides practical implications for policymakers and governments to improve the psychological well-being of people. in this regard, the government needs to regulate the model of equality in health care, education and price control in the country. pakistan is an underdeveloped country, and more than 25 million people lost their jobs after the pandemic, due to which financial stress has increased among people, having a negative effect on their psychological and physical health. the findings of this study suggest that the government needs to encourage investment and provide a chance to youth for business start-ups rather than focusing only on direct transfers. it will not only reduce unemployment but also expand the economy's size, which will also help reduce the stress about the economy and high prices in the country. the present study is limited to pakistan due to financial coverage and online surveys. moreover, this study is restricted to the role of government quality after the pandemic and financial stress. this study will help researchers in the future to extend the hypothesis on different factors, such as the mediation of social capital and the healthcare system. this study can be expanded all over the region, especially in those countries where people have lost trust and happiness. acknowledgements the authors pay special gratitude and thank the respondents of this survey, without whom this study could not be completed. the authors are also grateful to the editors of the european journal of government and economics for considering this paper for publication and reviewing process. we are also obliged to the reviewers for their valuable comments and suggestions to improve this draft. impact of government quality on post-covid subjective well-being references alamsyah, n., & zhu, y. q. (2022). we shall endure: exploring the impact of government information quality and partisanship on citizens' well-being during the covid-19 pandemic. government information quarterly, 39(1), 101646. https://doi.org/10.1016/j.giq.2021.101646 alfawaz, h.a., wani, k., aljumah, a.a., aldisi, d., ansari, m.g., yakout, s.m., . . . al-daghri, n.m. (2021). psychological well-being during covid-19 lockdown: insights from a saudi state university's academic community. journal of king saud university-science, 33(1), 101262. https://doi.org/10.1016/j.jksus.2020.101262 alola, a.a., alola, u.v., & sarkodie, s.a. (2021). the covid-19 and financial stress in the usa: health is wealth. environment, development and sustainability, 23, 9367-9378. https://doi.org/10.1007/s10668-020-01029-w aslam, n., shafique, k., & ahmed, a. (2021). exploring the impact of covid-19-related fear, obsessions, anxiety and stress on psychological well-being among adults in pakistan. the journal of mental health training, education and practice. https://doi.org/10.1108/jmhtep-10-2020-0074 baranov, v., bhalotra, s., biroli, p., & maselko, j. (2020). maternal depression, women's empowerment and parental investment: evidence from a randomized controlled trial. american economic review, 110(3), 824-859. https://doi.org/10.1257/aer.20180511 baranov, v., grosjean, p., khan, f.j., & walker, s. (2022). the impact of covid-related economic shocks on household mental health in pakistan. health economics, 31(10), 2208-2228. https://doi.org/10.1002/hec.4571 bareket-bojmel, l., shahar, g., & margalit, m. (2021). covid-19-related economic anxiety is as high as health anxiety: findings from the usa, the uk and israel. international journal of cognitive therapy, 14, 566-574. https://doi.org/10.1007/s41811-020-00078-3 bauman, a.e., reis, r.s., sallis, j.f., wells, j.c., loos, r.j., & martin, b.w. (2012). correlates of physical activity: why are some people physically active and others not? the lancet, 380(9838), 258-271. https://doi.org/10.1016/s0140-6736(12)60735-1 brooks, s.k., webster, r.k., smith, l. e., woodland, l., wessely, s., greenberg, n., & rubin, g.j. (2020). the psychological impact of quarantine and how to reduce it: rapid review of the evidence. the lancet, 395(10227), 912-920. https://doi.org/10.1016/s0140-6736(20)30460-8 brown, a., flint, s.w., kalea, a.z., o'kane, m., williams, s., & batterham, r.l. (2021). negative impact of the first covid-19 lockdown upon health-related behaviours and psychological well-being in people living with severe and complex obesity in the uk. eclinicalmedicine, 34, 100796. https://doi.org/10.1016/j.eclinm.2021.100796 cameron, a.c., & trivedi, p.k. (2005). microeconometrics: methods and applications. cambridge university press. https://doi.org/10.1017/cbo9780511811241 cardona-montoya, r.a., cruz, v., & mongrut, s.a. (2022). financial fragility and financial stress during the covid-19 crisis: evidence from colombian households. journal of economics, finance and administrative science, 27 (54), 376-393. https://doi.org/10.1108/jefas-01-2022-0005 cecchetto, c., aiello, m., gentili, c., ionta, s., & osimo, s.a. (2021). increased emotional eating during covid19 associated with lockdown, psychological and social distress. appetite, 160, 105122. https://doi.org/10.1016/j.appet.2021.105122 ceri, v., & cicek, i. (2021). psychological well-being, depression and stress during covid-19 pandemic in turkey: a comparative study of healthcare professionals and non-healthcare professionals. psychology, health & medicine, 26(1), 85-97. https://doi.org/10.1080/13548506.2020.1859566 cheng, t.c., kim, s., & koh, k. (2020). the impact of covid-19 on subjective well-being: evidence from singapore. https://doi.org/10.2139/ssrn.3695403 chew, n.w., ngiam, j. n., tan, b. y.-q., tham, s.-m., tan, c. y.-s., jing, m., . . . ahmad, a. (2020). asian-pacific perspective on the psychological well-being of healthcare workers during the evolution of the covid-19 pandemic. bjpsych open, 6(6). https://doi.org/10.1192/bjo.2020.98 chłoń-domińczak, a., & holzer-żelażewska, d. (2021). economic stress of people 50+ in european countries in the covid-19 pandemic-do country policies matter? european journal of ageing, 120. https://doi.org/10.1007/s10433-021-00662-2 cohen-louck, k., & levy, i. (2022). happiness during a mass trauma: predicting happiness during the covid-19 pandemic through function, stress, anxiety and coping. psychological trauma: theory, research, practice and policy. https://doi.org/10.1037/tra0001314 danish, m.h., & nawaz, s.m.n. (2022). does institutional trust and governance matter for multidimensional well-being? insights from pakistan. world development perspectives, 25, 100369. https://doi.org/10.1016/j.wdp.2021.100369 https://doi.org/10.1016/j.giq.2021.101646 https://doi.org/10.1016/j.jksus.2020.101262 https://doi.org/10.1007/s10668-020-01029-w https://doi.org/10.1257/aer.20180511 https://doi.org/10.1007/s41811-020-00078-3 https://doi.org/10.1016/s0140-6736(12)60735-1 https://doi.org/10.1016/s0140-6736(20)30460-8 https://doi.org/10.1016/j.eclinm.2021.100796 https://doi.org/10.1017/cbo9780511811241 https://doi.org/10.1108/jefas-01-2022-0005 https://doi.org/10.1016/j.appet.2021.105122 https://doi.org/10.1080/13548506.2020.1859566 https://doi.org/10.2139/ssrn.3695403 https://doi.org/10.1192/bjo.2020.98 https://doi.org/10.1007/s10433-021-00662-2 https://doi.org/10.1037/tra0001314 https://doi.org/10.1016/j.wdp.2021.100369 64 danish, ashfaq and azam de kock, j.h., latham, h.a., leslie, s j., grindle, m., munoz, s.-a., ellis, l., . . . o'malley, c.m. (2021). a rapid review of the impact of covid-19 on the mental health of healthcare workers: implications for supporting psychological well-being. bmc public health, 21(1), 1-18. https://doi.org/10.1186/s12889-020-10070-3 dragun, r., veček, n.n., marendić, m., pribisalić, a., đivić, g., cena, h., . . . kolčić, i. (2020). have lifestyle habits and psychological well-being changed among adolescents and medical students due to covid-19 lockdown in croatia? nutrients, 13(1), 97. https://doi.org/10.3390/nu13010097 godinić, d., & obrenovic, b. (2020). effects of economic uncertainty on mental health in the covid-19 pandemic context: social identity disturbance, job uncertainty and psychological well-being model. international journal of innovation and economic development, 6(1). https://doi.org/10.18775/ijied.1849-7551-7020.2015.61.2005 greyling, t., rossouw, s., & adhikari, t. (2020). happiness-lost: did governments make the right decisions to combat covid-19? glo discussion paper. hammami, a., harrabi, b., mohr, m., & krustrup, p. (2022). physical activity and coronavirus disease 2019 (covid-19): specific recommendations for home-based physical training. managing sport and leisure, 27(1-2), 26-31. https://doi.org/10.1080/23750472.2020.1757494 helliwell, j.f., layard, r., sachs, j.d., & neve, j.e.d. (2021). world happiness report 2021. holton, s., wynter, k., trueman, m., bruce, s., sweeney, s., crowe, s., . . . hitch, d. (2020). psychological wellbeing of australian hospital clinical staff during the covid-19 pandemic. australian health review, 45(3), 297-305. https://doi.org/10.1071/ah20203 huang, y., & zhao, n. (2020). generalized anxiety disorder, depressive symptoms and sleep quality during covid-19 outbreak in china: a web-based cross-sectional survey. psychiatry research, 288, 112954. https://doi.org/10.1016/j.psychres.2020.112954 johnson, m.s., skjerdingstad, n., ebrahimi, o.v., hoffart, a., & johnson, s.u. (2022). parenting in a pandemic: parental stress, anxiety and depression among parents during the government-initiated physical distancing measures following the first wave of covid-19. stress and health, 38(4), 637-652. https://doi.org/10.1002/smi.3120 kim, s., & kim, d. (2012). does government make people happy?: exploring new research directions for government's roles in happiness. journal of happiness studies, 13, 875-899. https://doi.org/10.1007/s10902-011-9296-0 lenzen, m., li, m., malik, a., pomponi, f., sun, y.-y., wiedmann, t., . . . geschke, a. (2020). global socioeconomic losses and environmental gains from the coronavirus pandemic. plos one, 15(7), e0235654. https://doi.org/10.1371/journal.pone.0235654 lesser, i.a., & nienhuis, c.p. (2020). the impact of covid-19 on physical activity behavior and well-being of canadians. international journal of environmental research and public health, 17(11), 3899. https://doi.org/10.3390/ijerph17113899 li, l.z., & wang, s. (2020). prevalence and predictors of general psychiatric disorders and loneliness during covid-19 in the united kingdom. psychiatry research, 291, 113267. https://doi.org/10.1016/j.psychres.2020.113267 lu, w., wang, h., lin, y., & li, l. (2020). psychological status of medical workforce during the covid-19 pandemic: a cross-sectional study. psychiatry research, 288, 112936. https://doi.org/10.1016/j.psychres.2020.112936 maselko, j., bates, l., bhalotra, s., gallis, j. a., o'donnell, k., sikander, s., & turner, e.l. (2018). socioeconomic status indicators and common mental disorders: evidence from a study of prenatal depression in pakistan. ssm-population health, 4, 1-9. https://doi.org/10.1016/j.ssmph.2017.10.004 muhammad gadit, a.a., & mugford, g. (2007). prevalence of depression among households in three capital cities of pakistan: need to revise the mental health policy. plos one, 2(2), e209. https://doi.org/10.1371/journal.pone.0000209 nguyen, j., & brymer, e. (2018). nature-based guided imagery as an intervention for state anxiety. frontiers in psychology, 9, 1858. https://doi.org/10.3389/fpsyg.2018.01858 petrovič, f., murgaš, f., & králik, r. (2021). happiness in czechia during the covid-19 pandemic. sustainability, 13(19), 10826. https://doi.org/10.3390/su131910826 rizvi, s.k.a., yarovaya, l., mirza, n., & naqvi, b. (2022). the impact of covid-19 on the valuations of nonfinancial european firms. heliyon, 8(6), e09486. https://doi.org/10.1016/j.heliyon.2022.e09486 rodrigues, m., silva, r., & franco, m. (2021). covid-19: financial stress and well-being in families. journal of family issues, 0192513x211057009. https://doi.org/10.1177/0192513x211057009 saalwirth, c., & leipold, b. (2021). well-being and sleep in stressful times of the covid-19 pandemic: relations to worrying and different coping strategies. stress and health, 37(5), 973-985. https://doi.org/10.1002/smi.3057 sallis, j.f., owen, n., & fisher, e. (2015). ecological models of health behavior. health behavior: theory, research and practice, 5(43-64). https://doi.org/10.1186/s12889-020-10070-3 https://doi.org/10.3390/nu13010097 https://doi.org/10.18775/ijied.1849-7551-7020.2015.61.2005 https://doi.org/10.1080/23750472.2020.1757494 https://doi.org/10.1071/ah20203 https://doi.org/10.1016/j.psychres.2020.112954 https://doi.org/10.1002/smi.3120 https://doi.org/10.1007/s10902-011-9296-0 https://doi.org/10.1371/journal.pone.0235654 https://doi.org/10.3390/ijerph17113899 https://doi.org/10.1016/j.psychres.2020.113267 https://doi.org/10.1016/j.psychres.2020.112936 https://doi.org/10.1016/j.ssmph.2017.10.004 https://doi.org/10.1371/journal.pone.0000209 https://doi.org/10.3389/fpsyg.2018.01858 https://doi.org/10.3390/su131910826 https://doi.org/10.1016/j.heliyon.2022.e09486 https://doi.org/10.1177/0192513x211057009 https://doi.org/10.1002/smi.3057 impact of government quality on post-covid subjective well-being sarmiento prieto, j.p., castro correa, c.p., arrieta, a., jerath, m., & arensburg, s. (2022). relevance of social capital in preserving subjective well-being in the face of the covid-19 pandemic. risk, hazards & crisis in public policy. https://doi.org/10.1002/rhc3.12260 stiglitz, j.e. (2021). the proper role of government in the market economy: the case of the post-covid recovery. journal of government and economics, 1, 100004. https://doi.org/10.1016/j.jge.2021.100004 tobin, j. (1958). estimation of relationships for limited dependent variables. econometrica: journal of the econometric society, 24-36. https://doi.org/10.2307/1907382 tran, b.x., nguyen, h.t., le, h.t., latkin, c a., pham, h.q., vu, l.g., . . . ta, n.t.k. (2020). impact of covid-19 on economic well-being and quality of life of the vietnamese during the national social distancing. frontiers in psychology, 11, 565153. https://doi.org/10.3389/fpsyg.2020.565153 wang, h., he, l., gao, y., gao, x., & lei, x. (2021). effects of physical activity and sleep quality on well‐being: a wrist actigraphy study during the pandemic. applied psychology: health and well-being, 13(2), 394-405. https://doi.org/10.1111/aphw.12255 werner, a., kater, m.j., schlarb, a.a., & lohaus, a. (2021). sleep and stress in times of the covid-19 pandemic: the role of personal resources. applied psychology: health and well-being, 13(4), 935-951. https://doi.org/10.1111/aphw.12281 xi, c., haiyan, g., yuchun, z., & fen, l. (2020). changes in psychological well-being, attitude and information-seeking behavior among people at the epicenter of the covid-19 pandemic: a panel survey of residents in hubei province, china. epidemiology and infection, 148, e201. https://doi.org/10.1017/s0950268820002009 zhao, q., sun, x., xie, f., chen, b., wang, l., hu, l., & dai, q. (2021). impact of covid-19 on psychological well-being. international journal of clinical and health psychology, 21(3), 100252. https://doi.org/10.1016/j.ijchp.2021.100252 zhao, x., liu, q., zhang, s., li, t., & hu, b. (2022). the impact of psychological capital and social capital on residents' mental health and happiness during covid-19: evidence from china. frontiers in psychology, 13. https://doi.org/10.3389/fpsyg.2022.962373 https://doi.org/10.1002/rhc3.12260 https://doi.org/10.1016/j.jge.2021.100004 https://doi.org/10.2307/1907382 https://doi.org/10.3389/fpsyg.2020.565153 https://doi.org/10.1111/aphw.12255 https://doi.org/10.1111/aphw.12281 https://doi.org/10.1017/s0950268820002009 https://doi.org/10.1016/j.ijchp.2021.100252 https://doi.org/10.3389/fpsyg.2022.962373 66 danish, ashfaq and azam appendix. the direct impact of regressors on fsi, happiness and ls (1) (2) (3) fsi happiness ls fsi 0.61*** 0.57*** (0.05) (0.05) gender: male -0.08 1.01 0.95 (0.08) (0.16) (0.15) social capital 0.18*** 0.67*** 0.76*** (0.04) (0.05) (0.06) government quality 0.01 1.14 1.20** (0.04) (0.09) (0.09) dietary habits 0.36*** 0.50** 0.48*** (0.13) (0.14) (0.13) covid-related stress 0.72*** 0.56** 0.71 (0.14) (0.16) (0.20) area: urban 0.03 1.37* 0.91 (0.08) (0.24) (0.16) education up to matriculation 0.11 0.85 2.47* (0.25) (0.44) (1.28) higher secondary 0.23 1.04 1.97 (0.23) (0.51) (0.95) graduation and above 0.27 0.90 2.36* (0.24) (0.46) (1.18) employment part-time -0.30** 1.49 1.36 (0.13) (0.40) (0.36) full-time -0.17* 1.27 1.12 (0.10) (0.25) (0.22) self-employed -0.15 1.02 1.06 (0.11) (0.23) (0.24) income 25-50 k -0.13 1.62* 2.02** (0.14) (0.46) (0.56) 50-75k -0.29** 1.84** 1.64* (0.14) (0.54) (0.48) 75-100k -0.49*** 2.30*** 2.49*** (0.14) (0.66) (0.71) more than 100k -0.74*** 2.46*** 2.37*** (0.13) (0.69) (0.67) economic stress -0.07* 1.10 1.02 (0.04) (0.09) (0.08) constant -0.22 (0.26) log likelihood -2064.8667 -2039.7902 observations 599 599 599 note: ***, **, and * indicate level of significance at 1%, 5% and 10% level. 1. introduction 2. literature review 3. material and methods 4. results 5. conclusions references ©the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 1 (2025), pages 71-95 https://doi.org/10.17979/ejge.2025.14.1.11247 submitted: sep 23, 2024 accepted: feb 20, 2025 published: jun 20, 2025 article comparative evaluation of quality of life indicators: towards developing a comprehensive standard model for measuring quality of life jalal hafeth ahmad abu-alrop,1,* 1 kazan federal university, russian federation *correspondence: jalalabualrop@gmail.com abstract. this study aims to improve the understanding of quality of life (qol) by evaluating key indicators, identifying influential factors, and assessing their relative weights. this study utilizes panel data from 181 cases across 32 european countries (2012-2017). the study employs exploratory and confirmatory factor analyses to create a comprehensive qol measurement model. four primary factors were found to influence qol in europe: economy, health, education, and governance quality. the study compares five common qol indicators—income, gdp, life expectancy, the human development index (hdi), and the legatum prosperity index—against this model. results show that the hdi is the most balanced indicator, while others exhibit biases. the study emphasizes the need for more precise and comprehensive qol measures and recommends applying exploratory and confirmatory factor analyses to enhance them. future research should validate the model in other regions and further improve qol measurement. keywords: quality of life; human development index; legatum prosperity index jel classification: i31; a1 1. introduction the concept of quality of life (qol) is inherently complex and multidimensional, making it difficult to define in a scientifically rigorous manner. this complexity arises from the open nature of the concept and the lack of a unified theoretical framework, which has led to ongoing debates about the possibility of developing a comprehensive qol theory. furthermore, the indicators used to measure qol are often criticized for being either too simplistic, failing to provide a complete representation of reality, or overly complicated, raising concerns about their validity. studies have revealed significant variations among these indicators, including their sensitivity to different variables, the diversity of approaches, and the weight assigned to each factor. given these challenges, it is essential to evaluate and compare these diverse indicators to better understand their strengths and weaknesses. such an analysis will not only clarify the qol concept but also strengthen its theoretical https://creativecommons.org/licenses/by-nc/4.0/ 72 jalal hafeth ahmad abu-alrop foundations, thereby improving the effectiveness of qol measurements. this study presents a novel standard model for measuring qol, developed using exploratory factor analysis (efa) and confirmatory factor analysis (cfa). the model identifies four core factors influencing qol in europe: economic quality, health quality, education quality, and governance quality. these factors are then used to evaluate five commonly employed qol indicators: income, gross domestic product (gdp), life expectancy at birth, the human development index (hdi), and the legatum prosperity index. through multiple regression analysis, the study examines the relative weights and biases of these indicators, highlighting the strengths and limitations inherent in each. the results provide critical insights, with hdi emerging as the most balanced indicator, while others, such as gdp and the legatum prosperity index exhibit distinct biases toward economic and governance factors, respectively. this empirical analysis provides evidence for the need to reassess how qol indicators are weighted and suggests improvements for more accurate and inclusive measurement. the significance of this research lies in its contribution to the development of a more comprehensive and balanced framework for assessing qol. the proposed standard model and the findings of this study offer valuable insights for policymakers and researchers seeking to refine qol measurements in europe and beyond. this study follows a structured approach to provide a comprehensive analysis of the topic of quality of life. after the abstract, the first chapter presents an introduction that outlines the general framework of the study, highlighting its significance and objectives. the second chapter reviews the relevant literature, focusing on the concept of quality of life, its theoretical foundations based on maslow's hierarchy of needs, and the methods used for its measurement. the third chapter details the methodology and data used in the analysis. this is followed by the fourth chapter, which presents the empirical results, beginning with exploratory factor analysis (efa) to identify the key determinants of quality of life in europe, followed by confirmatory factor analysis (cfa) to establish a standardized measurement model, and finally, multiple regression analysis to assess the impact of these factors on key quality of life indicators. the fifth chapter provides the main conclusions and recommendations derived from the study. finally, the research includes a list of references supporting the analysis and findings. 2. literature review 2.1 concept of quality of life brown, bowling, & flynn (2004) note that quality of life concept is often based on expert opinions rather than individual perspectives, which limits their relevance. nonetheless, quality of life isn't entirely subjective, especially politically, where satisfaction plays a crucial role. costanza, et al. (2007) define quality of life as the degree of satisfaction or dissatisfaction in various life areas. oxford university (2011) explains that satisfaction comes from fulfilling desires, expectations, or needs, representing a subjective judgment influenced by past experiences and current expectations (campbell, converse, & rodgers, 1976; campbell, 1981; morris & winter, 1978; potter & cantarero, 2014). community satisfaction pertains to how well individuals feel their community meets their needs (matarrita-cascante, 2010). deseran (1978) argues that society is subjectively reconstituted as an objective reality, comparative evaluation of quality of life indicators making it a multidimensional phenomenon. consequently, community satisfaction cannot always be objectively measured (christensen & levinson, 2003). marans & rodgers (1975) note that satisfaction depends on both objective conditions and individual values and expectations, meaning objective characteristics alone do not reflect personal feelings. studies indicate that individuals can be satisfied with their communities even in suboptimal environments (campbell, converse, & rodgers, 1976; gulick, bowerman, & back, 1962; hollingshead & rogler, 1963). furthermore, research shows that rural residents are often more satisfied with their communities than urban residents (campbell, 1981; marans & rodgers, 1975; theodori, 2001). the world health organization (2012) defines quality of life as individuals' perceptions of their position in life within their cultural and value contexts. malkina-pykh & pykh (2008) describe quality of life as reflecting overall experience and well-being. brown, bowling, & flynn (2004) identify two dimensions: societal (income, employment, housing, education, and environmental conditions) and individual (perceptions, values, well-being, happiness, and life satisfaction). thus, quality of life is a multidimensional concept that extends beyond economic status (eurostat, 2015) and includes social, biomedical, psychological, economic, and environmental factors (alan walker, 2005). this complex concept encompasses objective and subjective elements, communal and individual aspects, and both positive and negative interactions (lawton, 1991; von tesch-ro & motel-klingebiel, 2001). quality of life is a dynamic and complex concept, lacking a clear scientific definition and consensus (hajdouva, andrejovsky, gajdos, & andrejkovic, 2011; alan walker, 2005). this ambiguity arises from its open nature, individual focus, and lack of theoretical foundations. cultural values and preferences further complicate its universal definition and measurement (diener, oishi, & tay, 2018; walker & van der, 2004). the possibility of a unified theory of quality of life is debated due to its multidimensional nature (alan walker, 2005). however, such a theory could improve the application of quality of life measures in policy (noll, 2002). 2.2 maslow's theory and the hierarchy of needs as a basis for the concept of quality of life the complexity of the concept of quality of life stems from the complexity of human nature, as it revolves around fulfilling human needs and expectations, which, when met, bring satisfaction and happiness. assessments of quality of life vary among researchers due to the subjective nature of needs and priorities, making reliance on individual satisfaction insufficient. for instance, a poor person may feel content when receiving a free daily meal, whereas a billionaire may experience dissatisfaction despite their wealth. this illustrates that the feeling of satisfaction does not necessarily reflect actual quality of life, and material well-being alone does not appear to be a sufficient measure of quality of life. according to maslow (1943), as cited in green (2000), some individuals lose ambition under conditions of poverty, leading to the satisfaction and disappearance of less critical goals such as love, self-esteem, focusing solely on physiological needs, security, and stability, thereby distorting the concept of satisfaction. accordingly, the measurement of quality of life should be based on objective scientific criteria that rely on a deep understanding of human psychology and the mechanisms for achieving stability, happiness, and true satisfaction, rather than 74 jalal hafeth ahmad abu-alrop on varying personal opinions. additionally, according to maslow (1943), as cited in green (2000), desires and behaviors vary superficially across cultures, yet fundamental human needs remain universal. the more individuals engage with different societies, the greater their awareness of the similarity in human needs, while differences remain confined to behavioral patterns and their expression. therefore, it can be argued that the concept of quality of life is a singular notion that does not fundamentally differ across cultures, with variations existing only in behavioral patterns and their modes of expression. maslow’s (1943) theory of the hierarchy of human needs, as cited in green (2000), can serve as a solid scientific and objective foundation for the concept of quality of life. this theory posits that individuals seek to fulfill their needs hierarchically, beginning with physiological needs (food, water, shelter), followed by safety needs (financial and health stability), then social needs (belonging and social relationships), esteem needs (achievement and social recognition), and finally selfactualization (personal development and goal fulfillment). humans are inherently driven by desire, with their needs shaped by biological, cultural, and contextual factors. the emergence of one need depends on the satisfaction of a stronger, lower-level need in the hierarchy. a person cannot attain higher levels unless the lower levels are adequately met (maslow, 1943, as cited in green, 2000). physiological needs are the strongest of all needs. if these needs remain unmet, other needs, such as safety, freedom, and love, recede into the background. for example, an individual experiencing extreme hunger may define their ideal world as a place abundant in food, believing that securing food for life would bring complete happiness and eliminate all other desires. if physiological needs are well satisfied, safety needs emerge as the next dominant priority, with security and stability becoming the primary goals. once safety needs are fulfilled, social needs become the next focus, and a person may intensely feel the absence of friends or a romantic partner, even though they may have once dismissed love as trivial when struggling with hunger and poverty. upon satisfying social needs, esteem needs emerge, which are divided into two subcategories: the first involves selfconfidence, self-respect, and the desire for power, achievement, competence, independence, and freedom. the second encompasses the need for reputation, fame, recognition, and attention based on respect or appreciation from others. the failure to fulfill esteem needs leads to feelings of inferiority, weakness, and helplessness. once esteem needs are met, self-actualization needs emerge, representing the highest level of the hierarchy, where individuals strive to develop their abilities and achieve their full potential. self-actualization is not necessarily linked to creativity; rather, it reflects a person’s desire to become more and more of what they are (maslow, 1943, as cited in green, 2000). most individuals only partially satisfy their basic needs, with satisfaction levels decreasing as they ascend maslow's hierarchy. for example, an average person may achieve 85% satisfaction in physiological needs, 70% in safety needs, 50% in social needs, 40% in esteem needs, and only 10% in self-actualization. each new level emerges gradually as the lower level is satisfied, often unnoticed at first. these needs are often unconscious, with unconscious motivation playing a larger role, although this becomes more apparent among intellectually developed and educated individuals (maslow, 1943, as cited in green, 2000). other human needs are not ending in themselves but are interconnected sub-goals tied to fundamental needs. for example, cognitive needs—such as the pursuit of knowledge, understanding, and analysis—serve as a means to achieve security for the average person, while they represent a comparative evaluation of quality of life indicators form of self-actualization for highly intelligent individuals. thus, cognitive needs are an integral part of the fundamental hierarchy of needs. moreover, the hierarchy of needs is not entirely fixed; the sequence may be altered for some individuals due to psychological or social factors. for example, some people may view self-esteem as more important than love, leading them to exhibit aggressive behavior as a display of power. this can be explained by the fact that they actually lack love and seek it by imitating the traits of lovable individuals, who are often perceived as powerful. if fundamental needs remain unmet, this results in psychological imbalances that can be considered a form of "illness," just as nutritional deficiencies are regarded as health issues. a psychologically healthy person strives to realize their full potential, and if they are driven by secondary desires rather than fundamental needs, this reflects an unhealthy state. therefore, understanding quality of life necessitates focusing on the extent to which individuals can fulfill their essential needs, as this directly impacts their psychological and social well-being. when examining quality of life through the lens of the hierarchy of basic human needs, it becomes evident that it is influenced by the level of satisfaction of these needs. if an individual experiences deprivation in physiological needs, their quality of life will be low. conversely, achieving higher-level needs, such as self-actualization, should correspond to a high quality of life, even in the absence of some material resources. based on this premise, quality of life can be defined as the extent to which an individual is able to meet their fundamental needs, including physiological needs, safety needs, social needs, esteem needs, and finally, self-actualization needs. a society with a high quality of life is one that provides individuals with the appropriate environment to achieve self-actualization, where the primary motivation for a healthy person is the development of their full potential and capabilities. from this perspective, any deficiency in meeting these needs can be considered a form of illness, much like deficiencies in vitamins or minerals, which are recognized as health issues. 2.3 measuring quality of life given the complexity of the concept and differing perspectives, it is unsurprising that there is no consensus on defining and measuring quality of life. despite this, scientific research on the topic has significantly increased in citations (alan walker, 2005). though extensive research has been conducted, a consensus on conceptualization and measurement remains elusive. various organizations and authors propose sets of dimensions and composite indicators to measure progress and facilitate comparisons between countries (estes & sirgy, 2019). brown, bowling, & flynn (2004) identify eight models of quality of life: objective social indicators like living standards, health, and longevity (walker & van der, 2004); fulfillment of human needs (maslow, 1954; bigelow, mcfarland, & olson 1991); subjective indicators such as life satisfaction, psychological well-being, morale, self-respect, personal achievement, and happiness (clarke, marshall, ryff, & rosenthal, 2000); social capital in the form of personal resources (knipscheer, de jong-gierveld, van tilburg, & dykstra, 1995); environmental and neighborhood resources (scharf, phillipson, & smith, 2004); health and performance focusing on physical and mental capabilities (graham & kenealy, 2004); psychological models of factors (grundy & bowling, 1999); and interpretive approaches that emphasize individual values, interpretations, and perceptions (gabriel & bowling, 2004). quality of life is considered a dynamic, multifaceted concept that reflects the interplay 76 jalal hafeth ahmad abu-alrop between objective and subjective aspects, macro and micro levels, and both positive and negative influences. therefore, it is often measured through various domains (hughes, 1990; grundy & bowling, 1999). a multidimensional approach is necessary to capture the complexities of people's experiences and assess their well-being accurately. this approach helps policymakers understand the challenges faced by individuals or communities more comprehensively and develop targeted interventions (diener & biswas-diener, 2002). numerous models and measures of quality of life exist, ranging from simple single-variable models to complex multi-variable ones. for this paper, we have selected five commonly used indicators: income, gross domestic product (gdp), life expectancy at birth, human development index (hdi), and the legatum prosperity index. these indicators will be evaluated and analyzed to determine which best represents quality of life. 2.3.1 income as an indicator of quality of life income refers to the money individuals or households regularly receive from sources such as wages, salaries, investments, or government benefits. it is a crucial indicator of quality of life and economic well-being, providing insight into the financial resources available to meet basic needs and pursue personal growth (stiglitz et al. 2009). income is commonly used to reflect an individual or household’s financial resources and is considered a key factor in determining living standards (diener & biswas-diener, 2002). there is scientific consensus that income is positively associated with quality of life due to its influence on various aspects of well-being, such as better housing, healthcare, education, recreational activities, savings, and cultural experiences (diener & biswas-diener, 2002). higherincome individuals generally enjoy a higher standard of living. additionally, studies suggest that income can enhance quality of life when spent on meaningful activities, like social networks and charitable activities (dunn, aknin, & norton, 2008). research indicates that well-being increases with income, even above $75,000 a year (kahneman & deaton, 2010). however, the relationship is not linear; beyond a certain point, additional income does not significantly improve quality of life (easterlin, 1974). conversely, an excessive desire for wealth can negatively affect job satisfaction and quality of life (tang, 2007). thus, income's impact on quality of life varies depending on how it is perceived and used. income inequality and disparities in income distribution significantly impact the overall quality of life within societies. such disparities affect individuals' access to critical resources necessary for enhancing well-being. those with lower incomes often struggle to afford basic necessities like food, shelter, and healthcare, severely impacting their quality of life (diener & biswas-diener, 2002). moreover, wide gaps between high and low-income earners can lead to social fragmentation, higher crime rates, reduced community trust, and limited access to essential services such as education and healthcare (stiglitz, sen, & fitoussi, 2009). this inequality also restricts opportunities for social mobility and hinders the formation of social capital within communities. therefore, income serves as a complex and multidimensional indicator of quality of life, and its impact varies significantly depending on individual circumstances and broader societal contexts. comparative evaluation of quality of life indicators addressing income inequality is crucial for promoting more equitable access to resources and improving overall quality of life for all members of society. 2.3.2 gross domestic product (gdp) as an indicator of quality of life gross domestic product (gdp) is a widely recognized measure of a country's economic health and well-being. it represents the total value of all goods and services produced within a specific period. gross domestic product (gdp) is often used as an indicator of the standard of living, reflecting the material well-being and quality of life of a country's population. it is extensively utilized due to its simplicity, availability of data, and its ability to provide a comparative overview between countries, enabling policymakers to identify trends and patterns related to economic growth. higher gross domestic product (gdp) levels are frequently associated with increased access to resources such as healthcare, education, and infrastructure, thus implying an improvement in quality of life. 2.3.3 life expectancy at birth (lex) as an indicator of quality of life life expectancy at birth is a fundamental measure indicating the average lifespan individuals can expect, based on current mortality rates (world health organization, 2012). it serves as a critical indicator of quality of life by reflecting the overall health status and living conditions within communities and societies. higher life expectancy values typically signify advancements in public health policies, healthcare systems, living standards, and social conditions. these advancements contribute to longer and healthier lives, enabling individuals to potentially experience more fulfilling life events and make positive contributions to society. moreover, longer life spans offer individuals increased opportunities for personal growth and self-fulfillment (world health organization, 2012). life expectancy at birth varies widely among countries and regions due to several factors including economic development, healthcare systems, public health policies, prevalence of infectious diseases, social behaviors, and environmental conditions (eurostat, quality of life indicators measuring quality of life, 2024; raleigh, 2019). advances in healthcare, such as medical technologies, early diagnosis, disease treatment, and preventive measures, significantly contribute to increasing life expectancy and enhancing quality of life (world health organization, 2019; murray et al., 2020). education also plays a crucial role in improving both longevity and quality of life. educated individuals tend to make healthier lifestyle choices informed by knowledge and awareness, such as maintaining regular exercise routines, adopting balanced diets, and avoiding harmful habits like smoking or excessive alcohol consumption (cutler & lleras-muney, 2010). education empowers individuals to make informed decisions about their health needs and encourages proactive health management (baker et al., 2011). socioeconomic conditions are also influential factors in life expectancy and quality of life. higher socioeconomic status typically correlates with better access to healthcare services, improved living conditions, and greater opportunities for personal development. conversely, individuals from lower economic backgrounds often face barriers that limit their access to essential resources and opportunities for growth (marmot, friel, bell, houweling, & taylor, 2008). in summary, life expectancy at birth serves as a comprehensive indicator of quality of life by reflecting the interplay of healthcare, education, and socioeconomic factors that contribute to 78 jalal hafeth ahmad abu-alrop individuals' overall well-being and longevity. 2.3.4 human development index (hdi) as an indicator of quality of life the human development index (hdi) is a widely used composite statistical measure that evaluates the quality of life and human development in various countries. hdi encompasses three primary dimensions: a long and healthy life, knowledge, and a decent standard of living. this indicator aims to provide a comprehensive understanding of well-being beyond economic metrics alone, positing that people and their capabilities should be the ultimate criteria for assessing a country's development, not just economic growth. developed by pakistani economist mahbub ul haq, hdi is used by the human development report office of the united nations development program (elizabeth, 2007). hdi is a simple and easy-to-understand measure covering multiple dimensions of human development relevant to all countries, allowing for comparisons of human development between countries and regions and tracking progress over time. organizations like the world bank and the world economic forum use it to gauge countries' progress towards achieving sustainable development goals. before 2010, hdi was calculated using three dimensions: health (measured by life expectancy at birth), education (measured by adult literacy rate and gross enrollment ratio), and standard of living (measured by gross domestic product (gdp) per capita at purchasing power parity). the (undp, 2010) updated hdi calculation to include three dimensions: health (measured by the life expectancy index, with a maximum value of 1 at 85 years and a minimum value of 0 at 20 years), education (measured by the mean years of schooling index with a maximum of 15 years and the expected years of schooling index with a maximum of 18 years in most countries), and standard of living (measured by the income index, with a maximum value of 1 at $75,000 gni per capita and a minimum value of 0 at $100 gni per capita). the hdi is calculated as the geometric mean of normalized indices for each of the three dimensions. 2.3.5 legatum prosperity index (leg) as an indicator of quality of life the legatum prosperity index, developed by the legatum institute in 2007, is a comprehensive tool designed to assess the overall quality of life and prosperity in various countries. to provide a holistic evaluation of prosperity, the index divides dimensions into three main domains encompassing 12 pillars: inclusive societies, open economies, and empowered people. the inclusive societies domain consists of safety and security, personal freedom, governance, and social capital. the open economies domain includes investment environment, enterprise conditions, market access and infrastructure, and economic quality. the empowered people domain consists of living conditions, health, education, and the natural environment. these pillars are represented by 67 elements, each comprising a set of 300 indicators selected based on their ability to create a good representation of the relevant element, supported by conceptual and statistical reasons from academic literature. indicators are also chosen to be likely causes of wealth and well-being. after selecting indicators and filling in missing data points, the data undergo a normalization process and are aggregated to produce composite scores at the element level, then further aggregated at the pillar and domain levels, and finally for the index as a whole. comparative evaluation of quality of life indicators 2.3.6 criticisms of quality of life indicators the aforementioned indicators and others have been criticized individually and collectively. for instance, income indicators can be misleading due to inflationary effects or differences in the cost of living across regions or countries. life expectancy at birth does not account for the distribution of mortality across different age groups. gdp does not reflect income equality and wealth distribution. hdi can be sensitive to the choice of indicators, weights, and aggregation methods. the legatum prosperity index faces criticism over the selection and weighting of indicators, as different stakeholders might have varying opinions on the most important dimensions or how to measure them. while the legatum prosperity index has been criticized for its complexity, income indicators, life expectancy at birth, gdp, and hdi have been criticized for their simplicity, as they do not capture the complexity of the quality of life concept and overlook other crucial aspects such as wealth distribution, social justice, human rights, political freedom and participation, political stability, social support systems, social cohesion, psychological well-being, personal values, job security, work-life balance, environmental quality, and cultural differences (diener & seligman, 2004; stiglitz, sen, & fitoussi, 2009; world bank group, 2018; ryff & singer, 2008; world health organization, 2012). after reviewing the five indicators of quality of life and the related literature, significant differences in opinions can be observed between supporters and critics. criticism is directed at all indicators. some criticize simple indicators for not reflecting the complexity and multidimensionality of the concept of quality of life. others criticize composite and complex indicators, questioning the selection of variables and their relative weights. this reflects a lack of consensus on a clear theory of quality of life. the study of quality of life remains a branch of social science where definitive conclusions cannot be reached. each indicator for measuring quality of life has its strengths and weaknesses, highlighting the importance of comparing them to discover which is the most representative of the concept of quality of life. comparing these indicators allows for the observation of their differences and opens the door to their development or the creation of more representative indicators. beslerová & dzuričková (2014) observed significant differences between the human development index (hdi) and the legatum prosperity index. these differences are mainly due to the diversity and sensitivity of the sub-indicators used. the legatum prosperity index includes a wider range of sub-indicators compared to the hdi, reflecting a broader spectrum of information and various aspects. in contrast, the hdi does not adequately cover many dimensions of human development, such as income distribution equity, poverty, gender equality, housing, access to public services, human rights, and personal safety. based on this, the researchers suggested the need for further research on the sensitivity of sub-indicators to improve the measures used in developing these critical areas and supporting the overall development of countries. otoiu, titan, & dumitrescu (2014) conducted a cluster analysis to explore the variables that determine well-being and human progress in three well-known composite indices and found that the happy planet index was the least reliable compared to the hdi and the legatum prosperity index. they concluded that a robust classification of countries based on multidimensional well-being can be achieved using a small number of variables, reducing the risk of including unreliable or unavailable variables. hsieh (2013) studied the effect of non-inclusive domains on assessing the importance of 80 jalal hafeth ahmad abu-alrop weighting in quality of life measures. the results indicated that common methods for assessing weighting, such as correlation and moderate regression analysis, can lead to misleading results if quality of life measures are not constructed using a formative indicator approach. maričić (2019) pointed out three notable shortcomings in current composite indicators: the uneven inclusion of objective indicators, high instability, and potential bias in the weighting process. for this reason, she proposed the development of a new composite index known as the "european index of life satisfaction" (eils), which relies on a two-pronged approach: applying principal component analysis (pca) to determine the required number of dimensions, and the ess-cidi approach to propose unequal weights, focusing on satisfaction with time use as the most important indicator. similarly, wang, hopke, hancewicz, & zhang (2003) emphasized the necessity of using multiple weighting methods when determining weights in composite indicators. 3. methodology and data the aim of this study is to evaluate the key indicators used to measure quality of life and to identify the determining factors and their relative weights. to achieve this, confirmatory factor analysis (cfa) is employed to model the quality of life in european countries as a latent variable, which minimizes bias, inconsistency, and arbitrary weights of explanatory factors. this results in the development of a standardized model that optimally measures and describes quality of life. exploratory factor analysis (efa) is employed to reduce a set of 14 observed variables (x1, x2…, x14 ) into fewer unobserved latent factors (f1, f2, f3, f4) that explain the variance across the variables. the relationship can be expressed as: xi = λi1 f1 + λi2 f 2 + λi3 f3 + λi4 f4 + 𝟄𝟄i [1] where λij are the factor loadings representing the contribution of each factor to the observed variables, and 𝟄𝟄i is the error term for each variable xi. the goal of efa is to maximize the variance explained by the factors and to reduce the dimensionality of the data. each factor is formed independently using principal component analysis (pca) with varimax rotation to enhance interpretability. the resulting factor scores serve as input for confirmatory factor analysis (cfa), ensuring a structured and statistically validated model for measuring qol. confirmatory factor analysis (cfa) is used to confirm the factor structure derived from efa. in cfa, the qol is modeled as a latent variable fqol represented by the factors f1, f2, f3, f4. the model can be expressed as fqol= β1f1 + β2f2 + β3f3 + β4f4 + δ [2] where β1, β2, β3, and β4 are the factor loadings and δ is the error term for the latent qol variable. cfa ensures that the model fits the data well by estimating the relationships between the latent variables and observed indicators. this approach minimizes bias and inconsistency by allowing the model to estimate the optimal factor loadings rather than assigning arbitrary weights. the model is comparative evaluation of quality of life indicators estimated using maximum likelihood estimation (mle) to ensure statistical robustness. the final outcome is a standardized model for measuring and describing qol, along with the standardized relative weights of its components, forming the basis for constructing the quality of life index (qol). subsequently, the study employs multiple regression analysis using the ordinary least squares (ols) method to assess the impact of the four factors that constitute the standard model for measuring quality of life on commonly used quality of life indicators, aiming to determine their relative weights in each index. the study then uses the best model as a standardized benchmark for comparison with the main indicators used to measure quality of life in order to evaluate them. the study evaluates five indicators used to measure quality of life: income, gross domestic product (gdp), life expectancy at birth, human development index (hdi), and the legatum prosperity index. let yj represent one of the qol indicators (such as income, gdp, life expectancy, hdi, and the legatum prosperity index), and the regression equation can be written as yj = αj +β1jf1 + β2jf2 + β3jf3 + β4jf4 + 𝟄𝟄j [3] where αj is the intercept, βij are the regression coefficients, and 𝟄𝟄j is the error term. this model estimates the relative weights of the factors in influencing each qol indicator. the sample panel data consists of 181 cases, representing annual observations for 32 european countries over the period from 2012 to 2017. the selection of countries was based on data availability, resulting in a varying number of countries per year: 30 in 2012, 29 in 2013, 31 in 2014, 32 in 2015, 30 in 2016, and 29 in 2017. the dataset was compiled from four sources: our world in data1, world data2, the legatum institute3, and eurostat4. the countries included in the study are: austria, belgium, bulgaria, croatia, cyprus, czechia, denmark, estonia, finland, france, germany, greece, hungary, ireland, italy, latvia, lithuania, luxembourg, malta, netherlands, north macedonia, norway, poland, portugal, romania, serbia, slovakia, slovenia, spain, sweden, switzerland, and turkey. for data analysis, three software packages were used: ibm spss statistics 26.0, amos 26.0, and excel 2019. the data underpinning the analysis reported in this paper are deposited at harvard dataverse at https://doi.org/10.7910/dvn/mttfjr . (figure 1) illustrates the research process flowchart, outlining the key steps and analytical approach used in this study. 1 https://ourworldindata.org/ 2 https://www.worlddata.info/quality-of-life.php#tab 3 https://www.prosperity.com 4 https://ec.europa.eu/eurostat/en/ 82 jalal hafeth ahmad abu-alrop figure 1. research process flowchart. source: own design. 4. empirical results and discussion 4.1. exploratory factor analysis (efa) to establish factors that determine quality of life in europe exploratory factor analysis (efa) was conducted on 14 variables to derive 4 main factors determining quality of life. each factor was analyzed separately. (table 1) displays the initial variables and the factors resulting from efa, alongside quality of life indicators and their abbreviations. (table 2) presents the efa results. table 2 summarizes the key findings from the exploratory factor analysis (efa). the results highlight the importance of the kaiser-meyer-olkin (kmo) measure and bartlett's test in assessing the suitability of data for factor structure detection. comparative evaluation of quality of life indicators table 1. initial variables and factors from exploratory factor analysis (efa) with quality of life indicators and abbreviations. variables type initial variables and abbreviations factors and abbreviations independent variables equal rights governance quality (govq) elect free fair individual liberties and equality before the law freedom of expression judicial restrictions on executive power accountability transparency gross domestic product at market prices, euro per capita (gdp) economy quality (ecoq) median income, euro (inc) employment and activity, percentage of total population population growth rate research and development expenditure, percentage of gdp education quality (eduq) mortality rate health quality (heaq) depression rate smoking mortality rate dependent variables quality of life indicators the legatum prosperity index (leg) human development index (hdi) life expectancy at birth (lex) gross domestic product at market prices, euro per capita (gdp) median income, euro (inc) source: own design. the kmo measure assesses the sampling adequacy, indicating the extent to which variables share common variance that could be caused by underlying factors. a kmo value close to 1.0 suggests that factor analysis is likely appropriate for the data, whereas values below 0.5 indicate that factor analysis may not yield meaningful results. according to kaiser (1974), a minimum kmo value of 0.5 is recommended, with values between 0.7-0.8 considered acceptable, and values above 0.9 being excellent. bartlett's test of sphericity evaluates whether the correlation matrix of variables is an identity matrix, implying that variables are unrelated and unsuitable for factor analysis. a significance level (p-value) less than 0.05 typically indicates that factor analysis is suitable for the data. based on the results presented in table 2, both the kmo measure and bartlett's test indicate that the data are suitable and appropriate for detecting underlying factor structures. this suggests that the variables included in the analysis share sufficient common variance and are not so highly correlated that they would render factor analysis inappropriate. the "total variance explained" column shows the quality score for each component, called the eigenvalue. only components with high eigenvalues are likely to represent real underlying 84 jalal hafeth ahmad abu-alrop factors. the general rule is to choose components whose eigenvalues are at least 1. applying this rule to the table, our variables seem to measure factors. the percentage of variance indicates the amount of variance explained by each factor. the results indicate that all loadings are relatively high, ranging from 63% to 79%. table 2. summary of exploratory factor analysis (efa) results. factors and abbreviations number of variables kaiser-meyerolkin measure of sampling adequacy (kmo) bartlett's test (sig.) total variance explained eigenvalues % of variance governance quality (govq) 6 .89 <.001 4.79 79.82 economy quality (ecoq) 4 .71 <.001 2.85 71.21 education quality (eduq) 1 health quality (heaq) 3 .60 <.001 1.90 63.41 source: own calculations using ibm spss statistics 26.0. 4.2. confirmatory factor analysis for forming the standard model to measure quality of life in europe to form the best possible model for measuring and describing quality of life in europe, first-order confirmatory factor analysis (cfa) was conducted to identify the measurements that meet the requirements for the latent factor loading (quality of life). if the value is more than 0.5, the measurement is considered valid. it should be noted that the factor scores extracted from the exploratory factor analysis (efa) in section 4.1 were used as indicators to represent the variables related to governance quality, economy quality, education quality, and health quality. figure 2 presents the cfa model used, showing the four variables (governance quality, economy quality, education quality, health quality) that form the best possible model for measuring and describing quality of life in europe. figure 2. the best possible first-order confirmatory factor analysis (cfa) model for quality of life in europe. comparative evaluation of quality of life indicators notes: model fit measures: cmin= 0.001, df= 1, p= 0.974, cmin/df= 0.001, cfi= 1.000, srmr= 0.000, rmsea= 0.000, pclose= 0.979, regression weights: governance quality: estimate = 1. economy quality: estimate = 0.731, s.e.= 0.081, c.r.= 9.022, p= ***. health quality: estimate = 0.975, s.e.= 0.082, c.r.= 11.854, p= ***. education quality: estimate = 0.712, s.e.= 0.082, c.r.= 8.722, p= *** source: own calculations using amos 26.0. in such models, reliability and validity must be established before proving the model’s suitability. to establish the reliability and validity of this model, three types of validity are referenced: convergent validity, internal consistency, and composite reliability. table 3 displays the validity results. table 3. reliability and validity assessment of the quality of life model. measure composite reliability (cr) average variance extraction (ave) maximal reliability (maxr(h)) cronbach's alpha (ca) estimate 0.844 0.581 0.885 0.857 source: own calculations using amos 26.0. reliability was assessed using composite reliability (cr) and maximal reliability (maxr(h)). discriminant validity could not be tested in a single-factor model. composite reliability indicates each construct’s significance in the model. for quality of life, cr was 0.84, exceeding the threshold of 0.7, indicating good reliability. according to gaskin & lim (2016), maxr(h) values above 0.7 indicate an excellent model; here, it was 0.88. convergent validity was confirmed with average variance extracted (ave). alarcón & sánchez (2015) suggest ave should be ≥ 0.5; our model achieved 0.58 for quality of life. internal consistency, evaluated via cronbach's alpha, yielded a value of 0.85, surpassing the recommended threshold of 0.7 (alarcón & sánchez, 2015), indicating good internal consistency. thus, values in table 4 (cr, ave, and maxr(h)) affirm the model’s reliability and validity. model fit was assessed using criteria established by hu & bentler (1999). the criteria for an excellent fit are comparative fit index (cfi) > 0.95, standardized root mean square residual (srmr) < 0.08, and root mean square error of approximation (rmsea) < 0.06. our model met these criteria, indicating an excellent fit. (table 4) presents the detailed results of the model fit assessment. table 4. quality of life model fit measures. measure cmin df cmin/df cfi srmr rmsea pclose estimate 0.001 1 0.001 1.000 0.000 0.000 0.979 source: own calculations using amos 26.0. with the model's suitability, reliability, and validity established, it can effectively measure quality of life. next, the study will determine the relative impact of each variable on the quality of life index based on the model's findings. figure 3 illustrates the relative weights for the standard quality of life measurement model: governance quality (23%), health quality (22%), economic quality (16%), 86 jalal hafeth ahmad abu-alrop education quality (16%), and other factors (23%). notably, this model emphasizes governance and health quality more than economic and education quality, distinguishing it from other indicators. figure 3. relative weights of factors in the standard quality of life measurement model derived from multiple regression analysis. source: own calculations using amos 26.0 and excel 2019. 4.3. multiple regression analysis to measure the effect of factors on key quality of life indicators the study also uses multiple regression analysis using the ordinary least squares (ols) method to determine the relative weight of factors adopted by the standard model for measuring quality of life in forming key indicators used to measure quality of life. the regression equation that expresses the linear relationships between a single dependent variable and the independent variables is outlined in equation 4: y = α + β1x1 + β2x2 + β3x3 + β4x4 + 𝟄𝟄 [4] in equation 1, y is the predicted value of the dependent variable. the values of the independent variables are denoted as x1, x2, x3, x4. α is the constant, β1, β2, β3, β4 are the regression coefficients, and 𝟄𝟄 is a random factor. the constants' values are assigned based on the principle of least squares. leg, hdi, lex, gdp, and inc are the indices that measure quality of life influenced by 4 factors (inputs): govq, ecoq, eduq, and heaq. it should be noted that the factor scores extracted from the exploratory factor analysis (efa) in section 4.1 were used as indicators to represent the variables related to governance quality, economy quality, education quality, and health quality. by applying governance quality 23% economy quality 16% health quality 22% education quality 16% others 23% model fit measures: cmin= 0.001, df= 1, p= 0.974, cmin/df= 0.001, cfi= 1.000, srmr= 0.000, rmsea= 0.000, pclose= 0.979, regression weights: governance quality: estimate = 1. economy quality: estimate = 0.731, s.e.= 0.081, c.r.= 9.022, p= ***. health quality: estimate = 0.975, s.e.= 0.082, c.r.= 11.854, p= ***. education quality: estimate = 0.712, s.e.= 0.082, c.r.= 8.722, p= *** comparative evaluation of quality of life indicators the study variables to the above equation, 4 equations can be formed as follows. leg = α + β1 govq + β2 ecoq + β3 eduq + β4 heaq + 𝟄𝟄 [5] hdi = α + β1 govq + β2 ecoq + β3 eduq + β4 heaq + 𝟄𝟄 [6] lex = α + β1 govq + β2 ecoq + β3 eduq + β4 heaq + 𝟄𝟄 [7] gdp = α + β1 govq + β2 ecoq + β3 eduq + β4 heaq + 𝟄𝟄 [8] inc = α + β1 govq + β2 ecoq + β3 eduq + β4 heaq + 𝟄𝟄 [9] figures 4-8 below display the results of the five multiple regression models. according to the results in figure 4, the factors affecting inc include economic quality (82%), education quality (8%), health quality (8%), and other factors (-2%). this index notably excludes governance quality and heavily emphasizes economic factors. however, it offers broader coverage compared to the gross domestic product (gdp) index, incorporating more diverse factors, as detailed later in the study. figure 4. relative weights of factors in the income index (inc) derived from multiple regression analysis. source: own calculations using ibm spss statistics 26.0 and excel 2019. the results in figure 5 indicate that factors influencing the gross domestic product (gdp) index are primarily economic quality (85%), with minor impacts from education (-8%) and other factors (7%). this index lacks consideration for governance and health factors and is negatively influenced by education, showing a strong bias towards economic quality. economy quality 77% health quality 8% education quality 12% others -3% adjusted r square = 0.927, f= 765.883, sig.= <.001. (constant): b= -0.141, t= -1.016, sig.= 0.311. economy quality: b= 0.862, t= 31.705, sig.=0.000. health quality: b= 0.077, t= 3.041, sig.= 0.003. education quality: b= 0.085, t= 3.176, sig.= 0.002 88 jalal hafeth ahmad abu-alrop figure 5. relative weights of factors in the gross domestic product (gdp) index derived from multiple regression analysis. source: own calculations using ibm spss statistics 26.0 and excel 2019. the results in figure 6 show that the relative weights of the factors affecting lex are 48% for health quality, 28% for economic quality, -18% for governance quality, and 6% for other factors. the main advantage of this index compared to the standard model proposed by the study is that lex gives a high weight to health quality. its downsides are that it is not affected by education quality, is negatively affected by governance quality (the most important factor according to the standard model), and is heavily biased towards health quality. nevertheless, it is better than inc and gdp indices as it includes more factors. economy quality 85% education quality -8% others 7% adjusted r square = 0.914, f= 955.674, sig.= <.001. (constant): b= 0.463, t= 3.262, sig.= 0.001. economy quality: b= 1.010, t= 36.162, sig.=0.000. education quality: b= -0.090, t= -3.222, sig.= 0.002 comparative evaluation of quality of life indicators figure 6. relative weights of factors in the life expectancy index (lex) derived from multiple regression analysis. source: own calculations using ibm spss statistics 26.0 and excel 2019. the results in figure 7 show that the relative weights of the factors influencing hdi are 35% for economic quality, 21% for governance quality, 19% for education quality, 14% for health quality, and 11% for other factors. hdi is influenced by all factors. however, the downside of this index is that it assigns higher relative weights to economic quality and education quality than to health quality and governance quality, and it is also significantly biased towards economic quality. despite this, it is considered better than inc, gdp, and lex indicators because it includes all factors and is more balanced. it needs to increase the weights for governance quality and health quality. governance quality -18% economy quality 28% health quality 48% others 6% adjusted r square = 0.719. f= 154.785 , sig.= <.001. (constant): b= 0.581 , t= 2.173, sig.= 0.031. governance quality: b= -0.276 , t= -4.384, sig.=0.000. economy quality: b= 0.436, t= 8.836, sig.=0.000. health quality: b= 0.740, t= 11.898, sig.=0.000 90 jalal hafeth ahmad abu-alrop figure 7. relative weights of factors in the human development index (hdi) derived from multiple regression analysis. source: own calculations using ibm spss statistics 26.0 and excel 2019. the results in figure 8 show that the relative weights of the factors influencing leg are 35% for economic quality, 34% for governance quality, 22% for education quality, and 9% for other factors. leg assigns a high relative weight to governance quality. however, the downside of this index is that it is not influenced by the health quality factor, which is the second most important factor in the standard model, and it also assigns the highest relative weight to economic quality. this is noteworthy because leg includes in its composition a domain called "empowering people," which consists of four pillars, one of which is health. health is represented by six variables: longevity, physical health, mental health, care systems, behavioral risk factors, and preventive interventions. despite all these health variables, leg was not influenced by the health quality factor present in the standard model proposed by the study. nevertheless, leg remains better than inc and gdp indicators because it covers more factors, and it is also better than lex, which is negatively influenced by governance quality. leg needs to reconsider the variables representing health and reduce the weights for governance quality and economic quality. governance quality 27% economy quality 28% education quality 19% health quality 15% others -11% adjusted r square = 0.90. f= 435.705, sig.= <.001. (constant): b= --0.774, t= -4.846, sig.=0.000. governance quality: b= 0.271, t= 7.358, sig.=0.000. economy quality: b= 0.438, t= 13.968, sig.=0.000. health quality: b= 0.181, t= 5.001, sig.=0.000. education quality: b= 0.243, t= 7.847, sig.=0.000. comparative evaluation of quality of life indicators figure 8. relative weights of factors in the legatum prosperity index (leg) derived from multiple regression analysis. source: own calculations using ibm spss statistics 26.0 and excel 2019. 5. conclusions and recommendations this study aimed to evaluate the main indicators used to measure quality of life and determine their determining factors and relative weights. using data consisting of 181 cases, representing annual observations for 32 european countries over the period from 2012 to 2017, the study employed exploratory factor analysis (efa) to identify four key factors affecting quality of life in europe: economic quality, health quality, education quality, and governance quality. these key factors were validated using confirmatory factor analysis (cfa) to develop a standard model for measuring quality of life. the study utilized these foundational factors within the standard model to evaluate five indicators used for measuring quality of life: income, gross domestic product (gdp), life expectancy at birth, the human development index (hdi), and the legatum prosperity index (leg). confirmatory factor analysis (cfa) validated these factors as effective components of a standard quality of life model. analysis of quality of life indicators revealed the following biases and influences: • gross domestic product (gdp) indicator: biased towards economic quality, unaffected by governance and health quality. • income indicator (inc): not influenced by governance quality, but more comprehensive than gdp as it considers multiple factors. • life expectancy at birth indicator (lex): biased towards health quality, negatively impacted by governance quality, and unaffected by educational quality. governance quality 31% economy quality 28% education quality 19% health quality 11% others -11% adjusted r square = 0.89. f= 505.035, sig.= <.001. (constant): b= -0.642, t= -3.835, sig.=0.000. governance quality: b= 0.416, t= 13.394, sig.=0.000. economy quality: b= 0.431, t= 13.032, sig.=0.000. education quality: b=0.264, t= 8.088, sig.=0.000. 92 jalal hafeth ahmad abu-alrop • legatum prosperity index (leg): emphasizes governance quality but disregards health quality, with significant weight on economic quality. • human development index (hdi): influenced by all factors but biased towards economic and educational quality over health and governance quality. these findings underscore the different biases and strengths of each indicator in measuring quality of life, highlighting the need for balanced consideration of all relevant factors. the study emphasized the importance of considering multiple factors for a comprehensive representation of quality of life, highlighting the effectiveness of efa and cfa in determining relative weights and constructing integrated measurement models. the human development index (hdi) emerged as the most balanced and comprehensive indicator compared to others, despite its simplicity. the administrators of the legatum prosperity index (leg) are advised to reassess factor weights, particularly by enhancing the representation of health variables and reducing emphasis on economic and governance factors. for the hdi, improvements should include increased weighting for governance and health quality to enhance accuracy in measuring quality of life. researchers and policymakers are encouraged to utilize efa and cfa to develop more precise indicators. further research should expand to include diverse countries and regions to validate the proposed model and understand quality of life factors in different contexts. finally, the study proposes a definition of quality of life as the extent to which an individual is able to meet their fundamental needs, including physiological needs, safety needs, social needs, esteem needs, and, finally, self-actualization needs. a society with a high quality of life is one that provides individuals with the appropriate environment to achieve self-actualization, where the primary motivation for a healthy person is the development of their full potential and capabilities. from this perspective, any deficiency in meeting these needs can be considered a form of illness, much like deficiencies in vitamins or minerals, which are recognized as health issues. references abu-alrop, j. h. a. (2025). european european panel data: quality of life, governance, economy, education, and health (2012-2017) (version v1) [data file]. harvard dataverse. https://doi.org/10.7910/dvn/mttfjr alarcón, d., & sánchez, j. (2015). assessing convergent and discriminant validity in the adhd-r iv rating scale: user-written commands for average variance extracted (ave), composite reliability (cr), and heterotrait-monotrait ratio of correlations (htmt). spanish stata users group meeting, 1–39. madrid. https://www.stata.com/meeting/spain15/abstracts/materials/spain15_alarcon.pdf baker, d., leon, j., smith greenaway, e., collins, j., & movit, m. (2011). the education effect on population health: a reassessment. population and development review, 37(2), 307–332. https://doi.org/10.1111/j.1728-4457.2011.00412.x beslerová, s., & dzuričková, j. (2014). quality of life measurements in eu countries. procedia economics and finance, 12, 37–47. https://doi.org/10.1016/s2212-5671(14)00318-9 bigelow, d., mcfarland, b., & olson, m. (1991). quality of life of community mental health program clients: validating a measure. community mental health journal, 27(1), 43–55. https://doi.org/10.1007/bf00752714 https://www.stata.com/meeting/spain15/abstracts/materials/spain15_alarcon.pdf https://doi.org/10.1111/j.1728-4457.2011.00412.x https://doi.org/10.1016/s2212-5671(14)00318-9 https://doi.org/10.1007/bf00752714 comparative evaluation of quality of life indicators brown, j., bowling, a., & flynn, t. (2004). models of quality of life: a taxonomy, overview and systematic review of the literature. european forum on population ageing research. campbell, a. (1981). sense of well-being in america: recent patterns and trends. new york: mcgraw-hill. campbell, a., converse, p. e., & rodgers, w. l. (1976). the quality of american life. new york: russell sage foundation. christensen, k., & levinson, d. (eds.). (2003). encyclopedia of community: from the village to the virtual world. london: sage publications. https://doi.org/10.4135/9781412952583 clarke, p., marshall, v., ryff, c., & rosenthal, c. (2000). well-being in canadian seniors: findings from the canadian study of health and aging. canadian journal on aging, 19(2), 139–159. https://doi.org/10.1017/s0714980800013982 costanza, r., fisher, b., ali, s., beer, c., bond, l., boumans, r., ... snapp, r. (2007). quality of life: an approach integrating opportunities, human needs, and subjective well-being. ecological economics, 61(2– 3), 267–276. https://doi.org/10.1016/j.ecolecon.2006.02.023 cutler, d., & lleras-muney, a. (2010). understanding differences in health behaviors by education. journal of health economics, 29(1), 1–28. https://doi.org/10.1016/j.jhealeco.2009.10.003 deseran, a. f. (1978). community satisfaction as definition of the situation: some conceptual issues. rural sociology, 43(2), 235–249. diener, e., & biswas-diener, r. (2002). will money increase subjective well-being? a literature review and guide to needed research. social indicators research, 57(2), 119–169. http://www.jstor.org/stable/27526987 diener, e., & seligman, m. (2004). beyond money: toward an economy of well-being. psychological science in the public interest, 5(1), 1–31. https://doi.org/10.1111/j.0963-7214.2004.00501001.x diener, e., oishi, s., & tay, l. (2018). advances in subjective well-being research. nature human behaviour, 2(4), 253–260. https://doi.org/10.1038/s41562-018-0307-6 dunn, e., aknin, l., & norton, m. (2008). spending money on others promotes happiness. science, 319(5870), 1687–1688. https://doi.org/10.1126/science.1150952 easterlin, r. a. (1974). does economic growth improve the human lot? some empirical evidence. en p. a. david & m. w. reder (eds.), nations and households in economic growth: essays in honor of moses abramovitz (pp. 89–125). new york, ny: academic press. https://doi.org/10.1016/b978-0-12205050-3.50008-7 elizabeth, s. (2007). the human development index: a history (working paper no. 127). political economy research institute, university of massachusetts at amherst. eurostat. (2015). quality of life: facts and views (2015 edition). luxembourg: publications office of the european union. https://ec.europa.eu/eurostat/documents/3217494/6856423/ks-05-14-073en-n.pdf/742aee45-4085-4dac-9e2e-9ed7e9501f23?t=1433141890000 eurostat. (2024). quality of life indicators – measuring quality of life. https://ec.europa.eu/eurostat/statistics-explained/index.php?title=quality_of_life_indicators__measuring_quality_of_life gabriel, z., & bowling, a. (2004). quality of life from the perspectives of older people. ageing and society, 24(5), 675–691. https://doi.org/10.1017/s0144686x03001582 gaskin, j., & lim, j. (2016). master validity tool [amos plugin]. en gaskination’s statwiki. https://statwiki.gaskination.com graham, b., & kenealy, p. (2004). quality of life perceptions and social comparisons in healthy old age. ageing and society, 24(5), 755–769. https://doi.org/10.1017/s0144686x04002399 green, c. d. (2000). a theory of human motivation by a. h. maslow. en classics in the history of psychology. http://psychclassics.yorku.ca/maslow/motivation.htm grundy, e., & bowling, a. (1999). enhancing the quality of extended life years: identification of the oldest old with a very good and very poor quality of life. aging & mental health, 3(3), 199–212. https://doi.org/10.1080/13607869956154 gulick, j., bowerman, c., & back, k. (1962). newcomer enculturation in the city: attitudes and participation. en f. s. chapin jr. & s. weiss (eds.), urban growth dynamics in a regional cluster of cities (pp. 315– 358). new york, ny: john wiley & sons. hajdúová, z., andrejovský, p., gajdoš, j., & andrejkovič, m. (2011). the quality of life and the environment. journal of innovations and applied statistics, 1, 37–42. https://jias.euke.sk/wpcontent/uploads/2012/06/1-2011.pdf hollingshead, a., & rogler, l. (1963). attitudes toward slums and public housing in puerto rico. en l. j. duhl et al. (eds.), the urban condition. new york: basic books. hsieh, c.-m. (2013). issues in evaluating importance weighting in quality of life measures. social indicators research, 110(2), 681–693. https://doi.org/10.1007/s11205-011-9951-1 hu, l.-t., & bentler, p. (1999). cutoff criteria for fit indexes in covariance structure analysis: conventional criteria versus new alternatives. structural equation modeling: a multidisciplinary journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118 https://doi.org/10.4135/9781412952583 https://doi.org/10.1017/s0714980800013982 https://doi.org/10.1016/j.ecolecon.2006.02.023 https://doi.org/10.1016/j.jhealeco.2009.10.003 http://www.jstor.org/stable/27526987 https://doi.org/10.1111/j.0963-7214.2004.00501001.x https://doi.org/10.1038/s41562-018-0307-6 https://doi.org/10.1126/science.1150952 https://doi.org/10.1016/b978-0-12-205050-3.50008-7 https://doi.org/10.1016/b978-0-12-205050-3.50008-7 https://ec.europa.eu/eurostat/documents/3217494/6856423/ks-05-14-073-en-n.pdf/742aee45-4085-4dac-9e2e-9ed7e9501f23?t=1433141890000 https://ec.europa.eu/eurostat/documents/3217494/6856423/ks-05-14-073-en-n.pdf/742aee45-4085-4dac-9e2e-9ed7e9501f23?t=1433141890000 https://ec.europa.eu/eurostat/statistics-explained/index.php?title=quality_of_life_indicators_-_measuring_quality_of_life https://ec.europa.eu/eurostat/statistics-explained/index.php?title=quality_of_life_indicators_-_measuring_quality_of_life https://doi.org/10.1017/s0144686x03001582 https://doi.org/10.1017/s0144686x04002399 http://psychclassics.yorku.ca/maslow/motivation.htm https://doi.org/10.1080/13607869956154 https://jias.euke.sk/wp-content/uploads/2012/06/1-2011.pdf https://jias.euke.sk/wp-content/uploads/2012/06/1-2011.pdf https://doi.org/10.1007/s11205-011-9951-1 https://doi.org/10.1080/10705519909540118 94 jalal hafeth ahmad abu-alrop hughes, b. (1990). quality of life. en s. m. peace (ed.), researching social gerontology: concepts, methods and issues (pp. 46–58). london: sage. kahneman, d., & deaton, a. (2010). high income improves evaluation of life but not emotional well-being. proceedings of the national academy of sciences, 107(38), 16489–16493. https://doi.org/10.1073/pnas.1011492107 kaiser, h. (1974). an index of factorial simplicity. psychometrika, 39(1), 31–36. https://doi.org/10.1007/bf02291575 knipscheer, c., de jong-gierveld, j., van tilburg, t., & dykstra, p. (1995). living arrangements and social networks of older adults. amsterdam: vu university press. lawton, p. m. (1991). a multidimensional view of quality of life in frail elders. en j. e. birren, j. lubben, j. rowe, & d. deutchman (eds.), the concept and measurement of quality of life in the frail elderly (pp. 3–27). new york, ny: academic press. https://doi.org/10.1016/b978-0-12-1012755.50005-3 malkina-pykh, i., & pykh, y. (2008). quality-of-life indicators at different scales: theoretical background. ecological indicators, 8(6), 854–862. https://doi.org/10.1016/j.ecolind.2007.01.008 marans, w. r., & rodgers, w. (1975). toward an understanding of community satisfaction. en a. hawley & v. rock (eds.), metropolitan america in contemporary perspective (pp. 299–352). new york, ny: john wiley & sons. maričić, m. (2019). assessing the quality of life in the european union: the european index of life satisfaction (eils). statistical journal of the iaos, 35(2), 261–267. https://doi.org/10.3233/sji180481 marmot, m., friel, s., bell, r., houweling, t. a., & taylor, s. (2008). closing the gap in a generation: health equity through action on the social determinants of health. the lancet, 372(9650), 1661–1669. https://doi.org/10.1016/s0140-6736(08)61690-6 maslow, a. (1954). motivation and personality. new york: harper & row publishers. matarrita-cascante, d. (2010). changing communities, community satisfaction, and quality of life: a view of multiple perceived indicators. social indicators research, 98(1), 105–127. https://doi.org/10.1007/s11205-009-9520-z morris, e. w., & winter, m. (1978). housing, family, and society. new york: john wiley and sons. murray, c., aravkin, a., zheng, p., abbafati, c., abbas, k., abbasi-kangevari, m., ... acharya, d. (2020). global burden of 87 risk factors in 204 countries and territories, 1990–2019: a systematic analysis for the global burden of disease study 2019. the lancet, 396(10258), 1223–1249. https://doi.org/10.1016/s0140-6736(20)30752-2 noll, h.-h. (2002). towards a european system of social indicators: theoretical framework and system architecture. social indicators research, 58, 47–87. https://doi.org/10.1023/a:1015775631413 otoiu, a., titan, e., & dumitrescu, r. (2014). are the variables used in building composite indicators of well-being relevant? validating composite indexes of well-being. ecological indicators, 46, 575– 585. https://doi.org/10.1016/j.ecolind.2014.07.019 oxford university. (2011). oxford dictionary. oxford: oxford university press. potter, j., & cantarero, r. (2014). community satisfaction. en a. c. michalos (ed.), encyclopedia of quality of life and well-being research (pp. 1094–1099). dordrecht: springer. https://doi.org/10.1007/978-94-007-0753-5_480 raleigh, v. s. (2019). trends in life expectancy in eu and other oecd countries: why are improvements slowing? oecd health working papers, no. 108. paris: oecd publishing. https://doi.org/10.1787/223159ab-en ryff, c., & singer, b. (2008). know thyself and become what you are: a eudaimonic approach to psychological well-being. journal of happiness studies, 9(1), 13–39. https://doi.org/10.1007/s10902-006-9019-0 scharf, t., phillipson, c., & smith, a. (2004). poverty and social exclusion: growing older in deprived urban neighbourhoods. en c. walker (ed.), growing older: quality of life in old age (pp. 81–106). maidenhead: open university press. stiglitz, j., sen, a., & fitoussi, j. (2009). report by the commission on the measurement of economic performance and social progress. https://ec.europa.eu/eurostat/documents/8131721/8131772/stiglitz-sen-fitoussicommission-report.pdf tang, t. l.-p. (2007). income and quality of life: does the love of money make a difference? journal of business ethics, 72(4), 375–393. https://doi.org/10.1007/s10551-006-9176-4 tesch-römer, c., & motel-klingebiel, a. (2001). quality of life in the context of intergenerational solidarity. en s. o. daatland & k. herlofson (eds.), ageing, intergenerational relations, care systems and quality of life (pp. 63–73). oslo: nova. theodori, g. l. (2001). examining the effects of community satisfaction and attachment on individual wellbeing. rural sociology, 66(4), 618–628. https://doi.org/10.1111/j.1549-0831.2001.tb00087.x https://doi.org/10.1073/pnas.1011492107 https://doi.org/10.1007/bf02291575 https://doi.org/10.1016/b978-0-12-101275-5.50005-3 https://doi.org/10.1016/b978-0-12-101275-5.50005-3 https://doi.org/10.1016/j.ecolind.2007.01.008 https://doi.org/10.3233/sji-180481 https://doi.org/10.3233/sji-180481 https://doi.org/10.1016/s0140-6736(08)61690-6 https://doi.org/10.1007/s11205-009-9520-z https://doi.org/10.1016/s0140-6736(20)30752-2 https://doi.org/10.1023/a:1015775631413 https://doi.org/10.1016/j.ecolind.2014.07.019 https://doi.org/10.1007/978-94-007-0753-5_480 https://doi.org/10.1787/223159ab-en https://doi.org/10.1007/s10902-006-9019-0 https://ec.europa.eu/eurostat/documents/8131721/8131772/stiglitz-sen-fitoussi-commission-report.pdf https://ec.europa.eu/eurostat/documents/8131721/8131772/stiglitz-sen-fitoussi-commission-report.pdf https://doi.org/10.1007/s10551-006-9176-4 https://doi.org/10.1111/j.1549-0831.2001.tb00087.x comparative evaluation of quality of life indicators united nations development programme. (2010). human development report 2010: the real wealth of nations – pathways to human development. new york: undp. https://hdr.undp.org/content/human-development-report-2010 walker, a. (2005). a european perspective on quality of life in old age. european journal of ageing, 2(1), 2–12. https://doi.org/10.1007/s10433-005-0500-0 walker, a., & van der maesen, l. j. g. (2004). social quality and quality of life. en w. glatzer, s. von below, & m. stoffregen (eds.), challenges for quality of life in the contemporary world (social indicators research series, vol. 24). dordrecht: springer. https://doi.org/10.1007/978-1-4020-2903-5_2 wang, j.-h., hopke, p., hancewicz, t., & zhang, s. (2003). application of modified alternating least squares regression to spectroscopic image analysis. analytica chimica acta, 476(1), 93–109. https://doi.org/10.1016/s0003-2670(02)01369-7 world bank group. (2018). world bank annual report 2018. washington, dc: world bank publications. https://documents1.worldbank.org/curated/en/630671538158537244/pdf/the-world-bankannual-report-2018.pdf world health organization. (2012). the world health organization quality of life (whoqol). geneva: world health organization. https://www.who.int/publications/i/item/who-his-hsirev.2012.03 world health organization. (2019). primary health care on the road to universal health coverage: 2019 monitoring report. world health organization. https://www.who.int/publications/i/item/9789240029040 https://hdr.undp.org/content/human-development-report-2010 https://doi.org/10.1007/s10433-005-0500-0 https://doi.org/10.1007/978-1-4020-2903-5_2 https://doi.org/10.1016/s0003-2670(02)01369-7 https://documents1.worldbank.org/curated/en/630671538158537244/pdf/the-world-bank-annual-report-2018.pdf https://documents1.worldbank.org/curated/en/630671538158537244/pdf/the-world-bank-annual-report-2018.pdf https://www.who.int/publications/i/item/who-his-hsi-rev.2012.03 https://www.who.int/publications/i/item/who-his-hsi-rev.2012.03 https://www.who.int/publications/i/item/9789240029040 1. introduction 2. literature review 3. methodology and data 4. empirical results and discussion 5. conclusions and recommendations references assessing the optimality of euro adoption in romania european journal of government and economics 10(2), december 2021, 107-127 european journal of government and economics issn: 2254-7088 assessing the optimality of euro adoption in romania through shock correlations adrian bodea a, b *1 a european commission, directorate-general for economic and financial affairs, brussel, belgium. b faculty of business and economics, universidade da coruña, spain. * corresponding author at: a.bodea@udc.es abstract. the present paper is concerned with the prospect of euro adoption in romania. the study starts from the relevant literature of the optimum currency areas and identifies the most widely acknowledged meta property and methodological model for this purpose: the svar blanchard and quah decomposition for identifying the supply and demand shocks. employing the indicated model and the most recent data, we are able extract and analyse the underlying shocks that hit 34 european economic entities in the period 1995-2019, while also taking into account two crucial structural changes for the romanian economy – central bank independence and eu accession. after performing the pairwise correlations between romania and the rest of the economic entities for both the supply and demand disturbances, we map them on a bidimensional graph. we discover that while there is relevant integration and connectedness that ensures relatively high correlations between supply shocks, the politically-motivated monetary and fiscal policy disturbances that created ample and hectic demand side movements, are a factor of great concern for the prospect of single currency adoption in this eastern european country. the findings support the view that there is room for the conduct of macro policies to become more supportive to the process of euro adoption and that the respect of convergence criteria would help in this respect. to our knowledge, this is the first study performing pairwise shock correlations between romania and many other european economic entities, while also isolating the effect of post 2005 structural changes. keywords. optimum currency areas; svar; blanchard and quah decomposition; romania; euro adoption; shock correlation jel codes. e31; e32; e37. doi. https://doi.org/10.17979/ejge.2021.10.2.7907 1. introduction the main objective of this study is to provide an evidence-backed answer to whether, at this point in time, romania is prepared to adopt the european single currency. the empirical evaluation of this very complex issue is done using the most widely accepted methodology in the optimum currency areas (oca) literature: the svar blanchard and quah decomposition for the underlying demand and supply shocks (blanchard & quah, 1989). by correlating these shocks that hit the romanian economy and the ones that hit some other european economic entities (both national economies and eu/ea economic aggregates) in the past 25 years (1995-2019), a clear conclusion on the matter at hand can be inferred. 1 disclaimer. the information and views set out in this article are those of the author and do not necessarily reflect the opinion of the european commission. the present study has been conducted before the author joined the european commission. https://doi.org/10.17979/ejge.2021.10.2.7907 adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 108 the relevance of this paper stems from the fact that it is the first study of this kind involving romania taking into account the effect of two irreversible and structural changes that took place after 2005 (central bank independence and eu accession). as such, this study comes to reassess the situation after three decades of economic and political transition, after 14 years of central bank independence, and after 12 years of eu integration efforts. furthermore, the recent advances from neighbor countries on euro adoption, might also push the issue on the agenda of the executive from bucharest. given that the regional political and macroeconomic landscape has changed, a reassessment of the optimality of euro adoption is clearly demanded. the literature on the optimum currency areas (ocas) in europe has developed a great deal since the seminal paper of mundell (1961) and comprises both theoretical and empirical studies which shaped academic and the policy-oriented discussion on the creation and functioning of the single currency area we know today. at the same time and in parallel, the european single currency project advanced passing through various phases (european monetary system, ecu, euro and cash euro, etc.) since the werner report and the end of bretton-woods system in the 70’s. nowadays, the euro area encompasses the economies of 19 european countries (with other micro and small states using the euro as their de facto currency on the basis of agreements on monetary relations or due to the euroization of their economies), and close to 350 million citizens as daily users. on the basis of eu legislation, all eu countries without the euro (except denmark, which opted out) are required to adopt the single currency after having fulfilled the maastricht convergence criteria2. nevertheless, five countries (czech republic, hungary, poland, romania and sweden) preferred not relinquishing their monetary sovereignty at all, while other two (bulgaria and croatia) only recently expressed their intention to join the euro antechamber (i.e. erm ii)3. the most recent eurobarometers show very mixed feelings; while there are some eurooptimist member states, significantly low or decreasing popular support for the euro adoption and economic and monetary union are registered in several of these countries, with governments and central banks hesitant on advancements on the issue. these facts pose a conundrum for both academic and policy-making spheres with respect to the optimal size and the right composition of the euro area. this study, although touching upon all eu member states, focuses on the romanian case in the effort of making an empirical and measurable assessment of euro adoption optimality in this eastern european country. the rest of the paper is structured as follows. in the second section, an in-depth literature review centered on ocas is provided with subsections centered on its theoretical foundations, its properties and the most relevant empirical studies. the subsequent section details the economic and econometric model of choice, while also indicating the data source and the statistical tool. the fourth part consists of results and their interpretation, while the last one is reserved for conclusions and policy recommendations. 2 art. 140 (1) tfeu 3 “commission welcomes bulgaria and croatia's entry into the exchange rate mechanism ii”. european commission. brussels. press release. 12 july 2020. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 109 figure 1. sentiments regarding the emu in selected eu non-euro area countries. source: eurostat. question: “what is your opinion on each of the following statements? please tell me for each statement, whether you are for it or against it. a european economic and monetary union with one single currency, the euro.” dk/nas (“don’t know/no answer”) not included. 2. literature review 2.1. the foundations of the ocas one of the most relevant costs associated with adopting the euro is the relinquishment of the sovereign monetary policy instrument, which has been extensively used in the past, especially during the decade of the 90s by the romania government in concert with a non-independent national bank of romania (bodea & sánchez-santos, 2020). against the background of the hardships imposed by transition, monetary policy helped the executive in achieving both commercial policy (devaluation to buttress exports) and fiscal policy objectives (inflation to reduce public debt denominated in domestic currency and raise seigniorage and other indirect taxes). nevertheless, the full range of costs, as well as benefits of joining a monetary union, is not limited only to achieving commercial and fiscal policy objectives, the issue being at the core of 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 16-03-18 06-06-19 16-03-18 06-06-19 16-03-18 06-06-19 16-03-18 06-06-19 16-03-18 06-06-19 16-03-18 06-06-19 16-03-18 06-06-19 16-03-18 06-06-19 bu lg ar ia cz ec hi a de nm ar k cr oa tia hu ng ar y po la nd ro m an ia sw ed en against for adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 110 the literature of optimum currency areas (ocas). the theory of ocas, pioneered by robert mundell (1961) and mckinnon (1963), established the prerequisites for monetary integration: price and wage flexibility, mobility of factors of production, financial markets integration, economic openness, production and consumption diversification, inflation correlation and fiscal and political integration; achieving fiscal integration is only possible if there is enough political drive (issing, 2004). further studies focused on the cost-benefit analysis with a special focus on the need for real exchange rate adjustments in the absence of the fine-tuning option provided by the sovereign monetary policy (corden, 1972). moreover, the inability to steer the economy along the unemployment-inflation curve (i.e., the philips curve) coupled with some unrealistic assumptions of the oca theory might entail very high costs for a fixed exchange rate regime (ishiyama, 1975), especially in the context of asymmetric shocks (mongelli, 2002). the advent of the monetarist critique of the short-term phillips curve (i.e. wage bargaining takes into account expected, not current inflation) debilitated the argument for a sovereign monetary policy (mccallum, 1989; artis, 1991) and subsequent debates centered around the negative effects of higher inflation (association with higher unemployment levels and lower income per capita) (emerson, 1992) and on the credibility of the sovereigns. countries with historical higher inflation might suffer from credibility issues when pursuing inflation reduction policies since there is always the risk of reversing them. one way of gaining credibility is by “tying its hands” (also referred to as an “ulysses pact”), i.e. entering a monetary union with another lowinflation anchor country (giavazzi & giovannini, 1989; mongelli, 2002). 2.2. ocas meta property – shock correlations the studies cited above, although setting up the theoretical framework for the debate, lacked nevertheless the empirical dimension. the already advanced state of the debate, coupled with a renewed interest for ocas kindled by the integration push from the delors commission, and with methodological advancements, mainly in time series analysis, set the scene for the “empirical studies phase” in the literature of ocas (mongelli, 2002). price and wage flexibility, labor, factor, and financial markets integration, as well as the degree of economic openness and of political integration were all empirically assessed for a series of countries/regions (for a detailed literature review please refer to mongelli, 2002). a vein of the literature concentrated on a “catch-all” or meta oca property: the similarity of shocks; the present study falls in this category. the reason for which the similarity of shocks is envisaged as some sort of catch-all property for an oca is that in the presence of high correlation between the shocks affecting different economies, the need to have specific and divergent fiscal and monetary responses across the currency union drops, thus rendering the sovereignty in these areas less useful. a major impediment in assessing the correlation between the shocks was with respect to the source of them; some of the shocks might not be totally exogenous in the case of countries with sovereign monetary policy, since the said policy might act as a source of asymmetric shocks adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 111 (albeit temporary). in order to solve this issue, bayoumi and eichengreen (1993) implemented a methodology first developed by blanchard and quah (1989) in order to discriminate between permanent and temporary shocks (as the ones generated by a sovereign monetary policy). it consists of extracting from the time series of prices and output data the demand and supply shocks by first estimating a structural vector autoregressive model (svar) and then discerning between temporary (demand-side driven) and permanent (supply-side driven) shocks. a correlation between the series of demand and supply shocks is then computed for all the prospective countries of a monetary union, on the one side, and the average of the union (or an anchor country), on the other side, thus obtaining a quantifiable indication of the optimality of single currency adoption in each country (de grauwe, 2018). 2.3. empirical studies on shock correlations the firsts to discriminate between the supply and demand shocks on the basis of the effect they have on output and unemployment were blanchard and quah (1989), establishing that the former have a permanent effect on the gnp after two years, reaching a plateau after five, while the latter will determine a hump-shaped mirror image on both gnp and unemployment. bayoumi and eichengreen (1993) borrowed from the previously-cited study the methodological apparatus and used it in the framework of ocas literature. employing data spanning from 1960 to 1988 for output and inflation for 12 european community (ec) member states, among others, they established that the underlying shocks are considerably more idiosyncratic across these countries, than across the us, which corroborated with the lower factor mobility, lead the authors to infer the increased difficulty of operating a monetary union in ec. nevertheless, at the same time a cluster of countries concentrated around west germany was identified as the core of the future emu on the basis of increased shock synchronization. their findings were confirmed by bayoumi and eichengreen (1996) and demertzis et al. (2000). the finding that in europe there is more symmetry on the demand side, mainly manipulated by the policy intervention, made demertzis et al. (2000) conclude that the economic and monetary union, while not naturally an oca, is held together by policies targeted at synchronization. unsurprisingly, shock similarity between regions within the same country was found to be higher than the one between countries in europe, as the cases of us or germany prove (bayoumi & eichengreen, 1993; funke, 2000), but however, account for a big part of the variability across the euro area, with peripheral countries having important regions in the core and vice versa (forni & reichlin, 1997). studies assessing the shock similarity between central and eastern european countries (ceecs) and their western counterparts, are scanter than the ones solely concerned on western europe and are mostly concentrated in the period of early 2000, when the post-communist bloc was still struggling with economic transition and was in the process of eu accession. frenkel and nickel (2002), building on previous research (frenkel et al., 1999) with observations spanning from 1993 to 2001, found the highest correlations of demand shocks with adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 112 emu as a whole to be with hungary and poland, while on the supply side, the best candidates for euro adoption were hungary and slovenia. however, due to the lack of appropriate quarterly gdp data, romania was not included in the sample. the first study to include romania produced clear-cut evidence against euro adoption, as the correlation values with the euro area were virtually 0 (fidrmuc & korhonen, 2003) and brought again into attention that among the ceecs, hungary was by far the most aligned with the shocks in the single currency zone. the literature focused on this meta oca property lost momentum after the empirical researches conducted in late 90s and early 2000s. campos and macchiarelli made a relatively recent reassessment of the shock similarity situation in the euro area (reaching the conclusion that the core-periphery gap narrowed since the initial studies), but without including the ceecs (campos & macchiarelli, 2016). the fact that these empirical researches are concentrated in early 2000 is due to two main reasons (frenkel & nickel, 2002): data availability (the lack of relevant or reliable data for the decade of the 90s determined small sample sizes, affecting the estimation results and burdening the previous attempts), and political and economic uncertainty (when and by how much was the eu/emu to expand in central and eastern europe?). as such, the results obtained by these studies might not currently hold, as important structural changes might have taken place; ceecs had already known considerable periods of integration with their western counterparts since then. the most recent attempts to reassess the romanian situation (we are aware of) are detailed in the following paragraphs. in a 2013 paper which briefly touched upon the shock similarity aspect in the context of a wider ceecs sample and confirmed the initial intuition in the literature: the eu integration period structurally modified the economies of the new member states and synchronized their shocks (bobeica & manu, 2013). noteworthy is the fact that while the correlation of supply-side shocks reached 0.5, the demand-side one stayed close to 0, suggesting that in the transition period in romania, the problem might have lied with behavior in consumption, which the authors link with the policy induced temporary disturbances such as the allowance of credit boom in the pre-2008 period and the cut in public sector wages in 2010 (highly pro-cyclical policies) (bobeica & manu, 2013). empirical studies conducted in the last two years seems to point towards mixed and inconclusive results. deskar-škrbić et al. (2020) assessing the three candidate countries from the third wave of expansion (bulgaria, croatia, romania) found evidence that they are all fit for adopting the counter-cyclical ecb monetary policy, notwithstanding their different past exchange rate regimes. on the other hand, grimm et al. (2021) rejected the notion that these countries share a common cyclical response pattern with the ea aggregate (with the exception of sweden), at best, their business cycles exhibiting very weak codependence (given the spillovers from the ea). in the same line, arčabić & škrinjarić (2021), argue that due to the large spillovers, especially since 2007-2016 period of great recession, national stabilization policies proved ineffective, and a greater degree of international policy coordination is recommended. one common shortcoming of these above-mentioned empirical studies is that they do not take adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 113 into account two extremely important structural changes suffered by the romanian economy that are particularly relevant for the shock correlation aspect. (1) in 2005 the national bank of romania was granted a new statute, considerably more independent from political mixtures (bodea & sánchez-santos, 2020) and changed its mandate from monetary base targeting to inflation targeting (deskar-škrbić et al., 2020). (2) in 2007, romania’s eu accession happens, granting a more integrated approach towards the fiscal and monetary policy. as such, in order to control for these changes, we propose a competing model taking into account strictly the post-2005-time series of our variables of interest, being able to isolate the latest irreversible developments without “the noise” of the 90s and early 2000s period. noting the inclusiveness of the empirical researches on the similarity of shocks between romania and the euro area, as well as the lack of any study to account strictly for the effect of the past decade of eu integration policies in romania, the present paper comes to fill these gaps in the literature and reassess the situation at 30 years since the start of the economic and political transition, at more than 14 years since central bank independence, and at more than 12 years since eu accession. of high importance is also the evaluation of the overall effect of policy induced short term disturbances (identified in the literature as the main driver for the demand side correlation); if for the immediately anteand post-2008 financial crisis period, the economic policy has been highly pro-cyclical and resulted in null correlations with the overall euro area trends, where does it stand now? 3. methodology, data, and tools 3.1. the ad-as model the methodological framework starts from the basic aggregate demand and aggregate supply model (ad-as). the ad curve is downward slopping since demand is inversely related to the level of prices (lower prices boost demand). on the other side, short-run as curve is upward sloping indicating the direct relationship between price level and firm output and the wage stickiness (implying that higher prices mean lower real salaries). at the same time, long-run as curve is perfectly inelastic to changes in price level and real wages adjust to changes in prices in the long-run (bayoumi & eichengreen, 1993). an expansionary aggregate demand shock will shift both output and prices levels upwards in the short-run, but in the long run, the real output will come back to its previous level, while the prices will raise to a level permanently higher. on the other side, an aggregate long-run supplyside shock (such as a technology improvement) will result in the short-run in lower prices and higher output, since it will shift downwards the short-run as curve. in the long run, the new as curve will become increasingly inelastic, shifting the long-run as to the right and permanently increasing the output, while putting downward pressure on prices (bayoumi & eichengreen, 1993). the exercise within the ad-as model proves that demand shocks will have a short-living effect on real output, while the supply shocks will have a permanent one. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 114 3.2. the svar model for demand vs. supply shocks discrimination the methodology firstly proposed by blanchard and quah (1989) and later employed and extended by bayoumi and eichengreen (1993), is the procedure for decomposing permanent and temporary shocks. the model starts from the infinite moving average representation of the bivariate vector autoregressive system featuring gdp and inflation data as: 𝑋𝑋𝑡𝑡 = 𝐴𝐴0𝜀𝜀𝑡𝑡 + 𝐴𝐴1𝜀𝜀𝑡𝑡−1 + 𝐴𝐴2𝜀𝜀𝑡𝑡−2 + 𝐴𝐴3𝜀𝜀𝑡𝑡−3 + ⋯ = ∑ 𝐿𝐿𝑖𝑖𝐴𝐴𝑖𝑖𝜀𝜀𝑡𝑡∞ 𝑖𝑖=0 [1] where 𝑋𝑋𝑡𝑡 is the bivariate vector [∆𝑦𝑦𝑡𝑡 ,∆𝑝𝑝𝑡𝑡], (stationary output and prices first-differenced time series), 𝐿𝐿𝑖𝑖 is the lag operator, and the matrix 𝐴𝐴 represents the impulse response functions of the shocks to 𝑋𝑋. as such, expression (1) can be written in matrix form as: �∆𝑦𝑦𝑡𝑡∆𝑝𝑝𝑡𝑡 � = ∑ 𝐿𝐿𝑖𝑖∞ 𝑖𝑖=0 � 𝑎𝑎11𝑖𝑖 𝑎𝑎12𝑖𝑖 𝑎𝑎21𝑖𝑖 𝑎𝑎22𝑖𝑖� � 𝜀𝜀𝑑𝑑𝑡𝑡 𝜀𝜀𝑠𝑠𝑡𝑡 � [2] where � 𝜀𝜀𝑑𝑑𝑡𝑡 𝜀𝜀𝑠𝑠𝑡𝑡 � is the vector of output and price disturbances and 𝑎𝑎11𝑖𝑖 represents the element 𝑎𝑎11 for the country 𝑖𝑖𝑡𝑡ℎ in matrix 𝐴𝐴𝑖𝑖 and so on. in order to allow for the temporary effect of inflation on output (i.e. demand shocks have no permanent effect on gdp), one must impose the restriction that element 𝑎𝑎11𝑖𝑖 of matrix 𝐴𝐴𝑖𝑖 is equal to 0. besides this restriction, there are three others to allow for the identification of all four elements of matrix 𝐴𝐴𝑖𝑖: the orthogonality (independence) of the demand and supply shocks and two restrictions regarding the normalization of the disturbances terms (fidrmuc & korhonen, 2003; campos & macchiarelli, 2016). another way to look at this is by starting from the structural var expression of the system (i.e. allowing for contemporaneous relations between both endogenous variables). assuming a bivariate svar(1) – with one lag, the expression is as follows. 𝐴𝐴𝑋𝑋𝑡𝑡 = 𝛽𝛽0 + 𝛽𝛽1𝑋𝑋𝑡𝑡−1 + 𝑢𝑢𝑡𝑡 [3] which in linear forms is 𝑦𝑦𝑡𝑡 + 𝑎𝑎12𝑝𝑝𝑡𝑡 = 𝑏𝑏10 + 𝑏𝑏11𝑦𝑦𝑡𝑡−1 + 𝑏𝑏12𝑝𝑝𝑡𝑡−1 + 𝑢𝑢𝑦𝑦𝑡𝑡 𝑎𝑎21𝑦𝑦𝑡𝑡 + 𝑝𝑝𝑡𝑡 = 𝑏𝑏20 + 𝑏𝑏21𝑦𝑦𝑡𝑡−1 + 𝑏𝑏22𝑝𝑝𝑡𝑡−1 + 𝑢𝑢𝑝𝑝𝑡𝑡 [4] and in matrix form is � 1 𝑎𝑎12 𝑎𝑎21 1 � � 𝑦𝑦𝑡𝑡 𝑝𝑝𝑡𝑡� = �𝑏𝑏10𝑏𝑏20 � + �𝑏𝑏11 𝑏𝑏12 𝑏𝑏21 𝑏𝑏22 � � 𝑦𝑦𝑡𝑡−1 𝑝𝑝𝑡𝑡−1� + � 𝑢𝑢𝑦𝑦𝑡𝑡 𝑢𝑢𝑝𝑝𝑡𝑡� [5] adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 115 in order to solve for this system of equations, one must premultiply the whole expression in equation (3) with the inverse of the matrix 𝐴𝐴, which will result in the so-called reduced-form var, with the following expression: 𝐴𝐴−1𝐴𝐴𝑋𝑋𝑡𝑡 = 𝐴𝐴−1𝛽𝛽0 + 𝐴𝐴−1𝛽𝛽1𝑋𝑋𝑡𝑡−1 + 𝐴𝐴−1𝑢𝑢𝑡𝑡 [6] and by substituting 𝐴𝐴−1𝛽𝛽0 with 𝐺𝐺0, 𝐴𝐴−1𝛽𝛽1 with 𝐺𝐺1 and 𝐴𝐴−1𝑢𝑢𝑡𝑡 with 𝑒𝑒𝑡𝑡 we get 𝑋𝑋𝑡𝑡 = 𝐺𝐺0 + 𝐺𝐺1𝑋𝑋𝑡𝑡−1 + 𝑒𝑒𝑡𝑡 [7] the identification issue persists also in this case, since there are more unknowns than parameters to be estimated, so the most straightforward way is to proceed in the same manner and impose the restrictions already mentioned above; following fidrmuc and korhonen (2003) and campos and macchiarelli (2016), we impose four restrictions: two restrictions are just normalizations defining the variance of the supply and demand shocks (𝑢𝑢𝑦𝑦𝑡𝑡 and 𝑢𝑢𝑝𝑝𝑡𝑡, respectively), the third one is the assumption of orthogonality of the same two shocks, and the last one is that the term 𝑎𝑎12 = 0 in matrix 𝐴𝐴, i.e. in the above-discussed case, allow for contemporaneous effect of output to prices, but not viceversa. this will result in equation [4] to change to 𝑦𝑦𝑡𝑡 = 𝑏𝑏10 + 𝑏𝑏11𝑦𝑦𝑡𝑡−1 + 𝑏𝑏12𝑝𝑝𝑡𝑡−1 + 𝑢𝑢𝑦𝑦𝑡𝑡 −𝑎𝑎21𝑦𝑦𝑡𝑡 + 𝑝𝑝𝑡𝑡 = 𝑏𝑏20 + 𝑏𝑏21𝑦𝑦𝑡𝑡−1 + 𝑏𝑏22𝑝𝑝𝑡𝑡−1 + 𝑢𝑢𝑝𝑝𝑡𝑡 [8] noteworthy is the fact that by imposing this restriction, the inverse of matrix 𝐴𝐴, i.e. matrix 𝐴𝐴−1, will also change such that 𝐴𝐴−1 = � 1 0 −𝑎𝑎21 1� [9] and equation [7] in its matrix form will become � 𝑦𝑦𝑡𝑡 𝑝𝑝𝑡𝑡� = � 𝑔𝑔10 𝑔𝑔20� + � 𝑔𝑔11 𝑔𝑔12 𝑔𝑔21 𝑔𝑔22� � 𝑦𝑦𝑡𝑡−1 𝑝𝑝𝑡𝑡−1� + � 𝑒𝑒𝑦𝑦𝑡𝑡 𝑒𝑒𝑝𝑝𝑡𝑡� [10] which will be equivalent to � 𝑦𝑦𝑡𝑡 𝑝𝑝𝑡𝑡� = � 𝑏𝑏10 −𝑎𝑎21𝑏𝑏10 + 𝑏𝑏20 � + � 𝑏𝑏11 𝑏𝑏12 −𝑎𝑎21𝑏𝑏11 + 𝑏𝑏21 −𝑎𝑎21𝑏𝑏12 + 𝑏𝑏22 � � 𝑦𝑦𝑡𝑡−1 𝑝𝑝𝑡𝑡−1� + � 𝑢𝑢𝑦𝑦𝑡𝑡 −𝑎𝑎21𝑢𝑢𝑦𝑦𝑡𝑡 + 𝑢𝑢𝑝𝑝𝑡𝑡� [11] note again how in this case also the errors (shocks) relate to one another, in the sense that an output shock will not be affected by price fluctuations, but the price level will be negatively affected by output increases, as the economic intuition dictates. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 116 after the demand and supply shocks are recuperated from the system, we perform correlations between all the countries in the eu (with some exceptions due to missing data) and we map them on a bi-dimensional graph to have the overall picture regarding shock similarity (as in fidrmuc & korhonen, 2003; bobeica & manu, 2013; campos & macchiarelli, 2016). in this regard, noteworthy is the common shortcoming of the previous papers of using uninterrupted time series. arguably, this strategy is employed to obtain a larger sample, but the downside is quite serious – the ample shocks from 90s and early 2000s is camouflaged in the series. this pose the problem of having a past noise that is no longer relevant for the present outlook since irreversible changes (such as nbr independence and eu accession) took place. the only paper we are aware of using broken down time series (deskar-škrbić et al., 2020) made an debatable choice in this regard in our opinion. the choice of 2009 as the break point is not a good reference since it’s a common shock to all and is losing sight of 16 quarters of important reforms in romania. 3.3. data, data treatment, and statistical tool we employ quarterly country-specific time series from eurostat for two variables: real gdp (in 2015 euro) and price index, both calendar and seasonally adjusted. to ensure the validity of using an svar, we performed two different test: (1) a johansen cointegration tests for the variables in levels and (2) a stationarity test (augmented dickey fuller test). the results of these tests for romania and eu15 specific variables (presented only for these two economic entities given the space constrains) are detailed in the statistical annex section. the results of the johansen test recommend the use of an svar model, as opposed to a vector error correction model (vecm). noticing the non-stationarity of the output in levels, we took its natural logarithm and performed a first-order difference (inflation already is stationary as presented in table 2 and table 3) – a procedure also used in many other similar paper (campos & macchiarelli, 2016; deskar-škrbić et al., 2020; fidrmuc & korhonen, 2003). the time span of both time-series variables is 1995q1-2019q4 including (with the exception of czech republic and netherlands which start from 1996q1) and our sample covers 34 entities: eu28 countries (without slovakia and malta due to lack of relevant data), switzerland, norway, and eurostat-aggregated data for eu28, eu27, eu15, ea12, ea19 and ea (changing composition). however, in order to take into account the very important issue of structural changes in the economy (discussed above), we employ the overall sample and another subsample (1995q1 – 2019q4 and the subsample 2005q1 – 2019q4). data treatment and the econometric analysis were performed in r and the specific blanchard and quah decomposition was achieved by executing the bq{vars} function developed by pfaff (2021). the lag selection was done following campos and macchiarelli (2016), i.e. of order 2. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 117 4. results, discussions, and implications although eurostat data included observations for the period previous to the adoption of the euro, for all ea19 member states, for the sake of historical accuracy and due to continuous change in composition of the latter, the main comparison counterpart for romania is set to be eu15. we start from raw data representation. the graphs of quarterly (quarter-to-quarter change) output and inflation variation in romania and eu15 are shown below in figure 2. the variation in both variables is noticeably much higher for romania, in accordance with two facts. in the first place, the period of economic transition brought pronounced ups and downs and the catching-up process has led to relatively higher output increases. the second fact is related to the crucial role played by the national bank of romania (nbr); the period up to 2005, i.e. until the independence of nbr, prompted politically-determined massive inflation rates. once the independence status was adopted, these rates were brought to moderate levels; however, the lack of a clearly-stated inflation target still allowed for relatively higher inflationary and deflationary periods, even after 2005 (bodea & sánchez-santos, 2020). the correlations between inflation and output movements tell two different stories. the heat map of correlations for output variation shows similar growth patterns, since the correlation scores do not go below 0, and even the very small positive correlation coefficients are rare (figure 3). the most obvious outliers in this case are surprisingly (membership to the eurozone should have pushed in the direction of business cycle synchronization) two euro area countries: greece and italy on the lower side of the spectrum, and two other non-eurozone countries at the other: poland and switzerland (the latter not even being an eu member state). contrary to the image depicted in figure 3, figure 4 shows considerable misalignments in terms of inflationary trends, to the point that it would be more accurately to talk about clusters, instead of core vs. outliers. the price disturbances seem to be most aligned in two areas: eastern europe, uk, norway and sweden, on the one side, and in the area of what the literature identifies as the “core of the eurozone”: germany, france, netherlands, belgium (to a smaller degree) and surprisingly spain (country that has consistently been identified as being at the periphery). adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 118 figure 2. quarterly output and inflation variation in romania and eu15 1995q1-2019q4 including. source: own elaboration with data from eurostat. pairwise country correlations in terms of output variation, place romania among the most well integrated countries in europe, with high scores across the board, the smallest coefficients being recorded in relation with three peripheral countries: greece, italy and portugal (although the same three countries registered the smallest scores with the rest of the entities in our sample). however, when it comes to inflation dynamics, the similarities with the rest of the countries seem inexistent, the highest (although weak) correlation coefficients being with other eastern european countries: croatia, poland, bulgaria and latvia. such fact can be explained by their similar contemporaneous economic history and transition effects. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 119 figure 3. heat map matrix of pearson correlation coefficients between output growth rates for all the entities in the sample. source: own elaboration with data from eurostat. note: the labels denominate the output (y) plus the two-letter country codes (e.g. at – austria). figure 4. heat map matrix of pearson correlation coefficients between inflation rates for all the entities in the sample. source: own elaboration with data from eurostat. note: the labels denominate the inflation (infl) plus the two-letter country codes (e.g. at – austria). adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 120 in a second step of the analysis, we perform the typical blanchard and quah svar modelling detailed in section 3.2 the svar model for demand vs. supply shocks discrimination. we recover the responses of the output and inflation from the supply and demand shocks and map them on cumulative impulse response function (irf) graphs for both romania and eu15. such mapping, depicted in figure 5, conveys two very similar, albeit at different magnitudes, responses. in line with the economic theory and expectative, positive supply shocks will have a permanent effect on output only, while price levels will suffer from a deflationary trend that will dissipate after 4 to 6 quarters. at the same time, demand shocks have very little effect on output for both romanian and eu15 aggregate economy, but the response in price level is more apparent; initial negative demand shocks tend to cause deflationary periods for 1 to 4 quarters in romania and for 1 to 6 quarters in eu15, after which the price levels will return to their baseline levels. noteworthy is the very high magnitude of responses in the case of romania; with the exception of the output response to demand impulse, the rest of responses seem to have a much higher degree of variation than in the case of the eu15 counterpart. the cause of this behaviour might be related to the higher magnitude of the shocks suffered by the transition economies and it might indicate that the romanian economy is much more responsive to economic stimuli. finally, the bi-dimensional mapping of the countries according to their correlations with romania on supply and demand shocks series (figure 6), indicates large overall exogenous shock correlations with the other two newest members, bulgaria and croatia, and with the two aggregated entities of eu27 and eu28. besides these, and fortunately for the proponents of euro adoption in romania, on the supply side shocks are more correlated with some of the main countries at the core of the euro area: germany, france and netherlands and with the euro area (ea, ea12 and ea19), signalling the strong ties and similarities between structural production systems. at the opposite end of the spectrum, stand three non-euro area economies (norway, denmark and surprisingly poland) and luxembourg (a very service-intensive economy). on the demand side, correlations are strong with three eastern european economies (bulgaria, croatia and poland) and with the eu27 and eu28 aggregates, while the ones with the big five euro economies (germany, france, italy, spain and netherlands) are placed around the 𝜌𝜌 = ±0.1 band. given the large weight of these five countries, this is an indication of possible adjustment problems to the euro area monetary policy for the romanian economy; a euro areawide monetary policy, with inflation and interest rates largely influenced by developments in these member states, might not fit the romanian macroeconomic needs. overall, the least similarity is observed with a cluster of countries consisting of the peripheral euro-area member states (greece, portugal, spain, italy, ireland), switzerland and luxembourg. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 121 figure 5. mapping of the cumulative effects of positive supply and negative demand shocks on output and inflation rates for romania and eu15. source: own elaboration. as previously indicated in the literature (bobeica & manu, 2013), romanian structural production system is relatively well integrated with the european one and with the ones of some core countries. nevertheless, the politically-driven demand shocks make romania an unfit candidate for euro adoption, as shown by the very low correlation coefficients across the board. high-inflationary periods caused by a politically influenced central bank in the 90s, booms and busts in the economy prompted by increases and cuts in salaries in the decade of 2000s, and budget deficits in the late 2010s (which caused inquiries from the commission within the stability and growth pact framework), brought hectic and ample movements in the demand behaviour of the romanian economy, moving it further away from the prospect of euro adoption in good conditions. nonetheless, one should not forget that such demand shocks have a temporary nature; a more disciplined and more investment-oriented fiscal behaviour, with a long-term sustainable perspective, especially with respect to the public debt (which is increasingly accelerating in recent years), could help in the alignment of the romanian economy with the rest of euro area’s core countries. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 122 figure 6. bidimensional mapping of the correlation coefficients (ρ) between supply and demand shocks series recuperated from the bq model using the whole sample (1995-2019) – romania as anchor country. source: own elaboration. figure 7. bidimensional mapping of the correlation coefficients (ρ) between supply and demand shocks series recuperated from the bq model using the sample after the breaking point (2005-2019) – romania as anchor country. source: own elaboration. adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 123 this point is proven by the results obtained after running the same model only for the subsample starting from 2005, i.e. after the implementation of the two structural changes (nbr independence and eu accession). considering the distribution of the countries on the same bidimensional space from figure 7, one can notice how both on the supply and demand side, the correlations increase; on the supply side, strong coefficients (above 0.5) were registered with eu and ea aggregates, germany, france, netherlands (core countries), while on the demand side, the highest scores were in relation with other non-euro area economies (latvia, poland, hungary, united kingdom, norway). again, the fact that the coefficients were all but one lower than 0.5, points that lax fiscal policy and the sovereign monetary policy proved a factor of divergence in terms of demand shocks similarity. 5. conclusions and recommendations arguably, this study, like any other focused on the meta property of shock similarity based on time series analysis, has the flaw of being backward looking and unable to predict/asses the full impact of a change in a monetary regime as the one implied by the entry into erm ii or the adoption of the single currency, which could bring more integration. nevertheless, it has the merit of accurately capturing the effect of 12 years of european integration, of 14 years of central bank independence, and of roughly 3 decades of transitionary efforts on the alignment of romanian economy with other european economies. the preliminary analysis of raw data for output and inflation rates indicated much higher variations than the eu average and strong alignments in terms of the former, while for the latter the highest correlations (albeit weak in absolute terms) were registered with eastern european peers. these facts are indications of a still ongoing catching-up process, a relatively well connected and integrated economy, and an inflationary-prone monetary policy. the assessment and mapping of the similarities between the underlying demand and supply shocks suggest that on the supply side such shocks are correlated to a certain degree with some relevant core countries, results that are in line with findings in the previous studies and which might make a strong case for the euro-adoption proponents. furthermore, the increase in correlation coefficients observed in the model using just the post-2005 subsample, is an indication that the romanian economy heads into the right direction. nevertheless, the values are still low compared with the ones obtained by the biggest euro area countries in their pairwise correlations, which suggests that more efforts could be made in order to increase the similarity of the economic structure and the connectedness; in this respect the recovery endeavours aimed at investments made in the post-pandemic period could represent a good opportunity to steer the economic structure in this direction. the results obtained on the demand side are more unsatisfactory for a quick and optimal euro adoption; low correlation coefficients were observed all across the board, with the exception of just two small-sized eastern european economies. these feeble results could be linked to the politically-motivated inflationary disturbances in the 90s and to the expansionary fiscal shocks adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 124 consistently administered to the economy during the sample period. the post-2005 results confirm the above findings, although correlation coefficients appears somewhat higher. nevertheless, one should not lose sight of the fact that such demand shocks have a temporary nature and that a more disciplined and more prudential fiscal behaviour, compatible with the convergence criteria for euro adoption, would help mitigating the negative impact of demand shocks to the co-movement of the romanian economy with the euro area. acknowledgements the author would like to express his gratitude to elena bobeica from the european central bank for her valuable input and guidance, as well as to two anonymous referees for their very useful suggestions. references arčabić, v., & škrinjarić, t. (2021). sharing is caring: spillovers and synchronization of business cycles in the european union. economic modelling, 96(c), 25–39. https://doi.org/ 10.1016/j.econmod.2020.12.023 artis, m. j. (1991). one market, one money: an evaluation of the potential benefits and costs of forming an economic and monetary union. open economies review, 2(3), 315–321. https://doi.org/10.1007/bf01886149 bayoumi, t., & eichengreen, b. (1993). shocking aspects of european monetary integration (f. torres & f. giavazzi, eds.; pp. 193–235). cambridge university press. https://doi.org/ 10.1017/cbo9780511599231.014 bayoumi, t., & eichengreen, b. (1996). operationalizing the theory of optimum currency areas. in cepr discussion papers (no. 1484; cepr discussion papers). c.e.p.r. discussion papers. https://ideas.repec.org/p/cpr/ceprdp/1484.html blanchard, o. j., & quah, d. (1989). the dynamic effects of aggregate demand and supply disturbances. the american economic review, 79(4), 655–673. http://www.jstor.org/stable/ 1827924 bobeica, e., & manu, a. (2013). empirical analysis of business cycle synchronization and shock similarity between romania and the eurozone. emerging markets finance and trade, 49(4), 74–97. scopus. https://doi.org/10.2753/ree1540-496x490404 bodea, a., & sánchez-santos, j. m. (2020). seigniorage and inflation tax in romania. what is the executive giving up by adopting the euro? scientific annals of economics and business, 67(1), 75–91. https://doi.org/10.2478/saeb-2020-0004 https://doi.org/10.1016/j.econmod.2020.12.023 https://doi.org/10.1016/j.econmod.2020.12.023 https://doi.org/10.1007/bf01886149 https://doi.org/10.1017/cbo9780511599231.014 https://doi.org/10.1017/cbo9780511599231.014 https://ideas.repec.org/p/cpr/ceprdp/1484.html http://www.jstor.org/stable/1827924 http://www.jstor.org/stable/1827924 https://doi.org/10.2753/ree1540-496x490404 https://doi.org/10.2478/saeb-2020-0004 adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 125 campos, n. f., & macchiarelli, c. (2016). core and periphery in the european monetary union: bayoumi and eichengreen 25 years later. economics letters, 147, 127–130. https://doi.org/10.1016/j.econlet.2016.07.040 corden, w. m. (1972). monetary integration, essays in international finance. international finance section, 93, 1–45. de grauwe, p. (2018). economics of monetary union. oxford university press. demertzis, m., hallett, a. h., & rummel, o. (2000). is the european union a natural currency area, or is it held together by policy makers? weltwirtschaftliches archiv, 136(4), 657–679. https://www.jstor.org/stable/40440810 deskar-škrbić, m., kotarac, k., & kunovac, d. (2020). the third round of euro area enlargement. are the candidates ready? journal of international money and finance, 107, 102205. https://doi.org/10.1016/j.jimonfin.2020.102205 emerson, m. (1992). one market, one money. an evaluation of the potential benefits and costs of forming an economic and monetary union. oxford university press on demand. fidrmuc, j., & korhonen, i. (2003). similarity of supply and demand shocks between the euro area and the ceecs. in royal economic society annual conference 2003 (no. 77; royal economic society annual conference 2003). royal economic society. https://ideas.repec.org/p/ecj/ac2003/77.html forni, m., & reichlin, l. (1997). national policies and local economies: europe and the united states (cepr discussion paper no. 1632). c.e.p.r. discussion papers. https://econpapers.repec.org/paper/cprceprdp/1632.htm frenkel, m., & nickel, c. (2002). how symmetric are the shocks and the shock adjustment dynamics between the euro area and central and eastern european countries? (ssrn scholarly paper id 880912). social science research network. https://papers.ssrn.com/abstract=880912 frenkel, m., nickel, c., & schmidt, g. (1999). some shocking aspects of emu enlargement. ssrn electronic journal. https://doi.org/10.2139/ssrn.164551 funke, m. (2000). macroeconomic shocks in euroland vs the uk: supply, demand, or nominal? in quantitative macroeconomics working papers (no. 20001; quantitative macroeconomics working papers). hamburg university, department of economics. https://ideas.repec.org/p/ham/qmwops/20001.html giavazzi, f., & giovannini, a. (1989). limiting exchange rate flexibility. the mit press. https://mitpress.mit.edu/books/limiting-exchange-rate-flexibility grimm, l., steinkamp, s., & westermann, f. (2021). on optimal currency areas and common cycles. are the acceding countries ready to join the euro? (ssrn scholarly paper id 3830221). social science research network. https://papers.ssrn.com/abstract=3830221 ishiyama, y. (1975). the theory of optimum currency areas. a survey. staff papers (international monetary fund), 22(2), 344–383. https://doi.org/10.2307/3866482 https://doi.org/10.1016/j.econlet.2016.07.040 https://www.jstor.org/stable/40440810 https://doi.org/10.1016/j.jimonfin.2020.102205 https://ideas.repec.org/p/ecj/ac2003/77.html https://econpapers.repec.org/paper/cprceprdp/1632.htm https://papers.ssrn.com/abstract=880912 https://doi.org/10.2139/ssrn.164551 https://ideas.repec.org/p/ham/qmwops/20001.html https://papers.ssrn.com/abstract=3830221 https://doi.org/10.2307/3866482 adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 126 issing, o. (2004). economic and monetary union in europe. political priority versus economic integration? in b. ingo, c. volker, & s. bertram (eds.), political events and economic ideas. edward elgar publishing. https://econpapers.repec.org/repec:elg:eechap:3131_2 mccallum, b. t. (1989). monetary economics: theory and policy. macmillan ; collier macmillan. mckinnon, r. i. (1963). optimum currency areas. the american economic review, 53(4), 717– 725. https://www.jstor.org/stable/1811021 mongelli, f. p. (2002). “new” views on the optimum currency area theory. what is emu telling us? mundell, r. a. (1961). a theory of optimum currency areas. american economic review, 51(4), 657–665. pfaff, b. (2021). bq function. r documentation. https://www.rdocumentation.org/packages/vars/ versions/1.5-3/topics/bq https://econpapers.repec.org/repec:elg:eechap:3131_2 https://www.jstor.org/stable/1811021 https://www.rdocumentation.org/packages/vars/versions/1.5-3/topics/bq https://www.rdocumentation.org/packages/vars/versions/1.5-3/topics/bq adrian bodea / european journal of government and economics 10(2), december 2021, 107-127 127 statistical annex table 1. results of the johansen cointegration test for romania and eu15 specific variables (in levels). note: only the case of these two economic entities has been included due to space constraints. values of teststatistic and critical values of test (romania, maximum eigenvalue statistic): values of teststatistic and critical values of test (eu15, maximum eigenvalue statistic): test 10pct 5pct 1pct test 10pct 5pct 1pct r <= 1 | 11.41 7.52 9.24 12.97 r <= 1 | 8.60 7.52 9.24 12.97 r = 0 | 55.23 13.75 15.67 20.20 r = 0 | 36.23 13.75 15.67 20.20 table 2. results of the augmented dickey-fuller test for first-order difference (fod) gdp and inflation for romania and eu15 (1995-2019 sample). note: only the case of these two economic entities has been included due to space constraints. augmented dickey-fuller test fod gdp romania augmented dickey-fuller test inflation ro augmented dickey-fuller test fod gdp eu15 augmented dickey-fuller test inflation eu15 type 1: no drift no trend type 1: no drift no trend type 1: no drift no trend type 1: no drift no trend lag adf p.value lag adf p.value lag adf p.value lag adf p.value [1,] 0 -5.76 0.01 [1,] 0 -7.69 0.01 [1,] 0 -3.45 0.01 [1,] 0 -5.96 0.01 [2,] 1 -3.88 0.01 [2,] 1 -7.25 0.01 [2,] 1 -3.04 0.01 [2,] 1 -3.94 0.01 type 2: with drift no trend type 2: with drift no trend type 2: with drift no trend type 2: with drift no trend lag adf p.value lag adf p.value lag adf p.value lag adf p.value [1,] 0 -6.87 0.01 [1,] 0 -8.27 0.01 [1,] 0 -4.40 0.01 [1,] 0 -8.21 0.01 [2,] 1 -4.72 0.01 [2,] 1 -8.19 0.01 [2,] 1 -4.01 0.01 [2,] 1 -5.69 0.01 type 3: with drift and trend type 3: with drift and trend type 3: with drift and trend type 3: with drift and trend lag adf p.value lag adf p.value lag adf p.value lag adf p.value [1,] 0 -6.93 0.01 [1,] 0 -8.31 0.01 [1,] 0 -4.52 0.01 [1,] 0 -8.49 0.01 [2,] 1 -4.82 0.01 [2,] 1 -8.28 0.01 [2,] 1 -4.16 0.01 [2,] 1 -5.84 0.01 table 3. results of the augmented dickey-fuller test for first-order difference (fod) gdp and inflation for romania and eu15 (2005-2019 sample). note: only the case of these two economic entities has been included due to space constraints. augmented dickey-fuller test fod gdp romania augmented dickey-fuller test inflation ro augmented dickey-fuller test fod gdp eu15 augmented dickey-fuller test inflation eu15 type 1: no drift no trend type 1: no drift no trend type 1: no drift no trend type 1: no drift no trend lag adf p.value lag adf p.value lag adf p.value lag adf p.value [1,] 0 -3.90 0.01 [1,] 0 -6.69 0.01 [1,] 0 -3.03 0.01 [1,] 0 -4.72 0.01 [2,] 1 -2.84 0.01 [2,] 1 -4.58 0.01 [2,] 1 -3.04 0.01 [2,] 1 -3.50 0.01 type 2: with drift no trend type 2: with drift no trend type 2: with drift no trend type 2: with drift no trend lag adf p.value lag adf p.value lag adf p.value lag adf p.value [1,] 0 -4.91 0.01 [1,] 0 -7.23 0.01 [1,] 0 -3.31 0.0213 [1,] 0 -5.69 0.01 [2,] 1 -3.61 0.01 [2,] 1 -5.09 0.01 [2,] 1 -3.34 0.0198 [2,] 1 -4.33 0.01 type 3: with drift and trend type 3: with drift and trend type 3: with drift and trend type 3: with drift and trend lag adf p.value lag adf p.value lag adf p.value lag adf p.value [1,] 0 -4.86 0.0100 [1,] 0 -7.40 0.01 [1,] 0 -3.29 0.0814 [1,] 0 -5.63 0.01 [2,] 1 -3.58 0.0426 [2,] 1 -5.29 0.01 [2,] 1 -3.34 0.0732 [2,] 1 -4.25 0.01 assessing the optimality of euro adoption in romaniathrough shock correlations 1. introduction 2. literature review 3. methodology, data, and tools 4. results, discussions, and implications 5. conclusions and recommendations acknowledgements references ©the author(s) 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 120-135 https://doi.org/10.17979/ejge.2024.13.2.10853 submitted: may 31, 2024 accepted: oct 3, 2024 published: dec 3, 2024 article do rising interest rates matter for bank profitability? evidence from portuguese banks clara pires,1 carlos borralho,1, * ana cantarinha1 1 polytechnic institute of beja & ceos.pp, portugal *correspondence: cborralho@ipbeja.pt abstract. the purpose of this paper is to analyse the profitability determinants of seventeen banks operating in portugal from 2013 to 2023. the banking market has changed significantly, particularly since 2021, when euribor grew rapidly to control inflation target. methodologically, a hypothetical-deductive approach was used based on panel data collected from banks' published accounts. to generate results, grouped ordinary least squares were applied, as the breusch-pagan test confirms homoscedasticity, an essential assumption in this regression model. internal variables considered include credit quality, capital adequacy, management quality, financial margin, and bank size, alongside an external variable, the euribor. the findings reveal that credit risk, capital adequacy, management capacity, and euribor are the most statistically significant for both return on equity and return on assets, with euribor emerging as the greatest statistically significant variable. the analysis of euribor as an explanatory variable represents the key contribution of this study relative to the existing and reviewed literature. keywords: banks; profitability; euribor jel classification: g01 ; g21 ; g28 1. introduction studies around bank profitability have received a lot of attention from researchers. from the financial crisis that began in 2007 in the united states to subprime loans (triggered by the granting of high-risk mortgage loans to low-income families), the pandemic, and recent wars, portuguese banking has shown that it is not indifferent to the risk factors that have affected national banking at various levels, particularly in terms of profitability. for example, it is very present in the insolvency of bpn, banif, and bpp banks. the bes bank in 2014 had major state intervention so that the consequences of its collapse would not translate into a disaster for the portuguese economy. in this sense, banking supervision has become much more demanding, in compliance with more rigorous and robust ratios in terms of capital adequacy, liquidity, credit quality and market discipline; hence the importance of constant improvements and requirements imposed by the basel accords (bcbs, 2013; bcbs, 2014; bcbs, 2017). https://creativecommons.org/licenses/by-nc/4.0/ banks’ profitability in portugal from 2013 to 2023 according to the european central bank (bce), 2021 was the turning point in the monetary policy that followed over the last decade (european central bank, 2024). a period of negative interest rates has ended. in this regard, during the second half of 2023, the bce increased euribor (eur) by a total of 250 basis points. the eur increase occurred following rapid growth in inflation. in the eurozone, inflation rose to 8.4% on average in 2022, compared to 2.6% in 2021. according to the bce, the growth in inflation was due to two types of shock. on the supply side, there was a significant increase in the costs of production factors in all sectors of the economy. this increase in the costs of production factors was the result of import problems resulting from covid-19 and the energy crisis resulting from the russia-ukraine war. on the demand side, the reopening of the economy after the pandemic allowed companies to reflect increased costs in prices more quickly and sharply. the evolution of inflation, with projections above the 2% objective in the medium term, led the bce to increase eur considerably and at a constant pace as a way of dampening demand and ensuring that inflation remained under control. from this perspective, we analyse the impact of the increase in eur on the profitability of 17 banks operating in portugal in the period between 2013 and 2023. this sample represents about 98% of the portuguese banking product (appendix i). this paper is structured as follows. section 2 presents a brief literature review. the methodology, econometric approach, and data used are detailed in section 3. section 4 presents the results and discussions, and section 5 provides the main conclusions. 2. literature review there is a lot of research in this area, and it is quite diverse. there are researchers interested in this topic on all continents. in this sense, given the difference in legislation, supervisory rules, and cultural aspects, it is not always easy to carry out a comparative analysis of the results obtained. table 1 presents the main authors of papers where it was possible to obtain information and compare it with the results of our investigation. table 1. fundamental literature review: data, variables, and main conclusions. authors data variables conclusions dependent independent (carvalho & ribeiro, 2016) 29 banks in portugal between 2002 and 2012. roa. capital adequacy; non-performing assets and credit risk. the capital adequacy, non-performing assets and credit risk are statistically significative in profitability, measured by roa. haddad et al. (2019) 12 banks in jordan between 2009 and 2019. roa; roe and nim. assets size; capital adequacy; assets quality; liquidity; deposit; economic activity; inflation; interest rate. the results show that capital adequacy, asset quality, liquidity, deposit, net interest margin, and asset size are important factors that affect the profitability of the banking industry in jordan. (mota et al., 2019) 12 banks in portugal in the period 20062016. roa; roe and nim. sector concentration; costincome; liquidity risk; deposit growth; credit risk, cost-income, financial leverage and deposit growth stand out as the internal determinants of banks that best explain profitability. 122 pires et al. authors data variables conclusions dependent independent financial leverage and gdp. (neves et al., 2020) 66 portuguese and spanish banks were analysed. roa and roe. size; gdp; costincome; bank capitalization and annual deposit growth. the result shows a positive and negative non-linear relationship between bank size and their levels of profitability, measured by roa and roe. nunes (2021) 40 portuguese banks from the first quarter of 2010 to the last quarter of 2019. roe and roa. capital adequacy; loan loss provisions; bank size; costincome; loans; deposit ratio; gdp and euribor. the findings support the notion that banks with enough capitalization outperform those with insufficient capitalization. the size significantly reduces the profitability of portuguese banks when measured by roa and roe. loan loss provisions reduced the portuguese banks’ profitability. yuan et al. (2022) 20 banks in bangladesh and 20 banks in india from 2010 to 2020. roa and roe. deposit to asset; loan to deposit; bank size; debt to asset; inflation target and gdp. deposit to asset ratio and loan to deposit ratio were found to, significantly and negatively, impact roa for both countries. bank size, debt to asset ratio, inflation rate, and gdp were shown to have a favourable impact. deposit to asset ratio, inflation rate, and debt to equity ratio were negative and significant for roe. gdp and bank size were positive and significant. kasanaa et al. (2023) 50 international banks in the period (2008– 2020). roa; roe and nim. interest rate; spread; capital adequacy; non-performing assets; liquidity management; size; concentration and inflation target. the profitability of banks is positively correlated with the interest spread. while non-performing assets and concentration have a strong negative influence on roa, capital sufficiency, net interest margin and liquidity management have a considerable positive association. mashamba and chikutuma (2023) 11 zimbabwean banks in the period (2011 2020). roa; roe and nim. income diversification; liquidity; cost-toincome; capital adequacy; credit risk; gdp; inflation target; market structure and bank stability. this research considers a positive relationship between the cost-toincome ratio and bank profitability. the study reports that gdp and inflation do not have a significant effect on bank profitability in zimbabwe. puci et al. (2023) 12 albanian banks between 2011 and 2020. roa and nim. gdp growth rates; unemployment rates; interest rates; inflation target and loan loss reserves. the study looked at roa. although there has been some fluctuation in albania's financial industry over the past four years, the roa has improved dramatically. albanian banks' profitability was significantly negatively impacted by gdp growth, inflation target, and interest rates between 2011 and 2020. qehajakeka et al. (2023) 400 banks of kosovo and albania in the period (2010 2020). roa and roe. number of loans; number of employees; nonperforming loans and interest rate. the research concluded that number of workers, interest rate on loans, and percentage of loans in default affected the banks’ profitability. however, the number of bank employees did not. lamothe et al. (2024) 2091 banks from 110 countries roe; roa; nim. non-performing loans; efficiency ratio; gross margin; the findings show that default loans, efficiency, gross margin, capitalization, interest rates and gdp growth have a banks’ profitability in portugal from 2013 to 2023 authors data variables conclusions dependent independent (moody's database). cash liabilities ratio; customer deposits; customer loans; inflation target; capitalization; unemployment; gdp and interest rates. negative and statistically significant impact on banks’ profitability. martinho et al. (2024) 110 european banks (bankscope database). roa and nim. interest income; interest expenses; tier; cost-to-income; impairments; bank size; gdp growth; three-month euribor. substantial declines in average bank roa can be attributed to unfavourable economic situations and variations in interest rates. notably, bank profitability is positively impacted by gdp. the three-month euribor has a positive and significant impact on roa. source: own elaboration. the literature review found that, like our study, many authors used return on equity (roe) and return on assets (roa) to explain banking profitability (e.g., neves et al., 2020; nunes, 2021; yuan et al., 2022; qehaja-keka et al., 2023). usually, the dependent variables used to measure profitability are the return on equity (roe) and the return on assets (roa). while roe expresses the net return of the capital invested by the shareholders, the roa shows the net relative profit produced by the bank's total assets and is considered a measure of management efficiency (petria et al., 2015). in our study, we apply roe and roa to measure the profitability of portuguese banks (pires, basílio & borralho,2021). some researchers applied roe, roa and nim as the dependent variable (e.g. haddad et al., 2019; mota et al., 2019; kasanaa et al., 2023; mashamba and chikutuma 2023 and lamothe et al., 2024). on the other hand, about explanatory variables, most researchers in table 1 applied internal variables related to credit quality, bank size, and capital adequacy or bank solvability. regarding external variables, the literature review showed that inflation target and gdp are the most applied. in our study, we decided to apply the 12-month euribor rate. table 2 lists articles whose reading proved to be very important for understanding this topic but where it was difficult to establish a comparative analysis with the results obtained in this study. assessing the profitability of banks using panel data is a consolidated practice in the literature. although different econometric methods may be used, all have associated advantages and disadvantages. thus, in addition to the works contained in the two previous tables, it is possible to observe in these works of, among others, adem (2023), akther et al. (2023), dogan and yildiz (2022), jílková and kotěšovcová (2022), kotte et al. (2022), and lekpek and šabotić (2023). in this study, we chose to use the classic pooled ordinary least squares (ols) model, as it is one of the most tested and used, as detailed in the methodology. 124 pires et al. table 2. additional literature review: data, variables, and main conclusions authors data variables conclusions dependent independent jadah et al. (2020) 80 iraqi banks during the period (2005 2017). roa, roe and nim. bank size; credit risk; liquidity; total loans to total assets; gdp; inflation target; interest rate; unemployment; regulatory quality; political instability and government effectiveness. the results reveal that most bankspecific characteristics, economic conditions, and government variables have a statistically significant impact on the performance of iraqi commercial banks. the study suggests that the size of iraqi banks and the total equity to total assets ratio are among major drivers of iraqi banks’ profitability. boto-garcía et al. (2021) spanish banking sector during the period 1995– 2016. roa and roe. 12-month euribor; gdp and herfindahlhirschman (hh) index. the profitability rises with the longterm interest rate and falls with the 12-month euribor rate. the gdp variable increases roa. there is no proof that market concentration has a major impact on profitabil ity. hukwuogor t al. (2021) european banks from the period 1996 2019. nim. gdp rates; unemployment rates; interest rates; inflation rates and loan loss reserves. the findings show that the us economy's growth, inflation rate, net interest margin, percentage of nonperforming loans, and unemployment rate all considerably boost banks’ profitability. kozak and wierzbowska (2022) 40 european nations in the years 2019 and 2020. roa diversification; bank size; asset growth; loan-to-asset; tier; gdp growth; interest rate; lockdown; economic support; income support; debt relief and covid-19 pandemic. the findings show that the profitability of european banks was positively and statistically significantly impacted during the covid-19 epidemic by the growth in non-interest income as a percentage of total income. shahin et al. (2022) 22 kuwait banks. the sample period covers 2011 to 2020. roe interest rate; tier and bank size. bank profitability is lowered by low interest rate. capital levels above a certain threshold improve banks' profitability. most notably, the results demonstrate that banks with larger levels of prudential capital are less vulnerable to the negative impacts of low interest rates on profitability. (ho et al., 2023) 1231 banks of 90 countries, from 2008 to 2020. nim and roa. bank size; costincome; bank liquidity; loan-todeposit; nonperforming loan; market power; inflation target; gdp growth and money supply. the results indicate that bank size, roa, operational cost to operating income ratio, loan-to-deposit ratio, and non-performing loan positively affect the net interest margin of banks. son (2024) 13 banks vietnamese stock market between 2010 and 2022. nim credit quality; bank size; gdp and inflation target. the empirical findings show that increasing a bank's net interest margin (nim) is essential to increasing profits. greater profitability is indicated by a higher nim level. source: own elaboration. the following section outlines the institutions under study, the dependent and independent variables selected, and the methodology that allowed us to draw conclusions from our investigation. banks’ profitability in portugal from 2013 to 2023 3. data and methodology the methodology used is a panel data model, a pooled ols over the period 2013 2023. by using a balanced panel in the present study, it was possible to identify the variables of all companies throughout the entire period under analysis. the pooled model was chosen because it achieves the best results, given the small size of our sample. the software chosen to analyse the empirical model was r studio. 3.1 sample the main objective of this study is to analyse the impact of an increase in euribor on the profitability of banks operating in portugal over the period 2013 2023. the panel dataset is composed of annual data retrieved from the banks' annual reports on management efficiency, credit quality, capital adequacy, size, financial margin, banking product and profitability. the sample of 17 banks includes bpi, bcp, cgd, sicam, std, mtp, nb, bic, big, ctt, bai, fnt, ctl, bkt, db, best and atl, representing about 98% of the portuguese banking product (see appendix i). the variable euribor was obtained online via portugal bank´s databases. 3.2 study variables the dependent variables are return on equity (roe) and return on assets (roa). these variables are applied to measure the banks’ profitability. the independent variables are credit quality (icv and cvct), management efficiency (ctin), capital adequacy (tier 1), bank size (logat), financial margin (mf), banking product (pb) and euribor (eur). • in line with nunes (2021), impairment on non-performing loans (icv) is a proxy for credit risk. in practice, the higher this ratio, the lower the banks' credit quality, having a negative impact on their profitability. • in line with kasanaa et al. (2023), lamothe et al. (2024), qehaja-keka et al. (2023) and yuan et al. (2022), overdue loans over total loans (cvct) or non-performing loans are a proxy for credit risk, like a icv variable. typically, an increase in the ratio of non-performing loans will reduce banks' profitability. • the cost-income ratio (ctin) is defined as operating expenses over total gross profits. it is a proxy for management capacity and measures a bank's operating costs as a proportion of its total profits. in general, it measures the efficiency of a bank's management. an increase in operating expenses is expected to have a negative impact on banks' profitability if everything remains constant. this contrasts with mashamba and chikutuma (2023), where cost-income ratio had a positive impact on roa and roe. • the solvability ratio (tier 1) is defined as basic own funds or highest quality own funds over risk-weighted assets. this ratio is a proxy of capital adequacy and measures financial strength. in general, banks with higher solvability ratios are considered safer, with a positive impact on banks’ profitability (haddad et al., 2019; kasanaa et al., 2023). 126 pires et al. • bank size (logat) is calculated as the logarithm of the bank's total assets. this proxy has been widely used in previous literature, and its impact on bank profitability is not predictable. however, according to other studies (haddad et al., 2019; yuan et al., 2022), a positive correlation is expected between the size of the bank and its profitability. in contrast, nunes (2021) concluded that bank size has a positive impact on roa and roe. • financial margin (mf) is the term used in the analysis of banks and results from the difference between the interest charged on credits granted (calculated using the active interest rate) and the interest paid to holders of funds deposited in banks (calculated using the passive interest rate). it is expected to have a positive impact on banks’ profitability. • banking product (pb) corresponds to the gains achieved directly from banking activities. • euribor (eur) is an abbreviation for euro interbank offered rate; it represents the average interest rate charged on loans between a specific group of banks in euros. puci et al. (2023) show a negative impact on banks’ profitability before the year 2020. in our study, we expect a positive impact of euribor on banks’ profitability after 2021. it should be noted that the behaviour of euribor has changed significantly after this year. table 3 summarises the independent variables used and the expected effect on roe and roa. table 3. independent variables, definition, and expected effects variable definition expected effect icv impairment on non-performing loans − cvct overdue loans over total loans − ctin operating expenses over total gross profits − tier highest quality own funds over risk-weighted assets + logat logarithm of the bank's total assets + mf difference between active and passive interest rate + pb gains achieved directly from banking activities + eur average interest rates charged on loans + source: own elaboration. we expect a negative effect of credit quality (icv and cvct) and management quality (ctin). in contrast, we expect a positive impact of capital adequacy (tier), size and euribor. this expected effect has to do with the institutions that are currently active in portugal. the greatest requirements of supervisory rules resulting from the tree basel accords, particularly the third agreement, through which capital requirements, to address banking risks, namely liquidity, became much more stringent. as we know, we have lived through years in which euribor has presented negative values and banks have presented lower returns. however, families benefitted, particularly in the amount paid for loans obtained. since 2021, the euribor rate has been increasing. as a result, families have suffered thanks to an increased amount to be paid for loans obtained. however, we anticipate that banks expect the opposite a benefit resulting from an increased euribor. banks’ profitability in portugal from 2013 to 2023 3.3 econometric model we present balanced panel data because information is available for all 17 banks for the ten years under study. the model is as follows: 𝑌𝑌𝑖𝑖,𝑡𝑡 = 𝛽𝛽0 + �𝛽𝛽𝑘𝑘𝑋𝑋𝑘𝑘𝑖𝑖,𝑡𝑡 7 𝑘𝑘=1 + 𝜖𝜖𝑖𝑖,𝑡𝑡 where: 𝑌𝑌𝑖𝑖,𝑡𝑡 = 𝑅𝑅𝑅𝑅𝑅𝑅 and 𝑅𝑅𝑅𝑅𝑅𝑅 from bank i = 1, …, 17 and year t = 1, …,11 (2013–2023) β0 = constant βk = coefficients to be estimated by the model xki,t = vector of the explanatory variables 𝜖𝜖𝑖𝑖,𝑡𝑡= random error we applied pooled ols employing r studio. the breusch-pagan test was applied to analyse the existence of heteroscedasticity. based on the test carried out, we concluded that it does not exist. the random-effects model was therefore not chosen. the most appropriate model for our sample (17 banks) was the pooled ols (breusch & pagan, 1979; koenker, 1981). 4. results and discussion at this point, the statistical data, the results obtained and the comparative analysis with the literature review are presented. 4.1 statistical data our model considered ten variables. descriptions of these variables are presented in the following table 4. for each of the variables, the mean, standard deviation, minimum and maximum value are presented. on average, the 17 selected banks have a return on equity (roe) of 5.20% with a standard deviation of 12.74%. regarding return on assets (roa), the banks present an average indicator of 0.36% with a standard deviation of 2.38%. standard deviation shows large profitability heterogeneity among the banks in our sample in both roe and roa. the euribor variable (eur) presented negative values for several years (the minimum rate was -0.51%). this rate began to increase in 2021 to its maximum value in 2023, with a rate of 4.07%. the average euribor was 0.55% and the standard deviation 1.38%. 128 pires et al. table 4. summary statistics variable n mean std. dev. min max roa 187 0.36 2.38 -19.00 5.30 roe 187 5.20 12.74 -37.00 48.00 icv 187 86.41 62.78 27.60 430.20 cvct 187 4.78 4.38 3.2 18.70 ctin 187 56.08 15.64 27.60 100.60 tier 1 187 18.33 9.89 8.10 82.30 logat 187 3.91 1.00 1.49 8.01 mf 187 2.65 1.61 -0.74 6.12 pb 187 3.54 1.54 1.04 6.99 eur 187 0.55 1.38 -0.51 4.07 source: own elaboration employing r studio. according to table 5, we observe that only two variables (pb and mf) present values very close to 3. however, according to the literature, the variance inflation factor (vif) values would only be problematic if they were greater than 5 (fox & monette, 1992). in this sense, we verified that there are no collinearity problems between the independent variables. table 5. variance inflation factors icv cvct ctin tier logat mf pb eur 1.299 1.365 1.346 1.654 1.465 3.529 3.171 1.141 source: own elaboration employing r studio. according to the pearson's correlation matrix (table 6) for the independent variables, the banking product (pb) and financial margin (mf) variables had a high correlation (0.7972). the two variables that show the highest correlation also show the highest collinearity. it was decided to remove the pb variable from the explanatory model of banks’ profitability because our model improved without the pb variable. on the other hand, the mf variable is statistically more significant than the pb variable: this is the reason for eliminating the pb variable. banks’ profitability in portugal from 2013 to 2023 table 6. pearson's correlation matrix for the independent variables icv cvct ctin tier logat mf pb eur icv 1 cvct -0.1806 1 ctin 0.1252 0.1497 1 tier 0.0822 -0.2769 -0.1669 1 logat 0.0993 -0.0221 0.0530 -0.3554 1 mf 0.3898 0.2256 0.1936 -0.5085 0.4628 1 pb 0.3455 0.3208 0.1943 -0.4105 0.4572 0.7972 1 eur 0.1964 -0.0868 -0.0650 0.0669 0.1904 0.2215 0.2191 1 source: own elaboration employing r studio results. 4.2 estimation results the results are presented in table 7 and table 8. table 7 shows that the pooled model is considered appropriate given the results of the f statistic, which is significant at the 1% level (p-value: 7.4307e14). the r-squared is 64%, meaning that 64% of the variance of roe is explained by the set of exploratory variables applied. the test statistic bp=4.8074 follows a chi-square distribution with 7 degrees of freedom (number of independent variables) with a p-value of 0.6835. thus, we do not reject the null hypothesis, in this way being able to consider the validity of the premise of homoscedasticity and, as such, do not compromise the validity of the results obtained by the ols pool model. cvct (credit quality), ctin (management capacity), tier (capital adequacy) and eur (euribor) show great statistical significance (at 1% level). the cvct variable presents the expected sign and shows that banks with a higher level of non-performing loans will have lower profitability (roe). higher default also involves the establishment of a greater impairment level, which implies a lower net result. the ctin variable exhibits the expected sign and confirms that an increase in operating expenses has a negative impact on banks' profitability. the tier variable result confirms the idea that the higher the solvability ratio, the higher the banks’ profitability. usually, it shows that the bank with its robust own funds has higher roe. finally, when euribor increases by 1%, the bank's profitability (roe) improves by 2.303%, following the increase in the bank's financial margin. the euribor variable presents the expected sign and confirms that the increase since 2021 has had a very positive impact on the banks’ equity capital profitability. 130 pires et al. table 7. regression results y = roe coef. (β) cluster robust standard error t p-value icv -0.024 0.015 1.587 0.1143 cvct -0.852*** 0.020 -4.344 0.0000 ctin -0.228*** 0.051 -4.552 0.0000 tier 0.358*** 0.010 -3.563 0.0005 logat 2.445** 0.945 2.7034 0.0375 mf 1.816* 0.018 -2.499 0.1003 eur 2.303*** 0.727 3.843 0.0001 constant 20.56*** 5.379 3.382 0.0001 r-squared: 0.64084 p-value: f stat: 7.4307e-14 note: *, **, *** represent significance at the 10%, 5% and 1% level. source: own elaboration employing r studio. on the other hand, the logat (bank size) variable is statistically significant at the 5% level and the mf (financial margin) variable is statistically significant at 10%. the logat variable shows a positive correlation between the size of the bank and its profitability. this result was expected in our study. the mf variable has the lowest level of statistical significance. however, it is interesting to verify that our expectations regarding its behaviour were met. table 8 shows that the pooled model is considered appropriate, given the results of the f statistic, which is significant at the 1% level (p-value: 5.3199e-10). the r-squared is 54%, meaning that 54% of the variance of roa is explained by the set of independent variables applied. as for the test statistic, bp=9.5798 follows a chi-square distribution with 7 degrees of freedom (number of independent variables) with a p-value of 0.2958. thus, we do not reject the null hypothesis, consequently being able to consider the validity of the premise of homoscedasticity and as such do not compromise the validity of the results obtained by the ols pool model. banks’ profitability in portugal from 2013 to 2023 table 8. regression results y = roa coef. (β) cluster robust standard error t p-value icv 0.007** 0.003 2.419 0.0343 cvct -0.829** 0.039 -2.147 0.0298 ctin -0.012** 0.010 -1.212 0.0476 tier -0.119*** 0.020 -6.054 0.0000 logat 0.373** 0.178 2.093 0.0475 mf 0.302 0.143 -2.118 0.0892 eur 1.304** 0.118 2.570 0.0335 constant 2.197* 1.058 2.075 0.0106 r-squared: 0.54064 p-value: f stat: 5.3199e-10 note: *, **, *** represent significance at the 10%, 5% and 1% level. source: own elaboration employing r studio. of particular significance, at the 1% level, is the capital adequacy variable (tier). this variable result corroborates the idea that the higher the ratio, the lower the need for external funding, inducing lower asset profitability. the implication is that as capital adequacy increases, roa decreases. higher capitalisation has a negative impact on the asset capital profitability of banks. the credit risk variables (icv and cvct) are statistically significant at the 5% level. the cvct variable presents the expected sign to explain the roa. like the impact on roe, banks with a higher level of non-performing loans will have lower roa. on the other hand, the icv variable shows that an increase in impairment on non-performing loans has a positive impact on roa. the ctin variable is statistically less significant in roa than in roe. however, it presents the same sign: the increase in operating expenses has a negative impact on banks' profitability. the logat variable presents the same behaviour in roa and roe, as well as the same statistical significance (at the 5% level). it shows a positive correlation between the size of the bank and its return on assets. we show lower statistical significance in roa than in roe, the euribor variable present the expecting sign. this confirms that the increase after 2021 had a very positive impact on the asset capital profitability of banks. 132 pires et al. 4.3 comparative analysis by carrying out a comparative analysis between our results and the literature review, we have been able to verify the existence of some very similar conclusions. puci et al. (2023) show the negative impact that the decrease in interest rates had on albanian banks' profitability until 2020. our work demonstrates that euribor increased profitability after 2021. this variable has a positive impact on the profitability of portugal banks. in line with yuan et al. (2022), bank size has a positive and significant impact on roa and roe. in the same way, non-performing loans have a negative and significant impact on roe. according to haddad et al. (2019), capital adequacy (solvability), asset quality and bank size have a positive and significant impact on banks’ profitability. we arrived at the same conclusions. as in kasanaa et al. (2023) work, non-performing loans had a negative influence on roa. on the other hand, capital adequacy had a positive impact. lamothe et al. (2024) and qehaja-keka et al. (2023) reached the conclusion that non-performing loans had a negative impact on roa and roe. our research is not in line with that of mashamba and chikutuma (2023), where cost-income has a positive impact on roa and roe. in our work, we see that the cost-income variable (management assets) had a negative impact on profitability among portugal’s banks, measured by roa and roe. contrary to nunes (2021), bank size had a positive impact on roa and roe. however, the impairment of non-performing loans had a negative impact on banks’ profitability. similarly, martinho et al. (2024) show that interest rates have a positive and significant impact on roa. in terms of comparability of results, our work is in line with yuan et al. (2022), showing that bank size has a positive and significant impact on roa and roe. according to haddad et al. (2019), capital adequacy (solvability), asset quality and bank size have a positive and significant impact on banks’ profitability. like kasanaa et al. (2023), our work shows that non-performing loans had a negative influence on roa. on the other hand, capital adequacy had a positive impact. 5. conclusions the main goal of this study was to obtain the determinants of banks’ profitability in portugal over the period 2013 to 2023. above all, we sought to analyse the extent to which the increase in the euribor rate had an impact on portuguese banks’ profitability, measured by roe and roa. as in much of the literature (among others, haddad et al., 2019; lamothe et al., 2024; yuan et al., 2022), roe and roa were considered to measure the banks’ profitability. impairment of nonperforming loans, overdue loans over total loans, cost-income ratio, solvability ratio, bank size, financial margin and euribor were considered as explanatory variables. this choice is in line with the work of kasanaa et al. (2023), mashamba and chikutuma (2023), martinho et al. (2024), nunes (2021) and qehaja-keka et al. (2023). our results show that the variables that represent credit quality (cvct and icv), capital adequacy (tier 1), management capacity (ctin) and 12-month euribor present the expected sign. on the other hand, the range of exploratory variables in this study explains roe more adequately than roa. we were able to verify that 64% of the variance of roe is explained by the set of banks’ profitability in portugal from 2013 to 2023 independent variables applied. on the other hand, 54% of the variance of roa is explained by the set of exploratory variables employed. this conclusion is in line with the literature (petria, 2015). roe is computed as the ratio between the bank's net profit and equity: hence, to study the impact on banks' profitability, the measure of profitability chosen was roe. the roa shows the net profit generated by banks' total assets and is more appropriate for evaluating management efficiency. however, the main conclusion of this work concerns the euribor rate. when euribor increases by 1%, the banks’ profitability (roe) improves by 2.303%. in the same way, when euribor increases by 1%, the bank's profitability (roa) improves by 1.304%. so, despite the lower statistical significance for roa than for roe, the euribor variable presents the expected sign. this confirms that the euribor increase after 2021 had a very positive impact on the asset capital profitability of banks. this study sought to contribute to future research and show researchers the importance of this topic. in particular, the external variable (euribor) had not previously been tested in relation to portuguese banking profitability. compared to other studies focused on banking profitability, our significant contribution involves studying the effect of euribor on the portuguese banking sector. references adem, m. (2023). the impact of macroprudential policy and political institutions on bank credit risk in ethiopia: a dynamic ordinary least squares analysis. cogent business & management, 10(1), 1-20. https://doi.org/10.1080/23311975.2023.2257829 akther, t., rahman, m., & rahman, m. m. (2023). factors influencing commercial bank profitability in bangladesh: a panel data approach. future business journal, 9(66). https://doi.org/10.1186/s43093-023-00247-8 banco de portugal. (2023). relatório de estabilidade financeira. banco de portugal. https://bpstat. bportugal.pt/ bcbs. (2013). basel iii: the liquidity coverage ratio and liquidity risk monitoring. bis. bcbs. (2014). basel iii: the net stable funding ratio. bis. https://doi.org/10.5089/9781498346498.001 bcbs. (2017). basel iii: finalising post-crisis reforms. bis. boto-garcía, d., álvarez, a., & pino, j. f. (2021). the effect of euribor on banking profitability: evidence from the spanish banking system. european journal of government and economics, 10(1), 5-29. https://doi.org/10.17979/ejge.2021.10.1.7083 breusch, t. s., & pagan, a. r. (1979). a simple test for heteroscedasticity and random coefficient variation. econometrica, 47, 1287-1294. https://doi.org/10.2307/1911963 carvalho, j. f., & ribeiro, a. (2016). factors explaining the profitability of the banking sector: empirical evidence in portugal. estudos do isca, iv(13), 1-11. https://doi.org/10.34624/ei.v0i13.4729 chukwuogor, c., anoruo, e., & ndu, i. (2021). an empirical analysis of the determinants of the u.s. banks' profitability. banks and bank systems, 16(4), 209-217. https://doi.org/10.21511/bbs.16(4).2021.17 dogan, m., & yildiz, f. (2022). testing the factors that determine the profitability of banks with a dynamic approach: evidence from turkey. theoretical and applied economics, xxix(1), 215-232. https://doi.org/10.2478/jcbtp-2023-0010 european central bank. (2024). overview of monetary policy and markets. monetary policy & markets. https://www.ecb.europa.eu/mopo/html/index.en.html fox, j., & monette, g. (1992). generalized collinearity diagnostics. journal of the american statistical association, 87, 178-183. https://doi.org/10.1080/01621459.1992.10475190 haddad, h., al-qudah, l., almansour, b., & rumman, n. (2019). bank specific and macroeconomic determinants of commercial bank profitability in jordan from 2009–2019. montenegrin journal of economics, 18(4), 155-166. https://doi.org/10.14254/1800-5845/2022.18-4.13 ho, t. h., nguyen, d. t., luu, t. b., le, t. d., & ngo, t. d. (2023). bank performance during the covid-19 pandemic: does income diversification help? journal of applied economics, 26(1), 1-22. https://doi.org/10.1080/15140326.2023.2222964 https://doi.org/10.1080/23311975.2023.2257829 https://doi.org/10.1186/s43093-023-00247-8 https://bpstat.bportugal.pt/ https://bpstat.bportugal.pt/ https://doi.org/10.5089/9781498346498.001 https://doi.org/10.17979/ejge.2021.10.1.7083 https://doi.org/10.2307/1911963 https://doi.org/10.34624/ei.v0i13.4729 https://doi.org/10.21511/bbs.16(4).2021.17 https://doi.org/10.2478/jcbtp-2023-0010 https://www.ecb.europa.eu/mopo/html/index.en.html https://doi.org/10.1080/01621459.1992.10475190 https://doi.org/10.14254/1800-5845/2022.18-4.13 https://doi.org/10.1080/15140326.2023.2222964 134 pires et al. jadah, h. m., alghanimi, m. h., al-dahaan, n. s., & al-husainy, n. h. (2020). internal and external determinants of iraqi bank profitability. banks and bank systems, 15(2), 79-93. https://doi.org/10.21511/bbs.15(2).2020.08 jílková, p., & kotěšovcová, j. (2022). profitability management: how can macro and microeconomic determinants influence commercial bank profitability in the eu-27? tem journal, 11(3), 13001307. https://doi.org/10.18421/tem113-39 kasanaa, e., chauhanb, k., & sahoo, b. p. (2023). policy interest rate and bank profitability—scheduled commercial banks in india. financial markets and banks, 27(1), 138-149. https://doi.org/10.26794/2587-5671-2023-27-1-138-149 koenker, r. (1981). a note on studentizing a test for heteroscedasticity. journal of econometrics, 17(1), 107-112. https://doi.org/10.1016/0304-4076(81)90062-2 kotte, s., reddy, i. l., & bolagani, s. (2022). influence of bank-specific and macroeconomic factors on the profitability of indian commercial banks. the iup journal of applied economics, 21(4), 59-76. kozak, s., & wierzbowska, a. (2022). did the covid-19 pandemic amplify the positive impact of income diversification on the profitability of european banks? equilibrium quarterly journal of economics and economic policy, 17(1), 11-29. https://doi.org/10.24136/eq.2022.001 lamothe, p., delgado, e., solano, m. a., & fernández, s. m. (2024). a global analysis of bank profitability factors. humanities & social sciences communications, 11(124), 1-12. https://doi.org/10.1057/s41599-023-02545-6 lekpek, a., & šabotić, z. (2023). the impact of economic trends on bank profitability: the case of serbia and croatia. teme, xlvii(2), 317-336. https://doi.org/10.22190/teme210614021l martinho, r., oliveira, j., & oliveira, v. (2024). bank profitability and macroeconomic factors. banco de portugal. https://www.bportugal.pt/en/paper/bank-profitability-and-macroeconomic-factors mashamba, t., & chikutuma, c. n. (2023). determinants of bank profitability: evidence from the emerging economy. corporate & business strategy review, 4(4), 310-323. https://doi.org/10.22495/cbsrv4i4siart12 mota, c., silva, e. s., & silva, b. (2019). determinantes da rentabilidade bancária: evidências para os maiores bancos portugueses. european journal of applied business management, 5(2), 78-96. https://doi.org/10.58869/ejabm neves, m. e., proença, c., & dias, a. (2020). bank profitability and efficiency in portugal and spain: a nonlinearity approach. journal of risk and financial management, 13(284), 1-19. https://doi.org/10.3390/jrfm13110284 nunes, p. (2021). the determinants of bank profitability: evidence from portugal [master’s thesis, universidade católica portuguesa]. petria, n. c. (2015). determinants of banks' profitability: evidence from e.u. 27 banking systems. procedia economics and finance, 20, 518-524. https://doi.org/10.1016/s2212-5671(15)00104-5 pires, c., basílio, m., & borralho, c. (2021). determinants of portuguese banks' profitability: an update. tourism & management studies, 17(3), 63-70. https://doi.org/10.18089/tms.2021.170305 puci, j., draci, p., demi, a., & merja, z. (2023). an assessment of bank profitability: evidence from albania. international journal of applied economics, finance and accounting, 16(1), 86-96. https://doi.org/10.33094/ijaefa.v16i1.924 qehaja-keka, v., ahmeti, s., & aliu, m. (2023). bank profitability determinants: evidence from kosovo and albania. journal of liberty and international affairs, 9(2), 85-97. https://doi.org/10.47305/jlia2392370qk shahin, r., khalil, m., & sallam, h. (2022). the nexus between interest rate and bank profitability: does bank prudential capital matter? banks and bank systems, 17(2), 113-123. https://doi.org/10.21511/bbs.17(2).2022.10 son, n. n. (2024). analyze the impact of factors on vietnam banks' net interest margin: a panel data approach. journal of infrastructure, policy and development, 8(3), 1-18. https://doi.org/10.24294/jipd.v8i3.2796 yuan, d., gazi, m. a., harymawan, i., dhar, b. k., & hossain, a. i. (2022). profitability determining factors of banking sector: panel data analysis of commercial banks in south asian countries. frontiers in psychology, 13, 1-17. https://doi.org/10.3389/fpsyg.2022.1000412 https://doi.org/10.21511/bbs.15(2).2020.08 https://doi.org/10.18421/tem113-39 https://doi.org/10.26794/2587-5671-2023-27-1-138-149 https://doi.org/10.1016/0304-4076(81)90062-2 https://doi.org/10.24136/eq.2022.001 https://doi.org/10.1057/s41599-023-02545-6 https://doi.org/10.22190/teme210614021l https://www.bportugal.pt/en/paper/bank-profitability-and-macroeconomic-factors https://doi.org/10.22495/cbsrv4i4siart12 https://doi.org/10.58869/ejabm https://doi.org/10.3390/jrfm13110284 https://doi.org/10.1016/s2212-5671(15)00104-5 https://doi.org/10.18089/tms.2021.170305 https://doi.org/10.33094/ijaefa.v16i1.924 https://doi.org/10.47305/jlia2392370qk https://doi.org/10.21511/bbs.17(2).2022.10 https://doi.org/10.24294/jipd.v8i3.2796 https://doi.org/10.3389/fpsyg.2022.1000412 banks’ profitability in portugal from 2013 to 2023 appendix 1 bank banking product * bpi 1423 bcp 1245 cgd 1465 sicam 675 std 656 mtp 1920 nb 876 bic 1802 big 623 ctt 678 bai 276 fnt 897 ctl 697 bkt 988 db 998 best 487 atl 456 total of banking product 16495 total sample of banking product 16162 * millions of euros source: banco de portugal (2023). the sample used has a weight corresponding to 0.9798 of the total of banking product, concerning the year 2023 and indicated in the previous table. 1. introduction 2. literature review 3. data and methodology 4. results and discussion 5. conclusions references ©the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 1 (2025), pages 118-146 https://doi.org/10.17979/ejge.2025.14.1.11299 submitted: jul 7, 2024 accepted: jan 19, 2025 published: jun 20, 2025 article barriers and policy solutions for electric vehicle adoption in spain: a multidimensional analysis bianca cid,1,* maría teresa garcía-álvarez,1 1 universidade da coruña, elviña, 15071 a coruña, spain *correspondence: bianca.cid@udc.es abstract. in spain, the transport sector is one of the largest contributors to greenhouse gas (ghg) emissions, primarily due to the widespread use of fossil fuels. electric vehicles (evs) are a key component in transitioning towards sustainable mobility and transport decarbonization. the article presents novel insights into the interplay between economic, technical, regulatory, and social factors affecting ev uptake in spain, distinguishing itself from previous studies, using a multidimensional approach. it not only identifies the main challenges but also proposes actionable solutions based on successful international case studies. these include enhancing financial incentives, expanding nationwide charging networks, ensuring consistent regulatory frameworks and promoting public awareness campaigns to dispel misconceptions about evs, among others. by integrating these aspects, the research contributes significantly to the discourse on sustainable transport in spain, aiming to provide a roadmap for policymakers and stakeholders in achieving national climate targets. keywords: electric vehicles; sustainable mobility; greenhouse gas emissions; charging infrastructure; energy transition. jel classification: l62, r48, q42, q48, q54, o33, d12 1. introduction the transport sector in spain has historically been dominated by petroleum-based fuels, particularly diesel and gasoline. this supremacy has contributed to the provision of cheap and reliable transport services. however, it has made transport the largest emitter of greenhouse gas (ghg), the main cause of climate change. in 2022, this sector accounted for an estimated 30.4% of emissions, followed by industrial activities (19.0%), electricity generation (15.2%), agriculture and livestock as a whole (11.0%), fuel consumption in the residential, commercial and institutional sectors (7.9%) and waste management (4.9%). in addition, most sectors experienced an increase compared to the previous year, in the case of transport it was 4.4% (ministerio para la transición ecológica y el reto demográfico, 2023a). in spain, the predominant mode of transport for passengers and freight is road, accounting for more than 90% of journeys (ministerio de transportes y movilidad sostenible, 2024). in 2022, https://creativecommons.org/licenses/by-nc/4.0/ barriers and policy solutions for electric vehicle adoption in spain road transport alone accounted for an estimated 28.1% of total ghg emissions. as illustrated in figure 1, the evolution of ghg emissions in the transport sector has been marked by a steady increase from 1900 until 2007, when emissions peaked. subsequently, there was a decrease, mainly due to the spanish economic crisis. from 2013 onwards, there was again a sustained increase in emissions until 2020, when they fell by 17% due to the mobility restrictions imposed by the covid-19 pandemic. however, in 2021, the previous upward trend returned, with a year-on-year increase of 15% and 2.9% in 2022 (ministerio para la transición ecológica y el reto demográfico, 2023b). figure 1. ghg emissions measured in tonnes of co2 equivalent in the transport sub-sector in spain. source: own elaboration based on data from european environment agency (2024). spain ranks among the top five european union member states with the highest co2 emissions (see figure 2) (european parliament, 2023). this is due to a combination of structural, economic and energy factors that have shaped the country's emissions profile in recent decades. several elements explain why co₂ emissions in spain remain relatively high compared to other european countries. firstly, spain's economic structure has played an important role in shaping its co₂ emissions. the spanish economy has traditionally been dominated by sectors such as transport, construction, and industry, all of which are carbon-intensive. as highlighted above, the transport sector accounts for a significant proportion of co₂ emissions in spain, due to the fossil fuelbased model and the historical dependence on road transport. in addition, the extensive transport infrastructure network, including one of the largest road networks in europe, has facilitated a steady growth in the use of private vehicles, which contributes to keeping emissions high (european court of auditors, 2020). another key factor is the energy matrix. although in recent years there has been a significant increase in renewable energy production, particularly in the wind and solar sectors, the energy transition has progressed gradually. for much of the past decades, spain relied heavily on fossil fuels such as coal, natural gas and oil for electricity generation. this dependence has resulted in persistently high co₂ emissions, especially in comparison with european countries that have achieved a higher penetration of renewable energy and a more rapid reduction in the use of fossil fuels (rosales-asensio et al., 2024). 0.00 10,000,000.00 20,000,000.00 30,000,000.00 40,000,000.00 50,000,000.00 60,000,000.00 70,000,000.00 80,000,000.00 90,000,000.00 100,000,000.00 gh g em is si on s year maritime railway road aerial 120 cid & garcía-álvarez figure 2. total ghg emissions per european country in 2019 measured in kilotonnes of co2 equivalent. source: own elaboration based on data from european parliament (2023). in addition, spain's demographic and urban growth, especially in the last decades of the 20th century and early 21st century, has contributed to the increase in co₂ emissions. urban sprawl, often uncontrolled and characterized by dispersed territorial development, has led to a rise in energy and transport demand, thus exacerbating emissions. spatial and urban planning policies have not always favoured sustainable growth, resulting in carbon-intensive energy consumption patterns. finally, the implementation of climate and energy policies in spain has had a significant impact on co₂ emissions, but their effectiveness has been variable. while notable progress has been made in promoting renewable energies and improving energy efficiency, spanish policy has at times been marked by a lack of continuity and abrupt changes in strategies, which has made it difficult to consolidate a sustained trajectory towards emissions reductions (gonzález et al., 2024). given these circumstances and in line with the positions adopted by the european commission and most member states, spain has set itself the target of achieving ghg emission 0.00 200,000.00 400,000.00 600,000.00 800,000.00 1,000,000.00 germany france italy poland spain netherlands czech republic belgium romania greece austria hungary portugal ireland bulgaria finland sweden denmark slovakia croatia lithuania slovenia estonia latvia luxembourg cyprus malta ghg emissions co un tr y barriers and policy solutions for electric vehicle adoption in spain neutrality by 2050 (ministerio para la transición ecológica y el reto demográfico, 2020). to achieve this goal, a change in the transport sector is needed, as almost one out of every three tonnes of ghg emissions originates from this sector. this transformation necessitates replacing the current fossil fuel-based mobility model with an electric mobility framework. thus, the key to this metamorphosis lies in the mass adoption of evs (usmani & rösler, 2015). evs use one or more electric motors to generate the power needed for propulsion, unlike internal combustion engines that use fossil fuels. one of the primary advantages is that they do not emit exhaust gases at the point of operation (adnan et al., 2016). life cycle assessment is a tool that characterizes such impacts along the life cycle of vehicles. recent life cycle assessment studies consistently demonstrate that evs significantly reduce ghg emissions over their entire life cycle compared to conventional internal combustion engine vehicles powered by gasoline or diesel (knobloch et al., 2020; ajanovic & haas, 2019; moro & lonza, 2018). in addition to being the key player in transport decarbonization, evs are an efficient alternative. internal combustion vehicles have a modest thermodynamic efficiency of around 25% (chernyshev et al., 2017). however, in the case of evs, this figure rises to around 95% (albatayneh et. al., 2020). at the same time, through new technologies such as vehicle-to-grid (v2g) and vehicleto-home (v2h) chargers, evs are a vector for energy storage (izquierdo-monge et al., 2024; niu et al., 2024; kamran, 2022). in this context, this article aims to analyze the current situation of evs in spain, identifying the main barriers that hinder their mass adoption. this issue seems particularly interesting in this member state, due to its peculiarities. through a multidimensional approach, economic, infrastructural, regulatory and social factors that limit the growth of the ev market in spain are examined. furthermore, specific recommendations based on international experiences and case studies are proposed to improve public policies, encourage investment in charging infrastructure and foster greater public awareness of the environmental and economic benefits of this type of sustainable mobility. this article makes a significant contribution to the field of sustainable transportation by providing a thorough examination of the ev landscape in spain, a country that currently lags behind its european counterparts in ev adoption. unlike previous studies, this research uniquely integrates economic, regulatory, and social dimensions to provide a comprehensive understanding of the barriers to ev adoption in spain. the findings reveal critical insights into the complex interplay of these factors, highlighting how they collectively impede the transition to electric mobility. importantly, the article offers practical, evidence-based recommendations from successful international case studies, bridging the gap between theory and practice. by addressing the urgent need for cohesive public policies and targeted investment in charging infrastructure, the article positions itself as a vital resource for policymakers, industry stakeholders and researchers. ultimately, this work lays the groundwork for future studies and actions aimed at accelerating spain's transition to a more sustainable transport system, thereby contributing to broader european climate objectives. this paper is structured in several parts. first, it describes the importance of electric mobility in spain, highlighting its role and impact. next, it analyses the obstacles to the mass adoption of this type of mobility, both at the micro and macro levels. it then reviews existing measures to 122 cid & garcía-álvarez promote ev use and the targets set by the government for the coming years. finally, a series of recommendations is provided to policymakers to overcome the barriers identified at the macro and micro levels to improve policy implementation and promote ev deployment. 2. importance of electric mobility in spain the transport sector is the main cause of ghg emissions in spain, making it the largest source of atmospheric pollution, contributing to climate change (ministerio para la transición ecológica y el reto demográfico, 2023). thus, this is a critical time in which environmental awareness and the pursuit of sustainable solutions are imperative. in this context, electric mobility is a fundamental tool for addressing environmental challenges and transforming our relationship with transport. this shift towards evs is not simply a trend, but an urgent necessity in building a more sustainable future. one of the key pillars of electric mobility is its ability to reduce ghg emissions drastically. traditional fossil fuel-based vehicles are responsible for much air pollution and climate change. in contrast, evs run on electricity, meaning they do not directly emit pollutant gases during operation. however, it is important to emphasize that the environmental benefits of evs depend heavily on the source of electricity used for charging. to maximize their potential for reducing emissions, it is crucial that this electricity comes from renewable sources, as relying on fossil fuels could undermine the sustainability goals of electric mobility (maia et al., 2024; sousa & costa, 2022; longo et al., 2019). the transition to a large ev fleet would significantly reduce our environmental footprint and contribute to climate change mitigation (jeong et al., 2024). electric mobility not only contributes to environmental preservation but also reduces dependence on fossil fuels. in this respect, spain experienced an extreme shift from near selfsufficiency to total dependence. in the 1940s, spain imported only around 10% of the energy it consumed taking into account the important role played at that time by traditional energy sources (biomass, human and animal) a percentage that would increase to around 85% in 1975 and to an astonishing 90% at the beginning of the century (muñoz delgado & rubio vargas, 2015). this significant dependency remains unchanged, with only 25.65% of energy in 2022 sourced from domestic production (statista, 2024). oil prices' volatility and concerns about energy security have underscored the necessity of diversifying energy sources in spain. in this way, mass adoption of evs can reduce dependence on oil imports, improving the trade balance and strengthening energy autonomy (ross morrow et al., 2010). evs, powered by electricity generated from renewable sources, enable such diversification and the transition to a more sustainable and resilient energy system. the transition to electric mobility also presents significant economic opportunities. the manufacture, sale and installation of evs and their associated infrastructure generate employment and stimulate economic growth. thus, in addition to car manufacturers, energy companies and service providers will benefit from this shift as new jobs will be created by developing modernized technologies in different sectors (haddadian et al., 2015). in this context, spain has started to play a leading role, recognizing the importance of adopting evs to drive a significant change in mobility and contribute to building a more environmentally friendly barriers and policy solutions for electric vehicle adoption in spain society. like many other countries, spain faces environmental challenges from rising ghg emissions and air pollution. the transport sector, responsible for the largest contribution of these emissions, is at the centre of attention to achieve a transition towards cleaner and more sustainable mobility. in this regard, electric mobility is the pillar of sustainable and energy-efficient modes of transport (faria et al., 2014). evs offer several environmental, economic and social benefits. firstly, evs do not emit pollutant gases or co2 during operation, which contributes to reducing air pollution and carbon footprint. in addition, the electrification of transport can also help to reduce dependence on fossil fuels, promoting energy autonomy and generating new employment opportunities in the renewable energy sector (will et al., 2024). the analysis of ev development in spain is relevant within the context of the european union (eu) due to several interrelated reasons that underline both the importance of spain as a key player in the european landscape and the particularities of its economic, industrial and geographical situation. firstly, spain is one of the largest and most populated countries in the eu (eurostat, 2024), making its role in the transition to electric mobility crucial for the success of the european climate agenda (ministerio para la transición ecológica y el reto demográfico, 2020). moreover, spain holds a prominent position within the european automotive industry, being the second largest vehicle producer in the eu (international organization of motor vehicle manufacturers, 2024). the transformation of a traditional automotive industry towards the production and adoption of evs is a complex process that involves adjustments in the supply chain, in the production infrastructure and the adoption of new technologies. in this context, a comprehensive analysis of how spain is tackling this process would allow for a broader view of the challenges and opportunities facing the main industrial actors in europe. it is also essential to consider spain´s capacity for renewable energy production, particularly in the wind and solar sectors, which has direct implications for the sustainability of evs (raluy et al., 2021). integrating a growing ev fleet with an electricity grid dominated by renewable sources could maximize the environmental benefits of transport electrification, reducing dependence on fossil fuels (bastida-molina, 2020). spain, thus, serves as a model of how the energy transition and transport electrification can progress in tandem, providing a blueprint that may be replicated or adapted by other eu countries. although the prevalence of evs in spain is experiencing a sustained increase (see figure 3), its sales volume is still deficient. ev registrations grew by 38.6% in 2023 compared to the previous year. however, this growth is insufficient to meet the target set by the national integrated energy and climate plan, which aims for 5.5 million evs in circulation by 2030. currently, there are fewer than half a million evs in circulation (aedive, 2024; ganvam, 2024). 124 cid & garcía-álvarez figure 3. evolution of ev registrations in spain 2009-2023. source: own elaboration based on data from aedive (2024) and ganvam (2024) approximately one out of every 10 passenger cars registered in spain in 2023 was an electrified model, an increase of almost two and a half percentage points compared to the previous year. however, this figure is below the european average where two out of every 10 passenger cars registered were electric and far behind countries such as norway, iceland and sweden. in 2023, the norwegian country remained supreme, with more than 90% of vehicle registrations being electrified models (european automobile manufacturers' association, 2024). as in most european countries, gasoline vehicles remain the dominant choice for consumers in spain, accounting for a market share of 40.8% –35.70% in europe–. although diesel is losing the most presence -registrations in spain fell by almost five percentage points in 2023 compared to the previous year and in europe by nearly three percentage pointsthese two fuels still account for more than half of new registrations in spain (european automobile manufacturers' association, 2024). in europe, transport electrification is proceeding at very uneven rates. in 2022 and 2023, spain ranked 17th in the share of ev registrations (see figure 4). 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 ev r eg is tr at io ns year barriers and policy solutions for electric vehicle adoption in spain figure 4. registrations by energy source in europe. source: own elaboration based on data from the european automobile manufacturers´ association (2024) 22% 27% 26% 41% 39% 46% 38% 54% 22% 26% 31% 25% 22% 27% 24% 32% 35% 31% 35% 44% 22% 32% 10% 12% 56% 60% 22% 22% 56% 60% 89% 90% 26% 30% 51% 53% 23% 24% 23% 23% austria 2022 2023 belgium 2022 2023 denmark 2022 2023 finland 2022 2023 france 2022 2023 germany 2022 2023 ireland 2022 2023 luxembourg 2022 2023 malta 2022 2023 netherlands 2022 2023 portugal 2022 2023 spain 2022 2023 sweden 2022 2023 eu 2022 2023 iceland 2022 2023 norway 2022 2023 switzerland 2022 2023 efta 2022 2023 uk 2022 2023 eu + efta + uk 2022 2023 petrol diesel ev non-plug-in hybrids others 126 cid & garcía-álvarez 3. barriers to ev uptake in spain the transition to electric mobility represents a significant change in the transport industry, but the adoption of evs in spain faces several challenges at both the micro and macro levels. barriers at the micro level: high initial cost. one of the main barriers to electric mobility is the high purchase price (wood & jain, 2020). although battery cost –the most expensive component of the ev propulsion system– has dropped by nearly 90% (berckmans et al., 2017), evs still have a higher initial price than internal combustion vehicles. this represents a significant challenge for consumers who have to make a greater financial effort to purchase a car. over 50% of respondents in various studies (coffman et al., 2017; carley et al., 2013) identified the high initial costs as the primary barrier to the intention to purchase evs. to analyse the price difference, some vehicle models with their electric and internal combustion versions have been selected (see table 1). table 1. retail prices of selected vehicle models in spain. internal combustion engine version electric version peugeot 208 allure 20,950 € 35,000 € peugeot 2008 style 24,989.99 € 36,490 € peugeot 308 style 29,250 € 37,820 € peugeot rifter standard allure 26,873.07 € 32,850 € renault kangoo furgon 18,182 € 28,952 € renault master furgon 28,505 € 49,528 € source: own elaboration based on data from peugeot and renault (2024) evs are consistently higher priced across all models analysed –on average 40% more expensive–. notably, renault master furgon is 73.8% more expensive in its electric version than in its internal combustion version. limited range and range concerns. the limited range of some electric models and concerns about the distance they can travel on a single charge are determining factors for ev purchase (chakraborty et al., 2022; wicki et al., 2022; wood & jain, 2020; canepa et al., 2019; carley et al., 2013). although batteries have improved considerably in capacity and efficiency, the range remains a concern for some drivers. the fear of running out of charge on long journeys becomes a deterrent for many drivers to initiate the transition to electric mobility. 56% of spanish drivers would not buy an electric vehicle because they consider that the range is not sufficient (bnp paribas, 2024). insufficient charging infrastructure. the lack of a well-developed and accessible charging infrastructure is one of the main challenges facing the mass adoption of evs (yang & lin, 2024). the charging infrastructure in spain is still insufficient compared to other european countries (international energy agency, 2024). drivers need the assurance that they will be able to charge their vehicles properly. perceptions and lack of knowledge. user attitudes and perceptions are another barrier to adoption (wicki et al., 2022; singh et al., 2020; adnan et al., 2017). lack of awareness and understanding of the benefits and capabilities of evs are major determinants to be taken into account, as many consumers are still unfamiliar with the technology, the options available and the barriers and policy solutions for electric vehicle adoption in spain benefits they offer. in spain, many consumers are still unaware of key attributes of evs, such as longterm cost savings, environmental advantages and the availability of government incentives (higueras-castillo et al., 2020). uncertainty about batteries. uncertainty about the durability and lifespan of ev batteries is yet another obstacle for consumers. there is concern about the gradual degradation of battery capacity over time and with use, since as batteries are repeatedly discharged and recharged, they may lose energy storage capacity, resulting in reduced range and reduced vehicle performance. spain's warmer climate accelerates battery wear compared to countries with milder conditions, like norway, making it a less favorable environment for battery longevity (geotab, 2024). uncertainty about battery durability is also related to the cost and convenience of battery replacement. consumers fear that battery replacement is costly and may be needed early, which would significantly increase the total cost of ownership of the ev. in addition, the rapid evolution of battery technology may also lead to uncertainty about durability and lifetime, as battery technology advances may result in more efficient and durable models in the future, which could make current evs obsolete more quickly (adu-gyamfi et al., 2024). lack of model variety. the choice of ev models is minimal in size, style and functions compared to the wide range of internal combustion vehicle models, because this market is fully developed (international energy agency, 2023). it is estimated that in 2020 there were approximately 370 internal combustion vehicle models available on the global market, compared to approximately 140 ev models. this represents a ratio of roughly 1 ev model for every 2.6 internal combustion vehicle models (international council on clean transportation, 2022). although the supply of evs is increasing rapidly, it remains significantly lower than that of internal combustion vehicles, which may limit the options for consumers interested in transitioning to cleaner technologies. resistance to change. familiarity, resistance to change and fear of the unknown strongly influence transformation. drivers are used to internal combustion vehicles and may initially be reluctant to switch to new technology (krishna, 2021). 64% of spanish consumers do not consider buying an electric vehicle when renewing their car (alphabet, 2024). macro-level barriers changing policy incentives. political support is a key determinant of ev adoption (wood & jain, 2020; canepa et al., 2019; hardman et al., 2017; jenn et al., 2018). changing decisions on tax incentives, subsidies and regulations can create uncertainty among consumers and manufacturers. according to the empirical results, democracy and the quality of the government have a significant role in the success of environmental policy. implementing and enforcing environmental policies and ensuring they are based on community needs and objectives require democratic governance systems and effective government institutions (dokuzoğlu & güzel, 2024). policy stability is needed to encourage investment and long-term planning in the ev sector. limited public charging infrastructure. despite efforts to increase the number of charging stations, the current network may not be sufficient to support mass ev adoption. significant expansion is needed to ensure that drivers have convenient access to charging in both urban and 128 cid & garcía-álvarez rural settings (bastida-molina et al., 2022). electricity generation capacity. the increase in electricity demand associated with the mass adoption of evs poses challenges regarding electricity generation capacity (gilleran et al., 2021; anastasiadis et al., 2019). for every million evs, the impact on demand in the electricity system is estimated at 2,100 gwh (red eléctrica española, 2018). significant investments in electricity infrastructure are needed to ensure that it is robust and sustainable, especially considering the increasing focus on renewable energy sources (fernández, 2021). waste battery management. as the number of evs in circulation increases, the amount of discarded batteries is expected to increase considerably in the coming years. once ev batteries are degraded to 70–80% of their initial capacity, ev owners must replace them as the residual capacity becomes insufficient for automotive use (haram et al., 2021). however, there is a lack of infrastructure and capacity to recycle these batteries properly. lithium-ion batteries often contain valuable materials such as lithium, cobalt, nickel and aluminium, but without an effective recycling system, these materials can be lost or end up in landfills, representing a significant waste and environmental risk. toxic materials in batteries can leach into soil and water if not properly disposed of, which can have adverse effects on local ecosystems and human health. therefore, efficient recycling policies and systems are needed to address this problem and ensure that the transition to evs does not create new environmental problems (feng et al., 2024). competition with the traditional automotive industry. the transition to evs may face resistance from the traditional automotive industry, which has significant investments in producing internal combustion engine vehicles. ev manufacturing often requires different components and materials compared to conventional cars. policies are needed to encourage research and development and the transition to ev manufacturing. public policymakers can put entrepreneurs at the centre of their interests, but this may not be enough, as the entrepreneurial community is heterogeneous and has different interests (sahiti, 2023). uncertainty about the availability of raw materials. ev battery manufacturing depends on raw materials such as lithium, cobalt and nickel. dependence on these raw materials is a concern, as their supply may be limited, their extraction may have negative environmental impacts, and their price may be volatile due to increasing demand. several studies have analyzed the effects of ev penetration on lithium demand and compared it with data on lithium reserves and resources. some of these analyses also detected a potential premature depletion of current lithium reserves (pehlken et al., 2017; weil & ziemann, 2014; gruber et al., 2011; international energy agency, 2009). thus, uncertainty about the availability and sustainability of these raw materials may affect the scalability of ev production. 4. existing measures to promote the use of evs in spain and the targets set by the government spain has implemented a series of measures to encourage the adoption of evs and move towards cleaner and more efficient mobility. these initiatives reflect the country's commitment to transitioning to more sustainable transport and significantly reducing ghg emissions. barriers and policy solutions for electric vehicle adoption in spain financial assistance. one of the most important measures is implementing financial incentives to purchase evs. these incentives take the form of direct subsidies that reduce the purchase cost, making this option more attractive to consumers. these incentives aim not only to boost demand for evs but also to accelerate the transition to a more environmentally friendly vehicle fleet. in addition to direct subsidies, spain has implemented several tax benefits for ev owners. these include reductions in registration and circulation taxes, as well as tax exemptions in certain areas. these measures not only ease the financial burden for ev owners but also encourage the incorporation of clean technologies into the country's vehicle fleet. for more than a decade, the spanish government has been implementing a series of economic measures to encourage the adoption of evs as part of its transition strategy towards more sustainable mobility. in 2009, spain launched the movele program as one of the first financial support programs for the purchase of evs. this scheme offered direct subsidies to buyers of evs to promote their market penetration and contribute to the reduction of dependence on fossil fuels. in 2012, the pive program was implemented, which not only included incentives for the purchase of evs but also for the renewal of the spanish vehicle fleet, encouraging the retirement of old and polluting vehicles. finally, the moves i, moves ii and moves iii programs are the most relevant in promoting electric mobility in spain. these plans, in force from 2019 to 2024, have earmarked a considerable amount of economic resources to facilitate the acquisition of evs, as well as the installation of charging points and other charging infrastructures. the most important features of these plans are detailed below. the movele program had a budget of 8,000,000,000 € and was aimed at new passenger cars or commercial vehicles with a maximum authorized mass of less than 6,500 kg, motorbikes, heavy quadricycles and minibuses corresponding to evs and plug-in hybrid cars and range-extended evs with a range of more than 32 km. the amount of aid was determined as a percentage of the vehicle price with an absolute limit depending on the category concerned (see table 2). the percentage corresponded to 15% or 20% of the price (before tax), depending on whether the vehicle's technical data were below or above, respectively, the average energy efficiency curve established in the program (ministerio de industria, turismo y comercio, 2009). table 2. aid ceiling of the average energy efficiency curve for the different vehicle categories of the movele program upper limit lower limit passenger cars/commercials 7,000 € 5,000 € motorbikes 1,200 € 750 € heavy quadricycles 3,500 € 2,000 € minibuses 20,000 € 15,000 € source: own elaboration based on data from ministerio de industria, turismo y comercio (2009) the pive program had a budget of 75,000,000 €. it was aimed at new or used passenger cars and light commercial vehicles less than one year old since first registration and corresponding to conventional, hybrid, plug-in hybrid and extended-range electric cars (fully or partially powered by petrol or diesel internal combustion engines and electric), which were classified as class a or b in the 130 cid & garcía-álvarez idae database for fuel consumption and co2 emissions in new cars; pure electric vehicles and those powered by internal combustion engines that can use alternative fossil fuels and that could be certified as having co2 emissions of no more than 160 g/km. the price of the vehicles could not exceed 25,000 €, before vat, except in the case of pure electric, plug-in hybrid and extended-range cars, where this requirement did not apply. in addition, the vehicle to be scrapped had to have been permanently withdrawn from circulation, which had to be at least 12 years old in the case of passenger cars, and at least 10 years old in the case of light commercial vehicles. the total discount to be included in the purchase invoice under the program was at least 2,000 € (ministerio de industria, energía y turismo, 2012). the latest version of the moves program (moves iii), which runs until 31 july 2024, was approved in 2021 with an initial budget of 400,000,000,000 € and was subsequently increased to 1,200,000,000,000 €. the vehicles eligible for incentives are passenger cars, vans, motorbikes and electric quadricycles. the program is managed by the autonomous communities and the amount of aid can be up to 7,000 € (9,000 € with scrapping) for commercial vehicles up to 3,500 kg and up to 4,500 € (7,000 € with scrapping) for passenger cars. these amounts can be increased by 10% if the recipient is a disabled individual with reduced mobility, is registered in a municipality of fewer than 5,000 inhabitants or is a self-employed person using the vehicle for taxi or chauffeur-driven transport services (instituto para la diversificación y ahorro de la energía, 2024). eligible recharging infrastructure can be of any power and be intended for private use in the residential sector, including single-family homes; public use in the non-residential sector (public car parks, hotels, shopping centres, universities, hospitals, industrial estates, sports centres, etc.); private use in parking areas of private and public companies, to provide service to their fleet; public use in parking areas of private and public companies, to provide service to their employees and customers; public use on public roads, urban and interurban roads and public use on the road network, with a special interest in recharging infrastructure at service stations and petrol stations (ministerio para la transición ecológica y el reto demográfico, 2021; ministerio para la transición ecológica y el reto demográfico, 2023c). in addition to the moves iii program, other government initiatives are also being carried out. in june 2023, the ministry of transport, mobility and the urban agenda awarded 500,000,000 € of european next generation recovery funds to 120 municipalities to decarbonize their urban centres and promote sustainable mobility. 20% of the funds are earmarked to encourage bus zeroemission fleets and waste collection vehicles, as well as the charging infrastructure for their operation. this second call of the aid program to municipalities to implement low-emission zones complements the one awarded in 2022, which amounted to 1,000,000,000 € (ministerio de transportes y movilidad sostenible, 2023). two new deductions were also approved in june 2023 to promote the purchase of evs by individuals. in this sense, taxpayers can deduct 15% of the purchase value of the new ev and the amounts paid for installing ev battery charging systems in a property they own until 31 december 2024 (ministerio para la transición ecológica y el reto demográfico, 2023c). the main characteristics of the initiatives carried out by the spanish government to encourage the adoption of evs are summarised in table 3. barriers and policy solutions for electric vehicle adoption in spain table 3. summary table: main contributions of the key regulations and programs promoting ev adoption in spain. regulation/program year main contribution movele program 2009 provided direct subsidies for ev purchases, focusing on passenger cars, motorbikes, and minibuses to promote initial market penetration. pive program 2012 offered incentives for purchasing energy-efficient vehicles and retiring old, polluting vehicles, boosting fleet renewal. moves i, ii, iii program 20192024 allocated significant financial resources to subsidize ev purchases and develop charging infrastructure. moves iii expanded its budget and increased incentives for rural areas and low-income groups. low-emission zones 2022 used european next generation recovery funds to promote zero-emission bus fleets and urban decarbonization, focusing on municipal sustainable mobility projects. tax deductions 2023 allowed taxpayers to deduct part of the costs, including the installation of domestic charging points. source: own elaboration from ministerio de industria, turismo y comercio (2009), ministerio de industria, energía y turismo (2012), ministerio para la transición ecológica y el reto demográfico (2021), ministerio de transportes y movilidad sostenible (2023), ministerio para la transición ecológica y el reto demográfico (2023c) and instituto para la diversificación y ahorro de la energía (2024). norway’s tax strategy has proven to be the most effective incentive policy in europe, driving the country to lead the world in ev market share (european automobile manufacturers' association, 2024). unlike spain’s direct subsidies, norway focused on tax exemptions starting in the early 2000s, including a 100% exemption from vat and registration tax for evs, alongside free tolls, reduced parking fees, and access to bus lanes (mauritzen, 2024; rietmann & lieven, 2019). expansion of recharging points. deploying an efficient charging infrastructure is essential to drive the mass adoption of evs. recharging stations have been installed across urban areas and major roads in spain, facilitating convenient and rapid vehicle recharging (see figure 5). these infrastructure investments not only benefit ev owners but also contribute to economic growth and job creation in the sector (bastida-molina et al., 2022). figure 5. recharging points in spain. source: own elaboration based on data from the european alternative fuels observatory (2024) 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 re ch ar gi ng p oi nt s direct current alternating current 132 cid & garcía-álvarez in spain, at the beginning of 2020, there was one charging point for every 213 km2. by 31 december 2023, one for every 16.54 km2. the number of recharging points has risen substantially in recent years, positioning spain among the leading european countries. however, it still lags behind nations such as the netherlands, germany or france (see figure 6). the following is an analysis of the factors determining the number of recharging points in spain and europe. only 14.29% of the variation in the number of recharging points can be explained by the surface area of european countries (see figure 7). this low coefficient of determination suggests a weak relationship between the geographic size of a country and the extent of its charging infrastructure. larger countries do not necessarily have more recharging points. surface area alone is insufficient to predict recharging point distribution in european countries. therefore, geographic considerations should be viewed as secondary. figure 6. recharging points per country. source: own elaboration based on data from the european alternative fuels observatory (2024). 0.00 20,000.00 40,000.00 60,000.00 80,000.00 100,000.00 120,000.00 140,000.00 160,000.00 n et he rl an ds ge rm an y fr an ce un ite d ki ng do m be lg iu m ita ly sw ed en sp ai n n or w ay de nm ar k au st ri a sw itz er la nd fi nl an d tu rk ey po rt ug al po la nd cz ec h re pu bl ic h un ga ry gr ee ce ro m an ia ir el an d sl ov ak ia lu xe m bo ur g bu lg ar ia ic el an d sl ov en ia li th ua ni a cr oa tia es to ni a la tv ia cy pr us li ec ht en st ei n m al ta re ch ar gi ng p oi nt s barriers and policy solutions for electric vehicle adoption in spain figure 7. ratio between recharging points and surface area (km2) of european countries. source: own elaboration based on data from the european alternative fuels observatory (2024) and eurostat (2024) 64.87% of the variability in the number of recharging points across european countries is explained by gdp per capita (figure 8). this reflects a moderately strong relationship between economic wealth and investment in ev infrastructure, where wealthier countries are better positioned to allocate resources for expanding charging networks. this makes sense since higher gdp per capita typically correlates with greater fiscal capacity to subsidize green technologies, incentivize infrastructure development, and support ev adoption. figure 8. ratio between recharging points and gdp per capita (current prices, billions of euros) of european countries. source: own elaboration based on data from the european alternative fuels observatory (2024) and eurostat (2024). germany belgium spain finland france italy norway netherlands poland united kingdom romania sweden turkey r² = 0.1439 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 0 100,000 200,000 300,000 400,000 500,000 600,000 700,000 800,000 900,000 re ch ar gi ng p oi nt s surface area germany belgium spain france italy norway netherlands poland united kingdom sweden turkey r² = 0.6487 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 3,500,000 re ch ar gi ng p oi nt s gdp per capita 134 cid & garcía-álvarez 75.66% of the variability in the number of recharging points is explained by the number of electric vehicles in circulation (see figure 9). this very high r² suggests a strong direct correlation between the growth of evs and the expansion of charging infrastructure. it is logical that as ev adoption increases, the demand for charging infrastructure rises, compelling both public and private sectors to invest in expanding the network of recharging points. a strong r² like this one indicates that ev demand is the primary driver of infrastructure development. this highlights the interdependency between vehicle adoption and charging facilities, where supply (charging infrastructure) closely follows demand (ev uptake). in conclusion, the number of electric vehicles is the strongest predictor of charging infrastructure among the variables considered. policymakers should thus prioritize boosting ev adoption through incentives and awareness campaigns to simultaneously accelerate infrastructure growth. figure 9. ratio between recharging points and electric cars of european countries. source: own elaboration based on data from the european alternative fuels observatory (2024) and the european automobile manufacturers´ association (2024). only 14.27% of the variability in recharging points across spain's autonomous communities is explained by surface area (see figure 10). this low r² mirrors the result found in european countries, indicating that geographic size is not a significant factor in predicting the number of charging stations. in spain, smaller but highly urbanized regions (like madrid or catalunya) may have far more recharging points than larger, more rural regions. germany belgium spain france italy norway netherlands united kingdom sweden r² = 0.7566 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 180,000 0 500,000 1,000,000 1,500,000 2,000,000 2,500,000 3,000,000 re ch ar gi ng p oi nt s electric cars barriers and policy solutions for electric vehicle adoption in spain figure 10. ratio between recharging points and surface area (km2) of spanish autonomous communities. source: own elaboration based on data from aedive (2024) and instituto geográfico nacional (2024). only 10.04% of the variability in recharging points in spain’s autonomous communities is explained by gdp per capita (see figure 11). this is a surprisingly weak correlation, especially when compared to the european level (see figure 8). it suggests that regional economic wealth is not a strong determinant of charging infrastructure, perhaps due to spain’s centralized approach to ev policy or disparities in how regions allocate resources to ev infrastructure. unlike the european context, gdp per capita is not a strong predictor of ev infrastructure in spain. this points to the importance of local policies and government initiatives over economic factors in driving the growth of recharging points. figure 11. ratio between recharging points and gdp per capita (euros) of spanish autonomous communities. source: own elaboration based on data from aedive (2024) and instituto nacional de estadística (2024). andalucía aragón castilla la mancha castilla y león catalunya comunidad de madrid comunitat valenciana euskadi extremadura galicia r² = 0.1427 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000 re ch ar gi ng p oi nt s surface area andalucía aragón cantabria castilla y león catalunya comunidad de madrid comunidad foral de navarra comunitat valenciana euskadi galicia la rioja r² = 0.1004 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 20,000 22,000 24,000 26,000 28,000 30,000 32,000 34,000 36,000 38,000 40,000 re ch ar gi ng p oi nt s gdp per capita 136 cid & garcía-álvarez 48.1% of the variability in recharging points is explained by the number of electric vehicles in circulation (see figure 12). while this is a strong relationship, it is noticeably lower than the european level (see figure 9). this suggests that while ev adoption is still a significant driver of infrastructure, other factors such as regional policies, government support, or even geographical characteristics could also be at play. the number of electric vehicles remains an important predictor of charging infrastructure at the regional level in spain, but other factors, such as regional policies and public investment strategies, also play a significant role. boosting ev numbers would still help accelerate infrastructure expansion, but supplementary policies are also necessary to address the unexplained variability. figure 12. ratio between recharging points and electric cars in spanish autonomous communities. source: own elaboration based on data from aedive (2024) and ministerio de transportes y movilidad sostenible (2024). it should be noted that for the european countries and the spanish autonomous communities, the variable analyzed that explains a higher percentage of the variability in the number of recharging points is the number of electric cars in circulation. thus, the result of the analysis shows that there is a significant relationship between the number of electric cars in a country and the number of available recharging points. this underscores the importance of strategic planning for ev charging infrastructure. as the ev fleet expands, it is essential to ensure a sufficient number of strategically distributed recharging points to enhance the convenience and accessibility of charging for users. for the ev market to develop, investment in charging infrastructure by governments, businesses and other stakeholders is paramount. the availability of adequate charging infrastructure can, in turn, influence the continued adoption of evs by reducing range anxiety and improving the convenience of charging. andalucía castilla y león catalunya comunidad de madrid comunitat valenciana euskadi r² = 0.481 0 1,000 2,000 3,000 4,000 5,000 6,000 7,000 0 20,000 40,000 60,000 80,000 100,000 120,000 140,000 160,000 re ch ar gi ng p oi nt s electric cars barriers and policy solutions for electric vehicle adoption in spain table 4. ratio between recharging points, surface area, population density and electric cars of european countries. recharging points/ 1,000 km2 recharging points/ 10,000 inhabitants recharging points/ 100 electric cars austria 212.51 19.26 7.80 belgium 1,515.80 39.27 10.82 bulgaria 16.83 2.87 23.50 croatia 21.74 3.16 14.24 cyprus 36.69 3.67 13.66 czech republic 64.08 4.57 14.21 denmark 591.54 41.86 8.98 estonia 21.32 6.73 15.99 finland 43.12 23.59 6.82 france 206.21 19.20 8.20 germany 392.60 16.44 5.61 greece 24.98 3.12 9.52 hungary 40.53 3.85 5.56 iceland 17.29 48.13 6.16 ireland 44.32 5.86 3.15 italy 146.98 7.44 8.64 latvia 9.64 3.24 9.99 liechtenstein 793.75 32.52 20.13 lithuania 22.88 5.02 7.37 luxembourg 959.01 37.53 5.62 malta 322.68 1.86 1.33 netherlands 4,317.68 82.88 20.23 norway 95.39 57.22 3.32 poland 21.45 1.79 7.00 portugal 84.20 7.32 3.73 romania 13.71 1.69 6.56 slovakia 54.29 4.87 17.10 slovenia 86.67 8.25 12.13 spain 64.48 6.74 8.72 sweden 100.43 38.88 6.22 switzerland 381.26 18.16 5.91 turkey 13.91 1.30 12.53 united kingdom 318.79 11.62 5.03 source: own elaboration based on data from the european alternative fuels observatory (2024), eurostat (2024) and the european commission mobility and transport (2024). european countries are at different stages in the transition to electric mobility (see table 4). in the first category, which corresponds to an already developed state, are countries where evs have established a strong presence or have favourable conditions for their adoption shortly. this category includes austria, belgium, denmark, finland, france, germany, luxembourg, the netherlands, norway, sweden and the united kingdom. in the second category are countries with concrete interest in electrification, but where the transition is expected to take longer. this includes greece, hungary, ireland, italy, portugal, switzerland and spain. finally, the emerging category includes 138 cid & garcía-álvarez countries where affordability and availability barriers are difficult to overcome. in the latter category are bulgaria, croatia, czech republic, estonia, latvia, lithuania, poland, romania, slovakia, slovenia and turkey. 5. recommendations to policymakers to overcome the identified barriers based on the results obtained, a series of strategies are proposed that policymakers could follow to overcome the obstacles detected, both at the micro and macro levels. financial incentives and subsidies. implement stable government policies that offer financial incentives and subsidies for the purchase of evs to reduce the high initial cost barrier. this would be based on approving a new moves plan that would speed up and simplify its processing, advance the amounts to drivers, and establish criteria to favour the most disadvantaged groups so that individuals with lower incomes obtain higher amounts of aid. at the same time, a system of tax deductions could be established, such as reduced vat, which would be applied directly to the purchase price. expand freight infrastructure. make significant investments in expanding and improving public and private charging infrastructure to ensure an efficient and accessible nationwide network. as the demand for electric mobility increases, it is critical to have an extensive and accessible network of recharging points that meet users' needs and eliminate concerns about vehicle range. to achieve this goal, several comprehensive strategies need to be implemented: — prioritise installing recharging points in strategic locations, such as public car parks, commercial areas, service stations and densely populated residential areas. these public recharging points should offer various charging options, including fast charging and slow charging, to meet the different needs of users. — encourage public-private partnerships to accelerate the expansion of charging infrastructure. this may include incentives for private companies to invest in the installation of recharging points, as well as the simplification of administrative and regulatory procedures related to their installation and operation. — integrating charging infrastructure into urban planning is essential to ensure accessibility and convenience. local governments should consider including charging requirements in building codes and urban planning, as well as reserving space for installing recharging points in new construction and retrofitting. — encourage the installation of recharging points in homes and residential communities. this can be achieved through tax incentives or subsidies for the installation of residential recharging points, as well as the implementation of policies that facilitate their installation in residential buildings and community garages. up-to-date information on recharging infrastructure. a continuously updated publication of publicly available recharging points should be created. this can be done by creating centralized online platforms, such as mobile applications and websites, that provide detailed and up-to-date information on the location, availability, connector type, tariffs, and operating hours of ev recharging points across the country. these platforms can be managed by governmental entities, private barriers and policy solutions for electric vehicle adoption in spain companies, or non-profit organizations and should be easily accessible to users. collaborative mechanisms must be established between government entities, utilities, ev manufacturers, recharging point operators, and other relevant actors to collect and share up-to-date data on charging infrastructure. this could include creating shared databases and implementing information exchange standards. processes and procedures should be implemented to ensure continuous updating of data on the charging infrastructure, including regular verification of the availability and operational conditions of recharging points, as well as the addition of new recharging points as they are installed. user feedback on the charging infrastructure should also be facilitated through online platforms, allowing users to report on the availability, operational status and quality of charging services. this feedback can help identify problems and areas for improvement in the recharging infrastructure. in this regard, it is essential to implement notification and alert systems to inform users about changes in the availability or conditions of recharging points, as well as to provide real-time updates on events such as temporary closures or scheduled maintenance. awareness and education campaigns. launch campaigns to raise awareness and educate about the benefits of evs, addressing common concerns, and providing accurate and objective information. although ev technology has advanced significantly recently, a lack of public knowledge and understanding hinders its acceptance and adoption. therefore, it is essential to implement effective education and awareness strategies to inform citizens about the benefits of electric mobility and dispel common myths and concerns. to this end, awareness campaigns should be conducted at the national and local levels to highlight the environmental, economic and social benefits of evs. these campaigns can include media advertisements, community events, educational seminars, and collaboration with opinion leaders and celebrities to amplify the message. it is essential to provide clear, accessible and transparent information about evs, including their operation, costs, benefits and availability. this can be done through government websites, information brochures, outreach events and briefings at car dealerships. also, organizing ev testing and demonstration programs allows consumers to experience the technology and understand its benefits directly. these programs can include test-driving events, demonstrations at trade fairs and exhibitions, and collaborations with car rental companies to offer ev rental options. it is also important to provide education and training to automotive industry professionals, such as salespeople, mechanics, and technicians so that they are better prepared to respond to customer questions and needs regarding evs. this can include certification programs, training courses and hands-on workshops. improving battery technology. continue to invest in research and development to improve battery technology, increase range and reduce long-term costs. increasing the range of evs is essential to eliminate concerns about "range anxiety" and increase consumer acceptance. continued research and development are required to improve battery energy density and extend charge life, allowing evs to travel longer distances on a single charge. this includes exploring new materials for electrodes and electrolytes, as well as designing more efficient batteries. it is also important to establish efficient battery recycling systems to reduce waste and promote the reuse of materials to mitigate the environmental impacts associated with battery production and disposal while ensuring the availability of key materials to manufacture new batteries. thus, research incentives are essential, implementing tax incentives and funding programs for companies and institutions 140 cid & garcía-álvarez conducting research and development in battery technologies. this could include tax credits, grants for research and development projects, and the creation of centres of excellence in the battery field. develop innovative business models. explore new ways to offer and monetize services related to electric mobility, such as creating ev-sharing platforms that allow users to rent evs for short periods and facilitating flexible and sustainable mobility in urban environments. another alternative would be to introduce subscription and pay-as-you-go models for evs, where users pay a fixed monthly fee or for the actual use of the vehicle, rather than having to purchase the car traditionally. this can make ev adoption more accessible and affordable for many users. variety of models and options. stimulate the variety of ev models available by providing consumers with more options to suit their needs and preferences. this can be done by implementing fiscal and financial incentives for ev manufacturers and technology providers, such as tax credits, research and development subsidies, and preferential financing for the production of evs and related components, as well as by simplifying and streamlining the approval and certification processes for evs, which would facilitate the introduction of new models to the market and reduce the associated costs for manufacturers. development of electricity generation capacity. invest in electricity generation capacity to meet the growing demand associated with electric mobility, prioritizing renewable energy sources to ensure sustainability. the electricity grid needs to be modernized and strengthened to ensure its capacity to support the additional load from evs. this would include upgrading the transmission and distribution infrastructure, installing advanced monitoring and control equipment, and integrating smart load management technologies. it is also important to develop a smart charging infrastructure that enables efficient electricity demand management, especially during peak load periods when higher energy demand is expected due to ev charging. this would require implementing scheduled charging systems, dynamic tariffs and energy storage technologies. unified legislation and international collaboration. actively engage in global collaboration and initiatives that promote common standards, share best practices and facilitate technology transfer. set clear and ambitious targets for ev market penetration, as well as regulations that encourage the manufacture and marketing of evs, such as stricter emission standards and severe restrictions on internal combustion vehicles in urban areas. it is also necessary to work on harmonising ev-related regulations and standards at the international level, which would facilitate the manufacturing, approval and marketing of evs in different markets, reducing the complexity and associated costs for manufacturers and consumers. finally, it is essential to facilitate the exchange of practices and experiences with countries leading the way in ev adoption, such as norway, iceland and sweden. this may include the organization of conferences, seminars and workshops, as well as establishing networks and collaboration platforms. 6. conclusions spain faces significant environmental challenges, from rising temperatures to biodiversity loss. these problems are exacerbated by ghg emissions, mostly from the transport sector. electric mobility emerges as a key solution, as evs contribute to directly reducing pollutant emissions and barriers and policy solutions for electric vehicle adoption in spain improving air quality in cities. electric mobility goes beyond reducing emissions. in addition, evs provide higher energy efficiency, reduced noise pollution and lower maintenance costs. moreover, by relying on electricity, they open up opportunities to harness renewable energy sources, such as solar and wind, thus strengthening the country's energy resilience. the transition to evs takes place in the context of a transport system in spain that has historically been dominated by fossil fuels, mainly petrol and diesel. these have fuelled the majority of vehicles, which has led to the transport sector being the main ghg emitter in the country. the sustained growth of ghg emissions has put transport at the centre of the climate debate, making evident the need for structural change to decarbonize the sector. electric mobility appears to be one of the most promising solutions to reduce emissions, not only because of evs' energy efficiency but also because of their ability to operate without directly emitting polluting gases. however, the contextual analysis reveals that spain lags behind other european countries in ev adoption, with penetration levels below the european average and far behind leaders such as norway and the netherlands. the main objective of this paper is to analyze, from a multidimensional approach, the barriers that hinder the mass adoption of evs in spain. furthermore, it has been proposed that a diagnosis of existing policies and measures be carried out to promote the use of evs, as well as to offer specific recommendations based on international experiences and case studies. the final objective is to identify best practices to guide policymakers and private sector agents in implementing solutions to facilitate the transition towards electric mobility. this article presents a novel perspective on the adoption of evs in spain by offering a comprehensive analysis that integrates multiple dimensions of the challenges faced in this transition. unlike previous studies that may focus predominantly on singular aspects of ev adoption, this research uniquely identifies the interconnected barriers that impede progress. additionally, it introduces innovative, evidence-based solutions drawn from successful international case studies, emphasizing the need for cohesive public policies tailored to the spanish context. by addressing these dimensions holistically, this article not only contributes to the existing body of knowledge but also provides a roadmap for stakeholders and policymakers to accelerate the transition towards sustainable mobility in spain, thus filling a critical gap in the literature. the analysis results confirm that, although ev registrations have shown steady growth in recent years, this pace is insufficient to meet the ambitious targets of the national integrated energy and climate plan. the adoption of evs faces several common challenges globally, such as limited driving range, public scepticism toward ev technology and high initial costs. the latter remains one of the most important problems. the elevated upfront cost of evs, despite the decline in battery prices, continues to be a concern for consumers, particularly when compared to internal combustion vehicles, which are more cost-effective in the short term. however, the spanish case presents unique specificities that set it apart from the general landscape of ev adoption. one of the factors behind this gap is the lack of adequate infrastructure. although the number of charging points has increased significantly, spain still lags behind other european countries in terms of the availability and accessibility of charging stations. this infrastructure deficit is of particular concern in rural and some urban areas, where range anxiety 142 cid & garcía-álvarez remains a major barrier for potential ev users. policy inconsistency further complicates spain’s transition to electric mobility. unlike countries such as norway, which have implemented stable, long-term incentive strategies, spain has experienced frequent shifts in its subsidy programs and regulatory frameworks. these abrupt changes create uncertainty for consumers and investors, hindering the sustained growth of ev adoption. based on the findings, recommendations are made for policymakers and key stakeholders, focusing on improving financial incentives, developing a robust charging infrastructure, awarenessraising and education campaigns, research and development in battery technology, regulatory stability, and international collaboration. thus, the transition to evs in spain presents several significant challenges that require coordinated action by all actors involved: governments, the automotive industry, energy companies and citizens. the potential of electric mobility to reduce ghg emissions, diversify energy sources and stimulate economic growth is extraordinary. however, for spain to achieve its climate and sustainability goals, it is crucial that ambitious and coherent policies are implemented, and that adequate infrastructure is fostered to remove barriers to mass ev adoption. with the right measures, the country can position itself as a leader in the transition to sustainable mobility and set an example for other developing markets. references adnan, n., nordin, s. m., rahman, i., vasant, p., & noor, m. a. (2017). an integrative approach to study on consumer behavior towards plug-in hybrid electric vehicles revolution: consumer behavior towards plug-in hybrid electric vehicles. en r. ianole (ed.), applied behavioral economics research and trends (pp. 183–213). hershey, pa: igi global. https://doi.org/10.4018/978-1-5225-18266.ch010 adnan, n., nordin, s. m., rahman, i., vasant, p. m., & noor, m. a. (2016). a comprehensive review on theoretical framework-based electric vehicle consumer adoption research. international journal of energy research, 41(3), 317–335. https://doi.org/10.1002/er.3640 adu-gyamfi, g., song, h., nketiah, e., obuobi, b., & djanie, a. k. (2024). remedying driving range and refueling challenges in electric mobility: consumer adoption of battery-swappable electric vehicles. technology in society, 78, article 102661. https://doi.org/10.1016/j.techsoc.2024.102661 aedive. (2024, enero 2). las matriculaciones de vehículos electrificados cierran 2023 con una subida de más del 38%, rozando las 140.000 unidades. https://aedive.es/matriculaciones-vehiculoselectrificados-cierran-2023-subida-de-mas-del-38-rozando-140-000-unidades/ ajanovic, a., & haas, r. (2019). economic and environmental prospects for battery electricand fuel cell vehicles: a review. fuel cells, 19(5), 515–529. https://doi.org/10.1002/fuce.201800171 albatayneh, a., assaf, m. n., alterman, d., & jaradat, m. (2020). comparison of the overall energy efficiency for internal combustion engine vehicles and electric vehicles. environmental and climate technologies, 24(1), 669–680. https://doi.org/10.2478/rtuect-2020-0041 alphabet. (2024). radiografía de los hábitos de movilidad de los españoles. https://www.alphabet.com/eses/foro-de-movilidad-2024 anastasiadis, a. g., kondylis, g. p., polyzakis, a., & vokas, g. (2019). effects of increased electric vehicles into a distribution network. energy procedia, 157, 586–593. https://doi.org/10.1016/j.egypro.2018.11.223 bastida-molina, p., ribó-pérez, d., gómez-navarro, t., & hurtado-pérez, e. (2022). what is the problem? the obstacles to the electrification of urban mobility in mediterranean cities: case study of valencia, spain. renewable and sustainable energy reviews, 166, article 112649. https://doi.org/10.1016/j.rser.2022.112649 https://doi.org/10.4018/978-1-5225-1826-6.ch010 https://doi.org/10.4018/978-1-5225-1826-6.ch010 https://doi.org/10.1002/er.3640 https://doi.org/10.1016/j.techsoc.2024.102661 https://aedive.es/matriculaciones-vehiculos-electrificados-cierran-2023-subida-de-mas-del-38-rozando-140-000-unidades/ https://aedive.es/matriculaciones-vehiculos-electrificados-cierran-2023-subida-de-mas-del-38-rozando-140-000-unidades/ https://doi.org/10.1002/fuce.201800171 https://doi.org/10.2478/rtuect-2020-0041 https://www.alphabet.com/es-es/foro-de-movilidad-2024 https://www.alphabet.com/es-es/foro-de-movilidad-2024 https://doi.org/10.1016/j.egypro.2018.11.223 https://doi.org/10.1016/j.rser.2022.112649 barriers and policy solutions for electric vehicle adoption in spain bastida-molina, p., hurtado-pérez, e., peñalvo-lópez, e., & moros-gómez, m. c. (2020). assessing transport emissions reduction while increasing electric vehicles and renewable generation levels. transportation research part d: transport and environment, 88, article 102560. https://doi.org/10.1016/j.trd.2020.102560 berckmans, g., messagie, m., smekens, j., omar, n., vanhaverbeke, l., & van mierlo, j. (2017). cost projection of state of the art lithium-ion batteries for electric vehicles up to 2030. energies, 10, article 1314. https://doi.org/10.3390/en10091314 bnp paribas. (2024, marzo 21). estudio motor 2024. observatorio cetelem. https://elobservatoriocetelem.es/motor/estudio-motor-2024/ canepa, k., hardman, s., & tal, g. (2019). an early look at plug-in electric vehicle adoption in disadvantaged communities in california. transport policy, 78, 19–30. https://doi.org/10.1016/j.tranpol.2019.03.009 carley, s., krause, r. m., lane, b. w., & graham, j. d. (2013). intent to purchase a plug-in electric vehicle: a survey of early impressions in large us cities. transportation research part d: transport and environment, 18, 39–45. https://doi.org/10.1016/j.trd.2012.09.007 chakraborty, d., bunch, d. s., brownstone, d., xu, b., & tal, g. (2022). plug-in electric vehicle diffusion in california: role of exposure to new technology at home and work. transportation research part a: policy and practice, 156, 133–151. https://doi.org/10.1016/j.tra.2021.12.005 chernyshev, a. d., lisovskaya, t. a., & lisovskiy, r. a. (2017). comparative analysis of different electrical motor types as a traction drive part in electrical transmission. en 2017 international conference on industrial engineering, applications and manufacturing (icieam) (pp. 1–5). ieee. https://doi.org/10.1109/icieam.2017.8076311 coffman, m., bernstein, p., & wee, s. (2017). electric vehicles revisited: a review of factors that affect adoption. transport reviews, 37(1), 79–93. https://doi.org/10.1080/01441647.2016.1217282 dokuzoğlu, s., & güzel, a. e. (2024). democracy, governance, and environmental policy effectiveness: a cross-country analysis with sustainable governance indicators. european journal of government and economics, 13(1), 103–119. https://doi.org/10.17979/ejge.2024.13.1.9902 european alternative fuels observatory. (2024). infrastructure. https://alternative-fuelsobservatory.ec.europa.eu/transport-mode/road/spain/infrastructure european automobile manufacturers’ association. (2024, febrero 6). new car registrations: +13.9% in 2023; battery electric 14.6% market share. https://www.acea.auto/pc-registrations/new-carregistrations-13-9-in-2023-battery-electric-14-6-market-share european commission mobility and transport. (2024). electric vehicles. https://transport.ec.europa.eu/transport-themes/clean-transport/clean-and-energy-efficientvehicles/green-propulsion-transport/electric-vehicles_en european court of auditors. (2020). red básica de carreteras de la ue: se han reducido los tiempos de desplazamiento, pero todavía no es plenamente funcional. https://www.eca.europa.eu/lists/ecadocuments/sr20_09/sr_road_network_es.pdf european environment agency. (2024, abril 15). eea greenhouse gases-data viewer. https://www.eea.europa.eu/en/analysis/maps-and-charts/greenhouse-gases-viewer-dataviewers european parliament. (2023, marzo 23). emisiones de gases de efecto invernadero por país y sector. https://www.europarl.europa.eu/pdfs/news/expert/2018/3/story/20180301sto98928/201 80301sto98928_es.pdf eurostat. (2024). demography of europe – 2024 edition. https://ec.europa.eu/eurostat/web/interactivepublications/demography-2024 eurostat. (2024). regions in europe – 2024 edition. https://ec.europa.eu/eurostat/web/interactivepublications/regions-2024#gross-domestic-product eurostat. (2024). europe database. https://ec.europa.eu/eurostat/data/database faria, m. v., baptista, p. c., & farias, t. l. (2014). electric vehicle parking in european and american context: economic, energy and environmental analysis. transportation research part a: policy and practice, 64, 110–121. https://doi.org/10.1016/j.tra.2014.03.011 feng, j., guo, p., & xu, g. (2024). barriers to electric vehicle battery recycling in a circular economy: an interpretive structural modeling. journal of cleaner production, 469, article 143224. https://doi.org/10.1016/j.jclepro.2024.143224 fernández, r. a. (2021). stochastic analysis of future scenarios for battery electric vehicle deployment and the upgrade of the electricity generation system in spain. journal of cleaner production, 316, article 128101. https://doi.org/10.1016/j.jclepro.2021.128101 ganvam asociación nacional de vendedores de vehículos. (2024, julio 5). las matriculaciones de vehículos electrificados cierran 2023 con una subida de más del 38%, rozando las 140.000 unidades. https://ganvam.es/las-matriculaciones-de-vehiculos-electrificados-cierran-2023-con-unasubida-de-mas-del-38-rozando-las-140-000-unidades/ https://doi.org/10.1016/j.trd.2020.102560 https://doi.org/10.3390/en10091314 https://elobservatoriocetelem.es/motor/estudio-motor-2024/ https://doi.org/10.1016/j.tranpol.2019.03.009 https://doi.org/10.1016/j.trd.2012.09.007 https://doi.org/10.1016/j.tra.2021.12.005 https://doi.org/10.1080/01441647.2016.1217282 https://doi.org/10.17979/ejge.2024.13.1.9902 https://alternative-fuels-observatory.ec.europa.eu/transport-mode/road/spain/infrastructure https://alternative-fuels-observatory.ec.europa.eu/transport-mode/road/spain/infrastructure https://www.acea.auto/pc-registrations/new-car-registrations-13-9-in-2023-battery-electric-14-6-market-share https://www.acea.auto/pc-registrations/new-car-registrations-13-9-in-2023-battery-electric-14-6-market-share https://transport.ec.europa.eu/transport-themes/clean-transport/clean-and-energy-efficient-vehicles/green-propulsion-transport/electric-vehicles_en https://transport.ec.europa.eu/transport-themes/clean-transport/clean-and-energy-efficient-vehicles/green-propulsion-transport/electric-vehicles_en https://www.eca.europa.eu/lists/ecadocuments/sr20_09/sr_road_network_es.pdf https://www.eea.europa.eu/en/analysis/maps-and-charts/greenhouse-gases-viewer-data-viewers https://www.eea.europa.eu/en/analysis/maps-and-charts/greenhouse-gases-viewer-data-viewers https://www.europarl.europa.eu/pdfs/news/expert/2018/3/story/20180301sto98928/20180301sto98928_es.pdf https://www.europarl.europa.eu/pdfs/news/expert/2018/3/story/20180301sto98928/20180301sto98928_es.pdf https://ec.europa.eu/eurostat/web/interactive-publications/demography-2024 https://ec.europa.eu/eurostat/web/interactive-publications/demography-2024 https://ec.europa.eu/eurostat/web/interactive-publications/regions-2024#gross-domestic-product https://ec.europa.eu/eurostat/web/interactive-publications/regions-2024#gross-domestic-product https://ec.europa.eu/eurostat/data/database https://doi.org/10.1016/j.tra.2014.03.011 https://doi.org/10.1016/j.jclepro.2024.143224 https://doi.org/10.1016/j.jclepro.2021.128101 https://ganvam.es/las-matriculaciones-de-vehiculos-electrificados-cierran-2023-con-una-subida-de-mas-del-38-rozando-las-140-000-unidades/ https://ganvam.es/las-matriculaciones-de-vehiculos-electrificados-cierran-2023-con-una-subida-de-mas-del-38-rozando-las-140-000-unidades/ 144 cid & garcía-álvarez geotab. (2024). how long do electric car batteries last? what 10,000 electric vehicles tell us about ev battery life. https://www.geotab.com/blog/ev-battery-health/ gilleran, m., bonnema, e., woods, j., mishra, p., doebber, i., hunter, c., mitchell, m., & mann, m. (2021). impact of electric vehicle charging on the power demand of retail buildings. advances in applied energy, 4, article 100062. https://doi.org/10.1016/j.adapen.2021.100062 gonzález, p. f., presno, m. j., & landajo, m. (2024). tracking the change in spanish greenhouse gas emissions through an lmdi decomposition model: a global and sectoral approach. journal of environmental sciences, 139, 114–122. https://doi.org/10.1016/j.jes.2022.08.027 gruber, p. w., medina, p. a., keoleian, g. a., kesler, s. e., everson, m. p., & wallington, t. j. (2011). global lithium availability: a constraint for electric vehicles? journal of industrial ecology, 15(5), 760– 775. https://doi.org/10.1111/j.1530-9290.2011.00359.x haddadian, g., khodayar, m., & shahidehpour, m. (2015). accelerating the global adoption of electric vehicles: barriers and drivers. the electricity journal, 28(10), 53–68. https://doi.org/10.1016/j.tej.2015.11.011 haram, m. h. s. m., lee, j. w., ramasamy, g., ngu, e. e., thiagarajah, s. p., & lee, y. h. (2021). feasibility of utilising second life ev batteries: applications, lifespan, economics, environmental impact, assessment, and challenges. alexandria engineering journal, 60(5), 4517–4536. https://doi.org/10.1016/j.aej.2021.03.021 hardman, s., chandan, a., tal, g., & turrentine, t. s. (2017). the effectiveness of financial purchase incentives for battery electric vehicles – a review of the evidence. renewable and sustainable energy reviews, 80, 1100–1111. https://doi.org/10.1016/j.rser.2017.05.255 higueras-castillo, e., molinillo, s., coca-stefaniak, j. a., & liébana-cabanillas, f. (2020). potential early adopters of hybrid and electric vehicles in spain—towards a customer profile. sustainability, 12(11), article 4345. https://doi.org/10.3390/su12114345 instituto geográfico nacional. (2024). geoportal oficial del instituto geográfico nacional de españa. https://www.ign.es/web/ane-datos-geograficos/-/datosgeograficos/datospoblacion?tipobusqueda=ccaa instituto nacional de estadística. (2024). cuentas económicas. contabilidad regional de españa. https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736167628& menu=ultidatos&idp=1254735576581 international council on clean transportation. (2022). a global comparison of the life-cycle greenhouse gas emissions of combustion engine and electric passenger cars. https://theicct.org/publication/aglobal-comparison-of-the-life-cycle-greenhouse-gas-emissions-of-combustion-engine-andelectric-passenger-cars/ international energy agency. (2024). global ev outlook 2024. https://www.iea.org/reports/global-evoutlook-2024 international energy agency. (2023). global ev outlook 2023. https://www.iea.org/reports/global-evoutlook-2023 international energy agency. (2009). energy technology transitions for industry: strategies for the next industrial revolution. oecd publishing. instituto para la diversificación y ahorro de la energía. (2024). programa moves iii. movilidad eficiente y sostenible. https://www.idae.es/ayudas-y-financiacion/para-movilidad-y-vehiculos/programamoves-iii international organization of motor vehicle manufacturers. (2024). world motor vehicle production by country/region and type. https://www.oica.net/wp-content/uploads/by-country-region2023.pdf jenn, a., springel, k., & gopal, a. r. (2018). effectiveness of electric vehicle incentives in the united states. energy policy, 119, 349–356. https://doi.org/10.1016/j.enpol.2018.04.065 izquierdo-monge, o., bonilla, a. z. v., lafuente-cacho, m., peña-carro, p., & hernández-jiménez, á. (2024). heuristic method for electric vehicle charging in a spanish microgrid: leveraging renewable energy surplus. journal of power sources, 629, article 235945. https://doi.org/10.1016/j.jpowsour.2024.235945 jeong, j. w., lee, d. i., woo, s., lim, y., & lee, k. (2024). analysis of energy consumption efficiency and emissions according to urban driving of hybrid electric vehicles in korea. applied energy, 371, article 123686. https://doi.org/10.1016/j.apenergy.2024.123686 kamran, m. (2022). electric vehicles and smart grids. in fundamentals of smart grid systems (pp. 431460). london: academic press. https://doi.org/10.1016/b978-0-323-99560-3.00002-8 knobloch, f., hanssen, s., lam, a., pollitt, h., salas, p., chewpreecha, u., huijbregts, m. a. j., & mercure, j. f. (2020). net emission reductions from electric cars and heat pumps in 59 world regions over time. nature sustainability, 3, 437–447. https://doi.org/10.1038/s41893-020-0488-7 https://www.geotab.com/blog/ev-battery-health/ https://doi.org/10.1016/j.adapen.2021.100062 https://doi.org/10.1016/j.jes.2022.08.027 https://doi.org/10.1111/j.1530-9290.2011.00359.x https://doi.org/10.1016/j.tej.2015.11.011 https://doi.org/10.1016/j.aej.2021.03.021 https://doi.org/10.1016/j.rser.2017.05.255 https://doi.org/10.3390/su12114345 https://www.ign.es/web/ane-datos-geograficos/-/datos-geograficos/datospoblacion?tipobusqueda=ccaa https://www.ign.es/web/ane-datos-geograficos/-/datos-geograficos/datospoblacion?tipobusqueda=ccaa https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736167628&menu=ultidatos&idp=1254735576581 https://www.ine.es/dyngs/inebase/es/operacion.htm?c=estadistica_c&cid=1254736167628&menu=ultidatos&idp=1254735576581 https://theicct.org/publication/a-global-comparison-of-the-life-cycle-greenhouse-gas-emissions-of-combustion-engine-and-electric-passenger-cars/ https://theicct.org/publication/a-global-comparison-of-the-life-cycle-greenhouse-gas-emissions-of-combustion-engine-and-electric-passenger-cars/ https://theicct.org/publication/a-global-comparison-of-the-life-cycle-greenhouse-gas-emissions-of-combustion-engine-and-electric-passenger-cars/ https://www.iea.org/reports/global-ev-outlook-2024 https://www.iea.org/reports/global-ev-outlook-2024 https://www.iea.org/reports/global-ev-outlook-2023 https://www.iea.org/reports/global-ev-outlook-2023 https://www.idae.es/ayudas-y-financiacion/para-movilidad-y-vehiculos/programa-moves-iii https://www.idae.es/ayudas-y-financiacion/para-movilidad-y-vehiculos/programa-moves-iii https://www.oica.net/wp-content/uploads/by-country-region-2023.pdf https://www.oica.net/wp-content/uploads/by-country-region-2023.pdf https://doi.org/10.1016/j.enpol.2018.04.065 https://doi.org/10.1016/j.jpowsour.2024.235945 https://doi.org/10.1016/j.apenergy.2024.123686 https://doi.org/10.1016/b978-0-323-99560-3.00002-8 https://doi.org/10.1038/s41893-020-0488-7 barriers and policy solutions for electric vehicle adoption in spain krishna, g. (2021). understanding and identifying barriers to electric vehicle adoption through thematic analysis. transportation research interdisciplinary perspectives, 10, article 100364. https://doi.org/10.1016/j.trip.2021.100364 longo, m., foiadelli, f., & yaïci, w. (2019). electric vehicles integrated with renewable energy sources for sustainable mobility. in l. romeral martínez & m. delgado prieto (eds.), new trends in electrical vehicle powertrains (pp. 203-223). london: intechopen. https://doi.org/10.5772/intechopen.76788 maia, f., leitão, s., & marques, m. c. (2024). energy transition in portugal: the harnessing of solar photovoltaics in electric mobility and its impact on the carbon footprint. journal of cleaner production, 477, article 143834. https://doi.org/10.1016/j.jclepro.2024.143834 mauritzen, j. (2024). adoption of chinese cars and the shift to electric vehicles: early evidence from norway. transport policy, 158, 1–12. https://doi.org/10.1016/j.tranpol.2024.09.012 ministerio de industria, energía y turismo. (2012). resolución de 28 de septiembre de 2012, de la secretaría de estado de energía, por la que se publica la de 24 de septiembre de 2012, del consejo de administración del instituto para la diversificación y ahorro de la energía, por la que se establecen las bases reguladoras para la convocatoria de ayudas del programa de incentivos al vehículo eficiente. boletín oficial del estado, núm. 235, 29-09-2012. https://www.boe.es/buscar/doc.php?id=boe-a-2012-12216 ministerio de industria, energía y turismo. (2009). resolución de 26 de junio de 2009, del instituto para la diversificación y ahorro de la energía, por la que se establecen las bases reguladoras para la convocatoria única del programa de ayudas del idae para la adquisición y uso de vehículos eléctricos, en el marco del proyecto piloto de movilidad eléctrica, dentro del plan de activación del ahorro y la eficiencia energética 2008-2011 y del plan de acción 2008-2012 de la estrategia de ahorro y eficiencia energética en españa 2004-2012. boletín oficial del estado, núm. 160, 03-072009. https://www.boe.es/diario_boe/txt.php?id=boe-a-2009-11022 ministerio de transportes y movilidad sostenible. (2024). observatorio del transporte y la logística en españa. https://otle.transportes.gob.es/movilidad ministerio de transportes y movilidad sostenible. (2023). mitma adjudica 500 millones de euros a 120 municipios para descarbonizar sus núcleos urbanos y fomentar la movilidad sostenible. estrategia de movilidad segura, sostenible y conectada 2030. https://esmovilidad.transportes.gob.es/noticias/mitma-adjudica-500-millones-de-euros-120municipios-para-descarbonizar-sus-nucleos-urbanos ministerio para la transición ecológica y el reto demográfico. (2023a). nota informativa sobre el avance de las emisiones de gases de efecto invernadero correspondientes al año 2022. https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacionambiental/temas/sistema-espanol-de-inventario-sei-/avance_gei_2022.pdf ministerio para la transición ecológica y el reto demográfico. (2023b). informe de inventario nacional de emisiones de gases de efecto invernadero. https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacionambiental/temas/sistema-espanol-de-inventario-sei-/es_nir_edicion2023_tcm30-560374.pdf ministerio para la transición ecológica y el reto demográfico. (2023c). real decreto 406/2023, de 29 de mayo, por el que se modifican el real decreto 266/2021, de 13 de abril [...] y el real decreto 263/2019, de 12 de abril, por el que se regula el programa de ayudas para actuaciones de eficiencia energética en pyme y gran empresa del sector industrial. boletín oficial del estado, núm. 129, 3105-2023. https://www.boe.es/diario_boe/txt.php?id=boe-a-2023-12786 ministerio para la transición ecológica y el reto demográfico. (2021). real decreto 266/2021, de 13 de abril, por el que se aprueba la concesión directa de ayudas a las comunidades autónomas y a las ciudades de ceuta y melilla para la ejecución de programas de incentivos ligados a la movilidad eléctrica (moves iii) en el marco del plan de recuperación, transformación y resiliencia europeo. boletín oficial del estado, núm. 89, 14-04-2021. https://www.boe.es/buscar/doc.php?id=boe-a2021-5869 ministerio para la transición ecológica y el reto demográfico. (2020, enero 20). plan nacional integrado de energía y clima 2021–2030. https://www.miteco.gob.es/content/dam/miteco/images/es/pnieccompleto_tcm30508410.pdf moro, a., & lonza, l. (2018). electricity carbon intensity in european member states: impacts on ghg emissions of electric vehicles. transportation research part d: transport and environment, 64, 5– 14. https://doi.org/10.1016/j.trd.2017.07.012 muñoz delgado, b., & rubio vargas, m. (2015). la dependencia energética exterior de españa 1900–2010. en j. m. beneyto & j. c. pereira (eds.), historia de la política exterior española en los siglos xx y xxi. vol. i (pp. 461–490). madrid: ceu ediciones. https://doi.org/10.1016/j.trip.2021.100364 https://doi.org/10.5772/intechopen.76788 https://doi.org/10.1016/j.jclepro.2024.143834 https://doi.org/10.1016/j.tranpol.2024.09.012 https://www.boe.es/buscar/doc.php?id=boe-a-2012-12216 https://www.boe.es/diario_boe/txt.php?id=boe-a-2009-11022 https://otle.transportes.gob.es/movilidad https://esmovilidad.transportes.gob.es/noticias/mitma-adjudica-500-millones-de-euros-120-municipios-para-descarbonizar-sus-nucleos-urbanos https://esmovilidad.transportes.gob.es/noticias/mitma-adjudica-500-millones-de-euros-120-municipios-para-descarbonizar-sus-nucleos-urbanos https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/sistema-espanol-de-inventario-sei-/avance_gei_2022.pdf https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/sistema-espanol-de-inventario-sei-/avance_gei_2022.pdf https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/sistema-espanol-de-inventario-sei-/es_nir_edicion2023_tcm30-560374.pdf https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/sistema-espanol-de-inventario-sei-/es_nir_edicion2023_tcm30-560374.pdf https://www.boe.es/diario_boe/txt.php?id=boe-a-2023-12786 https://www.boe.es/buscar/doc.php?id=boe-a-2021-5869 https://www.boe.es/buscar/doc.php?id=boe-a-2021-5869 https://www.miteco.gob.es/content/dam/miteco/images/es/pnieccompleto_tcm30-508410.pdf https://www.miteco.gob.es/content/dam/miteco/images/es/pnieccompleto_tcm30-508410.pdf https://doi.org/10.1016/j.trd.2017.07.012 146 cid & garcía-álvarez niu, j., li, x., tian, z., & yang, h. (2024). uncertainty analysis of the electric vehicle potential for a household to enhance robustness in decision on the ev/v2h technologies. applied energy, 365, article 123294. https://doi.org/10.1016/j.apenergy.2024.123294 pehlken, a., albach, s., & vogt, t. (2017). is there a resource constraint related to lithium ion batteries in cars? the international journal of life cycle assessment, 22(1), 40–53. https://doi.org/10.1007/s11367-015-0925-4 peugeot. (2024, diciembre). https://www.peugeot.es/elige-un-coche/gama-peugeot.html raluy, r. g., guillén-lambea, s., serra, l. m., guadalfajara, m., & lozano, m. a. (2021). environmental assessment of central solar heating plants with seasonal storage located in spain. journal of cleaner production, 314, article 128078. https://doi.org/10.1016/j.jclepro.2021.128078 red eléctrica española. (2018). vehículo eléctrico: cuestiones a plantear. https://www.ree.es/sites/default/files/07_sala_prensa/20190114_presentacion_ve.pdf renault. (2024, diciembre). https://www.renault.es/configurador.html rietmann, n., & lieven, t. (2019). a comparison of policy measures promoting electric vehicles in 20 countries. in m. finger & m. audouin (eds.), the governance of smart transportation systems (pp. 125-145). cham, switzerland: springer. https://doi.org/10.1007/978-3-319-96526-0_7 rosales-asensio, e., borge, d. b., cabrera, p., & sarmento, p. (2024). effectiveness and efficiency of support schemes in promoting renewable energy sources in the spanish electricity market. international journal of electrical power & energy systems, 158, article 109926. https://doi.org/10.1016/j.ijepes.2024.109926 ross morrow, w., gallagher, k. s., collantes, g., & lee, h. (2010). analysis of policies to reduce oil consumption and greenhouse-gas emissions from the us transportation sector. energy policy, 38(3), 1305–1320. https://doi.org/10.1016/j.enpol.2009.11.006 sahiti, f. (2023). unleashing entrepreneurial potential in transition economies: a comparative analysis of the impact of macro and micro policies. european journal of government and economics, 12(2), 119–138. https://doi.org/10.17979/ejge.2023.12.2.9600 singh, v., singh, v., & vaibhav, s. (2020). a review and simple meta-analysis of factors influencing adoption of electric vehicles. transportation research part d: transport and environment, 86, article 102436. https://doi.org/10.1016/j.trd.2020.102436 sousa, c., & costa, e. (2022). types of policies for the joint diffusion of electric vehicles with renewable energies and their use worldwide. energies, 15(20), article 7585. https://doi.org/10.3390/en15207585 statista. (2024, mayo 30). dependencia energética exterior en españa 2006–2022. https://es.statista.com/estadisticas/493962/dependencia-de-las-importaciones-de-energia-deespana/ usmani, o., & rösler, h. (2015). policy recommendations and stakeholder actions towards effective integration of evs in the eu. https://resolver.tno.nl/uuid:e2ac5ae6-885a-4147-86c9c92a733d3fe1 weil, m., & ziemann, s. (2014). recycling of traction batteries as a challenge and chance for future lithium availability. en g. pistoia (ed.), lithium-ion batteries: advances and applications (pp. 509–528). amsterdam: elsevier. https://doi.org/10.1016/b978-0-444-59513-3.00022-4 wicki, m., brückmann, g., quoss, f., & bernauer, t. (2023). what do we really know about the acceptance of battery electric vehicles? – turns out, not much. transport reviews, 43(1), 62–87. https://doi.org/10.1080/01441647.2021.2023693 will, c., zimmermann, f., ensslen, a., fraunholz, c., jochem, p., & keles, d. (2024). can electric vehicle charging be carbon neutral? uniting smart charging and renewables. applied energy, 371, article 123549. https://doi.org/10.1016/j.apenergy.2024.123549 wood, j., & jain, a. (2020). raceways, rebates, and retrofits: an exploration of several american cities’ policies to facilitate electric vehicle purchase and usage. international journal of urban sustainable development, 13(2), 148–158. https://doi.org/10.1080/19463138.2020.1843467 yang, m., & lin, b. (2024). the development of consumer preferences for electric vehicle charging infrastructure in china: evidence from a questionnaire survey with a four-year interval. energy, 307, article 132671. https://doi.org/10.1016/j.energy.2024.132671 https://doi.org/10.1016/j.apenergy.2024.123294 https://doi.org/10.1007/s11367-015-0925-4 https://www.peugeot.es/elige-un-coche/gama-peugeot.html https://doi.org/10.1016/j.jclepro.2021.128078 https://www.ree.es/sites/default/files/07_sala_prensa/20190114_presentacion_ve.pdf https://www.renault.es/configurador.html https://doi.org/10.1007/978-3-319-96526-0_7 https://doi.org/10.1016/j.ijepes.2024.109926 https://doi.org/10.1016/j.enpol.2009.11.006 https://doi.org/10.17979/ejge.2023.12.2.9600 https://doi.org/10.1016/j.trd.2020.102436 https://doi.org/10.3390/en15207585 https://es.statista.com/estadisticas/493962/dependencia-de-las-importaciones-de-energia-de-espana/ https://es.statista.com/estadisticas/493962/dependencia-de-las-importaciones-de-energia-de-espana/ https://resolver.tno.nl/uuid:e2ac5ae6-885a-4147-86c9-c92a733d3fe1 https://resolver.tno.nl/uuid:e2ac5ae6-885a-4147-86c9-c92a733d3fe1 https://doi.org/10.1016/b978-0-444-59513-3.00022-4 https://doi.org/10.1080/01441647.2021.2023693 https://doi.org/10.1016/j.apenergy.2024.123549 https://doi.org/10.1080/19463138.2020.1843467 https://doi.org/10.1016/j.energy.2024.132671 1. introduction 2. importance of electric mobility in spain 3. barriers to ev uptake in spain 4. existing measures to promote the use of evs in spain and the targets set by the government 5. recommendations to policymakers to overcome the identified barriers 6. conclusions references ©the author(s) 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 225-239 https://doi.org/10.17979/ejge.2024.13.2.10584 submitted: mar 11, 2024 accepted: jun 24, 2024 published: dec 3, 2024 article key competencies and role profiles for sports managers: a multi-dimensional analysis daniel carranza-bautista,1, * manel valcarce-torrente,2 blanca esneda amaya-sánchez,3 claudia fernanda rojas-núñes 3 1 facultad de organización deportiva, universidad autónoma de nuevo león, méxico 2 facultad de business & tech, universidad alfonso x el sabio, spain 3 facultad de ciencias económicas y de la administración, iu escuela nacional del deporte, colombia *correspondence: daniel.carranzabt@uanl.edu.mx abstract. sports organizations, to be profitable and competitive, become structures that demand the integration of competent personnel with certain characteristics and specific attributes determined by the job profile. this position rests on the purpose of productivity through good performance to fulfill the mission, vision, policies, goals, or objectives of the organization. preceding from the above, there is diversification regarding the competencies, attributes, elements, and operational components that a sports manager must adopt to practice their profession successfully. in this sense, the aim of this study is to determine the elements and ideal holistic components that integrate the competencies and job profiles of the sports manager in the professional context. the research methodology uses nine variables in three dimensions: one related to six levels of mastery in sports management, another that refers to competencies, and one more that focuses on the job profile, resulting in 15 specific traits that a sports manager must develop to be professionally competent. keywords: job profile; competence; sports management; sports manager; management skills jel classification: z20; z22 1. introduction as an isolated concept, management aims to be understood by actions, competencies, and different performances required for proper company functioning. in its interaction, it is governed by logical and coordinated planning to promote efficiency in the results and better organizational performance (mestre, 2013). starting from this assertion, sports management, as a disciplinary area performed in organizations that provide sports services, is governed by strategic and operative strategies that set specific goals aligned with achieving organizational-level objectives (sierra et al., 2021). on the other hand, an epistemological point of view addresses the theories and methods applied to the administrative process (planning, organizing, executing, and controlling) to manage the available resources without being limited only to these elements, "done holistically to include administrative https://creativecommons.org/licenses/by-nc/4.0/ 226 carranza-bautista et al. tools, but also other areas of knowledge" (rozo rondón et al., 2022, p. 1019). it is also important to mention that sports administration is applied in different areas of action and currently forms a dualism with physical activities management that requires specialization and the design of profiles adapted to the needs of the organization (sandino rodríguez et al., 2022; carranza, 2021; blanco, 2016). that said, sports managers, in their daily performance, require certain work skills defined in the job profile and real work situations. in this sense, it is worth mentioning that job competency has a structure that integrates a series of elements, components, and attributes so that a person achieves a specific performance according to the needs of the position. for trujillo-segoviano (2014), these assertions were made with a holistic focus that requires not only action interactions, simultaneous tasks, or experience in specific situations but also knowledge, skills, attitudes, or values to achieve competent performance. according to carranza-bautista and valcarce-torrente (2024), their study identifies 10 essential competencies for sports managers, combining leadership, technical skills, and values while integrating conceptual, attitudinal, and procedural elements to optimize professional management in physical and sports activities. on the other hand, from the perspective of méndez (2015), aspects such as functions, profile, management styles, work area, and skills to be developed must be addressed to analyze a sports manager in their professional performance who focuses on being the executive responsible for sports management. furthermore, regarding the type of skills predominating in his conclusions, he highlighted three levels: technical skills at low levels, human skills at intermediate levels, and knowledge skills at higher levels. likewise, professional competencies respond to everyday situations or comprehensive problems in the required context, requiring people to apply certain principles of suitability that refer to training in terms of the relevance of personal resources, the efficiency with which they act and address the activities, and the favorable changes generated to achieve better performance. from this perspective, for tobón et al. (2010) and sandoval (2012), the concept addresses comprehensive actions in distinct settings where competencies respond to "knowing how to know" (knowledge), "knowing how to do” (skills, procedures, and techniques) and "knowing how to be" (attitudes and values). the job profile is usually a way to define the specific characteristics a person should possess in a job assignment. it integrates a series of elements that need to be specified according to the expectations of an organization that faces a changing and competitive environment. in this sense, the profiles in their structure articulate a series of elements to demand or generate abilities to meet a specific objective; thus, it is important to point out that there is a relationship between the profile of the positions, the competencies, the productivity, and the well-being of workers (núñez, 2019, may) these statements about the profile of the position are close to what was established by carranza (2021), who concluded that the sports or physical activities manager turns out to be a professional who can be located at different hierarchical levels depending on the position they hold within the structure of the sports system or organization, by stating that they perform integration and coordination functions in multiple disciplinary areas and mobilize resources in search of the fulfillment of plans, programs, and projects. he also established that their functions center on managerial skills and abilities corresponding to the hierarchical level in which they are located to key competencies and role profiles for sports managers maintain operational stability and promote favorable changes. based on the above, in theory, sports or physical activity managers pursue specific objectives, play a role in their position, and assume responsibilities to fulfill various functions, tasks, and activities that evaluate their abilities. these facts require demonstrating certain job skills. for león-alcérreca et al. (2021), the position of a sports manager, since its creation, has evolved since it requires administration knowledge and has also begun to explore other disciplinary fields. this position can involve a leading role as a leader, addressing planning, programming, administration, coordination, direction, and managing economic and financial resources to fulfill the stated purpose (bernaule, 2023). considering the above, it can be seen from a general context that in sports management, various authors point out the elements or criteria on which a sports manager must perform at work without projecting in a categorical and specific way the operational elements or attributes of suitability that are necessary to develop to be competent. based on the previous considerations, this work aims to determine the structural and operational elements of sports administration or management from the perspective of the sports leader, from a holistic and suitability approach, considering nine study variables, four that are identified with the components of competence and five with the position profile. 2. methodology the nature of the research was exploratory with a descriptive mixed-method approach using numeric, verbal, textual, visual, symbolic, and other classes of data to understand scientific problems (lieber & weisner, 2010). in this sense, the qualitative analysis sought to deepen the knowledge of the object of study in its natural environment, focusing on important elements on an abstract level (varela et al., 2014) and quantitatively using statistical analysis to manipulate variables related to the object of study. this research is cross-sectional by collecting data at a single moment since it seeks to specify important properties and characteristics and describe the trends of a group or population (hernández et al. 2014). 2.1 participant sample the authors considered studying a population of sports administrators and managers from various ibero-american countries. mexico, columbia, and spain were the main ones, referencing the six management levels presented by carranza (2021). this study used non-probabilistic convenience sampling, a technique that uses randomization to select the participants that meet the requirements (argibay, 2009). the sample consisted of 233 sports managers with the following distribution by country: colombia with 81, spain with 74, mexico with 72, and six from other countries. of the total, 71.24% were men and 28.76% were women. of the men, 39.76% claimed to have between 16 and more than 20 years of professional experience, while 56.72% of women had between 5 and 15 years. most of the participants had a university education with a specialty or a master's degree (table 1). 228 carranza-bautista et al. table 1. sample distribution by gender related to the range of professional experience and the level of studies related to academic training. gender n, (%) range of professional experience in years, n, (%) study level, n, (%) 1 a 5 5 a 10 11 a 15 16 a 20 < 20 pr ba sp ma d pos male 166 (71.24) 34 (20.48) 35 (21.08) 31 (18.67) 28 (16.87) 38 (22.89) 4 (2.41) 43 (25.90) 32 (19.28) 62 (37.35) 23 (13.86) 2 (1.20) female 67 (28.76) 15 (22.39) 19 (28.36) 19 (28.36) 8 (11.94) 6 (8.96) 3 (4.48) 11 (16.42) 17 (25,37) 27 (40.30) 7 (10.45) 2 (2.99) note: pr = preparatory, ba = bachelor's degree, sp = specialty, ma = maester's, d = doctorate, pos = postdoctorate. 2.2. instrument and procedure the instrument was a 16-item questionnaire designed based on the study's needs. the first section consisted of six questions corresponding to demographic data. the remaining questions corresponded to three dimensions with the following distribution: the first referred to management levels and had a multiple choice option; the other two were structured with nine open questions; four comprised the holistic components dimension of competency and five the job profile dimension. the instrument was validated by consultation with experts, with the participation of 10 judges selected according to the criteria of independence, professional solvency, research activity, and level of responsibility (cremades, 2017). each item was validated with the dimension it intended to analyze, through a matrix, with the following four criteria: 1) if the item responds to the variable it intends to measure, 2) the definition of the clarity of the item, 3) the information that is desired from the item, and 4) if it evaluates the dimension in which it is included. in addition, the contributions made by the experts regarding the content, form, and observations of the construct were addressed qualitatively to strengthen the properties of the instrument. 3. data analysis based on the previous procedure, the validity and reliability of the results provided by the validation matrix used by the experts were worked on to seek the psychometric solidity of the instrument. the query was processed with spss software for social and applied sciences research data analysis. the fleiss kappa coefficient, a reliability index, was used to evaluate the absolute agreement between the interobserver measurements of the possible agreements that occur in each dimension (picadoalavarado, 2008; gordillo et al. 2009; moya-mata, et al. 2018). in the first and only analysis to measure the instrument's validity through expert assessment, the fleiss kappa coefficient had an agreement strength of k=0.791, representing a good index. the dimension related to management levels had a k=0.662, showing good agreement strength. on the other hand, the dimension of holistic components showed an agreement strength of k=0.899, and the dimension related to the position profile showed an agreement strength of k=0.843 (table 2). key competencies and role profiles for sports managers table 2. scale of the interpretation of the fleiss kappa coefficient (altaman, 1991). k-value concordance strength <0.20 poor 0.21 0.40 weak 0.41 – 0.60 moderate 0.61 – 0.80 good 0.81 – 1.00 very good note: k = fleiss kappa index the following formula was used to determine the acceptance rate of the instrument and the dimensions related to the four validation criteria. var = [∑ 100 ∑𝑇𝑇𝑇𝑇 𝐼𝐼 𝑇𝑇𝐶𝐶 + ∑ 𝐷𝐷𝐼𝐼𝐼𝐼 ] 𝑀𝑀 + ∑ 𝐷𝐷𝑀𝑀 , where var = value acceptance rate, i = item, cw = criterion weighting, tc = total of the criteria, ir = item result, d = dimension and m = mean. the var, in general, was 88.4; the var of the management level dimension was 87.1; the var of the holistic components dimension was 90.8; and the job profile dimension was 89.8 (table 3). table 3. interpretation of the acceptance rate of expert assessment through the validation matrix. acceptance rate impact level < 50 poor 50 > 70 weak 70 > 80 moderate 80 > 90 favorable 90 > 100 very favorable in a second analysis of the data processed through the validation matrix used in the expert consultation, cronbach's alpha (α), which expresses the degree to which the items measure the same variable, was used to assess the instrument's internal consistency (quero, 2017). the data, in general, had an α=.949; regarding the management levels dimension, the α=0.926, the holistic components dimension was α=0.961, while the job profile dimension was α=.955, values that represent excellent reliability (table 4). table 4. reliability levels according to cronbach's alpha. index cronbach's alpha reliability level 1 > 0.9 excellent 2 > 0.8 good 3 > 0.7 acceptable 4 > 0.6 questionable 5 > 0.5 unacceptable source: own elaboration based on tuapanta et al. (2017). 230 carranza-bautista et al. the questionnaire was applied online with questionpro survey software. later, using atlas.ti version 8.4.5. software, the information collected was systematically processed using content analysis and the qualitative research technique that allows synthesizing information and formulating data and reproducible and valid inferences that can be applied to the context (varguillas, 2006). based on the mentioned methodology, the information was used to generate 3728 citations that refer to selecting a text relevant to the study. we later categorized and codified the data, obtaining 489 codes, and with this, we determined code rooting (gr), which is the number of times a code appears in a citation. the total sum of the codes was gr=8997. likewise, regarding the study variables, the level of co-occurrences (co), which refers to the number of times that a code coincides with a study variable, was determined. the index reported was co=7367. as an innovative part of the study, the relative code rooting of the variable (grrv), which refers to the level at which a code affects and belongs to a study variable or shows a dispersion between the various study variables, was determined by the rule of three (grrv= grv*100/gr). 4. results in a general analysis of the variables related to the dimensions of holistic components and the sports manager's job profile, the study had a balanced rooting of the variables, demonstrating stability in the responses provided by the participants. the contribution percentage that each variable produces in codes is calculated in relation to the total number of codes that the nine variables contribute to the study, which is 489. these are the codes that are present in each variable. the contribution percentage of the variables in code co-occurrences comes from the total of co-occurrences between the nine variables, which was 7367, representing the number of frequencies that a code interacts with the variable (table 5). table 5. contribution of the study variables to rooting, codes, and code co-occurrences. no. variables gr codes % co % holistic components dimension 1 sports manager knowledge 233 209 42.74 1278 17.35 2 sports manager skills 233 224 45.81 1198 16.26 3 required sports management values 233 173 35.38 1151 15.62 4 attributes that the sports manager must possess 233 187 38.24 704 9.56 job profile dimension 5 sports manager job functions 233 157 32.11 835 11.33 6 sports manager responsibilities 233 200 40.90 783 10.63 7 objective of the job position you hold 233 100 20.45 342 4.64 8 role of the sports manager 233 99 20.25 346 4.70 9 sports manager skills 233 182 37.22 730 9.91 note: the table presents the results of the qualitative analysis from the coding and quantification of the processed information. gr = variable rooting; co = co-occurrences. key competencies and role profiles for sports managers another relevant result is associated with the codes with greater relative roots in the variable "grrv." these codes stand out due to a high presence in the variable with which they are associated and little or no presence in other variables (figure 1). from this analysis, in the dimension of holistic components, sports legislation stands out as a specific knowledge of this variable, while negotiation stands out in skills. regarding values, loyalty is presented as the main code, and in attributes, the professionalism of a sports manager is the code with the highest incidence. regarding the dimension of the job profile of a sports manager, the main function refers to work supervision; in responsibilities, customer satisfaction stands out, while in objectives, the organization's profitability is ranked as the one with the greatest impact occupying first place. regarding the role of the sports manager, the main code refers to leading organizations, and in terms of capabilities, the work capacity of a sports manager stands out. figure 1. codes with a greater connection to the study variable considering the rooting of the relative code. note: the relative code rooting of the variable is determined with the rule of three by multiplying the cooccurrences that the code presents in the variable by one hundred and dividing them by the total code rooting (co*100/gr). grrv= relative code rooting in the variable, co= co-occurrences, gr= code rooting. 232 carranza-bautista et al. the management levels dimension considers carranza's classification (2021) as a reference. the sports managers participating in the study were identified mostly at level 2, with 31.76%, while levels 1 and 3 had very low participation. however, it is worth mentioning that the first three levels focus on representing the internal management of the organizational structure in an entity. together, they contribute 48.07% of the research data. regarding the other 51.93% of the information, 39.91% was distributed equally between levels 4 and 5, while level 6, the highest management level and the most complex, represented 12.02%. these three levels represent a managerial type of management with more professional characteristics and high demand levels in function performance (figure 2). the figure shows the relationship between each level and the different study variables. figure 2. analysis of the elements of the holistic and job profile dimensions in relation to management levels. source: own elaboration. co = co-occurrences, grrv = code rooting related to the variable. key competencies and role profiles for sports managers considering gr as the main indicator, it was possible to identify 15 relevant codes for this study, which, according to the opinion of the managers, can be associated with one or several variables, taking the grrv as a reference (figure 3). from this analysis, leadership can be established as a skill that must be developed first; however, it is also considered an attribute and an ability. assertive communication stands out more as a skill, without forgetting that, to a lesser extent, it is associated with a capacity that requires knowledge. on the other hand, planning is a function of the sports manager that requires knowledge and responsibility within the job profile. in this same analysis, empathy strongly impacts two variables: first highlighted as a value, second as a skill, and third, with less incidence, as an attribute. in this order of results, there is also problem-solving, which, in a balanced way, finds its strength as a capacity and ability later considered to a lesser extent as an attribute. figure 3. the 15 codes with greater contribution according to their gr in relation to the relative rooting with each study variable in general. note: the study produced 489 codes. many of these are considered emerging codes due to their low contribution and impact on the results. the 15 codes in the figure are the product of a general analysis that considers code rooting as the main indicator. gr = code rooting, grrv = code rooting related to the variable. 234 carranza-bautista et al. finally, in the analysis of each study variable, it was possible to determine the main codes with a greater presence, considering co as an indicator and grrv, which highlights the relevant codes as shown in figure 4. figure 4. variable analysis considering the three first codes from the co. source: own elaboration. gr = code rooting, co = co-occurrences, grrv = code rooting related to the variable. 5. discussion the purpose of the research focused on determining the holistic components of a competency and elements of the job profile, through nine study variables, resulting in 15 elements or components that are required in the competent performance of a sports manager. for burgos (2021), in a study identifying the profile of a sports manager in private sports centers in andalusia, concluded that personnel management, facility management, activity and service planning, managerial and administrative tasks, sales and marketing, and economic-financial management are the principal areas of responsibility of sports managers. in continuous manager training, he mentions that addressing marketing and sales first and coaching second is necessary. in a second block of interest, he mentions the administrative process, accounting, finance, and information technologies as relevant results. furthermore, he argues that key aspects for of the good performance of a private key competencies and role profiles for sports managers sports leader focus on "teamwork," while public managers lean toward the "budget." in general, these results coincide with planning, which was third overall in the research, followed by human resources management, which was sixth and similar to personnel management. regarding marketing, in the study, it ranked 55th; however, in the knowledge variable, it was the first element that a sports manager must know. salgado-barandela et al. (2019), in relation to the job profile of a manager in terms of capabilities and skills, presented similar characteristics in competencies by determining four primary components: organization, interaction, reflection, and direction. in the skills section, the results indicate decision, management, communication, initiative, and motivation as the most notable aspects. these results coincide mainly with codes such as assertive communication, which occupied the second position in the study. organization was ninth, efficient team management was fourth in the responsibilities variable, and ninth in the functions variable. gonzález-castro et al. (2021), in a study conducted on the impact of the future on the performance of the sports manager through a literary review, determined ten managerial skills that a sports manager must possess: leadership, communication, empathy, training, growth, teamwork, negotiation, creativity, flexibility, and ethics. these concur with our study, with five being the most relevant: leadership, assertive communication, empathy, teamwork, and negotiation as a skill with a high grrv. an investigation by bastías et al. (2023) evaluated the competencies of municipal sports managers in chile and analyzed the importance of identifying and associating professional skills with the functions of people who work in the sports field. the research concluded that carrying out this type of study leads to efficiency and quality in sports services, which coincides with the job profile dimension of a sports manager, in which client satisfaction stands out as a primary responsibility when managers of an organization have leadership and great work capacity, constituting a key position in the structure (rozo rondón et al., 2022; sandino rodríguez et al., 2022). burgos gil (2022) carried out a study focused on identifying the profile of sports managers in private centers in andalusia, presenting the capabilities and skills that a leader must have and highlighting that there is no unique sports manager figure since infrastructure, the number of people in charge, or the functions performed, lead to discrepancies with high management positions and operational responsibilities, and defining in parallel, diverse skills and competencies. similarly, this research showed that the sports manager's knowledge, skills, abilities, and roles are relevant for decisionmaking and achievement in sports management. however, attributes, such as a sports manager's professionalism, were the code with the greatest impact on the study results. the study by gambau i pinasa (2017) is notable and mentions responsibilities attributed to a sports director, such as managing processes to achieve financial objectives, infrastructure, human talent, and creating strategies to motivate and achieve productivity. however, these attributes are immersed in leadership, which is in agreement with this research. this attribute has been described as a primary skill that a manager must develop. likewise, it is important to mention the variables related to human talent management and financial activities, which lead to highlighting the leader's planning, optimization of resources, and achievement of the proposed goals. gutiérrez lillo (2014) conducted a study on the competencies of managers of the sports federations of the chilean olympic committee to identify the skills and knowledge that support sports management and evaluate the competencies that have the greatest impact on sports 236 carranza-bautista et al. federation leaders' performance. they demonstrated evident similarities between the main competencies and the analysis of this research, such as assertive communication with staff, effective problem-solving skills, managing financial resources, and coordinating and organizing operational aspects to conduct sports programs, which are variables of high impact in evaluating sports managers' competencies. 6. conclusions the reliability and agreement of the fleiss kappa analysis had, in general, "good" agreement strength, while in the dimension of holistic components, suitability, and job profile elements showed very strong expert agreement. regarding the acceptance rate of the var assessment, it was generally and very favorable in the dimension of holistic components. the assessment of internal consistency, using cronbach's alpha, indicated an excellent level of reliability in general and in all dimensions. this validation process of open questions of the questionnaire on criteria of clarity, obtaining information, evaluation of the dimension, and variable measurement proved to be a very solid methodology. this finding was reflected in the research by presenting in the citation of open questions a balanced general gr with stability in the responses that allowed obtaining, in the citation of open questions, a balanced general gr with stability in the responses, which allowed for the obtaining of very punctual and uniform information that facilitated the coding process of the questionnaires. regarding the analysis related to the relative rooting of the study variable grrv, the resulting codes reflect a greater belonging to the variable with which they stand out and hardly appear in the study due to their gr compared to other variables. this situation occurs because their co represents a high index of belonging to the variable they represent, approaching their general gr above the other codes. only the organization's profitability code, in the variable position objective, is the only one that appears in first place in both categories due to its high gr=43, co=28, and grrv= 65.12%. however, the code is considered part of the role of a sports manager and a responsibility in the job profile. in this order of conclusions, when referring to the dimension of management levels, the participation of managers in the type of sampling was not balanced individually; however, grouping the first three levels (n1, n2, and n3) that represent the management of the internal structure of an organization with the following three levels (n4, n5, n6), which represent the top administration management, a balance is found in participation that reveals in the results, the requirements regarding the study variables of sports management or physical activities. among these, in n1, assertive communication stands out; in n2, human resource management; in n3, problem-solving; in n4, planning; in n5, planning; and in n6, leadership. in this dimension, in general, the codes that predominate in the study are marketing as knowledge, assertive communication in skills, respect in the values variable, leadership as an attribute, planning in the functions and responsibilities variable, the profitability of the organization in the objective of manager's position, leadership as a role of the sports manager, and problem-solving as a skill. these same results coincide with the analysis of each variable to determine the holistic components and the elements of the job profile that predominate in sports or physical activity management through coding (figure 2). key competencies and role profiles for sports managers finally, to conclude the research in a general way, it was determined that 15 codes, due to their high gr, could be the basis for the generation of competencies that a sports or physical activity manager must have and develop, preferably in the variable that has the greatest impact, among these: leadership, assertive communication, planning, empathy, problem-solving, human resource management, teamwork, resource management, organization, responsibility, respect, financial management, honesty, coordinating activities, and knowledge in sports management (figure 5). figure 5. the 15 operational components or elements from a holistic and suitability approach that are required in the competent performance of a sports manager. source: own elaboration. gr = root coding. 6.1. limitations and future perspectives the study's authors considered sports managers active in the management or performance of a position during data collection as a study population. for this reason, the sample consisted of managers with the willingness and the scope of relationships and a contact network of researchers in each country. this fact made the sample size difficult, and thus, the study may not have considered a sector that could provide relevant information. furthermore, senior management positions in sports management are difficult to access and have little willingness, which limits participation in this type of research. 238 carranza-bautista et al. this work applied a qualitative analysis that lays the foundations for investigating the elements and components that can integrate or generate generic, specific, or concrete competencies in different sports management areas. likewise, it provides relevant information to consider in foundation studies, designs, and redesigns of educational programs at higher learning levels to obtain a degree or postgraduate degree in the disciplinary area of sports management. references altman, d. g. (1991). practical statistics for medical research. new york: chapman and hall. argibay, j. c. (2009). muestra en investigación cuantitativa. subjetividad y procesos cognitivos, 13(1), 1329. bastías, d. d., castillo, c. m., flandez, j., améstica, m. c., & ramírez, f. g. (2023). valoración de las competencias profesionales en gestores deportivos municipales de chile. retos: nuevas tendencias en educación física, deporte y recreación(50), 831-837. https://doi.org/10.47197/ retos.v50.100016 bernaule, j. (2023). los modelos de gestión del deporte universitario. sade-revista de ciencias de la salud y el deporte, 3, 45-56. blanco, p. e. (2016). tendencias en gestión deportiva. revista española de educación física y deportes, (414), 13-16. https://doi.org/10.55166/reefd.v0i414.477 burgos gil, m. á. (2022). identificación del perfil del gestor deportivo en los centros deportivos privados en andalucía. carranza, b. d. (2021). sports or physical activity management: global intervention model. retos: nuevas tendencias en educación física, deporte y recreación, 39, 961-972. carranza-bautista, d., & valcarce-torrente, m. (2024). diseño, validación y valoración de competencias del gestor de actividades físicas y deportivas. retos, 59, 527-536. https://doi.org/10.47197/ retos.v59.107515 cremades, r. (2017). validación de un instrumento para el análisis y evaluación de webs de bibliotecas escolares mediante el acuerdo interjueces. investigación bibliotecológica, 31(71), 127-149. https://doi.org/10.22201/iibi.0187358xp.2017.71.57813 gambau i pinasa, v. (2017). identificación profesional en el ámbito de la dirección y gestión deportiva. en el gestor deportivo en la organización del deporte en la sociedad actual (2ª ed., pp. 53-78). wanceulen editorial deportiva. gonzález-castro, j. c. a., morquecho, s. r., gadea, c. e. a., corrales, f. g. l., & tostado, a. j. m. (2021). impacto del futuro en el desempeño del gestor deportivo en méxico. revista de gerencia deportiva, 5(1), 35-49. gordillo, j. j. t., & rodríguez, v. h. p. (2009). cálculo de la fiabilidad y concordancia entre codificadores de un sistema de categorías para el estudio del foro online en e-learning. revista de investigación educativa, 27(1), 89-103. gutiérrez lillo, c. g. (2014). competencias de los gestores de las federaciones deportivas del comité olímpico de chile (tesis doctoral). universidad andrés bello. hernández sampieri, r., fernández-collado, c., & baptista-lucio, p. (2014). metodología de la investigación. méxico: mcgraw-hill. león-alcérreca, m. f., morquecho, s. r., gadea, c. e. a., medina, v. m., & pérez, g. j. a. (2021). el rol del gestor deportivo en la comunicación del futbol soccer femenil. revista de gerencia deportiva, 1(5), 10-24. lieber, e., & weisner, t. s. (2010). meeting the practical challenges of mixed methods research. en sage handbook of mixed methods in social and behavioral research (2nd ed., pp. 559-580). thousand oaks, ca: sage. https://doi.org/10.4135/9781506335193.n22 méndez, r. b. (2015). el directivo deportivo del siglo xxi: áreas de análisis. revista española de educación física y deportes: reefd, (410), 41-55. mestre, s. j. a. (2013). componentes de la gestión deportiva. una aproximación. viref revista de educación física, 2(2), 1-19. moya-mata, i., ruiz, j. m., sanchís, l. r., & ros, c. r. (2018). diseño, fiabilidad y validez de una herramienta para el análisis de las imágenes de los libros de texto de educación física. retos: nuevas tendencias en educación física, deporte y recreación(34), 240-246. https://doi.org/10.47197/ retos.v0i34.59497 https://doi.org/10.47197/retos.v50.100016 https://doi.org/10.47197/retos.v50.100016 https://doi.org/10.55166/reefd.v0i414.477 https://doi.org/10.47197/retos.v59.107515 https://doi.org/10.47197/retos.v59.107515 https://doi.org/10.22201/iibi.0187358xp.2017.71.57813 https://doi.org/10.4135/9781506335193.n22 https://doi.org/10.47197/retos.v0i34.59497 https://doi.org/10.47197/retos.v0i34.59497 key competencies and role profiles for sports managers núñez, a. r. (2019, mayo). la presencia de competencias informativas en el ámbito laboral: análisis en perfiles de puesto. en seminario en gestión del conocimiento y servicios informativos 2018. picado-alvarado, f. (2008). análisis de concordancia de atributos. revista tecnología en marcha, 21(4), ág29. quero virla, m. (2017). confiabilidad y coeficiente alpha de cronbach. rozo rondón, k. d., porras ramírez, k. a., bolívar silva, o. d., castro, j. a., & liévano, j. a. (2022). la gestión deportiva en latinoamérica: horizonte epistemológico y perspectivas actuales. retos, 46. https://doi.org/10.47197/retos.v46.92540 salgado-barandela, j., barajas, á., sánchez-fernández, p., gambau, v., & silva, r. (2019). perfil profesional del gestor deportivo municipal: capacidades y habilidades necesarias. revista española de educación física y deportes, (424), 45-56. https://doi.org/10.55166/reefd.vi424.722 sandino rodríguez, m. c., carranza bautista, d., rojas núñez, c. f., & gómez hincapié, á. j. (2022). factores clave de la gestión deportiva y de actividades físicas en latinoamérica: gestión deportiva. revista observatorio del deporte, 8(2), 1-18. https://doi.org/10.58210/odep280 sandoval, f., montaño, n., miguel, v., & ramos, e. (2012). gestión de perfiles de cargos laborales basados en competencias. revista venezolana de gerencia, 17(60), 660-675. https://doi.org/10.31876/ revista.v17i60.10937 sierra, g. l. g., morquecho, s. r., & gadea, c. e. a. (2021). la gestión deportiva y su implicación en la resolución de los conflictos. eirene estudios de paz y conflictos, 4(7), 163-182. tobón, s. t., prieto, j. h. p., & fraile, j. a. g. (2010). secuencias didácticas: aprendizaje y evaluación de competencias (vol. 1, p. 216). méxico: pearson educación. trujillo-segoviano, j. (2014). el enfoque en competencias y la mejora de la educación. ra ximhai, 10(5), 307-322. https://doi.org/10.35197/rx.10.03.e1.2014.26.jt tuapanta dacto, j. v., duque vaca, m. a., & mena reinoso, a. p. (2017). alfa de cronbach para validar un cuestionario de uso de tic en docentes universitarios. varela, c. t., suárez, a. l. s., soler, r. t., reyes, m. s., & assa, a. p. (2014). la modelación de los objetos y procesos como método para validar los resultados de la investigación científica. panorama cuba y salud, 9(1), 29-34. varguillas, c. (2006). el uso de atlas. ti y la creatividad del investigador en el análisis cualitativo de contenido. laurus, 12, 73-87. https://doi.org/10.47197/retos.v46.92540 https://doi.org/10.55166/reefd.vi424.722 https://doi.org/10.58210/odep280 https://doi.org/10.31876/revista.v17i60.10937 https://doi.org/10.31876/revista.v17i60.10937 https://doi.org/10.35197/rx.10.03.e1.2014.26.jt 1. introduction 2. methodology 3. data analysis 4. results 5. discussion 6. conclusions references ©the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 1 (2025), pages 96-117 https://doi.org/10.17979/ejge.2025.14.1.11272 submitted: sep 30, 2024 accepted: may 19, 2025 published: jun 20, 2025 article measuring the impact of democracy on gender equality: does choice of index matter? alka singh,1,* archna chaudhry 2 1 government college, bherian (pehowa), haryana, india 2 department of economics, kurukshetra university, kurukshetra, haryana, india *correspondence: alkasingh0426@gmail.com abstract. using a variety of indicators, the study tries to analyze the impact of democracy on gender equality for a sample of 159 countries across the world. for this purpose, the study used static panel data models and applied various tests of robustness to estimate the impact of democracy on gender equality. it is found that the effect of democracy on gender equality is uncertain and sensitive to the interchangeability of various gender equality indices. however, the replacement of various democracy indices does not cause any substantial change in the results. the impact of democracy on gender equality is almost insignificant, but certain components of democracy like political participation and government functioning have positive impacts in improving the levels of gender equality. among the control variables, economic growth and population size emerge as the positive determinants of gender equality. keywords: gender equality; democracy; index sensitiveness; static panel data models. jel classification: c23; d72; j16 1. introduction does democracy cause gender equality? the question has attracted much scholarly attention in recent times. theoretically, democracy may foster gender equality by expanding political rights, improving civil liberties, and enabling the representation of marginalized groups, including women. democracy, having universal voting rights, is theoretically supposed to enhance gender equality as a bottom-up process driven by democratic participation of females, who constitute fifty percent of the total electorate (franceschet and piscopo, 2013). while democracy increases the likelihood of descriptive representation of women, it does not guarantee substantive representation1 (lovenduski, 1997; mansbridge, 1999). pitkin (1972) argues that substantive representation is more important than descriptive representation. a ‘critical mass’ (around 30-35 percent women in 1 descriptive representation refers to representatives who share characteristics with their group, while substantive representation focuses on advocating for the group's interests, regardless of shared identity. https://creativecommons.org/licenses/by-nc/4.0/ measuring the impact of democracy on gender equality parliament) is necessary for influencing policy and institutional culture in favor of gender equality (dahlerup, 1988). however, institutional barriers, party dynamics, and cultural factors can limit women’s influence even when they reach critical mass. childs and krooks (2006) argue that it is not ‘critical mass’ but the ‘critical actors’ who drive the gender equality reforms. thus, it is not about numbers but about individual politicians who push for gender equality regardless of their gender (phillips, 1995). further, electoral accountability ensured by free and fair elections provides an opportunity for women to promote their interests and hold rulers accountable. since women constitute a significant proportion of the majority, theoretically, they should be able to defend their interests through elections (beer, 2009). in addition to bottom-up pressures, democracy may also enhance gender equality by implementing gender quotas through top-down state policies. most democracies use gender quotas and other affirmative action policies to counter historical and systematic barriers that have kept women under-represented (o’brien and piscopo, 2018). however, many authoritarian and hybrid regimes also enhance gender equality by implementing top-down state policies. rwanda, for example, has one of the highest proportions of women in parliament despite being classified as a hybrid regime. similarly, many left-wing authoritarian regimes have more women in their parliaments and cabinets than other regimes (reynolds, 1999). protection of civil rights is another mechanism that improves gender equality in democratic regimes. a democratic system provides a legal framework, an independent judicial system, and rule of law for protecting the rights of marginalized groups like women (ghai and cottrell, 2009). these protections are established through laws that promote equality, prevent discrimination, and support women's empowerment. women can seek justice more effectively through courts, legal institutions, and human rights bodies. democratic nations often comply with global treaties like cedaw. women can appeal to international bodies if their rights are violated and not being taken care of even after they report about the violation in their own countries. moreover, democracy plays a crucial role in advancing gender equality by protecting the freedom of speech, protest, and advocacy. these rights allow women and their allies to challenge discriminatory policies, push for legal reforms, and raise awareness about gender issues. freedom of expression allows discussion on gender norms, sexism, and traditional rules, leading to cultural transformation. against this background, the present study undertakes a thorough investigation of the impact of democracy on gender equality and differs from previous studies in the following ways: first, rather than relying on any single indicator, it uses widely accepted measures of gender equality— namely, the gender equality index (gei), global gender gap index (ggi), and gender development index (gdi)—as dependent variables to test the sensitivity of democracy’s impact to the interchangeability of these indices. second, we also test how the impact of democracy on gender equality changes when various democracy indicators are substituted for one another. third, the study takes into account country-specific fixed effects vis-à-vis other time-invariant factors such as political regime, economic status, religion, and region by dividing the sample into subsamples based on these characteristics. fourth, we also test the impact of various components of democracy on gender equality to examine their relative importance. finally, the study examines the impact of affirmative policies, such as the implementation of compulsory gender quotas in parliament. the major finding of this study is that democracy is not a robust predictor of gender equality, as its impact is uncertain and highly sensitive to the interchangeability among various gender equality indices. 98 singh & chaudhry however, gender quota in parliament, political participation, and government functioning play a positive role in this regard. the rest of the paper is arranged in following sections: section 2 discusses the existing empirical literature and research gaps to be addressed by this study. section 3 briefly outlines the methodology, data, and variables used in this paper. section 4 presents a detailed analysis of the empirical findings. finally section 5 concludes. 2. review of literature despite its theoretical plausibility, empirical investigation of the link between gender equality and democracy has received insufficient scholarly attention, with only a few attempts made in this area. inglehart, norris, and welzel (2003) were among the first to explore this link, using ols regression with cross-sectional data from 46 countries. they used the freedom house index of civil liberties and political rights as the measure of democracy and female representation in the lower house of parliament as the measure of gender equality. the authors concluded that the relation between these two variables is not direct; rather, they are linked through a third variable called cultural change. their main argument is that economic development tends to bring a gradual shift from survival to self-expression values, and these values in turn are conducive to both democracy and gender equality. thus, the authors discarded any causal link between these two variables. however, their analysis is based on cross-sectional data, which does not consider the role of time dynamics, which is very important to examine the nature of causation. further, rizzo et al. (2007) analyzed crosssectional data from 22,729 individuals in arab and non-arab muslim countries using the world values survey 2000. they estimated the relationship between support for gender equality and support for democracy using ols. their findings showed that non-arab muslim countries had higher levels of support for women's rights, and individuals who supported gender equality were more likely to support democracy. in contrast, the opposite was true for arab muslim countries. the authors argued that for a well-functioning democracy to emerge in the arab middle east, the rule of law protecting gender equality, minority rights, and citizen inclusion would need to be established. like inglehart et al., this research is not sufficient to establish causality between gender equality and democracy due to its cross-sectional nature and lack of time dynamics. building upon the analysis of inglehart et al., beer (2009) examined the impact of democracy on gender equality by estimating gee models for a variety of gender equality indicators related to health, education, nutrition, and employment based on data from 133 countries collected between 1960 and 2004. she found that countries with greater democracy and longer histories of women's suffrage exhibited higher levels of gender equality in health, education, and employment. beer's analysis neglects critical countryspecific fixed effects, rendering the estimated impact of democracy on gender equality vulnerable to endogeneity bias. however, she tries to capture these country-specific fixed effects by introducing dummies for regions. taking in view the impact of international conventions on gender equality, cho (2014) investigated the effect of cedaw on women’s rights in conjunction with democracy. using pooled ols on data from 126 countries between 1981 and 2007, cho found that the combined effect of measuring the impact of democracy on gender equality cedaw and democracy significantly enhanced women's social rights. however, neither cedaw nor democracy alone had a significant impact on gender equality. similar to beer (2009), cho's study could not address the endogeneity issue arising from the presence of country-specific time-invariant heterogeneity in the panel data model. in a similar fashion, hogstrom (2015) examined the link between gender equality and democracy. using the share of female ministers in the cabinet as the dependent variable and the freedom house democracy index as the independent variable, he analyzed data from 191 countries between 2000 and 2010 with pooled ols. the results demonstrate that economic development and democracy only affect gender equality in cabinets positively in specific contexts. development has a positive effect in developed democratic countries, but a negative effect in dictatorships, particularly military dictatorships. the level of democracy has a positive effect mainly in dictatorships, with the strongest effect in civilian dictatorships. further, utilizing the same methodology, blankenship and kubicek (2018) investigated the causality between gender equality and democracy using data from 41 sub-saharan african countries from 1990 to 2014. they constructed a composite gender equality index based on variables such as education, employment, and political representation and examined whether countries with stronger democratic records have better records with respect to gender equality. the authors found no such direct connection, as several non-democratic countries have also made significant progress on the variables in their gender equality index. the most recent attempt in this area is by andersen (2023). he uses v-dem's women's political empowerment index, which combines indicators of women's freedom of movement, freedom from forced labor, property rights, and access to justice. to measure democracy, he uses the dichotomous democracy measure developed by boix et al. (2013). additionally, he employs the unexpected second vatican council (1962-1965) as part of a shock-based identification strategy. andersen (2023) finds a significant causal effect of democracy on gender equality. however, the model may suffer from endogeneity, as the dependent and independent variables share some common characteristics. to address this issue, the author replaces the democracy variable with an instrumental variable constructed by multiplying the dummy variable for vatican ii (which takes the value 1 if the observation belongs to the post-vatican ii period and zero otherwise) by the share of catholics in a country. this instrument variable suffers from three limitations. first, the catholic share is constant over time, making the instrument also time-invariant and unable to capture changes in democracy over time. this raises questions about its relevance for establishing causality between gender equality and democracy. second, since the key explanatory variable in andersen's model is time-invariant and all time-invariant variables are wiped out in the fixed effects panel data model, it is unclear how this study takes into account country-specific fixed effects. finally, while vatican ii may have significantly promoted democratic values in christian societies, its impact on non-christian societies, which constitute a significant portion of the world population, remains uncertain. based on the above discussion, it is evident that majority of the studies use political participation of women—measured in terms of women’s share in parliament (inglehart et al. 2003), female ministers in the cabinet (hogstrom, 2015), and women’s political empowerment (andersen, 2023)—as the indicator of gender equality, while ignoring other aspects like health, education, and employment. they assumed that increased political representation of females will automatically reduce gender inequality in other spheres of society. therefore, rather than focusing solely on the 100 singh & chaudhry political dimension of gender inequality, the present study considers gender inequality in a broader sense. to this end, we use three widely accepted measures of gender (in)equality: gei, ggi, and gdi to account for multiple aspects of biases against women. we utilize all three measures rather than choosing one over the others because each measure aggregates and averages many individual dimensions of gender equality and is subject to errors of aggregation and measurement due to differences in their methodologies (vaccaro, 2021). it is also possible that one or more dimensions not included in one indicator may be considered by others. therefore, to present a holistic picture of gender equality and to ensure the robustness of the results, we consider all three measures of gender equality as dependent variables in our study. following the same reasoning, we use three democracy indices—eiu democracy index (eiud), freedom house index for civil liberties and political rights (fhd), and v-dem’s electoral democracy index (edi)—as measures of political regimes. apart from this, our study makes following departures from the previous literature on this issue: first, we try to deal with the problem of endogeneity bias by estimating fixed and random effects models according to the nature of country-specific individual heterogeneity for a variety of indicators for gender equality and democracy. second, the effects of time-invariant characteristics, such as region, religion, economic position, and political regime types, are taken into account without compromising the country-specific fixed effects by dividing the entire sample into sub-samples based on these characteristics and estimating a separate regression model for each sub-sample. third, we also investigate the impact of various aspects of democracy, such as free and fair elections, government functioning, political participation, political culture, and civil liberties, to understand the role of these mechanisms through which democracy can affect gender equality. 3. methodology 3.1 data and sample a sample of 159 countries worldwide was selected for the present study. the sample selection is based on the availability of data for the variables used in this study. the study covers a 14-year period, from 2006 to 2019, as data on key variables, such as the ggi and eiud, is available only for 2006 and onwards. a brief description of the variables, along with their respective sources, is provided in table 1. a variety of indicators are available for measuring gender equality, and the selection of a particular index is challenging, as the relationship between gender equality and democracy can be sensitive to the choice of measurement. to ensure precision, we do not rely on a single measure of gender equality. instead, we utilize three widely recognized indices: gei, ggi, and gdi. the gei is the inverse of undp’s gender inequality index (gii) which is calculated by deducing gii from 1. the gei ranges from 0 (indicating perfect inequality) to 1 (indicating complete equality). it is a composite index assessing gender equality across three dimensions: reproductive health (i.e., maternal mortality rate and adolescent birth rate), empowerment (i.e., percentage of females in higher education and parliament), and labor market participation. the ggi, using a different methodology, evaluates gender equality across four primary areas: economic participation and opportunity (i.e., labor force participation, gender-based wage differences, women’s income as a percentage of men’s, measuring the impact of democracy on gender equality female representation in leadership roles, and share of females in high-skill jobs); educational attainment (i.e., literacy rate and enrolment in primary, secondary, and tertiary education); health and survival (i.e., life expectancy and sex ratio at birth); and political empowerment (i.e., percentage of females in parliament, women in ministerial positions, and years with a female head of state). a higher ggi value denotes greater gender equality. the gdi, another index in this context, measures gender equality in health (i.e., life expectancy), education (i.e., mean years of schooling), and income (i.e., per capita income), with higher scores indicating greater gender equality. all indices of gender equality differ significantly in terms of their composition. ggi is a broader index that takes almost all the components of gii and gdi, excluding reproductive rights, which are the essential component of gii. however, ggi takes more comprehensive measures of economic participation and political empowerment as compared to gii and gdi. table 1. description of variables. variable indicator source gender equality gender equality index (gei) undp global gender gap index (ggi) the world economic forum gender development index (gdi) undp democracy eiu democracy index (eiud) the economists intelligence unit freedom house index for civil liberties and political rights (fhd) freedom house v-dem’ electoral democracy index (edi) varieties of democracy (v-dem) women’s share in parliament women’s share in parliament (wsp) world bank development indicators economic growth/income log of gdp per capita (usd) world bank development indicators gender quota in parliament constitutionally mandatory gender quartos in lower house international idea economic inequality gini-coefficient calculated from post tax net income world inequality database good governance corruption perception index (cpi) transparency international population size log of population world bank development indicators furthermore, democracy serves as our primary independent variable, hypothesized to positively impact gender equality. like gender equality, we use three popular indices of democracy: eiud, fhd, and edi. eiud includes political culture, civil liberties, and government functioning as essential components of democracy. it measures overall democratic health of a country, including political culture. in contrast, fhd focuses only on political rights and civil liberties, making it best suited for assessing freedom and human rights, particularly in authoritarian regimes. similarly, edi primarily focuses on political rights in terms of free and fair elections along with universal voting rights, making it ideal for analyzing electoral fairness and suffrage rights. therefore, eiud is a broader index that roughly encompasses all the components of fhd and edi. higher values of eiud and edi indicate greater levels of democracy. however, the opposite is true for fhd. to ensure consistency, we invert the scale of fhd so that higher values indicate higher levels of democracy. after normalization, the value of fhd ranges between 0 and 1. for ease of comparison, eiud and edi are also taken on the same scale. among the list of control variables, we have included per capita income, income inequality, good governance, and population size. all of the control variables excluding income inequality are expected to have positive impact on gender equality. 102 singh & chaudhry 3.2 model the studies discussed in section 2 either estimated cross-section regressions or pooled regressions to examine the impact of democracy on gender equality. the regression models based on crosssectional data are not sufficient to establish causality between these two variables since they do not consider variations over time, which play an important role in examining the nature of causality. on the other hand, the pooled regression model considers the variations in the variables over time but does not control for the unit-specific heterogeneity present in panel data, which might cause the endogeneity bias and impair the efficiency of ols estimates (wooldridge, 2016). in this regard, fixed effects (fe) and random effects (re) panel data regression models may be more suitable, as they capture not only cross-sectional variations in the data but also variations over time. additionally, these models effectively handle the impact of unobserved individual heterogeneity that might affect the relationship between democracy and gender equality. the basic regression model estimated in this study may be stated as: 𝐺𝐺𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1𝑋𝑋1𝑖𝑖1−𝑖𝑖 + 𝛽𝛽2𝑋𝑋2𝑖𝑖1−𝑖𝑖 + 𝛽𝛽3𝑋𝑋3𝑖𝑖1−𝑖𝑖 + ⋯+ 𝛿𝛿𝑖𝑖 + 𝑢𝑢𝑖𝑖1−𝑖𝑖 [1] where git is gender equality indicator for country i in year t, xs are set of explanatory variables taken at one year lag, 𝛿𝛿𝑖𝑖 is time-invariant individual heterogeneity, and uit is idiosyncratic error term. the ols estimation of equation (1) leads to biased estimates if 𝛿𝛿𝑖𝑖 is correlated with one or more of the xs. even if 𝛿𝛿𝑖𝑖 is uncorrelated with all of the xs, ols still produces inefficient estimates due to the positive serial correlation in the composite error term (𝛿𝛿𝑖𝑖+ uit). therefore, it would be appropriate to estimate equation (1) using fixed effects (fe) or random effects (re) models, depending on the correlation between 𝛿𝛿𝑖𝑖 and the xs. if they are correlated, fe estimation is appropriate as it effectively handles the problem of endogeneity. re estimation produces consistent and efficient estimates when 𝛿𝛿𝑖𝑖 is uncorrelated with all the explanatory variables. the choice between fe and re models is based on the hausman test, which follows a chi-square distribution with k-1 degrees of freedom and is based on the wald criterion. 𝑊𝑊 = χ2[k − 1] = [βfe−βre]′ψ−1[βfe−βre] [2] where 𝜓𝜓 = 𝑣𝑣𝑣𝑣𝑣𝑣[βfe−βre]. rejection of null hypothesis implies the selection of fe model suitable. the above specification of the model implies a static panel data model, which does not account for changes in the behavior of the dependent variable over time, as it does not include the lagged dependent variable among the set of explanatory variables. we also attempted to estimate equation (1) within the framework of dynamic panel data models using difference gmm and system gmm. however, due to the problem of invalid instruments, as indicated by the sargan test and hansen j test, we did not proceed with these methods. measuring the impact of democracy on gender equality 4. results and discussion 4.1 descriptive statistics table 2 presents the descriptive statistics of the dependent and independent variables used in this study. among gender equality variables, gei exhibits the highest variation (51.7 percent) compared to ggi (8.8 percent) and gdi (7.6 percent). regarding democracy variables, fhd is reported with 59 percent variation, followed by edi (48.9 percent) and eiud (40.2 percent). among the control variables, cpi displays the highest variation (66.3 percent), while the log of population shows the lowest (9.6 percent). table 2. summary statistics. variable total obs. mean median min max s.d. cv gei 1997 0.376 0.396 0.0250 0.819 0.194 0.517 ggi 1933 0.685 0.690 0.450 0.880 0.0606 0.088 gdi 2128 0.936 0.959 0.464 1.04 0.0717 0.076 eiud 2226 0.549 0.577 0.113 0.993 0.221 0.402 fhd 2226 0.562 0.583 0.00 1.00 0.332 0.590 edi 2211 0.530 0.521 0.0140 0.922 0.259 0.489 log (income per capita) 2189 8.54 8.54 5.12 11.7 1.49 0.174 gini coefficient 2042 0.508 0.537 0.153 0.723 0.123 0.241 cpi 2192 4.21 3.50 0.200 75.0 2.79 0.663 log (population) 2212 16.2 16.2 12.6 21.1 1.55 0.096 source: authors’ calculation table 3 measures the correlation between gender equality and democracy variables. gei is positively correlated with both ggi (0.555) and gdi (0.636), and these correlation coefficients are statistically significant at the one percent level. furthermore, the correlation between ggi and gdi is 0.644, which is also statistically significant. the correlation coefficients among gei, ggi, and gdi align with theoretical expectations that all gender equality indices should be positively correlated. however, their values are significantly less than one, indicating that these three measures of gender equality are not perfectly interchangeable. selecting a single indicator over others could lead to substantial variations in results. therefore, to enhance the robustness of our findings, we have considered all three gender equality indicators as dependent variables in this study. table 3. correlation matrix between gender equality and democracy indices. gei ggi gdi eiu fhd edi gei 1 0.555*** 0.636*** 0.606*** 0.509*** 0.514*** ggi 1 0.644*** 0.592*** 0.507*** 0.487*** gdi 1 0.489*** 0.428*** 0.372*** eiud 1 0.903*** 0.901*** fhd 1 0.915*** edi 1 source: authors’ calculation. note: ***, **, and * indicate level of significance at 1%, 5%, and 10%, respectively. furthermore, gei, ggi, and gdi are significantly associated with the democracy indices (i.e., eiud, fhd, and edi). the values of correlation coefficients among these indices range between 0.555 and 0.644. all of the gender equality indicators are positively correlated with all of the democracy indices. it implies that countries with higher levels of democracy also have higher levels of gender 104 singh & chaudhry equality. regarding correlations among the democracy indices, all are positively and significantly correlated with each other. the values of these correlation coefficients are substantially high and range from 0.901 to 0.915. this suggests that, unlike gender equality indices, substituting one democracy index for another is unlikely to substantially alter the results. table 4 compares various political regimes—full democracy, flawed democracy, hybrid regime, and authoritarian regime—using analysis of variance (anova). the economist intelligence unit defines full democracies as countries with substantial political freedom and civil liberties. these nations have robust political institutions, effective governance, independent judiciary, and free press. free and fair elections and respect for the rule of law are common characteristics of these regimes. in contrast, flawed democracies hold free elections but exhibit weaknesses in governance, political culture, and participation. while basic civil liberties are respected, issues such as media restrictions and corruption may prevail. hybrid regimes, a category following flawed democracies, display some democratic features but also significant authoritarian elements. elections might not be entirely free or fair in these countries. apart from this, they may also suffer from corruption, underdeveloped political culture, and weak rule of law. authoritarian regimes, the fourth category, exhibit limited political pluralism, with many lacking meaningful elections. civil liberties are restricted, and power is frequently concentrated in the hands of a single leader or party. table 4. gender equality across political regimes: anova. regime type gei ggi gdi full democracy 0.860 0.741 0.974 flawed democracy 0.651 0.697 0.970 hybrid democracy 0.520 0.668 0.916 authoritarian regime 0.528 0.642 0.894 f 421.4*** 278.6*** 214.9*** source: authors’ calculation. note: ***, **, and * indicate level of significance at 1%, 5%, and 10%, respectively. as clearly demonstrated by table 4, full democracies have the highest level of gender equality across all three indicators. the average scores of gei, ggi, and gdi for this category are 0.86, 0.741, and 0.974, respectively, over the period. following full democracies, flawed democracies perform better compared to the remaining two categories of political regime. authoritarian regimes are the worst performers across all three indicators of gender parity. the results of anova indicate that the four categories of political regimes significantly differ in terms of gei, ggi, and gdi. as the level of democratization increases, so does the level of gender equality. however, this does not confirm a causal connection between the two. 4.2 empirical findings this section presents a systematic analysis of democracy’s impact on gender equality, which starts from unconditional cross-section regression models and ends with conditional panel data models. the regression models estimated in this section can be divided into two categories: cross-section and panel data regressions. in each category, two types of regression models have been estimated: unconditional and conditional regression models. unconditional models do not account for control variables, whereas conditional models do. for this purpose, we have considered three dependent measuring the impact of democracy on gender equality variables, viz. gei, ggi and gdi, as indicators of gender equality. for each dependent variable, we have estimated three equations: model 1, model 2 and model 3. and for each model, a different democracy indicator is used while keeping other variables the same. the results of unconditional regression models demonstrate that all of the democracy variables have a significantly positive impact on all of the gender equality indicators. like inglehart et al. (2003) and rizzo et al. (2007), it implies that gender equality is positively associated with democracy in the cross-sectional data. table 5. conditional cross-section regression models. regression model independent variables dependent variable gei ggi gdi constant 0.950686 0.759991 0.782012*** eiud −0.0021385 0.0156806*** 0.011067*** income 0.02991*** −0.00474 0.019863*** model 1 inequality −0.520292*** −0.107158** 0.028634 cpi 0.0077567 0.00338 −0.006977 population 0.0063749 −0.003185 −0.008649** adjusted r2 0.837652 0.467599 0.385585 n 134 124 136 constant 0.949696 0.770658*** 0.775691*** fhd −0.011099 0.0708587*** 0.055722*** income 0.029848*** −0.004954 0.019857*** model 2 inequality −0.518189*** −0.128901*** 0.0167632 cpi 0.0075008 0.0062825* −0.005372 population −0.0065641 −0.0020433 −0.007603** adjusted r2 0.837571 0.397550 0.368229 n 134 124 136 constant 0.930551*** 0.765700*** 0.780891*** edi −0.04317 0.075906*** 0.0448628 income 0.030101*** −0.005412 0.0194056** model 3 inequality −0.537295*** −0.124458** 0.011281 cpi 0.009246 0.007097* −0.003469 population 0.006311 −0.002602 −0.00799** adjusted r2 0.839228 0.384012 0.350333 n 134 124 136 source: authors’ calculation. note: ***, **, and * indicate level of significance at 1%, 5%, and 10% respectively. table 5 demonstrates the results of conditional cross-section regression models. for this purpose, economic growth, income inequality, good governance, and population size are included as additional control variables. all of these variables are expected to have significant impact on gender equality indicators. table 5 reveals that when gei is taken as the dependent variable, all of the democracy indices have a statistically insignificant impact on gender equality. however, when ggi is used as the dependent variable, all of the democracy indices have a significantly positive impact on gender equality. and when gdi is used as the proxy for gender equality, two democracy variables, eiud and fhd, show a significantly positive impact, while the effect of edi is found to be statistically insignificant. it indicates that democracy has either an insignificant or a positive impact on gender equality. these results also suggest that the expected impact of democracy is dependent on the choice of a particular gender equality index. thus, it appears that the impact of democracy on gender equality is index-sensitive. however, replacing the democracy indicators with one another does not cause substantial variations in the impact of democracy on gei, ggi and gdi. 106 singh & chaudhry table 6. conditional panel data regression models. regression model independent variables dependent variable gei ggi gdi constant −4.16577 −0.246593*** 0.03406 eiud 0.000232 −0.00116 0.00282*** income 0.013605** 0.010554*** 0.00677*** model 1 inequality −0.255653*** 0.0009639 0.05943*** cpi 8.955e-05 0.00016 0.000152** population 0.21409*** 0.036816*** 0.03962*** hausman test 242.08*** 279.42*** 73.24*** estimated model fe fe fe constant 4.13987*** −0.23774*** −0.00129 fhd −0.007389 −0.00277 0.01373* income 0.01381*** 0.010417*** 0.00698** model 2 inequality −0.256917*** 0.001743 0.05657*** cpi 9.402e-05 0.00016 0.00015** population 0.21287*** 0.03592*** 0.042351*** hausman test 217.77*** 225*** 72.26 estimated model fe fe fe constant 4.20929*** −0.2373** 0.0291 edi 0.035817** −0.014145*** 0.01134** income 0.01302*** 0.010697*** 0.00703*** model 3 inequality −0.244552*** −0.00186 0.05866** cpi 7.362e-05 0.000162 0.00015*** population 0.21545*** 0.03629*** 0.04054*** hausman test 220.35*** 112.64 65.72*** estimated model fe fe fe source: authors’ calculation. note: ***, **, and * indicate level of significance at 1%, 5%, and 10%, respectively. as far as the impact of control variables is concerned, it is important to mention that the regression coefficients of these variables, as well as their levels of significance, follow a similar pattern across the three regression models demonstrated in table 5. this is because, as already mentioned, these three models differ only in terms of democracy variables. in terms of control variables, these models are the same. thus, it is appropriate to interpret the results of any one regression model (e.g., model 1) regarding the control variables. model 1 in table 5 shows that the economic growth has a profound impact on gender equality, as its regression coefficients are positive and highly significant in the case of gei (0.0229) and gdi (0.0198). this implies that countries with higher levels of per capita income also have higher levels of gender parity. like economic growth, income inequality also emerges as a robust predictor of gender equality, as its impact on two gender equality indicators, gei (-0.518) and ggi (-0.129), is negative and highly significant. this suggests that countries with higher levels of economic inequality also have higher levels of gender inequality. furthermore, population size is reported to have a significantly negative impact on gdi (-0.0086), indicating that countries with larger populations tend to have lower levels of gender equality. the cross-sectional regressions indicate a roughly positive association between democracy and gender equality. however, merely relying on cross-sectional data is not sufficient to establish causality between these two variables. in this regard, panel data regression models may be more suitable, as they capture not only cross-sectional variations in the data but also variations over time. additionally, these models effectively handle the impact of unobserved individual heterogeneity that might affect the relationship between democracy and gender equality. for this purpose, we have measuring the impact of democracy on gender equality estimated unconditional regression models taking the one-year lag2 of the democracy variables. the regression coefficients of the democracy variables—eiud, fhd, and edi—are found to be statistically insignificant regarding their impact on gender equality variables (i.e., gii, ggi, and gdi). thus, these results are not reported here. table 6 shows the results of conditional panel data regression models. based on the hausman test, it has been found that country-specific heterogeneity is significantly correlated with democracy in model 1, model 2, and model 3, as the chi-square values under this test are highly significant for all three models. in this case, the fe model produces consistent and efficient estimates of the regression coefficients compared to the re model. thus, the results presented in table 6 are estimated using the fe model. like conditional cross-section regression models, the results of the conditional panel data regressions also indicate that the impact of democracy on gender equality is not certain and is highly sensitive to the replacement of various gender equality indicators with one another, which casts doubt on the causation between gender equality and democracy. for instance, if gei is taken as a dependent variable, only edi is reported with a statistically significant impact on gender equality (i.e., 0.0358). if ggi is taken as a dependent variable, eiud and fhd again have an insignificant impact on gender equality; while edi is reported with a significantly negative impact (0.0141) on it. the negative effect of edi on ggi may be attributed to its methodology of ggi, as it measures the relative gender gaps rather than the absolute levels of development. it implies that a country with a low level of democracy can rank high in terms of ggi if the gender gap between men and women is small even if both men and women have low levels of overall development. for example, countries (e.g., rwanda, nicaragua, and bangladesh) with lower levels of democracy often rank higher in terms of gender equality than highly democratic nations like japan and the usa. interestingly, eiud and fhd have a significantly positive impact on gender equality if gdi is taken as a proxy of it. in contrast, if ggi and edi are taken as the proxy of gender equality and democracy, respectively, it might be concluded that democracy has a negative impact on gender equality. it implies that one may reach different conclusions depending on which indicator(s) is(are) used for gender equality and democracy. thus, the impact of democracy is not robust across the various indicators used for measuring gender equality. among the control variables, economic growth and population size are found to have a substantially positive impact on gender equality, as the effect of these two variables on gender equality remains consistent when the gender equality indicators are interchanged. the positive impact of per capita income on gender equality implies that economic growth leads to modernization by weakening traditional values and attitudes, which increases gender equality in social, political, and economic spheres (inglehart et al., 2003). additionally, economic growth enhances gender equality by providing more opportunities for females to acquire qualifications that are needed to serve top political and professional posts, which affects gender equality positively (högström, 2015). 2 we have also estimated unconditional as well as conditional panel data regression models by taking threeand five-year lags of democracy variables, assuming that the effect of democracy on gender equality may take a longer period to appear. however, since the regression coefficients for democracy were not significant, we have not reported the results of these models. 108 singh & chaudhry like economic growth, population size also has a positive impact on gender equality. in countries with larger populations, greater specialization and differentiation among the populace tend to lead to increased political diversity. this higher level of political diversity generates more demands for political equality, placing pressure on political parties and leaders to enhance political equality. in response to these demands, political parties and leaders strive to achieve greater political equality, which can result in outcomes such as increased representation of women in cabinets and consequently increased gender equality (högström, 2015). regarding income inequality, its impact on gender equality is uncertain because the sign of its regression coefficient changes depending on the gender equality indicator used. for instance, it reduces gender inequality when gei is taken as the dependent variable. on the other hand, the opposite result is observed when gdi is the dependent variable. this inconsistency in the impact of income inequality across various gender equality indicators makes its overall effect uncertain. table 7. impact of various components of democracy on gender equality. independent variables dependent variable gei ggi gdi constant 3.5416*** −0.00698 0.10242 electoral process & pluralism 0.00073 1.332e-05 0.00041 government functioning 0.00290** −0.00211 0.00326*** political participation 0.01044*** 0.00512*** 0.00150*** political culture −0.00696 −0.000457 −0.00244* civil liberties −0.01217 −0.00378 −0.00050 income 0.01248** 0.01131*** 0.00510 inequality −0.2333*** −6.424e-05 0.06365*** population 0.1838*** 0.02496*** 0.03691*** hausman test 211.53*** 122.47*** 65.83** estimated model fe fe fe source: authors’ calculation. note: ***, **, and * indicate level of significance at 1%, 5%, and 10% respectively. in table 7 we try to test the impacts of various components of democracy on gender equality indices since it is possible that some of the aspects of democracy may be more important than others as far as gender equality is concerned. while estimating the impacts of various aspects of democracy on gender equality, we continue to keep gei, ggi, and gdi as dependent variables, since the replacement of these variables for one another causes significant variations in the magnitude and direction of the effect of democracy on gender equality. however, we prefer eiud over fhd and edi as the proxy of democracy for two reasons: first, eiud includes electoral process and pluralism, functioning of government, political participation, civil liberties, and political culture as the essential components of democracy, whereas fhd includes the first four components of eiud and excludes political culture. similarly, edi is more focused on electoral process, political participation, and freedom of expression. in this respect, eiud is a broader index of democracy that covers almost all the aspects included in fhd and edi. second, eiud is highly correlated with both fhd and edi, and its substitution with them does not lead to substantial variations in the impact of democracy on gender equality indices. measuring the impact of democracy on gender equality table 8. impact of gender quota in parliament on gender equality. independent variables dependent variable gei ggi gdi constant −4.90407*** −0.726759*** −0.234226*** d*eiud 0.029004*** 0.0107069*** −0.0073467 eiud −0.034911*** −0.0148846*** 0.0110362*** income 0.0217*** 0.0183303*** 0.0112911*** urbanization 0.000495 0.0003743 9.103e-05 inequality −0.268646 −0.012332 0.045869*** cpi 0.0001797 0.000241 0.0001976*** population 0.273450*** 0.077355*** 0.0630053*** hausman test 463.9*** 227.39*** 184.56*** estimated model fe fe fe source: authors’ calculation. note: ***, **, and * indicate level of significance at 1%, 5%, and 10% respectively. d = 1, if a country implemented quota for females and zero otherwise. table 7 shows that political participation has a significantly positive impact on all of the gender equality indices. political participation measures how actively citizens and various social organizations engage in elections and form pressure groups or lobbies. it includes voter turnout rates, engagement in civic and political activities, and the role of civil society in shaping the policies. the positive regression coefficients for political participation imply that the active participation of voters, especially female voters, held leaders accountable through bottom-up pressures to form and implement the policies promoting gender equality. further, the government's functioning has a significantly positive impact on gei (0.0029) and gdi (0.0033)3. its impact on ggi is reported to be statistically insignificant. government functioning includes accountability of government to the people, lack of corruption, and balance of power between the executive, legislature, and judiciary. it indicates that the countries where democratic institutions are strong and accountable to the public are likely to have higher levels of gender equality. the other components of eiud, like electoral process and pluralism, political culture, and civil liberties, have statistically insignificant impact on all of the gender equality indices. thus, it can be said that merely having the free and fair election and protection of civil liberties are not enough for creating the bottom-up pressures for the policy formulation to promote gender equality without active participation of masses and strong democratic institutions. the impact of democracy on gender equality is not as straightforward as it may seem theoretically. it is possible that democracy may require a longer time horizon to bring about positive changes in gender equality through bottom-up pressures. these changes cannot be tested effectively by the data of a time span of 14 years. however, we tried to test the impact of top-down state policies like the quota of women in parliament mandated directly by the national constitution. these quotas are harder to change since constitutional amendments require complex legal procedures. to include the gender quota in the regression model, we have created a dummy variable that takes the value of one if a particular country has implemented a gender quota in parliament and zero otherwise. here, it is important to mention that while estimating the fe models, all the time-invariant variables like the presence or absence of quota will be eliminated due to time demeaning of the variables. thus, the impact of the quota cannot be estimated directly by the fe specification of the model. for this purpose, we have interacted the dummy variable for quota with the democracy variable. the 3 while estimating the regression model in table 7, we do not include cpi (i.e., transparency) among the control variables since it is already accounted for in government functioning. 110 singh & chaudhry interaction coefficient measures the additional impact of democracy on gender equality when the quota for women in parliament is present. these results are presented in table 8. the interaction coefficient is statistically significant and positive when gei and ggi are taken as dependent variables. interestingly, the impact of democracy on gei and ggi turns significantly negative after the inclusion of gender quota among the list of control variables. the significantly positive values of the interaction coefficient imply that the implementation of gender quotas along with democracy will significantly improve gender equality in terms of gei and ggi. this result also provides an explanation of the statistically insignificant impact of eiud on gei and ggi in table 6. the gross impact of democracy on gender equality is split into two mutually exclusive components: the interaction coefficient and the coefficient of eiud. when a significantly positive interaction coefficient is combined with a significantly negative net effect of eiud, the gross impact of democracy is statistically insignificant, as shown in table 6. further, the interaction coefficient is reported as statistically insignificant while the coefficient of eiud is found to be significantly positive in the case of gdi. however, the coefficients of the other control variables remain almost unchanged from those in table 6. 4.3 robustness tests many cultural, historical, and geographical factors may have a significant impact on the relationship between gender equality and democracy. these factors may weaken or strengthen the magnitude of this relationship and even change the direction of their association. we include political regime type, economic status, religion, and region among these factors. all of these factors are time invariant and cannot be included directly as explanatory variables in the fe model, because the time-demeaned fe model eliminates all the time-invariant variables along with unobserved country-specific heterogeneity. on the other hand, lsdv estimation of the fe model may encounter the problem of multicollinearity between country dummies and the dummies used for other time-invariant variables. therefore, to take into account the impact of these time-invariant factors, we have divided the entire sample into subsamples based on the characteristics of these variables, and then estimated a separate regression equation for each subsample. as in tables 7 and 8, while estimating these regression equations, we continue to use gei, ggi, and gdi as dependent variables and eiud as the proxy for democracy. the results calculated on the basis of sub-samples are presented below. type of political regime although we estimated the impact of political regime on gender equality in the previous section, where we measured the regime type on the ordinal or ratio scale using the entire sample, here we take it as a categorical variable. based on the relative strength of democracy, the economist intelligence unit divides the countries across the world into four categories: full democracy, flawed democracy, hybrid democracy, and authoritarian regime. as a country moves from full democracy to an authoritarian regime, its level of democracy declines. the difference among these political regimes is attributed to the strength and quality of democratic institutions, political culture, and good governance. therefore, we try to take into account the impact of institutional quality on the measuring the impact of democracy on gender equality relationship between gender equality and democracy by estimating a separate regression equation for each political regime. the results are presented in table 9. table 9. impact of democracy on gender equality across political regimes. regime type independent variable dependent variable gei ggi gdi full democracy constant 4.4870*** −0.8603 0.46150*** eiud 0.0071 0.00953 0.00270 income 0.00244 0.00822 0.00807*** inequality -0.05533 −0.14258*** −0.00135 cpi 1.22e-05 0.000322 1.292e-05 population 0.18025*** 0.035474 0.011550 hausman test 79.24*** 21.42*** 16.78*** estimated model fe fe fe flawed democracy constant 1.32161*** −0.9594*** −0.3221 *** eiud 0.000832 −0.0069* −0.00045 income 0.04927*** 0.00324 0.00335 ** inequality 0.000329 0.07496* −0.01832 cpi -0.00026 −0.00017 6.769e-05 population 0.00378 0.08476*** 0.07286*** hausman test 49.29*** 51.60*** 101.78*** estimated model fe fe fe hybrid democracy constant 0.92742*** −0.15459*** 0.6554*** eiud -0.00162 −0.00227 0.00178 income 0.04843*** 0.026522*** 0.00445 inequality -0.26561*** 0.052354* 0.04963 cpi 0.01045** 0.00553** 0.00433** population 0.02268** 0.028452* −0.00514 hausman test 13.07* 28.31*** 11.92 estimated model re fe re authoritarian regime constant 5.2846*** −0.2781*** −0.1375 eiud 0.00402* 0.00320** 0.00399** income -0.004711 0.00512** 0.00551** inequality -0.580915*** 0.07888 0.20015*** cpi 0.00485* 0.001768 0.00289** population 0.31307*** 0.033784*** 0.04358*** hausman test 130.43*** 60.47*** 30.59*** estimated model fe fe fe source: authors’ calculation. note: ***, **, and * level of significance at 1%, 5%, and 10% respectively. most of the countries experiencing full democracies have a strong historical heritage of democratic values, developed democratic institutions, and political cultures conducive to democracy. most of these countries are high-income european nations. table 9 shows that democracy does not cause any significant improvement in gender equality in full democracies as well as in flawed democracies. however, this result may be interpreted in a different way: these countries already have achieved the highest peak of democracy as well as gender equality, and there is a lesser possibility of further improvement in these variables. this, as a consequence, can make the impact of democracy on gender equality insignificant. similarly, the effect of democratization is not significant for any of the gender equality indices in hybrid regimes. on the other hand, democracy has a positive impact on ggi (0.0032) and gdi (0.004) at the 5 percent level of significance in authoritarian regimes. this implies that these countries are making significant progress toward higher levels of democracy and gender equality over time. furthermore, economic growth and population size continue to have a positive impact on gender equality across all types of political regimes. 112 singh & chaudhry economic status economic status, or the level of economic development, is an important factor that characterizes the modernization of society. the transformation of society from an agriculture-based traditional society to a modern industrialized society produces significant cultural changes like democratization and changes in the traditional roles of women. industrialization leads to occupational specialization, rising educational levels, and increasing levels of income. thus, modernization is an important factor that affects the level of gender equality in a society. for this purpose, we have divided the countries into four subsamples according to the level of per capita income as per the definition adopted by the world bank in 2019 and extended this definition for the preceding years, since the income categories classified by the world bank fairly remained constant over the time considered in the study. the world bank classifies the countries into four categories on the basis of per capita income: high-, upper-, middle-, and low-income countries. table 10. impact of democracy on gender equality across income categories. economic status independent variables dependent variable gei ggi gdi high income countries constant 7.0254*** −0.5346*** 0.2352*** eiud 0.00127 0.00162 −0.00153 income 0.00184 0.01268** 0.00685** inequality −0.34573*** −0.04775 0.00369 cpi 0.000124 0.00010 2.66e-05 population 0.35029 0.01313 0.04232*** hausman test 190.40*** 51.98*** 122.04*** estimated model fe fe fe upper middle income countries constant 2.4822*** 0.03975 0.1527*** eiud 0.00156 0.00227 0.0043*** income 0.02705*** 0.00131 0.0102*** inequality 0.05899 0.04350*** 0.0690*** cpi 0.00769 0.00657 0.0009 population 0.0787*** 0.01809 0.0340*** hausman test 51.66*** 64.19 27.91*** estimated model fe fe fe lower middle income countries constant 2.9003*** −1.1819*** −0.6972*** eiud 0.00505* −0.00174 −0.0010 income 0.03753*** 0.01124*** 0.00784*** inequality −0.38379** 0.04886 0.01177 cpi 0.00078 −4.75e-05 0.00044 population 0.11920*** 0.09613*** 0.08626*** hausman test 18.07** 49.60*** 53.82*** estimated model fe fe fe low income countries constant 2.0537** * −1.2127*** 0.3035 eiud −0.00531*** −0.00133 0.0048 income −0.00617*** 0.01512*** −0.0092 inequality 0.09051 −0.04279 0.4510*** cpi 0.00535** 0.00095 0.0259*** population 0.06719*** 0.10861*** 0.0060 hausman test 14.69** 237.20*** 7.57 estimated model fe fe re source: authors’ calculation. note: ***, **, and * level of significance at 1%, 5%, and 10% respectively. table 10 shows that democracy has an insignificant impact on gender equality in highincome and lower-middle-income countries. in upper-middle-income countries, democracy has a positive impact only on gdi (0.0043), whereas its impact on gei and ggi is statistically insignificant. measuring the impact of democracy on gender equality on the other hand, its impact on gei is significantly negative in low-income countries. this implies that the relationship between democracy and gender equality is not robust across various income categories. similar to political regimes, economic growth and population size consistently have significant and positive impacts on gender equality indicators across all income categories. religion religion also seems to be a major reason why many nations with a strict islamic background have often ranked at the bottom of the list worldwide in terms of gender equality (inglehart et al. 2003), including even the more affluent arab societies like kuwait and saudi arabia, as well as egypt, jordan, and lebanon (abu-zayad, 1998 and rizzo, 2007). post-industrial societies with a historical prevalence of catholicism are considered to represent more traditional attitudes towards women and families than protestant religions (rule 1987). thus, it is suspected by scholars that religion may have a significant impact on the democracy-gender equality relationship. in order to investigate the magnitude of this relationship across various religions, we have divided the sample into three subgroups: christian countries, muslim countries, and non-christian-non-muslim countries, based on the religion of the majority in a particular country. table 11. impact of democracy on gender equality across religions. religion independent variables dependent variable gei ggi gdi christian countries constant 1.6533*** −0.74785*** −0.1957** euid −0.01238* −0.00403 −0.0003 income 0.02924*** 0.01510** 0.00876*** inequality −0.14595*** −0.02004 0.06319*** cpi −6.50e-05 0.000157 8.75e-05** population 0.04907** 0.07449** 0.05843*** hausman test 113.51*** 81.37** 138.29*** estimated model fe fe fe muslim countries constant 5.1017*** −0.09978 0.13459 euid 0.00022 −0.00032 0.00626*** income −0.01192 0.00575** 0.00223 inequality −0.53631*** 0.05229 0.10831*** cpi 0.02385* 0.00314 0.01217*** population −0.2894*** 0.01154* 0.02149*** hausman test 32.64*** 61.68*** 34.41*** estimated model fe fe fe non-christian non-muslim countries constant 3.2999*** −0.7560* 0.8352*** euid 0.0201*** 0.00218 −0.00065 income −0.0091 −0.01118* 0.00887*** inequality 0.08321 0.07448* 0.01540 cpi −0.00045 −0.00518* −0.00052 population 0.08450** 0.05364** −0.00499 hausman 59.04*** 18.91** 13.55* estimated model fe fe re source: authors’ calculation. note: ***, **, and * level of significance at 1%, 5%, and 10% respectively. the results of the regression models for religion-specific groups of countries are demonstrated in table 11. the table shows that the effect of democracy and gender equality is not robust across the subsamples based on religions. in christian countries, democracy has an insignificant impact on gei, ggi, and gdi. in muslim countries, its impact is significant only on gdi (0.0063). similarly, democracy’s impact is significant only on gei (0.0201) in non-christian-non114 singh & chaudhry muslim countries. among the control variables, the effect of per capita income on gender equality is not homogeneous across all religions. the regression coefficients of per capita income are found to be statistically significant and positive only in christian countries across all gender equality indices. it implies that economic growth leads to positive changes in gender equality in these countries. in the case of the remaining two categories, the impact of economic growth on gender equality is insignificant for at least two gender equality indicators. region the region or geographical location of a country is a time-invariant characteristic that may have a significant impact on the gender equality-democracy conundrum. it is a fact that countries sharing boundaries also share history and cultural traditions. for example, the foundations of asian societies rest on collectivism, whereas european societies are relatively more individualistic (oyserman and kemmelmeier, 2002). these common historical and cultural factors, along with geography, play a significant role in shaping this relationship. therefore, to take into account these characteristics, we have divided the countries into four regions: asia, africa, europe, and latin america. it is important to mention that we have included the usa, canada, australia, and new zealand in the subsample containing european countries, since these countries are culturally and historically more close to europe than other regions. table 12 shows the impact of democracy on gender equality in these four regions. democracy’s impact is not homogeneous across all regions and is not robust across all gender equality indices. in asia, democracy has a significantly positive impact on two out of three gender equality indicators (e.g., gei and gdi). in europe and north america, democracy has a significantly positive impact only on ggi, while its impact is insignificant on the other two indices. in africa and latin america, democracy’s impact on all gender inequality indices is found to be statistically insignificant. further, economic growth positively affects at least two gender equality indices in asia and africa. in europe and north america, its impact is significantly positive only on gdi. in the case of latin america, all the regression coefficients of per capita income are reported as statistically insignificant. as far as size of population is concerned, its impact on gender equality is almost positive on gender equality indices across all regions except europe and north america. in this region, the level of significance as well as the sign of the coefficient of population size changes significantly as we replace one gender equality indicator with the other. thus, the empirical findings of this study suggest that the impact of democracy on gender equality is not certain. in most cases, this impact is reported as statistically insignificant. in some cases, the effect of democracy is reported as positive, while in others, it is negative. it suggests that the impact of democracy is highly sensitive to the interchangeability of various gender equality indices with one another, which may cause significant variations in effect of democracy. however, it also may be suspected that democracy may produce different effects on gender equality in different environments like type of political regime, economic status or level of modernization, and culture (which is highly affected by region and geographical location). thus, to check the robustness of the effect of democracy on gender equality, we have estimated the regression equations for various gender equality indices in section 4.3 by dividing the sample into subsamples based on these measuring the impact of democracy on gender equality characteristics. even after accounting for these characteristics, the results suggest that the effect of democracy on gender equality remains uncertain and highly sensitive to the selection of the dependent variable. table 12. impact of democracy on gender equality across regions. region independent variables dependent variable gei ggi gdi asia constant 5.8555*** −0.0923 0.23185*** eiud 0.02022*** 0.00112 0.00388** income 0.02267*** 0.00771*** 0.00178 urbanization −2.23e-05 0.00071* 0.00028 health 0.01792*** 0.00464*** 0.00299*** inequality −0.18701** 0.02676 0.07302*** cpi 0.00655* −0.00192 0.00229** population 0.25842*** 0.01929*** 0.02320*** hausman test 100.99*** 35.88*** 99.47*** estimated model fe fe fe africa constant 3.7176*** −1.2931*** −0.32296* eiud 0.0038* −0.00042 0.00154 income 0.0077 0.01007** 0.01265*** urbanization 0.0004 0.00162** 0.00133** health 0.0016*** 0.00023 0.00118*** inequality −0.0308 0.02920 0.08458** cpi −0.00138 0.00333 0.00687*** population 0.18337 0.11003*** 0.05857*** hausman test 103.92*** 64.66*** 31.39*** estimated model fe fe fe europe and north america constant 3.2191*** −0.8756 0.4677*** eiud −0.0049 0.01420*** 0.00059 income −0.0011 −0.00096 0.00748*** urbanization 0.00033 −9.13e-05 −0.00017 health 0.02192*** 0.01211*** −0.00028 inequality 0.02355 −0.00629 0.01055 cpi 7.55e-05 0.00011 1.86e-05 population −0.08681*** 0.03526 0.02844*** hausman test 53.84*** 50.41*** 48.72*** estimated model fe fe fe latin america constant 8.6512*** −4.9365*** −0.36196** eiud −0.00055 −0.00124 −0.00049 income −0.00207 −0.00287 0.004054 urbanization −0.00258* −0.00154 −0.00032 health 0.00858*** 0.00124 0.00177* inequality −0.12699 0.11028 0.06285** cpi −0.00622** −0.00074 −0.00092 population 0.46378*** 0.33056*** 0.068324 hausman test 69.43*** 80.71*** 35.092*** estimated model fe fe fe source: authors’ calculation. note: ***, **, and * level of significance at 1%, 5%, and 10% respectively. our findings indicate that in the modern world, every society (whether democratic or undemocratic) recognizes the importance of human rights and, therefore, values women as equally important as men. today, every country is an open society in the sense that there is a free flow of ideas from one society to another. in this scenario, any demand that is rational and logical on the grounds of human rights cannot be easily discarded by the government. moreover, if the government of a country does not take the issue of human rights seriously, it may face criticism and pressure not only from its citizens but also from the international community. no rational government, irrespective of its regime type, wants to be seen as a villain in the eyes of its citizens and the 116 singh & chaudhry international community. moreover, if the neighboring countries are democratic and have more gender-inclusive policies, an authoritarian state may have to adopt similar policies to establish its image as a liberal state in front of its citizens. therefore, it may be argued that as the world democratizes as a whole and follows more humanitarian policies, authoritarian states will try to compete with democratic states on the basis of their policies. thus, policies of democratic governments effectively shape the policies of non-democratic governments. in this respect, both types of governments promote the overall well-being of their citizens and pay equal attention to issues regarding gender equality. 5. conclusion the study presents a systematic analysis of democracy’s impact on gender equality by estimating several cross-sectional and panel data regression models for various gender equality indicators, such as the gei, ggi, and gdi. the study also applies several robustness tests by dividing the sample into subsamples based on political regimes, income categories, religions, and regions. the cross-sectional regressions indicate a roughly positive association between democracy and gender equality; however, the results of the conditional panel data regressions indicate that the impact of democracy on gender equality is not certain and is sensitive to the replacement of various gender equality indices with one another, which casts doubt on the causation between gender equality and democracy. however, replacing the democracy indicators with one another does not cause substantial variations in the impact of democracy on gei, ggi, and gdi. the results also reveal that political participation, government functioning, and implementation of gender quotas in parliament positively affect the society’s level of gender equality. among the control variables, economic growth and size of population have a positive role in enhancing the level of gender equality. the positive impact of economic growth seems to support the importance of modernization in gender equality. the scope of this study is limited to measuring the impact of democracy on gender equality rather than democratization. it is possible that the process of democratization is more important than the absolute levels of democracy in promoting gender equality. however, democratization is a long-term process that cannot be adequately measured using data spanning only fourteen years. future research should focus on measuring the impact of democratization using data from a longer time period. measuring the impact of democracy on gender equality references abu-zayd, g. (1998). in search of political power: women in parliament in egypt, jordan and lebanon. in a. karam (ed.), women in parliament: beyond numbers (pp. 43-54). stockholm: international idea. andersen, t. b. (2023). does democracy cause gender equality? journal of institutional economics, 19(2), 210–228. https://doi.org/10.1017/s1744137422000236 beer, c. (2009). democracy and gender equality. studies in comparative international development, 44(2), 212–227. https://doi.org/10.1007/s12116-009-9043-2 blankenship, j., & kubicek, p. (2018). democratization and gender equality in sub-saharan africa. the journal of the middle east and africa, 9(1), 27–50. https://doi.org/10.1080/21520844.2018.1449458 boix, c., miller, m., & rosato, s. (2013). a complete data set of political regimes, 1800–2007. comparative political studies, 46(12), 1523–1554. https://doi.org/10.1177/0010414012463905 childs, s., & krook, m. l. (2006). gender and politics: the state of the art. politics, 26(1), 18–28. https://doi.org/10.1111/j.1467-9256.2006.00247.x cho, s. y. (2014). international women's convention, democracy, and gender equality. social science quarterly, 95(3), 719–739. https://doi.org/10.1111/ssqu.12069 dahlerup, d. (1988). from a small to a large minority: women in scandinavian politics. scandinavian political studies, 11(4), 275–298. https://doi.org/10.1111/j.1467-9477.1988.tb00372.x franceschet, s., & piscopo, j. m. (2013). equality, democracy, and the broadening and deepening of gender quotas. politics & gender, 9(3), 310–316. https://doi.org/10.1017/s1743923x13000184 ghai, y., & cottrell, j. (2009). the rule of law and access to justice. in y. ghai & j. cottrell (eds.), marginalized communities and access to justice (pp. 11–32). london: routledge-cavendish. https://doi.org/10.4324/9780203866405 högström, j. (2015). do development and democracy positively affect gender equality in cabinets? japanese journal of political science, 16(3), 332–356. https://doi.org/10.1017/s1468109915000225 inglehart, r., norris, p., & welzel, c. (2003). gender equality and democracy. in r. l. inglehart (ed.), human values and social change (pp. 91–115). leiden: brill. https://doi.org/10.1163/9789047404361_007 kenworthy, l., & malami, m. (1999). gender inequality in political representation: a worldwide comparative analysis. social forces, 78(1), 235–268. https://doi.org/10.1093/sf/78.1.235 lovenduski, j. (1997). gender politics: a breakthrough for women? parliamentary affairs, 50(4), 708–719. https://doi.org/10.1093/oxfordjournals.pa.a028766 mansbridge, j. (1999). should blacks represent blacks and women represent women? a contingent "yes." the journal of politics, 61(3), 628–657. https://doi.org/10.2307/2647821 o’brien, d. z., & piscopo, j. m. (2018). electing women to national legislatures. in a. c. alexander, c. bolzendahl & f. jalalzai (eds.), measuring women's political empowerment across the globe: strategies, challenges, and future research (pp. 139–163). cham, switzerland: springer. https://doi.org/10.1007/978-3-319-64006-8_7 oyserman, d., coon, h. m., & kemmelmeier, m. (2002). rethinking individualism and collectivism: evaluation of theoretical assumptions and meta-analyses. psychological bulletin, 128(1), 3–72. https://doi.org/10.1037/0033-2909.128.1.3 phillips, a. (1995). the politics of presence. oxford: clarendon press. pitkin, h. (1972). the concept of representation. berkeley, ca: university of california press. reynolds, a. (1999). women in the legislatures and executives of the world: knocking at the highest glass ceiling. world politics, 51(4), 547–572. https://doi.org/10.1017/s0043887100009254 rizzo, h., abdel-latif, a.-h., & meyer, k. (2007). the relationship between gender equality and democracy: a comparison of arab versus non-arab muslim societies. sociology, 41(6), 1151–1170. https://doi.org/10.1177/0038038507082320 rule, w. (1987). electoral systems, contextual factors and women's opportunity for election to parliament in twenty-three democracies. western political quarterly, 40(3), 477–498. https://doi.org/10.1177/106591298704000307 vaccaro, a. (2021). comparing measures of democracy: statistical properties, convergence, and interchangeability. european political science, 20(4), 666–684. https://doi.org/10.1057/s41304021-00328-8 wooldridge, j. m. (2016). introductory econometrics: a modern approach (6th ed.). mason, oh: cengage learning. https://doi.org/10.1017/s1744137422000236 https://doi.org/10.1007/s12116-009-9043-2 https://doi.org/10.1080/21520844.2018.1449458 https://doi.org/10.1177/0010414012463905 https://doi.org/10.1111/ssqu.12069 https://doi.org/10.1111/j.1467-9477.1988.tb00372.x https://doi.org/10.1017/s1743923x13000184 https://doi.org/10.4324/9780203866405 https://doi.org/10.1017/s1468109915000225 https://doi.org/10.1163/9789047404361_007 https://doi.org/10.1093/sf/78.1.235 https://doi.org/10.1093/oxfordjournals.pa.a028766 https://doi.org/10.2307/2647821 https://doi.org/10.1007/978-3-319-64006-8_7 https://doi.org/10.1037/0033-2909.128.1.3 https://doi.org/10.1017/s0043887100009254 https://doi.org/10.1177/106591298704000307 1. introduction 2. review of literature 3. methodology 4. results and discussion 5. conclusion references european journal of government and economics 12(1), june 2023, 5-38 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 wagner’s hypothesis in europe: a causality analysis with disaggregated data ivan d trofimova * a kolej yayasan saad (kys) business school, malaysia * corresponding author at: ivan.trofimov@kysbs.edu.my abstract. this paper examines wagner hypothesis of the growth of public expenditure alongside the growth of economic activity for a panel of 28 european economies during the 1995-2018 period. the hypothesis is verified using pesaran (2007) panel unit root and westerlund (2007) cointegration tests that account for cross-sectional dependence in the series, and three panel causality tests (toda-yamamoto, dumitrescu-hurlin and juodis-karavias-sarafidis) that are suitable for mixed order of series’ integration, heterogeneous balanced panels and cases of limited evidence of cointegration. the empirical results suggested that expenditure and output variables were non-stationary in levels and stationary in the first differences; the cointegration among the variables was present; the causality was principally uni-directional (from output to public expenditure), in line with wagner’s hypothesis, or bi-directional; the causality from public expenditure to output along keynesian lines was limited. keywords. wagner’s hypothesis; panel cointegration; panel causality, europe jel codes. c23; e60; h50; n44 doi. https://doi.org/10.17979/ejge.2023.12.1.9146 1. introduction this paper empirically tests wagner’s hypothesis,1 which assumes a faster rate of growth of government expenditure (in absolute and in relative senses) than economic growth, as well as expansion of the government spending (and, more generally, activity) at the expense of private sector (wagner, 1912). the consideration of wagner’s hypothesis was diverse in terms of the definition and measurement of the public expenditure variable, the selection of independent variables in addition to gdp, the country coverage, econometric methods and tests, as well as the findings (the following section provides a cursory review of the empirical studies). in contrast to many previous studies, the focus of this paper is on the relationship between output and expenditure at both aggregate and disaggregated levels. the consideration of the aggregate level of analysis is justified by the fiscal sustainability, budget deficit and public debt problems that plague 1 in the absence of universally acceptable terms in the empirical studies, we use ‘public’ and ‘government’ expenditure terms interchangeably, and, given the conflicting findings of the empirical research, refer to wagner’s ‘hypothesis’, as opposed to wagner’s ‘law’. https://creativecommons.org/licenses/by-nc/4.0/ mailto:ivan.trofimov@kysbs.edu.my https://doi.org/10.17979/ejge.2023.12.1.9146 ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 6 the developed economies particularly strongly (koester & priesmeier, 2013): the validity of wagner’s hypothesis may provide a solid theoretical explanation of these problems (in addition to other explanatory factors). on the other hand, while wagner’s hypothesis may hold at the aggregate expenditure level, this may not necessarily be the case for the individual expenditure categories, e.g. the growth of health expenditure may exceed output growth, but the growth of expenditure on agriculture may lag behind it (as put by peacock & scott, 2000, the references to the differing pace and speed of expansion of the government may be found in wagner’s original works). additionally, the focus on the aggregate level may be unsound from an econometric perspective: rajaguru (2002) and kucukkale et al. (2012: 1-2) note the distortionary and bias-inducing effects of expenditure aggregation on cointegration and causality relationships, while granger (1988) mentions the incongruity of the unit root properties of the aggregate and component (disaggregated) series. given the specific aim of the paper (to make certain generalisations as to the validity of wagner’s hypothesis in a sufficiently large group of economies) and the lack of public expenditure data with a sufficient time series dimension (that is typically available only for a limited number of economies), we use a panel data set that covers 28 european economies over the 1995-2018 period. due to the economic and political integration that has been underway in europe since the 1950-60s, it is reasonable to assume that the panel data would be characterized by the cross-sectional dependence (stemming principally from intra-regional trade and investment flows, implementation of common economic policies, and the public expenditure decisions made at supra-national level) and thus use appropriate panel data econometric techniques. the rest of the paper is organized as follows. section 2 contains a review of the theoretical basis of wagner’s hypothesis and of the relevant empirical studies. section 3 provides the modeling framework, the data and econometric methodology, and the description of the panel data. section 4 presents the empirical results. section 5 concludes the paper, discusses the policy implications and outlines the avenues for future research. 2. theoretical framework and empirical literature certain ambiguity exists regarding the exact definition of wagner’s hypothesis, its applicability in a specific socio-economic setting and the formulation for the modelling purposes. peacock and scott (2000, pp. 2-3), based on a thorough examination of wagner’s original works, note that wagner preferred to call the concept an ‘empirical observed uniformity’ rather than law; tended to apply the concept both to the traditional services of the government (e.g. defence, law and order) and to the newer functions (welfare provision); included the expenditure by the central and local government as well as the activity of public enterprises; noted the problem of public sector expansion at the expense of the private sector growth; and conceptualised public sector growth not purely as a quantitative but also as a qualitative phenomenon (manifested in sophistication and greater extent of government regulation). ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 7 timm (1961) argued that wagner formulated the hypothesis for the analysis of the 19th-century governments but did not restrict the applicability of the concept, stating that government size would grow and involvement would extend as long as “cultural and economic progress” continues. thus, the hypothesis may be equally applicable to the economies at the early stage of capitalist development (as were european economies in the 19th century), and to other economies at various development stages.2 the public expenditure growth at a faster rate than economic growth has been attributed to a number of objective factors. firstly, the industrialisation, production on a large scale, and application of science and technology necessitate the accompanying growth of the organising force, embodied in government and manifested in greater spending (north & wallis, 1982, pp. 336; oxley, 1994, pp. 287). secondly, in contrast to the earlier epochs (e.g. 18th and 19th centuries, when the role of the government was restricted to the provision of public order, law and basic services), the 20th century witnessed much broader roles of the government. the popularity of socialist ideas and the accompanying class struggle, the rise of central or indicative planning (along with socialist economic doctrine), the two world wars and the cold war, the failures of the capitalist economies at certain junctures (e.g. great depression), led to greater public expenditure in such areas as welfare, education, health, and defense. thirdly, the sophistication of economic activity and transactions, and scientific and technological progress required greater regulation and oversight by the government in such functional areas as the design and implementation of legislation, improvement of economic and political institutions, consumer protection from monopolies, prevention of unfair competition, and also justified greater spending to support scientific and technological capability. the economic theory provides a number of explanations behind wagner’s hypothesis. the demand for education, health, social protection has high-income elasticity, hence the expenditure in these categories is more likely to conform to wagner’s predictions. public spending is an outcome of a public decision making process that is subject to influence by vested interests and special interest groups through their lobbying effort, bringing the expansion of the government size and greater public spending in a number of categories (mueller, 1987). the relative cost of government services may rise over time, resulting in spending growth in nominal terms (baumol, 1967). lastly, as stated by the bureaucracy theory of the government, the proliferation of the policies and the growth of spending (frequently at the expense of economic efficiency) result from the immanent features of bureaucracy and bureaucratic concerns that may be summarized as striving for power and prestige, policy and position preservation, and expansion of the bureaucratic apparatus (niskanen, 1971; oxley, 1994, pp. 286). an alternative hypothesis of reverse causality that runs from government expenditure to output is rooted in keynesian economics and is also supported by other lines of economic theory. in the shortrun, during the cyclical downswing, the increase in government expenditure may stimulate output, provided that there exists idle capacity and the economy is below the potential output (ray & pal, 2022). this view was challenged in a number of studies: while the causality originating from government expenditure may be present, the effects on output may be negative. barro (1991) notes the inefficiencies that are caused by the public sector expansion and that potentially slowdown economic 2 an alternative argument is the exclusive applicability of wagner’s hypothesis to the society that undergo rapid industrialisation, and not to the preor post-industrial social settings (jaen garcia, 2004, p. 13). ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 8 growth. tobin (2005) argues that the strength and sign of the expenditure-output effect hinges on the specific role of the state as a modernisation agent (as opposed to a product of vested interests and political pressures, or a source of bureaucratic failure and slack). the positive effects of expenditure on output are, therefore, not guaranteed. the empirical testing of wagner’s hypothesis was diverse in many respects. with regard to the definition of government expenditure, the empirical studies chose between the nominal and real expenditure values (for instance, gandhi, 1971, and lin, 1995, are examples of the study that used expenditure in current prices, while bohl, 1996, is the study that defined expenditure in constant prices). in a number of studies, the defence expenditure was excluded to isolate the expenditure-output relationships in a civil economy (abizadeh, gray, 1985). the empirical research focused on both the expenditure by the central government (mann, 1980) and the spending at sub-national level (e.g. narayan et al, 2012 in the study of wagner’s hypothesis for a group of indian states), but tended, in the absence of disaggregated expenditure data, to examine the aggregate expenditure (few studies, such as the analysis of the hypothesis in canada by biswal et al, 1999, stand as exception). regarding the selection of regressors, the following variables were used in addition to gdp (per capita): the sectoral shares of output to account for structural influences on ‘expenditure-output’ nexus (mann, 1980), the values of exports and imports to account for external economy influences of government spending decisions (abizadeh, gray, 1985); permanent income and deviation of current demand from the trend (courakis et al, 1993); country size and population density (alesina, wacziarg, 1998; dao, 1995). in terms of geographic coverage, a large number of studies focused on a single economy: wagner’s hypothesis in usa (islam, 2001), spain (jaen garcia, 2004), italy (magazzino, 2012), canada (ahsan et al, 1996), egypt (ghazy et al, 2021), among others. selected studies looked at regional or level of development groups (developing economies by diamond, 1977; 14 european economies by afonso, alves, 2017), economies with specific characteristics (e.g. the study of the hypothesis for petroleum exporters, where oil export revenues constitute a major source of government revenue and thus influence fiscal policy significantly, burney, 2002), or random sets of economies (e.g. the comparative study of spain and armenia by sedrakyan and varela-candamio, 2019; or of the three african economies, ansari et al, 1997). in terms of econometric methods and tests, the earlier studies tended to rely on descriptive statistics analysis (bairam, 1992), the use of linear regression (lin, 1995), or specification of the simultaneous equation models for the demand and supply of public expenditure (dollery, singh, 2000). such methods suffered from the ‘spurious regression’ and identification problems and ignored the non-stationarity of the time series. to address these problems, a more recent empirical analysis started to utilise cointegration and causality tests (oxley, 1994; legrenzi, 2000), vector autoregression models (benavides et al, 2013), panel data techniques (afonso, jalles, 2014), or non-linear models (karagianni, pempetzoglou, 2009). the empirical research delivered conflicting results. a number of studies indicated the absence of causality between expenditure and gdp in either direction (huang, 2006; sinha, 2007). the unidirectional causality that supports wagner’s hypothesis was identified in the studies by courakis et ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 9 al (1993), tang (2001), chang et al (2004), jaen garcia (2004), sideris (2007), lamartina and zaghini (2011) and, in selected specifications, benavides et al (2013). the causality along keynesian lines was demonstrated by iyare and lorde (2004) and babatunde (2011). the bi-directional causality was indicated by narayan et al (2008), ziramba (2009), and yay and tastan (2009). given a variety of empirical outcomes in the studies that focused at aggregate expenditure and individual economies (or narrow groups of economies), it is advantageous to consider the specific expenditure categories and a broader panel that encompassed the maximum possible number of countries. 3. methodology model following loizides and vamvoukas (2005), a bivariate representation of wagner’s hypothesis is used, with no inclusion of additional variables. the five specifications of the hypothesis suggested by peacock and wiseman (1979), mann (1980), musgrave (1969), gupta (1967), and goffman (1968) were considered: specification 1 ln ln tg yα β ε= + + (1) specification 2 ln( ) ln tg y yα χ ε= + + (2) specification 3 ln( ) ln( ) tg y y pα δ ε= + + (3) specification 4 ln( ) ln( ) tg p y pα φ ε= + + (4) specification 5 ln ln( ) tg y pα ϕ ε= + + , (5) where g , y and p respectively represent government, expenditure, real gdp and population. it is assumed that causality runs from the right to the left-hand side of the above equations under wagner’s hypothesis, and in the opposite direction under keynesian hypothesis. data the paper uses the gdp and public expenditure data released by eurostat. both gdp and expenditure were represented in millions of euro in chain-linked volume measures of gdp (with the base of the chain-linked volume index set at 2010). the constructed panel dataset used in the empirical analysis covered 1995-2018 period and included 28 european economies: austria, belgium, bulgaria, cyprus, czech republic, denmark, estonia, finland, france, germany, greece, hungary, ireland, italy, latvia, lithuania, luxembourg, netherlands, norway, poland, portugal, romania, slovakia, slovenia, spain, ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 10 sweden, switzerland, and the uk. the public expenditure categories followed classification of the function of the government (cofog), version 1999 (oecd, 2011, pp. 194-5). in addition to total expenditure, the following expenditure categories are delineated: defence; economic affairs; education; environmental protection; general public services; health; housing and community amenities; public order and safety; recreation, culture and religion; social protection. econometric methods initially, we examine the integration order and the unit root properties of the expenditure and output. there is a possibility of cross-sectional dependence in the series (the correlation of the panel members in the cross-section due to various unobserved common factors, such as policy and political integration or policy learning). as a result, we conduct pesaran (2004, 2015) cross-sectional dependence test that is flexible with regard to panel data structure (specifically when time series dimension is smaller than cross-sectional dimension, t n< ), and pesaran (2007) cross-sectionally augmented ips test (cips) that is robust to the cross-sectional correlation (in lieu of the first-generation panel unit root tests that ignore the phenomenon). for the variables that have (1)i integration order and contain unit roots, we consider the possibility of cointegration and conduct westerlund panel cointegration tests (westerlund, 2007). westerlund test tackles cross-sectional dependence, removes common factor restriction, and specifies the errorcorrection model as follows (persyn& westerlund, 2008, pp. 233): ( )1 1 1 1 i i i p p it i t i it it ij it ij it j it j j q y d y z y zδ α β α γ ε− − − − = =− ′ ′∆ = + − + ∆ + ∆ +∑ ∑ , (6) where td is a deterministic component, and iα is a coefficient that measures the speed at which the system returns to the equilibrium ( 1 1it ity zβ− −′− ). the null hypothesis is of the absence of error-correction and cointegration ( 0iα = for all i ); an alternative hypothesis is of the presence of error correction and cointegration for some of i , or all i ’s ( 0iα < in the test’s group-mean statistics or 0iα α= < in the panel statistics). as a next step, in order to ascertain causality between public expenditure and output, we employed toda-yamamoto (ty) causality test (toda & yamamoto, 1995). the test is applicable for the situations when the variables have different order of integration, for instance, the combination of (0)i and (1)i order (the data characteristic that renders the conventional granger causality test unreliable), or when the cointegration tests give conflicting indications. the ty test is invariant to the integration orders (allows any combination of the orders) and is valid irrespective of the presence or absence of cointegration. implementation-wise, the ty test establishes the maximum integration order of the series, maxd , uses it in the estimation of the augmented var model in levels (where the optimal lag ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 11 order is determined using the conventional lag selection criteria, and the selected lag for each variable is increased by maxd ) and constructs modified wald ( mwald ) statistics to confirm causality (similarly to the conventional granger causality test). the inclusion of lags is necessary, due to the delayed (not instantaneous) effects of gdp on government expenditure. for the case of public expenditure-output relationship, the augmented var model is represented as: max max 0 1 1 1 1 1 1 1 1 1 d dk k t i t i i t j i t j i t j t i j k j j k y y y x xα φ ϕ γ λ ε− − − − = = + = = + = + + + + +∑ ∑ ∑ ∑ (7) max max 1 2 2 2 2 2 1 1 1 1 d dk k t i t i i t j i t j i t j t i j k j j k x x x y yα φ ϕ γ λ ε− − − − = = + = = + = + + + + +∑ ∑ ∑ ∑ (8) the causality in the sense of granger from x to y in equation (7) exists when 1 0iγ ≠ , i∀ , and the like causality from y to x in equation (8) is present when 2 0iγ ≠ , i∀ . for the purpose of robustness check, we conducted dumitrescu-hurlin (2012) test of causality in a panel of stationary data (i.e. to implement the test the (1)i variable were converted to the first difference form). the estimated coefficients are time-invariant but are allowed to differ across the panel members (the test is thus suitable for heterogeneous balanced panels, lopez & weber, 2017, pp. 3-4). the lag order ( k ) that is uniform across panel members is determined using the usual information criteria, with the first maxk eliminated during the lag selection process. for the case of two stationary variables, the dumitrescu-hurlin model is set as (lopez& weber, 2017, pp. 3-4): 1 1 k k it i ik it k ik it k it k k x x yα δ γ ε− − = = = + + +∑ ∑ , (9) for 1, ,i n=  and 1, ,t t=  . the null hypothesis is of no causality for any cross-sections in the panel: 0 1: 0i ikh γ γ= = = for 1, ,i n∀ =  (10) under an alternative hypothesis the causality in some of the cross-sections is allowed: 1: 0a i ikh γ γ= = = for 11, ,i n∀ =  and (11) 1 0iγ ≠ or  0ikγ ≠ for 1 1, ,i n n∀ = +  (12) for each i , the wald test is conducted for the hypothesis 0ikγ = and the wald test statistics for each panel member and the aggregate panel are obtained as: ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 12 itw and 1 1 n it i w n w = = ∑ (13) as a last step the standardized z and z statistics are calculated: ( ) , (0,1) 2 d t n nz w k k →∞= − →ν and (14) 3 5 3 3 (0,1) 2 2 3 3 1 d n n t k t kz w k n k t k t k →∞ − − − − = ⋅ ⋅ ⋅ − → − − − −   (15) as a last step we perform the causality test proposed by juodis et al. (2021). in contrast to dumitrescu-hurlin test that has the highest power with large t dimension and 2/n t approaching zero, it is suitable for the panels with a moderate time series dimension. the test performs well in both homogeneous and heterogeneous panels, assumes that granger causality parameters are equal to zero under the null, and thus allows using pooled estimator for granger causality parameters (xiao et al., 2021, p. 4). the pooled estimators have faster rate of convergence while at the same time suffering from the ‘nickel bias’, the problem corrected in the test using the half-panel jackknife method (xiao et al., 2021, p. 3). the test equation is given as (xiao et al., 2021, p. 3): 0 1 1 p p it i pi it p pi it p it p p x x yφ φ γ ε− − = = = + + +∑ ∑ , (16) where 1, ,i n=  , 1, ,t t=  , 1, ,p p=  , 0iφ are individual specific effects, piφ are heterogeneous autoregressive coefficients, and piγ are heterogeneous feedback coefficients. ity is assumed to be a scalar. the null hypothesis assumes the absence of granger causality 0 : 0pih γ = for all i and p , while the alternative hypothesis assumes that 0piγ ≠ for some i and p . 4. empirical results figure 1 (appendix) shows the dynamics of cross-sectional means of gdp and expenditure series. both gdp and gdp per capita exhibited an upward trend throughout the period (with the exception of the global financial crisis / gfc years of 2008-09). with regard to absolute values of expenditures, the upward trend during the whole period (1995-2018) was observed for the total, education, health, public order and safety, social protection and recreation, culture and religion categories, while other categories were characterised by stabilisation of expenditure (economic affairs, general public services, and environmental protection), decline (defence and housing and community amenities) in the post-gfc years. the expenditure shares of gdp did not follow any specific trend (possibly with the exception of defence, and general public services). for the expenditure per capita, the deterministic patterns were observed only for education, health, public order and safety, social protection and recreation, culture and religion categories. the visual inspection is supplemented for formal unit root tests; the former method is deemed misleading (cuddington et al, 2002: 21). ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 13 as demonstrated in table 1, according to the pesaran 2004 and 2015 tests, the null hypotheses of cross-sectional independence and weak dependence were respectively rejected: each of the variables in question was found to be characterised by strong cross-sectional correlation. the pattern is not unusual, given the high degree of economic and political integration in europe: the previous research indicated convergence in public expenditure across european economies (apergis et al., 2013) and also highlighted gdp and business cycle comovement in the region (azcona, 2022). the incorporation of the cross-sectional dependence into unit root and cointegration tests is therefore warranted. the results of the pesaran cadf unit root test are presented in table 1. the test was implemented with a trend component in order to account for the growth of public expenditure and gdp in the longrun. the test was also run on the levels and the first differences of the variables in order to determine the integration order of the variables, which is instrumental for toda-yamamoto causality test (that requires the knowledge of the maximum order of integration). the test indicates that gdp and gdp per capita variables were trend stationary in both levels and the first differences (in the latter case at 5% and 10% significance levels), and are thus (0)i order variables. all other variables in the level form contained unit root and the test’ null hypothesis was not rejected (the exceptions were the absolute level of total, education, health, public order and safety, environmental protection and housing and community amenities expenditure; the per capita expenditure on education, public order and safety; and the public expenditure as a proportion of gdp in general public services and environmental protection). in the first difference form, all expenditure variables were trend stationary at 5% level (social protection expenditure as a proportion of gdp at 10% level). overall, we conclude that the variables are the mix of (0)i and (1)i order of integration, and the maximum order is (1)i . the westerlund cointegration test was implemented with a maximum of two lags, one lead and a short kernel window of two (in line with recommendations for panels with small time-series dimension, persyn & westerlund, 2008) for each of the five specifications of wagner’s hypothesis. alternative laglead and kernel width combinations were also tried, yielding rather similar results (results available ob request). a total of 220 test statistics were obtained: four statistics, ( tg , ag , tp and ap ) for five specifications (peacock-wiseman, mann, musgrave, gupta and goffman), and 11 expenditure categories. the test failed to reject the null hypothesis of no cointegration in 41 instances (i.e. by 41 statistics), thus suggesting that overall the public expenditure and gdp were cointegrated. the greatest number of non-rejections was indicated in the social protection expenditure category, suggesting that a long-term equilibrium relationship between public expenditure and gdp was less likely in this category. out of four westerlund test statistics, the greatest number of non-rejections of the null was indicated for the ag statistics. out of five specifications, mann and musgrave models (specifications 2 and 3) gave somewhat weaker support for cointegration. while the evidence of cointegration relationships is generally strong, we relied on toda-yamamoto causality test that is invariant to the presence of the long-run equilibrium relationship and the (non-) stationarity of the variables. the pesaran cadf unit root test unequivocally indicated the highest order of integration of one. the optimal lag length for the test was selected based on a combination of information criteria (akaike, schwarz-bayesian and hannan-quinn). given that the too-short lag length ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 14 causes bias, while longer than optimal lag length results in test’ inefficiency (caporale & pittis, 1999; ayad & belmokaddem, 2017) we considered alternative lag structures and, where applicable, reported the results with the highest and the second highest lag length. the toda-yamamoto test was run for 100 cases (given five possible specifications of wagner’s hypothesis and alternative lag structures for the test). the causality running unidirectionally from public expenditure to gdp was not indicated in a single case, while bi-directional causality was noted in 37 cases (table 3). no causality in either direction was demonstrated in 14 cases, while causality from gdp to public expenditure in line with wagner’s hypothesis was shown in 49 cases. overall, toda-yamamoto test does not support the keynesian hypothesis of the uni-directional influence of public expenditure on gdp. as part of the robustness check, we conducted dumitrescu-hurlin causality test on each specification, with variables expressed as first-differences. the optimal lag length was determined by bayesian-schwarz criterion from a maximum of eight lags, and in all cases was found to be one. table 4 contains an asymptotically well-behaved average wald statistics (w ), the standardised z and approximated standardised z statistics that follow normal distribution (the latter used for the panels with fixed time-series dimension), together with the respective p-values. we note that in a number of instances z and z statistics gave conflicting results; however, based on z , the test rejected the null hypothesis that public expenditure does not granger-cause gdp only in five cases (three specifications for the general public services category, and two specifications for the defence category). in contrast, the null hypothesis that gdp does not granger-cause public expenditure was rejected in a total of 26 cases, while bi-directional causality was indicated in 7 cases. lastly, given the relatively short n and t dimensions of the panel, we implemented the biascorrected granger causality test proposed by xiao et al. (2021). the test was run on the first differences of the series with optimal lag selected based on bayesian-schwarz criterion (similarly to the previous test, in all cases, the optimal lag length stood at one). the results are presented in table 5: the halfpanel jackknife (hpj) wald test statistics and the estimator, both with the respective p-values. the test rejected the null of no unidirectional causality from gdp to public expenditure in a total of 29 out of 55 cases, but rejected the null of no unidirectional causality from public expenditure to gdp in only 2 cases (mann and musgrave specifications for the public order and safety category). the bi-directional causality was indicated in 11 cases. robustness checks in addition to causality testing, the study has experimented with cross-sectional analysis of the data. for each year of the period (1995-2018), the cross-sectional average was calculated for each output and expenditure variable. the wagner’s hypothesis was then verified through the significance of the coefficient in the regression that includes expenditure as independent variable and gdp (output) as regressor. table 6 in the appendix demonstrates the coefficient significance in most expenditure categories in specifications 1 and 4, in the majority of categories in specification 5, and in many categories in other specifications. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 15 the toda-yamamoto causality results were consistent over time. two checks that capture the effect of the business cycle on the causality and that distinguish causality in the expansion and downturn were implemented. the first check introduced the cyclical dummy variable in the toda-yamamoto relationship (the dummy takes the value one when gdp has negative growth and zero value when the growth rate is positive). as indicated in table 7 in the appendix, the inclusion of the dummy variable did not alter the results substantially, with bi-directional causality replacing wagner hypothesis causality in few categories (specifications) and vice versa. importantly, the keynesian-type causality from expenditure to output was not identified in any of the cases, similar to the baseline model. the second check included the specification of output in expansionary and recessionary stages (the calculation of the growth rate of gdp, the separation of instances of positive and negative growth, and calculation of the cumulative sums of positive and negative growth). the check was performed for specification 1, for the whole period and the two sub-periods prior to and after gfc (the maximum lags were tried only for the whole period, given a limited number of observations in the sub-periods). for the whole period, bidirectional causality was identified in 10 out of 40 cases, while wagner type causality was observed in 25 cases (table 8 in the appendix). for the post-gfc period, the number of wagner type causality cases stood at 11 out of 22, and for the pre-gfc period at 9 out of 22. bi-directional or no causality were observed in other instances. keynesian causality was identified only in two instances in pre-gfc period, however at higher lags this causality does not hold. overall, the robustness checks give results that are very similar to the baseline model. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 16 table 1. pesaran cd test and cadf test results. expenditure area variable pesaran (2004) cd test pesaran (2015) cd test cadf (trend, levels) cadf (trend, first differences) stat p-value stat p-value stat p-value stat p-value gdp y 95.240 0.000 85.585 0.000 -3.124 0.000 -2.629 0.035 gdp per capita y/p 95.232 0.000 81.140 0.000 -3.080 0.000 -2.583 0.060 total g/y 95.214 0.000 24.075 0.000 -2.693 0.015 -3.424 0.000 total g/p 95.229 0.000 52.590 0.000 -2.789 0.003 -3.642 0.000 total g 95.236 0.000 83.761 0.000 -2.847 0.001 -3.880 0.000 economic affairs g/y 93.318 0.000 8.802 0.000 -2.523 0.113 -4.011 0.000 economic affairs g/p 95.131 0.000 14.079 0.000 -2.512 0.126 -3.906 0.000 economic affairs g 95.175 0.000 34.250 0.000 -2.698 0.014 -4.003 0.000 defense g/y 33.609 0.000 27.440 0.000 -2.286 0.555 -3.846 0.000 defense g/p 95.057 0.000 8.194 0.000 -2.379 0.348 -4.086 0.000 defense g 95.124 0.000 5.827 0.000 -2.598 0.051 -4.141 0.000 education g/y 95.045 0.000 16.610 0.000 -2.262 0.608 -3.588 0.000 education g/p 95.208 0.000 48.748 0.000 -2.844 0.001 -3.461 0.000 education g 95.228 0.000 71.561 0.000 -2.997 0.000 -3.652 0.000 health g/y 94.694 0.000 46.563 0.000 -2.572 0.068 -3.110 0.000 health g/p 95.188 0.000 76.444 0.000 -2.634 0.033 -3.166 0.000 health g 95.214 0.000 82.476 0.000 -2.940 0.000 -3.431 0.000 public order and safety g/y 64.675 0.000 4.521 0.000 -2.331 0.452 -3.263 0.000 public order and safety g/p 95.177 0.000 51.373 0.000 -2.825 0.002 -3.436 0.000 public order and safety g 95.203 0.000 70.858 0.000 -3.029 0.000 -3.762 0.000 social protection g/y 95.121 0.000 27.647 0.000 -2.064 0.919 -2.584 0.060 social protection g/p 95.220 0.000 70.212 0.000 -2.183 0.765 -2.911 0.000 social protection g 95.230 0.000 88.382 0.000 -1.948 0.980 -2.752 0.006 general public services g/y 94.471 0.000 29.048 0.000 -2.989 0.000 -3.819 0.000 general public services g/p 95.166 0.000 1.752 0.080 -2.545 0.090 -3.478 0.000 general public services g 95.201 0.000 17.338 0.000 -2.606 0.046 -3.539 0.000 environmental protection g/y 42.025 0.000 5.969 0.000 -2.815 0.002 -3.851 0.000 environmental protection g/p 94.405 0.000 21.722 0.000 -2.762 0.005 -3.563 0.000 environmental protection g 94.859 0.000 40.309 0.000 -2.840 0.001 -3.910 0.000 recreation, culture, religion g/y 9.367 0.000 8.040 0.000 -2.278 0.572 -3.388 0.000 recreation, culture, religion g/p 95.066 0.000 36.176 0.000 -2.556 0.080 -3.372 0.000 recreation, culture, religion g 95.164 0.000 57.551 0.000 -2.707 0.012 -3.421 0.000 housing and community amenities g/y 29.495 0.000 18.457 0.000 -2.757 0.005 -3.740 0.000 housing and community amenities g/p 94.292 0.000 5.234 0.000 -2.735 0.008 -3.860 0.000 housing and community amenities g 94.905 0.000 6.287 0.000 -2.868 0.001 -3.736 0.000 note. all variables are in the logarithm form. the pesaran cadf test critical values at 10%, 5% and 1% levels are -2.580, -2.660, and -2.810 respectively. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 17 table 2. westerlund cointegration test results. expenditure area variable specification 1 specification 2 specification 3 specification 4 specification 5 peacock-wiseman mann musgrave gupta goffman d(g) d(y) d(g/y) d(y) d(g/y) d(y/p) d(g/p) d(y/p) d(g) d(y/p) total gt -3.147 0.000 -3.147 0.000 -3.108 0.000 -2.593 0.000 -2.874 0.000 ga -12.747 0.000 -6.083 0.848 -6.272 0.801 -8.719 0.063 -11.625 0.000 pt -15.051 0.000 -14.431 0.000 -14.502 0.000 -14.359 0.000 -13.517 0.000 pa -10.429 0.000 -5.479 0.068 -5.531 0.061 -9.286 0.000 -8.156 0.000 education gt -2.385 0.000 -2.385 0.000 -2.532 0.000 -2.251 0.003 -2.235 0.004 ga -10.305 0.001 -5.272 0.966 -5.993 0.868 -10.102 0.002 -8.833 0.050 pt -10.370 0.003 -10.191 0.006 -11.235 0.000 -10.849 0.001 -11.435 0.000 pa -8.331 0.000 -5.207 0.122 -5.845 0.027 -8.976 0.000 -7.918 0.000 economic affairs gt -2.636 0.000 -2.636 0.000 -2.640 0.000 -2.676 0.000 -2.646 0.000 ga -11.200 0.000 -9.484 0.011 -10.038 0.002 -11.439 0.000 -10.633 0.000 pt -15.750 0.000 -15.670 0.000 -15.946 0.000 -15.404 0.000 -15.369 0.000 pa -11.689 0.000 -11.182 0.000 -11.605 0.000 -12.243 0.000 -10.838 0.000 defense gt -2.324 0.000 -2.324 0.001 -2.228 0.004 -2.153 0.014 -2.185 0.008 ga -6.541 0.721 -7.666 0.305 -7.063 0.531 -6.815 0.625 -6.973 0.566 pt -11.095 0.000 -11.142 0.000 -11.014 0.000 -10.213 0.005 -10.457 0.003 pa -7.071 0.000 -6.982 0.001 -6.848 0.001 -6.220 0.009 -6.608 0.002 health gt -3.021 0.000 -3.021 0.000 -3.041 0.000 -2.597 0.000 -2.744 0.000 ga -15.795 0.000 -7.471 0.375 -6.924 0.584 -11.931 0.000 -12.402 0.000 pt -14.370 0.000 -12.472 0.000 -12.773 0.000 -13.233 0.000 -13.933 0.000 pa -13.334 0.000 -9.349 0.000 -8.813 0.000 -10.820 0.000 -11.121 0.000 social protection gt -2.630 0.000 -2.630 0.000 -2.522 0.000 -2.408 0.000 -2.489 0.000 ga -6.039 0.858 -2.905 1.000 -2.841 1.000 -6.099 0.845 -5.663 0.925 pt -10.889 0.001 -9.029 0.084 -8.594 0.174 -11.481 0.000 -10.817 0.001 pa -6.109 0.013 -3.283 0.871 -2.876 0.947 -5.033 0.169 -4.681 0.296 ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 18 table 2. (cont). public order and safety gt -2.593 0.000 -2.593 0.000 -2.566 0.000 -2.089 0.034 -2.618 0.000 ga -9.733 0.006 -7.253 0.457 -7.390 0.405 -7.565 0.341 -9.064 0.031 pt -12.577 0.000 -13.962 0.000 -13.940 0.000 -11.949 0.000 -12.971 0.000 pa -7.873 0.000 -7.909 0.000 -7.641 0.000 -6.629 0.002 -7.326 0.000 general public services gt -2.369 0.000 -2.369 0.000 -2.192 0.007 -2.113 0.024 -2.178 0.009 ga -7.489 0.368 -6.942 0.577 -5.688 0.921 -7.175 0.487 -6.468 0.744 pt -12.926 0.000 -12.820 0.000 -12.971 0.000 -12.342 0.000 -12.845 0.000 pa -8.278 0.000 -7.714 0.000 -7.800 0.000 -8.006 0.000 -8.067 0.000 environmental protection gt -2.339 0.001 -2.339 0.001 -2.464 0.000 -2.357 0.000 -2.608 0.000 ga -8.991 0.036 -8.058 0.187 -8.336 0.123 -8.523 0.090 -9.969 0.003 pt -11.330 0.000 -11.317 0.000 -10.769 0.001 -12.300 0.000 -11.051 0.000 pa -7.516 0.000 -6.883 0.001 -6.393 0.005 -8.480 0.000 -7.053 0.000 recreation, culture, religion gt -2.001 0.094 -2.001 0.094 -2.103 0.028 -1.862 0.310 -2.278 0.002 ga -8.151 0.164 -5.426 0.952 -5.975 0.872 -7.149 0.497 -8.488 0.096 pt -10.135 0.006 -10.099 0.007 -10.134 0.006 -9.094 0.075 -10.454 0.003 pa -7.613 0.000 -6.229 0.009 -6.541 0.003 -5.995 0.018 -7.857 0.000 housing and community amenities gt -2.862 0.000 -2.862 0.000 -2.615 0.000 -2.588 0.000 -2.848 0.000 ga -8.446 0.103 -8.193 0.154 -8.154 0.163 -7.591 0.331 -8.551 0.085 pt -12.283 0.000 -12.329 0.000 -11.711 0.000 -13.035 0.000 -12.140 0.000 pa -8.131 0.000 -7.855 0.000 -7.664 0.000 -9.089 0.000 -8.204 0.000 note. the statistics in bold indicates non-rejection of the null hypothesis. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 19 table 3. toda-yamamoto panel causality test results expenditure area specification 1 specification 2 specification 3 specification 4 specification 5 peacock-wiseman mann musgrave gupta goffman d(g) d(y) d(y) d(g/y) d(y) d(y) d(g/y) d(g/y) d(y/p) d(y/p) d(g/y) d(g/p) d(y/p) d(y/p) d(g/p) d(g) d(y/p) d(y/p) d(g) 1 chi square 123.284 27.356 45.688 27.356 40.079 24.389 33.964 0.281 86.866 28.584 prob 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.869 0.000 0.000 lag 4 4 4 4 4 4 2 2 4 4 bi-directional bi-directional bi-directional bi-directional bi-directional chi square 133.328 23.566 54.452 23.566 48.766 23.720 68.708 13.088 98.053 24.664 prob 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.011 0.000 0.000 lag 5 5 5 5 5 5 4 4 5 5 bi-directional bi-directional bi-directional bi-directional bi-directional 2 chi square 40.058 10.212 31.179 10.212 31.835 10.133 31.660 7.525 29.158 16.343 prob 0.000 0.037 0.000 0.037 0.000 0.038 0.000 0.111 0.000 0.003 lag 4 4 4 4 4 4 4 4 4 4 bi-directional bi-directional bi-directional y→g bi-directional 3 chi square 15.608 1.191 5.734 1.191 5.491 1.451 12.011 3.187 13.582 1.351 prob 0.001 0.755 0.125 0.755 0.139 0.694 0.003 0.203 0.004 0.717 lag 3 3 3 3 3 3 2 2 3 3 y→g no causality no causality y→g y→g chi square 39.619 9.280 35.848 9.280 34.014 11.722 10.529 2.748 38.216 10.650 prob 0.000 0.233 0.000 0.233 0.000 0.110 0.015 0.432 0.000 0.222 lag 7 7 7 7 7 7 3 3 8 8 y→g y→g y→g y→g y→g 4 chi square 106.126 0.847 18.663 0.847 18.242 1.757 83.357 5.503 76.628 1.211 prob 0.000 0.932 0.001 0.932 0.001 0.780 0.000 0.139 0.000 0.876 lag 4 4 4 4 4 4 3 3 4 4 y→g y→g y→g y→g y→g chi square 65.742 1.974 prob 0.000 0.741 lag 4 4 y→g ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 20 table 3. (cont). 5 chi square 116.408 11.018 38.008 11.018 32.330 10.884 61.118 10.971 97.801 11.314 prob 0.000 0.026 0.000 0.026 0.000 0.028 0.000 0.027 0.000 0.023 lag 4 4 4 4 4 4 4 4 4 4 bi-directional bi-directional bi-directional bi-directional bi-directional chi square 99.358 21.883 50.920 21.883 38.204 12.770 63.230 15.889 71.674 12.500 prob 0.000 0.005 0.000 0.005 0.000 0.047 0.000 0.044 0.000 0.052 lag 8 8 8 8 6 6 8 8 6 6 bi-directional bi-directional bi-directional bi-directional bi-directional 6 chi square 24.435 0.384 0.575 0.384 7.317 0.735 18.662 0.013 9.018 0.258 prob 0.000 0.825 0.750 0.825 0.026 0.692 0.000 0.994 0.011 0.879 lag 2 2 2 2 2 2 2 2 2 2 y→g no causality y→g y→g y→g chi square 50.145 4.343 10.492 4.343 5.750 7.161 34.687 4.209 33.511 4.038 prob 0.000 0.362 0.033 0.362 0.219 0.128 0.000 0.520 0.000 0.401 lag 4 4 4 4 4 4 5 5 4 4 y→g y→g no causality y→g y→g 7 chi square 107.205 10.899 13.445 10.899 26.908 10.355 75.663 5.104 123.616 16.208 prob 0.000 0.012 0.004 0.012 0.000 0.066 0.000 0.403 0.000 0.006 lag 3 3 3 3 5 5 5 5 5 5 bi-directional bi-directional bi-directional y→g bi-directional chi square 155.005 12.198 32.839 12.198 9.300 10.229 50.810 1.501 88.522 14.007 prob 0.000 0.032 0.000 0.032 0.026 0.017 0.000 0.682 0.000 0.003 lag 5 5 5 5 3 3 3 3 3 3 bi-directional bi-directional bi-directional y→g bi-directional 8 chi square 1.720 1.124 23.417 1.124 25.006 0.831 2.024 0.612 1.484 0.838 prob 0.423 0.570 0.000 0.570 0.000 0.660 0.363 0.737 0.476 0.658 lag 2 2 2 2 2 2 2 2 2 2 no causality y→g y→g no causality no causality chi square 2.611 5.814 13.948 5.814 15.238 4.550 1.289 2.225 2.615 4.468 prob 0.456 0.121 0.003 0.121 0.002 0.208 0.732 0.527 0.455 0.215 lag 3 3 3 3 3 3 3 3 3 3 no causality y→g y→g no causality no causality ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 21 table 3. (cont). 9 chi square 17.189 0.301 2.176 0.301 1.408 0.370 18.387 0.912 13.608 0.074 prob 0.000 0.861 0.337 0.861 0.495 0.831 0.000 0.634 0.001 0.964 lag 2 2 2 2 2 2 2 2 2 2 y→g no causality no causality y→g y→g chi square 18.435 1.259 3.333 1.259 2.346 2.023 8.614 3.022 15.974 24.433 prob 0.000 0.739 0.343 0.739 0.504 0.568 0.035 0.388 0.007 0.000 lag 3 3 3 3 3 3 3 3 5 5 y→g no causality no causality y→g bi-directional 10 chi square 39.510 1.100 15.197 1.100 15.293 1.029 36.183 0.832 33.692 1.501 prob 0.000 0.577 0.001 0.577 0.001 0.598 0.000 0.660 0.000 0.472 lag 2 2 2 2 2 2 2 2 2 2 y→g y→g y→g y→g y→g chi square 39.449 6.960 11.257 6.960 11.097 7.227 45.733 1.348 32.579 9.442 prob 0.000 0.138 0.024 0.138 0.026 0.124 0.000 0.718 0.000 0.051 lag 4 4 4 4 4 4 3 3 4 4 y→g y→g y→g y→g bi-directional 11 chi square 29.725 3.015 15.849 3.015 14.745 6.003 8.567 17.115 22.147 8.894 prob 0.000 0.389 0.001 0.389 0.002 0.111 0.036 0.001 0.000 0.031 lag 3 3 3 3 3 3 3 3 3 3 y→g y→g y→g bi-directional bi-directional chi square 25.658 2.969 11.814 2.969 10.254 7.086 17.014 8.491 prob 0.000 0.563 0.019 0.563 0.036 0.131 0.002 0.075 lag 4 4 4 4 4 4 4 4 y→g y→g y→g bi-directional note. the expenditure area numbers represent respectively: total expenditure (1), economic affairs (2), defense (3), education (4), health (5), public order and safety (6), social protection (7), general public services (8), environmental protection (9), recreation, culture, religion (10), and housing and community amenities (11). ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 22 table 4. dumitrescu-hurlin panel causality test results. expenditure area specification 1 specification 2 specification 3 specification 4 specification 5 peacock-wiseman mann musgrave gupta goffman d(g) d(y) d(g/y) d(y) d(y) d(g/y) d(g/y) d(y/p) d(y/p) d(g/y) d(g/p) d(y/p) d(y/p) d(g/p) d(g) d(y/p) d(y/p) d(g) 1 w-bar 3.241 0.968 0.895 0.968 0.990 0.866 3.248 0.744 3.319 0.945 z-bar 8.386 -0.120 -0.392 -0.120 -0.036 -0.501 8.412 -0.957 8.677 -0.207 p-value 0.000 0.905 0.695 0.905 0.972 0.617 0.000 0.339 0.000 0.836 z-bar tilde 6.490 -0.457 -0.680 -0.457 -0.389 -0.768 6.511 -1.141 6.727 -0.528 p-value 0.000 0.648 0.497 0.648 0.698 0.442 0.000 0.254 0.000 0.597 y→g no causality no causality y→g y→g 2 w-bar 2.233 1.586 1.575 1.586 1.601 1.476 2.327 1.312 2.232 1.531 z-bar 4.614 2.191 2.153 2.191 2.247 1.779 4.963 1.167 4.608 1.986 p-value 0.000 0.029 0.031 0.029 0.025 0.075 0.000 0.243 0.000 0.047 z-bar tilde 3.409 1.430 1.399 1.430 1.476 1.094 3.694 0.593 3.404 1.262 p-value 0.001 0.153 0.162 0.153 0.140 0.274 0.000 0.553 0.001 0.207 y→g no causality no causality y→g y→g 3 w-bar 1.696 1.916 1.393 1.916 1.430 1.881 2.379 1.744 1.657 1.883 z-bar 2.605 3.426 1.472 3.426 1.609 3.295 5.161 2.783 2.456 3.303 p-value 0.009 0.001 0.141 0.001 0.108 0.001 0.000 0.005 0.014 0.001 z-bar tilde 1.768 2.439 0.842 2.439 0.955 2.332 3.856 1.914 1.647 2.338 p-value 0.077 0.015 0.400 0.015 0.340 0.020 0.000 0.056 0.100 0.019 bi-directional g→y g→y bi-directional bi-directional 4 w-bar 5.129 1.511 1.458 1.511 1.296 1.303 3.363 1.307 4.773 1.363 z-bar 15.450 1.911 1.714 1.911 1.107 1.132 8.842 1.150 14.119 1.358 p-value 0.000 0.056 0.087 0.056 0.268 0.258 0.000 0.250 0.000 0.174 z-bar tilde 12.260 1.201 1.040 1.202 0.545 0.565 6.863 0.580 11.172 0.749 p-value 0.000 0.230 0.298 0.230 0.586 0.572 0.000 0.562 0.000 0.454 y→g no causality no causality y→g y→g 5 w-bar 3.285 0.790 1.332 0.790 1.191 0.772 3.293 0.925 3.192 0.737 z-bar 8.551 -0.785 1.242 -0.785 0.715 -0.853 8.580 -0.279 8.203 -0.983 p-value 0.000 0.432 0.214 0.432 0.474 0.394 0.000 0.780 0.000 0.326 z-bar tilde 6.625 -1.001 0.655 -1.001 0.225 -1.056 6.648 -0.588 6.341 -1.162 p-value 0.000 0.317 0.513 0.317 0.822 0.291 0.000 0.557 0.000 0.245 y→g no causality no causality y→g y→g ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 23 table 4. (cont). 6 w-bar 2.680 1.362 1.119 1.362 1.105 1.556 2.414 1.373 2.523 1.577 z-bar 6.286 1.354 0.445 1.354 0.393 2.082 5.290 1.395 5.700 2.158 p-value 0.000 0.176 0.656 0.176 0.695 0.037 0.000 0.163 0.000 0.031 z-bar tilde 4.775 0.746 0.004 0.746 -0.039 1.341 3.961 0.780 4.296 1.403 p-value 0.000 0.455 0.997 0.456 0.969 0.180 0.000 0.436 0.000 0.161 y→g no causality no causality y→g y→g 7 w-bar 3.252 1.590 2.376 1.590 2.607 1.761 1.954 1.969 3.111 1.621 z-bar 8.428 2.206 5.150 2.206 6.013 2.848 3.570 3.627 7.900 2.324 p-value 0.000 0.027 0.000 0.027 0.000 0.004 0.000 0.000 0.000 0.020 z-bar tilde 6.524 1.443 3.847 1.443 4.552 1.966 2.556 2.603 6.093 1.538 p-value 0.000 0.149 0.000 0.149 0.000 0.049 0.011 0.009 0.000 0.124 y→g y→g bi-directional bi-directional y→g 8 w-bar 1.950 2.596 0.913 2.596 0.991 2.371 1.404 2.147 2.027 2.555 z-bar 3.555 5.970 -0.326 5.970 -0.036 5.129 1.510 4.290 3.844 5.817 p-value 0.000 0.000 0.744 0.000 0.972 0.000 0.131 0.000 0.000 0.000 z-bar tilde 2.544 4.517 -0.626 4.517 -0.389 3.830 0.874 3.145 2.780 4.391 p-value 0.011 0.000 0.531 0.000 0.698 0.000 0.382 0.002 0.005 0.000 bi-directional g→y g→y g→y bi-directional 9 w-bar 2.119 0.750 1.352 0.750 1.212 0.769 2.614 1.402 1.909 0.789 z-bar 4.187 -0.936 1.316 -0.936 0.791 -0.863 6.040 1.505 3.403 -0.789 p-value 0.000 0.349 0.188 0.349 0.429 0.388 0.000 0.132 0.001 0.430 z-bar tilde 3.060 -1.124 0.715 -1.124 0.287 -1.064 4.574 0.870 2.420 -1.004 p-value 0.002 0.261 0.475 0.261 0.774 0.287 0.000 0.385 0.016 0.316 y→g no causality no causality y→g y→g 10 w-bar 3.285 1.021 1.740 1.021 1.667 1.032 3.060 0.661 3.279 1.072 z-bar 8.549 0.077 2.768 0.077 2.495 0.121 7.709 -1.270 8.529 0.269 p-value 0.000 0.939 0.006 0.939 0.013 0.904 0.000 0.204 0.000 0.788 z-bar tilde 6.624 -0.297 1.901 -0.297 1.679 -0.261 5.937 -1.397 6.607 -0.140 p-value 0.000 0.767 0.057 0.767 0.093 0.794 0.000 0.163 0.000 0.889 y→g y→g y→g y→g y→g 11 w-bar 1.408 1.073 1.013 1.073 0.948 1.088 1.413 1.207 1.347 1.060 z-bar 1.526 0.272 0.050 0.272 -0.196 0.329 1.546 0.776 1.299 0.224 p-value 0.127 0.786 0.961 0.786 0.845 0.742 0.122 0.438 0.194 0.823 z-bar tilde 0.887 -0.138 -0.319 -0.138 -0.520 -0.091 0.903 0.274 0.702 -0.177 p-value 0.375 0.891 0.750 0.891 0.603 0.927 0.366 0.784 0.483 0.860 no causality no causality no causality no causality no causality note. as per table 3. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 24 table 5. bias-corrected granger causality test results. expenditure specification 1 specification 2 specification 3 specification 4 specification 5 area peacock-wiseman mann musgrave gupta goffman d(g) d(y) d(g/y) d(y) d(y) d(g/y) d(g/y) d(y/p) d(y/p) d(g/y) d(g/p) d(y/p) d(y/p) d(g/p) d(g) d(y/p) d(y/p) d(g) 1 hpj 33.811 1.270 0.682 1.270 2.935 0.647 19.080 0.821 24.607 1.504 p-value 0.000 0.260 0.409 0.260 0.087 0.421 0.000 0.365 0.000 0.220 estimator 0.385 -0.027 -0.066 -0.027 -0.136 -0.020 0.305 0.020 0.326 -0.030 p-value 0.000 0.260 0.409 0.260 0.087 0.421 0.000 0.365 0.000 0.220 y→g no causality y→g y→g y→g 2 hpj 0.137 0.482 6.632 0.482 8.524 0.220 0.523 0.063 0.000 0.335 p-value 0.711 0.487 0.010 0.487 0.004 0.639 0.470 0.801 0.986 0.563 estimator 0.093 -0.004 -0.672 -0.004 -0.755 -0.003 -0.179 -0.001 0.005 -0.003 p-value 0.711 0.487 0.010 0.487 0.004 0.639 0.470 0.801 0.985 0.563 no causality y→g y→g no causality no causality 3 hpj 8.584 0.032 0.064 0.032 0.057 0.108 7.205 0.879 6.324 0.041 p-value 0.003 0.857 0.800 0.857 0.812 0.742 0.007 0.348 0.012 0.840 estimator 0.559 0.002 0.049 0.002 -0.046 0.003 0.452 0.009 0.475 0.002 p-value 0.003 0.180 0.800 0.857 0.812 0.742 0.007 0.348 0.012 0.839 y→g no causality no causality y→g y→g 4 hpj 76.201 0.429 2.222 0.429 0.912 1.000 40.331 7.728 70.472 0.334 p-value 0.000 0.513 0.136 0.513 0.340 0.317 0.000 0.005 0.000 0.563 estimator 0.642 0.015 0.121 0.015 0.077 0.023 0.484 0.054 0.611 0.013 p-value 0.000 0.513 0.136 0.513 0.339 0.317 0.000 0.005 0.000 0.563 y→g no causality no causality bi-directional y→g 5 hpj 52.275 20.878 5.965 20.878 3.099 19.524 35.574 18.975 47.383 17.096 p-value 0.000 0.000 0.015 0.000 0.078 0.000 0.000 0.000 0.000 0.000 estimator 0.620 0.084 0.234 0.084 0.168 0.082 0.506 0.076 0.582 0.077 p-value 0.000 0.000 0.015 0.000 0.078 0.000 0.000 0.000 0.000 0.000 bi-directional bi-directional bi-directional bi-directional bi-directional ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 25 table 5. (cont). 6 hpj 30.772 3.555 0.065 3.556 0.174 2.841 20.409 5.291 26.481 2.579 p-value 0.000 0.059 0.799 0.059 0.676 0.092 0.000 0.021 0.000 0.108 estimator 0.731 0.025 0.033 0.025 -0.053 0.022 0.496 0.034 0.668 0.021 p-value 0.000 0.059 0.799 0.059 0.676 0.092 0.000 0.021 0.000 0.108 bi-directional g→y g→y bi-directional y→g 7 hpj 39.533 0.229 0.490 0.229 0.112 0.161 9.266 4.865 30.563 0.038 p-value 0.000 0.633 0.484 0.633 0.737 0.688 0.002 0.027 0.000 0.845 estimator 0.336 -0.015 0.063 -0.015 -0.031 0.013 0.198 0.057 0.293 -0.006 p-value 0.000 0.632 0.484 0.632 0.737 0.688 0.002 0.027 0.000 0.845 y→g no causality no causality bi-directional y→g 8 hpj 15.101 0.801 0.008 0.801 0.110 0.811 10.690 0.288 13.703 0.934 p-value 0.000 0.371 0.931 0.371 0.740 0.368 0.001 0.592 0.000 0.334 estimator 0.610 -0.009 -0.014 -0.009 -0.053 -0.009 0.523 -0.005 0.575 -0.010 p-value 0.000 0.371 0.931 0.371 0.740 0.368 0.001 0.592 0.000 0.334 y→g no causality no causality y→g y→g 9 hpj 13.539 1.559 14.192 1.559 11.572 1.800 9.443 7.081 13.781 1.584 p-value 0.000 0.212 0.000 0.212 0.001 0.180 0.002 0.008 0.000 0.208 estimator 0.996 0.008 0.976 0.008 0.877 0.008 0.875 0.016 0.995 0.008 p-value 0.000 0.212 0.000 0.212 0.001 0.180 0.002 0.008 0.000 0.208 y→g y→g y→g bi-directional y→g 10 hpj 26.708 1.568 4.004 1.568 2.209 1.375 15.501 0.994 23.595 1.863 p-value 0.000 0.211 0.045 0.211 0.137 0.241 0.000 0.319 0.000 0.172 estimator 0.743 -0.015 0.274 -0.015 0.202 -0.014 0.528 0.012 0.688 -0.016 p-value 0.000 0.211 0.045 0.210 0.137 0.241 0.000 0.319 0.000 0.172 y→g y→g no causality y→g y→g 11 hpj 20.952 0.145 6.709 0.145 4.520 0.109 7.875 5.655 17.461 0.164 p-value 0.000 0.703 0.010 0.703 0.034 0.742 0.005 0.017 0.000 0.686 estimator 1.222 -0.002 0.677 -0.002 0.551 -0.002 1.007 -0.010 1.103 -0.002 p-value 0.000 0.703 0.010 0.703 0.033 0.742 0.005 0.017 0.000 0.686 y→g y→g y→g bi-directional y→g note. as per table 3. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 26 5. discussion and conclusion the paper examined alternative hypotheses that describe the relationship between public expenditure and output in a panel of 28 european economies over the 1995-2018 period using the panel data methods. the respective variables were found to be cross-sectionally dependent (expected result, given the tight and extensive political and economic integration between european economies), and thus, the appropriate unit root and cointegration tests (pesaran cadf and westerlund) were used. the variables were characterised by a mix of (1)i and (0)i integration order, and the long-run equilibrium relationship between public expenditure and output was established in every specification and expenditure category. the strong evidence of cointegration is an indication of robust results, as wagner’s hypothesis presumes a strong equilibrium relationship between public expenditure and gdp at earlier stages of development, and a weaker relationship at later stages. this result is in line with other studies of wagner’s hypothesis for the advanced and industrialised economies in the post-wwii period (the study of g7 economies by kolluri et al., 2000; of six european countries by thornton, 1999; and of italy by magazzino, 2012). the evidence of cointegration was somewhat weaker in the social protection expenditure category. while, as put by shelton (2007), the population ageing in european countries has necessitated greater social security and protection expenditure and thus stronger relationship with gdp (per capita), the previously strong relationship between social expenditure and gdp in the well-established welfare states of europe has weakened in recent decades, due to the structural change of the economy and the reforms of the social security and welfare systems that these economies have been undergoing in the recent decades (carter, 2003; kuckuck, 2014, p. 150). this likely affected the social protection category the most. the causality running from output to public expenditure along wagner’s lines was identified in the majority of cases. the result was consistent across the spending categories. given the robustness checks performed, it was also consistent across the periods as well as in the presence of cyclical dummies. the result supports the findings of previous research in european context, e.g. magazzino (2011) and lamartina and zaghini (2011). there were a number of instances, when there was no causality in either direction between gdp and public expenditure (14.0% of all cases in toda-yamamoto test, and 30.9% and 23.6% in dumitrescu-hurlin and juodis-karavias-sarafidis tests). the absence of causality, however, was not a characteristic of a particular expenditure category, and thus can be attributed to the properties of the causality test or the specification of wagner’s hypothesis. the keynesian hypothesis of sole influence of public expenditure on output and the absence of reverse causality did not find support (zero cases in toda-yamamoto test, and three and two cases in dumitrescu-hurlin and juodis-karavias-sarafidis tests, respectively, in general public services and public order and safety expenditure categories). the missing causality in these two categories may be explained by the structure of the panel: the expansion of the public order and general administration activities alongside output growth is typical for developing economies with a nascent state apparatus, which is clearly not the case of european countries that ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 27 experienced such expansion in the 19th century. the bilateral causality was encountered in 37.0%, 10.9% and 20.0% of cases under the three tests. overall, while bi-directional relationship between public expenditure and output was present, which to certain extend contradicts wagner’s hypothesis, the keynesian hypothesis of sole unidirectional influence of public expenditure on output can be rejected (in this present study). there are multiple policy implications of the results. firstly, the strong evidence of cointegration between output and expenditure and the absence of government expenditure to output causality along keynesian lines may imply limited effectiveness of the short-term spending cuts and rises in government expenditure, given that expenditure will tend to return to a long-run equilibrium level. in this context, as noted by magazzino (2011) and akitoby et al (2006), structural reforms would be needed to help achieve the social and economic objectives that fiscal policy is supposed to accomplish. secondly, the strong output-to-expenditure causality along wagner’s lines and sustained public expenditure was observed despite the slowdown of european economic growth in recent decades. the factors that helped to maintain strong long-term relationships and causality along wagner’s lines included increased revenue collection of the governments, a decrease in the size of the shadow economy, and the growing demand for public expenditure in certain functional areas, e.g. increase in health spending due to new health threats, or the projected expansion of the defence expenditures due to uncertain geopolitical situation in europe. thirdly, as argued by afonso et al (2005), the wagner’s relationship is necessarily weakened over time, due to the improvement of the quality of goods and services provided by the public sector, to the higher efficiency of the sector and slower increase (and perhaps) decrease in spending in the industrial economies. the analysis conducted in this paper for the sub-periods did not demonstrate any weakening of the ‘output-spending’ cointegration in later periods. fourthly, as noted by benavides et al (2013, pp. 71-72), the causality along wagner’s as opposed to keynes’ lines points to the limited role of political factors and political decision-making in the determination of economic outcomes. this empirical result would be contrary to the actual experience of the developed economies in the postwwii period (expansion of public policies, growth of bureaucracy and the predominance of redistribution as opposed to efficiency objectives and logic in the public policies), and the premises of the public choice and new political economy theories (buchanan, tullock, 1977). lastly, the identified weakness of keynesian hypothesis may be attributed to the crowding out of private investment by public spending, a phenomenon that has been well documented in the literature (ashauer, 1989, erden, holcombe, 2005). the relevant policy implication is the absence of any automatic positive effect of greater public spending on growth and productivity. in a related vein, the weakness of keynesian causality may be explained by the weakening of the positive effects of public spending on total factor productivity that has been observed starting from the 1980s. with regard to both phenomena, lachler and aschauer (1998) stressed the importance of savings-financed increase in public spending and capital (as opposed to an increase in public consumption that merely brings greater public debt and higher current and future taxes) for the full manifestation of positive gains of fiscal effort. future research may modify the present model and setting by considering the cyclical components of government expenditure and output and alternative channels through which the expenditure is directed in the economy, by incorporating the effects of discretionary fiscal policy, and by investigating ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 28 wagner’s hypothesis at the level of sub-national finances (kucukkale, yamak, 2012; inshauspe et al, 2020). future empirical studies may consider the role of public enterprises (in addition to the level of spending), the role of government regulation and legislative activity as a proxy or measure of the government size, the breaks and non-linearities in the relationship (in line with armey-rahn hypothesis of optimal size of government). they may also continue to experiment with alternative functional forms of the relationship and alternative definitions of dependent and independent variables (peacock, scott, 2000). the bulk of the research on wagner’s hypothesis has been on variables (e.g. public spending or the scope of legislative activity of the government) that are amenable to quantitative analysis. a more important issue is the nature of the government itself. as noted by lamartina and zaghini (2011), the pursuit of selfor vested-interests (alongside corruption and moral hazard behaviour) by the bureaucracy would render expansion of the government involvement in social and economic affairs undesirable, notwithstanding the growth of the economy and objective need for ‘bigger’ government. in a related vein, the growth of the spending on public services and administration that is motivated by the self-interest in the expansion of bureaucratic prominence and functions or proliferation of government programs due to lobbying efforts are examples of the growth of public spending that takes place irrespective of output or economic growth (niskanen, 1971). wagner’s hypothesis is thus a partial explanation of the public sector growth. references abizadeh, s., gray, j. (1985). wagner's law: a pooled time series, cross-section analysis. national tax journal 38, 209-218. https://doi.org/10.1086/ntj41792010 afonso, a., alves, j. (2017). reconsidering wagner's law: evidence from the functions of the government. applied economics letters 24(5), 346-350. https://doi.org/10.1080/13504851 .2016.1192267 afonso, a., jalles, j. (2014). causality for the government budget and economic growth. lisboa school of economics and management working paper no. wp07/2014/de/uece, pp. 1-5. https://doi.org/10.2139/ssrn.2429909 afonso, a., schuknecht, l., tanzi, v. (2005). public sector efficiency: an international comparison. public choice 123, 321-347. https://doi.org/10.1007/s11127-005-7165-2 ahsan, m., kwan, a., sahni, b. (1996). cointegration and wagner's hypothesis: time series evidence for canada. applied economics 28(8), 1055-1058. https://doi.org/10.1080/000368496328182 akitoby, b., clement, b., gupta, s., inchauste, g. (2006). public spending, voracity, and wagner's law in developing countries. european journal of political economy 22(4), 908-924. https://doi.org/10.1016/j.ejpoleco.2005.12.001 alezina, a., wacziarg, r. (1998). openness, country size and government. journal of public economics 69(3), 305-321. https://doi.org/10.1016/s0047-2727(98)00010-3 ansari, m., gordon, d., akuamoah, c. (1997). keynes versus wagner: public expenditure and national income for three african countries. applied economics 29(4), 543-550. https://doi.org/ 10.1080/000368497327038 apergis, n., christou, c., hassapis, c. (2013). convergence in public expenditures across eu countries: evidence from club convergence. economic and finance research: an open access journal 1(1), 45-59. https://doi.org/10.1080/21649480.2013.862735 aschauer, d. (1989). public investment and productivity growth in the group of seven. economic perspectives, federal reserve bank of chicago 13, 17-25. https://doi.org/10.1086/ntj41792010 https://doi.org/10.1080/13504851.2016.1192267 https://doi.org/10.1080/13504851.2016.1192267 https://doi.org/10.2139/ssrn.2429909 https://doi.org/10.1007/s11127-005-7165-2 https://doi.org/10.1080/000368496328182 https://doi.org/10.1016/j.ejpoleco.2005.12.001 https://doi.org/10.1016/s0047-2727(98)00010-3 https://doi.org/10.1080/000368497327038 https://doi.org/10.1080/000368497327038 https://doi.org/10.1080/21649480.2013.862735 ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 29 ayad, h., belmokaddem, m. (2017). financial development, trade openness and economic growth in mena countries: tydl panel causality approach. theoretical and applied economics 24(1), 233246. azcona, n. (2022). business cycle co-movement in europe: trade, industry composition and the single currency. open economies review 33(1), 121-139. https://doi.org/10.1007/s11079-021-09625-7 babatunde, m. (2011). a bound testing analysis of wagner's law in nigeria 1970-2006. applied economics 43(21), 2843-2850. https://doi.org/10.1080/00036840903425012 bairam, e. (1992). variable elasticity and wagner's law. public finance 3, 491-495. barro, r. (1991). economic growth in a cross section of countries. quarterly journal of economics 106(2), 407-443. https://doi.org/10.2307/2937943 baumol, w. (1967). macroeconomics of unbalanced growth: the anatomy of urban crisis. american economic review 57 (3), 415-426. benavides, d., venegas-martinez, f., santiago, v. (2013). la ley de wagner versus la hipotesis keynesiana: el case de mexico, 1950-2009. investigacion economica lxxii(283), 69-98. https://doi.org/10.1016/s0185-1667(13)72587-6 biswal, b., dhawan, u., lee, h.-y. (1999). testing wagner versus keynes using disaggregated public expenditure data for canada. applied economics 31(10), 1283-1291. https://doi.org/ 10.1080/000368499323490 bohl, m. (1996). some international evidence on wagner's law. public finance 51(2), 185-200. buchanan, j., tullock, g. (1977). the expanding public sector: wagner squared. public choice 31, 147-150. https://doi.org/10.1007/bf01718982 burney, n. (2002). wagner's hypothesis: evidence from kuwait using cointegration tests. applied economics 34(1), 49-57. https://doi.org/10.1080/00036840010027540 caporale, g., pittis, n. (1999). unit root testing using covariates: some theory and evidence. oxford bulletin of economics and statistics 61(4), 583-595. https://doi.org/10.1111/1468-0084.00145 carter, m. (2003). three essays on empirical patterns in functional distribution for a cross section of countries: evidence from the extended penn world tables, 1963-1996. new school university, ph.d. dissertation. chang, t., liu, w., caudill, s. (2004). a re-examination of wagner's law for ten countries based on cointegration and error-correction modelling techniques. applied financial economics 14(8), 577589. https://doi.org/10.1080/0960310042000233872 courakis, a., moura-roque, f., tridimas, g. (1993). public expenditure growth in greece and portugal: wagner's law and beyond. applied economics 25(1), 125-134. https://doi.org/10.1080/ 00036849300000121 cuddington, j., ludema, r., jayasuriya, s. (2002). prebisch-singer redux. washington: us international trade commission. dao, m. (1995). determinants of government expenditures: new evidence from disaggregative data. oxford bulletin of economics and statistics 57(1), 67-76. https://doi.org/10.1111/j.14680084.1995.tb00027.x diamond, j. (1977). wagner's "law" and the developing countries. the developing economies 15(1), 37-59. https://doi.org/10.1111/j.1746-1049.1977.tb00369.x dollery, b., singh, s. (2000). explaining the real size of government in australia: an application of the ferris and west model. economic analysis and policy 30(2), 157-173. https://doi.org/10.1016/ s0313-5926(00)50018-6 dumitrescu, e.-i., hurlin, c. (2012). testing for granger non-causality in heterogeneous panels. economic modelling 29(4), 1450-1460. https://doi.org/10.1016/j.econmod.2012.02.014 erden, l., holcombe, r. (2005). the effects of public investment on private investment in developing economies. public finance review 33(5), 575-602. https://doi.org/10.1177/1091142105277627 gandhi, v. p. (1971). wagner's law of public expenditure: do recent cross-section studies confirm it? public finance 26, 44-56. ghazy, n., ghoneim, h., paparas, d. (2021). the validity of wagner's law in egypt from 1960-2018. review of economics and political science, 6(2), 98-116. https://doi.org/10.1108/reps-0120200004 goffman, j. (1968). on the empirical testing of wagner's law: a technical note. public finance 3(3), 359-364. granger, c.w.j. (1988). aggregation of time series variables a survey. federal reserve bank of minneapolis, discussion paper no. 1, pp. 1-26. https://doi.org/10.21034/dp.1 gupta, s. (1967). public expenditure and economic growth: a time series analysis. public finance 22(4), 423-461. https://doi.org/10.1007/s11079-021-09625-7 https://doi.org/10.1080/00036840903425012 https://doi.org/10.2307/2937943 https://doi.org/10.1016/s0185-1667(13)72587-6 https://doi.org/10.1080/000368499323490 https://doi.org/10.1080/000368499323490 https://doi.org/10.1007/bf01718982 https://doi.org/10.1080/00036840010027540 https://doi.org/10.1111/1468-0084.00145 https://doi.org/10.1080/0960310042000233872 https://doi.org/10.1080/00036849300000121 https://doi.org/10.1080/00036849300000121 https://doi.org/10.1111/j.1468-0084.1995.tb00027.x https://doi.org/10.1111/j.1468-0084.1995.tb00027.x https://doi.org/10.1111/j.1746-1049.1977.tb00369.x https://doi.org/10.1016/s0313-5926(00)50018-6 https://doi.org/10.1016/s0313-5926(00)50018-6 https://doi.org/10.1016/j.econmod.2012.02.014 https://doi.org/10.1177/1091142105277627 https://doi.org/10.1108/reps-01-20200004 https://doi.org/10.1108/reps-01-20200004 https://doi.org/10.21034/dp.1 ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 30 huang, c.-j. (2006). government expenditures in china and taiwan: do they follow wagner's law? journal of economic development 31(2), 139-148. inchauspe, j., kobir, m., macdonald, g. (2020). wagner's law and the dynamics of government spending in indonesia. bulletin of indonesian economic studies 58(1), 1-22. https://doi.org/10.1080/00074918.2020.1811837 islam, a. (2001). wagner's law revisited: cointegration and exogeneity tests for usa. applied economics 8, 509-515. https://doi.org/10.1080/13504850010018743 iyare, s., lorde, t. (2004). co-integration, causality and wagner's law: tests for selected caribbean economies. applied economics letters 11(13), 815-825. https://doi.org/10.1080/ 1350485042000254881 jaen garcia, m. (2004). la ley de wagner: un analisis sintetico. instituto de estudios fiscales, working paper no. 6/04, pp. 1-59. juodis, a., karavias, y., sarafidis, v. (2021). a homogeneous approach to testing for granger noncausality in heterogeneous panels. empirical economics 60, 93-112. https://doi.org/10.1007/ s00181-020-01970-9 karagianni, s., pempetzoglou, m. (2009). evidence for non-linear causality between public spending and income in the european union countries. journal of applied business research 25, 69-82. https://doi.org/10.19030/jabr.v25i1.1049 koester, g., priesmeier, c. (2013). does wagner's law ruin the sustainability of german public finances? finanzarchiv / public finance 69(3), 256-288. https://doi.org/10.1628/ 001522113x671128 kolluri, b., panik, m., wahab, m. (2000). government expenditure and economic growth: evidence from g7 countries. applied economics 32(8), 1059-1068. https://doi.org/10.1080/000368400322110 kuckuck, j. (2014). testing wagner's law at different stages of economic development. finanzarchiv / public finance analysis 70(1), 128-168. https://doi.org/10.1628/001522114x679183 kucukkale, y., yamak, r. (2012). cointegration, causality and wagner's law with disaggregated data: evidence from turkey, 1968-2004. mpra paper no. 36894. lachler, u., aschauer, d. (1998). public investment and economic growth in mexico. world bank, policy research working paper no. 1964. lamartina, s., zaghini, a. (2011). increasing public expenditure: wagner's law in oecd countries. german economic review 12(2), 149-164. https://doi.org/10.1111/j.1468-0475.2010.00517.x legrenzi, g. (2000). the long-run relationship between public sector size and economic growth: income-elasticity and causality of the italian general government expenditure (1861-1998). rivista internazionale di scienze economiche e commerciali 47(3), 415-437. lin, c.-a. (1995). more evidence on wagner's law for mexico. public finance 50(2), 267-277. loizides, j., vamvoukas, g. (2005). government expenditure and economic growth: evidence from trivariate causality testing. journal of applied economics 8(1), 125-152. https://doi.org/10.1080/15140326.2005.12040621 lopez, l., weber, s. (2017). testing for granger causality in panel data. university of neuchatel, institute for economic research working paper no. 17-03, pp. 1-13. https://doi.org/10.1177/1536867x1701700412 magazzino, c. (2011). disaggregated public spending, gdp and money supply: evidence for italy. european journal of economics, finance and administrative sciences 41, 118-131. magazzino, c. (2012). wagner versus keynes: public spending and national income in italy. journal of policy modeling 34(6), 890-905. https://doi.org/10.1016/j.jpolmod.2012.05.012 mann, a. (1980). wagner's law: an econometric test for mexico, 1925-1970. national tax journal 33, 189-201. https://doi.org/10.1086/ntj41862301 mueller, d. (1987). the growth of government: a public choice perspective. imf staff papers 34(1), 115-149. https://doi.org/10.2307/3867026 musgrave, r. (1969). fiscal systems. new haven: yale university press. narayan, p., nielsen, i., smyth, r. (2008). panel data, cointegration, causality and wagner's law: empirical evidence from chinese provinces. china economic review 19(2), 297-307. https://doi.org/10.1016/j.chieco.2006.11.004 narayan, s., rath, b., narayan, p. (2012). evidence of wagner's law from indian state. economic modelling 29(5), 1548-1557. https://doi.org/10.1016/j.econmod.2012.05.004 niskanen, w. (1971). bureaucracy and representative government. chicago: aldine-atherton. north, d., wallis, j. (1982). longer-run determinants of the economic role of the us government. american economic review (papers and proceedings) 72(2), 336-340. organisation for economic cooperation and development / oecd (2011). government at a glance, 2011. paris: oecd publishing. https://doi.org/10.1080/00074918.2020.1811837 https://doi.org/10.1080/13504850010018743 https://doi.org/10.1080/1350485042000254881 https://doi.org/10.1080/1350485042000254881 https://doi.org/10.1007/s00181-020-01970-9 https://doi.org/10.1007/s00181-020-01970-9 https://doi.org/10.19030/jabr.v25i1.1049 https://doi.org/10.1628/001522113x671128 https://doi.org/10.1628/001522113x671128 https://doi.org/10.1080/000368400322110 https://doi.org/10.1628/001522114x679183 https://doi.org/10.1111/j.1468-0475.2010.00517.x https://doi.org/10.1080/15140326.2005.12040621 https://doi.org/10.1177/1536867x1701700412 https://doi.org/10.1086/ntj41862301 https://doi.org/10.2307/3867026 https://doi.org/10.1016/j.chieco.2006.11.004 https://doi.org/10.1016/j.econmod.2012.05.004 ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 31 oxley, l. (1994). cointegration, causality and wagner's law: a test for britain 1870-1913. scottish journal of political economy 41(3), 286-298. https://doi.org/10.1111/j.1467-9485.1994.tb01127.x peacock, a., wiseman, j. (1979). approaches to the analysis of government expenditure growth. public finance quarterly 7(1), 3-23. https://doi.org/10.1177/109114217900700101 peacock, a., scott, a. (2000). the curious attraction of wagner's law. public choice 102, 1-17. https://doi.org/10.1023/a:1005032817804 persyn, d., westerlund, j. (2008). error-correction-based cointegration tests for panel data. the stata journal 8(2), 232-241. https://doi.org/10.1177/1536867x0800800205 rajaguru, g. (2002). the distortionary effects of temporal aggregation on granger causality. bond university business school publications, pp. 1-37. ray, p., pal, p. (2022). fiscal stimulus and the ghost of keynes: an evolutionary chronicle. in n. yoshino, r. paramanik, & a. kumar (eds.), studies in international economics and finance. singapore: springer. https://doi.org/10.1007/978-981-16-7062-6_4 sedrakyan, g., varela-candamio, l. (2019). wagner's law vs. keynes' hypothesis in very different countries (armenia and spain). journal of policy modeling 41(4), 747-762. https://doi.org/10.1016/j.jpolmod.2019.02.011 shelton, c. (2007). the size and composition of government expenditure. journal of public economics 91(11-12), 2230-2260. https://doi.org/10.1016/j.jpubeco.2007.01.003 sideris, d. (2007). wagner's law in 19th century greece: a cointegration and causality analysis. bank of greece working paper no. 64, pp. 1-18. https://doi.org/10.2139/ssrn.4164000 sinha, d. (2007). does the wagner's law hold for thailand? a time series study. mpra paper no. 2560. tang, t.c. (2001). testing the relationship between government expenditure and national income in malaysia, analisis 8(1/2), 37-51. thornton, j. (1999). cointegration, causality and wagner law in 19th century europe. applied economics letters 6(7), 413-416. https://doi.org/10.1080/135048599352916 timm, h. (1961). das gesetz der wachsenden staatsausgaben. finanzarchiv nf 21, 201-247. tobin, d. (2005). economic liberalization, the changing role of the state and "wagner's law": china's development experience since 1978. world development 33(5), 729-743. https://doi.org/10.1016/j.worlddev.2004.12.001 toda, h., yamamoto, t. (1995). statistical inference in vector autoregressions with possibly integrated processes. journal of econometrics 66(1-2), 225-250. https://doi.org/10.1016/03044076(94)016168 wagner, a. (1912). les fondements de l'economie politique. paris: girard & briere. westerlund, j. (2007). testing for error correction in panel data. oxford bulletin of economics and statistics 69(6), 709-748. https://doi.org/10.1111/j.1468-0084.2007.00477.x xiao, j., juodis, a., karavias, y., sarafidis, v. (2021). improved tests for granger non-causality in panel data. munich: mpra paper no. 107180, pp. 1-11. yay, t., tastan, h. (2009). growth of public expenditures in turkey during the 1950-2004 period: an econometric analysis. romanian journal of economic forecasting 4, 101-118. ziramba, e. (2009). wagner's law: an econometric test for south africa, 1960-2006. south african journal of economics 76(4), 596-606. https://doi.org/10.1111/j.1813-6982.2008.00218.x https://doi.org/10.1111/j.1467-9485.1994.tb01127.x https://doi.org/10.1177/109114217900700101 https://doi.org/10.1023/a:1005032817804 https://doi.org/10.1177/1536867x0800800205 https://doi.org/10.1016/j.jpolmod.2019.02.011 https://doi.org/10.1016/j.jpubeco.2007.01.003 https://doi.org/10.2139/ssrn.4164000 https://doi.org/10.1080/135048599352916 https://doi.org/10.1016/j.worlddev.2004.12.001 https://doi.org/10.1016/0304-4076(94)016168 https://doi.org/10.1016/0304-4076(94)016168 https://doi.org/10.1111/j.1468-0084.2007.00477.x https://doi.org/10.1111/j.1813-6982.2008.00218.x ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 32 appendix figure 1. dynamics of output and expenditure categories 11.7 11.8 11.9 12.0 12.1 12.2 12.3 12.4 1995 2000 2005 2010 2015 y 9.6 9.7 9.8 9.9 10.0 10.1 10.2 1995 2000 2005 2010 2015 y/p 10.9 11.0 11.1 11.2 11.3 11.4 11.5 95 00 05 10 15 total 8.9 9.0 9.1 9.2 9.3 95 00 05 10 15 total/p 3.72 3.76 3.80 3.84 3.88 95 00 05 10 15 total/y 8.7 8.8 8.9 9.0 9.1 9.2 9.3 95 00 05 10 15 econ 1.45 1.50 1.55 1.60 1.65 1.70 95 00 05 10 15 econ/y 6.7 6.8 6.9 7.0 7.1 95 00 05 10 15 econ/p 7.5 7.6 7.7 7.8 7.9 95 00 05 10 15 def .0 .1 .2 .3 .4 .5 95 00 05 10 15 def/y 5.45 5.50 5.55 5.60 5.65 5.70 95 00 05 10 15 def/p 8.7 8.8 8.9 9.0 9.1 9.2 9.3 95 00 05 10 15 edu 1.52 1.56 1.60 1.64 1.68 1.72 95 00 05 10 15 edu/y 6.7 6.8 6.9 7.0 7.1 95 00 05 10 15 edu/p 8.6 8.8 9.0 9.2 9.4 9.6 95 00 05 10 15 heal 1.5 1.6 1.7 1.8 1.9 95 00 05 10 15 heal/y 6.6 6.8 7.0 7.2 7.4 95 00 05 10 15 heal/p ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 33 7.4 7.6 7.8 8.0 8.2 8.4 95 00 05 10 15 pub .44 .48 .52 .56 .60 95 00 05 10 15 pub/y 5.5 5.6 5.7 5.8 5.9 6.0 6.1 95 00 05 10 15 pub/p 9.8 10.0 10.2 10.4 10.6 95 00 05 10 15 soc 2.65 2.70 2.75 2.80 2.85 95 00 05 10 15 soc/y 7.8 7.9 8.0 8.1 8.2 8.3 95 00 05 10 15 soc/p 9.15 9.20 9.25 9.30 9.35 9.40 9.45 95 00 05 10 15 gener 1.6 1.7 1.8 1.9 2.0 2.1 95 00 05 10 15 gener/y 7.08 7.12 7.16 7.20 7.24 7.28 95 00 05 10 15 gener/p 6.6 6.8 7.0 7.2 7.4 95 00 05 10 15 env -.6 -.5 -.4 -.3 95 00 05 10 15 env/y 4.4 4.6 4.8 5.0 5.2 95 00 05 10 15 env/p 7.0 7.2 7.4 7.6 7.8 8.0 95 00 05 10 15 recr .00 .05 .10 .15 .20 .25 95 00 05 10 15 recr/y 5.1 5.2 5.3 5.4 5.5 5.6 5.7 95 00 05 10 15 recr/p 6.8 6.9 7.0 7.1 7.2 7.3 7.4 95 00 05 10 15 hous -.8 -.7 -.6 -.5 -.4 -.3 -.2 95 00 05 10 15 hous/y 4.80 4.85 4.90 4.95 5.00 5.05 5.10 95 00 05 10 15 hous/p note. the series are expressed in natural logarithms. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 34 table 6. cross-sectional results. expenditure area coefficient specification 1 specification 2 specification 3 specification 4 specification 5 peacock-wiseman mann musgrave gupta goffman coeff. p-value coeff. p-value coeff. p-value coeff. p-value coeff. p-value total c -1.319 0.000 3.286 0.000 3.106 0.000 -1.310 0.001 1.066 0.767 cross-sect. 1.041 0.000 0.041 0.020 0.068 0.058 1.053 0.000 1.027 0.008 economic affairs c -2.446 0.000 2.159 0.000 2.525 0.000 -1.855 0.001 0.485 0.886 cross-sect. 0.953 0.000 -0.047 0.074 -0.094 0.073 0.887 0.000 0.865 0.016 defense c -4.801 0.000 -0.196 0.773 2.109 0.054 -2.338 0.043 0.069 0.986 cross-sect. 1.036 0.000 0.036 0.515 -0.188 0.084 0.799 0.000 0.771 0.061 education c -2.778 0.000 1.827 0.000 0.952 0.046 -3.474 0.000 -1.088 0.745 cross-sect. 0.984 0.000 -0.016 0.491 0.068 0.148 1.054 0.000 1.026 0.005 health c -3.937 0.000 0.669 0.187 0.678 0.432 -3.673 0.000 -1.362 0.722 cross-sect. 1.086 0.000 0.086 0.044 0.103 0.237 1.082 0.000 1.062 0.010 public order and safety c -3.611 0.000 0.994 0.024 3.261 0.000 -1.230 0.011 1.221 0.735 cross-sect. 0.961 0.961 -0.039 0.262 -0.275 0.000 0.717 0.000 0.683 0.068 social protection c -2.969 0.000 1.636 0.000 0.927 0.080 -3.473 0.000 -1.113 0.764 cross-sect. 1.092 0.000 0.092 0.001 0.183 0.001 1.167 0.000 1.142 0.005 general public services c -3.447 0.000 1.158 0.015 1.231 0.115 -3.285 0.000 -0.809 0.828 cross-sect. 1.058 0.000 0.058 0.128 0.063 0.414 1.057 0.000 1.021 0.011 environmental protection c -5.816 0.000 -1.211 0.054 -0.996 0.333 -5.287 0.000 -3.035 0.431 cross-sect. 1.067 0.000 0.067 0.183 0.060 0.556 1.033 0.000 1.019 0.013 recreation, culture, religion c -3.998 0.000 0.607 0.220 0.013 0.987 -4.303 0.000 -2.027 0.548 cross-sect. 0.960 0.000 -0.040 0.327 0.012 0.886 0.987 0.000 0.970 0.008 housing and community amenities c -4.587 0.000 0.018 0.982 1.765 0.172 -2.586 0.059 -0.274 0.942 cross-sect. 0.968 0.000 -0.032 0.633 -0.215 0.100 0.762 0.000 0.744 0.058 ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 35 table 7. toda-yamamoto panel causality test with cyclical dummy results. expenditure specification 1 specification 2 specification 3 specification 4 specification 5 area peacock-wiseman mann musgrave gupta goffman d(g) d(y) d(y) d(g/y) d(y) d(y) d(g/y) d(g/y) d(y/p) d(y/p) d(g/y) d(g/p) d(y/p) d(y/p) d(g/p) d(g) d(y/p) d(y/p) d(g) 1 lag 4 4 4 4 4 4 2 2 4 4 bi-directional bi-directional bi-directional y→g bi-directional lag 5 5 5 5 5 5 4 4 5 5 bi-directional bi-directional bi-directional bi-directional bi-directional 2 lag 4 4 4 4 4 4 4 4 4 4 bi-directional bi-directional bi-directional bi-directional bi-directional 3 lag 3 3 3 3 3 3 2 2 3 3 y→g no causality no causality y→g y→g lag 7 7 7 7 7 7 3 3 8 8 y→g y→g y→g y→g y→g 4 lag 4 4 4 4 4 4 3 3 4 4 y→g y→g y→g y→g y→g lag 4 4 y→g 5 lag 4 4 4 4 4 4 4 4 4 4 y→g y→g bi-directional bi-directional bi-directional lag 8 8 8 8 6 6 8 8 6 6 bi-directional bi-directional bi-directional bi-directional y→g 6 lag 2 2 2 2 2 2 2 2 2 2 y→g no causality bi-directional y→g bi-directional lag 4 4 4 4 4 4 5 5 4 4 y→g y→g no causality y→g y→g 7 lag 3 3 3 3 5 5 5 5 5 5 bi-directional bi-directional bi-directional y→g bi-directional lag 5 5 5 5 3 3 3 3 3 3 bi-directional bi-directional bi-directional y→g bi-directional 8 lag 2 2 2 2 2 2 2 2 2 2 no causality y→g y→g no causality no causality lag 3 3 3 3 3 3 3 3 3 3 no causality y→g y→g no causality no causality ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 36 table 7. (cont). 9 lag 2 2 2 2 2 2 2 2 2 2 y→g no causality no causality y→g y→g lag 3 3 3 3 3 3 3 3 5 5 y→g no causality no causality y→g bi-directional 10 lag 2 2 2 2 2 2 2 2 2 2 y→g y→g no causality y→g y→g lag 4 4 4 4 4 4 3 3 4 4 y→g y→g bi-directional y→g bi-directional 11 lag 3 3 3 3 3 3 3 3 3 3 y→g y→g bi-directional bi-directional bi-directional lag 4 4 4 4 4 4 4 4 y→g y→g bi-directional bi-directional note. as per table 3. the causality patterns that are different from the toda-yamamoto procedure with no cyclical dummy (table 3) are highlighted in bold. ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 table 8. toda-yamamoto panel causality test with positive and negative output components (whole period and subperiods) whole period 2009-2018 1995-2008 expenditure specification 1 specification 1 specification 1 area peacock-wiseman peacock-wiseman peacock-wiseman d(g) d(y) d(y) d(g) d(y) d(y) d(g) d(y) d(y) 1 lag 4 4 4 4 4 4 y+ bi-directional bi-directional y→g ybi-directional bi-directional g→y lag 5 5 y+ bi-directional ybi-directional 2 lag 4 4 4 4 4 4 y+ bi-directional bi-directional y→g yy→g y→g no causality 3 lag 3 3 3 3 3 3 y+ y→g y→g no causality yy→g y→g no causality lag 7 7 y+ bi-directional yy→g 4 lag 4 4 4 4 4 4 y+ y→g y→g bi-directional yy→g bi-directional y→g 5 lag 4 4 4 4 4 4 y+ y→g y→g y→g ybi-directional bi-directional y→g lag 8 8 y+ y→g ybi-directional 6 lag 2 2 2 2 2 2 y+ y→g y→g no causality yy→g y→g bi-directional lag 4 4 y+ y→g yy→g 7 lag 3 3 3 3 3 3 y+ y→g bi-directional y→g ybi-directional bi-directional no causality lag 5 5 y+ bi-directional yy→g 8 lag 2 2 2 2 2 2 y+ no causality no causality no causality yno causality y→g g→y lag 3 3 y+ y→g yno causality ivan d trofimov / european journal of government and economics 12(1), june 2023, 5-38 38 table 8. (cont). 9 lag 2 2 2 2 2 2 y+ y→g no causality y→g yno causality no causality no causality lag 3 3 y+ y→g yno causality 10 lag 2 2 2 2 2 2 y+ y→g no causality y→g yy→g y→g bi-directional lag 4 4 y+ y→g yy→g 11 lag 3 3 3 3 3 3 y+ y→g y→g y→g yy→g y→g bi-directional lag 4 4 y+ y→g yy→g ©the authors 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 240-261 https://doi.org/10.17979/ejge.2024.13.2.10168 submitted: dec 13, 2023 accepted: jul 8, 2024 published: dec 3, 2024 article do geopolitical risks affect stock market returns and volatilities: an analysis based on the tvp-var model ahlem lamine,1,* sirine zribi 1 1 university of economic science and management of sfax, tunisia *correspondence: ahlemlaminefakhfakh@gmail.com abstract. this study examines the effects of geopolitical risk (gpr) shocks on stock market returns and volatility across g7, brics, and gulf countries, using a time-varying parameter vector autoregression (tvpvar) model. by analyzing responses over short, medium, and long-term horizons, our findings reveal significant variations in how geopolitical risks impact stock markets across different countries and timelines. we observe that gpr-related impacts on stock returns weaken over time, while volatility effects tend to strengthen, suggesting persistent risks for investors in these markets. these insights provide new perspectives for portfolio management and investment strategies during times of geopolitical uncertainty. keywords: geopolitical risks; g7 countries; brics countries; gulf countries; stock market returns, volatility; tvp-var model jel classification: g11 ; g17 1. introduction war and border disputes have a harmful effect on financial markets (kumari et al., 2023). the start of the russian-ukrainian conflict on february 24, 2022, caused a sharp increase in geopolitical risk facing regional and international financial markets. intuitively, this risk harms financial markets directly and indirectly (umar et al., 2022). indeed, the factors influencing the dynamics of financial markets are not limited to economic and financial factors. they also include shocks induced by uncertainty (antonakakis et al., 2017). among these shocks is geopolitical risk, which covers geopolitical tensions, war risk, terrorist attacks, and military threats (alqahtani et al., 2020). geopolitical risk is defined by caldara and iacoviello (2022) as "the risks associated with wars, acts of terrorism, and tensions between states that affect the normal and peaceful course of international relations". this risk is a key determinant of stock market dynamics and investment decisions (caldara and iacoviello, 2022, baur and smales, 2020). effectively, the unpredictable disclosure of geopolitical events will harm investor sentiment (drakos, 2010). it delays the decisionmaking process of market players (salisu et al., 2022) by pushing them to postpone or divest their equity investments (antonakakis et al., 2017). this can cause a massive sell-off of stocks by investors https://creativecommons.org/licenses/by-nc/4.0/ do geopolitical risks affect stock market returns and volatilities seeking a stable future characterized by safer financial instruments (apergis and apergis, 2016). this situation causes large variations in stock volatility and a decrease in stock returns (drakos, 2010, jeribi et al., 2015, wang et al., 2020). numerous studies in the literature have examined the effects of geopolitical events, including terrorist attacks (corbet et al., 2018; papakyriakou et al., 2019), wars, and political tensions (omar et al., 2017; hudson and urquhart, 2015) on stock markets. following the russian-ukrainian conflict, governments, investors and academics are more concerned with examining the impact of the sharply rising gpr on financial markets. our study aims to extend the existing literature by investigating the impact of gpr geopolitical risk on stock market returns and volatility during a period marked by global financial crises, including the health crisis and the russian-ukrainian conflict, leading to significant changes and sudden geopolitical risks. the remaining paper is organized as follows. section 2 details the literature review. the data and methodology are presented in section 3. section 4 presents empirical results and analysis. the discussion is presented in section 5. the last section presents concluding remarks. 2. literature review based on the gpr index developed by caldara and iacoviello (2022), studies are conducted to examine the relationships between gprs and financial market dynamics. bourras et al. (2019) study the role of gpr on the volatility of 18 emerging markets and find that gpr has a significant impact on the volatility of emerging stock markets. examining the link between gpr and stock market volatility in emerging economies, salisu et al. (2022) find that emerging market volatility responds positively to gpr. based on the garch-midas approach, ndako et al. (2021) show that the gpr would increase the volatility of islamic stocks in indonesia and malaysia. other studies find that gpr has a significant impact on the commodity market (cunado et al., 2020; plakandaras et al., 2019; gkillas et al., 2020). aysan et al. (2019) demonstrate that gpr induces negative returns for bitcoin and positive price volatility. following the russian-ukrainian conflict, recent literature is more concerned with examining the impact of the gpr on financial markets. boungou and yatié (2022) reveal that the russian-ukrainian war had a negative impact on the stock market, especially for the countries bordering these warring nations. in studying the impact of the russian-ukrainian conflict on global stock markets, boubaker et al. (2022) find that the war had a negative impact on developed economies compared to emerging countries. zhang et al. (2023) use the bias-corrected lsdv estimator to study the effect of gpr on stock market volatility for 32 countries and regions. they claim that gpr has a significant positive effect on stock market volatility. by applying the wavelet coherence approach, będowska-sojka et al. (2022) analyze the impact of geopolitical risk on different types of securities. they argue that different asset classes can provide the best hedge against geopolitical risk. boungou et al. (2022) analyze the dynamic connectivity between russia, europe, the united states, and global commodity markets to see the impact of the russian-ukrainian war on global financial markets. by providing insight into the vulnerability of the constituent companies of the main stock market indices of the g7 countries to war events, abbassi et al. (2022) show that stock 242 ahlem lamine and sirine zribi prices are fragile in the face of gpr geopolitical risk and create negative abnormal returns. although a significant number of studies have examined the effects of gpr geopolitical risk on financial market dynamics, there are limitations to be explored. first, previous studies assumed that the relationship between gprs and financial market dynamics was time-invariant and used the event study approach, var, svar, or garch. however, in reality, following spikes linked to major geopolitical events such as the invasion of iraq, the gulf war and the russian-ukrainian crisis, gprs evolve over time (caldara and iacoviello, 2022). the responses of stock markets to variations in the gpr are, therefore, heterogeneous over time. secondly, most studies focus on the study of the gpr influence on a specific area and over a short period, which does not allow conclusions and comparisons to be made between the impact of different peaks linked to geopolitical events on the different stock markets. the contributions of our article are multiple: first, we focus on the time-varying responses of the stock indices of the g7, brics and golf countries to gpr shocks by distinguishing between the transitory and persistent effects of the crisis in several important respects. second, we analyze the gprs' dynamic effects on stock markets at different times and time horizons through the construction of tvp-var models. third, our estimation sample covers a longer period than previous studies and is marked by global financial crises, including the health crisis and the russian-ukrainian conflict, leading to significant and sudden changes in geopolitical risk. to the best of our knowledge, this is the first attempt to study in depth the impact of multiple pbr shocks on the dynamics of different stock markets using the tvp-var model. the lack of empirical studies may be mainly due to the absence of major geopolitical events that characterize past periods and, consequently, to a lack of data. in our case, the period under study is essentially characterized by the russian-ukrainian conflict, which can be considered a major geopolitical event. a second possible explanation is that prior to this study, the database lacked indicators to measure the importance of geopolitical events. the creation of this type of indicator made our task easier. 3. data and methodology 3.1 data specifications our research provides a monthly dataset of the geopolitical risk index (gpr). this risk is represented using the daily gpr index of caldara and iacoviello (2022). this gpr index reflects various risks resulting from changes in government, civil unrest, threats of war, military conflicts, terrorist attacks, and any tension between states and political actors that affect the peaceful course of international relations. according to caldara and iacoviello (2022), the gpr index is constructed by counting the number of articles mentioning words related to geopolitical tensions in 11 major national and international newspapers (share of the total number of press articles). gpr index data are obtained from the caldara and iacoviello webpage (https://www.matteoiacoviello.com/gpr.htm). we also consider monthly frequency data for stock index prices in the g7 countries (usa, uk, germany, france, japan, italy, canada), the brics countries (brazil, russia, india, china, south africa) and the gulf states (ksa, oman, qatar, kuwait, bahrain, uae). the stock index price data are obtained from datastream. we have chosen the sp500, ftse, nikkei, dax40, cac40, ftse mib and https://www.matteoiacoviello.com/gpr.htm do geopolitical risks affect stock market returns and volatilities s&p/tsx indices to represent the us, uk, japanese, german, french, italian, and canadian stock markets, respectively. for brics countries, china's sse, russia's rtsi, india's bse 30, brazil's bvsp and south africa's jtopi indices are used. for the gulf states, the bahraini, omani, qatari, saudi and uae stock markets are represented respectively by the bax, msm30, qeas, tasi and adx indices. we believe that these indices provide a very good representation of stock market trends in the countries in question, and we can be confident in the results found with this data. in addition, we have chosen this group of countries (g7, brics and gulf) because we believe it covers a large part of the current political scene and that these countries are most affected by the geopolitical events under consideration. we calculate monthly returns by considering the difference in the logarithmic values of two consecutive prices ri,t= ln (pi,t/pi,t-1) × 100, where ri,t represents the monthly percentage returns for index i at time t, while pi,t represents the price of index i at time t .these monthly r turns were then annualized. moreover, the volatility series are not directly observable and must be estimated. the garch(1,1) model is the simplest version of the garch models, where the autoregressive and arch components are both of order one. this model assumes that past squared error and past volatility also influence current volatility. the parameters of the garch(1,1) model are estimated using maximum likelihood. although simple, the garch (1,1) model can adequately capture the dynamics of volatility in many financial time series. what's more, the garch(1,1) model is preferred by many economists over other stochastic volatility models because of its relative ease of implementation. in fact, because this model is given by discrete-time stochastic difference equations, the likelihood function is easier to manipulate and consequently, the estimates are more accurate than continuoustime models. and since financial data is generally collected at discrete intervals, garch models are the most appropriate models for representing volatility in financial markets. indeed, despite its simplicity, we believe that the garch (1,1) model is optimal for predicting the volatility of stock market index returns for the countries selected in our sample. we apply this model to determine the conditional variance of each index i as follows: ( )       ++= → += −− 2 1, 2 1,0, 2 , 2 ,, ,0,, ,0 tiitiiiti titi tiiti c n ar σβεασ σε ε [1] where 2 tσ is the conditional variance of the residuals with the conditions α. β≥0 and α+β<1. the study period extends from january 2016 to april 2023. our data sampling period is marked by unexpected events such as the covid-19 pandemic crisis and the russian-ukrainian conflict. 244 ahlem lamine and sirine zribi 3.2 methodology the methodology used in this study is time-varying parameter vector autoregression (tvp-var). it is an innovation of the traditional var framework, which assumes that all model parameters are constant over time (antonakakis and gabauer, 2017[27]). the tvp-var approach, as a statistical technique used to estimate a vector autoregressive (var) model with time-varying parameters, eliminates the need to define an arbitrary window length for the estimation. in effect, it contributes to individual point estimates for each of the periods of the study instead of a single average point estimate for the whole period of the sample. which can capture the dynamic behavior of variables more accurately. this method is useful primarily in modeling economic and financial time series data, where relationships between variables may change over time. the tvp-var model developed by primiceri (2005) is defined by ttttpttptttt uxuybybby +θ=++++= −− ' ,1,1,0 ...... [2] [ ].',......,',1 1 ' pttt yyx −−= [3] where yt is a vector (n × 1) of observed dependent variables and b0,t...bp,t are matrices of time-varying coefficients (n × n) translated into the form of a matrix θt. xt is the (n×k) matrix, including the ordinates and offsets of the endogenous variables. ut is the independent structural shock with dimension (n × 1) assumed to be a normally distributed heteroscedastic disturbance term with zero mean and a time-varying variance-covariance matrix ωt. this matrix can be broken down as follows: ( )'11 −−=ω tttt aha where at is a lower triangular matrix, ht is a diagonal matrix.                 = 1 01 001 0001 00001 ,54,53,52,51 ,43,42,41 ,32,31 ,21 tttt ttt tt t ta αααα ααα αα α                 = t t t t t t h h h h h h ,5 ,4 ,3 ,2 ,1 0000 0000 0000 0000 0000 following primiceri (2005) (equations [5]-[7]) the time-varying parameters are assumed to depend on the following random walk process as: do geopolitical risks affect stock market returns and volatilities ),0( 1 qnvv tttt →+θ=θ − [5] ),0( 1 sntttt →+= − τταα [6] )1,0( )ln()ln( ,,1,, nhh titiititi →+= − ηησ [7] according to primiceri (2005), it is assumed that the coefficients of contemporaneous relationships between variables evolve independently in each order of the equation to simplify inference and increase estimation efficiency. thus, the error terms of the measurement equation and the transition equations (the parameters of the at matrix) are independent. given stochastic volatility, parameters must be defined with maximum likelihood estimation. markov chain monte carlo (mcmc) based on bayesian inference was also used to simulate sampling. technically, the tvp-var model has two main advantages. first, random volatility is taken into account in the estimation of the model, which significantly improves the quality of parameter estimation and avoids the problem of heteroskedasticity. second, the tvp-var model allows us to identify in a meaningful way whether the influence of exogenous factors on endogenous variables shows a structural change. indeed, in a tvp-var model as proposed by primiceri (2005), the coefficients evolve with structural changes. therefore, in order to test for time-varying effects between variables, the tvp-var model is considered as a flexible and powerful approach. it also provides a novel dynamic way of looking at the evolution of the relationship by identifying differences in the influence of various events. 4. empirical results and discussion 4.1 preliminary analyses table 1 shows the descriptive statistics for the gpr index, the monthly returns (panel a) and the conditional variance series (panel b) of the g7, brics and gulf stock market indices. as shown in table 1, all the stock market indices studied show similar behavior in terms of returns, and the range of variation in these returns is small. indeed, the bvsp index (flagship index of brazil's são paulo stock exchange) has the highest average return, followed by the bse (indian stock exchange index) and the abu dhabi stock exchange (uae) index. on the other hand, only the oman stock market index posted a negative average return over the study period. the chinese index had the lowest positive average return. in terms of standard deviation, investment in the bahrain stock market can be considered the safest, while investment in the russian stock market is very risky. the latter can be explained by the russian invasion of ukraine. skewness values are low but negative for all return series. kurtosis values are greater than three for virtually all return series (with the exception of japan, south africa, ksa, and qatar). these results indicate that the probability distribution of returns in the sample (with the exception of japan, south africa, ksa, and qatar) is asymmetric and leptokurtic and rejects the normality confirmed by jarque-bera (j.-b.) statistics. 246 ahlem lamine and sirine zribi table 1. descriptive statistics. mean maximum minimum std. dev. skewness kurtosis jb obs gpr 102.8013 325.4400 60.60164 35.82020 3.300504 19.43118 1149.709 88 panel a: indices returns g7 stock market indices usa 0.094314 1.357056 -1.518952 0.552216 -0.565972 3.751062 6.766434 88 uk 0.034732 1.386537 -1.688554 0.432372 -0.773560 6.122911 44.53588 88 germany 0.056387 1.664915 -2.040667 0.589801 -0.461879 4.193598 8.352693 88 japan 0.059130 1.668301 -1.264153 0.544833 -0.338623 3.429915 2.359461 88 france 0.068117 2.182283 -2.146030 0.573665 -0.189864 5.939288 32.20655 88 italy 0.037416 2.459574 -2.887717 0.718877 -0.470959 6.086181 38.17632 88 canada 0.060512 1.170430 -2.218500 0.459590 -1.376000 9.187664 168.1559 88 brics stock market indices china 0.002973 1.699654 -2.335511 0.555897 -0.486206 5.993300 36.31991 88 russia 0.031388 2.189547 -6.509287 1.129673 -2.499401 14.80033 602.1975 88 india 0.110502 1.530670 -2.977927 0.593425 -1.354873 9.943830 203.7182 88 brasil 0.114218 1.756908 -4.037607 0.810932 -1.529734 9.365449 182.8907 88 south africa 0.065096 1.510438 -1.343617 0.532145 0.130942 3.304914 0.592371 88 gulf stock market indices bahraïn 0.062394 1.019529 -1.957794 0.389424 -1.484343 10.44053 235.3070 88 kuwait 0.061857 1.393309 -2.530908 0.526312 -1.534887 8.832443 159.2833 88 oman -0.017718 1.127539 -1.915278 0.404916 -0.649262 7.736067 88.42715 88 qatar 0.027650 1.644088 -1.282169 0.556983 -0.102878 2.845795 0.242420 88 ksa 0.064957 1.728884 -1.848417 0.653920 -0.427031 3.342267 3.104079 88 panel b: indices volatility g7 stock market indices usa 0.349883 1.195069 0.056675 0.270802 0.922936 3.028674 12.49625 88 uk 0.194598 1.697595 0.019423 0.223137 4.866163 29.22011 2868.112 88 germany 0.339060 1.575335 0.198275 0.188541 4.120157 24.23298 1902.054 88 japan 0.288312 0.534708 0.266000 0.036826 4.547489 27.36638 2480.278 88 france 0.369096 3.516855 0.114625 0.483547 5.080932 30.89545 3231.873 88 italy 0.512339 0.826530 0.076468 0.111342 -1.404599 6.664227 78.16656 88 canada 0.266981 4.430728 0.056621 0.526840 6.121145 46.50111 7488.141 88 brics stock market indices china 0.285438 1.888137 0.206153 0.229790 5.592734 35.57790 4350.259 88 russia 1.647087 43.11977 0.302961 4.704396 8.027124 70.44637 17624.76 88 india 0.390518 7.259795 0.080986 0.768153 8.281832 74.38137 19688.74 88 brasil 0.674802 6.796237 0.411888 0.701214 7.726937 67.57642 16166.10 88 south africa 0.277656 0.859246 0.077403 0.127576 2.699911 11.33494 361.6410 88 gulf stock market indices bahraïn 0.180558 1.910986 0.071991 0.230610 5.495450 38.72667 5123.048 88 kuwait 0.289881 2.777954 1.72e-07 0.300635 6.754604 54.97319 10573.61 88 oman 0.150564 0.260339 0.073515 0.019824 1.689719 16.89985 750.2970 88 qatar 0.307896 0.856196 0.181924 0.131581 1.870552 6.776140 103.6020 88 ksa 0.416821 1.307038 0.308581 0.166302 3.346162 15.98800 782.7428 88 uae 0.263502 2.432828 0.115890 0.248337 7.767309 67.80945 16285.83 88 do geopolitical risks affect stock market returns and volatilities the descriptive statistics of the conditional variance series obtained by the garch (1,1) model are presented in panel b of table 1. this panel shows that the highest average conditional variance for the g7 countries is observed for the italian stock market index (0.512339). the highest average conditional variances for the brics and gulf countries are obtained for the russian (1.647087) and saudi (0.416821) markets, respectively. furthermore, the jarque-bera test shows that the null hypothesis of normality is rejected for all return and volatility series examined. in addition, the augmented dickey-fuller (adf) and phillips and perron (pp) tests were used to examine the stationarity of the variables. table 2 shows that all variables (both yields and volatilities) are stationary in terms of levels at the 1% significance level for both the adf and pp tests. consequently, the level data were used for the following empirical study. table 2. unit root tests variables adf test pp test gpr -3.876951*** -3.931328*** return volatility return volatility g7 stock market indices usa -10.98700*** -2.951146** -11.16284*** -2.948086** uk -9.082738*** -6.105537*** -9.096397*** -5.798166*** germany -9.991293*** -7.520084*** -9.991293*** -7.453225*** japan -10.28373*** -12.08317*** -10.30959*** -12.08317*** france -9.811079*** -7.931754*** -9.813727*** -7.898927*** italy -10.59588*** -4.062610*** -10.60848*** -4.089589*** canada -10.55670*** -5.784601*** -10.56286*** -5.674474*** brics stock market indices china -10.62961*** -19.01510*** -10.97546*** -22.73520*** russia -9.694159*** -9.037548*** -9.804686*** -9.037548*** india -9.988251*** -8.736843*** -10.01856*** -8.736843*** brasil -8.420072*** -7.516881*** -8.639780*** -7.520713*** south africa -9.653704*** -5.180757*** -9.656024*** -5.271113*** gulf stock market indices bahraïn -6.375853*** -5.371185*** -5.984285*** -5.380160*** kuwait -8.097271*** -8.997574*** -8.207790*** -8.992385*** oman -9.997543*** -6.857102*** -10.03244*** -6.911374*** qatar -9.372080*** -4.796342*** -9.372080*** -4.647228*** ksa -9.644327*** -7.207502*** -9.642112*** -8.064956*** uae -9.576446*** -8.066089*** -9.575057*** -8.035662*** ***. ** and * denote significance at 1%. 5% and 10% level respectively. 4.2 causality test results we begin our analysis with the granger linear causality test. this test will answer our research question: does geopolitical risk affect the returns and volatility of stock markets in g7, brics and gulf countries? table 3 presents the results of this test. by considering the geopolitical risk index gpr as the causal variable, the results show that the null hypothesis of granger non-causality cannot be rejected in a relatively large number of cases. indeed, for the g7 countries, geopolitical risk only affects the returns of the american and canadian markets in the granger sense. for brics countries, only chinese and indian stock market returns are strongly influenced by geopolitical risk. on the other hand, the null hypothesis of non-causality in the granger sense is rejected in the case of returns on all gulf markets. regarding stock market volatility, the linear causality test reveals a single significant causal relationship between the gpr index and stock market investment risk (extracted via a garch 248 ahlem lamine and sirine zribi model). otherwise, the alternative hypothesis of the granger test is only accepted in the case of russia. this result can be attributed to the russian-ukrainian war. however, the granger causality test is a linear test that gives a general and aggregate view of the nature of the relationship between two variables. this link is constant over the entire period and only superficially characterizes the causality between markets. additionally, as we pointed out in the statistical description, streaks are associated with fat tails, excessive kurtosis, and nonnormality. given this evidence, it is imperative to further study the dynamic relationships between the gpr index and different stock markets. the non-linear tvp-var model, therefore, constitutes the appropriate model for studying the time-varying and continuous effect of geopolitical risks on the different stock markets of the g7, brics, and gulf countries. table 3. linear causality test results: gpr as the causal variable return volatility f-stat. prob. f-stat. prob. h0 : gpr does not granger cause g7 market usa 4.29701** 0.0168 2.08703 0.1307 uk 1.10224 0.3371 0.49736 0.6100 germany 0.80586 0.4503 0.16012 0.8523 japan 0.49740 0.6100 0.51507 0.5994 france 1.19582 0.3077 0.15965 0.8527 italy 1.66218 0.1961 0.46223 0.6315 canada 2.99262* 0.0557 0.13545 0.8735 h0 : gpr does not granger cause brics stock market china 2.37190* 0.0926 0.52163 0.5955 russia 0.43239 0.6504 11.8990*** 3.e-05 india 3.18673** 0.0465 0.34763 0.7074 brasil 1.13239 0.3273 0.15544 0.8563 south africa 0.94908 0.3914 2.08703 0.1307 h0 : gpr does not granger cause gulf stock market bahraïn 6.51155*** 0.0024 0.16510 0.8481 kuwait 4.03717** 0.0213 0.06123 0.9407 oman 2.32776* 0.0960 0.10609 0.8995 qatar 5.76246*** 0.0046 0.16517 0.8480 ksa 2.87802* 0.0620 0.47725 0.6222 uae 2.32736* 0.0944 0.21362 0.8081 ***. ** and * denote the rejection of the nullhypothesis of granger-non-causality at 1%. 5% and 10% significance level respectively. 4.3 time-varying effects of the gpr index on stock market returns and volatilities we used the mcmc method based on the bayesian framework to estimate the tvp-var model (primiceri 2005, kang et al. 2015, degiannakis et al. 2018 and toparlı et al. 2019). table 4 shows the results of the parameters estimating selected in the tvp-var model. for all the zones considered (g7, brics, golf) and for both returns and volatilities, the mean of the estimated parameters lies within the confidence interval. what's more, the inefficiency factors are relatively low, and the geweke statistics show that the parameter converges to the posterior distribution. we can therefore conclude that the mcmc algorithm efficiently draws the posterior distribution. do geopolitical risks affect stock market returns and volatilities table 4. estimation results of the main parameters in the tvp-var model parameter mean std. dev. 95% confidence interval geweke inef. estimation for the set (gpr, g7 stock market indices return) (σβ)1 0.0023 0.0003 [0.0018, 0.0029] 0.259 6.06 (σβ)2 0.0023 0.0003 [0.0018, 0.0028] 0.216 6.08 (σα)1 0.0047 0.0010 [0.0031, 0.0072] 0.055 19.26 (σh)1 0.0058 0.0018 [0.0034, 0.0102] 0.004 32.37 (σh)2 0.0055 0.0016 [0.0034, 0.0096] 0.692 28.82 estimation for the set (gpr, g7 stock market indices volatility) (σβ)1 0.0023 0.0003 [0.0018, 0.0029] 0.862 6.06 (σβ)2 0.0023 0.0003 [0.0018, 0.0029] 0.695 4.21 (σα)1 0.0055 0.0044 [0.0032, 0.0105] 0.000 20.89 (σh)1 0.0058 0.0021 [0.0034, 0.0110] 0.014 32.73 (σh)2 0.3027 0.1423 [0.0042, 0.4594] 0.000 340.07 estimation for the set (gpr, brics stock market indices return) (σβ)1 0.0023 0.0003 [0.0018, 0.0029] 0.975 4.60 (σβ)2 0.0023 0.0003 [0.0018, 0.0029] 0.786 4.52 (σα)1 0.0046 0.0010 [0.0031, 0.0070] 0.294 14.97 (σh)1 0.0056 0.0016 [0.0034, 0.0099] 0.728 29.56 (σh)2 0.0056 0.0019 [0.0034, 0.0102] 0.611 31.86 estimation for the set (gpr, brics stock market indices volatility) (σβ)1 0.6693 0.2087 [0.4614, 1.2824] 0.000 307.79 (σβ)2 0.0023 0.0003 [0.0018, 0.0029] 0.320 5.15 (σα)1 2.1744 2.1985 [0.0026, 6.8471] 0.000 341.56 (σh)1 0.1476 0.3499 [0.0036, 1.1731] 0.000 322.95 (σh)2 0.0059 0.0046 [0.0033, 0.0112] 0.501 25.21 estimation for the set (gpr, gulf stock market indices return) (σβ)1 0.0102 0.0564 [0.0018, 0.0054] 0.220 167.39 (σβ)2 0.0023 0.0003 [0.0018, 0.0029] 0.370 4.40 (σα)1 0.0046 0.0010 [0.0031, 0.0069] 0.0031 25.37 (σh)1 0.0058 0.0021 [0.0035, 0.0114] 0.166 26.40 (σh)2 0.0056 0.0016 [0.0034, 0.0098] 0.063 32.71 estimation for the set (gpr, gulf stock market indices volatility) (σβ)1 0.0023 0.0003 [0.0018, 0.0029] 0.020 6.85 (σβ)2 0.0023 0.0003 [0.0018, 0.0028] 0.186 5.27 (σα)1 0.0060 0.0141 [0.0033, 0.0072] 0.168 15.71 (σh)1 0.0056 0.0017 [0.0034, 0.0099] 0.299 22.34 (σh)2 0.0071 0.0098 [0.0034, 0.0233] 0.007 36.23 4.3.1 time-varying effects of the gpr index on g7 stock market returns figure 1 shows the responses of stock index returns in the g7 countries to geopolitical risk shocks (gpr index) at different time horizons. we have analyzed impulse responses over the short (1 month), medium (6 months), and long term (12 months). interval analysis of impulse responses enables us to simulate impulse responses more effectively and reveal differences between different terms. in general, responses vary over time. as shown in figure 1, over a 1-month time horizon, geopolitical risks mainly exert positive effects on the returns of virtually all g7 stock market indices. otherwise, a shock to the gpr index mainly triggers short-term increases in returns in the g7 countries. in the case of the usa and the uk, the gpr has a negative effect on returns from mid-2020, resulting in lower returns for the stock market indices of these two countries. in the medium term, however, yield responses in the g7 countries are weak. they are mostly negative, with temporary positive responses. over the 12-month horizon, responses are small and almost non-existent. on the other hand, the largest positive effects are seen in the period between mid-2017 and mid-2020, and the largest negative impacts occur around early 2020 and late 2022. we also find that yield responses 250 ahlem lamine and sirine zribi to geopolitical risk are strongest in the short term, much weaker in the medium-term lag, and negligible in the long term. we can therefore conclude that the effect of the gpr index on the returns of the g7 country indices is seen more in the short term. it is interesting to note that g7 country index returns react to the gpr with little difference over the entire study period and for all three time horizons. our findings are in line with the article by salisu et al. (2022), which illustrates that gpr is an important predictor of stock returns in advanced economies and that their markets are adversely affected by gpr threats (such as threats of war and terrorism). figure 1. time-varying responses of g7 country indices returns to the gpr at different time horizons. note: gpr refers to the gpr benchmark. us, uk, gr, jp, fr, it and ca refer to the returns of the benchmark equity indices in the usa, uk, germany, japan, france, italy and canada, respectively. our results show that the impulse responses of stock index returns following a shock to the gpr index vary over time. to deepen this analysis, we propose to study the dynamic effect of geopolitical risk on specific dates. our study period (2016-2023) is not characterized by significant geopolitical events, and even the evolution of the gpr index registers only one peak in 2022 with the russian invasion of ukraine. we propose three dates: january 2017, january 2020, and february 2022. these three periods correspond to the election of donald trump as president of the united states, the first wave of the health crisis, and the war between russia and ukraine, respectively. as shown in figure 2, g7 return responses are highly volatile, frequently shifting from positive to negative at all three predefined dates. as a result, we can say that the effects of the gpr index on g7 stock market returns have no regular direction during specific geopolitical events. indeed, in the face of unpredictable and significant geopolitical risks, investors' behavior towards stock market investments can go either way, depending on their expectations and risk aversion. some investors choose to buy stocks, and others choose to sell based on their expectations for the future, triggering a corresponding rise or fall in stock returns. in addition, the biggest return reactions were seen during the health crisis and the russian invasion of ukraine. during the health crisis and the war between russia and ukraine, the return responses of the g7 stock indices (except japan) are similar, passing through a negative-positivedo geopolitical risks affect stock market returns and volatilities negative alteration process. in the case of japan, the effect of the gpr index on nikkei returns is always positive during the covid-19 crisis and negative during the russian invasion of ukraine. this result is in line with the study by abbasi et al. (2022), which shows that the stock prices of g7 countries are fragile in the face of war events, creating negative abnormal returns. figure 2. time-varyingresponses of g7 country indices returns to the gpr at differentdates. note: gpr refers to the gpr benchmark.us, uk, gr, jp, fr, it and ca refer to the returns of the benchmark equity indices in the usa, uk, germany, japan, france, italy and canada, respectively. 4.3.2 time-varying effects of the gpr index on g7 stock market volatilities in addition to studying the effect of political risk on stock returns, we also propose to analyze the effect of the gpr index on stock market volatility. as already mentioned, these volatilities have been extracted using garch models. figure 3 shows the impulse responses of g7 stock index return volatilities to short-, mediumand long-term gpr index shocks. as in the case of returns, volatility responses are dynamic and change over time. as shown in figure 3, in the short, medium and long term, geopolitical risks mainly exert positive effects on the volatilities of almost all g7 stock market indices and over the entire study period (except for the uk and italy). for these two countries, we observe mainly positive reactions, with temporary negative reactions for the period mid-2019 and end of 2020. overall, we can conclude that the gpr index has a positive effect on the volatility of stock market indices in the g7 countries, which translates into higher risks for this type of investment. this finding confirms the conclusions of balcilar et al. (2018), bouras et al. (2019) and das et al. (2019), who also found positive gpr effects on stock volatility. we also note that, unlike returns, volatility responses to geopolitical risk are strongest in the long term, much weaker in the medium term and negligible in the short term. we can therefore conclude that the effect of the gpr index on the volatilities of the g7 indices is more pronounced in the long term. there is therefore some persistence in the effect of gpr indice on volatilities. 252 ahlem lamine and sirine zribi figure 3. time-varying responses of g7 country indices volatilities to the gpr at different time horizons. note: gpr refers to the gpr benchmark. usv, ukv, grv, jpv, frv, itv and cav refer to the volatilities of the benchmark equity indices in the usa, uk, germany, japan, france, italy and canada, respectively. as shown in figure 4, the reactions of g7 volatilities are, in the majority of cases, positive on all three predefined dates. consequently, we can say that geopolitical risk increases the risk of investing in g7 stock markets. in addition, the strongest reactions were observed in 2022, corresponding to the russian invasion of ukraine. during the war between russia and ukraine, the volatility of the g7 stock indices (with the exception of italy) reacted largely positively to the gpr index shock. figure 4. time-varying responses of g7 country indices volatilities to the gpr at different dates. note: gpr refers to the gpr benchmark. usv, ukv, grv, jpv, frv, itv and cav refer to the volatilities of the benchmark equity indices in the usa, uk, germany, japan, france, italy and canada, respectively. 4.3.3 time-varying effects of the gpr index on brics stock market returns the impulse responses of brics stock index returns to geopolitical risk shocks (gpr index) at different time horizons are plotted in figure 5. as shown in figure 5, at a one-month horizon, do geopolitical risks affect stock market returns and volatilities geopolitical risks mainly affect returns in different ways across countries. in the case of china, a shock to the gpr index mainly triggers short-term declines in stock index returns. in contrast, the gpr index has a positive impact on the indian stock index. in the case of russia, the gpr has had a negative impact on returns since mid-2020, leading to falling stock index returns for both countries. in the medium term, however, the return responses in the g7 countries are weak. they are mostly negative, with temporary positive reactions. over the 12-month horizon, the reactions are weak and almost non-existent. moreover, the largest positive effects are observed between mid-2017 and mid-2020, and the largest negative effects occur between early 2020 and late 2022. we also find that yield responses to geopolitical risk are strongest in the short term, much weaker in the medium term, and negligible in the long term. we can therefore conclude that the effect of the gpr index on brics country index returns is most pronounced in the short run. it is interesting to note that brics index returns react to the gpr with little difference over the entire period studied and for all three time horizons. figure 5. time-varying responses of brics country indices returns to the gpr at different time horizons. note: gpr refers to the gpr benchmark.ch, ru, in, br, and sa refer to the returns of the benchmark equity indices in china, russia, india, brazil, and south africa, respectively. figure 6 shows that the return reactions of the brics countries are highly volatile, frequently changing from positive to negative on the three predefined dates. as a result, the impact of the gpr index on brics stock returns does not have a regular direction during specific geopolitical events. our results are in line with the study of balcilar et al. (2018), who studied the effect of geopolitical uncertainty on the return and volatility dynamics of brics stock markets and asserted that the geopolitical risks (gpr) effect is heterogeneous on stock markets and does not affect the returns dynamics in these markets in a uniform manner. in fact, the largest return reactions were observed during the russian invasion of ukraine. in this sense, the return reactions were consistently negative. the russian invasion of ukraine negatively affected stock returns in the brics countries. similarly, during the health crisis, the reactions of stock index returns in china, russia, and india were similar. they are sometimes negative and sometimes positive. the effect of the grp index on stock market returns is always negative in the case of brazil and positive in the case of south africa. 254 ahlem lamine and sirine zribi figure 6. time-varying responses of brics country indices returns to the gpr at different dates. note: gpr refers to the gpr benchmark.ch, ru, in, br, and sa refer to the returns of the benchmark equity indices in china, russia, india, brazil, and south africa, respectively. 4.3.4 time-varying effects of the gpr index on brics stock market volatilities as shown in fig. 7, the impulse responses of brics stock index volatilities to shocks linked to geopolitical risks (gpr index) at different time horizons are positive overall, with some temporary negativities (with the exception of china). this finding is similar to the results found in the case of g7 countries. in addition, the strongest positive effects are observed between late-2020 and mid-2022, a period characterized mainly by the health crisis and the russian-ukrainian conflict. in the case of china, volatility reactions following a shock to the gpr index are negative for all horizons studied. we also note that, as in the case of volatilities in the g7 countries, volatility reactions to geopolitical risk are strongest in the long term, much weaker in the medium term and negligible in the short term. we can therefore conclude that the effect of the gpr index on brics country index volatilities is most pronounced in the long term. we can conclude that there is a persistence of gpr index shocks, which explains the greater long-term effect on volatility. as shown in figure 8, the volatility responses of brics stock indices are broadly positive at all three predefined dates. this result confirms the findings of salisu et al. (2022), who assert that the stock market volatility of these countries reacts more positively to geopolitical risks. the effects of geopolitical risk increase the risks of investing in brics stock markets. furthermore, the strongest reactions were observed in 2020, a year characterized primarily by the first wave of the covid-19 pandemic. the greatest impact of this crisis was seen in china. this result seems logical, given that the coronavirus was first detected in wuhan. in addition, during the russian invasion of ukraine, the volatility of the brics stock indices also reacted positively to a shock to the gpr index. the magnitude of this effect, however, was smaller than that of the health crisis. do geopolitical risks affect stock market returns and volatilities figure 7. time-varying responses of brics country indices volatilities to the gpr at different time horizons. note: gpr refers to the gpr benchmark.chv, ruv, inv, brv, and sav refer to the volatilities of the benchmark equity indices in china, russia, india, brazil, and south africa, respectively. figure 8. time-varying responses of brics country indices volatilities to the gpr at different dates. note: gpr refers to the gpr benchmark. chv, ruv, inv, brv, and sav refer to the volatilities of the benchmark equity indices in china, russia, india, brazil, and south africa, respectively. 4.3.5 time-varying effects of the gpr index on gulf stock market returns as shown in figure 9, the impulse responses of gulf yields are heterogeneous across time horizons. at the one-month horizon, the responses are generally positive. thus, when geopolitical events occur, investors in the gulf countries choose to buy rather than sell stocks, leading to higher stock index returns. at the eight-month horizon, the responses are mostly negative. gulf investors tend to sell stocks on the stock exchanges, leading to lower stock index returns. however, at the 12-month horizon, the responses oscillate between positive and negative at low levels. the functional strength of the responses also varies over time. the most significant responses occur in the short term, and the least significant effects occur in the long term. as in the case of the g7 and brics countries, the effects of the gpr index on gulf returns weaken over time. 256 ahlem lamine and sirine zribi figure 9. time-varying responses of gulf country indices returns to the gpr at different time horizons. note: gpr refers to the gpr benchmark.bh, ku, om, qa, ksa, and uae refer to the returns of the benchmark equity indices in the bahrain, kuwait, oman, qatar, ksa, uae, respectively. as in the case of the g7 and brics countries, the feedback from the gulf countries is very volatile, often swinging from positive to negative on the three predefined dates (figure 10). therefore, there is no regular direction for the impact of the gpr index on gulf stock returns during specific geopolitical events. our results do not coincide with the study by alqahtani et al. (2022), which asserts that the time-varying conditional correlation between gulf stock market returns and geopolitical risk is systematically negative. during the health crisis, the reactions of the gulf stock index returns were similar. they are initially negative, then positive, and then close to zero. during the russian-ukrainian conflict, the responses alternated between positive and negative, but the negative responses are more significant and have larger amplitudes. figure 10 also shows that, in contrast to the g7 and brics countries, returns generally reacted positively to donald trump's victory in the us election. figure 10. time-varying responses of gulf country indices returns to the gpr at different dates. note: gpr refers to the gpr benchmark.bh, ku, om, qa, ksa, and uae refer to the returns of the benchmark equity indices in the bahrain, kuwait, oman, qatar, ksa, uae, respectively. do geopolitical risks affect stock market returns and volatilities 4.3.6 time-varying effects of the gpr index on gulf stock market volatilities as shown in figure 11, the impulse responses of return volatilities in the gulf region are very volatile, frequently changing from positive to negative in all three time horizons. indeed, in the face of unpredictable and significant geopolitical risks, investors' behavior with respect to stock market investments may vary according to their expectations and risk aversion. some investors may choose to buy stocks, while others may choose to sell based on their expectations for the future, leading to higher or lower stock market risk. as in the case of the g7 and brics, the gpr index impact on gulf returns diminishes over time. we also find that, as in the case of g7 and brics volatilities, the volatility responses to the gpr index shocks increase with time horizons. they are strongest in the long term, much weaker in the medium term, and negligible in the short term. this more pronounced effect on long-term volatility can be attributed to the persistence of shocks in the gpr index. figure 11. time-varying responses of gulf country indices volatilities to the gpr at different time horizons. note: gpr refers to the gpr benchmark.bhv, kuv, omv, qav, ksav, and uaev refer to the volatilities of the benchmark equity indices in the bahrïn, kuwait, oman, qatar, ksa, uae, respectively. the gulf indices volatility reactions are similar on the three predefined dates (figure 12). during a health crisis, the reactions are first positive and then negative. for the russian-ukrainian conflict, the gpr index has a negative effect on the volatility, then the reaction is reversed. on the other hand, the effect of the gpr index during the 2017 us elections on the stock market volatilities of the gulf countries does not have a regular direction and often goes from positive to negative. 258 ahlem lamine and sirine zribi figure 12. time-varyingresponses of gulf country indices volatilities to the gpr at differentdates. note: gpr refers to the gpr benchmark.bhv, kuv, omv, qav,, ksav, and uaevrefer to thevolatilities of the benchmark equity indices in the bahrïn, kuwait, oman, qatar, ksa,uae, respectively. 5. discussion this paper uses a time-varying parameter vector autoregression (tvp-var) model to examine the impact of the gpr index shock, which measures geopolitical risk, on stock index returns and volatility in g7, brics, and gulf countries at three time horizons and three points in time. the results of this study show that the impact of geopolitical risk on stock index returns in g7 and brics countries is significant but greater in the short run, which is consistent with salisu et al. (20-22). moreover, geopolitical risk has a positive impact on the volatility of g7 and brics stock markets. in other words, investing in these markets becomes riskier during periods characterized by a high gpr index, which is consistent with balcilar et al. (2018), bouras et al. (2019), and das et al. (2019). in contrast to returns, the effect of the gpr index on the volatility of the g7 and brics stock indices was stronger in the long run. thus, there seems to be some persistence in the way the gpr index affects volatility. furthermore, we find that returns and volatility of g7 stock indices are more sensitive to geopolitical risk during the russia-ukraine war, which is consistent with the study by abbasi et al. (2022). for the brics countries, on the other hand, the impact of geopolitical risk was greater during the covid-19 pandemic. as with the g7 and brics stock indices, the impact of geopolitical risk on gulf returns decreases over time, and the volatility response to geopolitical risk shocks is more pronounced in the long run. this more significant effect on long-term volatility is due to the persistence of the shocks. the results of this study will certainly have an impact on financial investment strategies. indeed, the addition of precious metals (especially gold) and cryptocurrencies to a portfolio of stock market indices can provide investors with an advantage in terms of diversification or hedging during geopolitical crises. do geopolitical risks affect stock market returns and volatilities 6. conclusions and policy implications from a global perspective, this paper explores the gpr index shock effects on the stock market indices returns and volatilities in three different country groupings (g7, brics, and gulf countries) at three time horizons and three points in time. this study is based on the tvp-var models. the three-time horizons refer to lags of 1 month (short term), 6 months (medium term), and 12 months (long term), and the three time periods include 2017, 2020, and 2022 (corresponding to donald trump's victory in the us election, the first wave of covid-19 and the russian invasion of ukrania). from this study, we can conclude that, first, the gpr index effects on stock index returns vary over time and differ across countries, different lag periods, and different times. in addition, the effect of the gpr index on the returns of the g7, brics, and gulf indices weakens over time. it is more pronounced in the short run. moreover, stock market returns show a high vulnerability to geopolitical events, in particular, the russian invasion of ukraine. second, our study shows that the gpr index effects on the stock market indices volatility are generally positive for the different countries, lag periods and different dates chosen. geopolitical risks increase the risk of investing in financial markets. we also find that this effect increases with time horizons for virtually all countries. in other words, the main impulse responses to shocks to the gpr index are observed in the long run. we can therefore conclude that there is a persistence of the gpr index shocks. furthermore, it is clear that the russian-ukrainian war has the most significant impact on stock market risk. the empirical findings of this study have policy implications and provide insightful information for portfolio managers and investors. given that stock markets are affected by global geopolitical events such as the russian-ukrainian crisis, we suggest that investors diversify their portfolios by adding stocks that have proven their effectiveness in terms of hedging and risk reduction. these assets can be seen as safe havens, providing investors with hedging and diversification benefits. in that sense, investors turn to these assets as a hedge against risk in times of geopolitical crisis, particularly russia's evasion of ukraine. adding precious metals and cryptocurrencies to a stock portfolio may help diversify during times of crisis. the policy implications of this research are as follows. the establishment of a dynamic warning platform is necessary due to the unpredictability and complexity of crises. furthermore, an information-sharing mechanism should be implemented to mitigate blind decisions due to information asymmetry. this will prevent speculators from taking advantage of the uneven distribution of information and allow investors to receive timely information on changes caused by crises. in addition, encouraging the participation of major investment institutions is crucial to enhancing investor confidence and thereby ensuring financial market stability. to mitigate the consequences of crises such as the covid-19 health crisis or the russian-ukrainian crisis, we also recommend international cooperation. several lines of research can be considered to refine this work. first, we examine the impact of the russian-ukrainian war on stock, bond, and commodity markets using alternative methods such as the quantile connection approach or copulas. in addition, given the growing interest in digital assets, including cryptocurrencies, nfts, and defi, we may examine the impact of geopolitical risk on these digital assets. in this context, it is interesting to compare the diversification benefits offered by commodities with those offered by digital assets. 260 ahlem lamine and sirine zribi references abbassi, w., kumari, v., & pandey, d. k. (2023). what makes firms vulnerable to the russia-ukraine crisis? journal of risk finance, 24, 24-39. https://doi.org/10.1108/jrf-05-2022-0108 alqahtani, a., bouri, e., & vo, x. v. (2020). predictability of gcc stock returns: the role of geopolitical risk and crude oil returns. economic analysis and policy, 68, 239-249. https://doi.org/10.1016/j.eap.2020.09.017 alqahtani, a., hammoudeh, s., & selmi, r. (2022). relationship between different sources of geopolitical risks and stock markets in the gcc region: a dynamic correlation analysis. review of behavioral finance, 14, 296-316. https://doi.org/10.1108/rbf-07-2019-0099 antonakakis, n., & gabauer, d. (2017). refined measures of dynamic connectedness based on tvp-var. mpra paper 78282, university library of munich, germany. antonakakis, n., gupta, r., kollias, c., & papadamou, s. (2017). geopolitical risks and the oil-stock nexus over 1899–2016. finance research letters, 23, 165-173. https://doi.org/10.1016/j.frl.2017.07.017 apergis, e., & apergis, n. (2016). the 11/13 paris terrorist attacks and stock prices: the case of the international defense industry. finance research letters, 17, 186-192. https://doi.org/10.1016/j.frl.2016.03.002 aysan, a. f., demir, e., gozgor, g., & lau, c. k. m. (2019). effects of the geopolitical risks on bitcoin returns and volatility. research in international business and finance, 47, 511-518. https://doi.org/10.1016/j.ribaf.2018.09.011 balcilar, m., bonato, m., demirer, r., & gupta, r. (2018). geopolitical risks and stock market dynamics of the brics. economic systems, 42, 295-306. https://doi.org/10.1016/j.ecosys.2017.05.008 baur, d. g., & smales, l. a. (2020). gold and geopolitical risk. ssrn paper 3109136. https://doi.org/10.2139/ssrn.3109136 będowska-sójka, b., demir, e., & zaremba, a. (2022). hedging geopolitical risks with different asset classes: a focus on the russian invasion of ukraine. finance research letters, 50, 103192. https://doi.org/10.1016/j.frl.2022.103192 boubaker, s., goodell, j. w., pandey, d. k., & kumari, v. (2022). heterogeneous impacts of wars on global equity markets: evidence from the invasion of ukraine. finance research letters, 48, 102934. https://doi.org/10.1016/j.frl.2022.102934 boungou, w., & yatié, a. (2022). the impact of the ukraine-russia war on world stock market returns. economics letters, 215, 110516. https://doi.org/10.1016/j.econlet.2022.110516 bouras, c., christou, c., gupta, r., & suleman, t. (2019). geopolitical risks, returns, and volatility in emerging stock markets: evidence from a panel garch model. emerging markets finance and trade, 55(8), 1841-1856. https://doi.org/10.1080/1540496x.2018.1507906 caldara, d., & iacoviello, m. (2022). measuring geopolitical risk. american economic review, 112(4), 11941225. https://doi.org/10.1257/aer.20191823 corbet, s., gurdgiev, c., & meegan, a. (2018). long-term stock market volatility and the influence of terrorist attacks in europe. the quarterly review of economics and finance, 68, 118-131. https://doi.org/10.1016/j.qref.2017.11.012 cunado, j., gupta, r., lau, c. k. m., & sheng, x. (2020). time-varying impact of geopolitical risks on oil prices. defence and peace economics, 31, 692-706. https://doi.org/10.1080/10242694.2018.1563854 das, d., kannadhasan, m., & bhattacharyya, m. (2019). do the emerging stock markets react to international economic policy uncertainty, geopolitical risk and financial stress alike? the north american journal of economics and finance, 48, 1-19. https://doi.org/10.1016/j.najef.2019.01.008 degiannakis, s., & filis, g. (2017). forecasting oil price realized volatility using information channels from other asset classes. journal of international money and finance, 76, 28-49. https://doi.org/10.1016/j.jimonfin.2017.05.006 drakos, k. (2010). terrorism activity, investor sentiment, and stock returns. review of financial economics, 19, 128-135. https://doi.org/10.1016/j.rfe.2010.01.001 gkillas, k., gupta, r., & wohar, m. e. (2018). volatility jumps: the role of geopolitical risks. finance research letters, 27, 247-258. https://doi.org/10.1016/j.frl.2018.03.014 hudson, r., & urquhart, a. (2015). war and stock markets: the effect of world war two on the british stock market. international review of financial analysis, 40, 166-177. https://doi.org/10.1016/j.irfa.2015.05.015 jeribi, a., fakhfekh, m., & jarboui, a. (2015). tunisian revolution and stock market volatility: evidence from fiegarch model. managerial finance, 41(10), 1112-1135. https://doi.org/10.1108/mf-12-2014-0310 https://doi.org/10.1108/jrf-05-2022-0108 https://doi.org/10.1016/j.eap.2020.09.017 https://doi.org/10.1108/rbf-07-2019-0099 https://doi.org/10.1016/j.frl.2017.07.017 https://doi.org/10.1016/j.frl.2016.03.002 https://doi.org/10.1016/j.ribaf.2018.09.011 https://doi.org/10.1016/j.ecosys.2017.05.008 https://doi.org/10.1016/j.frl.2022.103192 https://doi.org/10.1016/j.frl.2022.102934 https://doi.org/10.1016/j.econlet.2022.110516 https://doi.org/10.1080/1540496x.2018.1507906 https://doi.org/10.1257/aer.20191823 https://doi.org/10.1016/j.qref.2017.11.012 https://doi.org/10.1080/10242694.2018.1563854 https://doi.org/10.1016/j.najef.2019.01.008 https://doi.org/10.1016/j.jimonfin.2017.05.006 https://doi.org/10.1016/j.rfe.2010.01.001 https://doi.org/10.1016/j.frl.2018.03.014 https://doi.org/10.1016/j.irfa.2015.05.015 https://doi.org/10.1108/mf-12-2014-0310 do geopolitical risks affect stock market returns and volatilities kang, w., ratti, r. a., & yoon, k. h. (2015). time-varying effect of oil market shocks on the stock market. journal of banking & finance, 61, 150-163. https://doi.org/10.1016/j.jbankfin.2015.08.027 kumari, v., kumar, g., & pandey, d. k. (2023). are the european union stock markets vulnerable to the russia-ukraine war? journal of behavioral and experimental finance, 37, 100793. https://doi.org/10.1016/j.jbef.2023.100793 ndako, u. b., salisu, a. a., & ogunsiji, m. o. (2021). geopolitical risk and the return volatility of islamic stocks in indonesia and malaysia: a garch-midas approach. asian economics letters, 2(3). https://doi.org/10.46557/001c.24843 omar, a. m., wisniewski, t. p., & nolte, s. (2017). diversifying away the risk of war and cross-border political crisis. energy economics, 64, 494-510. https://doi.org/10.1016/j.eneco.2016.02.015 papakyriakou, p., sakkas, a., & taoushianis, z. (2019). the impact of terrorist attacks in g7 countries on international stock markets and the role of investor sentiment. journal of international financial markets, institutions and money, 61, 143-160. https://doi.org/10.1016/j.intfin.2019.03.001 plakandaras, v., gupta, r., & wong, w. k. (2019). point and density forecasts of oil returns: the role of geopolitical risks. resources policy, 62, 580-587. https://doi.org/10.1016/j.resourpol.2018.11.006 primiceri, g. e. (2005). time varying structural vector autoregressions and monetary policy. the review of economic studies, 72, 821-852. https://doi.org/10.1111/j.1467-937x.2005.00353.x salisu, a. a., ogbonna, a. e., lasisi, l., & olaniran, a. (2022). geopolitical risk and stock market volatility in emerging markets: a garch-midas approach. north american journal of economics and finance, 62, 101755. https://doi.org/10.1016/j.najef.2022.101755 toparlı, e. a., çatık, a. n., & balcılar, m. (2019). the impact of oil prices on the stock returns in turkey: a tvp-var approach. physica a: statistical mechanics and its applications, 535, 122392. https://doi.org/10.1016/j.physa.2019.122392 umar, z., polat, o., choi, s. y., & teplova, t. (2022). the impact of the russia-ukraine conflict on the connectedness of financial markets. finance research letters, 48, 102976. https://doi.org/10.1016/j.frl.2022.102976 wang, y., & young, m. (2020). terrorist attacks and investor risk preference: evidence from mutual fund flows. journal of financial economics, 137, 491-514. https://doi.org/10.1016/j.jfineco.2020.02.008 zhang, y., he, j., he, m., & li, s. (2023). geopolitical risk and stock market volatility: a global perspective. finance research letters, 53, 103620. https://doi.org/10.1016/j.frl.2022.103620 https://doi.org/10.1016/j.jbankfin.2015.08.027 https://doi.org/10.1016/j.jbef.2023.100793 https://doi.org/10.46557/001c.24843 https://doi.org/10.1016/j.eneco.2016.02.015 https://doi.org/10.1016/j.intfin.2019.03.001 https://doi.org/10.1016/j.resourpol.2018.11.006 https://doi.org/10.1111/j.1467-937x.2005.00353.x https://doi.org/10.1016/j.najef.2022.101755 https://doi.org/10.1016/j.physa.2019.122392 https://doi.org/10.1016/j.frl.2022.102976 https://doi.org/10.1016/j.jfineco.2020.02.008 https://doi.org/10.1016/j.frl.2022.103620 1. introduction 2. literature review 3. data and methodology 4. empirical results and discussion 5. discussion 6. conclusions and policy implications references © the author(s) 2024 this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 1 (2024), pages 85-102 https://doi.org/10.17979/ejge.2024.13.1.9788 submitted: june 21, 2023 accepted: december 2, 2023 published: june 6, 2024 article tourism, growth, and carbon emissions in sub-saharan africa: a balancing act merith ifeoma anaba,1,* jayanthi r. alaganthiran,2 kafilah lola gold,3 folorunso obayemi tamitope obasuyi 4 1 university of malaya, malaysia; veritas university abuja, nigeria. 2 university of malaya; cabi, p.o.box 210, 43400 upm serdang, selangor, malaysia. 3 dsi/nrf south african research chair in industrial development, university of johannesburg, south africa; kwara state college of education, ilorin, nigeria 4 bamidele olumilua university of education science and technology, ekiti, nigeria. *correspondence: merithifeoma@yahoo.com abstract. tourism is one of the major determinants of global economic growth, creating jobs within the sector, and africa is no exception. the target of the sub-saharan african (ssa) countries is to consider tourism as an alternative means of economic expansion. however, tourism is a means of environmental imbalance. this study investigates the complex relationship between tourism, economic growth, and carbon emissions in 47 sub-saharan african countries from 2005 to 2020. while economic growth significantly increases carbon emissions, tourism revenue shows a potential mitigating effect. trade openness also contributes to emissions, while employment shows a negative correlation. these findings highlight the need for stricter environmental regulations and policies that leverage the region's labor surplus for sustainable tourism practices. implementing such measures is crucial for minimizing the environmental damage associated with foreign direct economic activities and ensuring long-term sustainability. keywords: tourism revenues; carbon dioxide emissions; economic growth; trade openness; sub-saharan africa jel classification : z330, q540, o4, f100, o55 1. introduction tourism affects the economy in three major ways, namely, its contribution to growth, leisure and carbon emissions (wijesekara et al., 2022; du, lew, & ng, 2016; haller et al., 2021). regarding growth, tourism destinations increase huge revenue through investment in several sectors of the industry. also, globally, evidence shows that the tourism industry remains one of the most important contributors of carbon dioxide emissions to the environment (chen, thapa, & yan, 2018; jong, soh, & puah, 2022). this has been due to the normative effects that tourism usually brings to the environment. predominantly, given the tourism industry's various sectors, including hotel and https://creativecommons.org/licenses/by-nc/4.0/ 86 anaba et al. lodging, aviation and airline, tourist attractions, and food and drink, the tourism industry is known to be extremely energy-dependent (adedoyin, et al., 2021). therefore, tourism activities are known for their high cause of climate change because of environmental degradation and a negative rise in energy consumption (gyamfi, et al., 2021a; kyara et al., 2022). tourism has continued to have direct contributions to the growth of the world economies through tourist arrivals and receipt which could have effect on the physical environment. in addition, tourism contributes to economic growth and reduces poverty (kyara et al., 2022) however, the complexity of the relationship between environmental sustainability and tourism has continued to be explored in diverse ways among scholars (pal & mitra, 2017). africans have been facing excessive carbon emissions, which are scientifically and empirically proven to contaminate the atmosphere and harm human beings and the natural environment in sub-sahara africa (ssa). one of the sustainable development goals (sdg’s) is focused on climate change. thus, there is a need to research and identify numerous solutions to combat climate change issues effectively among ssa countries. across countries, carbon dioxide emissions are continuously causing externalities issues to human existence and environmental inequality. for example, previous studies confirmed that air pollution accounted for 1.1 million deaths across african countries in 2019 (fisher et al., 2021) and approximately 7 million premature deaths in 2017 in eastern africa (wipfli et al., 2021) as well as bronchial asthma increased due to air pollution among ssa countries (ku et al., 2021). it has been empirically established that the african continent is suffering because of contaminated air (fisher, et al., 2021). the particulate matter (pm) 2.5 was highly recorded in kampala and uganda (awokola et al., 2020) and the dispersion modelling and spatial analysis confirmed that pm10 concentrations were higher during the day and eventually distributed to wide areas at night (tshehla & wright, 2019). in particular, economic indicators have an impact on carbon dioxide emissions. in africa alone, empirical research has confirmed that foreign direct investment, economic growth, tourism and governance positively increased carbon dioxide emissions (agyeman, et al., 2022). similarly, findings by djellouli et al. (2022) reveal that economic growth and foreign direct investment among twenty ssa countries determined carbon dioxide emissions between 2000 and 2015. besides, low socioeconomic status caused women to be exposed to poor air quality in adama, ethiopia (flanagan et al., 2022). furthermore, al-mulali & binti che sab (2012) contend that energy consumption in the thirty ssa economies played a significant role in boosting economic growth and financial development, but at the cost of excessive pollution. along with various other factors, the level of income generally correlates with an increase in co2 emissions. delving into these previous studies, however, as a matter of importance, these existing findings on the relationships between co2 emissions and economic growth in the ssa countries, have rarely produced any study on tourism's impact on economic expansion in the region. simply, examining the relationships between tourism revenue, economic growth and carbon dioxide is uncommon among scholars despite the gains of growth as well as contributions of tourism to growth. thus, this study examines the impact of tourism revenue and economic growth on carbon dioxide emissions in the 47 ssa countries. specifically, the study examines the linear associations among tourism arrivals, receipts, environmental pollution and economic growth in the ssa countries’ tourism industry. the study estimates the correlation between carbon dioxide and confounding tourism revenue, economic growth and carbon dioxide emissions nexus variables, including gdp, ec, tourism revenue, trade openness and labour surplus. from the empirical findings of this study, the selected ssa countries’ economic growth significantly contributed to carbon dioxide emissions. moreover, african countries’ tourism revenue shows a potential indication that this sector could reduce carbon dioxide emissions. the rest of this study are structured as follows: the first section reviews previous research on the topic that is both current and pertinent; section three discusses empirical technique; section four gives the findings and discussion; and section five closes the study. 2. literature review 2.1. tourism and economic growth in most nations globally, including africa, the tourism sector is without a doubt one of the energyintensive sectors that significantly boosts national gdp and creates jobs. to strengthen an economy, many developing nations, especially those in ssa, are increasingly focusing on the growth of tourism (el menyari, 2021; adedoyin & bekun, 2020). kyara et al. (2022) used time series data from 1995 to 2017 to examine the environmental impacts of tourism growth in tanzania. the study employs autoregressive distributed lag bounds testing, vector error correction model (vecm), and granger causality test for analysis and the wild bootstrap approach to check the accuracy of the computed statistics. hence, the vecm granger causality test indicates that environmental degradation in tanzania is compacted by foreign tourist arrivals and trade openness, while it is accelerated by urbanization and primary energy use. furthermore, although the variables have long-term cointegration, the environmental kuznets curve hypothesis was not ascertained in tanzania. su et al. (2023) study examined the environmental effect of financial stability in iceland from 1995 to 2019. using the nonlinear ardl and fourier-based techniques. the results of nonlinear bound tests and fourier-based approaches show that co2 emissions and financial stability frequently cointegrate. the nardl results demonstrated that a positive variation in financial stability reduces co2 emissions, whereas a negative variation has no effect. additionally, positive income variation causes co2 emissions, whereas negative income variation has no influence on co2 emissions. conversely, a decrease in trade openness has increasing effects on co2 emissions, whereas an increase in co2 emission-mitigation effects is a positive development. also, sun et al. (2022) reviewed 81 environmental kuznets curve research published between 2013 and 2021 to determine whether tourism has an impact on carbon emissions and its consequences on development plans. however, none of the researchers examined international aviation emissions. nonetheless, the results show that there is a paradoxical relationship between tourism and emissions, with divergent findings reported across nations, income levels, and the sector's economic significance. indicating the need to critically re-evaluate the ways in which tourism and carbon are related, as well as the techniques employed in empirical research. the world travel and tourism council (2019) confirm that the tourism industry provides 330 million job opportunities for the world generally and increases the world gross domestic product (gdp) by us$8.9 trillion, representing 10.3% of the world (gdp). this has shown that tourism creates jobs and eventually increases the economic growth of such nations. according to the european union 88 anaba et al. (eu, 2012), tourism plays an important role in most sectors of the economy, including the creation of jobs and a source of economic development. also, it contributes positively to a country’s balance of payment. an increase in both international and domestic tourism arrivals has boosted the country’s revenue, which has indirectly led to growth in the energy consumption sector (dogru & bulut, 2018). for instance, the channel of growth has been by increasing tourism activities such as transportation facilities and hotel stays. however, if a country is an export industry, tourism creates export revenues and contributes to economic growth. furthermore, a country can experience an economic growth process mainly through tourism activities (nyasha et al., 2021). in comparison, china and turkey have experienced tourism-led growth over time, while russia and spain experienced growth-led tourism (ongan, et al., 2017). 2.2 tourism and environmental degradation in light of the aforementioned, adedoyin and bekun (2020) study indicated that, as of 2019, the tourist sector had surpassed the building sector to rank among the top environmental polluters in the world, accounting for 8% of global co2 emissions. however, the transportation sector, like air transportation, contributes significantly to the increase in energy consumption and emissions. the study by isik, et al. (2020) found that tourism has a negative impact on the environment in greece. suess et al. (2020) found that tourism has both negative and positive impacts on emissions in different countries. using the world bank data indicators database from 1990 to 2016, bekun (2022) investigates the impact of economic growth, investment in the energy sector, non-renewable energy, renewable energy, and renewable energy on co2 emissions in india. the study employs fully modified least squares (fmols), dynamic least squares (dols), and canonical cointegration regression (ccr) techniques, and the empirical analysis indicates that a positive relationship exists between co2 emissions and non-renewable and gdp growth. however, a negative relationship exists between co2 emissions and renewable energy. for the granger analysis, the findings show a oneway causality among renewable energy and co2 emissions, economic development, and energy investment. using a different technique and sector-specific findings of bekun et al. (2019) from the pooled mean group-autoregressive auto regressive distributive lag model (pmg-ardl estimations show that “overdependence on natural resource rent affects environmental sustainability if conservation and management options are ignored”. this is also reflected in the non-renewable energy consumption and economic growth, both of which contribute to carbon dioxide emissions in the panel countries. kiracı and bakır (2019) used panel data covering low, lower-middle, upper-middleand high-income countries from 1995-2015. a fully modified ordinary least square (fmols) was used, and it was found that tourism and corruption are the main contributors to c02 emissions. however, the contributions of the c02 emissions have more effect in the lesser-income nations than in highincome nations. furthermore, anser et al. (2020) utilize dynamic gmm and granger causality estimate with panel data spanning from 1665-2018 covering 132 countries comprising algeria, albania, angola, benin, belgium, chile, canada, egypt, ecuador, georgia, ghana and iraq etc. the tourism revenue, economic growth and carbon dioxide emissions nexus result shows that the cost incurred on c02 emission decreases inbound tourism and international tourist receipts. in other words, the c02 emission translates to increasing international tourist expenditure in these selected countries. in nyashay et al. (2020), panel data covering ssa from 20022018 was used to explore tourism and economic growth using the generalized method of moments (gmm). the findings reveal that tourism expenditure has a negative effect on economic growth, while tourism receipt has a positive effect on economic growth. while tourism receipts are robust in lowincome countries, tourism expenditures are robust in the middle-income sub-sample countries. also, using panel smooth transition regression (pstr), as recommended by nosheen et al. (2021), discover the link between tourism, growth, and environmental degradation, which is important in the current era. the study demonstrates that when tourism development declines, environmental deterioration rises. second, as tourism grows, the environment is not being harmed as much. however, as the populace increases, the environment degenerates. while urbanisation has a temporary and large positive impact on environmental degradation, it was later changed to negative by using panel data from 1995-2017 covering 20 countries austria, canada, china, france, germany, greece, hong-kong, italy, japan, malaysia, mexico, netherland, poland, portugal, russian, federation spain, thailand, turkey, and usa. 2.3. tourism and economic indicators it is imperative to understand that foreign direct investment (fdi) in tourism-hosting countries usually affects tourism development. hence, by testing for long and short-run effects and examining the environmental kuznets curve using time series data spanning from 1971 through 2012 in india and china, pal and mitra (2017) establish a long-run effect on economic activity and trade openness. also, the short-run effect of energy use on c02 emission was discovered to be positive. furthermore, instead of the u-shaped expected, the result placed emphasis on the n-shaped relationship between c02 emission and per capita gdp. they argued in favour of the sampled countries that, as per capita gdp increases, the c02 emission also increases, but co2 emission decreases as per capita gdp reaches a certain level. yusuf et al. (2023) examined australia’s energy use and its relatedness with trade liberalisation, co2 emission, gdp and industrialisation using autoregressive distributed lag (ardl) technique and a vector error correction model (vecm). the findings indicate that energy use and gdp positively and significantly affect co2 emissions. however, trade liberalization has a significantly adverse influence on emissions. likewise, the relationship between industrialisation and co2 emissions is insignificant but positive in australia. bekun et al. (2022) examine the function of international tourism influx on e7 countries, an ekc environment economy. its nexus on income, trade, and institutional quality on co2 emission is also assessed from 1995 to 2016. the study utilised data from the world bank development indicators database and employs second generational panel estimator with the driscoll-kraay robust estimator to analyse the data. the findings indicate that non-renewable energy and per capita gdp diminish the environment quality. as well, co2 emissions in e7 economies is due to an increase in non-renewable energy and tourism influx. the quality of institutions improves the quality of the environment. qin et al. (2023) employ 90 anaba et al. the time-varying parameter-stochastic volatility-vector auto-regression (tvp-sv-var) model to examine the interactions among blockchain market (bcm), green finance (gf) and carbon neutrality in china (cnp). the study, from june 2017 to august 2022, used weekly time series data to achieve this objective. the results show that blockchain market development creates long-term inducement that has both positive and negative effects on carbon neutrality in china. green finance development can continually enhance carbon neutrality, though not at the same pace as the blockchain market. likewise, green finance has a positive influence on bcm, as compared to bcm on gf. in shaheen et al. (2019) and ozpolat et al. (2021), both studies discovered a link between rising fdi as a result of tourism, energy use, and co2 emissions. the authors reaffirmed that there is a convincing relationship between the growth of foreign tourism and an increase in energy consumption, which has a direct detrimental outcome on the environment. in a similar study, hanif (2018) examines the relationship between environmental degradation and gdp in sub-saharan africa using gdp, consumption of fossil fuels, sustainable energy, and carbon emissions. it was found that south africa, madagascar, nigeria, mauritius, ghana, uganda, and cameroon are among the topranked developing nations of sub-saharan africa, increasing environmental pollution, co2 and ghg emissions. this correlates with the world health organization (2019) position that a half million people die each year in sub-saharan africa as a result of this increase in environmental pollution. this accounts for the increase in serious biological damage. evidence also revealed a mixed relationship between co2 and other environmental pollutants, energy use, gdp, and foreign direct investment (bataka, 2020). furthermore, considering the importance of co2 emissions on economic growth, particularly in african regions, kyara et al. (2022) used time series data from 1995 to 2017 to examine the environmental impacts of tourism growth in tanzania. however, as important as their study is, the scope is limited to a single country in the african region solely. other recent studies of sun et al. (2022), su et al. (2023), su et al. (2022), qin et al. (2023), bekun et al. (2022), yusuf et al. (2023), and bekun (2022) emphasise the importance of co2 emissions, trade liberalisation, technology innovation, blockchain market, industrialisation and energy usage on economic growth and tourism in other regions and countries, except on the tourism's impact on economic expansion in african region. also, bataka (2021) applies the panel specification with the estimation approach by hoechle. it is used to account for auto-correlation spatial dependency and heteroscedasticity. plane data were used to cover the sub-saharan african countries from 1980 to 2017, and c02 emission data was sourced from the emission database of the global atmospheric research base (cegar). however, the impact of energy consumption on tourism cannot be underestimated because the policy perspective clarifies the significant effect of energy consumption on economic growth, because it serves as a preliminary stage for industrial society. it provides facilities for household consumption, industrial production, resource mining, and transportation sectors, which have shown that economic growth and development cannot be achievable without the proper significant of energy. hence, this study aims to look into the following objectives: 1. to examine whether economic expansion in the ssa countries contributes to carbon dioxide emissions 2. to investigate whether the tourism industry’s revenue contributes to the carbon dioxide emissions in the ssa countries.3. to analyse the extent of tourism inducement on environmental pollution in the ssa countries. 4. to determine if labour-intensive technique through employment generation contributes to the rising of tourism revenue, economic growth and carbon dioxide emissions nexus greenhouse emissions in the ssa countries. the following section of this paper discusses and explains the empirical techniques and econometrics used in this investigation's analysis and presentation. therefore, it is deemed crucial to comprehend how tourism revenue and economic growth contribute to carbon dioxide emissions in sub-saharan african countries. thus, within the context of the ssa economies, this study provides a clearer empirical analysis of the topic. 3. methodology 3.1 data and source this study explores the contributions of tourism revenue and economic growth impacted on carbon dioxide in the 47 ssa countries (see appendix of the list of countries used in the study). in other words, the study examines the nexus between greenhouse gas emissions and economic growth between 2009 and 2020 using a panel data set. the study used co2 emissions, trade openness, international tourism revenue, gdp, tourism receipts, fossil fuels, electricity generation and labour force in ssa countries. the labour force was selected because there is labour surplus in the ssa countries that could be used instead of capital-intensive equipment capable of generating emissions. the carbon dioxide emission in this study represents greenhouse gas emission, and it measures thousands of tons. in detail, carbon dioxide emissions are derived from burning fossil fuels and consumption of solid, liquid and gas fuels and gas flaring. the study focused on the economic growth variable, gdp, which is measured in billions of u.s. dollars. moreover, the study introduced regressors, namely energy consumption, tourism revenue, trade openness, and labour. hence, the data were downloaded from the global economy website on the stated indicators for 47 ssa countries. the list of the 47 sampled countries of the ssa is in appendix i. table 1 describes the variables’ symbols, measurements and expected signs. table 3. operationalization of factors variables description measurement expected sign dependent variable co2e carbon dioxide emissions thousands of tonnes kt independent variables: gdp gross domestic product billions of u.s. dollars + ec fossil fuels electricity generation billion kilowatt hours + torismr international tourism revenue million usd tradeo trade openness percent + labour labour force million people + 92 anaba et al. furthermore, previous studies have shown neoclassical growth theory, which confirmed economic growth association with environmental degradation, namely air pollution and water pollution (gao, et al., 2021). also, econometric analysis employs the cobb-douglas production function to examine the effect of economic growth on carbon dioxide emissions (chaabouni & saidi, 2017; gao, et al., 2021). thus, this study replicated the above similar theoretical framework to examine economic growth and carbon dioxide emissions. again, the prior study focused on only the top ten tourism countries (destek & aydın, 2022). although the tourism sector has contributed to economic growth, it is a factor that causes environmental degradation through many channels. unlike other studies, this study classified ssa countries into regions, namely eastern africa, middle africa, north africa, south africa and western africa. this is to cover the gap in ohajionu et al. (2022) who urged future researchers to examine potential variables that might cause environmental degradation in ssa countries. model estimation this section explains various econometrics estimation techniques to examine the relationships between the tourism industry and greenhouse emissions. there are three level estimations in this study. first, the study determines the correlation between the examined variables to understand the extent to which each variable affects one another. also, the basic correlation analysis provides some basic hints on variable association regardless of the dependent and independent variables' status. second, the study analysed pooled ordering least square (pols) which allows small variation in estimation that obeys best linear unbiased estimation. in this instance, we assumed that there are no unobservable entity-specific effects. simply, we proposed that all the data set of member countries of ssa had the same underlying characteristics during the period. finally, the study employed least square dummy variable (lsdv), with the assumption that each region in the ssa countries could control its carbon dioxide emission levels differently. correlation analysis this study performed pairwise correlation analysis to determine the correlation coefficient range among variables and identify coefficient signs. pooled ols the study applies the pooled panel ordinary least square (ols) to measure the overall influence of economic growth on carbon dioxide emissions without controlling for countries' heterogeneity. the study considered the poole ols, considering that all the data set had the same underlying characteristics during the period. the pooled ols model is specified in equation [1]. 𝐶𝐶𝐶𝐶2𝐸𝐸𝑖𝑖𝑖𝑖 = 𝜙𝜙𝑖𝑖 + 𝜙𝜙1𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝜙𝜙2𝐸𝐸𝐶𝐶𝑖𝑖𝑖𝑖 + 𝜙𝜙3𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 + 𝜙𝜙4𝑇𝑇𝐶𝐶𝑖𝑖𝑖𝑖 + 𝜙𝜙5𝐿𝐿𝑖𝑖𝑖𝑖 + ɛ𝑖𝑖𝑖𝑖 [1] tourism revenue, economic growth and carbon dioxide emissions nexus in model 1, 𝐶𝐶𝐶𝐶2𝐸𝐸𝑖𝑖𝑖𝑖 is the dependent variable and 𝜙𝜙𝑖𝑖is the intercept, 𝜙𝜙denotes slope coefficients and independent variables are 𝐺𝐺𝐺𝐺𝐺𝐺 represents gross domestic product, 𝐸𝐸𝐶𝐶 denotes energy consumption, 𝑇𝑇𝑇𝑇 refers tourism revenue, 𝑇𝑇𝐶𝐶 describes trade openness and 𝐿𝐿is labor. on the other hand, pooled ols estimation subjects biased and inefficient outcomes. finally, the study employed the least square dummy variable (lsdv). we assumed that each region in ssa could be determining carbon dioxide emission levels differently. also, different ssa region groups might have different patterns regarding economic growth, energy consumption, and tourism revenue generation. this section is dedicated to formulating lsdv estimation methodology that pooled 47 ssa countries data information from qualitative variables in econometric model. the model contains dummy variables to measure qualitative influence by coding the different possible outcomes with continuous variables. the dummy variable has dichotomized the possible outcomes and assigned the values of 0 and 1. thus, ssa regions have been coded as 𝐺𝐺𝐷𝐷𝐷𝐷𝑖𝑖𝑖𝑖= 1 for middle africa, 0 otherwise; 𝐺𝐺𝐷𝐷𝐷𝐷𝑖𝑖𝑖𝑖= 1 for northern africa, 0 otherwise; 𝐺𝐺𝐷𝐷𝐷𝐷𝑖𝑖𝑖𝑖= 1 for northern africa, 0 otherwise.𝐺𝐺2012𝑖𝑖𝑖𝑖= 1 for 2012, 0 otherwise and 𝐺𝐺2018𝑖𝑖𝑖𝑖= 1 2018, 0 otherwise. study has derived regression model that included dummy variables as below 𝐶𝐶𝐶𝐶2𝐸𝐸𝑖𝑖𝑖𝑖 = 𝜙𝜙𝑖𝑖 + 𝜙𝜙1𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝜙𝜙2𝐸𝐸𝐶𝐶𝑖𝑖𝑖𝑖 + 𝜙𝜙3𝑇𝑇𝑇𝑇𝑖𝑖𝑖𝑖 + 𝜙𝜙4𝑇𝑇𝐶𝐶𝑖𝑖𝑖𝑖 + 𝜙𝜙5𝐿𝐿𝑖𝑖𝑖𝑖 + 𝜙𝜙6𝐺𝐺𝐷𝐷𝐷𝐷𝑖𝑖𝑖𝑖 + 𝜙𝜙7𝐺𝐺𝐷𝐷𝐷𝐷𝑖𝑖𝑖𝑖 + 𝜙𝜙8𝐺𝐺𝐷𝐷𝐷𝐷𝑖𝑖𝑖𝑖+ 𝜙𝜙9𝐺𝐺𝑊𝑊𝐷𝐷𝑖𝑖𝑖𝑖 + 𝜙𝜙10𝐺𝐺2012𝑖𝑖𝑖𝑖 + 𝜙𝜙11𝐺𝐺2018𝑖𝑖𝑖𝑖 + ɛ𝑖𝑖𝑖𝑖 [2] the descriptive statistics of the data collected is presented in table 2. table 2. descriptive statistics 4. results and discussions the results and discussions section is divided into three sub-sections, namely, the pairwise correlation coefficient and pooled ols and lsdv estimations. variable obs mean std. dev. min max co2 658 1.002 1.907 .02 11.68 gdp 735 31.342 74.52 .14 546.68 ec 713 6.316 32.421 0 233.05 tourismr 551 4.455 7.292 0 42.18 tradeop 683 73.996 37.632 9.96 311.35 labour 736 7.727 10.849 .05 63.23 the descriptive statistics show that the data used in this study is unbalanced. this was accommodated by the software used. 94 anaba et al. 4.1 the pairwise correlation coefficient this sub-section presents the results of the pairwise correlation coefficient and the p-value to understand the significant level of the relationship. the results id presented in table 3. table 3. pairwise correlations variables (1) (2) (3) (4) (5) (6) (1) co2 1.000 (2) gdp 0.306* 1.000 (0.000) (3) ec 0.534* 0.683* 1.000 (0.000) (0.000) (4) tourismr 0.464* -0.154* -0.048 1.000 (0.000) (0.000) (0.259) (5) tradeop 0.401* -0.215* -0.078* 0.591* 1.000 (0.000) (0.000) (0.046) (0.000) (6) labour -0.042 0.702* 0.245* -0.210* -0.366* 1.000 (0.286) (0.000) (0.000) (0.000) (0.000) * significant at p <0.05. we analysed pairwise correlation analysis to identify the strength of the partial relationship between carbon dioxide emissions, economic growth, energy consumption, tourism revenue, trade openness and labour. table 3 presents pairwise correlation results. thus, we have possible and negative correlations across variables. for example, energy consumption has shown a positive and substantial correlation with carbon dioxide emissions and gdp because ec coefficients lie between 0.50 to 0.69. besides, tourism revenue has shown a positive coefficient at 0.464 and a moderate correlation with carbon dioxide emissions. however, tourism revenue has indicated negligible and negative correlation with gdp as well as insignificant correlation with energy consumption. 4.2 pooled ols results and discussion table 4 presents pooled ols model results that comprised 47 ssa countries. in detail, pooled ols estimation has shown five statistically significant associations among economic growth, energy consumption, tourism revenue, trade openness and labour. although the results have passed diagnostic tests, the model suffered from some limitations. we considered the limitations to be insufficient to explain the carbon dioxide emissions phenomenon without compromising the heterogeneity of the countries. hence, the finding shows that a 1% increase in gdp increases carbon dioxide emissions by about 0.004%. basically, growth often comes with intensive use of capital, where electricity and fossil fuels are considerably used. earlier, we hypothesized that economic growth will increase carbon dioxide emissions in ssa countries, as validated in previous findings. the positive association is in line with the findings of jiang et al. (2022), liu et al. (2022) and ouyang et al. (2019). the study has supported prior literature and neoclassical growth theory that decomposed economic growth that consisted of additional factors that could reduce the tourism revenue, economic growth and carbon dioxide emissions nexus environmental quality in ssa countries. however, prior literature lists have always concentrated on non-linear estimation that emphasized ekc theory in countries such as china and oecd countries. furthermore, researchers have focused on the causality between air pollution indicators such as pm 2.5 and gdp. this study has added to existing literature that 47 ssa countries air quality was reduced because of their lower economic activities. the fact that developing countries have fewer industrial activities compared with developed countries, their little economic activities have always neglected environmental quality, thereby breaching environmental law. as such, the ssa countries economic activities, legally or illegally, are expected to emit carbon dioxide that affects the environment. table 4. linear regression: pooled ols co2 coef. st.err. t-val p-val [95% conf interval] sig gdp 0.004 0.001 4.91 0.000 0.002 0.005 * ec 0.03 0.001 25.23 0.000 0.028 0.033 * tourismr 0.03 0.007 4.61 0.000 0.017 0.043 * tradeop 0.008 0.001 7.66 0.000 0.006 0.01 * labour -0.03 0.004 -7.04 0.000 -0.039 -0.022 * constant 0.086 0.083 1.03 .305 -0.078 0.249 diagnostic tests: heteroscedasticity 12.11 multicollinearity 2.40 normality pass mean dependent var 0.828 sd dependent var 1.454 r-squared 0.809 number of obs 481 f-test 402.993 prob > f 0.000 akaike crit. (aic) 939.076 bayesian crit. (bic) 964.132 *significant at p<.05 the ssa african countries energy consumption reduced air quality. again, the results in table 4.2 showed that a 1% increase in fossil fuel consumption in the sampled ssa countries increased carbon dioxide emissions by 0.03%. in other words, the expected signs and findings show that fossil fuel usage deteriorates environmental quality (chien, et al., 2021; khan, et al., 2016). again, and in no small measure, energy used by industries and households will emit harmful gases into the air, thereby reducing air quality. to this end, we introduced another important variable, i.e. electricity consumption, which used fossil fuel to generate power supply. the finding reveals that the higher the electricity consumption, the higher the air pollution. further, on the assumption that cross-country trade activities will cause greenhouse gas emissions, we found that a 1% increase in trade openness increases carbon dioxide emissions by about 0.008%. although the contribution of trade openness to greenhouse emissions is extremely reduced, the positive relationship between trade openness and co2 is consistent with previous findings (kukla-gryz, 2009; lin, et al., 2014). most importantly, the ssa low energy consumption contributes more to carbon dioxide emissions (agyeman, et al., 2022), as reflected in the developed countries, is accountable to the low carbon dioxide emissions in the sampled countries of ssa. finally, there are two positions regarding the estimated labour variable. first, labour variable was introduced to understand that labour-intensive production might be better than using 96 anaba et al. capital-intensive techniques. the more equipment used, the higher the emissions. whereas using labour intensive techniques would reduce greenhouse emissions. second, we argue that urbanization tends to increase emissions through pollution channels. the migration from rural to urban centres to join the accumulation of industries effect, co2 would increase considerably. however, the result showed a negative relationship. although at a minimal percentage, a unit increase in the labour indicator caused 0.03% reduction in carbon dioxide emissions in the selected ssa countries. this rather supports the labour-intensive technique of production (li et al., 2020) against the assumption that labour increases air pollution. 4.2 least square dummy variable (lsdv) approach this section presents the results of the least square dummy variable (lsdv). hence, we performed lsdv regression as presented in table 5. table 5. lsdv regression results in table 5, we computed the lsdv to explain the increment in carbon dioxide emissions levels in ssa countries between 2005 and 2020. as shown there, there are three parts of the lsdv analysis comprising continuous variables association, countries interaction with economic growth and dummy for years. first, the lsdv estimation produced results similar to those of pooled ols except for the tourism revenue. the tourism revenue supported the earlier hypothesis that the tourism co2 coef. st. err. t-value p-value [95% conf interval] sig gdp 0.007 .003 2.54 0.012 .002 .013 ** ec 0.047 .008 5.77 0.000 .031 .063 *** tourismr -0.039 .013 -3.04 0.003 -.064 -.014 *** tradeop 0.001 .001 1.40 0.163 0 .003 labour -0.057 .014 -3.92 0.000 -.085 -.028 * region1#co: base 1 0 . . . . . middle africa -0.007 .003 -2.00 0.047 -.013 0 * east africa -0.005 .006 -0.83 0.405 -.017 .007 south africa -0.01 .003 -3.06 0.002 -.017 -.004 * west africa -0.009 .003 -2.97 0.003 -.015 -.003 * 2009b 0 . . . . . 2010 0.016 .036 0.46 0.645 -.054 .086 2011 0.035 .036 0.95 0.341 -.037 .106 2012 0.061 .037 1.67 0.095 -.011 .134 * 2013 0.078 .038 2.06 0.04 .003 .152 * 2014 0.097 .038 2.52 0.012 .021 .172 * 2015 0.091 .039 2.34 0.02 .014 .168 * 2016 0.112 .04 2.78 0.006 .033 .191 * 2017 0.147 .042 3.49 0.001 .064 .229 * 2018 0.173 .045 3.86 0.000 .085 .262 * constant 0.964 .143 6.75 0.000 .683 1.245 * mean dependent var 0.854 sd dependent var 1.457 r-squared 0.214 number of obs 342 f-test 4.316 prob > f 0.000 akaike crit. (aic) -373.141 bayesian crit. (bic) -300.280 *significant at p<.05. tourism revenue, economic growth and carbon dioxide emissions nexus sector could improve environmental quality. simply, tourism revenue demonstrates a negative relationship with carbon dioxide emissions at 1% significance level. it shows that 1% increase in tourism revenue decreased carbon dioxide emissions in about 0.039%. second, we decompose the ssa countries into regions to determine the regional dummy interaction with economic growth. unlike previous studies that lumped a large list of countries into single estimation, we categorized ssa countries into sub-regions comprising eastern africa, central africa, southern africa and western africa where eastern african region remains the base dummy to avoid dummy trap. table 4.3, the constant value explains that carbon dioxide emissions from the controlled group eastern africa is 0.964. the gdp coefficient has shown marginal effects of economic growth for the control group di= eastern africa. in other words, eastern africa contributed an additional 0.07 carbon dioxide emissions mt / % for an additional year of economic growth. likewise, the central africa interaction with economic growth demonstrated that countries from this region emitted carbon dioxide [0.964(0.007=0.957 units) for extra years of economic growth. as such, central africa region emitted carbon dioxide lesser units of 0.007 below the 0.964 threshold. moving to south africa region, considering the south africa region interaction with gdp, we established that the region emitted [0.964(0.01=0.95 units] for extra years of economic growth. this shows that south africa region emitted carbon dioxide lesser units of 0.01 below the threshold. finally, in the western african region, the economic growth interaction with west african countries indicated that those countries from the region emitted [0.964(0.009=0.955 units] for extra years of economic growth. this simply shows that the west african countries’ carbon dioxide emissions contributed lesser units 0.01 below the threshold. third, we introduced a dummy year variable to explain the period that the carbon dioxide emissions actually hit the selected countries. we made 2009 the base year. hence, the dummy year estimation reveals that the average carbon dioxide emissions were 0.016 mt / % higher in 2010 but not significant. after that, in 2012, carbon dioxide emission was given as 0.061, which was higher than 2010. in general, the coefficients of the dummy year have indicated a statistically significant increment in carbon dioxide emissions in 2013 (0.078 mt), 2014 (0.097 mt), 2015 (0.091 mt), 2016 (0.11 mt), 2017 (0.147 mt) and 2018 (0.173 mt). while other dummy years’ results were high, the 2015 revealed that carbon dioxide decreased slightly to about 0.091, but later increased to 0.178 in 2018. this is to understand that the dummy year coefficients explained that holding dependent and independent variables constant, on average, carbon dioxide had greater emissions between 2012 and 2018 in the ssa countries. 5. conclusions and policy implications the study investigates the relationship between tourism’s revenue, economic growth and carbon dioxide emissions in the ssa countries. it further decomposes the ssa region into sub-regions and examines countries’ interaction with economic growth and dummy years to understand each region's contributions to the accumulation of greenhouse emissions in ssa countries. finally, the study determines the labour force involvement in greenhouse emissions across the sampled countries. as such, four independent conclusions were drawn. first, although the economic activities 98 anaba et al. of the ssa countries are low compared with some asia and europe economies, the revenue realized from the tourism industry contributes greatly to the total emissions of the region. this is accountable to the reinvestment of growth gains into the tourism industry, which triggers the uncontrolled fossil fuels being used around the year and the trade openness through fdi channel. however, the world market competition does not and should not discourage considerable investment into the tourism industry in the ssa countries, otherwise, the region would remain undeveloped. second, with the decomposition of the region into sub-regions, we conclude that, although at varying degrees, the sub-regions jointly contributed to the high rate of greenhouse emissions in the ssa countries. for example, southern african countries contributed the highest percentage of greenhouse emissions compared to the other sub-regions due to heavy use of industrial equipment and high use of electricity. while the ssa sub-regions should not restrain from heavy investment into the real sector, the government could integrate into the global value chains for optimal use of equipment and resources. third, before the digital age (ir4.0), countries relied solely on capital-intensive techniques in the production process. this period of ir3.0 was a means of capital-intensive use of machines that de-emphasized the labour-intensive technique, which created a labour surplus in the long run. ., although at a minimal rate, we conclude in favour of the labour-intensive technique as the result of this study showed, where it triggered the reduction of greenhouse emissions in the ssa countries rather than the capital-intensive use of production. fourth, developing sustainable tourism is one of the 2030 sustainable development goals (sdg) has an indirect effect on poverty reduction. this realization that the positive effect of tourism cannot be altered due to the infinitesimal incremental effect of tourism revenue on co2 in the ssa countries. thus, rather than curtailing tourism expansion in the ssa countries, the governments of the sampled countries should study those elements to increase emissions by developing or adopting alternatives to emission reduction. for example, if electricity consumption increases co2 in a country or sub-region, it is imperative that the government consider the installation of solar resources as a strong alternative policy to electricity consumption to mitigate against excess greenhouse emissions in such a country or sub-region. finally, the government of the sampled ssa countries should carefully consider acceptance of foreign direct investment (fdi) absorption into the economic system. this is especially necessary that bringing in an industry that would trigger carbon emissions into the country should be technically avoided. rather, the government may consciously integrate itself into the digital age movement through intensive knowledge transfer in the space/digital exploration and intensive computer educational courses (e.g. artificial intelligence (ai), cloud computing, etc.) that allow for digital marketing, business development and physical services and construction. crucially, the empirical results demonstrated that the economic expansion of african nations had a major impact on carbon dioxide emissions. furthermore, the tourism industry's potential to reduce carbon dioxide emissions is indicated by the revenue it generates for african nations. therefore, it is recommended that the countries included in the sample implement stringent policies that prioritize environmental regulations. this will help to mitigate the negative impact of foreign direct economic activities on both the natural and human environments. tourism revenue, economic growth and carbon dioxide emissions nexus limitations of the study and future research the study analyses how tourism revenue and economic growth contributed to carbon dioxide emissions in selected 47 (ssa) between 2005 and 2020. other studies should look at how tourism revenue and economic growth contribute to carbon dioxide in more (ssa) countries on a wider range. although the study employed pooled ordinary least square (ols) and least square dummy variable (lsdv) as the study econometric technique, we observed that the study did not address the long-run relationships of the variables, which would have attracted panel cointegration. the future study may proceed to test for stationary to detect the best econometric method, such as panel autoregressive distributed lag (panel ardl) model, panel vector error correction model, pooled mean group (pmg), and mean group (mg). references adedoyin, f.f., agboola, p.o., ozturk, i., bekun, f.v., & agboola, m.o. (2021). environmental consequences of economic complexities in the eu amidst a booming tourism industry: accounting for the role of brexit and other crisis events. journal of cleaner production. 305:127117. https://doi.org/10.1016/j.jclepro.2021.127117 adedoyin, f.f., bekun, f.v., (2020). modelling the interaction between tourism, energy consumption, pollutant emissions and urbanization: renewed evidence from panel var. environmental science pollution research. 27(31):38881-38900. https://doi.org/10.1007/s11356-020-09869-9 agyeman, f.o., zhiqiang, m., li, m., sampene, a.k., dapaah, m.f., kedjanyi, e.a.g., buabeng, p., li, y., hakro, s., & heydari, m. (2022). probing the effect of governance of tourism development, economic growth, and foreign direct investment on carbon dioxide emissions in africa: the african experience. energies, 15(13), 4530 https://doi.org/10.3390/en15134530 al-mulali, u., & binti che sab, c.n. (2012). the impact of energy consumption and co2 emission on the economic growth and financial development in the sub-saharan african countries. energy, 39(1), 180-186. https://doi.org/10.1016/j.energy.2012.01.032 awokola, b.i., okello, g., mortimer, k. j., jewell, c. p., erhart, a., & semple, s. (2020). measuring air quality for advocacy in africa (ma3): feasibility and practicality of longitudinal ambient pm2. 5 measurement using low-cost sensors. international journal of environmental research and public health, 17(19), 7243. https://doi.org/10.3390/ijerph17197243 bataka, h. (2020). globalization and environmental pollution in sub-saharan africa. african journal of economic review. 9(1):191-205. bekun, f.v. (2022). mitigating emissions in india: accounting for the role of real income, renewable energy consumption and investment in energy. international journal of energy economics and policy. https://doi.org/10.32479/ijeep.12652 bekun, f.v., adedoyin, f.f., etokakpan, m.u., & gyamfi, b.a. (2022). exploring the tourism-co2 emissionsreal income nexus in e7 countries: accounting for the role of institutional quality. journal of policy research in tourism, leisure and events, 14(1), 1-19. https://doi.org/10.1080/19407963 .2021.2017725 bekun, f.v., alola, a.a., & sarkodie, s.a. (2019). toward a sustainable environment: nexus between co2 emissions, resource rent, renewable and non-renewable energy in 16-eu countries. science of the total environment, 657, 1023-1029. https://doi.org/10.1016/j.scitotenv.2018.12.104 chaabouni, s., & saidi, k. (2017). the dynamic links between carbon dioxide (co2) emissions, health spending and gdp growth: a case study for 51 countries. environmental research, 158, 137-144. https://doi.org/10.1016/j.envres.2017.05.041 chen, l., thapa, b., & yan, w. (2018). the relationship between tourism, carbon dioxide emissions, and economic growth in the yangtze river delta, china. sustainability, 10(7), 2118. https://doi.org/10.3390/su10072118 chien, f., sadiq, m., nawaz, m.a., hussain, m. s., tran, t.d., & le thanh, t. (2021). a step toward reducing air pollution in top asian economies: the role of green energy, eco-innovation, and environmental taxes. journal of environmental management, 297, 113420. https://doi.org/10.1016/ j.jenvman.2021.113420 https://doi.org/10.1016/j.jclepro.2021.127117 https://doi.org/10.1007/s11356-020-09869-9 https://doi.org/10.3390/en15134530 https://doi.org/10.1016/j.energy.2012.01.032 https://doi.org/10.3390/ijerph17197243 https://doi.org/10.32479/ijeep.12652 https://doi.org/10.1080/19407963.2021.2017725 https://doi.org/10.1080/19407963.2021.2017725 https://doi.org/10.1016/j.scitotenv.2018.12.104 https://doi.org/10.1016/j.envres.2017.05.041 https://doi.org/10.3390/su10072118 https://doi.org/10.1016/j.jenvman.2021.113420 https://doi.org/10.1016/j.jenvman.2021.113420 100 anaba et al. destek, m.a., & aydın, s. (2022). an empirical note on tourism and sustainable development nexus. environmental science and pollution research, 29(23), 34515-34527. https://doi.org/10.1007 /s11356-021-18371-9 djellouli, n., abdelli, l., elheddad, m., ahmed, r., & mahmood, h. (2022). the effects of non-renewable energy, renewable energy, economic growth, and foreign direct investment on the sustainability of african countries. renewable energy, 183, 676-686. https://doi.org/10.1016/ j.renene.2021.10.066 dogru, t., & bulut, u. (2018). is tourism an engine for economic recovery? theory and empirical evidence. tourism management, 67, 425-434. https://doi.org/10.1016/j.tourman.2017.06.014 du, d., lew, a.a., & ng, p.t. (2016). tourism and economic growth. journal of travel research, 55(4), 454464. https://doi.org/10.1177/0047287514563167 el menyari, y. (2021). the effects of international tourism, electricity consumption, and economic growth on co2 emissions in north africa. environmental science and pollution research, 1-11. https://doi.org/10.21203/rs.3.rs-224555/v1 fisher, s., bellinger, d.c., cropper, m.l., kumar, p., binagwaho, a., koudenoukpo, j.b., park, y., taghian, g., & landrigan, p.j. (2021). air pollution and development in africa: impacts on health, the economy, and human capital. the lancet planetary health, 5(10), 681-688. https://doi.org/10.1016/s2542-5196(21)00201-1 flanagan, e., oudin, a., walles, j., abera, a., mattisson, k., isaxon, c., & malmqvist, e. (2022). ambient and indoor air pollution exposure and adverse birth outcomes in adama, ethiopia. environment international, 164, 107251. https://doi.org/10.1016/j.envint.2022.107251 gao, c., ge, h., lu, y., wang, w., & zhang, y. (2021). decoupling of provincial energy-related co2 emissions from economic growth in china and its convergence from 1995 to 2017. journal of cleaner production, 297, 126627. https://doi.org/10.1016/j.jclepro.2021.126627 gyamfi, b.a, adebayo, t. s, bekun, f.v, agyekum, e.b., kumar, n.m., alhelou, h.h., & al-hinai, a. (2021). beyond environmental kuznets curve and policy implications to promote sustainable development in mediterranean. energy republic. energy reports 7, 6119-6129. https://doi.org/10.1016/j.egyr.2021.09.056 haller, a.p., ionela butnaru, g., tacu hârșan, g.d., & ştefănică, m. (2021). the relationship between tourism and economic growth in the eu-28. is there a tendency towards convergence? economic research-ekonomskaistraživanja, 34(1), 1121-1145. https://doi.org/10.1080/1331677x.2020.1819852 hanif, i. (2018). impact of economic growth, nonrenewable and renewable energy consumption, and urbanization on carbon emissions in sub-saharan africa. environmental science pollution research. 25(15):15057-15067. https://doi.org/10.1007/s11356-018-1753-4 işik, c., kasımatı, e., & ongan, s. (2017). analyzing the causalities between economic growth, financial development, international trade, tourism expenditure and/on the co2 emissions in greece. energy sources, part b: economics, planning, and policy, 12(7), 665-673. https://doi.org/10.1080/15567249.2016.1263251 jiang, s., tan, x., hu, p., wang, y., shi, l., ma, z., & lu, g. (2022). air pollution and economic growth under local government competition: evidence from china, 2007-2016. journal of cleaner production, 334, 130231. https://doi.org/10.1016/j.jclepro.2021.130231 jong, m. c., soh, a.n., & puah, c.h. (2022). tourism sustainability: climate change and carbon dioxide emissions in south africa. international journal of energy economics and policy, 12(6), 412-417. https://doi.org/10.32479/ijeep.13662 khan, m.m., zaman, k., irfan, d., awan, u., ali, g., kyophilavong, p., shahbaz, m., & naseem, i. (2016). triangular relationship among energy consumption, air pollution and water resources in pakistan. journal of cleaner production, 112, 1375-1385. https://doi.org/10.1016/ j.jclepro.2015.01.094 kiracı, k. & bakır, m. (2019). causal relationship between air transport and economic growth: evidence from panel data for high, upper-middle, lower-middle and low-income countries. https://doi.org/10.5782/2223-2621.2019.22.3.24 kukla-gryz, a. (2009). economic growth, international trade and air pollution: a decomposition analysis. ecological economics, 68(5), 1329-1339. https://doi.org/10.1016/j.ecolecon.2008.09.005 ku, g., da silveira, v., kegels, g., develtere, p., nemery, b., doussou, j., bossyns, p., & gyselinck, k. (2021). tackling bad air and promoting prevention and care for chronic lung diseases in sub-saharan africa. european journal of public health, 31(supplement_3), ckab165. 189. https://doi.org/ 10.1093/eurpub/ckab165.189 kyara, v.c., rahman, m.m., & khanam, r. (2022). investigating the environmental externalities of tourism development: evidence from tanzania. https://doi.org/10.1016/j.heliyon.2022.e09617 https://doi.org/10.1007/s11356-021-18371-9 https://doi.org/10.1007/s11356-021-18371-9 https://doi.org/10.1016/j.renene.2021.10.066 https://doi.org/10.1016/j.renene.2021.10.066 https://doi.org/10.1177/0047287514563167 https://doi.org/10.21203/rs.3.rs-224555/v1 https://doi.org/10.1016/s2542-5196(21)00201-1 https://doi.org/10.1016/j.envint.2022.107251 https://doi.org/10.1016/j.jclepro.2021.126627 https://doi.org/10.1016/j.egyr.2021.09.056 https://doi.org/10.1080/1331677x.2020.1819852 https://doi.org/10.1007/s11356-018-1753-4 https://doi.org/10.1080/15567249.2016.1263251 https://doi.org/10.1016/j.jclepro.2021.130231 https://doi.org/10.32479/ijeep.13662 https://doi.org/10.1016/j.jclepro.2015.01.094 https://doi.org/10.1016/j.jclepro.2015.01.094 https://doi.org/10.5782/2223-2621.2019.22.3.24 https://doi.org/10.1016/j.ecolecon.2008.09.005 https://doi.org/10.1093/eurpub/ckab165.189 https://doi.org/10.1093/eurpub/ckab165.189 https://doi.org/10.1016/j.heliyon.2022.e09617 tourism revenue, economic growth and carbon dioxide emissions nexus li, x., wang, y., zhou, h., & shi, l. (2020). has china's war on pollution reduced employment? quasiexperimental evidence from the clean air action. journal of environmental management, 260, 109851. https://doi.org/10.1016/j.jenvman.2019.109851 lin, j., pan, d., davis, s. j., zhang, q., he, k., wang, c., streets, d. g., wuebbles, d. j., & guan, d. (2014). china's international trade and air pollution in the united states. proceedings of the national academy of sciences, 111(5), 1736-1741. https://doi.org/10.1073/pnas.1312860111 liu, h., cui, w., & zhang, m. (2022). exploring the causal relationship between urbanization and air pollution: evidence from china. sustainable cities and society, 80, 103783. https://doi.org/ 10.1016/j.scs.2022.103783 nosheen, m., iqbal, j., & khan, h.u. (2021). analyzing the linkage among co 2 emissions, economic growth, tourism, and energy consumption in the asian economies. environmental science and pollution research, 28, 16707-16719. https://doi.org/10.1007/s11356-020-11759-z nyasha, s., odhiambo, n.m., & asongu, s.a. (2021). the impact of tourism development on economic growth in sub-saharan africa. the european journal of development research, 33, 1514-1535. https://doi.org/10.1057/s41287-020-00298-5 ohajionu, u.c., gyamfi, b.a., haseki, m.i., & bekun, f.v. (2022). assessing the linkage between energy consumption, financial development, tourism and environment: evidence from method of moments quantile regression. environmental science and pollution research, 29(20), 3000430018. https://doi.org/10.1007/s11356-021-17920-6 ongan, s., işik, c., & özdemir, d. (2017). the effects of real exchange rates and income on international tourism demand for the usa from some european union countries. economies, 5(4), 51. https://doi.org/10.3390/economies5040051 ouyang, x., shao, q., zhu, x., he, q., xiang, c., & wei, g. (2019). environmental regulation, economic growth and air pollution: panel threshold analysis for oecd countries. science of the total environment, 657, 234-241. https://doi.org/10.1016/j.scitotenv.2018.12.056 ozpolat, a., ozsoy, f.n., & destek, m.a. (2021). investigating the tourism originating co 2 emissions in top 10 tourism-induced countries: evidence from tourism index. in strategies in sustainable tourism, economic growth and clean energy (pp. 155-175). springer, cham. https://doi.org/10.1007/978-3-030-59675-0_9 pal, d., & mitra, s.k. (2017). the environmental kuznets curve for carbon dioxide in india and china: growth and pollution at crossroad. journal of policy modeling, 39(2), 371-385. https://doi.org/10.1016/j.jpolmod.2017.03.005 qin, m., zhang, x., li, y., & badarcea, r.m. (2023). blockchain market and green finance: the enablers of carbon neutrality in china. energy economics, 118, 106501. https://doi.org/ 10.1016/j.eneco.2022.106501 shaheen, k., zaman, k., batool, r., khurshid, m.a., aamir, a., shoukry, a.m., & gani, s. (2019) dynamic linkages between tourism, energy, environment, and economic growth: evidence from top 10 tourism-induced countries. environmental science pollution research 26(30), 31273-31283. https://doi.org/10.1007/s11356-019-06252-1 su, c.-w., pang, l.-d., tao, r., shao, x., & umar, m. (2022). renewable energy and technological innovation: which one is the winner in promoting net-zero emissions? technological forecasting and social change, 182, 121798. https://doi.org/10.1016/j.techfore.2022.121798 su, c.-w., umar, m., kirikkaleli, d., awosusi, a.a., &altuntaş, m. (2023). testing the asymmetric effect of financial stability towards carbon neutrality target: the case of iceland and global comparison. gondwana research. https://doi.org/10.1016/j.gr.2022.12.014 sun, y.-y., gossling, s., & zhou, w. (2022). does tourism increase or decrease carbon emissions? a systematic review. annals of tourism research, 97, 103502. https://doi.org/10.1016/ j.annals.2022.103502 tshehla, c. e., & wright, c.y. (2019). spatial and temporal variation of pm10 from industrial point sources in a rural area in limpopo, south africa. international journal of environmental research and public health, 16(18), 3455. https://doi.org/10.3390/ijerph16183455 wijesekara, c., tittagalla, c., jayathilaka, a., ilukpotha, u., jayathilaka, r., & jayasinghe, p. (2022). tourism and economic growth: a global study on granger causality and wavelet coherence. plos one, 17(9), e0274386. https://doi.org/10.1371/journal.pone.0274386 wipfli, h., kumie, a., atuyambe, l., oguge, o., rugigana, e., zacharias, k., simane, b., samet, j., & berhane, k. (2021). the geohealth hub for eastern africa: contributions and lessons learned. geohealth, 5(6), e2021gh000406. https://doi.org/10.1029/2021gh000406 world health organization. (2019). world health statistics overview 2019: monitoring health for the sdgs, sustainable development goals (no. who/dad/2019.1). world health organization. yusuf, m.s., musibau, h.o., dirie, k.a., & shittu, w.o. (2023). role of trade liberalization, industrialisation and energy use on carbon dioxide emissions in australia: 1990 to 2018. environmental science and pollution research, 1-16. https://doi.org/10.1007/s11356-023-27825-1 https://doi.org/10.1016/j.jenvman.2019.109851 https://doi.org/10.1073/pnas.1312860111 https://doi.org/10.1016/j.scs.2022.103783 https://doi.org/10.1016/j.scs.2022.103783 https://doi.org/10.1007/s11356-020-11759-z https://doi.org/10.1057/s41287-020-00298-5 https://doi.org/10.1007/s11356-021-17920-6 https://doi.org/10.3390/economies5040051 https://doi.org/10.1016/j.scitotenv.2018.12.056 https://doi.org/10.1007/978-3-030-59675-0_9 https://doi.org/10.1016/j.jpolmod.2017.03.005 https://doi.org/10.1016/j.eneco.2022.106501 https://doi.org/10.1016/j.eneco.2022.106501 https://doi.org/10.1007/s11356-019-06252-1 https://doi.org/10.1016/j.techfore.2022.121798 https://doi.org/10.1016/j.gr.2022.12.014 https://doi.org/10.1016/j.annals.2022.103502 https://doi.org/10.1016/j.annals.2022.103502 https://doi.org/10.3390/ijerph16183455 https://doi.org/10.1371/journal.pone.0274386 https://doi.org/10.1029/2021gh000406 https://doi.org/10.1007/s11356-023-27825-1 102 anaba et al. appendix i. list of countries in the panel data s/n country s/n country 1 angola 25 malawi 2 benin 26 mali 3 botswana 27 mauritania 4 burkina faso 28 mauritius 5 burundi 29 mozambique 6 cameroon 30 namibia 7 cape verde 31 niger 8 central african republic 32 nigeria 9 chad 33 republic of the congo 10 comoros 34 rwanda 11 democratic republic of the congo 35 sao tome and principe 12 equatorial guinea 36 senegal 13 eritrea, 37 seychelles 14 ethiopia 38 sierra leone 15 gabon 39 somalia 16 gambia 40 south africa 17 ghana 41 sudan 18 guinea 42 swaziland 19 guinea-bissau 43 tanzania 20 ivory coast 44 togo 21 kenya 45 uganda 22 lesotho 46 zambia 23 liberia 47 zimbabwe 24 madagascar 1. introduction 2. literature review 3. methodology 4. results and discussions 5. conclusions and policy implications references © the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 2 (2025), pages 164-184 https://doi.org/10.17979/ejge.2025.14.2.11363 submitted: nov 2, 2024 accepted: nov 3, 2025 published: dec 17, 2025 article do natural resources hinder or help? exploring economic development in eu countries through human and institutional lenses mert gül,1,* mustafa batuhan tufaner,1 1 istanbul beykent university, economics (english) department, istanbul, türkiye. *correspondence: mertgul@beykent.edu.tr abstract. this study investigates the impact of natural resource dependence and abundance on economic development in 21 eu countries from 1996 to 2019, focusing on critical gaps in the literature on resourcedriven development. using the generalized method of moments (gmm), we address issues of endogeneity, heterogeneity, and cross-sectional dependence issues to capture nuanced relationships. findings indicate that natural resource dependence and abundance are negatively associated with development in the eu; economic growth, however, positively reinforces it. additionally, the observed patterns consistent with the resource curse are associated with lower development outcomes. considering the impact of natural resources alongside economic growth, human capital, and institutional quality factors, this study provides valuable insights for eu policymakers, highlighting the need for sustainable and inclusive resource management policies. in a broader sense, the findings are robust for resource-rich countries aimed at achieving sustainable development. keywords: human development index; natural resources; economic growth; institutional quality; panel data; gmm jel classification: o15; o13; o47; o43 ; c23 1. introduction since the united nations development programme (undp, 1990) introduced the human development index (hdi), there has been continuous debate regarding the link between natural resources and development outcomes. hdi is a widely recognized composite metric used to evaluate overall development, considering three fundamental dimensions: life expectancy, educational attainment, and standard of living. as a pivotal measure of human well-being, policymakers are increasingly attuned to their countries’ hdi outcomes in pursuit of improving societal welfare beyond conventional indicators such as gross domestic product (gdp). the nexus between macroeconomic indicators and economic development is well-documented (khan et al., 2023), yet it remains a complex and multifaceted topic of study. historically, the hdi emerged as part of a growing recognition that economic growth alone does not guarantee improvements in human welfare (haq, https://creativecommons.org/licenses/by-nc/4.0/ do natural resources hinder or help? 1995, p. 24), in other words, as a response to the limitations of gdp in measuring the well-being of societies (sen, 1999). over the last three decades, development economists have emphasized the importance of people’s choices and freedoms while improving their health and education levels. hdi gained worldwide recognition with periodically adjusted metrics and a methodology to measure it, yet the fundamentals have continued to center on the multidimensional features of human welfare (fukuda-parr, 2003). a key strand of development economics focuses on natural resource dependence (nrd), defined as the degree to which an economy relies on revenues or exports related to natural resources. dependence can heighten vulnerability to commodity price shocks, limit economic diversification (matsen & torvik, 2005), and weaken incentives for human capital accumulation (gylfason, 2001; petermann et al., 2007). these negative impacts of nrd align with the resource curse hypothesis (auty, 1994; sachs & werner, 2001; zhang et al., 2008). high nrd may also deepen inequality (bardhan, 2005), concentrating benefits among elites while leaving broader society underdeveloped, contributing to social instability (kim et al., 2020) and unrest (dabla-norris et al., 2015; stiglitz, 2012). weak institutional structures further amplify these risks by enabling corruption and misallocation of resource revenues (cust & mihalyi, 2017; rodrik et al., 2004). another perspective emphasizes natural resource abundance (nra) as typically measured as the size of a country’s natural endowments or the share of resources in total output. when supported by strong institutions and effective governance, abundant resources can stimulate economic growth (singh et al., 2024), reduce poverty, and finance investments in health and education, thereby enhancing human capital (cust & mihalyi, 2017; li et al., 2023). however, nra may also be associated with rent-seeking behavior, institutional fragility, and volatility in public revenues, all of which reinforce the mechanisms of the resource curse (mehlum et al., 2006; mousavi & clark, 2021; ross, 2012; sachs & warner, 2001; timbe et al., 2024). in this view, the presence of resource abundance itself may create incentives for political capture and mismanagement, thereby constraining sustainable development. contemporary empirical evidence shows mixed outcomes. the nra has been found to hinder development for african countries (debonheur et al., 2023), salahojaev et al. (2024) presented a u-shaped relationship where hdi improves once resource rents exceed a threshold (42.8% of gdp) and are managed efficiently in belt and road initiative (bri) countries. these contradictory findings suggest that the nra is not inherently detrimental, but instead its effects depend on governance, institutional quality, and the allocation of revenues towards human capital and economic diversification (van der ploeg & poelhekke, 2017). taken together, both nrd and nra are linked to the resource curse, though through distinct mechanisms. resource dependence reflects economic reliance and exposure to volatility, while abundance emphasizes structural endowments and potential for mismanagement. understanding these dynamics is crucial for designing policies for channeling resource revenues toward sustainable development progress. in this context, the sustainable development goals1 (sdgs) provide a valuable framework to assess the broader developmental implications for natural resources (undp, 2015). the links between nrd, nra, and specific sdgs are detailed in the literature review section, enabling policymakers to evaluate how resource endowments affect economic development across 1 sustainable development goals (sdgs) are a set of global objectives adopted by the un for improving human development levels all by 2030 (undp, 2015). 166 gül and tufaner multiple dimensions (eurostat, 2022). this study examines the impact of nrd and nra along with several other control variables on hdi using panel data for 21 highly developed eu countries from 1996 to 2019. the generalized method of moments (gmm) estimator is employed to address endogeneity and unobserved heterogeneity (farhadi et al., 2015). our findings indicate a negative association between natural resource rents and human development in eu countries, suggesting that resource dependence and abundance may constrain progress when not supported by growth, human capital, and institutional quality. specifically, the resource curse is linked with lower hdi levels. these results can inform eu policymakers on how to ensure that natural resource rents are utilized appropriately to support development. this study reviews the literature on the relationship between the dependent variable, hdi, and proxy variables, nra and nrd, with control variables in the next section. section 3 outlines the data sources and methodology. section 4 presents the gmm estimation results. the study is concluded in the final section. 2. literature review despite the substantial progress in development in the eu, the impact of natural resources on hdi remains an ongoing inquiry (radulescu et al., 2025). some eu countries have abundant resources, whereas others are resource-dependent and mostly rely on imports for raw materials and energy. this diversity presents an opportunity to examine the influence of resources, growth, human capital, and institutions on development. 2.1 economic development and natural resource dependence nrd is defined as an economy’s overreliance on resource revenues for exports, government budgets, or gdp. resource curse theory (gelb, 1988; sachs & warner, 1995) suggests that heavy dependence on volatile natural resource rents often crowds out investment in human capital, increases rentseeking behavior, and exposes public spending for social benefits to commodity price shocks (arezki & brückner, 2011), resulting in slower growth for countries. furthermore, nrd harms development as resource inflows appreciate the real exchange rate, reduce competitiveness in other tradable sectors, and foster de-industrialization (corden & neary, 1982). the asymmetric growth in both resource allocation and income distribution is the primary reason for limiting investments in crucial areas such as education and healthcare, which hinders development (stijns, 2005; mohamed, 2020). this problem is illustrated by nigeria’s (sala-i-martin & subramanian, 2013; ross, 2012) and venezuela’s heavy reliance on oil export revenues (karl, 1997; corrales & penfold, 2015), as decades of oil dependence have been accompanied by widespread poverty, fiscal mismanagement, and stagnant hdi levels despite substantial petroleum reserves. additionally, weaker institutional structures among resource-dependent countries are one do natural resources hinder or help? of the main reasons for hindered development in individuals and society (sdgs2 10, 16). the reliance on natural resources may lead to volatility rather than stable development progress, underinvestment in other industrial sectors, and governance issues (saud et al., 2023) that negatively impact social outcomes at the education level and public health (sdgs 3, 4, 8, 9). bulte et al. (2005) reveal that resource-dependent developing countries tend to invest less in public services or other non-resource-dependent industries that can adversely impact their hdi rankings (sdgs 1, 10). in more contemporary evidence, cust and mihalyi (2017) echo these findings, showing that resourcedependent countries experience challenges transforming resource wealth into economic development, especially without an effective governance system (sdg 16). 2.2 economic development and natural resource abundance on the contrary, nra refers to the existence of substantial natural resource endowments that can have both beneficial as well as detrimental effects on development progress, depending on how authorities manage them (sdgs 1, 3, 4). while resource abundance can reduce poverty, expand healthcare spending, and improve education (sdgs 1, 3, 4, 16) to increase human capital if allocated effectively by the existence of strong institutions and strategic investments (li et al., 2023), but it may also foster rent capture (sdgs 10, 16, 17) by elites and governance failures in contexts of weak institutions (ross, 2001; mehlum et al., 2006). van der ploeg and poelhekke (2011) provide evidence that the main effect of resource abundance is to increase growth volatility, which in turn reduces the long-term average growth rate. results from african economies indicate that resource abundance harms development outcomes (debonheur et al., 2023), yet salahojaev et al. (2024) demonstrate that a u-shaped relationship for 51 bri countries increases hdi when a country’s share of natural resources exceeds 42.8% of gdp. their results imply that resource-rich countries, such as norway’s management of oil revenues (mehlum et al., 2006) or botswana’s diamond revenues (acemoglu et al., 2003), alongside greater financial development and efficient institutions, may channel their resource wealth, leading to higher hdi results. thus, the developmental impact of nra is not predetermined but rather depends on governance, distributional policies, and the ability to channel resources into human capital. 2.3 economic development and control variables the nexus between development and growth is well-established. sen's (1999) capability approach offers a theoretical framework for understanding this relationship, asserting that economic growth is essential but insufficient for human development. to truly improve hdi, growth must be inclusive and directed toward improving vital social indicators, including education, health, and income 2 the related sdgs to resource endowments and economic development are sdg1: no poverty, sdg3: good health and well-being, sdg4: quality education, sdg8: decent work and economic growth, sdg9: industry, innovation and infrastructure, sdg10: reduced inequalities, sdg16: peace, justice and strong institutions, sdg17: partnership for the goals. 168 gül and tufaner distribution. empirical research by ranis et al. (2000) confirms that countries experiencing sustained economic growth tend to improve their hdi, but only when the gains are channeled into public services. abraham and ahmed (2011) applied an error correction model, and their findings show that growth has sustained negative impact in the short run, while in the long run, this impact is reversed. hashmat et al. (2023) presented the adverse outcome of hdi due to the gdp per capita income disparities in their multi-country dataset from 1996 to 2021, applying gmm analysis. the increase in population within a country should not be considered a significant indicator of human capital growth on its own. sustainable population growth, achieved in an environment where access to education and social health indicators are improving (schultz, 1961; becker, 1962), is essential for countries to increase their human capital (hc) and perform better in their hdi rankings. hc is directly linked to two critical components of the hdi: education and life expectancy. countries prioritizing education and healthcare systems are better positioned to improve their hdi scores. barro and lee (2013) empirically found a strong positive correlation between educational attainment and improvements in hdi, particularly in life expectancy and income per capita. hanushek and woessmann (2021) also underscore the importance of human capital investments in achieving sustainable development in the long term. their study emphasizes that even in advanced economies, where educational systems are well-established, continued investment in human capital is critical for maintaining high hdi scores, especially in the face of technological change and evolving labor market demands. institutions play a fundamental role in determining economic development outcomes. according to acemoglu and robinson (2012), inclusive institutions that enforce the law, ensure political stability and promote good governance are crucial for converting economic gains into improvements in development. strong institutions help to manage resources efficiently, reduce corruption, and ensure that the benefits of economic growth are distributed equitably across society (torvik, 2009). countries with transparent and effective governance mechanisms may be better equipped to respond to crises (e.g., financial crises, global pandemics, etc.) and protect the economic development gains obtained in the past. empirical evidence provided by rodrik et al. (2004) supports this argument, showing that countries with more robust institutional frameworks experience higher levels of development. hong (2017) analyzed data from 1972 to 2008 and confirmed that natural resource dependence, diminished bureaucratic incentives to invest in human capital through spending in markets characterized by low institutional quality. radulović (2020) found a positive relationship between the quality of institutions and growth for eu countries in the see region by applying the ardl estimator only in the long run from 1996 to 2017. 2.4 contemporary empirical framework panel data models are widely utilized in empirical research to investigate the dynamic relationships between natural resource endowment and development. these models offer several advantages, including the control for unobserved heterogeneity across countries and over time, leading to more robust and consistent estimations (baltagi, 2021; van der ploeg & poelhekke, 2011). badeeb et al. (2017) suggest that panel frameworks provide a clear picture of the concept the resource curse. do natural resources hinder or help? hsiao (2014) argues that panel data models mitigate potential biases arising from omitted variables and measurement errors, providing more accurate insights into the complex interactions among the variables. therefore, this study employs a panel data model to examine the multifaceted impacts of natural resources and control variables on development outcomes. several studies employed alternative models to investigate the impact of natural resources on development. sinha (2019) investigated the relationship between resource rents and development by applying co-integration and vector error correction models (vecm) to study asiapacific countries, highlighting the moderating effect of globalization. destek et al. (2022), utilizing continuously-updated and fully modified (cup-fm) and continuously-updated and bias-corrected (cup-bc), fully modified ols (fmols), found that an inverted u-shaped relationship where resource dependence is not a barrier to growth but constitutes a curse for sustainable development by analyzing 28 countries from 1990 to 2017. aljarallah (2020) analyzed the resource-rich gulf countries and their economic activities by applying ardl from 1984 to 2014, and the results showed that resource abundance increases gdp. zhang et al. (2024) employed an autoregressive distributed lag (ardl) model for eight south asian countries from 1996 to 2022. they found that natural resource rents have a negative impact on economic performance in the short run. the european experience provides additional insights into the relationship between natural resources and development. studies on norway, a resource-rich country, highlight how its strong institutional structure and sovereign wealth funds have allowed resource abundance to be channeled into higher hdi levels (cappelen & mjøset, 2009; holden, 2013). in contrast, resource-rich southern and eastern european countries, as well as post-soviet states with weaker governance, struggle to convert their resource dependence into sustainable human development reflecting rent-seeking behavior and a fragile institutional structure (alexeev & conrad, 2009; esanov et al., 2001). empirical research on eu countries indicates that resource rents tend to have a limited direct effect on hdi; however, they operate indirectly through growth and institutional channels (boschini et al., 2013; mavratos et al., 2001; papyrakis & gerlagh, 2007). more recently, simionescu et al. (2024) found that the eu’s green economy transition substantially affects development. their study analyzes eu member countries from 2008 to 2022, while natural resource rents detriment development in the region. several studies have applied gmm regression (blundell & bond, 1998) with various macroeconomic indicators. farhadi et al. (2015) examined the relationship between the impact of resource rents on growth in 99 countries from 1970 to 2010 on their path of economic development. the study revealed that the countries with higher freedom of choice levels turn the adverse growth effects of natural resources into positive ones for development. awoa et al. (2024) analyzed how economic complexity moderates the negative impact on income inequality across 111 countries, finding that economic complexity reduces income inequality linked to resource dependence. additionally, countries with less income inequality problems have smoother development paths. the empirical results of debonheur et al. (2023) show that resource abundance has a negative influence on development outcomes for 41 african countries covering from the period from 1996 to 2019. farooq et al. (2025) examined 48 asian countries from 1996 to 2019 and found that while natural resources negatively affect growth, strong governance moderates this effect, transforming the resource curse into a resource blessing and enhancing growth. 170 gül and tufaner 3. data and methodology 3.1 data description this study examines the relationship between nrd, nra, and development for 21 eu member countries3 between 1996 and 2019. the choice of variables is based on both theoretical reasoning and established empirical practice. hdi is the dependent variable as a composite measure encompassing life expectancy, educational attainment, and standard of living. both nrd (total natural resource rents, as a percentage of gdp) and nra (natural capital per capita, nonrenewable assets, constant 2018 us$) are the proxy variables that their data are derived from the world bank’s the changing wealth of nation dataset (cwn). nrd captures the extent to which an economy relies on resource revenues relative to its overall output, which is central to the “resource curse” hypothesis and nra reflects the stock dimension of natural wealth available to a country, thereby operationalizing the abundance concept. economic growth (eg) from the world bank’s world development indicators (wdi), human capital (hc) from the penn world table (pwt), and institutional quality (iq) from transparency international (ti) are included as control variables. in the gmm estimation, internal instruments are generated from the lagged values of endogenous regressors (e.g., lagged hdi and nrd/nra), rather than using eg, hc, or iq as external instruments. definitions and sources of the variables are given in table 1 below. table 1. definitions and resources of variables. variables definition source dependent variable economic development hdi human development index undp proxy variables natural resource dependence nrd total natural resources rents (%gdp) world bank (cwn) natural resource abundance nra natural capital per capita, nonrenewable assets (constant 2018 us$) world bank (cwn) control variables economic growth eg gdp growth (annual %) world bank (wdi) human capital hc human capital index pwt institutional quality iq corruption perceptions index ti source: undp (2024), world bank (2021, 2024), pwt (feenstra et al., 2015), ti (transparency international, 2023). 3.2 preliminary investigation tables 2 and 3 present statistics and the correlation matrix of the variables. according to descriptive statistics, the highest hdi among the sample is 0.951, the lowest is 0.711, and the mean is 0.865. the highest nrd is 5.714, the lowest is 0.008, and the mean is 0.702. therefore, there is a significant 3 austria, belgium, czechia, denmark, estonia, finland, france, germany, greece, hungary, italy, latvia, lithuania, luxembourg, netherlands, poland, portugal, slovakia, slovenia, spain, sweden. do natural resources hinder or help? difference between the nrd of the countries in the sample. the same goes for the nra. while the highest nra value is 7270.7, some countries in the analysis have no natural resources for at least one year, while the mean value of nra is 1384.133. development is adversely affected by the nrd and eg factors, as seen by the correlation results in table 3. on the other hand, nra, hc, and iq positively affect economic development. the variable with the highest interaction with hdi is nra, with a coefficient of 0.3858, while the variable with the least interaction is hc, with 0.1860. table 2. descriptive statistics, variable observation mean std. dev. min max hdi 504 0.8654544 0.0489737 0.711 0.951 nrd 504 0.5325826 0.7025445 0.0088736 5.714163 nra 483 870.2907 1384.133 0 7270.7 eg 504 2.431929 3.141529 -14.83861 13.05001 hc 504 2.98 7.70 2.92 3.82 iq 504 6.413512 1.942724 3 10 source: authors’ own calculations table 3. correlation matrix. hdi nrd nra eg hc iq hdi 1.0000 nrd -0.2105 1.0000 nra 0.3858 0.3781 1.0000 eg -0.2706 0.2051 -0.1398 1.0000 hc 0.1860 0.1205 0.1569 0.0202 1.0000 iq 0.7322 -0.1731 0.3734 -0.1128 0.1366 1.0000 source: authors’ own calculations when performing panel data analysis, it is essential to first test for cross-sectional dependence. if there is no cross-sectional dependence, first-generation panel data tests can be employed. conversely, if cross-sectional dependence exists, second-generation panel data tests should be utilized. there are various tests to detect cross-sectional dependence. since t (24) > n (21) in our model, the breusch-pagan lm test should be applied. h0 of the test states that there is no crosssectional dependence, whereas h1 states that there is. the results presented in table 4 below indicate the existence of cross-sectional dependence. due to cross-sectional dependence, unit root tests to be applied in the later stages of the analysis should be selected among techniques that account for crosssectional dependence. table 4. cross-sectional dependence tests. model 1: nrd model 2: nra test stats. prob. stats. prob. lm 802.2 0.0000 880.5 0.0000 lm adj* 46.33 0.0000 50.46 0.0000 lm cd* 21.52 0.0000 23.09 0.0000 source: authors’ own calculations. 172 gül and tufaner in this study, we apply three variants of the breusch-pagan lagrange multiplier (lm) test for cross-sectional dependence. the standard breusch-pagan lm test (breusch & pagan, 1980) evaluates the null hypothesis of no cross-sectional correlation in the residuals. however, when the panel dimensions differ (e.g., moderate t relative to n), the test can be biased. pesaran et al. (2008) proposed adjusted versions to improve finite-sample properties. the lm adj corrects the mean and variance of the lm statistic under finite samples, while the lm cd (cross-section dependence test) standardizes the statistic for panels where n and t are both large. reporting all three ensures robustness: in our case, all versions reject the null, confirming cross-sectional dependence. it is also important to perform the endogeneity test to ensure the robustness of the estimates. the durbin (1954) score and wu-hausman (wu, 1973; hausman, 1978) tests were applied to test endogeneity. the h0 of the test is “variables are exogenous”. the test results are shown in table 5 below. table 5. endogeneity tests. model 1: nrd model 2: nra test statistic probability statistic probability durbin (score) 202.472 0.0000 151.81 0.0000 wu-hausman 345.721 0.0000 219.562 0.0000 source: authors’ own calculations. based on the test results, h0 is rejected, and endogeneity is present. when the independent variables and the error term are correlated, panel data analysis encounters the endogeneity issue. this correlation causes bias and inconsistency in the estimates that undermine the reliability of the analysis. the endogeneity issue is a common problem in panel data analysis, and gmm is widely used in dynamic panel models to resolve it by deriving instrumental variables for endogenous variables. regression assumptions must be tested to apply regression analysis effectively. in this context, heteroskedasticity (breusch-pagan, 1979) and autocorrelation tests (arellano & bond, 1991) were conducted. the null hypotheses of the tests are established as "there is no heteroscedasticity" and "there is no autocorrelation", respectively. according to the test results, the model has heteroscedasticity but no autocorrelation. the tests’ results are given in table 6 below. table 6. preliminary tests. model 1: nrd model 2: nra test statistic probability statistic probability breusch-pagan (1979) 228.78 0.0000*** 208.12 0.0000*** arellano-bond (1991) -1.70 0.089* -1.71 0.087* source: authors’ own calculations note: ***, * represent the significance levels at %1 and %10. do natural resources hinder or help? 3.3 model specification the study focuses on the relationship between development, nrd, and nra. growth, human capital, and institutional quality are included as instrumental variables. we use two different models for comparison: 𝐻𝐻𝐻𝐻𝐻𝐻𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1𝑖𝑖𝑖𝑖𝑁𝑁𝑁𝑁𝐻𝐻 + 𝛽𝛽2𝑖𝑖𝑖𝑖𝐸𝐸𝐸𝐸 + 𝛽𝛽3𝑖𝑖𝑖𝑖𝐻𝐻𝐻𝐻 + 𝛽𝛽4𝑖𝑖𝑖𝑖𝐻𝐻𝐼𝐼 + 𝜀𝜀𝑖𝑖𝑖𝑖 [1] 𝐻𝐻𝐻𝐻𝐻𝐻𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1𝑖𝑖𝑖𝑖𝑁𝑁𝑁𝑁𝑁𝑁 + 𝛽𝛽2𝑖𝑖𝑖𝑖𝐸𝐸𝐸𝐸 + 𝛽𝛽3𝑖𝑖𝑖𝑖𝐻𝐻𝐻𝐻 + 𝛽𝛽4𝑖𝑖𝑖𝑖𝐻𝐻𝐼𝐼 + 𝜀𝜀𝑖𝑖𝑖𝑖 [2] the gmm approach used in this study was chosen to consider both the dynamic structure of the model and the potential endogeneity problem among the variables. hdi is a variable affected by past values and can exhibit bidirectional relationships between institutional quality, human capital, and natural resource indicators. therefore, fixed effects or classical panel estimators cannot provide reliable results due to endogeneity. the gmm method mitigates endogeneity by using lagged values of the variables as internal instruments; eg, hc, and iq are included as control variables rather than external instruments. furthermore, the 21 eu countries covered in the study exhibit heterogeneous structures: gmm controls for country fixed effects, limiting the distorting effects of these differences on the estimates. in practice, the differential gmm estimator suggested by arellano and bond (1991) was preferred. the primary reason for this was the moderate size of our data set (n=21, t=24) and the possibility that the system gmm could lead to over-instrumentation in small samples. the validity of the model was tested using the sargan test, which assesses the adequacy of the instruments employed, and the arellano-bond ar(2) test, which verifies the absence of second-order autocorrelation. the results confirmed the robustness of the method used. therefore, the gmm method was considered the most appropriate method for revealing the effects of natural resource dependence and abundance on development. 3.3.1 panel unit root analysis the panel unit root tests are reported separately from the preliminary diagnostics because they address a different econometric property. while the breusch-pagan lm tests focus on cross-sectional dependence in the residuals of the regression model, the cross-sectionally augmented ips (cips) test (pesaran, 2007) is specifically designed to assess stationarity of the variables under cross-sectional dependence. for clarity, we therefore present the cross-sectional dependence results first (table 4), and the unit root test results in a dedicated section (table 7). this separation avoids conflating the diagnostics of error dependence with the time-series properties of the variables. in order to perform regression analysis, the stationarity of the variables must first be tested. the series’ stationarity is crucial for the reliability of analysis results. since there is a correlation between units in the series, second-generation panel unit root tests must be performed (breitung & das, 2005). therefore, the cross-sectionally augmented im-pesaran-shin (cips) test will be used (im et al., 2003). since the cips test is an extension of the adf regression with lagged cross-sectional 174 gül and tufaner means (pesaran, 2007), cross-sectional dependence is eliminated. 𝑌𝑌𝑖𝑖𝑖𝑖 = (1 − 𝑓𝑓𝑖𝑖)𝑚𝑚𝑖𝑖 + 𝑓𝑓𝑖𝑖𝑌𝑌𝑖𝑖,𝑖𝑖−1 + 𝑢𝑢𝑖𝑖 [3] uit can be stated as follows: 𝑈𝑈𝑖𝑖𝑖𝑖 = 𝑔𝑔𝑖𝑖𝑓𝑓𝑖𝑖 + 𝑒𝑒𝑖𝑖𝑖𝑖 [4] ∆𝑌𝑌𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑖𝑖 + 𝑟𝑟𝑖𝑖𝑌𝑌𝑖𝑖𝑖𝑖−1 + 𝑔𝑔𝑖𝑖𝑓𝑓𝑖𝑖 + 𝑒𝑒𝑖𝑖𝑖𝑖 [5] the cips statistic, as the cross-sectional extended type of the ips test, is the average of the cadf statistic: 𝐻𝐻𝐻𝐻𝐶𝐶𝐶𝐶 = 𝑁𝑁−1 ∑ 𝐻𝐻𝑁𝑁𝐻𝐻𝐷𝐷𝑖𝑖𝑁𝑁 𝑖𝑖=1 [6] the h0 of the cips test is "the series is not stationary". 3.3.2 panel regression analysis economic actors are under the influence of past behaviors in their decision-making processes. therefore, when an economic model is established, the explanatory power of the model is increased by adding the lagged values of the variables to the model. dynamic panel data models account for the lagged values of the variables. the dynamic model equation is established as follows: 𝑌𝑌𝑖𝑖𝑖𝑖 = 𝛼𝛼𝑌𝑌𝑖𝑖,𝑖𝑖−1 + 𝛽𝛽𝚤𝚤𝑖𝑖́ 𝑋𝑋 + 𝜇𝜇𝑖𝑖𝑖𝑖 i = 1,…,,n; t = 1,…..,t [7] in equation 1, the scalar term 𝑎𝑎 stands for the matrices in dimensions. it is assumed that 𝜇𝜇𝑖𝑖𝑖𝑖 follows the one-way error component model (baltagi, 2021). 𝜇𝜇𝑖𝑖𝑖𝑖 = 𝑢𝑢𝑖𝑖 + 𝑣𝑣𝑖𝑖𝑖𝑖 [8] 𝑢𝑢𝑖𝑖 and 𝑣𝑣𝑖𝑖 in equation 2 are independent error terms. in equation 1, the dependent variable’s lagged value (𝑌𝑌𝑖𝑖,𝑡𝑡−1) is correlated with the error term (𝜇𝜇𝑖𝑖,𝑡𝑡−1). since 𝑌𝑌𝑖𝑖𝑡𝑡 is a function of 𝜇𝜇𝑖𝑖, 𝑌𝑌𝑖𝑖,𝑡𝑡−1 is also a function of 𝜇𝜇𝑖𝑖. hence, 𝑌𝑌𝑖𝑖,𝑡𝑡−1 is also correlated with 𝜇𝜇𝑖𝑖 (baltagi, 2021). the existence of such a correlation may lead to a deviation from the exogeneity assumption. balestra and nerlove (1966) suggested using instrumental variables to solve this problem. according to tatoglu (2012), instrumental variables are highly correlated with the variable they replace and have no correlation with the error term. when using lagged values of independent variables as instrumental variables, the estimates will be biased if a unit effect is not considered. therefore, instead of a random effects model in dynamic models, it is more appropriate to use fixed effects and first difference estimators that allow unit effects and explanatory variables to be correlated. the arellano and bond (1991) gmm uses the first difference model. the generalized least squares estimates this method by transforming the first difference model with the instrumental variable matrix. the first difference model is expressed with matrices as in equation 3: do natural resources hinder or help? 𝑍𝑍′∆𝑌𝑌 = 𝑍𝑍′∆𝑌𝑌−1𝑌𝑌𝛾𝛾 + 𝑍𝑍′∆𝑋𝑋𝛽𝛽 + 𝑍𝑍′∆𝑢𝑢 [9] the gmm estimator is given in equation 4 below: �̂�𝛿𝐺𝐺𝐺𝐺𝐺𝐺 = (∆𝑋𝑋′𝑍𝑍�𝑍𝑍′ω�𝑍𝑍�−1𝑍𝑍′∆𝑋𝑋)−1 [10] here, �̂�𝛺 is the error term’s covariance matrix. the sargan test is proposed by arellano and bond (1991) to test the validity of the instrumental variables used in gmm. the exogeneity of the instrumental variables used in the model is investigated. if exogeneity exists, the model residuals are not correlated with the independent variables. the sargan test is tested against the h0 of “overidentification restrictions are valid.” 𝑠𝑠 = ∆𝑢𝑢�𝑍𝑍(∑ 𝑧𝑧𝑖𝑖′∆𝑢𝑢�𝑖𝑖𝑁𝑁 𝑖𝑖=1 𝑢𝑢�𝑖𝑖′𝑍𝑍𝑖𝑖)−1𝑍𝑍′∆𝑢𝑢�~χ𝑝𝑝−𝐾𝐾−12 [11] in equation 5, 𝛥𝛥𝑢𝑢 ̂ represents the residuals obtained from the two-stage estimation, and p represents the number of columns of z. the test statistics are distributed as 𝜒𝜒2 in p-k-1 degrees of freedom (tatoglu, 2012). the gmm estimator should be effective without second-difference autocorrelation. arellano and bond (1991) use the residuals from a first-difference model to test for autocorrelation. the null hypothesis is “there is no autocorrelation.” 𝑠𝑠𝑎𝑎𝑎𝑎 = 𝑢𝑢�−2𝑢𝑢� 𝑢𝑢� 1 2 ~𝑁𝑁(0,1) [12] 3.3.3 panel causality analysis the causality test reveals whether there is a unidirectional or bidirectional causality relationship between variables and the direction of these relationships. the dumitrescu-hurlin (d-h) (2012) causality test was used to determine the causality relationship. the d-h (2012) test has several advantages. first, it considers the cross-sectional dependence among the countries in the dataset. second, it is not sensitive to the difference in magnitude between the time and the cross-sectional dimensions. the test can be explained with the help of the following equation: 𝑌𝑌𝑖𝑖,𝑖𝑖 = 𝛼𝛼𝑖𝑖 + ∑ 𝛾𝛾𝑖𝑖,𝑖𝑖 (𝑘𝑘)𝛾𝛾𝑖𝑖,𝑖𝑖−𝑘𝑘 + ∑ 𝛽𝛽𝑖𝑖 (𝑘𝑘)χ𝑖𝑖,𝑖𝑖−𝑘𝑘 + 𝜀𝜀𝑖𝑖,𝑖𝑖𝐾𝐾 𝑘𝑘=1 𝐾𝐾 𝑘𝑘=1 [13] k represents the lag length for all cross-sections. to test h0 and h1 in the d-h test, individual wald statistics are calculated for the cross-sections, and the wald statistics of the panel are obtained by taking the average of these statistics. 176 gül and tufaner 4. empirical findings table 7. panel unit root test results. variable t-bar cv10 cv5 cv1 z[t-bar] prob. hdi -1.668 -2.070 -2.150 -2.300 0.381 0.649 ∆hdi -3.221 -2.070 -2.150 -2.300 -6.881 0.000 nrd -2.060 -2.070 -2.150 -2.300 -1.449 0.074 ∆nrd -4.196 -2.070 -2.150 -2.300 -11.438 0.000 nra -1.759 -2.070 -2.150 -2.300 -0.043 0.483 ∆nra -2.231 -2.070 -2.150 -2.300 -2.247 0.012 eg -2.962 -2.070 -2.150 -2.300 -5.668 0.000 ∆eg -3.791 -2.070 -2.150 -2.300 -9.546 0.000 hc -2.542 -2.070 -2.150 -2.300 -3.705 0.000 ∆hc -3.639 -2.070 -2.150 -2.300 -8.834 0.000 iq -2.053 -2.070 -2.150 -2.300 -1.416 0.078 ∆iq -2.927 -2.070 -2.150 -2.300 -5.505 0.000 note: the lag length was selected as 1 based on the aic. source: authors’ own calculations table 7 above shows the cips test results. upon investigating the cips unit root test results, it becomes clear that the eg and hc variables are stationary at the level. the hdi, nrd, nra, and iq variables become stationary at their first differences are taken. table 8 shows the gmm test results for eu member countries. according to wald test results, both models are significant. the sargan test indicates that the over-identifying restrictions implied by the internal lag instruments are valid, supporting the exogeneity of the instrument set. when examining the results in table 8, a negative relationship is observed between nrd and development in eu member countries. a one-unit increase in nrd results in a 0.016-unit decrease in economic development. economies overly dependent on natural resources often suffer from resource price volatility, low investment diversity, and delays in the development of innovative sectors. many european countries, which mainly depend on resources such as oil and natural gas, may experience difficulties in achieving industrial diversification for sustainable economic growth. this relationship is the same for nra in the second model. a one-unit increase in nra decreases development by 0.0000578 units. high natural resource revenue can make other export sectors uncompetitive by appreciating the domestic currency. this can lead to shrinkage in the industrial and agricultural sectors, as well as slow economic growth, particularly for the developed economies in the eu. eg has a significant positive effect on hdi for both models. since economic growth is a key pillar of the hdi, higher growth rates will lead to an increase in the hdi. the effect of hc and iq variables on hdi is positive but insignificant. gmm results show that natural resource dependence (nrd) and abundance (nra) reduce hdi in the short term in the eu sample, whereas growth increases hdi (table 8). the negative and significant nrd coefficient indicates that price volatility, loss of competitiveness due to exchange rate appreciation, and the overshadowing of resource revenues by other sectors suppress welfare indicators in the short term. similarly, the negative relationship between increases in nra and hdi suggests that how resource revenues are channeled into public investment and sectoral diversification is a determining factor. high environmental standards and transformation costs may strengthen this effect in the eu. do natural resources hinder or help? table 8. gmm test results. dependent variable: hdi model 1: nrd model 2: nra hdi (l1) 0.0566897 (0.448) 0.3653504* (0.056) nrd -0.0162052*** (0.005) -- nra ---0.0000578** (0.025) eg 0.0006828*** (0.000) 0.0005171** (0.025) hc 3.81 (0.362) 3.64 (0.440) iq 0.0004783 (0.483) 0.0027557 (0.246) constant 0.0012061 (0.332) 0.0009407 (0.478) wald test 40.05*** (0.000) 27.45*** (0000) sargan test 0.30 (0.587) 0.06 (0.814) arellano-bond test -1.70* (0.089) -1.71* (0.087) observations 462 441 note: the lag length was selected as 2; probability values are shown in parentheses. *, ** and *** represent the significance levels of %10, %5 and %1, respectively. source: authors’ own calculations it is worth noting an apparent contradiction here. while simple correlations show a positive relationship between nra and hdi (0.386) in table 3, dynamic gmm estimations find adverse shortrun effects (-0.0000578) in table 8. this difference stems from the conceptual difference between cross-level correlations and the gmm difference estimation, which focuses on annual within-country changes. developed countries may have historically had both higher hdi and more natural capital; however, annual resource shocks and exchange rate effects can slow the progression of hdi in the short term. thus, the long-run level correlation and the short-run marginal effect may not be in the same direction. the positive and significant finding of eg in the gmm results (table 8) is an expected outcome, despite the negative appearance of the eg-hdi relationship (-0.271) in the correlation matrix (table 3). because the correlation captures both the sharp growth contractions during crisis periods and the slow-moving hdi dynamics, it can create confusion in signs. however, gmm isolates the causal short-term effect of growth on hdi by controlling for lagged hdi and contemporaneous endogeneity. this difference highlights the distinction between correlation and causality. the coefficients for hc and iq are positive but statistically insignificant (table 8). hc and iq are slow-moving indicators. since the difference between gmm removes a country’s fixed effects and utilizes annual changes, these variables remain limited in terms of within-time variance. 178 gül and tufaner table 9. causality test results. causality statistic probability nrd→hdi 1.4431 0.1490 hdi→nrd 1.2477 0.2121 nra→hdi -0.9719 0.3311 hdi→nra 3.8160 0.0001*** eg→hdi 8.8059 0.0000*** hdi→eg -1.4455 0.1483 iq→hdi 2.3463 0.0190*** hdi→iq 5.3121 0.0000*** hc→hdi 0.5157 0.0061*** hdi→hc 5.2638 0.0000*** note: the lag length was selected as 2 based on the aic. *** represents the significance level at 5%. source: authors’ own calculations according to the causality test results (see in table 9), there is a unidirectional causality from hdi to nra and from eg to hdi. a unidirectional causality from hdi to nra implies that the development level of countries impacts the discovery or development of natural resources. during development, progress in industry and technology enables natural resources to be discovered, extracted, and processed more effectively. the development increases elements such as infrastructure, capital and knowledge, allowing countries to utilize their natural resource reserves more efficiently. unidirectional causality from growth to development arises from the fact that higher growth rates enable countries to provide the necessary conditions for achieving development. causality tests do not indicate a direct causal effect of nrd/nra on hdi; thus, the negative coefficients should be interpreted as associations rather than causal effects. there are also bidirectional causalities between hdi and iq, hdi and hc. the bidirectional causality between development and institutional quality creates a sustainable development cycle. while strong institutions support development, increasing development also ensures the development of institutions. this cycle creates a significant advantage, especially for developing countries, because both economic growth and institutional structure are strengthened during the development process. the bidirectional causality between development and human capital creates a cycle that supports sustainable development. as human capital develops, economic growth accelerates; as growth increases, more investment is made in human capital. this cycle increases the competitiveness of countries and ensures long-term development. finally, the unidirectional directionality of eg→hdi and hdi→nra in the causality analysis clarifies policy channels. while inclusive growth pushes the hdi upward in the short term, the efficient use of resources increases as the level of development rises. these findings support the thesis that resource revenues should be converted into hdi through human capital and institutional quality (table 9). 5. conclusion the path of development is one of the most critical issues for all countries. in literature, numerous studies have been conducted on attaining development. a crucial factor in determining a country’s development is the availability of its natural resources. for this reason, the connection between do natural resources hinder or help? development and natural resources has become more attractive lately. this study aims to reveal the effects of natural resource dependence and natural resource abundance on development. in this context, 21 eu countries were examined from 1996 to 2019 using gmm and panel causality estimations. our results indicate that natural resource abundance can negatively impact hdi, provided it is coupled with weak institutions and low levels of human capital (corrocher & deshaires, 2020). in contrast, excessive dependence on natural resources without adequate investment in education and institutional quality is associated with weaker development outcomes (sachs & warner, 1995). economic growth, particularly when inclusive and sustained, plays a vital role in improving hdi, but the quality of growth and its distribution across society ultimately determine human progress. our findings underscore the importance of institutional frameworks that prioritize human capital development and transparent governance, particularly in resource-rich eu countries (boubaker & sghaier, 2020). although policy effects were not directly tested, our results may serve as an input for european policymakers concerned with sustainable and inclusive resource management. in this context, eu policymakers should manage natural resources more transparently and accountably, encourage the participation of local communities, and channel resource revenues to areas that support development, such as education, health, and infrastructure. in addition, investing in innovative sectors and green technologies will increase economic diversification, prevent the overexploitation of natural resources, enhance social welfare and promote sustainable development. future studies could use mixed methods to further investigate the relationship between natural resource dependency and development by applying time-varying methodologies. in particular, case studies that analyze the social, economic, and environmental impacts of natural resources on local communities would help us better understand this relationship better in longerdated periods. furthermore, comparative analyses across european countries could provide an important basis for evaluating the impacts and successes of various policies in different regions. such studies would contribute to tailoring sustainable development strategies and support better decision-making processes in natural resource management. references abraham, t. w., & ahmed, u. a. (2011). economic growth and human development index in nigeria: an error correction model approach. international journal of administration and development studies, 2(1), 239–254. acemoglu, d., & robinson, j. a. (2012). why nations fail: the origins of power, prosperity, and poverty. crown publishers. acemoglu, d., johnson, s., & robinson, j. a. (2003). an african success story: botswana. in d. rodrik (ed.), in search of prosperity: analytic narratives on economic growth (pp. 80–119). princeton university press. https://doi.org/10.1515/9781400845897-006 alexeev, m., & conrad, r. (2009). the elusive curse of oil. the review of economics and statistics, 91(3), 586–598. https://doi.org/10.1162/rest.91.3.586 https://doi.org/10.1515/9781400845897-006 https://doi.org/10.1162/rest.91.3.586 180 gül and tufaner aljarallah, r. a. (2020). the economic impacts of natural resource dependency in gulf countries. international journal of energy economics and policy, 10(6), 36–52. https://doi.org/10.32479/ijeep.9836 arellano, m., & bond, s. (1991). some tests of specification for panel data: monte carlo evidence and an application to employment equations. the review of economic studies, 58(2), 277–297. https://doi.org/10.2307/2297968 arezki, r., & brückner, m. (2011). food prices and political instability. imf working paper, 11(62). international monetary fund. https://doi.org/10.5089/9781455221066.001 awoa, p. a., ondoa, h. a., & tabi, h. n. (2024). natural resources and income inequality: economic complexity is the key. environment and development economics, 29, 127–153. https://doi.org/10.1017/s1355770x2300013x badeeb, r. a., lean, h. h., & clark, j. (2017). the evolution of the natural resource curse hypothesis: a critical literature survey. resources policy, 51, 123–134. https://doi.org/10.1016/j.resourpol.2016.10.015 balestra, p., & nerlove, m. (1966). pooling cross-section and time-series data in the estimation of a dynamic model: the demand for natural gas. econometrica, 34(3), 585–612. https://doi.org/10.2307/1909771 baltagi, b. h. (2021). econometric analysis of panel data (6th ed.). springer. https://doi.org/10.1007/9783-030-53953-5 bardhan, p. k. (2005). scarcity, conflicts, and cooperation: essays in the political and institutional economics of development. mit press. barro, r. j., & lee, j. w. (2013). a new data set of educational attainment in the world, 1950–2010. journal of development economics, 104, 184–198. https://doi.org/10.1016/j.jdeveco.2012.10.001 becker, g. s. (1962). investment in human capital: a theoretical analysis. journal of political economy, 70(5), 9–49. https://doi.org/10.1086/258724 blundell, r., & bond, s. (1998). initial conditions and moment restrictions in dynamic panel data models. journal of econometrics, 87(1), 115–143. https://doi.org/10.1016/s0304-4076(98)00009-8 boschini, a. d., pettersson, j., & roine, j. (2013). the resource curse and its potential reversal. world development, 43, 19–41. https://doi.org/10.1016/j.worlddev.2012.10.007 boubaker, s., & sghaier, a. (2020). institutional quality and economic growth: evidence from european countries. international journal of economics and finance, 12(2), 64–76. breitung, j., & das, s. (2005). panel unit root tests under cross-sectional dependence. statistica neerlandica, 59(4), 414–433. https://doi.org/10.1111/j.1467-9574.2005.00299.x breusch, t. s., & pagan, a. (1979). a simple test for heteroscedasticity and random coefficient variation. econometrica, 47(5), 1287–1294. https://doi.org/10.2307/1911963 breusch, t. s., & pagan, a. r. (1980). the lagrange multiplier test and its applications to model specification in econometrics. the review of economic studies, 47(1), 239–253. https://doi.org/10.2307/2297111 bulte, e. h., damania, r., & deacon, r. t. (2005). resource intensity, institutions, and development. world development, 33(7), 1029–1044. https://doi.org/10.1016/j.worlddev.2005.04.004 cappelen, a., & mjøset, l. (2009). can norway be a role model for natural resource abundance? wider working paper, 2009(23). unu-wider. corden, w. m., & neary, j. p. (1982). booming sector and de-industrialization in a small open economy. the economic journal, 92(368), 825–848. https://doi.org/10.2307/2232670 https://doi.org/10.32479/ijeep.9836 https://doi.org/10.2307/2297968 https://doi.org/10.5089/9781455221066.001 https://doi.org/10.1017/s1355770x2300013x https://doi.org/10.1016/j.resourpol.2016.10.015 https://doi.org/10.2307/1909771 https://doi.org/10.1007/978-3-030-53953-5 https://doi.org/10.1007/978-3-030-53953-5 https://doi.org/10.1016/j.jdeveco.2012.10.001 https://doi.org/10.1086/258724 https://doi.org/10.1016/s0304-4076(98)00009-8 https://doi.org/10.1016/j.worlddev.2012.10.007 https://doi.org/10.1111/j.1467-9574.2005.00299.x https://doi.org/10.2307/1911963 https://doi.org/10.2307/2297111 https://doi.org/10.1016/j.worlddev.2005.04.004 https://doi.org/10.2307/2232670 do natural resources hinder or help? corrales, j., & penfold, m. (2015). dragon in the tropics: venezuela and the legacy of hugo chávez (2nd ed.). brookings institution press. https://doi.org/10.5040/9780815750956 cust, j., & mihalyi, d. (2017). evidence for a presource curse? oil discoveries, elevated expectations, and growth disappointments. oxcarre working paper, 193. oxford centre for the analysis of resource rich countries. dabla-norris, e., ji, y., townsend, r. m., & filiz unsal, d. (2015). distinguishing constraints of financial inclusion and their impact on gdp and inequality (nber working paper no. 20821). national bureau of economic research. https://doi.org/10.3386/w20821 debonheur, k. d., desire, a., & ouedraogo, i. (2023). the effect of natural resources rents on human development in selected african countries. natural resources forum, 48(3), 803–837. https://doi.org/10.1111/1477-8947.12341 destek, m. a., aydın, s., & destek, g. (2022). investigating an optimal resource dependency to prevent natural resource curse: evidence from countries with the curse risk. resources policy, 79, 102981. https://doi.org/10.1016/j.resourpol.2022.102981 dumitrescu, e.-i., & hurlin, c. (2012). testing for granger non-causality in heterogeneous panels. economic modelling, 29(4), 1450–1460. https://doi.org/10.1016/j.econmod.2012.02.014 durbin, j. (1954). errors in variables. review of the international statistical institute, 22(1–3), 23–32. https://doi.org/10.2307/1401917 esanov, a., raiser, m., & buiter, w. (2001). nature’s blessing or nature’s curse: the political economy of transition in resource-based economies (zef discussion paper no. 81). university of bonn, center for development research (zef). eurostat. (2022). human development index across eu countries. european union. farhadi, m., islam, m. r., & moslehi, s. (2015). economic freedom and productivity growth in resource-rich economies. world development, 72, 109–126. https://doi.org/10.1016/j.worlddev.2015.02.014 farooq, u., tabash, m. i., & al-faryan, m. a. s. (2025). natural resources and economic growth in asia: the moderating role of governance. transnational corporations review. advance online publication. https://doi.org/10.1016/j.tncr.2025.200143 feenstra, r. c., inklaar, r., & timmer, m. p. (2015). the next generation of the penn world table. american economic review, 105(10), 3150–3182. https://doi.org/10.1257/aer.20130954 fukuda-parr, s. (2003). the human development paradigm: operationalizing sen’s ideas on capabilities. feminist economics, 9(2–3), 301–317. https://doi.org/10.1080/1354570022000077980 gelb, a. (1988). oil windfalls: blessing or curse? oxford university press. gylfason, t. (2001). natural resources, education, and economic development. european economic review, 45(4–6), 847–859. https://doi.org/10.1016/s0014-2921(01)00127-1 hanushek, e. a., & woessmann, l. (2021). education and economic growth. oxford research encyclopedia of economics and finance. https://doi.org/10.1093/acrefore/9780190625979.013.651 haq, m. ul. (1995). reflections on human development. oxford university press. https://doi.org/10.1093/oso/9780195101911.001.0001 hashmat, a., ghouse, g., & ahmad, n. (2023). impact of economic and environmental dynamics on human development: an analysis of hdi classifications. bulletin of business and economics, 12(3), 738– 751. https://doi.org/10.61506/01.00112 hausman, j. a. (1978). specification tests in econometrics. econometrica, 46(6), 1251–1271. https://doi.org/10.2307/1913827 holden, s. (2013). avoiding the resource curse: the case of norway. energy policy, 63, 870–876. https://doi.org/10.1016/j.enpol.2013.09.010 https://doi.org/10.5040/9780815750956 https://doi.org/10.3386/w20821 https://doi.org/10.1111/1477-8947.12341 https://doi.org/10.1016/j.resourpol.2022.102981 https://doi.org/10.1016/j.econmod.2012.02.014 https://doi.org/10.2307/1401917 https://doi.org/10.1016/j.worlddev.2015.02.014 https://doi.org/10.1016/j.tncr.2025.200143 https://doi.org/10.1257/aer.20130954 https://doi.org/10.1080/1354570022000077980 https://doi.org/10.1016/s0014-2921(01)00127-1 https://doi.org/10.1093/acrefore/9780190625979.013.651 https://doi.org/10.1093/oso/9780195101911.001.0001 https://doi.org/10.61506/01.00112 https://doi.org/10.2307/1913827 https://doi.org/10.1016/j.enpol.2013.09.010 182 gül and tufaner hong, j. y. (2017). does oil hinder social spending? evidence from dictatorships, 1972–2008. studies in comparative international development, 52(4), 457–482. https://doi.org/10.1007/s12116-0179237-y hsiao, c. (2014). analysis of panel data (3rd ed.). cambridge university press. https://doi.org/10.1017/cbo9781139839327 im, k. s., pesaran, m. h., & shin, y. (2003). testing for unit roots in heterogeneous panels. journal of econometrics, 115(1), 53–74. https://doi.org/10.1016/s0304-4076(03)00092-7 karl, t. l. (1997). the paradox of plenty: oil booms and petro-states. university of california press. https://doi.org/10.1525/9780520918696 khan, y., liu, f., & hassan, t. (2023). natural resources and sustainable development: evaluating the role of remittances and energy resources efficiency. resources policy, 80, 103214. https://doi.org/10.1016/j.resourpol.2022.103214 kim, d.-h., chen, t.-c., & lin, s.-c. (2020). does oil drive income inequality? new panel evidence. structural change and economic dynamics, 55, 137–152. https://doi.org/10.1016/j.strueco.2020.08.002 li, y., pang, d., & cifuentes-faura, j. (2023). time-varying linkages among financial development, natural resources utility, and globalization for economic recovery in china. resources policy, 82, 103498. https://doi.org/10.1016/j.resourpol.2023.103498 matsen, e., & torvik, r. (2005). optimal dutch disease. journal of development economics, 78(2), 494–515. https://doi.org/10.1016/j.jdeveco.2004.09.003 mavratos, g., murshed, s. m., & torres, s. (2011). natural resource dependence and economic performance in the 1970–2000 period. review of development economics, 15(1), 124–138. https://doi.org/10.1111/j.1467-9361.2010.00597.x mehlum, h., moene, k., & torvik, r. (2006). institutions and the resource curse. the economic journal, 116(508), 1–20. https://doi.org/10.1111/j.1468-0297.2006.01045.x mohamed, e. s. e. (2020). resource rents, human development and economic growth in sudan. economies, 8(4), article 99. https://doi.org/10.3390/economies8040099 mousavi, a., & clark, j. (2021). the effects of natural resources on human capital accumulation: a literature survey. journal of economic surveys, 35(4), 1073–1117. https://doi.org/10.1111/joes.12441 papyrakis, e., & gerlagh, r. (2007). resource abundance and economic growth in the united states. european economic review, 51(4), 1011–1039. https://doi.org/10.1016/j.euroecorev.2006.04.001 pesaran, m. h. (2007). a simple panel unit root test in the presence of cross-section dependence. journal of applied econometrics, 22(2), 265–312. https://doi.org/10.1002/jae.951 pesaran, m. h., ullah, a., & yamagata, t. (2008). a bias-adjusted lm test of error cross-section independence. the econometrics journal, 11(1), 105–127. https://doi.org/10.1111/j.1368423x.2007.00227.x petermann, a., guzmán, j. i., & tilton, j. e. (2007). mining and corruption. resources policy, 32(2), 91–103. https://doi.org/10.1016/j.resourpol.2007.08.003 rădulescu, m., simionescu, m., kartal, m. t., mohammed, k. s., & balsalobre-lorente, d. (2025). the impact of human capital, natural resources, and renewable energy on achieving sustainable cities and communities in european union countries. sustainability, 17(5), 2237. https://doi.org/10.3390/su17052237 https://doi.org/10.1007/s12116-017-9237-y https://doi.org/10.1007/s12116-017-9237-y https://doi.org/10.1017/cbo9781139839327 https://doi.org/10.1016/s0304-4076(03)00092-7 https://doi.org/10.1525/9780520918696 https://doi.org/10.1016/j.resourpol.2022.103214 https://doi.org/10.1016/j.strueco.2020.08.002 https://doi.org/10.1016/j.resourpol.2023.103498 https://doi.org/10.1016/j.jdeveco.2004.09.003 https://doi.org/10.1111/j.1467-9361.2010.00597.x https://doi.org/10.1111/j.1468-0297.2006.01045.x https://doi.org/10.3390/economies8040099 https://doi.org/10.1111/joes.12441 https://doi.org/10.1016/j.euroecorev.2006.04.001 https://doi.org/10.1002/jae.951 https://doi.org/10.1111/j.1368-423x.2007.00227.x https://doi.org/10.1111/j.1368-423x.2007.00227.x https://doi.org/10.1016/j.resourpol.2007.08.003 https://doi.org/10.3390/su17052237 do natural resources hinder or help? radulović, m. (2020). the impact of institutional quality on economic growth: a comparative analysis of the eu and non-eu countries of southeast europe. economic annals, 65(225), 163–182. https://doi.org/10.2298/eka2025163r ranis, g., stewart, f., & ramirez, a. (2000). economic growth and human development. world development, 28(2), 197–219. https://doi.org/10.1016/s0305-750x(99)00131-x rodrik, d., subramanian, a., & trebbi, f. (2004). institutions rule: the primacy of institutions over geography and integration in economic development. journal of economic growth, 9(2), 131–165. https://doi.org/10.1023/b:joeg.0000031425.72248.85 ross, m. l. (2012). the oil curse: how petroleum wealth shapes the development of nations. princeton university press. https://doi.org/10.1515/9781400841929 sachs, j. d., & warner, a. m. (1995). natural resource abundance and economic growth. nber working paper, 5398. https://doi.org/10.3386/w5398 sachs, j. d., & warner, a. m. (2001). the curse of natural resources. european economic review, 45(4–6), 827–838. https://doi.org/10.1016/s0014-2921(01)00125-8 salahojaev, r., djalilov, b., bakieva, i., & kobiljonov, i. (2024). the relationship between natural resource abundance and human development in belt and road initiative countries: the role of financial development. international journal of energy economics and policy, 14(1), 45–52. https://doi.org/10.32479/ijeep.14494 sala-i-martin, x., & subramanian, a. (2013). addressing the natural resource curse: an illustration from nigeria. journal of african economies, 22(4), 570–615. https://doi.org/10.1093/jae/ejs033 saud, s., haseeb, a., zafar, m. w., & li, h. (2023). articulating natural resource abundance, economic complexity, education and environmental sustainability in mena countries: evidence from advanced panel estimation. resources policy, 80, 103261. https://doi.org/10.1016/j.resourpol.2022.103261 schultz, t. w. (1961). investment in human capital. the american economic review, 51(1), 1–17. sen, a. (1999). development as freedom. oxford university press. simionescu, m., rădulescu, m., & cifuentes-faura, j. (2024). the role of natural resources rents and egovernment in achieving sustainable development in the european union. resources policy, 97, 105278. https://doi.org/10.1016/j.resourpol.2024.105278 singh, s., sharma, g. d., rădulescu, m., balsalobre-lorente, d., & bansal, p. (2024). do natural resources impact economic growth? an investigation of p5+1 countries under sustainable management. geoscience frontiers, 15(3), 101595. https://doi.org/10.1016/j.gsf.2023.101595 stiglitz, j. e. (2012). the price of inequality: how today’s divided society endangers our future. w. w. norton. stijns, j.-p. c. (2005). natural resource abundance and economic growth revisited. resources policy, 30(2), 107–130. https://doi.org/10.1016/j.resourpol.2005.05.001 tatoglu, f. y. (2012). i̇leri panel veri analizi: stata uygulamalı. beta yayınları. timbe, g., flavio, l. p., bam, w., hartmann, d., & de bruyne, k. (2024). is natural resource abundance a curse or an opportunity? economic complexity, fdi, and industrial policies in mozambique. resources policy, 98, 105326. https://doi.org/10.1016/j.resourpol.2024.105326 torvik, r. (2009). why do some resource-abundant countries succeed while other do not? oxford review of economic policy, 25(2), 241–256. https://doi.org/10.1093/oxrep/grp015 transparency international. (2023). corruption perceptions index 2022. united nations development programme (undp). (1990). human development report 1990. oxford university press. https://doi.org/10.2298/eka2025163r https://doi.org/10.1016/s0305-750x(99)00131-x https://doi.org/10.1023/b:joeg.0000031425.72248.85 https://doi.org/10.1515/9781400841929 https://doi.org/10.3386/w5398 https://doi.org/10.1016/s0014-2921(01)00125-8 https://doi.org/10.32479/ijeep.14494 https://doi.org/10.1093/jae/ejs033 https://doi.org/10.1016/j.resourpol.2022.103261 https://doi.org/10.1016/j.resourpol.2024.105278 https://doi.org/10.1016/j.gsf.2023.101595 https://doi.org/10.1016/j.resourpol.2005.05.001 https://doi.org/10.1016/j.resourpol.2024.105326 https://doi.org/10.1093/oxrep/grp015 184 gül and tufaner united nations development programme (undp). (2015). transforming our world: the 2030 agenda for sustainable development. united nations. united nations development programme (undp). (2024). human development report 2023–2024. van der ploeg, f., & poelhekke, s. (2011). volatility and the natural resource curse. oxford economic papers, 61(4), 727–760. https://doi.org/10.1093/oep/gpp027 van der ploeg, f., & poelhekke, s. (2017). the impact of natural resources: survey of recent quantitative evidence. the journal of development studies, 53(2), 205–216. https://doi.org/10.1080/00220388.2016.1160069 world bank. (2021). the changing wealth of nations 2021: managing assets for the future. https://www.worldbank.org/en/publication/changing-wealth-of-nations/data world bank. (2024). world development indicators. https://databank.worldbank.org/source/worlddevelopment-indicators wu, d. m. (1973). alternative tests of independence between stochastic regressors and disturbances. econometrica, 41(4), 733–750. https://doi.org/10.2307/1914093 zhang, j., muhammad, t., dai, w., khan, q. r., & ahmad, m. (2024). how does the resource curse influence economic performance? exploring the role of natural resource rents and renewable energy consumption in south asia. sustainability, 16(11), 11138. https://doi.org/10.3390/su162411138 zhang, x., xing, l., & fan, s. (2008). resource abundance and regional development in china. economics of transition, 16(1), 7–29. https://doi.org/10.1111/j.1468-0351.2007.00318.x https://doi.org/10.1093/oep/gpp027 https://doi.org/10.1080/00220388.2016.1160069 https://www.worldbank.org/en/publication/changing-wealth-of-nations/data https://databank.worldbank.org/source/world-development-indicators https://databank.worldbank.org/source/world-development-indicators https://doi.org/10.2307/1914093 https://doi.org/10.3390/su162411138 https://doi.org/10.1111/j.1468-0351.2007.00318.x 1. introduction 2. literature review 3. data and methodology 4. empirical findings 5. conclusion references has covid-19 changed the correlation between cryptocurrencies and stock markets? © the author(s) 2023. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 12, no. 2 (2023), pages 139-156 https://doi.org/10.17979/ejge.2023.12.2.9960 submitted: oct 5, 2023 accepted: nov 14, 2023 published: dec 5, 2023 article has covid-19 changed the correlation between cryptocurrencies and stock markets? ines abdelkafi,1,* youssra ben romdhane,2 sahar loukil 3 1 uramef, esc, university of sfax, tunisia 2 led, fseg, university of sfax, tunisia 3 artige, fseg, university of sfax, tunisia *correspondence: ines.abdelkafi@escs.usf.tn abstract. the covid-19 pandemic has challenged the notion that cryptocurrencies are uncorrelated with traditional asset markets. this study uses var-ols techniques to investigate the time-varying correlation between bitcoin and three major european stock market indices from january 4, 2016, to february 26, 2021. our results show that cryptocurrencies and stock markets are dependent during crisis periods, but not during non-crisis periods. this confirms the time-varying correlation between cryptocurrencies and stock markets, which depends on the extent and persistence of responses to own and cross shocks. to improve the robustness of our results, we also test the impact of government measures on bitcoin and stock market indices, and find that they are both affected by these measures. our study adds to the literature by examining the impacts of pandemics on the correlations between bitcoin returns and the stock market, oil, and gold index returns, which have so far been unaddressed. keywords: covid-19; financial markets; bitcoin; stock indices jel classification:i1; d53; g15; g12; c22 1. introduction the concept of safe haven value for investment is driven by investor loss aversion (tversky & kahneman, 1991), where investors are more concerned about avoiding losses than the associated potential gains (hwang & satchell, 2010). this loss aversion encourages investors to seek safe haven assets, i.e. assets that are not correlated or correlated negatively with traditional assets in times of market turbulence (baur & lucey, 2010). various safe haven assets have been established at short to medium horizons, including gold (bredin et al., 2015), currency (ranaldo & söderlind, 2010), longterm treasury bills (flavin et al., 2014) and, more recently, cryptocurrencies. the role of alternative investments in improving the returns of traditional equity bond portfolios has long been the subject of academic research. given the shortage of “traditional” alternative investment classes, the role of the increasingly important cryptocurrency markets becomes relevant. cryptocurrency is a digital or https://creativecommons.org/licenses/by-nc/4.0/ 140 abdelkafi et al. virtual currency that is exchanged between peers without the need for a third party. since the introduction of bitcoin in 2008, academic research has highlighted the low correlation between bitcoin and traditional financial markets (baur et al., 2018; corbet et al., 2018). this correlation, however, became stronger after the introduction of futures contracts on bitcoin in december 2017 (matkovskyy & jalan, 2019; sami & abdallah, 2021). this discovery led to a more indepth investigation into the hedging and diversification properties of bitcoin compared to traditional financial assets (urquhart & zhang, 2019; guesmi et al., 2018; bouri et al., 2017; cretarola et al., 2021). other authors suggest that the assets become more correlated during economic downturns. thus, many studies have explored the response of cryptocurrency markets to the covid-19 pandemic as well as changes in interactions between cryptocurrencies and other traditional asset classes (lahmiri & bekiros, 2020; conlon & mcgee, 2020; mnif et al., 2020; ali et al., 2020; goodell & goutte, 2021; ji et al., 2020). indeed, the covid-19 global health crisis has the potential to slow down the global economy and increase the level of volatility in financial markets. baker et al. (2020) have explained the stronger impact of the covid-19 pandemic on equity markets than previous outbreaks of infectious diseases. this crisis is an opportunity to study much more about the evolution of the crypto-currency market as well as the interdependencies between the crypto-currency market and the traditional asset classes. the objective of this article is to examine the correlation between crypto-currencies and stock markets. we study the impact of the introduction of bitcoin into investors' portfolios on their efficiency in the context of the covid-19 pandemic. the remainder of this document is organized as follows. section 2 presents the literature review. section 3 describes the methodology. section 4 examines the empirical results. section 5 is devoted to discussion and the final section presents the main conclusions. 2. literature review the usefulness of alternative investments is confirmed in the literature which documents that a planned asset allocation between various super form asset classes systematically involves both a strategic commodity portfolio and a full equity portfolio (conover et al., 2010; ciner et al., 2013; li & lucey, 2017; gao & nardari, 2018) recent studies suggest that there now appears to be a shortage of alternative assets that can be used to reduce the risk of declining equity investments (bouri et al., 2019; shahzad et al., 2019). regarding portfolio diversification with cryptocurrencies, chen and vivek (2014) show that bitcoin can play an important role in improving the efficiency of the portfolio. based on an analysis of traditional assets and alternative investments, bouri et al. (2017) show that investing in bitcoin offers significant diversification benefits. bouri et al. (2019) document that hedging stocks with cryptocurrencies is beneficial. inci and lagasse (2019) explain the role of crypto-currencies in improving investment portfolios. they found that private and listed companies recorded huge profits in these assets. they confirmed that ownership of crypto-currencies varies between companies, ranging from investment objectives to supporting future plans to accept digital currencies as a means of payment for goods and services. kajtazi and moro (2019) demonstrate that the addition of bitcoin, has covid-19 changed the correlation between cryptocurrencies and stock markets? despite its speculative characteristics, results in an improvement in the performance of asset portfolios in the us, china and europe. chan et al. (2019) suggested strong bitcoin hedging properties against five international stock indexes, including the s&p 500. urquhart and zhang (2019) investigate the hedging, diversification and value-safe-haven properties of bitcoin against global currency fluctuations. their results support the assumption that alternative investments, in this case cryptocurrencies, add value by improving the performance of traditional financial assets. platanakis et al. (2019) suggest that investors should include bitcoin in their portfolios as it generates significantly higher risk-adjusted returns. in line with this thinking, frankovic et al. (2021) analysed the relationship between the share prices of australian companies holding crypto-currencies and their prices. they found that these companies adopt positions that are sensitive to fluctuations in crypto-currency prices. similarly, xu et al. (2022) examined the strategic reasons for crypto-currency integration by corporate treasury departments and explored the risk-return outcomes of these decisions and strategies. as a result, they confirmed in their study that the emergence of jumps in crypto-currencies increases the probability of jumps in the returns of said us companies. empirical studies have also emerged to explore the response of cryptocurrency markets to the covid-19 pandemic as well as changes in interactions between cryptocurrencies and other traditional assets. in the us context, sharif et al. (2020) examined the reaction of crypto-currencies during the period of the health crisis. they found that the correlation between price volatility shocks on economic policy uncertainty and the oil market is also related to the spread of the covid-19 pandemic in the us. they confirmed that the health crisis is a geopolitical threat. these results are confirmed by the study by umar and gubareva (2020). al-awadhi et al. (2020) found evidence of a significant negative impact of covid-19 on the equity returns of all companies included in the hang seng index and the shanghai composite index. these results are confirmed with studies by sharif et al. (2020) and zhang et al. (2020). lahmiri and bekiros (2020) suggest that the cryptocurrency market has been relatively more volatile than international equity markets during the covid-19 pandemic. the empirical results show the variable relationship over time between the cryptocurrency market and the us stock market or the price of the gold market. recent data has shown that there has been a positive relationship that has varied over time between these two markets since covid-19. mishra et al. (2022) demonstrated a reduction in market linkage during a recession as compared to expansion. hung (2019) suggested that portfolio managers need to adjust their asset allocations in times of turbulence or crisis when asset volatility shifts from one market to another. in this sense, kim et al. (2020) examined the relationship of major financial assets, bitcoin, gold and s&p 500 with garch models. they show the relationship of bitcoin with gold and s&p 500. they also analyzed the relationships between the conditional correlation varying over time with the volatility of bitcoin and the volatility of the s&p 500 by a marginal regression of the gaussian copula (gcmr). the empirical results show that the s&p 500 and gold prices are statistically significant for bitcoin in terms of log-back and volatility. 142 abdelkafi et al. 3. methods we examine the nexus between cryptocurrencies and stock markets before and after the covid-19 pandemic. our study retrieved global daily-frequency data from 4 january 2016 to 26 february 2021. we divide our sample to before covid-19 and covid-19 period around december 31, 2019 which corresponds to the date the first cases were discovered by chinese authorities. our choice is justified since investors are more sensitive to negative information. data on cryptocurrencies and stock markets were collected respectively from coinbase, retrieved from federal reserve economic database, and the data of oil and gold price are extracted from dtastream. the data of governmental measures are extracted from oxford data base. these governmental measures are used herein as a proxy for reducing actions of global pandemic uncertainty. moreover, we calculate the continuous compounded daily returns for all the series as 100 × ln ( pt pt−1 ), where pt represents the daily closing price of each asset. descriptive statistics results clearly show that during the covid-19 outbreak, all indices return downturn dramatically. the mean returns of all stock market indices are very small, among which dax30 is positive, while cac40 and ftsemib turn out to be negative with a significant increase in volatility. as expected, the bitcoin has high mean returns and standard deviations (std. dev.) this result suggests its potential hedge role to investors (kajtazi & moro, 2019; kim et al., 2020). besides, we can easily notice the free fall in oil price return and the remarkable increase in volatility. finally, according to the results, gold’s return slightly decreased in crisis period amid a slight decrease in volatility. thus, we confirm the previous findings of bouri et al. (2020), who suggests the hedging role of gold, especially in mitigated periods. for government measures, we can easily notice that all countries are seriously implementing controlling the disease with a high index (at least 79%) and that the most applied measures are school closure, workplace closures, cancel public events, testing policy and contract tracing. on the other hand, stay at home, and restrictions on internal movement are comparatively less applied. this result could be explained by the difficulty of forcing people to stay at home and outlaw movements since it is socially difficult to accept and harmful for economic activities. the first confinement is marked by a drop in wages. france has a drop in wages of the poorest among the lowest in europe. its decline is lower than in germany. this impact is particularly present in local businesses that suffer from this effect: confinement causes a considerable decrease in the number of people who use the shops (restaurants, small markets, bakeries, tourism, etc.). table 1. descriptive statistics for bitcoin and stock indices before and during covid-19 source: authors calculations before covid-19 during covid-19 variable obs mean std. dev. min max obs mean std. dev. min max lnbitcoin 1028 0.0027301 0.0462737 -0.247405 0.221747 297 0.0062591 0.0520257 -0.497278 0.193671 dax30 1028 1.000292 0.0097051 0.931767 1.03506 297 0.0001463 0.0195434 -0.130549 0.104143 lncac40 1028 0.0002763 0.0095263 -0.083844 0.040604 297 -0.0001585 0.0193948 -0.130983 0.080561 lnftsemib 1028 0.0001299 0.0128752 -0.133314 0.049111 297 -0.000109 0.0212854 -0.185411 0.085495 lngold 1028 0.0005163 0.02189 -0.082336 0.136944 297 0.0004485 0.0121427 -0.058928 0.042968 lnwti 1028 1.000376 0.007489 0.968084 1.04022 297 0.0000159 0.064684 -0.601676 0.319634 has covid-19 changed the correlation between cryptocurrencies and stock markets? table 2. variables measurement and descriptive statistics for governmental responses germany france italy variable obs mean std. dev. min max mean std. dev. min max mean std. dev. min max lnschoolcl~g 294 .811 .404 0 2 .791 .412 0 2 .817 .386 0 1 lnworkplace 294 .828 .381 0 1.5 .846 .374 0 2 .898 .327 0 2 lncancelev~s 294 .874 .342 0 2 .874 .350 0 2 .891 .312 0 1 lnstayhome 294 .495 .520 0 2 .573 .501 0 2 .686 .473 0 2 lnmouvements 294 .760 .450 0 2 .597 .504 0 2 .566 .509 0 2 lntestingp~s 294 .950 .259 0 3 .956 .226 0 2 .935 .260 0 2 lncontactt~g 294 .962 .203 0 2 .954 .223 0 2 .932 .252 0 1 δcumcases 294 .0861 .322 0 4 .099 .410 0 5 .086 .259 0 2.575 source: authors calculations according to insee, “73% of companies report a decrease in their sales of more than 10%. and 35%, a decrease of more than 50% during the trigger period.” to get more detailed results, we present below the time series of each variable in figure 1. ftsemib dax30 bitcoin cac40 wti figure 1. cac40, dax30, wti and bitcoin return over the whole period. source: own elaboration. 144 abdelkafi et al. the trends of cac 40, dax30 and ftse mib stock market indices and prices of bitcoin, oil and gold are similarly affected by covid-19 in the first quarter of 2020 with clearly persistent high volatility for stock indices, gold and bitcoin, while oil return almost rejoin it trend as before covid-19 period. concretely, during the outbreak of covid-19 from 1rst january 2020 to 17 september 2020, the prices of cryptocurrencies and the stock indices of the world fell sharply first and then rose moderately. all series show a downward trend. 4. results 4.1 pearson matrix table 3 presents the pearson’s correlations between bitcoin return, stock market, oil and gold indices returns. table 3. pearson matrix before covid-19 during covid-19 lnbitc~n dax30 lncac40 lnftse~b gold lnwti lnbitc~n dax30 lncac40 lnftse~b gold lnwti lnbitcoin 1 1 dax30 0.004 1 0.390* 1 0.903 0.000 lncac40 -0.003 0.917* 1 0.384* 0.957* 1 0.931 0 0 0 lnftsemib -0.008 0.792* 0.829* 1 0.491* 0.908* 0.914* 1 0.795 0 0 0 0 0 gold 0.049 -0.268* -0.272* -0.240* 1 0.262* 0.163* 0.108 0.117* 1 0.118 0 0 0 0 0.005 0.062 0.043 lnwti -0.011 0.254* 0.302* 0.311* -0.034 1 0.161* 0.285* 0.292* 0.283* 0.074 1 0.730 0 0 0 0.264 0.005 0 0 0 0.202 note: * p-value <0.05. source: authors calculations. before covid-19, it can be easily observed that in addition to the weak and positive correlations between the stock market returns of the three countries and bitcoin, there is a positive correlation between bitcoin and gold, proving their similarity. therefore, we employ the var approach to study the heterogeneous relationships between these variables. next, we check the robustness of our results and test whether bitcoin and stock market indices are similarly affected by government actions using a gls regression. 4.1 var model we intend to figure out the existence of a potential link between cryptocurrencies and stock returns before and during covid-19 period through var model. the lagged bitcoin returns, stock returns, and oil and gold are used as exogenous factors to detect the mean spillover effect within the studied indices. the stationary tests of adf and pp with the null hypothesis of having a unit root confirm that all indices are stationary at first level as indicated in table 4. the lags orders are selected according to fpe, aic, hqic and sbic criteria allowed in tables 5, 6. and 7. has covid-19 changed the correlation between cryptocurrencies and stock markets? table 4. stationary tests before covid-19 during covid-19 stationary level i(0) stationary at first difference i(1) stationary level i(0) stationary at first difference i(1) adf test pp test adf test pp test adf test pp test adf test pp test case 1: model without trend case 1: model without trend case 1:model without trend case 1: model without trend ftse mib -1.934 -1.934 -13.942 *** -29.949*** -2.08 -2.08 -13.941*** -29.949*** dax30 -1.720 -1.720 -14.777*** -27.550*** -2.08 -2.09 -14.765*** -27.536*** cac40 -1.518 -1.518 -14.625*** -25.483*** -1.67 -1.67 -25.489*** -25.489*** bitcoin 2.990 2.990 -15.949*** -29.233*** 2.20 2.20 -29.429*** -29.429*** gold -1.698 -1.698 -30.922*** -30.922*** -1.86 -1.86 -30.957*** -30.957*** wti -2.851 -2.851 -29.718*** -29.718*** -2.84 -2.84 -29.712*** -29.712*** note: *** statistical significance at 1%. source: authors calculations table 5. selection-order criteria before and after covid-19 for italy selection-order criteria before covid lag ll lr df p fpe aic hqic sbic 0 2049.12 8.2e-15* -21.0837* -21.0564* -21.0163* 1 2063.69 29.157 16 0.023 8.3e-15 -21.069 -20.9326 -20.7321 2 2080.01 32.637 16 0.008 8.3e-15 -21.0723 -20.8267 -20.4659 3 2091.51 22.995 16 0.114 8.7e-15 -21.0259 -20.6712 -20.15 4 2109.72 36.422* 16 0.003 8.5e-15 -21.0487 -20.5849 -19.9032 selection-order criteria during covid-19 lag ll lr df p fpe aic hqic sbic 0 535.012 4.8e-14 -19.3095 -19.2531 -19.1635* 1 563.308 56.592 16 0.000 3.1e-14 -19.7566 -19.4744* -19.0267 2 574.879 23.143 16 0.110 3.7e-14 -19.5956 -19.0875 -18.2817 3 602.427 55.095 16 0.000 2.5e-14 -20.0155 -19.2816 -18.1177 4 619.753 34.652* 16 0.004 2.5e-14* -20.0637* -19.104 -17.5819 source: authors calculations table 6. selection-order criteria before and after covid-19 for france selection-order criteria before covid lag ll lr df p fpe aic hqic sbic 0 2113.24 4.2e-15* -21.7447* -21.7175* -21.6774* 1 2123.72 20.954 16 0.180 4.5e-15 -21.6878 -21.5514 -21.3509 2 2138.02 28.605 16 0.027 4.6e-15 -21.6703 -21.4248 -21.0639 3 2147.04 18.045 16 0.321 4.9e-15 -21.5984 -21.2437 -20.7225 4 2163.06 32.037* 16 0.010 4.9e-15 -21.5986 -21.1347 -20.4531 selection-order criteria during covid-19 lag ll lr df p fpe aic hqic sbic 0 529.101 6.0e-14 -19.0946 -19.0381 -18.9486* 1 554.908 51.615 16 0.000 4.2e-14 -19.4512 -19.1689* -18.7213 2 570.998 32.18 16 0.009 4.2e-14 -19.4545 -18.9464 -18.1406 3 596.131 50.266 16 0.000 3.1e-14* -19.7866 -19.0527 -17.8887 4 612.474 32.686* 16 0.008 3.2e-14 -19.799* -18.8393 -17.3173 source: authors calculations 146 abdelkafi et al. table 7. selection-order criteria before and after covid-19 for germany selection-order criteria before covid-19 lag ll lr df p fpe aic hqic sbic 0 2120.86 3.9e-15* -21.8233* -21.796* -21.7559* 1 2131.18 20.634 16 0.193 4.1e-15 -21.7647 -21.6283 -21.4278 2 2145.85 29.346 16 0.022 4.2e-15 -21.751 -21.5055 -21.1446 3 2154.89 18.073 16 0.320 4.5e-15 -21.6792 -21.3245 -20.8033 4 2170.5 31.24* 16 0.013 4.5e-15 -21.6753 -21.2115 -20.5299 selection-order criteria during covid-19 lag ll lr df p fpe aic hqic sbic 0 535.6 4.7e-14 -19.3309 -19.2744 -19.1849* 1 560.191 49.183 16 0.000 3.5e-14 -19.6433 -19.361* -18.9134 2 574.783 29.184 16 0.023 3.7e-14 -19.5921 -19.084 -18.2782 3 598.892 48.218 16 0.000 2.8e-14* -19.887* -19.1531 -17.9891 4 612.624 27.465* 16 0.037 3.2e-14 -19.8045 -18.8448 -17.3227 source: authors calculations then, the granger causality and var results quality are validated in tables 10, 11 and 12 for italy, france and germany, respectively before and after the covid-19 pandemic crisis. it is confirmed that all the eigenvalues lie inside the unit circle. consequently, var satisfies stability condition for france, italy and germany before and during covid-19. after running the var, we support our findings by implementing the impulse function in figure 2. table 8. granger causality for italy italy before covid after covid equation excluded prob> chi2 prob> chi2 dlnδftsemib dlnδbitcoin 0.864 0.018** dlnδftsemib dlnδgold 0.265 0.587 dlnδftsemib dllnδwti 0.402 0.009** dlnδftsemib all 0.596 0.003** dlnδbitcoin dlnδftsemib 0.944 0.000*** dlnδbitcoin dlnδgold 0.24 0.095* dlnδbitcoin dllnδwti 0.093* 0.007** dlnδbitcoin all 0.192 0.000*** dlnδgold dlnδftsemib 0.72 0.002** dlnδgold dlnδbitcoin 0.81 0.033** dlnδgold dllnδwti 0.048** 0.013** dlnδgold all 0.193 0.005** dllnδwti dlnδftsemib 0.055* 0.080* source: authors calculations has covid-19 changed the correlation between cryptocurrencies and stock markets? table 9. granger causality for france france before covid after covid equation excluded prob> chi2 prob> chi2 dlnδcac40 dlnδbitcoin 0.657 0.033 dlnδcac40 dlnδgold 0.378 0.287 dlnδcac40 dllnδwti 0.285 0.018 dlnδcac40 all 0.57 0.005 dlnδbitcoin dlnδcac40 0.36 0 dlnδbitcoin dlnδgold 0.175 0.083 dlnδbitcoin dllnδwti 0.061 0.006 dlnδbitcoin all 0.134 0 dlnδgold dlnδcac40 0.961 0.013 dlnδgold dlnδbitcoin 0.818 0.09 dlnδgold dllnδwti 0.039 0.014 dlnδgold all 0.204 0.021 dllnδwti dlnδcac40 0.302 0.002 dllnδwti dlnδbitcoin 0.747 0.248 dllnδwti dlnδgold 0.02 0.244 dllnδwti all 0.042 0.013 source: authors calculations table 10. granger causality for germany germany before covid after covid equation excluded prob> chi2 dlnδdax30 dlnδbitcoin 0.824 0.078* dlnδdax30 dlnδgold 0.054* 0.484 dlnδdax30 dllnδwti 0.098* 0.012** dlnδdax30 all 0.102 0.012** dlnδbitcoin dlnδdax30 0.984 0.000*** dlnδbitcoin dlnδgold 0.251 0.149 dlnδbitcoin dllnδwti 0.093 0.010*** dlnδbitcoin all 0.192 0.000*** dlnδgold dlnδdax30 0.648 0.001** dlnδgold dlnδbitcoin 0.796 0.043** dlnδgold dllnδwti 0.045 0.008** dlnδgold all 0.187 0.003** dllnδwti dlnδdax30 0.462 0.014** dllnδwti dlnδbitcoin 0.737 0.329 dllnδwti dlnδgold 0.017 0.34 dllnδwti all 0.054 0.066* notes: *p< 0.1, ** p < 0.05, *** p < 0.01 source: authors calculations table 11. stability test for italy eigenvalue modulus be fo re c o vi d .603695 + .03151711i .604517 .603695 .03151711i .604517 .4107565 + .01133293i .410913 .4107565 .01133293i .410913 all the eigenvalues lie inside the unit circle. var satisfies stability af te r co vi d .8291133 .829113 .4813131 + .07542815i .487188 .4813131 .07542815i .487188 .4038571 .403857 all the eigenvalues lie inside the unit circle. var satisfies stability source: authors calculations -1 -.5 0 .5 1 im ag in ar y -1 -.5 0 .5 1 real roots of the companion matrix -1 -.5 0 .5 1 im ag in ar y 1 5 0 5 1 roots of the companion matrix 148 abdelkafi et al. table 12. stability test for france eigenvalue modulus be fo re c o vi d .5998576 .599858 .5304404 .53044 .4077736 + .01516735i .408056 .4077736 .01516735i .408056 all the eigenvalues lie inside the unit circle. var satisfies stability af te r co vi d .7005232 .700523 .5066378 + .09266491i .515042 .5066378 .09266491i .515042 .3528257 .352826 all the eigenvalues lie inside the unit circle. var satisfies stability diti source: authors calculations table 13. stability test for germany eigenvalue modulus be fo re c o vi d -.118457 0 -.0539891 0 .0078151 .028414 .0078151 -.028414 all the eigenvalues lie inside the unit circle. var satisfies stability af te r co vi d .0033594 .1356476 .0033594 -.1356476 -.1243459 0 .0044669 0 all the eigenvalues lie inside the unit circle.var satisfies stability condition. source: authors calculations -1 -.5 0 .5 1 im ag in ar y -1 -.5 0 .5 1 real roots of the companion matrix -1 -.5 0 .5 1 im ag in ar y 1 5 0 5 1 roots of the companion matrix -1 -.5 0 .5 1 im ag in ar y -1 -.5 0 .5 1 real roots of the companion matrix -1 -.5 0 .5 1 im ag in ar y roots of the companion matrix has covid-19 changed the correlation between cryptocurrencies and stock markets? figure 2. impulse response functions. source: the authors. germany france italy be fo re c o vi d -1 9 d ur in g co vi d -1 9 -.5 0 .5 1 -.5 0 .5 1 0 5 10 0 5 varbasic, lnbitcoin, lnbitcoin varbasic, lnbitcoin, lndax30 varbasic, lndax30, lnbitcoin varbasic, lndax30, lndax30 95% ci impulse response function (irf) step graphs by irfname, impulse variable, and response variable -.5 0 .5 1 -.5 0 .5 1 0 5 10 0 5 varbasic, lnbitcoin, lnbitcoin varbasic, lnbitcoin, lncac40 varbasic, lncac40, lnbitcoin varbasic, lncac40, lncac40 95% ci impulse response function (irf) step graphs by irfname, impulse variable, and response variable -.5 0 .5 1 -.5 0 .5 1 0 5 10 0 5 10 varbasic, lnbitcoin, lnbitcoin varbasic, lnbitcoin, lnftsemib varbasic, lnftsemib, lnbitcoin varbasic, lnftsemib, lnftsemib 95% ci impulse response function (irf) step graphs by irfname, impulse variable, and response variable -.5 0 .5 1 -.5 0 .5 1 0 5 10 0 5 varbasic, lnbitcoin, lnbitcoin varbasic, lnbitcoin, lndax30 varbasic, lndax30, lnbitcoin varbasic, lndax30, lndax30 95% ci impulse response function (irf) step graphs by irfname, impulse variable, and response variable -.5 0 .5 1 -.5 0 .5 1 0 5 10 0 5 varbasic, lnbitcoin, lnbitcoin varbasic, lnbitcoin, lncac40 varbasic, lncac40, lnbitcoin varbasic, lncac40, lncac40 95% ci impulse response function (irf) step graphs by irfname, impulse variable, and response variable -.5 0 .5 1 -.5 0 .5 1 0 5 10 0 5 varbasic, lnbitcoin, lnbitcoin varbasic, lnbitcoin, lnftsemib varbasic, lnftsemib, lnbitcoin varbasic, lnftsemib, lnftsemib 95% ci impulse response function (irf) step graphs by irfname, impulse variable, and response variable bitcoin-bitcoin bitcoin-bitcoin bitcoin-bitcoin bitcoin-bitcoin bitcoin-bitcoin bitcoin-bitcoin bitcoin-index 30 bitcoin-index 30 index 30-bitcoin index 30-bitcoin index 30-index 30 index 30-index 30 bitcoin-cac40 bitcoin-cac40 cac40-bitcoin cac40-bitcoin cac40-cac40 cac40-cac40 bitcoin-ftse mib bitcoin-ftse mib ftse mib-bitcoin ftse mib-bitcoin ftse mib-ftse mib ftse mib-ftse mib 150 abdelkafiet al. 4.2 impulse response function results figure 2 provides the responses of the variables to their own and cross-shocks during the two periods. the impulse–response graph places one impulse in each row and one response variable in each column. the horizontal axis is time. the vertical axis is in units of the variables measured in percentage points. 4.3 ls and gls regressions in order to confirm our findings, we test if, in the case of covid-19, both stock indexes and bitcoin are affected by governmental actions to control the pandemic through an ols regression. if so, we confirm the existence of coherency between bitcoin and stock indices during mitigated periods. we test for and residual autocorrelation. since results confirm the existence of such problems (see table 14), we turn to estimate gls regression for the following models (table 15): bitcoin = f (governmental measures), ftse mib = f (governmental measures of italy) cac= f(governmental measures of france), dax30= f(governmental measures of germany) table 14. estimation of ols regression italy france germany lnδbitcoin lnδftsemib lnδbitcoin lnδbitcoin lnδbitcoin lnδcac40 coeff coeff coeff coeff coeff coeff lnschoolclosing .0445947*** .0258763*** .0278783** .006253 -.011966 .0059655 lnworkplace -.0177996 -.003466 -.0208287 .0080479 .0615218*** .0231909** lncancelevents -.0199816 -.0279787*** -.0012075 -.0109665 -.039791** -.0244232*** lnstayhome -.0125133 .0004167 .0065614 .0015334 .0102935 .0004281 lnmouvements .0166239* .0013323 -.0083567 -.0013301 -.0037976 -.0006247 lntestingpolicies .038037 .0123604 .0022607 -.0132201 .0111725 .0020282 lncontacttracing -.0514426 -.0066714 -.0091534 .0096287 -.038255* -.0037589 δcumcases .0044208 -.0023159 -.0043617 -.0034895 .0022145 .0032284 _cons .0151017 .0006138 .0114601 .0014226 .0239506 -.0008544 r-sq 0.0516 0.1022 0.0319 0.0415 0.0579 0.1036 hetheroscedasticity test 0.0000 0.0000 0.0000 0.0000 0.0000 0.0000 durbin' 0.0107 0.0134 0.0212 0.8027 0.0067 0.4364 notes: *p< 0.1, ** p < 0.05, *** p < 0.01. source: authors calculations table 15. estimation of gls regression italy france germany lnδbitcoin lnδftsemib lnδbitcoin lnδcac40 lnδbitcoin lnδdax30 lnschoolclosing .0445947*** .0258763*** .0278783** .006253 -.011966 .0059655 lnworkplace -.0177996 -.003466 -.0208287 .0080479 .0615218*** .0231909** lncancelevents -.0199816 -.0279787** -.0012075 -.0109665 -.039791** -.0244232*** lnstayhome -.0125133 .0004167 .0065614 .0015334 .0102935 .0004281 lnmouvements .0166239*** .0013323 -.0083567 -.0013301 -.0037976 -.0006247 lntestingpolicies .038037 .0123604 .0022607 -.0132201 .0111725** .0020282 lncontacttracing -.0514426 -.0066714 -.0091534 .0096287 -.038255 -.0037589 δcumcases .0044208 -.0023159 -.0043617 -.0034895 .0022145 .0032284 _cons .0151017 .0006138 .0114601 .0014226 .0239506* -.0008544 notes: *p<0.1, ** p < 0.05, *** p < 0.01. source: authors calculations has covid-19 changed the correlation between cryptocurrencies and stock markets? 5. discussion 5.1 theoretical implications the results of pearson matrix illustrate a negative correlation between gold and returns of stocks and oil, confirming the hedging role of gold acknowledged by yousaf et al. (2021) and salisu et al. (2021). the outcomes of soomro et al. (2022) show the results of developed further intention and trust of investors towards cryptocurrency adoption. during covid-19, all correlations are positive and particularly, the correlations between bitcoin and the indices are stronger than those between gold and indices. this result seems to imply that a cryptocurrency is more likely to be a diversifier or a weak hedge for stock markets rather than a strong hedge during crisis period. however, some studies show that there are heterogeneous relationships between cryptocurrencies and stock market indices (bouri et al., 2017; feng et al., 2018; shahzad et al., 2019) and the bds1 test indicates that all series have nonlinear structures. according to results shown in table 8, before covid-19, we notice the absence of causality between bitcoin and the stock index in italy, while there is a unidirectional causality from oil to bitcoin and from ftse mib to oil and bidirectional causality between oil and gold. but things changed after the first discovered cases of covid-19. we interestingly find bidirectional causality between bitcoin and the stock index, oil and stock index and bitcoin and gold. table 8 shows that before covid-19, in france, almost like italy, there was a unidirectional causality from oil to bitcoin and a bidirectional causality between oil and gold. after the announcement of the first covid-19 cases, we found that causality relations significantly changed. in fact, we prove the existence of directional causality between bitcoin and cac40, oil and cac40 and bitcoin and gold (table 8). for the case of germany, we notice the absence of causality between dax30 and bitcoin before covid-19, suggesting their independence, while gold and oil caused bitcoin variation, in line with aysan (2021). after the outbreak of the pandemic, the results confirm the existence of bidirectional causality between dax30 and bitcoin. furthermore, interestingly, gold is affected by the stock index, bitcoin and oil (see table 8). our results join huang et al. (2021). finally, we note bidirectional causality between the stock index and oil, confirming previous studies of mariana et al. (2021). to conclude, we clearly notice the significant pattern change of correlations between studied assets and the strengthening of directional causalities during crisis periods, according to mishra et al. (2022). besides, we confirm the previous results of corbet et. al (2018) supporting the common consensus regarding weak correlations between cryptocurrencies and stock market and re-examine it during a crisis period. thus, we join the line of thoughts of jiang et al. (2021); bouri et al. (2017) and kristoufek (2015). at last, investors and decision-makers would reconsider their investment strategies in mitigated periods since cryptocurrencies cannot be used as a strong hedge against the risks of stock indices. according to the results, school closing and movement restrictions measures to control the 1 the broock, dechert and scheinkman (bds) test (broock et al. 1996) to test the linearity of all the returns series. 152 abdelkafiet al. spread of covid-19 enter a significant and positive impact on the ftse mib, reflecting its reducing effect on uncertainty and consequently rebuilding investor’s confidence in the italian stock market. bachman (2020) and sarkis et al. (2020) prove that while trying to save lives, some governmental measures controlling covid-19 spread are economically efficient, whereas others lack financial efficiency. thus, we prove the previous results of corbet et al. (2018), who prove turning to the impact on the cryptocurrency market. we find that the “stay at home” measure and the growth of the number of confirmed cases have a negative impact on bitcoin. in the case of france, international travel restricting measures, testing policy and the growth number of confirmed cases have a negative impact on the stock return cac40. these measures should generate negative impacts (see tables 7 and 8). figure 2 details that responses of bitcoin to the effect of a one-standard-deviation impulse of stock return (dax30, cac40 and ftse mib) are more pronounced during covid-19. specifically, it declines slightly after one day and then peaks at one percentage point increase before declining. it should be noted that the bitcoin response is not persistent. conversely, stock return response to choc due to bitcoin is barely significant. besides, responses to own chocks show that an impulse to bitcoin and each stock return causes a decline by about one percentage point over the following day. they respond strongly to their own shock. our findings imply that the investors in bitcoin and stock markets during the covid-19 pandemic would face an abnormal initial impact on shock after the pandemic compared to stable periods. 5.2 policy implications the oecd (2020) announced that the health crisis linked to covid-19 has severely affected the tourism sector in france. the losses are 20 billion euros in 2020, and it has lost between 70% and 80% of its turnover. finally, covid-19 cases are a negative sign for investors worldwide, and france was among the worst-affected countries. as for cac40, bitcoin is also negatively affected by international travel restricting measures. this is quite understandable since most bitcoin transactions are for international travelers. for germany, we find that workplace closing has a positive and significant effect on bitcoin and dax30 while the cancel events measure has a negative and significant effect. besides, testing policy measures has a positive and significant effect on bitcoin. in doing so, both bitcoin and stock indices are influenced by governmental measures. the results are useful for portfolio diversification and risk management. cardona-montoyaet al. (2022) point out that the pandemic has taught us to improve biosecurity measures and that financial strength, remote working and income diversification are key factors in dealing with negative shocks. valerio roncagliolo and villamonte blas (2022) argued that the stock market index could serve as a precautionary measure against possible crises in the financial market and thus inform measures to reduce the financial stress impact of on economies. thus, policymakers would have to reduce uncertainties in financial markets by reducing policy inconsistencies and enhancing monetary and fiscal policy coordination that would guarantee the effective implementation of policy decisions that would reduce the impact of the pandemic on the global economy. has covid-19 changed the correlation between cryptocurrencies and stock markets? 5.3 future research agenda the results are useful for portfolio diversification and risk management. future studies may consider a larger sample covering europe and north america. in addition, the overall level of stock market indices used could mask the potential heterogeneity of the gold hedging ability across stock market indices. therefore, future studies could expand our analysis by considering the level of stock market indices by sector of activities. another possible direction is to adopt the artificial neural network approach to study the relationship between cryptocurrencies and stock market indices. 6. conclusions previous studies have provided strong evidence of the hedging and safe-haven properties of commodities relative to equity indices in times of stress. the unprecedented outbreak of covid-19 has had a negative impact on human health and caused economic gridlock and uncertainty in financial markets around the world. due to the recent evidence of a stronger impact of the covid-19 pandemic on stock markets than previous epidemics and the lack of related empirical studies on the link between gold, crypto-currencies and stock markets, we analyse this missing insight for france, italy and germany before and during the covid-19 epidemic. specifically, we use the var model to understand the relationship between crypto-currencies and stock market returns before and during the covid-19 period. the main results are summarised below: in the italian and french contexts and for the period from 4 january 2016 to 31 december 2019, we show the non-existence of causality between bitcoin and the stock market index. instead, we note unidirectional causality from oil to bitcoin and from ftse mib to oil and bidirectional causality between oil and gold. in the german context, we find no causality between the dax30 and bitcoin prior to covid-19, suggesting their independence, while gold and oil cause bitcoin volatility index. in contrast, for the period from 31 december 2019 to 26 february 2021, we find that things change in france and italy as we find bidirectional causality between bitcoin and the stock market index, oil and the stock market index and bitcoin and gold. in germany, the empirical results confirm the existence of bidirectional causality between the dax30 and bitcoin. it is also interesting to note that gold is affected by the stock market index, bitcoin and oil. finally, we find bidirectional causality between the stock market index and oil, which confirms previous studies by mariana et al. (2021). second, we support our findings by implementing the impulse function to study both own and cross shocks for bitcoin and each stock market before and during the global covid-19 pandemic. the results show that bitcoin's responses to the effect of an increase in stock market performance at one standard deviation (dax30, cac40 and ftse mib) are more pronounced during the covid-19 pandemic. third, we test, during the pandemic, which stock market indices and bitcoin are affected by government actions to control the pandemic via an ols regression. we find that bitcoin's response to the effect of an increase in stock market performance (dax30, cac40 and ftse mib) is more pronounced during the covid-19 period. 154 abdelkafiet al. specifically, it declines slightly after one day before peaking at a one percentage point increase, before declining. consequently, bitcoin's reaction is not persistent. on the other hand, the reaction of stock market returns to the bitcoin shock is not very significant. investors are now better informed about the role of gold in hedging the risk of certain asian stock markets, not only in normal times, but also during the catastrophic event of the covid-19 epidemic. therefore, the results have important policy implications for investors and authorities. the analysis highlights the importance of clarifying the link between crypto-currencies and stock markets in the short and long term in order to establish policies aimed at stabilising stock markets. in addition, they are useful for portfolio diversification and market risk management. references al-awadhi, a.m., al-saifi, k., al-awadhi, a, & alhamadi, s. (2020). death and contagious infectious diseases: impact of the covid-19 virus on stock market returns. journal of behavioral and experimental finance,27(c), 100326. https://doi.org/10.1016/j.jbef.2020.100326 ali, m., alam, n., aun, s, & rizvi, r. (2020). coronavirus (covid-19) an epidemic or pandemic for financial markets. journal of behavioral and experimental finance, 27(c), 1-9. https://doi.org/10.1016/j.jbef.2020.100341 aysan, a. f., bergigui, f, & disli, m. (2021). blockchain -based solutions in achieving sdgs after covid-19. journal of open innovation: technology, market and complexity, 7 (2), 1-16. https://doi.org/10.3390/joitmc7020151 bachman, d. (2020). covid-19 could affect the global economy in three main ways”, available at: https://www2.deloitte.com/us/en/insights/economy/covid-19/economic-impact-covid19.html (accessed 10 april 2023). baker, s.r., bloom, n., davis, s.j., & terry, s.j. (2020). "covid-induced economic uncertainty," nber working papers 26983, national bureau of economic research, inc. baur, d.g., &dimpfl, t. (2018). the asymmetric return-volatility relationship of commodity prices. energy economics, 76 (c), 378-387. https://doi.org/10.1016/j.eneco.2018.10.022 baur, d.g., hong, k, & lee, a.d. (2018). bitcoin: medium of exchange or speculative assets? journal of international financial markets institutions and money, 54, 177-189. https://doi.org/10.1016/ j.intfin.2017.12.004 baur, d.g, & lucey, b.m. (2010). is gold a hedge or a safe haven? an analysis of stocks, bonds and gold. the financial review, 45 (2), 217-229. https://doi.org/10.1111/j.1540-6288.2010.00244.x bouri, e., gupta, r., tiwari, a, & roubaud, d. (2017). does bitcoin hedge global uncertainty? evidence from wavelet-based quantile-in-quantile regressions. finance research letters, 23, 87-95. https://doi.org/10.1016/j.frl.2017.02.009 bouri, e., shahzad, s.j.h., roubaud, d., kristoufek, l., & lucey, b. (2020). bitcoin, gold. and commodities as safe havens for stocks: new insight through wavelet analysis. the quarterly review of economics and finance, 77(c), 156-164. https://doi.org/10.1016/j.qref.2020.03.004 bouri, e., shahzad, s.j.h, & roubaud, d. (2019). cryptocurrencies as hedges and safe-havens for us equity sectors. the quarterly review of economics and finance, 75(c), 294-307. https://doi.org/ 10.1016/j.qref.2019.05.001 bredin, d., conlon, t, & poti, v. (2015). does gold glitter in the long-run? gold as a hedge and safe haven across time and investment horizon. international review of financial analysis, 41, 320-328. https://doi.org/10.1016/j.irfa.2015.01.010 cardona-montoya, r.a., cruz, v., & mongrut, s.a. (2022). financial fragility and financial stress during the covid-19 crisis: evidence from colombian households. journal of economics,finance and administrative science, 27(54), 376-393. https://doi.org/10.1108/jefas-01-2022-0005 chan, w.h., le, m., & wu, y.w. (2019). holding bitcoin longer: the dynamic hedging abilities of bitcoin. the quarterly review of economics and finance, 71 (c), 107-113. https://doi.org/ 10.1016/j.qref.2018.07.004 chen. y. wu, & vivek, p.k. (2014). the value of bitcoin in enhancing the efficiency of an investor’s portfolio. journal of financial planning, 27, 44-52. available at : https://www.financialplanning association.org/article/journal/sep14-value-bitcoin-enhancing-efficiency-investors-portfolio https://doi.org/10.1016/j.jbef.2020.100326 https://doi.org/10.1016/j.jbef.2020.100341 https://doi.org/10.3390/joitmc7020151 https://www2.deloitte.com/us/en/insights/economy/covid-19/economic-impact-covid-19.html https://www2.deloitte.com/us/en/insights/economy/covid-19/economic-impact-covid-19.html https://ideas.repec.org/p/nbr/nberwo/26983.html https://ideas.repec.org/s/nbr/nberwo.html https://ideas.repec.org/s/nbr/nberwo.html https://doi.org/10.1016/j.eneco.2018.10.022 https://doi.org/10.1016/j.intfin.2017.12.004 https://doi.org/10.1016/j.intfin.2017.12.004 https://doi.org/10.1111/j.1540-6288.2010.00244.x https://doi.org/10.1016/j.frl.2017.02.009 https://doi.org/10.1016/j.qref.2020.03.004 https://doi.org/10.1016/j.qref.2019.05.001 https://doi.org/10.1016/j.qref.2019.05.001 https://doi.org/10.1016/j.irfa.2015.01.010 https://www.emerald.com/insight/search?q=ra%c3%bal%20armando%20cardona-montoya https://www.emerald.com/insight/search?q=vivian%20cruz https://www.emerald.com/insight/search?q=samuel%20arturo%20mongrut https://www.emerald.com/insight/publication/issn/2218-0648 https://www.emerald.com/insight/publication/issn/2218-0648 https://doi.org/10.1108/jefas-01-2022-0005 https://doi.org/10.1016/j.qref.2018.07.004 https://doi.org/10.1016/j.qref.2018.07.004 https://www.financialplanningassociation.org/article/journal/sep14-value-bitcoin-enhancing-efficiency-investors-portfolio https://www.financialplanningassociation.org/article/journal/sep14-value-bitcoin-enhancing-efficiency-investors-portfolio has covid-19 changed the correlation between cryptocurrencies and stock markets? (accessed 10 april 2023). ciner, c., gurdgiev, c, & lucey, b.m. (2013). hedges and safe havens: an examination of stocks, bonds, gold, oil and exchange rates. international review of financial analysis, 29, 202-211. https://doi.org/ 10.1186/s40854-020-00199-w conover, m., jensen, g., johnson, r. & mercer, j. (2010). is now the time to add commodities to your portfolio? the journal of investing, 19(3), 10-19. https://doi.org/10.3905/joi.2010.19.3.010 conlon, t, & mcgee, r. (2020). safe haven or risky hazard? bitcoin during the covid-19 bear market. finance research letters, 35, 101607. https://doi.org/10.1016/j.frl.2020.101607 corbet, s., meegan, a., larkin, c., lucey, b, & yarovaya, l. (2018). exploring the dynamic relationships between cryptocurrencies and other financial assets. economics letters, 165, 28-34. https://doi.org/10.1016/j.econlet.2018.01.004 cretarola, a., figà-talamanca, g. & grunspan, c. (2021). blockchain and cryptocurrencies: economic and financial research. decisions in economics and finance, 44, 781–787. https://doi.org /10.1007/s10203-021-00366-3 feng, w., wang, y., & zhang, z. (2018). can cryptocurrencies be a safe haven: a tail risk perspective analysis. applied economics, 50(44), 4745-4762. https://doi.org/10.1080/00036846 .2018.1466993 flavin, t., morley, c.e, & panopoulou, e. (2014). identifying safe haven assets for equity investors through an analysis of the stability of shock transmission. journal of international financial markets, institutions and money, 33, 137-154. https://doi.org/10.1016/j.intfin.2014.08.001 frankovic, j., liu, b, & suardi, s. (2021). on spillover effects between cryptocurrency-linked stocks and the cryptocurrency market: evidence from australia. global finance journal, 54(c), 100642. https://doi.org/10.1016/j.gfj.2021.100642 gao, x, & nardari, f. (2018). do commodities add economic value in asset allocation? new evidence from time-varying moments. journal of financial and quantitative analysis, 53(1), 365-393. https://doi.org/10.1017/s002210901700103x goodell, j.w, & goutte, s. (2021). co-movement of covid-19 and bitcoin: evidence from wavelet coherence analysis. finance research letters, 38, 101625. https://doi.org/10.1016/j.frl.2020.101625 guesmi, k., saadi, s., abid, i, & ftiti, z. (2018). portfolio diversification with virtual currency: evidence from bitcoin”, international review of financial analysis, 63, 431-437. https://doi.org/10.1016/ j.irfa.2018.03.004 huang, y., duan k., & mishra t. (2021). is bitcoin really more than a diversifier? a pre-and post-covid-19 analysis. finance res lett, 34(c), 102016. http://dx.doi.org/10.1016/j.frl.2021.102016 hung, n.t. (2019). return and volatility spillover across equity markets between china and southeast asian countries. journal of economics, finance and administrative science, 24 (47), 66-81. https://doi.org/10.1108/jefas-10-2018-0106 hwang, s., & satchell, s.e. (2010). how loss averse are investors in financial markets? journal of banking and finance, 34(10), 2425-2438, http://dx.doi.org/10.2139/ssrn.383942 inci, a.c., & lagasse, r. (2019). cryptocurrencies: applications and investment opportunities. journal of capital markets studies, 3 (2), 98-112. https://doi.org/10.1108/jcms-05-2019-0032 ji, q., zhang, d, & zhao, y. (2020). searching for safe-haven assets during the covid-19 pandemic. international review of financial analysis,71(c), 101526. https://doi.org/10.1016/j.physa .2019.04.124 kajtazi, a., & moro, a. (2019). the role of bitcoin in well diversified portfolios: a comparative global study. international review of financial analysis, 61, 143-157. https://doi.org/10.1016/j.irfa. 2018.10.003 kim, j.m., kim, s.t, & kim, s. (2020). on the relationship of cryptocurrency price with us stock and gold price using copula models. mathematics, 8(11), 1-15. https://doi.org/10.3390/math8111859 kristoufek, l. (2015). what are the main drivers of the bitcoin price? evidence from wavelet coherence analysis. plos one, 10(4), e0123923. https://doi.org/10.1371/journal.pone.0123923 lahmiri, s., & bekiros, s. (2020). the impact of covid-19 pandemic upon stability and sequential irregularity of equity and cryptocurrency markets. chaos, solitons & fractals, 138(c), 109936. https://doi.org/10.1016/j.chaos.2020.109936 li, s., & lucey, b.m. (2017). reassessing the role of precious metals as safe havens–what colour is your haven and why? journal commodity markets,7(c), 1-14. https://doi.org/10.1016/j.jcomm .2017.05.003 mariana, c.d., ekaputra, i.a, & husodo, z.a. (2021). are bitcoin and ethereum safe-havens for stocks during the covid-19 pandemic? finance research letters, 38(c), 101798, 1-6. https://doi.org/10.1016/ j.frl.2020.101798 matkovskyy, r., & jalan, a. (2019). from financial markets to bitcoin markets: a fresh look at the contagion effect. finance research letters, 31(c), 93-97. https://doi.org/10.1016/j.frl. 2019.04.007 https://doi.org/10.1186/s40854-020-00199-w https://doi.org/10.1186/s40854-020-00199-w https://doi.org/10.3905/joi.2010.19.3.010 https://doi.org/10.1016/j.frl.2020.101607 https://doi.org/10.1016/j.econlet.2018.01.004 https://doi.org/10.1007/s10203-021-00366-3 https://doi.org/10.1007/s10203-021-00366-3 https://doi.org/10.1080/00036846.2018.1466993 https://doi.org/10.1080/00036846.2018.1466993 https://doi.org/10.1016/j.intfin.2014.08.001 https://doi.org/10.1016/j.gfj.2021.100642 https://doi.org/10.1017/s002210901700103x https://doi.org/10.1016/j.frl.2020.101625 https://doi.org/10.1016/j.irfa.2018.03.004 https://doi.org/10.1016/j.irfa.2018.03.004 http://dx.doi.org/10.1016/j.frl.2021.102016 https://doi.org/10.1108/jefas-10-2018-0106 http://dx.doi.org/10.2139/ssrn.383942 https://doi.org/10.1108/jcms-05-2019-0032 https://doi.org/10.1016/j.physa.2019.04.124 https://doi.org/10.1016/j.physa.2019.04.124 https://doi.org/10.1016/j.irfa.2018.10.003 https://doi.org/10.1016/j.irfa.2018.10.003 https://doi.org/10.3390/math8111859 https://doi.org/10.1371/journal.pone.0123923 https://doi.org/10.1016/j.chaos.2020.109936 https://doi.org/10.1016/j.jcomm.2017.05.003 https://doi.org/10.1016/j.jcomm.2017.05.003 https://doi.org/10.1016/j.frl.2020.101798 https://doi.org/10.1016/j.frl.2020.101798 https://doi.org/10.1016/j.frl.2019.04.007 https://doi.org/10.1016/j.frl.2019.04.007 156 abdelkafiet al. mishra, a.k., & patwa, j.a. (2022). return and volatility spillover between india and leading asian and global equity markets: an empirical analysis. journal of economics, finance and administrative science, 27 (54), 294-312. https://doi.org/10.1108/jefas-06-2021-0082 mnif, e., jarboui, a., & mouakhar, k. (2020). how the cryptocurrency market has performed during covid 19? a multifractal analysis. finance research letters, 36(c), 101647. https://doi.org/10.1016 /j.frl.2020.101647 oecd (2020). global financial markets policy responses to covid-19. available at: https://www.oecd.org/coronavirus/policy-responses/global-financial-markets-policyresponses-to-covid-19-2d98c7e0/ (accessed 10 april 2023). platanakis, e., sakkas, a., & sutcliffe, c. (2019). harmful diversification: evidence from alternative investments. the british accounting review, 51(1), 1-23. https://doi.org/10.1016/ j.bar.2018.08.003 ranaldo, a., & söderlind, p. (2010). safe haven currencies. review of finance, european finance association, 14 (3), 385-407. http://dx.doi.org/10.2139/ssrn.999382 salisu, a.a., vo, x.v, & lawal, a. (2021). hedging oil price risk with gold during covid19 pandemic. resources policy,70(c),101897. https://doi.org/10.1016/j.resourpol.2020.101897 sami, m., & abdallah, w. (2021). how does the cryptocurrency market affect the stock market performance in the mena region? journal of economic and administrative sciences, 37 (4), 741753. https://doi.org/10.1108/jeas-07-2019-0078 sarkis, j., dewick, p., strauss, s., schroeder, p.m., vazquez-brust, d., & kruger, r. (2020). ce forumpostcovid. summary report. https://doi.org/10.13140/rg.2.2.25795.43047 shahzad, s.j.h., bouri, e., roubaud, d., kristoufek, l., & lucey, b. (2019). is bitcoin a better safe-haven investment than gold and commodities? international review of financial analysis, 63(c), 322330. https://doi.org/10.1016/j.irfa.2019.01.002 sharif, a., aloui, c., & yarovaya, l. (2020). covid-19 pandemic, oil prices, stock market, geopolitical risk and policy uncertainty nexus in the us economy: fresh evidence from the wavelet-based approach. international review of financial analysis, 70(c), 101496 https://doi.org/10.1016 /j.irfa.2020.101496 soomro, b.a., shah, n., & abdelwahed, n.a.a. (2022). intention to adopt cryptocurrency: a robust contribution of trust and the theory of planned behaviour. journal of economic and administrative sciences, vol. ahead-of-print no. ahead-of-print. https://doi.org/10.1108/jeas10-2021-0204 tversky, a., & kahneman, d. (1991). loss aversion in riskless choice: a reference-dependent model. the quarterly journal of economics, 106(4), 1039-1061, https://doi.org/10.2307/2937956 umar, z., & gubareva, m. (2020). a time–frequency analysis of the impact of the covid-19 induced panic on the volatility of currency and cryptocurrency markets. journal of behavioral and experimental finance. 28, 100404. https://doi.org/10.1016/j.jbef.2020.100404 urquhart, a., & zhang, h. (2019). is bitcoin a hedge or safe haven for currencies? an intraday analysis. international review of financial analysis, 63(c), 49-57. https://doi.org/10.1016/j.irfa .2019.02.009 valerio roncagliolo, f.c., & villamonte blas, r.n. (2022). impact of financial stress in advanced and emerging economies. journal of economics, finance and administrative science, 27 (53), 6885. https://doi.org/10.1108/jefas-05-2021-0063 xu, f., bouri, e., & cepni, o. (2022). blockchain and crypto-exposed us companies and major cryptocurrencies: the role of jumps and co-jumps”, finance research letters, 50. https://doi.org/10.1016/j.frl.2022.103201 yousaf, i., ali, s., naveed, m., & adeel, i. (2021). risk and return transmissions from crude oil to latin american stock markets during the crisis: portfolio implications. sage open, 11 (2), 1-11. https://doi.org/10.1177/21582440211013800 zhang, d., hu, m., & ji, q. (2020). financial markets under the global pandemic of covid19”, “finance research letters. 36 (c), 101528. https://doi.org/10.1016/j.frl.2020.101528 https://www.emerald.com/insight/search?q=aswini%20kumar%20mishra https://www.emerald.com/insight/search?q=jash%20ashish%20patwa https://www.emerald.com/insight/publication/issn/2218-0648 https://www.emerald.com/insight/publication/issn/2218-0648 https://doi.org/10.1108/jefas-06-2021-0082 https://doi.org/10.1016/j.frl.2020.101647 https://doi.org/10.1016/j.frl.2020.101647 https://www.oecd.org/coronavirus/policy-responses/global-financial-markets-policy-responses-to-covid-19-2d98c7e0/ https://www.oecd.org/coronavirus/policy-responses/global-financial-markets-policy-responses-to-covid-19-2d98c7e0/ https://doi.org/10.1016/j.bar.2018.08.003 https://doi.org/10.1016/j.bar.2018.08.003 http://dx.doi.org/10.2139/ssrn.999382 https://doi.org/10.1016/j.resourpol.2020.101897 https://www.emerald.com/insight/search?q=mina%20sami https://www.emerald.com/insight/search?q=wael%20abdallah https://www.emerald.com/insight/publication/issn/1026-4116 https://doi.org/10.1108/jeas-07-2019-0078 https://doi.org/10.13140/rg.2.2.25795.43047 https://doi.org/10.1016/j.irfa.2019.01.002 https://doi.org/10.1016/j.irfa.2020.101496 https://doi.org/10.1016/j.irfa.2020.101496 https://www.emerald.com/insight/search?q=bahadur%20ali%20soomro https://www.emerald.com/insight/search?q=naimatullah%20shah https://www.emerald.com/insight/search?q=nadia%20a.%20abdelmegeed%20abdelwahed https://www.emerald.com/insight/publication/issn/1026-4116 https://www.emerald.com/insight/publication/issn/1026-4116 https://doi.org/10.1108/jeas-10-2021-0204 https://doi.org/10.1108/jeas-10-2021-0204 https://doi.org/10.2307/2937956 https://doi.org/10.1016/j.jbef.2020.100404 https://doi.org/10.1016/j.irfa.2019.02.009 https://doi.org/10.1016/j.irfa.2019.02.009 https://www.emerald.com/insight/search?q=flavio%20c%c3%a9sar%20valerio%20roncagliolo https://www.emerald.com/insight/search?q=ricardo%20norberto%20villamonte%20blas https://www.emerald.com/insight/publication/issn/2218-0648 https://doi.org/10.1108/jefas-05-2021-0063 https://doi.org/10.1016/j.frl.2022.103201 https://doi.org/10.1177/21582440211013800 https://doi.org/10.1016/j.frl.2020.101528 1. introduction 2. literature review 3. methods 4. results 5. discussion 6. conclusions references european journal of government and economics 11(1), june 2022, 5-6 © 2022. this work is licensed under a cc by-nc 4.0 license. european journal of government and economics issn: 2254-7088 editorial statement: ten years of life of ejge josé manuel sánchez santos a, giacomo benedetto b a universidade da coruña, spain b royal holloway, university of london, united kingdom abstract. in this editorial statement, we present a balance of the first decade of life of the european journal of government and economics. we discuss the main difficulties an independent journal faces in maintaining the policy of open access and publication free of charge. we also renew the commitment to the journal's spirit, particularly with the editorial rigour and best practices. finally, we identify the next steps in the consolidation of ejge as an academic journal that attempts to keep an integrated approach between the fields of political science and economics. keywords. academic journal, economics, indexing, open access, political science. doi. https://doi.org/10.17979/ejge.2022.11.1.9174 with the publication of volume 11 (issue 1) in june 2022, the european journal of government and economics has just completed its first decade. the ejge’s founding spirit is to integrate approaches between the fields of political science and economics. ten years ago, our founding editor diego varela launched this project. initially, giacomo benedetto and josé manuel sánchez as associate editors, and later paolo rungo as managing editor, have tried to give continuity to the legacy of the founding editor. this was a great challenge because the sustainability of an independent academic journal is a complex task. in the editorial statement of the first issue of ejge, we explained the reasons for the creation of the european journal of government and economics. our success over the last ten years is the best proof that the creation of a journal such as ejge was the right decision. since 2012 we have published over 100 articles by authors from across the globe. at first glance, it may seem that the number is not very high, but our philosophy and editorial policy from the beginning have been to prioritize quality over quantity. it is only fair to take this opportunity to acknowledge all the people who have contributed to the success of this editorial initiative. we want to give particular thanks to all of those involved in the editorial process of ejge, who have served it with dedication and unique competence. we are very grateful to the authors that have chosen ejge for publishing their papers. the journal has also benefited from the effort of the reviewers, whose work has enabled the editorial team to select the best papers for publication. we really appreciate the availability of the reviewers because their help as has been invaluable to us. https://creativecommons.org/licenses/by-nc/4.0/deed.en https://doi.org/10.17979/ejge.2022.11.1.9174 european journal of government and economics 11(1), june 2022, 31-50 6 we express our gratitude to the members of editorial board for their dedication to the journal. the editorial board is composed of outstanding theoretical and empirical economists and political scientists. the team covers the main areas of economics and political science and is geographically balanced. since the first issue was launched, we have worked hard to consolidate ejge, and now it is time to provide a new impulse. we have accumulated a wealth of experience that will help us to improve our work as editors. in this line, we envision further progress in the indexing process in the near future. in this direction, a next step is the entry into the emerging sources citation index (esci). looking ahead, we shall update the membership of the editorial board to accommodate scholars who will help give the journal a boost and enable it to face future challenges and developments. to summarize, after ten years it is a good time to renew and reinforce our commitment to ejge. it is an open-access academic journal with the mission of providing free dissemination of scientific findings on current issues regarding the interrelationship between the fields of government and economics. the journal is committed to continue a strict open-access, nocharge policy. furthermore, we shall ensure that ejge continues to meet the highest editorial standards of rigor and quality. we are aware that it is not easy to maintain an independent journal while respecting the standards of academic quality, but we are optimistic, and we hope that our authors and reviewers will continue to trust our project. european journal of government and economics 11(2), december 2022, 210-233 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 the impact of gender inequality on economic growth: an explanatory sequential mixed methods study of female labour participation in the civil service andrew jia yi kama , jerry sankay oboha *, zarina othmanb, jusoh sufiana a institute of malaysian and international studies, universiti kebangsaan malaysia, malaysia b pusat citra universiti, universiti kebangsaan malaysia, bangi, malaysia * corresponding author at: jerrysankayoboh@gmail.com abstract. gender inequality is recognised as an impediment to economic growth and development. in nigeria, several policies have been enacted to eliminate gender inequality. however, studies show that women are grossly marginalised, exploited, and discriminated against in their socioeconomic life. this paper examines the impact of gender inequality and female labour participation in the nigerian civil service on the economic growth of nigeria using mix-methods. first, an estimation of the impact of gender inequality on economic growth using panel data regression, followed by contentand consistency-based analysis that examines the factors responsible for gender inequality and discrimination issues women experience in the nigerian civil service using contentand consistency-based analysis. the findings suggest that gender inequality negatively impacts economic growth and distorts the pool of human capital available to the nigerian civil service and recommends the use of education and outsourcing of some employment functions to eliminate it effectively. keywords. gender inequality, female labour participation, civil service, culture and tradition, economic growth. jel codes. o11; h70;j7. doi. https://doi.org/10.17979/ejge.2022.11.2.8951 1. introduction the significance of gender equality as a sine qua non for achieving economic growth and development is globally recognised. the united nations sustainable development framework (unsdf: 2017) on human rights put gender equality and women’s economic empowerment at the core of all development strategies, with the world bank endorsing it as critical to sustainable and inclusive growth (devadas & kim, 2020). it is, therefore, imperative to understand the dynamics of gender inequality and its implication for female labour participation and economic growth. the civil service as part of the public service contributes to economic growth and development via its contribution to gross domestic product (gdp) from revenue generated, women empowerment, policy formulation and implementation. with about 11 gender equality policies enacted, the nigerian government have demonstrated its willingness to ensure that women are given equal access and opportunities as men. however, studies still suggest that women are grossly marginalised, continuously exploited and discriminated against in many areas of their socioeconomic life (adegoke et al., 2016 & oloni, 2015). in the nigerian civil service, women are https://creativecommons.org/licenses/by-nc/4.0/ https://doi.org/10.17979/ejge.2022.11.2.8951 andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 211 still challenged by structural barriers such as the glass ceiling and discrimination in development opportunities that limit them from fully exploring their productive capabilities (oboh et al., 2022 & 2021). a report from the canadian international development agency (cida) showed that in nigeria, women make up about 24% of civil servants, while men are 76%. women constitute less than 14% of management and decision-making roles, 17.5% in well-paid and highly skilled roles, while men constitute 86% and 82.5%, respectively (cida, 2012). the poor representation of women and the lack of opportunities in decision-making positions in the public service imply that they are under-represented in the country’s key decision-making process. this coincides with part of the concerns highlighted in the beijing platform for action1 (national bureau of statistics (nbs), 2018). the civil service puts more men in senior management positions as a way of reinforcing the existing glass ceiling while keeping men in leadership (fatile & adejuwon, 2010). this is a concern because the public sector is strategic not just for providing women with employment but also for giving women quality employment. having more women in paid employment (which is associated with public service) improves their economic well-being, and mitigates gender inequality (oboh et al., 2022 & 2021). thus, the role of public service in curtailing gender inequality is significant. despite the importance of the issue, the empirical literature on gender inequality and economic growth in nigeria is sparse, and even lesser are the studies that have been conducted using mixmethod. therefore, the objective of the study is to examine the determinants and the impact of gender inequality on economic growth in nigeria. this study contributes to the sparse empirical literature by quantifying the magnitude of the impact of female labour participation and gender inequality in the civil service on the economic growth of nigeria. in further identifying the determinants, the paper further identifies the gender disparity issues and factors that contribute to gender inequality by coding a gender inequality nexus from face-to-face interviews. this is to provide a new measurement of the forms of gender disparities that influence the civil service. using explanatory sequential mixed methods that combine panel data regression and content analysis, the study estimates the impact of gender inequality on economic growth and uncovers the dynamism of gender inequality on female labour participation in the civil service. this combination of econometricand content analysis is designed to explore a subject matter vertically (depth) and horizontally (breadth), followed by a verification process (crestwell, 2012). to explore the subject matter vertically, the explanatory sequential mixed methods utilise the within-methods (i.e., the synthesising of the qualitative studies to establish the gender inequality nexus). for horizontal analysis, the between-methods is deployed (i.e., the triangulation of the quantitative and qualitative studies to give a robust insight and a better understanding of gender inequality in the nigerian civil service and how it impacts economic growth). 1 the 1995 beijing declaration and platform for action was one of the outcomes of the fourth world conference on women that requires participating countries to acknowledged the voices of women, ensure equal gender participation in the public service and governance, advance gender equality, continuously addressing any obstacle, reduce the increasing poverty rate against women, fostering women advancement and empowerment etc. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 212 the next section examines existing literature and empirical studies to provide a background of knowledge to the study. 2. literature review a lot of emphasis on gender inequality has been attributed to culture, religion and other belief systems, which form the core of gender socialisationa term where women and men are defined to fit certain roles in society. past literature mainly challenged the definition of roles. when it is biased, it forms the notion of gender inequality or discrimination. in gender studies literature, there are two major perspectives on the sources of inequality. qualitative studies by scholars in the radical feminist discipline argued that gender inequality is rooted in biological differences with a focus on procreation and sexuality, while liberal feminists ascribed the relegation of women to culture, lack of educational opportunities and legal constraints as the impediments to women’s success in the society (enyew & mihrete, 2018; tong, 2014). society, by tradition, socialises men to take up leadership, decision-making and major household roles, while women are cast as unpaid domestic labour and low-earning jobs. therefore, not much attention is given to women’s formal learning and development, which forms the basis of female empowerment (enyew & mihrete, 2018: tong 2014). this is consistent in the nigerian context, where masculinity and femininity consist of both cultural and social phycological perspectives. the male gender roles are the major emphasis in various societies, while women's cultural values and roles are determined by religion, culture and customary laws that are presided over by men (okongwu, 2020). the public service sector has the potential to empower women and avail them of quality employment. this emphasis on empowerment gave rise to the 1994 beijing platform for action, which advocated for 30% women’s participation in the public service sector (nbs, 2018). a study by kpanja & umar (2018) showed that gender inequality in the nigerian civil service incapacitates the smooth delivery of government service machinery. this is also consistent with oboh et al. (2019) empirical analysis whereby gender inequality impacts the revenue generation of the nigerian civil service negatively. the mechanism in which gender inequality impedes economic growth lies in the underutilization of human capital in a country. by discriminating employment opportunities based on gender, the full potential in the pool of a country’s human capital is not fully utilized. even though the present regulation for entrance, transfer and promotion in the nigerian civil service is via a competitive examination, cases of nepotism and corruption have impeded the implementation (unodc, 2020). this further contributes to one of the reasons women are still discriminated against and deprived of employment, quality employment and relegated to background roles irrespective of their educational qualifications (dfid, 2012). ogunyemi (2014) explained that gender relations issues in nigeria have no regard for performance. rather, it makes cultural melancholy a gender identification that is polarised by internalised prohibition and could lead to the relegation or displacement of the female gender. this is refined in economics and feminism in management, where the attention of feminist andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 213 scholars is focused on visible sources of discrimination against women in the labour markets and wages. policy emphases are focused on women attaining equal status as men in employment, career path, government, and traditional institutions (nienaber & moraka, 2016). theoretical connections between gender equality and economic growth are linked by decomposing the solow model whereby labour input is decomposed into skills and non-skills, and differentiated by gender. the gender element was initially brought into growth discourse partly due to the argument that women have a higher propensity to save compared to men in their household consumption, hence increasing the country’s gdp (lofstrom, 2008). it was also initially driven by concerns on humanitarian grounds that gender inequality may be detrimental to economic growth because it would breed poverty and disempowerment in a specific gender group (brummet 2008; morrison et al., 2007). distortions in gender and skills level may constitute labour distortions, which subsequently impact economic growth, holding other factors constant. empirical evidence confirmed this connection and showed that gender inequality distorts the pool of talent available to employers via the gender gap in employment (oboh et al., 2019; klasen, 2018; cuberes & teignier, 2016). conversely, al-shammari & al-rakhis (2017), ali (2015) and braunstein (2012) reported otherwise. these studies have shown that gender inequality serves as a stimulus to growth in situations where inequality lowers the cost of labour and not the quality of output. the results are inconclusive, and an extensive search for literature shows that empirical studies are scarce. nevertheless, the paper hypothesized that ensuring gender equality would lead to higher economic growth, improve women’s financial and physical wellbeing and their household bargaining power etc. (devadas & kim, 2020). the studies recognize women as active agents of transformation and development, thus, drawing attention to women's subordination, oppression, and other challenges imposed by their various societies and places of work and how it affects the production of goods and services in an economy over time. in terms of measuring the dependent variable, literature shows that economic growth is commonly measured by gross national product (gnp) and gross domestic product (gdp). only a small number of studies have explored other forms of national output as indications of economic growth and economic output. otteson (2004) explained that adam smith posited in his wealth of nations that an economy’s annual revenue would always be equal to the exchangeable value of the whole total output from its industries or the same as the exchangeable value. smith also recommended that capital should be invested in a way that generates revenue for the economy. this premise suggests that annual revenue generated by an economy is also an indication of the economy’s level of productivity. alternatively, oboh et al. (2019) have shown that government capital expenditure and female labour participation positively impact stategenerated revenue. the methodology that links both quantitative and qualitative literature will be discussed in the next section. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 214 3. methodology the research utilises explanatory sequential mixed-mode methods that comprise quantitative and qualitative analyses. figure 1. an overview of the explanatory sequential mixed-model methods. the method follows a procedure starting with the quantitative analysis that estimates the relationship between gender inequality and economic growth using a panel data regression technique before the qualitative analysis that uncovers the factors responsible for gender inequality and the gender disparity issues in the civil service using a contentand consistencybased analytical technique as depicted in figure 1 below. the findings from both the quantitative and qualitative analyses were synthesised using a between-methods triangulation technique, while the gender inequality nexus in the qualitative analysis was derived using a within-methods triangulation technique. the triangulation method increases the credibility and validity of research findings since the study uses multiple techniques at multiple levels. for example, the empirical study uses official macro-level data (i.e. employment, education level, etc.) while the discourse analysis collects more micro-data level data. results from macro-level estimation enable the study to identify key (statistically significant) variables, which will be further verified (triangulated) at the micro-level interviews. the micro-level interviews, in return, will further provide inputs to re-confirm the state of gender inequality at the country level. in other words, the qualitative study provides support and extends the findings from the empirical model. combining both methods provide depth and insights into the impact of gender inequality on economic growth and female labour participation in the nigerian civil service. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 215 3.1 quantitative data and model the study uses the solow growth model and used interstate cross-sectional data from 36 states in nigeria from 2008 to 2016. the solow growth model was augmented to account for human capital following mankiw et al. (1992): 𝑌𝑌𝑡𝑡 = 𝐾𝐾𝑡𝑡𝛼𝛼𝐻𝐻𝑡𝑡 𝛽𝛽(𝐴𝐴𝑡𝑡𝐿𝐿𝑡𝑡)1−𝛼𝛼−𝛽𝛽 [1] ht is the human capital stock, β and 1-α-β are the elasticities of human capital and effective labour, respectively. kt is physical capital. yt is output. the assumption of decreasing return to scale for investment to output is α+β < 1. ht depreciates at the same rate as kt. similar to the solow-swan, a fraction of output unconsumed or saved in each period syt is invested in both physical and human capital. therefore, s = sk + sh, resulting in two fundamental dynamic equations in the model. the reduced forms of physical capital and human capital are as follows: 𝑘𝑘𝑡𝑡∗ = 𝑠𝑠𝑘𝑘𝑦𝑦𝑡𝑡 − (𝑛𝑛 + 𝑔𝑔 + 𝛿𝛿)𝑘𝑘𝑡𝑡 [2] ℎ𝑡𝑡∗ = 𝑠𝑠ℎ𝑦𝑦𝑡𝑡 − (𝑛𝑛 + 𝑔𝑔 + 𝛿𝛿)ℎ𝑡𝑡 [3] sk and sh are fractions of saved output invested in physical and human capital, respectively. imposing a steady-state of 𝑘𝑘𝑡𝑡∗ = ℎ𝑡𝑡∗ = 0, and taking the natural log of equation 1, the final specification is as follows: 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺ℎ = 𝑙𝑙𝑛𝑛𝐴𝐴(0) + 𝑔𝑔𝐺𝐺 + 𝛼𝛼 1−𝛼𝛼−𝛽𝛽 𝑙𝑙𝑛𝑛(𝑠𝑠𝑘𝑘) + 𝛽𝛽 1−𝛼𝛼−𝛽𝛽 𝑙𝑙𝑛𝑛(𝑠𝑠ℎ) − 𝛼𝛼+𝛽𝛽 1−𝛼𝛼−𝛽𝛽 𝑙𝑙𝑛𝑛(𝑛𝑛 + 𝑔𝑔 + 𝛿𝛿) [4] in this form, growth is determined by physical capital, human capital and effective labour. human capital in this study is proxy by junior or senior cadre, and (in)efficiency of labour is assumed by the disparity in the skill levels in gender compositions – which is grossly interpreted in the literature as gender inequality in education or employment positions. a(0) or technology is retained from the original specification of the solow model. therefore, the final specification model is specified as: 𝑙𝑙𝑛𝑛𝑙𝑙𝑙𝑙𝑙𝑙𝑡𝑡 = 𝐿𝐿𝐿𝐿𝑡𝑡 + 𝛼𝛼𝑙𝑙𝑛𝑛𝑙𝑙𝐺𝐺𝛼𝛼𝛼𝛼𝑡𝑡 + 𝛽𝛽1𝑙𝑙𝑛𝑛𝑙𝑙𝑙𝑙𝐿𝐿𝑡𝑡 + 𝛽𝛽2𝑙𝑙𝑛𝑛𝑙𝑙𝑙𝑙𝐿𝐿𝑡𝑡 + 𝛽𝛽3𝑙𝑙𝑛𝑛𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝐿𝐿𝑡𝑡 + 𝛽𝛽4𝑙𝑙𝑛𝑛𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝐿𝐿𝑡𝑡 + 𝜀𝜀𝑡𝑡 [5] where the dependent variable is the logarithm of the growth rate of total state revenue (lntsr) (see oboh et al., 2019); it is revenue generated by the states annually. it is proxied as economic growth due to the unavailability of state-level gross domestic product (gdp or output) data in nigeria. as highlighted in the literature review, government revenue can be a mirror equivalence of the monetary value of all the goods and services produced in an economy. it consists of revenue generated via levelling of fees, penalties or fines, taxes, property income derived from assets ownership, sales of goods and services, licenses and transfers imposed by government andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 216 units, amongst others (adenuga & chike, 2013). the data for female labour participation in the junior and senior cadre of the civil service (jfp/sfp) were sourced from the national bureau of statistics statistical report on women and men, while data for state government capital expenditure (sgce) and annual total state revenue (tsr) were from the central bank of nigeria annual report from. the proxy for gender inequality (djmfp, dsmfp) and labour productivity (lp) were computed using the data from the aforementioned sources. lp is the proxy for technology within the model. 3.2 qualitative data and methods content-and consistency-based analysis introduced by oboh et al. (2021) was used to analyse individuals’ subjective experiences, views, and opinions from the interview that was conducted in 2018. the process involves the coding of the interview transcripts, counting of concepts, and frequency analysis. the participants twenty-seven participants were selected using purposive and convenience sampling techniques. the participants comprised eighteen female and nine male civil servants between the ages of thirty-five and sixty-three years2. thirteen of the participants are from the junior cadre, and fourteen are from the senior cadre. twenty-two of the participants were stationed in the federal capital territory (fct), while the remaining five participants were from five other states that visited the fct for training. the participants were from ten ministries and were selected based on the criteria below. • employment: the participant must be an employee of the nigerian civil service. • tenure: the participant must be employed in the nigerian civil service for a minimum of 10 years. • qualification: the participants are expected to have a minimum qualification of a first leaving school certificate for employees in the junior cadre and a first degree for employees in the senior cadre. using these criteria, the study assumed that the selected participants are familiar with and possess extensive experience pertinent to gender inequality in the civil service. also, due to the sensitivity of the subject matter, the participants were selected voluntarily. the focus of the qualitative study is on examining women’s experiences. the gender distribution of the participants is of the ratio of two-to-one, with women constituting two and men one. the study account for the gender ratio of women and men based on their responses in 2 the disparity in the number of female and male participant is not an issue as our research objective is not about identifying the differences between male or female responses. we leave that for future research. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 217 percentages, i.e., the total number of female or male civil servants that suggested the coded categories out of the total number of their gender representation, respectively. data collection we collected the data by administering face-to-face interviews with semi-structured interview questions. the interview questions follow a repetitive pattern that addresses the factors that contribute to gender inequality and gender disparity issues. two major questions were asked of the participants. each of the questions was asked three times3 using different sentence structures (meaning they were paraphrased). thus, bringing it to a total of six questions. the responses from the interviews were recorded using a voice recorder. the reason for the repetition in questioning is to ensure consistency. in this study, we introduced a new consistency measure for accounting and validating responses from the interview sessions. to avoid double-counting and overemphasizing the participant responses, the study assumed that: • the participant would be consistent in all three questions. if that is the case, the consistency would be measured as 3/3. • if the participant answers no to two interview questions and yes to one that addresses the same interview question, the level of consistency would be measured as 1/3. the results are converted to percentages by multiplying the outcomes (i.e. 2/3) by 1004. 4. findings the empirical analysis is divided into three sections: quantitative, qualitative and triangulation. the quantitative section presents the findings from the panel data regression, the qualitative section presents the content analysis in the form of the gender inequality nexus while the triangulation presents a causal relationship via the gender inequality nexus. 4.1 empirical analysis: panel-data estimation a hausman test was conducted, and the fixed-effect estimator was used. the estimation also included a robust fixed-effects panel data regression to control for heteroscedasticity5. the results can be seen in table 1. issues of multicollinearity are not severe, as seen in appendix a2. the findings from the empirical model suggested that female labour force participation at the senior cadre (sfp) and its gender inequality indicator (dsmfp) are not significant, while female 3 our pilot study showed that some civil servants are always trying to be politically correct and tend to only give credible answers after going through a series of similar questions. this pattern is similar to participant’s behaviour exhibited in focused group interviews, thus, the need to paraphrase the questions. 4 the total number of responses that identified gender inequality is calculated from the 3 interview questions. it is then divided by the “expected outcome” and multiplied by 100. example: if 5 participants say there is gi the expected consistent outcome will be 15 (3 x 5 “yes” responses). however, if some participants answered no even though the questions were only paraphrases, the consistency only show at 66.7 % (as in 10/15 x 100). 5 data limitations has prevented the study to include dynamics component. hence a panel fixed-effect estimator is used in this study. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 218 labour participation at the junior cadre (jfp), gender inequality indicator for the junior cadre (djmfp), state government capital expenditure (sgce) and labour productivity (lp) are significant6. jfp is significant at a 95% confidence level which suggests that a percentage increase in female labour participation would result in a 0.06% increase in total revenue. therefore, employing and empowering more women in the junior cadre of nigerian civil service would be beneficial to the economy. this is not the case for gender inequality in the junior cadre, it is significant at a 90% confidence level, suggesting that forgoing women’s employment in the junior cadre in favour of men by a percentage would make economic growth worse off by 0.04%. sfp and dsmfp are insignificant. this may be a data-related problem, severe gender imbalance in employment, poor employment management or the ghost worker syndrome7 that has plagued the public service for several decades. the level of technological advancement and government capital expenditure were significant at a 95% confidence level, suggesting that a percentage advancement in technology and an increase in government spending on capital spending would result in a 5% and 0.21% increase in economic growth, respectively. to ensure the model is robust, a robustness check on the model is conducted. this was done by reversing the calculation on the gender inequality indicator from the original (female subtracting male) to (male subtracting female) in different cadres. we anticipate the latter to be insignificant because the differences in male and female participation are the main contribution to inequality and not vice-versa. the result is in appendix a3. although only lp, sgce and jfp are significant, all other variables are correctly signed and consistent with the results in table 1 – thus, suggesting that the empirical model is robust. the key variable of interest, djfmp is insignificant in the robustness check model, while its mirror djmfp is significant. this is also consistent with our assumptions, which further strengthened the robustness of the model. table 1. panel data regression: robust fixed effect model. lntsr coef. p>|t| lp 0.0548 0.0160 lnsgce 0.2073 0.0000 lnjfp 0.0576 0.0350 lnsfp 0.0039 0.8920 lndjmfp -0.0399 0.0970 lndsmfp 0.0328 0.3050 cons 3.2273 0.0000 n = 148 prob > f = 0.0000 sigma_u = 0.1819 sigma_e = 0.3123 rho = 0.2534 6 panel unitroot (levinlinchu) test shows that test statistic t = −2.0015 is significantly less than zero (p < 0.00005), so we reject the null hypothesis of a unit-root in favor of the alternative that lntsr is stationary (that is, that φ < 0). 7 agboola, 2018; igbokwe-ibeto et al., 2015. this is an act of stealing from the state and federal government at large under the disguise of paying wages to employees, and, these employees do not exist in a person. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 219 4.2 qualitative analysis the results from the contentand consistency-based analysis are divided into three parts. since this section follows a qualitative framework, the analysis is based on inductive reasoning. it starts with identifying the factors responsible for gender inequality. using these factors, the study identifies gender disparity issues that are related to the identified factors. finally, we form a conjecture or conclusion which represents the gender inequality nexus in the nigerian civil service. this new finding is depicted in figure 2 (which will be discussed later). the consistency tables, the number of responses and the participants used for the content analysis are in appendices b1, b2 and b3. factors responsible for gender inequality the study revealed that the factors fuelling gender inequality in the nigerian civil service are seen from three perspectives. those with origins stemmed from external factors (exogenous factors), internal (endogenous factors) and both (exo-endogenous factors). a. exogenous factors exogenous factors encapsulate two key sources in which gender discrimination against women strives in the nigerian civil service. these sources are the deficit in education and religion, with consistency rates of 88.0% and 81.0%, respectively. the findings suggest that deficits in education as well as religious biases impact women’s availability for employment negatively. both of these factors affect women’s skill set and the quality of employment they received, as narrated in the responses in both female (fp) and male participants (mp) excerpts below. …majority of us women don’t have the opportunity…in education. we do more house chores and don’t have enough time to study… (fp18) … there are some religious backgrounds where the women in the north are restricted from going out… (mp23) b. endogenous factors the endogenous factors comprise two main factors and two subcategories which are i) stigmatisation of women and ii) the operational structure of the civil service. it is further decomposed into femininity, decision-makers, and employees’ preferences. these internal factors fuel gender discrimination against women from within the civil service. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 220 with a consistency rate of 67.7%, 63.9% for the main factors, and 81.5% and 61.1% for the sub-factors, respectively (appendix b2), the participant suggested that endogenous factors allow for the elevation of the male gender identity and discrimination against women in the workplace. this includes employment opportunities, empowerment, privileges, and job security in the civil service. examples are narrated in the excerpts below. …i was denied employment into the…because they thought such opportunities were meant for men… (fp9) …a woman cannot stay late at work with her male boss…because everyone…would say that she is having a secret affair with him… (fp15) …we still have both men and women directors and heads…who intentionally want to work with only men and would always choose men over women… (fp22) masculinity (commonly associated with men’s biological traits) is a culture in nigeria that emphasises the role of men and their superiority over women. women are “gender socialised” (orientated) to see themselves as supports and subordinates to men. they are not given fair opportunities and privileges to participate and contribute as much as men, and they face unfair criticism for working very closely with their male counterparts. at times they are stigmatized with negative terms and social labels when they are deemed more successful than their peers. c. exo-endogenous factors the exo-endogenous factors comprise three main factors and five sub-factors. these main factors are as follows: i) sex appeal/sexual exploitation as a criterion for employment. ii) marriage and childbearing which imply availability for interstate transfers/training, and work flexibility. iii) culture/traditions are categorised into superiority syndrome and breadwinner syndrome from families' and employers’ perspectives. with a consistency rate of 66.7%, 100% and 69.1%, respectively (appendix b2), the participants suggested that the three main factors above create employment discrimination against women in their workplace, as cited in the excerpts below. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 221 …we men are very spoilt, just because women have “companies8” that can never fail…may decide to select the woman over the man or any other woman... (mp10) one of the factors affecting women’s employment is marriage, and pregnancy…that is why preference is given to men… (fp13) …our culture where men are masters and women are expected to be in the kitchen…our president said, women’s office is next to the restroom.… (fp25) these discriminations reduce women’s career advancement, dignity, employment access/opportunities and security. based on the excerpts above, women are not given equal support from their families and the civil service administration. identifying gender disparity issues the study presents gender disparity issues from two perspectives, systemic and individualdriven gender disparity issues. these are classified based on the identified exogenous and endogenous factors in the previous section. the first is the level of discrimination linked to the civil service hiring and employment strategy. the second one is based on individual relations and assessments in employment. a. systemic gender disparity the systemic form of gender disparity consists of two main categories and four subcategories. the first is gender disparity in recruitment, and the second is a glass ceiling9 and portfolio management10. while the former has no subcategories, the latter is further decomposed in gender disparity in career progression (86.67%), skill management (81.25%), development opportunities (60%), and employment portfolio (85.71%). with a consistency rate of 94.20%, the participants responded that discrimination against women in the nigerian civil service recruitment process is an ongoing issue that impedes women’s employment opportunities. women find it difficult to gain employment in the civil service as they only comprise 30% to 35% of total civil service employment, as indicated in the excerpt below. …men are about 65 to 70% of the total civil service employment…there has never been a time when the civil service employed more women than men… (fp25) 8 ability to provide sexual pleasure. 9 in the nigeria civil service, glass ceiling is a form systemic discrimination that limits mostly qualified women's opportunities to advance within the hierarchies of the various ministries and parastatals. 10 portfolio management refers to the level of discrimination in aligning women’s skills, experiences and qualification to roles that are appropriate for them. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 222 the participants, with a consistency rate of 80.95%, revealed that the civil service undermines women's potential by imposing a glass ceiling and discrimination in portfolio management. in summary, both the female and male participants made it clear that men are given better employment opportunities with stronger decision-making roles and positions of authority that matches their qualifications. women at times we relegated to minor decision-making roles and experienced mismatch in the fields of their expertise as expressed in the excerpts below. …women do not go further beyond their initial duties… some women may stay in a role for their entire tenure in the service even though they get promoted or they finish their training course…(fp18) … and even though there are women, most of them are in roles that are the complete opposite of their area specializations…(mp14) …if a man has to climb 10 stairs to get to where he wants to go, a woman will need to climb 20… we are mostly underrated... (fp18) …we only have 2 women as directors and the rest are all men, say about 16 plus...are you telling me that all the women died or resigned…(fp25) gender discrimination has indeed distorted the pool of human capital as women were not able to contribute to their fullest and optimal potential. another way to interpret the finding is that the role of women is primarily to provide support in ensuring the success of men in their careers. b. individual-driven gender disparity issues this form of gender discrimination is broken into one main and four subcategories. the main category is ego and non-compliance with regulations11. the other four subcategories are bullying of women (61.90%), inter-gender conflict (61.11%), sexual harassment (66.67%) and victimisation of women (80%). with a consistency rate of 73.33%, the participant explained that the values of a patriarchal society are well vested in the nigerian civil service. these values reinforce men’s perception (or ego) that they are superior to women. this is exacerbated by the lack of enforcement of gender discrimination policies within the civil service and thus allowing men to exercise certain levels of authority over women. some put constraints on their women counterparts to gain leverage in their career progression. this is shown in the following excerpts on the individual-driven gender disparity issues below. 11 ego and non-compliance to regulation is a situation where superordinate in the civil service would implement their personal agendas and sops rather than following or abiding to the regulations of the civil service. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 223 …they will intimidate you because…or makes people question their trustworthiness… (fp24) …sexual harassment is a big problem for women…it becomes a serious case for you… (fp13) …i have seen an instance when a man rebuke having a woman as a leader…they will make her life difficult, trust me... (fp24) the narratives showed that women are bullied, suppressed, and sexually harassed by their superordinate peers. their contributions are unappreciated, leaving them demoralized and demotivated. this is why many women left before they reached their full potential. these forms of gender disparities are imposed by individuals for personal reasons. forming conjectures/conclusions: gender inequality nexus the gender inequality nexus is the outcome of the qualitative study. linking gender disparity issues women experience in the civil service to the factors contributing to it using a within-method triangulation technique. the study identified several conjectures depicted in figure 2 below and discussed the findings thematically. the nexus between a, b and c (the gender disparity issues women experience in the civil service) and the numbers 1 to 7 (the factors in which gender inequality strives) were inductively identified based on the consistencies in the participants' narratives. a. the deficit in education12 the participants' narratives suggested that a deficit in education is the reason women struggled in components a and b. they highlighted the intrinsic role education play in the preparation and availability of women for employment. in nigeria, particularly in the north, women do not have equal access as men to education, and as a result, they are not well-equipped with the required skill set to compete with men. thus, allowing women to be discriminated against in recruitment, limiting their opportunities to break the glass ceiling and secure a better employment portfolio. b. religion religion was also linked to components a and b. the participants' responses showed that religious text strongly justifies and positions men as women’s superior, heads of families, spiritual 12 deficit in education can easily be a subfactor of culture and religion but we presented it as a factor on its own in this study. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 224 leaders, and leaders of social groups/tribes. women, on the other hand, are put in supporting roles such as that of a housewife. while a lot has changed and the interpretations have evolved, some practices persist. religion in the northern part of nigeria deprives women of social life and formal employment benefits. it moderates their role and influence in society. figure 2. the gender-inequality nexus. c. the operational structure of the civil service the participants perceived this component as a factor responsible for components a, b and c. they explained that the structure of civil service employment is biased towards men. it encourages male participation and treats masculinity as men’s biological trait. therefore, it recognises men as the preferred employees. as mentioned earlier, this allows them to obtain more employment opportunities and thus impedes the progress of female participation. d. stigmatisation against women the participants' response shows that negative assumptions and stigmas about women’s sexual morality and fidelity create the environment for components b and c to strive. it creates a toxic andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 225 working environment where women have to monitor themselves to avoid being called derogatory terms. this affects their work efficiency and performance because some of the jobs are important to them, and the negativity affects their personal lives (marriages and social status). e. sex appeal/exploitation as a criterion for employment factors 3, 5, and 7 have more relevance to female labour participation. it revealed that decisionmaking on employment is over-dominated by some men who make questionable hiring and administrative decisions. this factor does not affect every woman’s opportunity and experience in employment, but it encompasses all the gender disparity components (a, b and c, see figure 2). some female respondents also attributed their lack of career advancement to standing up against sexual harassment. the other silent few are willing to quit their jobs to avoid the repercussions. f. marriage and childbearing a woman’s career path is impeded due to structural failures that allow systemic restriction on women in the civil service. for example, the ability to resume service after maternity leave is often factored in as an employment criterion. marriage and childbearing provide favourable conditions for components a and b, as suggested by the participants. however, some women also opted to leave their service due to family obligations (individual factors) which may not be due to the structure of civil service. g. culture and tradition the participants’ narratives suggested that components a, b and c manifest in the civil service because of culture and tradition. the findings show that despite nigeria’s proclaimed commitment to a non-discriminatory decree in section 2 of the 1999 constitution, the civil service is yet to achieve gender equality in recruitment and employment. the discriminatory culture that is based on gender has continuously prevailed within the walls of the civil service. the next section integrates the findings from the gender inequality nexus (qualitative analysis) with the finding from panel data regression (quantitative study). andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 226 4.3 triangulation: combining findings from quantitative and qualitative sections the constraints imposed by gender inequality in the nigerian civil service and on economic growth stem from exogenous, endogenous, and exo-endogenous factors. while the exogenous factors distort the pool of human capital available for employment, it affects the quality of the female labour force and their participation, therefore eliminating the benefits of perfect competition in the labour market. the results showed that gender biases are in favour of men’s employment irrespective of their capabilities and competencies, thus, men with lesser capabilities have better chances than women with better qualifications. the effect of the endogenous factors is more on women’s career prospects, empowerment, and ability to realise their potential. the civil service is not fully utilising and benefiting from the pool of human capital and resources available to it. the underlining theory from the triangulation exercise can be seen in figure 3 below. figure 3. gender inequality and economic growth causal relationship examining the left arrows of figure 3, our theory states that when the factors responsible for gender inequality increase, it will also promote gender disparity. this, in return, will negatively impact female labour participation in the civil service and further affect economic growth negatively. conversely, when there is a decrease or decline in the factors responsible for gender inequality via policy measures, the gender disparities will decrease, thereby improving women's participation and experience in the civil service and, overall, increasing economic growth. in detail, the empirical study shows that when female labour participation in the junior cadre increases, economic growth also increases. however, when less capable men are employed at women's expense, it would negatively impact economic growth13. although the contribution of female labour participation to economic growth is positive, it is at a low rate compared to the necessitating percentage increase in the women’s workforce. findings from the interview show that women are not allowed to explore their potential actively and fully, they are relegated to background work, and not given equal opportunities to develop themselves and earn as much as men. this results in some women being demotivated, oppressed and relegated to less impacting 13 the findings in the qualitative study under the operational structure of the civil service on page 24 showed that men are given more employment than women as a result of their gender and misconception of masculinity and femininity together with the result on gender inequality from the quantitative study on page 13 to 19. this result is then compared to the findings in the quantitative section where female labor participation increases economic growth. therefore, by transitive property, we arrive to the aforementioned conclusion. andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 227 roles in the civil service. this will make them leave the labour market, hence reducing female labour participation. again, this negatively affects economic growth. an important caveat from the empirical study is that only women in the junior cadre have statistically contributed to economic growth. this, however, does not mean women in the senior cadre did not contribute. qualitative findings suggest that women may get promoted, and receive an increment in income, but do not experience significant changes in their job functions. they are still discriminated against and positioned in roles lower than their expected level of job functions. factoring in incidences of sexual harassment and family obligations, the cycle repeats as they leave their employment due to demotivation and, thus, are unable to realize their full potential in contributing to economic growth within the civil service. 5. conclusion and recommendations the study employed explanatory sequential mixed-mode methods to investigate the impact of gender inequality in the nigerian civil service on the economic growth of nigeria. panel-data estimation was used to examine the link between gender inequality and economic growth, while qualitative analysis accounts for additional dimensions of gender inequality that were not accounted for in the empirical model. the study found that although an increase in female labour participation would positively impact economic growth, gender inequality would negatively impact economic growth. with systemic and individual-driven gender-based discrimination in effect, women are unable to participate and contribute as much as they would to economic growth. the gender inequality nexus showed that the operational structure of the civil service, sex appeal/exploitation as a criterion for employment, and culture and tradition impose higher challenges on female labour participation. to manage and mitigate the negative effects of gender inequality in the civil service, the study recommends that the government of nigeria go after the root of the problem by putting in place a well-coordinated policy framework that will continuously encourage female education and ensure women's employment security. empowering women through formal education is important. having gender-sensitive school curriculums also helps mitigate the issues of gender inequality in the long run since the study shows that the deficit in education constitutes a condition in which inequality strives. it should ensure that women are particularly aware of their rights with an emphasis on gender equality while making any form of gender discrimination in employment a punishable offence. the government should consider outsourcing its recruitment and appraisal processes to a third party, with gender equality being their key performance indicator. systems enabling women to return to work after maternity leave or employment gap due to family obligations should also be part of the workplace policy. while some factors such as religion, culture and tradition may be andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 228 difficult to alter, efforts to ensure a change in mindset can be made through public service campaigns, social media and also academic content in schools and universities. in the wake of the digitalization phenomenon, women are increasingly being empowered by utilising technology to increase productivity. this aspect was also not captured in this study. the way forward, therefore, is to study enablers of women empowerment in an increasingly digitalized economy. this study also only sheds light on the role of public services in empowering women. the study did not account for women’s experiences in other sectors. constraints on the accessibility of state-level output data and the availability of disaggregated time series data at longer periods made restricted the use of different statistical approaches. therefore, the study can be extended to other sectors/industries to allow more data to be pooled and used as a controlled variable to allow for more comparative analysis. references adegoke, f., adegoke, o. j. & oyedele, j. b. (2016). factors influencing women participation in home ownership in nigeria. property management, 34(1), pp. 67-78. https://doi.org/10.1108/pm-12-2014-0053 adenuga, a. a., & chike, f. o. (2013). the effect of internal revenue generation on infrastructural development. a study of lagos state internal revenue services. journal of education and social research, 3(2), 419 436. https://doi.org/10.5901/jesr.2013.v3n2p419 agboola, t. o. (2018). effectiveness of integrated personnel and payroll information system in addressing ghost worker syndrome in nigerian public sector. global journal of human social science (h), vol 18, no 2-h. no. 2-h. 2018. ali, m. (2015). effect of gender inequality on economic growth. case of pakistan. journal of economics and sustainable development, 6(9), 10-10. al-shammari, n. & al-rakhis, m., (2017). impact of gender inequality on economic growth in the arab region. research in applied economics, macrothink institute. braunstein, e. (2012). neoliberal macroeconomics. a consideration of its gendered employment effects. unrisd research paper 2012-1, geneva: unrisd brummet, q. 2008. the effect of gender inequality on growth: a cross-country empirical study. honours projects paper 88, http://digitalcommons.iwu.edu/econ_honproj/88 central bank of nigeria (2018; 2016; 2015; 2013; 2012; 2011; 2009). annual report. central bank of nigeria. cida (2012). gender inequality issues. new york, cida. crestwell, j. w. (2012). educational research: planning, conducting and evaluating quantitative and qualitative research. 4th ed boston, ma pearson. cuberes, d. & teignier, m. (2016). aggregate effects of gender gaps in the labour market: a quantitative estimate. journal of human capital, 10(1), 1-32. https://doi.org/10.1086/683847 dfid (2012). human right approach to school protection. oxford university press. https://doi.org/10.1108/pm-12-2014-0053 https://doi.org/10.5901/jesr.2013.v3n2p419 http://digitalcommons.iwu.edu/econ_honproj/88 https://doi.org/10.1086/683847 andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 229 devadas, s. & kim, y. e. (2020). exploring the potential of gender parity to promote economic growth. research & policy briefs, world bank malaysia hub. world bank group. enyew, e. b. & mihrete, g. a. (2018). liberal feminism: assessing its compatibility and applicability in ethiopia context. international journal of sociology and anthropology. fatile, j. o. & adejuwon, k. d. (2010). gender issues in human resources management in the nigerian public service. african journal of political science and international relations 5(3), 112-119. igbokwe-ibeto, j., florence, a., & osawe, c. (2015). work content in the nigerian civil service and its implication on sustainable development. singaporean journal of business economics, and management studies, 4(4). https://doi.org/10.12816/0017706 klasen, s. (2018). the impact of gender inequality on economic performance in developing countries. annual review of resource economics. https://doi.org/10.1146/annurev-resource100517-023429 kpanja k. l. & umar u. s. (2018). impact of gender imbalance on civil service productivity: implication for counselling. international journal of innovative social sciences and humanities research, 6(2), 45-51. lofstrom, a. (2008). gender equality, economic growth and employment. swedish ministry of integration and gender equality. mankiw, n. g. romer, d. & weil, d. n. (1992). a contribution to the empirics of economic growth. the quarterly journal of economics, 107 (2), 407-437. https://doi.org/10.2307/2118477 morrison, a., raju, d., & sinha, n. (2007). gender equality, poverty and economic growth (policy research working paper no. wps4349). washington dc: world bank. https://doi.org/10.1596/1813-9450-4349 nbs (2013; 2014; 2015; 2016; and 2018). statistical report on women and men in nigeria. national bureau of statistics. nienaber, h. & moraka, nv (2016). feminism in management research: a route to justly optimise talent. acta commercii, 16(2), 139-163. https://doi.org/10.4102/ac.v16i2.417 oboh, s. j., othman, z., kam, a. j. y. & jusoh s. (2022). does gender inequality matter in employment? a contentand consistency-based analysis of female labour participation in the nigerian civil service. african journal of gender, society & development, 11(2), 79-102. https://doi.org/10.31920/2634-3622/2022/v11n2a4 oboh, s. j., othman, z., kam, a. j. y. & jusoh s. (2021). gender inequality nexus in employment: a contentand consistency-based analysis of female labour participation in the civil service. african journal of business and economic research (ajber), vol. 16(2). https://doi.org/10.31920/1750-4562/2021/v16n2a13 oboh, s. j., kam, a. j. y., zarina, o. & sufian, j. (2019). gender inequality and economic growth in the nigerian civil service. international journal of innovative technology and exploring engineering, 9(2). ogunyemi, c. b. (2014). the configuration of gender and identity in nigerian literatures. international journal of gender and women's studies, 2(2), 43-59. https://doi.org/10.12816/0017706 https://doi.org/10.1146/annurev-resource-100517-023429 https://doi.org/10.1146/annurev-resource-100517-023429 https://doi.org/10.2307/2118477 https://doi.org/10.1596/1813-9450-4349 https://doi.org/10.31920/2634-3622/2022/v11n2a4 https://doi.org/10.31920/1750-4562/2021/v16n2a13 andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 230 okongwu, o. (2020). are laws the appropriate solution: the need to adopt non-policy measures in aid to the implementation of sex discrimination laws in nigeria? international journal of discrimination and the law, 21(1), 26-46. https://doi.org/10.1177/1358229120978915 oloni, e. f. (2015). gender inequalities and employment in nigeria research. research gate. otteson, j. r. (2004). adam smith selected philosophical writings. library of scottish philosophy. imprint academic. tong, r. (2014). feminist thought: a more comprehensive introduction 4th ed. boulder: westview press. united nations office on drugs and crime (2020). gender and corruption in nigeria. unodc research. united nations sustainable development framework 2018-2030 (2017). government of jordan and the united nations. ministry of planning and international cooperation. https://doi.org/10.1177/1358229120978915 andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 231 appendix a1. descriptive statistics appendix a2. correlation table appendix a3. fixed effects model (robustness check) andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 232 appendix b1. gender disparity issues consistency table code descriptions appendix b2. factors responsible for inequality consistency table andrew jia yi kam et al. / european journal of government and economics 11(2), december 2022, 210-233 233 code descriptions appendix b3. gender inequality nexus consistency table abstract 1. introduction 2. literature review 3. methodology 4. findings 5. conclusion and recommendations references appendix a1. descriptive statistics 231appendix a1. descriptive statisticsappendix a2. correlation table appendix a3. fixed effects model (robustness check) appendix b1. gender disparity issues consistency table appendix b2. factors responsible for inequality consistency table appendix b3. gender inequality nexus consistency table long-term relationship between investment and economic growth: a cointegration analysis of oecd countries © the author(s) 2023. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 12, no. 2 (2023), pages 175-195 https://doi.org/10.17979/ejge.2023.12.2.9909 submitted: sep 3, 2023 accepted: nov 14, 2023 published: dec 5, 2023 article long-term relationship between investment and economic growth: a cointegration analysis of oecd countries fisnik morina, 1 valdrin misiri,1,* fitore gashi 1 1 university “haxhi zeka”, peja, republic of kosovo, albania *correspondence: valdrin.misiri@unhz.eu abstract. this article investigates the impact of domestic investments on economic growth in oecd countries from 2000 to 2020. it identifies factors affecting economic growth and analyzes the relationship between domestic investments and economic growth using econometric models. empirical data from the world bank, imf, and oecd reports supports the positive impact of domestic investments on economic growth in oecd countries. the study also finds a long-term causality between gdp and gross fixed capital formation (gfcf). these findings offer valuable insights into investment dynamics and their effects on economic growth, informing governments and policymakers in oecd countries. keywords: investments, economic growth, oecd countries jel classification: e2; g11; e42; f3; f43 1. introduction investments are the process of engaging and mobilizing money or other productive resources at the present moment with the hope and expectation of securing some anticipated benefits in the future. in other words, investment is any expenditure that is made expecting benefits in the future. according to the well-known theorist massé (1962), investments represent the exchange of a safe good, the consumption of which is given up, with the hope we have in the investment object. similarly, sharpe (1998) states that investment is the sacrifice of a present value for a possible future value. according to the same authors, the difference between investment and saving is made. where saving is presented as giving up consumption, investment is made to increase output in the future. economic growth can be defined as an expansion in the potential gdp of a country. in other words, economic growth is defined as the ability of a country's economy to produce more and more goods and services that consumers want. it is one of the central macroeconomic stabilizing objectives because this fact is closely related to the standard of living of a country's population. an increase in the population's standard of living necessarily requires an increase in the production of goods and services. https://creativecommons.org/licenses/by-nc/4.0/ 176 morina, misiri and gashi the first theories of economic growth by smith (1776) and malthus (1820) started from the hypothetical situation when there was no private ownership of land and no capital accumulation, which implies that prices were formed only based on the labor consumed. the national product aligned with population dynamics, while the actual income per worker was constant. in particular, malthus (1820) claimed that population pressures would break the economy to the point where the workers would find themselves at the minimum threshold of existence. therefore, the primary purpose of this research is to study the cointegration (causal relationship) between domestic investment and economic growth in oecd countries. another goal of this study is to analyze other macroeconomic and macro-financial factors that influence the causal relationship between investment and economic growth. we suggest the following research hypotheses to investigate the relationship between investment and economic growth: • h1: capital investments (gfcf) positively impact economic growth in oecd countries. • h2: investments by sector positively impact economic growth in oecd countries. • h3: investments according to assets positively impact economic growth in oecd countries. this study is divided into the following sections. the first section of the study contains a review of the literature on the effects of domestic investments on economic growth; the second section includes a meta-analysis of the relationship between investments and economic growth; the fourth and fifth sections cover the data methodology and empirical results and findings of the study; and the last parts contain the conclusions, discussions, and references. 2. literature review in the last decades, several studies have been carried out by different authors, who have analyzed and identified the impact and relationships of investments in the economic growth of a country or several countries, giving different views. several studies from the empirical literature have found favorable correlations between investment and economic growth (starker & khan, 2020; abbes et al., 2015; klonowski, 2010; islam et al., 2020; fuente-mella et al., 2019; dinh et al., 2019; almfraji et al., 2014; brown & ulijn, 2004). an investment is an asset or item acquired to generate income or appreciation. appreciation refers to an increase in the value of an asset over time. when an individual buys a good as an investment, the intention is not to consume the good but to use it in the future to create wealth. an investment is always about spending some capital today time, effort, money, or an asset hoping for a greater return in the future than what was initially put in (hayes, boyle & rathburn, 2021). specifically, an investment is the actual commitment of money for some time in order to obtain future payments that will compensate the investor for the time the funds are committed (reilly, keith, & sanford, 2020). according to starker & khan (2020), foreign direct investment (fdi) is among the most critical factors influencing economic growth. in the absence of adequate domestic investment, fdi has been withdrawn from industrially advanced countries to accelerate the path of industrialization, long-term relationship between investment and economic growth promote and maintain sustainable economic growth, and reduce unemployment. furthermore, the effectiveness of fdi in host countries depends on the efficiency of domestic investment. today, the importance of fdi has increased through technology transfer and market networks that can result in efficient production and sales globally. indeed, fdi is increasingly sought by both developed and developing countries and is no longer considered a factor of dominance but a significant channel for technology transfer and innovation. thus, the global economy has been completely transformed in recent years. it operates in an increasingly confused environment as free trade and free movement of capital and goods become hallmarks, where fdi is increasingly seen as a new way of financing economic growth (abbes, mostéfa, seghir, & zakarya, 2015). developing new entrepreneurial ventures plays a crucial role in shaping any national economy. entrepreneurial firms are a source of growth and innovation in the industry for owners and provide jobs for the local population. they are also believed to offset economic downturns and help restructure existing industries. venture capital is generally defined as capital provided to private businesses to accelerate a firm's development through access to capital and a wide range of business support services (klonowski, 2010). investments can arise for various reasons, but mainly, a company will invest to achieve a return on that investment in the form of profit shares or dividends (collings, 2016). savings and investments are two macroeconomic aggregates that are important in gdp growth in the long run. these factors relate to the allocation of resources between different periods. one method of increasing future product productivity is to allocate current resources to increase the capital stock, which is achieved by saving a portion of their current income to finance investment to grow. however, there is evidence that capital accumulation increases productivity and a consensus that higher levels of investment accelerate economic growth. increasing the capital stock increases productivity and accelerates gdp growth (fuente-mella, vallina-hernandez, & fuentes-solis, 2019). the importance of savings in developing countries depends on the long-debated economic theory that the rate of economic growth is a function of the rate of investment and that the rate of domestic savings limits investment. savings contribute to economic growth by freeing up resources that can then be used to increase the economy's productive capacity by increasing the amount of capital equipment, machinery, and buildings (joshi, pradhan, & bist, 2019). foreign direct investment (fdi) has emerged as a significant economic linkage between developed and developed and developing economies (strange, 2003). in theory, fdi directly affects growth through capital accumulation and incorporating new inputs and foreign technologies into the host country's production function (almfraji, mohammad, & khalid, 2014). moreover, economic growth can be a source of foreign direct investment, a country's infrastructure development, and a solid tax base (brown & ulijn, 2004). the relationship between foreign direct investment (fdi) and economic growth has attracted much attention from academics and governments of developing countries. since economic growth is one of their main focuses, policies related to attracting fdi have been prioritized during economic growth and development. from another point of view, it can be affirmed that fdi is one of the decisive factors for economic integration, as it increases benefits and long-term ties between different countries (dinh, vo, the vo, & nguyen, 2019). 178 morina, misiri and gashi foreign direct investment (fdi) is often seen as a massive economic boon. fdi has long been widely accepted as an enabler for the sustainable growth of an economy. given the importance of fdi, economies – especially developing ones – should attract more fdi. although many factors as attractors of fdi have been highlighted, financial development's (fd) impact has been the least explored in the finance and fdi literature. while financial development is valued as a country's increased provision of financial goods and services to its citizens and enterprises, a developed financial system primarily symbolizes confidence for foreign investors. most importantly, the financial system functions as an allocative resource, provides information, and functions as a costreduction mechanism (islam, khan, popp, sroka, & oláh, 2020). 3. meta-analysis: analysis of the cointegration between investment and economic growth this section presents a comprehensive meta-analysis consolidating a substantial body of research conducted by various researchers concerning the relationship between investments and their impact on economic growth. table 1 provides a concise overview of these studies, detailing the variables employed and the specific econometric methodologies utilized by the respective authors. the summary encapsulates the pivotal findings and results of each study, elucidating the nuanced dynamics between investments and economic growth. furthermore, the subsequent section delves into an in-depth examination of select studies by other esteemed researchers, offering a detailed scrutiny of their findings and results within the context of this thematic domain. this critical analysis aims to provide a comprehensive understanding of the nuances and complexities inherent in the relationship between investments and economic growth as perceived by other scholars. by undertaking this meticulous review, this research seeks to contribute to the existing discourse and offer valuable insights that can inform future research directions and policy formulation in the field of economic development. table 1. existing literature authors years variables methods findings dinh, vo, the vo, & nguyen (2019) 20002014 real gdp per capita, foreign capital flows, domestic credit to the private sector, total domestic investments, human capital panel-based unit root test, johansen cointegration test, vector error correction model (vecm), and fully modified ols (fmols) the results of this study show that fdi helps stimulate economic growth in the long run, although it has a negative impact in the short term for the countries in this study. other macroeconomic factors also play an essential role in explaining economic growth in these countries. abbes, mostéfa, seghir, & zakarya (2015) 19802010 gdp: gross domestic product fdi: foreign direct investment unit root tests in panel series, cointegration tests, vector long-term cointegration, and panel granger causality test the results show a disparity in the relationship between panel study cointegration. the results also show a unidirectional causality from fdi to gdp, which can be an excellent tool to prioritize allocating resources across sectors to promote foreign direct investment. long-term relationship between investment and economic growth authors years variables methods findings szkorupová (2014) 20012010 gdp, foreign direct investments and exports. johansen test for cointegration, vector error correction (var) model. the results confirm the existence of long-term causal relationships between the studied variables in slovakia. the study also reveals a positive impact of foreign direct investment and export on gross domestic product. sunde (2017) 19942013 gdp, foreign direct investments and exports. unit root tests, bounds test for cointegration and causality within ardl modeling approaches. the article confirmed the cointegration between economic growth, foreign direct investment, and exports. causality analysis found unidirectional causality between economic growth and fdi, unidirectional causality between foreign direct investment and exports, and bidirectional causality between economic growth and exports. kumari & sharma (2018) 19812013 electricity consumption (elc), foreign direct investment (fdi) and economic growth (gdp). adf unit root test, johansen cointegration approach, granger causality and vecm. the study findings show that the three variables are stationary at the first difference level from the johansen cointegration approach, confirming two cointegrated equations between the study variables. the granger causality analysis confirms two unidirectional causalities, going from elc to gdp and gdp on fdi, and a bidirectional causality between elc and fdi. pegkas (2015) 20022012 gross domestic product and fdi. descriptive statistics, stationary panel tests, panel cointegration tests, fixed and random effects methods. empirical results show a positive relationship between fdi and economic growth. the empirical analysis reveals a positive long-run cointegrating relationship between the stock of fdi and economic growth. also, the fixed and random methods of assessing country effects show that foreign direct investment is an essential factor that positively affects economic growth. abdelhafidh (2013) 19702008 domestic savings, foreign capital inflows and economic growth. augmented dickey– fuller (adf) test and causality tests. based on the study's findings, growth performance can be directly enhanced by a more significant mobilization of various foreign capital inflows, ranging from grants, fdi, long-term loans, short-term loans, bilateral loans, multilateral loans, and bank loans. pece, andreea, oros & olivera (2015) 20002013 gdp, number of patents, number of trademarks, research and development expenditure eur/capita, research and development expenditure. analysis was performed using multiple regression models. the results show a positive relationship between economic growth and innovations. furthermore, we found that foreign direct investment significantly impacts economic growth through the transfer of knowledge and the improvement of technological processes. 180 morina, misiri and gashi authors years variables methods findings gobalet & diamon (2014) 19501980 economic development, specialization in mining, internal capital formation, dependence on investment. panel regression. from the research results, investment-dependent countries have tended to experience relatively less subsequent economic growth. the most substantial adverse effects depend on the period, and the negative consequences of fdi can lie in the structural distortions of the country's economy, which appear slowly but prove harmful in the long term. hong (2014) 19942010 foreign direct investments (fdi), economic growth (gdp). system evaluation results – gmm. based on the generated research results, it was found that fdi has a positive impact on economic development. moreover, economies of scale, human capital, infrastructure, wage levels, and regional differences actively interact with fdi and drive economic growth in china, while trade openness does not significantly drive fdi. ribaj & mexhuani (2021) 20102017 savings and economic growth. augmented dickeyfuller tests, johansen cointegration tests and ganger causality test. the regression results showed that deposits have a significant positive impact on the economic growth of kosovo because savings stimulate investments, production, and employment and consequently generate more significant economic growth. in addition, loans and remittances also help grow kosovo's economy through their direct impact on investment. joshi, pradhan & bist (2019) 19752016 savings, investments and economic growth. johansen, gregory– hansen and ardl cointegration tests. the study's empirical results show a long-term stable relationship between savings, investment, and economic growth. similarly, the long-run estimates of the ardl model show that investment has a significant positive impact on economic growth. wigren & wilhelmsson (2007) 19802004 gdp, total investment, residential, buildings and infrastructure. cointegration and error correction model. from the research results, investments in the construction of housing or other types of buildings, together with investments in infrastructure, are assumed to have a direct and indirect effect on economic growth. therefore, it is essential to analyze its effect on economic growth. meyer & sanusi (2019) 19952016 economic growth, internal investments (gross fixed capital formation), employment register and export. adf and pp unit root tests, johansen cointegration tests, vec granger causality test and pairwise granger causality test. empirical research findings suggest a long-term relationship between domestic investment, employment, and economic growth, with causality from economic growth to investment and not vice versa. the results also show that investments have a positive long-term impact on employment. empirical evidence further suggests bidirectional causality between employment and economic growth. long-term relationship between investment and economic growth authors years variables methods findings nguyen & nguyen (2021) 20002020 economic growth, public investment, private investment, fdi, regular expenditure, total value of imports and exports and employment. descriptive statistics, correlation and robustness test. the study results show that factors such as labor and trade openness negatively impact economic growth in the short term. public investment harms economic growth in the long run, while domestic private investment, foreign direct investment, trade openness, and labor positively affect economic growth. tabassum & ahmed (2014) 19722011 economic growth, foreign direct investment, domestic investment, import and export. multiple regression method. the research results show that domestic investments positively impact economic growth, while foreign direct investments and trade openness are less critical. sothan (2015) 19802013 foreign direct investment (fdi), export and economic growth. panel unit root tests, panel granger causality, panel cointegration analysis and the fmols approach. the study's findings strongly confirm a long-run steady-state relationship between the selected countries' fdi, exports, and gdp. based on the granger panel causality analysis, a long-run bidirectional causality exists between fdi and gdp and exports and gdp. from this, it can be concluded that fdi and exports have a causal impact on longterm growth. alshehry (2015) 19702012 economic growth, foreign direct investment, gross fixed capital formation, and international trade. granger causality test, johansen cointegration tests, unit root tests and descriptive statistics. the study's empirical results suggest the existence of long-term unidirectional causality and longterm relationship between fdi and economic growth, a bidirectional causal relationship between capital investment and economic growth in the long run, and the one-way causality flowing from fdi to domestic capital investment. therefore, increased foreign and domestic investment promotes economic growth. owusu (2020) 19902016 foreign direct investment, credit to the private sector, trade openness, gross national expenditure and economic growth. ardl method of cointegration and uecm and granger causality test. the study results find a solid bidirectional causal relationship between fdi inflows and economic growth in the short and long run. the study also finds that the flow of fdi, credit for private international trade, and national spending promote economic growth both in the short and long term. bila (2020) 20022018 foreign direct investments (fdi) and gross domestic product (gdp). descriptive statistics, crosssectional dependence test statistics, panel cointegration tests (ardl model) and granger causality analysis. the study results found evidence that there is a long-term balance between gdp and foreign direct investment. also, the panel causality test showed that a relationship between gdp growth rate and fdi growth rate is only indirect. finally, only weak evidence showed that fdi had a statistically significant impact on gdp. source: own elaboration. 182 morina, misiri and gashi the meta-analysis reflected in table 1 contains a summary of several papers by different authors on the topic of investments and their impact on economic growth. from the analysis of their empirical results, a conclusion is reached that investments positively impact economic growth and a long-term positive cointegrating relationship between them. however, these studies have analyzed other macroeconomic factors and have impacted economic growth, including domestic credit, export, domestic savings, import, and human capital, implying that these interact actively with investments and promote economic growth. 4. research methodology and data 4.1 research methodology the primary purpose of this research is to analyze the cointegration (causal relationship) between domestic investments and economic growth of oecd countries. so, the main focus of this study is to identify the fact that this relationship exists and what is the impact of investments on economic growth. authors such as starker & khan (2020), abbes et al. (2015), bayar (2014), dinh et al. (2019), trinh & nguyen (2015), omri & kahouli (2014), naz et al. (2015), sunde (2017), and pegkas (2015), as well as many other studies by other authors, have analyzed the impact of investments on the economic growth of oecd countries. the primary importance of carrying out this study lies in the fact that the oecd countries, taking into account that they are countries with a relatively developed economy, then investment can be considered as a vital component for the economic growth of these countries, but also for indirect effects that are caused in the economies of developing countries. this research reflects how domestic investments (gfcf) have influenced economic growth in oecd countries. what is the impact of investments by sectors of the real economy on gdp in oecd countries, and what effect do investments by assets have on economic growth in oecd countries? the main hypotheses of this study are: • h1: capital investments (gfcf) positively impact economic growth in oecd countries. auxiliary hypotheses:  h1a: a positive correlation exists between capital investment (gfcf) and economic growth.  h1b: a positive causal relationship exists between capital investment (gfcf) and economic growth. • h2: investments by sector positively impact economic growth in oecd countries. auxiliary hypotheses:  h2a: a positive correlation exists between sector investment and economic growth.  h2b: there is a positive causal relationship between investment by sector and economic growth. • h3: investments according to assets positively impact economic growth in oecd countries. long-term relationship between investment and economic growth auxiliary hypotheses:  h3a: a positive correlation exists between investment by assets and economic growth. • h3b: there is a positive causal relationship between investment by assets and economic growth. quantitative methods have been applied to test these hypotheses based on the secondary data of the oecd, the imf, and the world bank annual reports. these data are mainly annual data presented in time series and numerical data. to analyze the impact of the domestic investments on economic growth, it was necessary to consider many other factors that explain the causal relationship between these variables. our econometric model is derived from various studies analyzing the impact of investments on the economic growth of oecd countries, conducted by different authors. the analysis of this research includes data from 37 oecd countries for twenty years (2000 2020). the data are processed in the stata program, and to test the validity of the hypotheses of this study, the following statistical tests were applied: var analysis (vector autoregression model), granger causality wald test, johansen cointegration test, multiple linear regression, effect random, fixed effect, hausman – taylor regression, gmm model – arellano bond estimation and generalized estimating equations (gee model) and nelson's e-garch. table 2. description of the variables included in the econometric model variables acronym measure evidence data source gross domestic product gdp total, us dollars/capita – oecd (abbes, mostéfa, seghir, & zakarya, 2015), (sunde, 2017), (wigren & wilhelmsson, 2007), (meyer & sanusi, 2019), (bilas, 2020). oecd annual reports (2000 – 2020) gross fixed capital formation gfcf total quarterly growth rate – oecd (dinh, vo, the vo, & nguyen, 2019), (pegkas, 2015), (abdelhafidh, 2013), (gobalet & diamond, 2014), (meyer & sanusi, 2019), (ugochukwu & chinyere, 2013) , (alshehry, 2015). oecd annual reports (2000 – 2020)) investments by sectors is percentage of gross fixed capital formation – oecd (bouchoucha & ali, 2019), (bayar, 2014), (naz, sabir, & ahmed, 2015), (tabassum & ahmed, 2014). oecd annual reports (2000 – 2020) investments by assets ia percentage of gross fixed capital formation – oecd (trinh & nguyen, 2015), (wigren & wilhelmsson, 2007), (joshi, pradhan, & bist, 2019), (dinh, vo, the vo, & nguyen, 2019). oecd annual reports (2000 – 2020) inflation inf annual growth rate – oecd (trinh & nguyen, 2015), (ugochukwu & chinyere, 2013), (naz, sabir, & ahmed, 2015). oecd annual reports (2000 – 2020) interest rate ir total, % per year – oecd (ugochukwu & chinyere, 2013). oecd annual reports (2000 – 2020) domestic credit dc percentage of gdp – world bank (owusu, 2020), (dinh, vo, the vo, & nguyen, 2019). oecd and world bank annual reports (2000 – 2020) source: own elaboration 184 morina, misiri and gashi to test the validity of the hypotheses of this study, we have built this econometric model as follows: 𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 = 𝛽𝛽0 + 𝛽𝛽1 𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛽𝛽2 𝐼𝐼𝐼𝐼𝑖𝑖𝑖𝑖 + 𝛽𝛽3 𝐼𝐼𝐼𝐼𝑖𝑖𝑖𝑖 + 𝛽𝛽4 𝐼𝐼𝐼𝐼𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛽𝛽5 𝐼𝐼𝐼𝐼𝑖𝑖𝑖𝑖 + 𝛽𝛽6 𝐺𝐺𝐺𝐺𝑖𝑖𝑖𝑖 + 𝛾𝛾𝑖𝑖𝑖𝑖 , where: gdp – gross domestic product gfcf – investments is – investments by sectors. ia – investments by assets inf – inflation ir – interest rate dc – domestic credit for the private sector γ – stochastic variables (other factors not considered in the model) i – code and t – time period. 4.2 descriptive statistics in this part of the econometric analysis, all the results of the descriptive statistics of the statistical tests for the variables that are part of the econometric model are presented. these descriptive statistics include minimum, maximum, mean, and standard deviation values. from the data generated from the descriptive statistics table (table 3), we can conclude that variables such as gdp, credit, and investments have the highest standard deviation value. in contrast, the variables with the smallest standard deviation values are the interest rate, inflation, investments by assets, and investments by sector. from this, we can conclude that the variables with a higher standard deviation value are more spread than the mean value. in comparison, the variables with a lower standard deviation value are considered to be distributed closer to the mean value. table 3. variable definitions and descriptive statistics variables obs. minimum maximum mean std. deviation gdp 777 6709 117721 35184.76 16786.22 gfcf 772 -20.4 313.9 1.555052 12.75924 is 709 3.11 51.23 24.1159 7.478691 ia 735 4.23 43.69 20.67887 7.249532 inf 777 -4.48 54.92 2.658443 3.892529 ir 713 -0.52 22.5 3.887363 2.640092 dc 689 0.2 304.6 96.28142 46.1832 source: own elaboration. long-term relationship between investment and economic growth 5. empirical results table 4 shows the regression analysis results using panel data and the assessment of the impact of investments on the economic growth of oecd countries. table 4 uses a variety of econometric approaches or measures, with the study's findings reflecting the same results from almost all panel data models that estimate the investment effect or relationship with economic growth in these countries. moreover, an essential element of this research is analyzing the causal relationship between investments and economic growth. time series model approaches were also used to analyze and prove the validity of the proposed hypotheses. these results show that all the variables included in this econometric model are significant at the 0.05 and 0.10 levels. therefore, we have based the results of the multiple linear regression model for interpretation purposes. according to the data presented in table 4, we can observe that some of the variables are statistically significant at the 95% and 90% confidence levels. the parameter coefficient b0= 39043.37 considers that if other factors are constant, the gdp value will be $39043.37 (unit). if the primary independent variable (gfcf) increases by 1 unit while keeping other independent variables constant, gdp will increase by 178.16 units. this statement is correct as the significance value (pvalue = 0.000 < 0.05) is at a statistical significance level. based on this result, we can observe that investments have positively influenced the economic growth of oecd countries. the primary hypothesis is rejected, and the alternative hypothesis is accepted. so, with the increase in investments, the oecd countries increase the capital stock, and according to this, the increase in the amount of capital available to an economy contributes to economic growth. table 4. results of the regression analysis variables linear regression random effects – gls regression fixed – effects regression hausman taylor regression gee model gmm model gdp 0.978821*** (0.000) gfcf 178.1594*** (0.000) 100.552*** (0.000) 97.92486*** (0.000) 98.89147*** (0.000) 100.3164*** (0.000) 2.766009 (0.548) is 269.7156** (0.025) 1468.535*** (0.000) 1681.159*** (0.000) 1551.936*** (0.000) 1486.855*** (0.000) -34.95403 (0.607) ia -231.1823* (0.062) 1198.149*** (0.000) 1408.499*** (0.000) 1278.423*** (0.000) 1216.183*** (0.000) 11.72929 (0.864) inf 257.3447 (0.521) 204.0757 (0.307) 200.3417 (0.310) 204.3697 (0.299) 203.8267 (0.355) 137.0001** (0.014) ir 3592.182*** (0.000) 2469.269*** (0.000) 2387.309*** (0.000) 2439.206*** (0.000) 2462.165*** (0.000) -100.3756* (0.065) dc 90.86103*** (0.000) 89.22334*** (0.000) 85.03547*** (0.000) 87.80784*** (0.000) 88.87247*** (0.000) 25.87507*** (0.000) const. 39043.37*** (0.000) 47950.86*** (0.000) 49404.94*** (0.000) 51394.49*** (0.000) 48031.39*** (0.000) r square 0.4258 0.6241 0.6253 adj. r2 0.4198 0.4174 0.3650 note: * p < 0.1; ** p<0.05; *** p < 0.01. source: own elaboration. 186 morina, misiri and gashi if investments by sector increase by 1 unit while keeping other variables constant, then gdp will grow by 269.72 units. this statement is correct since the level of statistical significance is above 90% (0.025<0.05). investment in different sectors includes portfolio diversification and risk management, where with risk management, investors tend to have a better performance in that investment, and this chain effect causes a positive impact on the economic growth of these countries. if investments according to assets increase by 1 unit while keeping other variables constant, then gdp will decrease by -231.18 units. this statement is correct since the significance value is below the level of statistical significance (0.062<0.10). in order to explain even more the impact of investments on the economic growth of the oecd countries, the effect of the inflation rate, the interest rate, and domestic loans were also taken into account in the analysis. if the inflation rate increases by 1 unit, keeping other factors constant, gdp will increase by 257.34 units. this statement is incorrect, considering that the significance level is above the level of statistical significance (p-value = 0.521 > 0.05). since we know that inflation is a continuous increase in the general level of prices, the increase in the inflation rate reduces the economic growth of the oecd countries. with higher inflation, employment in oecd countries decreases. consumers need more money to buy goods and services, so the economy of these countries can be low. if interest rates increase by 1 unit while keeping other variables constant, gdp will decrease by -3592.18 units. this statement is correct since the significance value is below the level of statistical significance (p-value = 0.000 < 0.01). interest rates not only affect economic growth but also the level of investment since businesses and consumers will reduce their spending when they increase. also, this will cause incomes to fall and thus affect the economic growth of these countries. the last independent variable is internal credits, where we mean that with the increase in internal credits by 1 unit, keeping all other variables constant, gdp will increase by 90.86 units. this statement is correct since the significance value is below the level of statistical significance (p-value = 0.000 < 0.01). in order to verify the validity of the hypotheses presented in this study and to give more support to the econometric results presented in the study's findings, we have reflected some additional analyses related to the analysis of the var model (vector autoregressive model). initially, the results of the three statistical tests were reflected in this part as: • johansen test for cointegration; • vector autoregressive models (var); • wald tests of granger causality. the first test, the johansen test for cointegration, analyzes whether the main variables in this study, such as "gross domestic product" (gdp) and "investment" in oecd countries, are integrated. in the second analysis, through the var method, we see whether the main variables explain each other. finally, the "granger causality wald test" as part of the time series model is reflected to verify whether these variables have long-term or short-term causality. long-term relationship between investment and economic growth johansen test • h0: there is no dynamic correlation and co-integrating relationship between gdp and investments • h1: there is a dynamic correlation and co-integrating relationship between gdp and investments table 5. johansen tests for cointegration trend: constant number of obs = 774 sample: 4 – 777 lags = 3 maximum rank parms ll eigenvalue trace statistic 5% critical value 0 10 -11058.774 . 214.2017 15.41 1 13 -10972.572 0.19968 41.7974 3.76 2 14 -10951.673 0.05257 maximum rank parms ll eigenvalue max statistic 5% critical value 0 10 -11058.774 . 172.4043 14.07 1 13 -10972.572 0.19968 41.7974 3.76 2 14 -10951.673 0.05257 source: own elaboration. table 5 reflects that the value of "trace statistics" is greater than the "critical value 5%" for ranking 0. we can say there is a co-integration between these variables – gross domestic product (gdp) and investment (gfcf). so, in this case, the primary hypothesis is rejected, and the alternative hypothesis (h1) is accepted. also, since the value of the "max statistic" is greater than the "critical value 5%", we can say that there is a co-integration between these two variables (gdp and gfcf). so, in this case, the alternative hypothesis (h1) is accepted. since the two main variables in this study are integrated, we can use the var model in the econometric analysis. based on the fact that gdp and gfcf were integrated, we can conclude that there is a long-term causality between gross domestic product and gfcf investments in oecd countries. below, the analysis through the var model is reflected, which proves the validity of the hypotheses if gdp, in the long term, explains the gfcf. vector autoregressive model (var) • h0: there is no long-term causality between gdp and investments; • h1: there is a long-run causality between gdp and investment; from table 6, we can conclude that gdp depends on the values of this variable in the periods (lag_1) in the long term since the significance values are at the standard level of 5%. • gdp-lag_1: (p-value = 0.000<0.05) • gdp-lag_2: (p-value = 0.673 > 0.10) • gdp-lag_3: (p-value = 0.124 > 0.10) 188 morina, misiri and gashi table 6. vector autoregressive model (var) coef. std. err. z p > |z| 95% conf. interval gdp gdp l1. .9772 .035899 27.22 0.000 .9068392 1.047561 l2. -.020772 .0492747 -0.42 0.673 -.1173486 .0758046 l3. -.0530826 .0345103 -1.54 0.124 -.1207217 .0145564 gfcf l1. 74.43553 20.46765 3.64 0.000 34.31967 114.5514 l2. -164.5353 21.5386 -7.64 0.000 -206.7502 -122.3205 l3. -.5613106 21.31825 -0.03 0.979 -42.34432 41.22169 _cons 3576.681 598.8376 5.97 0.000 2402.981 4750.381 source: own elaboration. based on the coefficient values, we can conclude that if gdp (lag_1) in the previous year increases by 1 unit, then gdp in the current year will increase by 0.9772 units. this statement is correct since the significance value is below the level of statistical significance (0.000<0.05). also, in (lag_2), if gdp increases by 1 unit, then actual gdp will decrease by -0.020772 units. this statement is incorrect, as the significance value is above the level of statistical significance (0.673>0.10). while in the period (lag_3), if the gdp will increase by 1 unit, then the current gdp will have a negative value of -0.05308 units. also, this statement is incorrect since the p-value is above the level of statistical significance (0.124> 0.10). meanwhile, the results of the var model show that gdp in oecd countries depends on the values of gfcf in the periods (lag_1) (lag_2) in the long term. gdp–gfcf (lag_1) is a significant variable, and this explains the dependent variable (gdp), as the significance value is less than 0.05 (p-value = 0.000 < 0.05). likewise, the variables gdp-gfcf (lag_2) reflect a significant variable that explains the dependent variable (gdp), considering the level of significance, which has a value of less than 5% (p-value = 0.000<0.05). such a result provides more empirical evidence for the integration of these two variables and the existence of causality in the long term. while for the variable gdp-gfcf (lag_3), there is no significant correlation since the significance level is above the standard level of 10% (p-value = 0.979 > 0.10). we applied the granger causality wald test to verify the hypothesis of any long-term causality between gdp and gfcf for oecd countries. granger causality wald tests • • h0: there is no long-term causality between gdp and investments; • • h1: there is a long-term causality between gdp and investments; long-term relationship between investment and economic growth table 7. granger causality wald tests equation excluded chi2 df prob > chi2 gdp gfcf 67.265 3 0.000 gdp all 67.265 3 0.000 source: own elaboration based on the results of the granger causality wald tests (see table 7), we can conclude that there is a long-run causality between gdp and investment – gfcf. so, we say there is a long-run causality between gdp and gfcf, so the primary hypothesis is rejected, and the alternative hypothesis is accepted (p-value = 0.000 < 0.05). the chi-square value between the two variables is 67.265. also, from the granger causality wald test results, we can observe a long-term causality between gdp and other variables. so, we can say there is a long-term causality between gdp – all since the significance value is below the 5% level (0.000<0.05). based on the econometric results of statistical tests, johansen test for cointegration, vector autoregressive test, and wald causality granger test, it is proved that the two main variables of this study (gdp and gfcf) are integrated. likewise, they are significant and explainable to each other, implying that there is an express causality between them in the long term. such a result is consistent with the studies of the authors bilas (2020), owusu (2020), alshehry (2015), sothan (2015), nguyen & nguyen (2021), meyer & sanusi (2019), pegkas (2015), kumari & sharma (2018), sunde (2017). table 8 presents the econometric results of nelson's e-garch statistical test. this econometric model has analyzed whether capital investments (gfcf) have a positive or negative impact on the volatility of economic growth in oecd countries. according to the econometric results of this statistical test, we can conclude that capital investments (gfcf) have positively influenced economic growth in oecd countries (p-value = 0.000 < 0.05). the effect of gdp is present in the independent variable (capital investments – gfcf) in the dynamic time lag l1. table 8. econometric results of nelson's e garch model between gdp and investments gdp coef. std. error z p > |z| 95% conf. interval 95% conf. interval gfcf 246.1699 28.24278 8.72 0.000 190.815 301.5247 _cons. 37527.42 184.4446 203.46 0.000 37165.92 37888.93 arch l1. .3175469 .1265185 2.51 0.012 .0695752 .5655187 earch l1. 1.282838 .246562 5.20 0.000 .7995853 1.766091 e-garch (l1) .9120341 .1983063 4.60 0.000 .5233608 1.300707 e-garch (l2) .0107963 .3079794 0.04 0.972 -.5928323 .6144248 e-garch (l3) -.2087136 .2129475 -0.98 0.327 -.6260831 .2086559 e-garch (l4) .2002932 .2046487 0.98 0.328 -.2008109 .6013974 e-garch (l5) -.0500757 .1952083 -0.26 0.798 -.4326769 .3325255 _cons. 2.196741 1.178594 1.86 0.062 -.11326 4.506742 source: own elaboration. in the constant (l1), we have a positive and significant correlation (p-value = 0.000 < 0.05). so, when capital investment in 2019 increases by 1 unit, then gdp volatility in 2020 has increased by 0.912 units. this economic phenomenon shows that the oecd countries must follow an efficient strategy in managing these investments because the growth of these investments in 2020 has influenced the increase in the volatility of economic growth. an increase in volatility means that a sudden increase in the risk of these investments negatively affects the economic growth of these countries. 190 morina, misiri and gashi figure 1. volatility of time series data for gdp and investment variables. source: own elaboration. in the constant (l2), we have a positive correlation, but not significant (p-value = 0.972 > 0.05). consequently, when capital investments in 2018 have increased by 1 unit, the volatility of economic growth has increased by 0.010 units. this statement is incorrect since the significance value is above the level of statistical significance. this economic phenomenon shows that these countries in the dynamic time delay (l2) can manage the risks that may occur in those investments. according to this result, the oecd countries, in the most extended periods, have carefully managed these risks in these investments, which has influenced the economic growth of these countries. referring to figure 1, we can notice that the time series data for gdp and investments have an accumulation of volatility because the periods when oecd countries are associated with high risk of these investments, then in these periods, these countries are characterized by a high volatility of economic growth. in continuation of nelson's e-garch analysis, it has been analyzed whether investments by sector have a positive or negative impact on the volatility of economic growth in oecd countries (table 9). table 9. econometric results of the nelson's e garch model between gdp and investments by sector gdp coef. std. error z p > |z| 95% conf. interval 95% conf. interval is 129.9488 16.5479 7.85 0.000 97.51552 162.3821 _cons. 33450.66 437.8684 76.39 0.000 32592.46 34308.87 arch l1. .4085706 .1727945 2.36 0.018 .0698997 .7472415 earch l1. 1.636798 .2993979 5.47 0.000 1.049989 2.223607 e-garch (l1) .7671487 .2045074 3.75 0.000 .3663216 1.167976 e-garch (l2) .0497486 .3463328 0.14 0.886 -.6290512 .7285484 e-garch (l3) -.0028908 .2341155 -0.01 0.990 -.4617488 .4559672 _cons. 3.277742 1.218685 2.69 0.007 .8891627 5.666321 source: own elaboration. according to the results of nelson's e-garch model, we can conclude that investments by sector influence gdp growth. so, with the increase in investments according to the sector, the economic growth in the oecd countries is positively affected. in the constant (l1), we have the presence of gdp and investments by sector. in the constant (l1), we have a positive and significant correlation. long-term relationship between investment and economic growth therefore, when investments by sector with a dynamic time lag (1) increase by 1 unit, the value of economic growth volatility will increase by 0.767 units. so, based on this econometric result, it can be concluded that the oecd countries have not effectively managed these investments because, as a result of the growth of these investments in various sectors, these countries have had an increase in the volatility of economic growth for one year. such an effect does not result in the constant (l2) and (l3). these constants have a positive and negative correlation but are not significant since the significance values are above the level of statistical significance. as a result, when investments by sector with a dynamic time lag (2) and (3) increase by 1 unit, the value of economic growth volatility will decrease by 0.0028 units. based on this econometric result, the oecd countries have followed a policy on investment risk management in the three years because, despite the risks that these investments may have, they positively affect the economic growth of these countries. table 10. econometric results of nelson's e – garch model between gdp and investments by assets gdp coef. std. error z p > |z| 95% conf. interval 95% conf. interval ia 366.0884 18.49781 19.79 0.000 329.8834 402.3435 _cons. 28805.92 420.4492 68.51 0.000 27981.85 29629.98 arch l1. .3944389 .1521463 2.59 0.010 .0962377 .6926402 earch l1. 1.798244 .2313236 7.77 0.000 1.344858 2.25163 e-garch (l1) .8118304 .1609492 5.04 0.000 .4963757 1.127285 e-garch (l2) .0006715 .2524377 0.00 0.998 -.4940972 .4954403 e-garch (l3) .0204547 .1994363 0.10 0.918 -.3704333 .4113428 _cons. 2.696959 1.07553 2.51 0.012 .5889583 4.80496 source: own elaboration. based on the econometric results of nelson's e-garch model (see table 10), we can conclude a positive correlation between investments according to assets and gdp. therefore, if investment by assets increases by 1 unit, then the value of gdp will increase by 366.08 units. the effect of gdp exists in the constant (l1). in the constant (l1), we have a positive and significant correlation. therefore, when investments according to assets with a dynamic time delay (1) increase by 1 unit, the value of the volatility of economic growth will increase by 0.811 units. based on this econometric result, the oecd countries have not effectively managed these investments, which has influenced these countries to have an increase in the volatility of economic growth for one year. such an effect does not result in the constant (l2) and (l3). these constants have a positive correlation but are not significant since the significance values are above the level of statistical significance. as a result, when investments according to assets with a dynamic time lag (2) and (3) increase by 1 unit, the value of economic growth volatility will increase. the statement is incorrect since the significance value in the three years is above statistical significance. from this, we can conclude that in more extended periods, oecd countries manage investments more effectively according to assets. 6. discussion researchers explore two hypotheses concerning the correlation between investments and economic growth, as investments essentially entail allocating a portion of existing capital, anticipating a 192 morina, misiri and gashi subsequent increase in value. the first hypothesis is that investments positively impact economic growth. the second is a positive correlation between investment and economic growth. based on all the statistical tests we have presented, we can conclude that investments positively affect economic growth, from which the first hypothesis was verified. the support of this conclusion is also found in the author's research (szkorupová, 2014), which, in order to prove the first hypothesis showing that investments have a positive impact on economic growth, used the data for the country of slovakia during the period 2001-2010. the study examines whether investment is strongly linked to economic growth. the author identifies a positive influence between investments and economic growth. evidence that generally strongly supports the view that investments have a positive effect on economic growth includes research by authors such as (pegkas, 2015; dinh et al., 2019; tabassum & ahmed, 2014). their data show a positive relationship between investments and economic growth, and they consider investments as one of the most critical factors that positively affect the economic growth of a country. as for the second hypothesis, which reflected that there is a positive correlation between investments and economic growth. in its confirmation was the research of authors (owusu, 2020) and (sothan, 2015), where from the results obtained, they proved a two-way solid causal relationship between foreign direct investments and economic growth. in confirmation of this hypothesis, it shows that in addition to investments affecting economic growth, it is also a fact that economic growth affects investments, which means that the countries with a higher economic development attract more investors to invest in that country. the growth of more investments means a more significant reason for reducing unemployment in that country and its economic growth. 7. conclusions and recommendations in conclusion, the entire research provides a detailed summary regarding the impact of investments on economic growth. the empirical results of this study have reflected various statistical tests that have been applied to examine the impact of investments on economic growth for oecd countries. based on oecd and world bank time series data covering all 37 oecd countries for 20 years, the research concluded that capital investment (gfcf) has positively influenced the economic growth of oecd countries, as well as being integrated, are also significant, explainable and there is a causality expressed in the long term. considering the impact of investments by sector and investments by assets, the empirical findings of this study state that this category of investments positively affects the economic growth of oecd countries. since these countries, through investments in different sectors, manage to diversify their portfolio and manage risks, the better risk management made the investors of these countries have a better performance in that investment, and from this chain, the effect has an impact on the economic growth of these countries. therefore, according to the findings, investments are a strong basis for a country's economy, especially countries that are still developing, and need to offer favorable conditions for investors not to hesitate or be reluctant to invest in those countries. therefore, as a recommendation, it is long-term relationship between investment and economic growth recommended that oecd countries implement policies to attract inflows of foreign direct investments, which will bring positive effects to the economic growth of these countries. we recommend that the governments of the oecd countries work on creating good economic stability so that investments are adequate; as a result, this would reduce unemployment and provide new jobs. also, the oecd countries would have to provide stability in price volatility, security in investments, fighting corruption, and other negative aspects that would hinder the economic development of a country. future research should focus on exploring the intricate dynamics between investments and economic growth within the evolving global market landscape and geopolitical climate. a more nuanced investigation into the impacts of different investment sectors and asset categories can provide valuable insights for policymakers and investors. however, it is crucial to acknowledge the limitations of this study, such as the reliance on secondary data and constraints associated with the chosen econometric models. additionally, the focus on oecd countries may restrict the generalizability of the findings. to address these limitations, future research should utilize diverse data sources and encompass a broader spectrum of economies, thereby enhancing our understanding of the complex relationship between investments and economic growth. references abbes, s.m., mostéfa, b., seghir, g.m., & zakarya, g.y. (2015). causal interactions between fdi, and economic growth: evidence from dynamic panel co-integration. procedia economics and finance, 276-290. https://doi.org/10.1016/s2212-5671(15)00541-9 abdelhafidh, s. (2013). potential financing sources of investment and economic growth in north african countries: a causality analysis. journal of policy modeling, 150-169. https://doi.org/10.1016/j.jpolmod.2012.01.001 almfraji, a., mohammad, a., & khalid, m. (2014). foreign direct investment and economic growth literature review from 1994 to 2012. procedia social and behavioral sciences, 206-213. https://doi.org/10.1016/j.sbspro.2014.03.668 alshehry, a.s. (2015). foreign direct investments and economic growth in saudi arabia: a cointegration analysis. iiste, 69-76. bayar, y. (2014). effects of foreign direct investment inflows and domestic investment on economic growth: evidence from turkey. international journal of economics and finance, 69-78. https://doi.org/10.5539/ijef.v6n4p69 bilas, v. (2020). fdi and economic growth in eu13 countries: cointegration and causality tests. journal of competitiveness, 47-63. https://doi.org/10.7441/joc.2020.03.03 bouchoucha, n., & ali, w. (2019). the impact of fdi on economic growth: evidence from tunisia. journal of smart economic growth, 23-46. brown, t.e., & ulijn, j. (2004). innovation, entrepreneurship and culture: the interaction between technology, progress and economic growth. usa: edward elgar publishing. https://doi.org/10.4337/9781845420550 collings, s. (2016). financial investments. john wiley & sons, ltd. https://doi.org/10.1002/ 9781119283393.ch15 dinh, t.t.-h., vo, d. h., the vo, a., & nguyen, t.c. (2019). foreign direct investment and economic growth in the short run and long run: empirical evidence from developing countries. risk and financial management, 176. https://doi.org/10.3390/jrfm12040176 fuente-mella, h. d., vallina-hernandez, a. m., & fuentes-solis, r. (2019). stochastic analysis of the economic growth of oecd countries. economic research-ekonomska istraživanja, 18489664. gobalet, j. g., & diamond, l. j. (2014). effects of investment dependence on economic growth. jstor terms and conditions, 412-444. hayes, a., boyle, m.j., & rathburn, p. (2021, august 20). investment. retrieved march 03, 2022, from investopedia: https://www.investopedia.com/terms/i/investment.asp https://doi.org/10.1016/s2212-5671(15)00541-9 https://doi.org/10.1016/j.jpolmod.2012.01.001 https://doi.org/10.1016/j.sbspro.2014.03.668 https://doi.org/10.5539/ijef.v6n4p69 https://doi.org/10.7441/joc.2020.03.03 https://doi.org/10.4337/9781845420550 https://doi.org/10.1002/9781119283393.ch15 https://doi.org/10.1002/9781119283393.ch15 https://doi.org/10.3390/jrfm12040176 https://www.investopedia.com/terms/i/investment.asp 194 morina, misiri and gashi hong, l. (2014). does and how does fdi promote the economic growth?evidence from dynamic panel data of prefecture city in china. ieri procedia, 57 62. https://doi.org/10.1016/j.ieri.2014.03.010 islam, m.a., khan, m.a., popp, j., sroka, w., & oláh, j. (2020). financial development and foreign direct investment-the moderating role of quality institutions. sustainability, 1-22. https://doi.org/10.3390/su12093556 joshi, a., pradhan, s., & bist, j.p. (2019). savings, investment, and growth in nepal:an empirical analysis. financial innovation, 1-13. https://doi.org/10.1186/s40854-019-0154-0 klonowski, d. (2010). the venture capital investment process. united states america: palgrave macmillan. https://doi.org/10.1057/9780230110076 kumari, a., & sharma, a. (2018). causal relationships among electricity consumption, foreign direct investment and economic growth in india. the electricity journal, 1-6. https://doi.org/ 10.1016/j.tej.2018.08.004 malthus, t. r. (1820). principles of political economy. england. massé, p. (1962). optimal investment decisions: rules for action and criteria for choice. united states: prentice hall; first edition (january 1, 1962). meyer, d.f., & sanusi, k. a. (2019). a causality analysis of the relationships between gross fixed capital formation, economic growth and employment in south africa. sciendo, 33-44. https://doi.org/10.2478/subboec-2019-0003 naz, s., sabir, g. m., & ahmed, m. (2015). impact of foreign direct investment on economic growth: empirical results from pakistan. international institute for science, technology and education, 101-105. nguyen, k., & nguyen, h. (2021). the impact of investments on economic growth: evidence from vietnam. korea distribution science association, 345-353. omri, a., & kahouli, b. (2014). the nexus between foreign investment, domestic capital and economic growth: empirical evidence from the mena region. research in economics, 257-263. https://doi.org/10.1016/j.rie.2013.11.001 owusu, e. l. (2020). the relationship between foreign direct investment and economic growth: a multivariate causality approach from namibia. wiley, 1-8. pece, m., andreea, s., oros, e., & olivera, s. f. (2015). innovation and economic growth: an empirical analysis for cee countries. procedia economics and finance, 461 467. https://doi.org/10.1016/s2212-5671(15)00874-6 pegkas, p. (2015). the impact of fdi on economic growth in eurozone countries. the journal of economic asymmetries. https://doi.org/10.1016/j.jeca.2015.05.001 reilly, f., keith, b., & sanford, l. (2020). investment analysis and portfolio management. boston. ribaj, a., & mexhuani, f. (2021). the impact of savings on economic growth in a developing country (the case of kosovo). journal of innovation and entrepreneurship. https://doi.org/10.1186/ s13731-020-00140-6 sharpe, w.c. (1998). investments. new jeresy: prentice hall. smith, a. (1776). the wealth of nations. london: w. strahan and t. cadell, london. sothan, s. (2015). foreign direct investment, exports, and long-run economic growth in asia: panel cointegration and causality analysis. international journal of economics and finance, 26. https://doi.org/10.5539/ijef.v8n1p26 starker, b., & khan, f. (2020). nexus between foreign direct investment and economic growth in bangladesh: an augmented autoregressive distributed lag bounds testing approach. financial innovation. https://doi.org/10.1186/s40854-019-0164-y strange, r.s. (2003). trade and investment in china. usa and canada: routledge. sunde, t. (2017). foreign direct investment and economic growth: adrl and causality analysis for south africa. research in international business and finance, 1-23. https://doi.org/10.1016 /j.ribaf.2017.04.035 szkorupová, z. (2014). a causal relationship between foreign direct investment, economic growth and export for slovakia. procedia economics and finance, 123-128. https://doi.org/ 10.1016/s2212-5671(14)00458-4 tabassum, n., & ahmed, s.p. (2014). foreign direct investment and economic growth: evidence from bangladesh. international journal of economics and finance, 117-135. https://doi.org/ 10.5539/ijef.v6n9p117 trinh, n.h., & nguyen, q.a. (2015). the impact of foreign direct investment on economic growth: evidence from vietnam. international institute for science, technology and education, 1-9. ugochukwu, u.s., & chinyere, u.p. (2013). the impact of capital formation on the growth of nigerian economy. research journal of finance and accounting, 36-42. https://doi.org/10.1016/j.ieri.2014.03.010 https://doi.org/10.3390/su12093556 https://doi.org/10.1186/s40854-019-0154-0 https://doi.org/10.1057/9780230110076 https://doi.org/10.1016/j.tej.2018.08.004 https://doi.org/10.1016/j.tej.2018.08.004 https://doi.org/10.2478/subboec-2019-0003 https://doi.org/10.1016/j.rie.2013.11.001 https://doi.org/10.1016/s2212-5671(15)00874-6 https://doi.org/10.1016/j.jeca.2015.05.001 https://doi.org/10.1186/s13731-020-00140-6 https://doi.org/10.1186/s13731-020-00140-6 https://doi.org/10.5539/ijef.v8n1p26 https://doi.org/10.1186/s40854-019-0164-y https://doi.org/10.1016/j.ribaf.2017.04.035 https://doi.org/10.1016/j.ribaf.2017.04.035 https://doi.org/10.1016/s2212-5671(14)00458-4 https://doi.org/10.1016/s2212-5671(14)00458-4 https://doi.org/10.5539/ijef.v6n9p117 https://doi.org/10.5539/ijef.v6n9p117 long-term relationship between investment and economic growth wigren, r., & wilhelmsson, m. (2007). construction investments and economic growth in western europe. journal of policy modeling, 439-451. https://doi.org/10.1016/j.jpolmod.2006 .10.001 https://doi.org/10.1016/j.jpolmod.2006.10.001 https://doi.org/10.1016/j.jpolmod.2006.10.001 1. introduction 2. literature review 3. meta-analysis: analysis of the cointegration between investment and economic growth 4. research methodology and data 5. empirical results 6. discussion 7. conclusions and recommendations references european journal of government and economics 11(1), june 2022, 7-30 european journal of government and economics issn: 2254-7088 immigration, ethnic fractionalization, and the fiscal burden in the oecd federico guerrero a, elliott parker a * a department of economics, university of nevada, reno, nv 89557 usa * corresponding author at: eparker@unr.edu abstract. this study considers the impact of immigration and ethnic diversity on government spending in 31 oecd countries over 25 years and compares the marginal effects for expenditures and revenues to approximate the fiscal burden. results suggest that ethnic fractionalization, not immigration itself, has a negative impact on spending in the oecd. on the whole immigrants tend to contribute more in taxes than they cause in expenditures, at least relative to the averages for the population as a whole, but this effect is reversed for immigrants from poorer countries. keywords. immigration; ethnic diversity; fractionalization; government revenues; government expenditures. jel codes. f22; h41; p50 doi. https://doi.org/10.17979/ejge.2022.11.1.8601 1. introduction immigration continues to be a salient and contentious issue in most countries. dempster et al. (2020) examine studies of global sentiment, and report that in 26 of 27 countries surveyed, more respondents wanted immigration into their countries reduced rather than increased. these same surveys, however, find that a plurality, and sometimes even a majority, remain in the “conflicted middle.” they recognize the economic value of immigrants, but they are hesitant about cultural and linguistic differences and also worry about labor market competition from immigrants and its effect upon wages. in addition to concerns about employment competition for native workers, there is also concern about the fiscal burden of immigration for the public sector. immigrants may increase the costs of education and other government services, particularly if they are from poorer countries, and the tax revenue they generate may take a generation to make up for it. dustmann and frattini (2014) estimated the marginal effects of immigration on the government budget, both in revenue and expenditures in the uk and concluded that immigrants from member states of the european economic area, particularly those from eastern europe, contributed a net fiscal surplus. however, they found that immigrants from other, poorer regions were a net fiscal burden, at least for the first generation. increased immigration may lead to a greater need for certain types of government spending, even though immigrants are often excluded from receiving benefits from the social safety net. © 2022. this work is licensed under a cc by-nc 4.0 license. mailto:eparker@unr.edu https://doi.org/10.17979/ejge.2022.11.1.8601 https://creativecommons.org/licenses/by-nc/4.0/ federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 8 immigration also directly increases the population of a country, and without an equal increase in government spending would lead to a decline in spending per capita. but immigration is also the chief source of rising ethnic diversity in many countries, and there is a literature arguing that voters are less likely to support many government services, and the taxes that pay for them, when they perceive those services being reallocated to other ethnic groups. of course, it may depend on the type of service, with less support for education, redistribution, and public infrastructure, and more support for policing and public safety. member states of the organization for economic cooperation and development (oecd) govern only 13% of the world’s population, but host over half of the world’s foreign-born residents and make up most of the countries surveyed for global sentiment. even eastern europe, which has far more emigrants than immigrants, is host to more than 20 million foreign-born residents (parker, 2020). the average income in these countries is almost three times that of the rest of the world, and they tend to have democratic governments that protect civil liberties and consider the preferences of voters when setting spending levels. these countries have other characteristics as well that help make them desirable destinations for migrants, and as a result they host four times as many foreign-born residents per capita as the rest of the world. in this study, we examine annual aggregate data for 31 oecd countries from 1995 to 2019, looking for evidence in country-level data that immigration alters government spending per capita. we also consider whether ethnic diversity, which we measure with the index of ethnic fractionalization, could be a better indicator of the impact. in the next section, we review some of the literature on how immigration and ethnic fractionalization may affect government spending. in section 3, we discuss our data and its sources, and we explain our regression approach. in section 4, we report our results. these results begin with a comparison between ols, between-country estimates, and both fixed and random effects. checking for consistency due to the correlation between immigration and fractionalization, we then use the random effects model to consider the differential effects of immigration on revenue and three broad categories of government spending. in section 5, we check our estimates for robustness, and we report estimates for marginal revenues and expenditures per immigrant. section 6 concludes the study. 2. the economic and fiscal impact of immigration many studies find that the benefits of immigration exceed its costs for most recipient countries. for example, edwards and ortega (2017) estimated a substantial economic impact for the us, as unauthorized immigrants contributed over 3% to gdp and would have contributed almost 5% if existing immigrants were legalized and more easily able to switch jobs. docquier et al. (2013) found that immigration into oecd countries increased the wages of less-educated natives and did not decrease native wages overall. using bilateral migration data for 194 countries, ortega and peri (2009) found that migration increases the differences in income per capita between origin federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 9 and host countries, and also increases employment in host countries without crowding out nativeborn workers. nonetheless, there is significant opposition to immigration in recipient countries. card et al. (2009) noted that although trade and migration have similar economic effects, public opposition to immigration is much stronger than it is for trade since it includes potential externalities on natives. using the european social survey, individual attitudes appear to be based mostly on these “compositional amenities,” especially for less-educated natives. can part of this antiimmigrant sentiment come from the fiscal burden that immigrants might create for the government? 2.1 fiscal costs of immigration even if immigration increases gdp for the host country, and even if more native workers may benefit from immigration than are harmed by it, there may still be other costs that lead to economic harm and political opposition. jaime-castillo et al. (2016) referred to this as the fiscal burden model, and using data from the european social survey found that anti-immigrant attitudes are strongly associated with the availability of social welfare benefits. based on a study of three latin american countries, noy and voorend (2016) argued that immigration causes a conflict between the specific rights of citizens and the social rights granted by the state to all residents. immigrants pay taxes and contribute to social welfare systems, and increased immigration may lead to a greater need for certain types of government spending even though immigrants are often excluded from receiving benefits from the social safety net. some types of spending may be for nonrival public goods that do not increase with population, while other public goods may be congestible, e.g., roads and highways that become crowded as the population grows. smith and edmonston (1997) reported estimates of the net fiscal burden of immigrants in both california and new jersey, with public expenditures rising in excess of revenues for state and local governments. this was because immigrants tend to have more children and thus need more public education services, and because immigrants tend to be poorer than average and thus pay less in taxes to regional governments and receive more in income transfers. however, immigrants tend to pay more to the federal government than they receive back, and they and their children earn more income over time. it also depends on the origin country, as immigrants from europe and canada are more likely to provide a net fiscal surplus than immigrants from central america. two decades later, a careful follow-up analysis that relied on accounting for age cohorts over time and the marginal costs of public goods estimated that immigrants in the us imposed a substantial fiscal burden, in the first, second, and even third generations (national academies of sciences, engineering, and medicine, 2017). the primary costs were for educating immigrant children and supporting impoverished retirees, while immigrants between the ages of 25 and 65 more than paid for themselves through their taxes. orrenius (2017) objected that this study estimated of the fiscal cost of immigrants by assigning average government spending per capita federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 10 to them. because a significant share of this cost comes from congestible public goods, the marginal cost of an immigrant was likely much lower, and the fiscal burden reversed. they noted, however, that education spending is the primary exception. hanson (2005) argued that low-skill immigrants in the us hurt a state’s fiscal balance because they contribute less to tax revenues and their families put a heavy demand on public services, and this would be felt most in states with a stronger social safety net. furthermore, public opposition to immigration in the us appeared to fit the fiscal burden model, in that the 1996 federal welfare reform act, which made it harder for immigrants to access the social safety net, reduced the opposition to immigration from higher-skilled native workers. however, lee and miller (2000) argued that the fiscal consequences of immigration should be too small to matter, relative to other costs and benefits, in setting immigration policy. they did note that immigration has distributional consequences between different levels of government, and between local levels when their immigration levels are different. auerbach and oreopoulos (1999) estimated the fiscal impact of immigrants via a generational accounting method; in the long run immigrants reduce the fiscal burden for natives, they found, but in the short run they may add to it. a generational accounting study of immigrants in austria found an overall positive fiscal balance (mayr, 2005). hansen et al. (2017) examined the impact of immigrants into denmark and found that immigrants from developed economies tended to have a positive fiscal impact, while immigrants from poorer countries had a negative impact that diminished by the second generation. dustmann et al. (2010) reported that immigrants into the uk from eastern europe were much less likely to receive welfare benefits or live in social housing that native residents. still, almost half of europeans surveyed believed that new immigrants benefit more than they are taxed (dustmann and frattini, 2014), and only 8% believe they should have access to those benefits before their first full year of residence. in their study of immigration into the uk, dustmann and frattini carefully accounted for taxes, housing costs, and non-congestible public goods, and estimate that immigrants from the european economic area (eea) countries contribute more to the fiscal balance than native workers, while immigrants from non-eea countries tend to contribute less. 2.2 immigrants and fractionalization immigration is the chief source of rising ethnic diversity in many countries, and there is a significant literature arguing that voters are less likely to support many government services, and the taxes that pay for them, when they perceive those services being reallocated to other ethnic groups. salient ethnic differences may undermine social cohesion and trust, particularly trust of the government itself. of course, it may depend on the type of service, with less support for education, redistribution, and public infrastructure, and more support for policing and public safety. after developing an index of fractionalization to measure ethnic, linguistic, or religious diversity federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 11 (alesina et al., 2003), alesina and ferrara (2005) found significant international evidence that ethnic diversity reduces public support for funding the public sector. in sub-saharan africa, ethnic fractionalization – also called fragmentation in the literature – is associated with reduced public infrastructure, fewer years of schooling, and political instability (easterly and levine, 1997). jofremonseny et al. (2016) studied immigration in spain prior to the great recession and found municipalities with greater concentrations of immigrants were less likely to increase public spending that benefited immigrants. alesina et al. (2001) argued that the greater diversity of the united states relative to other oecd countries helps to explain why there is less public support for a european-style welfare state. the reluctance of voters to support public spending that may benefit other ethnic groups is relatively prominent in the us (rugh and trounstine, 2011), particularly as diversity rises over time, once cross-sectional differences are accounted for (hopkins, 2009). sunding and zwane (2004) found that ethnic fractionalization was negatively correlated with support for local public goods in california. mohanty et al. (2005) found that health care expenditures in the us were substantially lower for immigrants, at least before the affordable care act was passed. diversity may lead to a rebalancing of public expenditures from public infrastructure to other goods such as public safety, which may or may not change total expenditures overall (lee et al., 2016). highfill and o’brien (2015) examined the effect of ethnic diversity on nine categories of municipal spending in a 2002 cross-section of 417 us cities and found that the effects depended on the category of spending, and on whether diversity was measured narrowly and broadly. alvarado and creedy (1998) examined immigration in australia and found that it increased social spending per capita. a study of immigrants in the eu-15 found they took sick leave more often, and used relatively more social welfare benefits (rasmussen, 1997). national-level data has some advantages. trounstine (2013) notes that the correlation between government spending and ethnic diversity may be complicated at the local level by a type of tiebout (1956) sorting, in which people with conservative racial views tend to prefer living in homogenous communities. even after accounting for that, however, voters responded to increased ethnic diversity in their community by reducing their support for public expenditures, particularly when they believe that spending benefits minorities. while the children of immigrants may lead to rising demand for education, cutler et al. (1993) argued that voter preferences have a much larger effect on us public education spending per student. poterba (1997) found that public school funding declined when the ethnic diversity of students exceeded that of the elderly in their states. figlio and fletcher (2012) found similar evidence. in a study of us county-level data from the 2000 census, ajilore (2006) found that ethnic diversity leads to lower education spending per student. coen-pirani (2011) estimated that education spending in california is almost a quarter less than it would have been without the increased immigration of the last several decades. in a study of the effect on immigration on schools in spain, tanaka et al. (2018) found that immigration led to a large increase in public school enrollment accompanied by a shift of native schoolchildren into private schools and a federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 12 reduction in public school funding. in a cross-country study, davide (2008) found that immigrant children led to more upper-income parents placing their children into private schools, declining tax revenues, and more crowding in public schools. mavisakalyan (2011) argued that crosssectional studies were appropriate because expenditure and immigrant shares don’t respond quickly over time; he analyzed a cross section of 82 countries and found that immigration leads to reduced public school funding and thus increases private school enrollment. stichnoth and van der straeten (2013) updated the evidence from cross-country studies, subnational studies, surveys, and experiments, and concluded the evidence was mixed, and the issue more complicated. the political climate, among other things, tended to matter, and a reduced lack of support from voters did not always translate into less spending. schaeffer (2013) went further and performed a meta-analysis of 172 different empirical studies to evaluate the evidence. the evidence, they found, tended to support the hypothesis that ethnic diversity reduced social cohesion and support for spending on public goods more often than not, but it depended on many other factors. 3. model and data the above literature review suggests two competing possibilities. on one hand, immigrants pay taxes and receive some government services. if those services are non-congestible public goods, we should expect to see a marginal expenditure on immigrants that is lower than the average cost of those services for the existing native population. if certain transfer payments are unavailable to immigrants, however, then those marginal costs might be very low. however, some services may be more expensive to provide to immigrants, due to language, cultural barriers, or even preexisting health and educational deficits for immigrants from much poorer countries. this approach assumes that the oecd countries continue to provide immigrants with the similar levels of government services on the margin. on the other hand, the oecd countries tend to be democracies in which voters may have some say on the level of government services provided, and if those voters are less willing to provide public services to a population that becomes ethnically more diverse, then we should see a reduced level of spending on the margin that is independent of the actual cost of providing those services. both of these competing possibilities could be simultaneously true, and so our task is to decide which should be given greater weight. we do this through regressions for government expenditures and revenues that considers the impact of immigrant shares and ethnic fractionalization, and then calculating the implied marginal expenditures and revenues of an additional immigrant. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 13 table 1. oecd data, average 1995-2019, part 1. general government share of gdp frac index 2017 dollar ppp gdp per capita unemp. rate country exp soc hed oth rev (f) (inc) (u) usa 37.9 7.4 14.0 16.5 32.6 0.488 $53,538 5.7 australia 37.6 10.1 12.3 15.2 34.7 0.222 $42,757 6.0 japan 37.3 14.5 10.2 12.6 31.6 0.017 $37,616 4.0 rep. korea 28.0 4.8 7.4 15.7 29.7 0.064 $30,691 3.7 israel 44.6 11.1 12.2 21.3 40.7 0.356 $33,737 8.7 switzerland 32.9 12.9 7.0 13.0 32.8 0.350 $61,709 4.0 norway 46.3 17.5 13.0 15.8 56.2 0.126 $60,216 3.8 iceland 46.3 9.2 15.0 22.1 46.3 0.028 $46,086 3.9 germany 46.6 20.0 10.9 15.7 45.0 0.174 $46,428 7.3 france 54.7 22.5 13.2 19.1 51.1 0.147 $41,461 9.6 italy 48.8 18.6 10.8 19.4 45.6 0.100 $42,337 9.9 belgium 52.1 18.2 13.0 20.9 50.1 0.585 $46,101 7.8 netherlands 44.7 16.0 11.5 17.2 43.0 0.272 $50,000 4.8 luxembourg 41.6 17.0 9.8 14.9 43.5 0.298 $102,100 4.3 uk 40.7 14.9 11.6 14.1 37.0 0.380 $41,356 6.0 ireland 36.2 12.7 10.2 13.4 33.5 0.158 $55,630 8.3 denmark 53.8 23.3 14.3 16.3 54.2 0.141 $50,854 5.7 greece 49.1 16.5 9.6 22.9 42.6 0.138 $31,069 14.5 spain 42.0 14.9 9.9 17.2 38.2 0.649 $36,367 16.7 portugal 45.8 15.4 12.4 17.9 41.0 0.155 $30,800 8.4 austria 51.5 20.8 12.6 18.1 49.1 0.217 $49,791 5.0 sweden 52.6 21.3 13.2 18.2 52.6 0.205 $45,263 7.3 finland 52.7 22.3 12.9 17.5 53.1 0.135 $43,193 9.8 poland 44.2 17.0 9.8 17.3 40.3 0.055 $21,648 11.3 hungary 49.3 15.7 10.5 23.1 44.5 0.193 $23,799 7.6 czechia 42.7 12.5 11.8 18.4 39.7 0.295 $31,160 5.9 slovakia 43.5 14.4 10.3 18.8 38.9 0.244 $22,367 13.2 estonia 37.8 11.5 11.1 15.2 38.0 0.475 $25,320 9.2 latvia 37.8 11.7 9.5 16.6 35.8 0.564 $20,323 12.5 lithuania 37.3 11.7 11.0 14.6 34.6 0.301 $22,753 11.5 slovenia 47.6 17.8 12.7 17.1 44.4 0.242 $30,967 6.8 our regression model is: 𝑙𝑙𝑙𝑙 � 𝑋𝑋𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � = 𝛽𝛽0 + 𝛽𝛽1 � 𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽2𝐹𝐹(𝐼𝐼)𝑖𝑖,𝑡𝑡 + 𝛽𝛽3 � 𝑌𝑌𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽4 � 𝑃𝑃𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + �𝛾𝛾𝑗𝑗𝑍𝑍𝑖𝑖,𝑡𝑡 6 𝑗𝑗=1 + 𝛾𝛾𝑡𝑡𝑡𝑡 + 𝑒𝑒𝑖𝑖,𝑡𝑡 where xi,t is real public revenue or expenditure for country i in year t, n is the total resident population, and the primary variables of interest are: i, the stock of foreign-born residents; f(i), the index of ethnic fractionalization, itself an implicit function of i; yi, the number of immigrants below the age of 18; and pi, the number of immigrants from poor, less-developed countries. both yi and pi are subsets of i. in addition, z is a vector of income and demographic variables that affect government spending per capita. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 14 our dataset is a balanced panel for 31 countries, out of the 38 oecd member states, for 25 years from 1995 to 2019. general government revenue (rev) and expenditures (exp) are reported by the oecd as a share of gdp. there is a multi-year gap in revenue for both iceland and japan, which we interpolated using the unr-wider dataset (united nations university, 2021). total expenditures are divided into categories based on the classifications of the functions of government, or cofog (united nations, 1991). on average, the largest spending shares are for social protection, public services, health care, education, and economic affairs, a function which includes spending on items such as agricultural support programs and government support for energy production, transportation, and communication. spending on other functions includes the functions of national defense, public order and safety, environmental protection, housing and community amenities, and the combination of recreation, culture, and religion. we group these into three broad categories: social expenditures (soc), health and education (hed), and all other (oth) expenditures. the first is primarily transfer payments, the second investments in human capital, and the third is presumably dominated by the provision of public goods. these shares are reported in table 1 by country, averaged over the 1995-2019 sample period. in this table, we also report 2019 gdp per capita (inc), measured in 2017 international dollars that adjust for inflation and purchasing power parity (ppp), and the average unemployment rate (u) over the sample period. these latter two variables come from world bank (2021). finally, table 1 shows the average value for the index of fractionalization, which is essentially the probability that two randomly selected individuals in any population are from different groups (alesina et al., 2003). the historical index of ethnic fractionalization (f) comes from the harvard dataverse (drazanova, 2019). this index is calculated as: 𝐹𝐹𝑖𝑖,𝑡𝑡 = 1 −��𝑠𝑠𝑖𝑖,𝑡𝑡,𝑘𝑘� 2 𝐾𝐾 𝑘𝑘=1 , where si,t,k is the population share of ethnic group k in country i and year t, out of k total ethnic groups identified in that country. essentially, it is a reverse herfindahl–hirschman index of ethnic concentration, scaled over the unit interval [0,1]. the number of foreign-born residents for each country are estimated and reported quintennially by the un (united nations, 2019). we interpolate for the years in between and divide by population to estimate the stock of foreign-born immigrants (i). because the un estimates include both origin and destination countries, as well as age groups, we are able to estimate the number of immigrants each country receives from less-developed countries as well as the number of immigrants younger than 18. these latter two variables are also divided by the country’s total population to get the number of poor (pi) and young (yi) immigrants residing in each country. the control variables are in the vector z = [ln(inc), u, y/n, e/n, r/n, ln(n)], and they all come from world bank (2021). table 2 reports the mean values for most of these variables by country and averaged over the sample period. of course, ln(inc) is the log of real gdp per capita, adjusted for inflation and ppp. the inc variable was already used to convert the government spending shares into total government federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 15 spending, and if government spending shares are invariant to per-capita income, the coefficient for this variable should be equal to one. the unemployment rate (u) should be higher during recessions that increase the demand for government services for native and foreign-born residents alike. demographic variables include: the ratio y/n, the share of total population below the age of 18; e/n, the elderly share of the population over the age of 65; and r/n, the rural share of the population. we expect that a higher share of children should increase the demand for education as well as for social support services, a higher share of elderly should increase spending on pensions and health care; and a more rural population should increase the average costs of many government services. finally, we include ln(n) to help capture any scale effects due to a larger population, particularly for non-congestible public goods. table 2. oecd data, average 1995-2019, part 2. millions percentage of population total pop total immig immig young imm poor imm elderly young rural country (n) (i) (i/n) (yi/n) (pi/n) (e/n) (y/n) (r/n) usa 299.98 40.80 14.0 1.6 1.3 13.0 20.0 20.0 australia 21.29 5.49 26.0 2.4 1.0 13.0 20.0 15.0 japan 127.16 1.98 2.0 0.2 0.1 22.0 14.0 14.0 rep. korea 48.76 0.67 1.0 0.1 0.1 10.0 18.0 19.0 israel 7.24 1.91 27.0 2.5 1.9 10.0 28.0 8.0 switzerland 7.68 1.96 25.0 3.5 1.4 16.0 16.0 26.0 norway 4.79 0.49 10.0 1.5 1.2 16.0 19.0 22.0 iceland 0.31 0.03 9.0 1.9 0.1 12.0 22.0 7.0 germany 81.99 9.69 12.0 1.0 0.4 19.0 14.0 24.0 france 63.66 7.05 11.0 1.0 1.1 17.0 19.0 22.0 italy 58.59 4.26 7.0 0.9 0.4 20.0 14.0 32.0 belgium 10.72 1.48 14.0 1.7 1.5 17.0 17.0 3.0 netherlands 16.41 1.77 11.0 1.2 0.7 15.0 18.0 16.0 luxembourg 0.50 0.18 36.0 4.4 0.1 14.0 18.0 13.0 uk 61.76 6.52 10.0 1.2 1.0 17.0 18.0 19.0 ireland 4.28 0.58 13.0 2.6 0.1 12.0 21.0 39.0 denmark 5.50 0.48 9.0 1.2 0.7 16.0 18.0 14.0 greece 10.88 1.17 11.0 1.4 0.3 19.0 15.0 25.0 spain 44.02 4.20 9.0 1.2 0.2 17.0 15.0 22.0 portugal 10.37 0.75 7.0 1.0 2.7 18.0 16.0 41.0 austria 8.31 1.24 15.0 1.6 0.3 17.0 16.0 41.0 sweden 9.30 1.32 14.0 1.7 1.6 18.0 18.0 15.0 finland 5.31 0.22 4.0 0.7 0.4 17.0 17.0 17.0 poland 38.21 0.73 2.0 0.2 0.0 14.0 17.0 39.0 hungary 10.05 0.40 4.0 0.5 0.1 16.0 16.0 32.0 czechia 10.39 0.34 3.0 0.2 0.1 15.0 16.0 26.0 slovakia 5.40 0.14 3.0 0.3 0.0 13.0 17.0 45.0 estonia 1.35 0.23 17.0 0.6 0.0 17.0 17.0 31.0 latvia 2.19 0.36 16.0 0.5 0.0 17.0 16.0 32.0 lithuania 3.22 0.18 6.0 0.3 0.1 16.0 17.0 33.0 slovenia 2.03 0.23 11.0 0.8 0.2 16.0 15.0 48.0 federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 16 4. regression results we begin by comparing approaches to our regressions. table 3 reports these regression results for total expenditures (ln(exp/n), the log of total real general government expenditures per capita) as the dependent variable. we estimate four regressions: ordinary least squares (ols) of pooled data in column 1, a purely cross-sectional between country (bc) regression in column 2, a country-specific fixed effects (fe) regression in column 3, and a country-specific random effects (re) regression in column 4. some studies argue that cross-section datasets are more appropriate for a study such as this because expenditure and immigrant shares change slowly over time (mavisakalyan, 2011), while other studies argue that cross-sectional differences should be removed to better account for changes over time (hopkins, 2009). it is clear from the first three columns of this table that some of the results may depend on this. the ols estimate of pooled data in column 1 of table 3 provides the basic multivariate correlation embedded in the entire panel dataset containing both cross-sectional and over-time variation. the ols estimates of pooled data are usually criticized because they do not control for unobservable country characteristics that could potentially bias the estimated coefficients. since expenditures, immigration shares, and fractionalization indices change slowly over time, column 2 shows the bc estimates with a purely cross-sectional regression that uses average data for each country over all years, eliminating the within-country variation with just one observation per country. to control for country-specific unobservable characteristics, we then use two different panel data methods. if the unobserved effects are correlated with the error term of the regression, then fe in column 3 is appropriate, but if the unobserved effects are uncorrelated with the error term of the regression, then re in column 4 are appropriate. the fe regression is essentially an ols regression on the pooled data with country-specific dummy variables appended, and as is conventional in the literature these dummy estimates are not reported to save space. the information exploited in the fe estimation exploits only the over-time or within variation of the data, as the cross-sectional variation is absorbed by the set of country specific dummy variables. the estimates reported in column 3 are thus the opposite of those reported in column 2, in that they only show the effects of within-country variation over time. as columns 2 and 3 show, most of the variation in the data is contained in the over-time or within dimension. we consider then whether the statistically significant fe results in column 3 hold when combined with cross-sectional variation in a pooled regression that controls for unobserved cross-sectional heterogeneity. in other words, will the results of column 3 stay similar if the relevant covariates happen to change only slowly? to answer that question, we report the results of a random effects regression in column 4, an approach that exploits the full information in the dataset while at the same time controlling for unobservable country characteristics. this is our preferred specification. in a sense, the re estimates are a weighted average of the between and within regressions in columns 2 and 3, thus nicely summarizing the information contained in the dataset. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 17 table 3. ols, between, fixed effects & random effects regressions. dependent variable is ln(exp), with standard errors (1) (2) bc (3) fe (4) re variables ols between fixed effects random effects i/n -0.227* 0.381 -0.442** -0.244 immigrant shr. (0.128) (0.843) (0.217) (0.180) f -0.216*** -0.257 -0.302*** -0.264*** fractionalization (0.037) (0.202) (0.101) (0.083) yi/n 4.383*** 15.630 -3.795*** -3.286*** young imm. shr. (1.183) (9.386) (0.846) (0.828) pi/n 6.035*** 7.271 0.299 0.970 poor imm. shr. (0.865) (5.008) (1.200) (1.142) ln(inc) 1.190*** 1.387*** 0.825*** 0.902*** income (0.024) (0.168) (0.047) (0.035) u 1.876*** 2.609** 1.072*** 1.164*** unemployment (0.146) (1.153) (0.117) (0.111) y/n 1.230*** 1.757 -1.855*** -1.320*** young pop. (0.303) (1.921) (0.405) (0.354) e/n 2.646*** 2.614 0.536 0.738** elderly pop. (0.277) (1.872) (0.376) (0.306) r/n -0.061 0.014 -0.334* -0.394*** rural pop. (0.055) (0.289) (0.188) (0.143) ln(n) -0.034*** -0.045** 0.002 -0.017 scale (0.004) (0.020) (0.072) (0.015) t -0.006*** 0.001 -0.0004 trend (0.001) (0.002) (0.001) c -3.336*** -5.564*** 0.410 constant (0.272) (1.747) (0.443) observations 775 775 775 775 r2 0.910 0.934 0.864 note: significance *** at 1%, ** at 5%, * at 10%. 4.1 baseline results for total expenditures the ols regression in table 3 suggests there is enough variation in the pooled data, as shown by the statistical significance of many of the covariates. total expenditures do not seem to increase together with immigration; in fact, the coefficient on immigration (i/n) is negative and statistically significant at 10%, but given caveats on the potential endogeneity bias discussed above, we prefer not to place much weight on the point precision of this result. the federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 18 fractionalization index (f) seems to be negatively correlated with total expenditures, while the shares of both young immigrants (yi/n) and poor immigrants (pi/n) seem to move together with total expenditures. these effects do not hold up once we account for the endogeneity bias in columns 3 and 4 by controlling for unobserved heterogeneity. in the control variables, the results appear to make intuitive sense. government expenditures per capita increase with income, unemployment, the share of young people in the population, and the share of elderly people. the coefficient on income is greater than one, consistent with wagner’s law that the government’s share of economic activity rises with income (wagner, 1890), but the size of this coefficient appears to be biased upwards in the ols model. there even appears to be a negative scale effect, so that the larger countries (e.g., the usa) spend less per capita, and the time trend in spending is negative. the results in column 2 illustrate that most of the variation is within countries, not between. the coefficient for income elasticity is even larger, as is the coefficient for the unemployment rate. similarly, the scale effect appears to be driven by between-country differences, not changes over time. all other coefficients in this regression are statistically insignificant. column 3 shows the within-country estimates. in contrast with column 2, more estimated coefficients are statistically significant at conventional levels. in particular, immigration, fractionalization, and the share of young immigrants all display negative correlation with total government expenditures per capita. the coefficient on income remains significant and positive, but appears to be significantly less than one, and the correlation between unemployment and expenditures is also significant but smaller once the between-country variation is removed. a greater share of young people in the population has a negative coefficient, as does the effect of rural population share, but the scale and trend effects disappear. column 4 displays the random effects estimates which nicely summarize the information in the data uncontaminated by unobserved heterogeneity. the share of immigrants in the population does not correlate positively with total government expenditures per capita and is therefore unlikely to be a source of expansion of the welfare state as feared. fractionalization, however, has a negative and statistically significant coefficient, as does the share of young immigrants, the share of young people in the population, and the share of rural residents. the coefficient for income elasticity remains statistically significant and positive, but less than one, suggesting that government spending in this sample does not keep pace with economic growth. the coefficient for the unemployment rate remains positive and mostly unchanged from the fixed effects model. there does not seem to be a population scale effect, and nor do poor immigrants appear to pose a significant burden on public finances. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 19 table 4. alternative random effect regressions. dependent variable is ln(exp), with standard errors variables (1) (2) (3) (4) i/n 0.162 immigrant shr. (0.143) f -0.286*** -0.275*** fractionalization (0.084) (0.082) yi/n -2.792*** -2.675*** young imm. shr. (0.703) (0.699) pi/n 0.273 0.572 poor imm. shr. (1.101) (1.097) ln(inc) 0.974*** 0.928*** 0.948*** 0.917*** income (0.032) (0.033) (0.032) (0.033) u 1.187*** 1.163*** 1.178*** 1.165*** unemployment (0.112) (0.111) (0.112) (0.111) y/n -0.789** -1.139*** -0.948*** -1.230*** young pop. (0.342) (0.351) (0.340) (0.348) e/n 0.950*** 0.686** 0.998*** 0.889*** elderly pop. (0.309) (0.286) (0.286) (0.286) r/n -0.222 -0.441*** -0.188 -0.368*** rural pop. (0.136) (0.145) (0.131) (0.141) ln(n) -0.015 -0.018 -0.013 -0.015 scale (0.016) (0.016) (0.015) (0.015) t -0.002** -0.001 -0.002** -0.001 trend (0.001) (0.001) (0.001) (0.001) c -0.564 0.154 -0.302 0.194 constant (0.409) (0.423) (0.392) (0.416) note: significance *** at 1%, ** at 5%, * at 10%. 4.2 the sensitivity of covariates immigration is correlated with fractionalization. more precisely, the four variables of interest (i/n, f, yi/n, and pi/n) are positively correlated with each other, with correlation coefficients of 0.39 between i/n and f, 0.33 between i/n and yi/n, and 0.37 between i/n and pi/n. given this correlation, is it possible that the inclusion of all four variables is biasing the coefficient for immigration towards zero? table 4 shows the random effects model for expenditures per capita for four alternative specifications of these four variables. in the first, only i/n is included. in the second, only f is included. in the third, only yi/n and pi/n are included. in the last, only i/n is excluded. the results are consistent with each other, and with the results reported in column 4 of table 3. the immigrant share of population does not have a statistically significant effect on expenditures per capita in federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 20 these regressions, while both fractionalization and the share of young immigrants both have a negative effect. other coefficients remain largely unchanged, except that the coefficient on the time trend becomes statistically significant (and negative) when f is excluded, and the rural share becomes insignificant. 4.3 results for government revenue in the first column of table 5, we show the results of a random-effects regression for general government revenue. once again, immigration is uncorrelated with the fiscal variable, while fractionalization shows a negative correlation. the quantitative size of the fractionalization coefficient is roughly half the size estimated for total expenditures.: -0.149 in column 1 of table 5 versus -0.264 in column 4 of table 3). young immigrants are associated significantly with a reduction of revenue. other effects are mostly similar. the income elasticity is estimated to be positive but less than one, while the unemployment rate seems to have a positive effect, ceteris paribus. this seems counterintuitive, but it is only a quarter of the size of the coefficient for expenditures and may be the consequence of the combined contributions of employees and employers to unemployment insurance programs. a larger share of young people has a marginally negative coefficient, while the population share of elderly has a larger and more significant positive coefficient and a more rural population has a negative one, possibly the result of taxes on savings accumulated in retirement and pension funds. finally, there appears to be both a negative scale effect and a negative time trend. 4.4 results for different types of expenditures columns 2, 3 and 4 of table 5 report the random-effect regression results of covarying immigration and fractionalization on three different types of expenditures. using the spending functions reported by the oecd, we divide these into social protection expenditures (soc), health and education expenditures (hed), and all other expenditures (oth). among the covariates of social spending in column 2, we can see that immigration is not statistically significant whereas fractionalization, the share of young immigrants, the share of young population, and the share of poor-country immigrants are all negatively and significantly correlated. the levels of real per-capita income and the share of elderly both co-move positively with social spending, as expected. for health and education expenditures, which are an approximation of human capital investments, the covariates are shown in column 3. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 21 table 5. random effect regressions for different fiscal variables. dependent variables: (1) total revenue; (2) social expenditures; (3) health + education; (4) other variables (1) rev (2) soc (3) hed (4) oth i/n 0.152 -0.116 -0.125 -0.628** immigrant shr. (0.139) (0.249) (0.187) (0.262) f -0.149** -0.356*** 0.002 -0.255** fractionalization (0.066) (0.118) (0.086) (0.118) yi/n 0.507 -3.717*** -0.822 -4.477*** young imm. shr. (0.611) (1.080) (0.865) (1.264) pi/n -0.104 -5.025*** 4.384*** 2.905* poor imm. shr. (0.852) (1.510) (1.191) (1.714) ln(inc) 0.920*** 0.779*** 0.940*** 0.941*** income (0.028) (0.051) (0.036) (0.049) u 0.251*** 1.898*** 0.171 1.204*** unemployment (0.083) (0.146) (0.116) (0.169) y/n -0.450* -2.975*** -2.536*** -0.782 young pop. (0.272) (0.486) (0.368) (0.519) e/n 1.707*** 1.753*** -0.406 0.287 elderly pop. (0.232) (0.414) (0.319) (0.453) r/n -0.272** -0.252 -0.510*** -0.272 rural pop. (0.116) (0.210) (0.148) (0.198) ln(n) -0.059*** 0.009 -0.019 -0.026 scale (0.015) (0.029) (0.016) (0.019) t -0.002** 0.005*** 0.002 -0.006*** trend (0.001) (0.002) (0.001) (0.002) c 0.006 0.554 -0.941** -0.885 constant (0.357) (0.648) (0.460) (0.621) note: significance *** at 1%, ** at 5%, * at 10%. once again, the immigrant share of population is not significantly correlated, but the coefficient for both ethnic fractionalization and the share of young immigrants are also not statistically significant. however, a larger share of immigrants from poorer countries is positively related to spending on health and education, and the difference with social protection spending is notable. both a younger population and a more rural population are inversely associated with health and education expenditures, while the income effect remains significant but less than one. finally, column 4 of table 5 has a negative and significant coefficient for the share of immigrants on all other expenditures, as does ethnic fractionalization and the share of young immigrants. the coefficient for the share of poor-country immigrants is marginally positive. other expenditures are also significantly associated with higher per-capita income and a higher unemployment rate, but not with the demographic variables or the country population. there does appear to be a negative time trend. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 22 5. interpreting the results how can we interpret these results? our first task is to determine whether our estimated coefficients are robust to specification. we then use our estimated coefficients to calculate the marginal effects of an additional immigrant on government revenues and expenditures, in order to determine whether or not immigrants are a fiscal burden on oecd countries. 5.1 robustness the results presented in tables 3, 4, and 5 do not take into account either the strong inertial behavior of total government expenditures (or revenue) from year to year or the possibility that some or all of the covariates may themselves be endogenously explained by the dependent variable. the strong inertia in government spending is well known and is present in our dataset, as about 72% of current year expenditures can be explained by the previous year’s expenditure. excluding the lagged dependent variable may lead to apportioning explanatory power to variables that may not truly have it. the dynamic panel estimation procedure of arellano and bond (1991) is an econometric specification that helps control for both inertia in the dependent variable and also controls for the endogeneity of the set of right-hand side covariates. the second lags of the right-hand-side covariates act as instruments, and the first lag of the dependent variable is included as a righthand-side regressor. table 6 presents the results from the arellano-bond regressions. instead of just the level of real government expenditures per capita, the dependent variable can be thought of as a quasi-first difference of these expenditures, i.e., 𝑙𝑙𝑙𝑙 �𝐸𝐸𝐸𝐸𝐸𝐸𝑡𝑡 𝑁𝑁𝑡𝑡 � − 0.72 𝑙𝑙𝑙𝑙 �𝐸𝐸𝐸𝐸𝐸𝐸𝑡𝑡−1 𝑁𝑁𝑡𝑡−1 �, across all four columns. column 1 of table 6 includes all four of our variables of interest. column 2 excludes all but immigration, column 3 excludes all but fractionalization, and column four includes only young and poor immigrant ratios. these results stay qualitatively similar to the ones presented so far. the coefficient for immigration is negative but insignificant, regardless of whether the other three variables are excluded, while fractionalization is negative and statistically significant either way. both the share of young immigrants and the share of poor-country immigrants have statistically insignificant coefficients, whether or not immigration rates and fractionalization indices are included. for the remaining control variables, there are some differences. the income elasticity remains positive and significant but is much smaller, while the effect of the unemployment rate reverses sign. the share of young people has a negative coefficient that is consistently significant, while the share of elderly is either insignificant or marginally significant at the 10 percent level. rural shares are insignificant in the arellano-bond estimates, as are scale effects and time trends. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 23 as a second way to deal with inertia in the lag dependent variable (and the lack of independence across different observations over time), we tried to run regressions in which the time dimension is collapsed, averaging observations every five-year period for each country, i.e., 1995, 2000, 2005, 2010, and 2015 only. this was especially relevant given how the annual number of foreign-born residents was interpolated from quinquennial statistics. unfortunately, since most of the variation in our dataset is within countries in the time dimension (recall the insignificance of between regressions in table 3), this procedure simply drops too much information by the side, preventing us from estimating the covariates on the right-hand side with any degree of precision at conventional statistical levels. we make those regressions available upon request but do not include them here, given space constraints. table 6. arellano & bond dynamic panel regressions. dependent variable: total expenditures, with standard errors variables (1) (2) (3) (4) ln(exp)-1 0.711*** 0.721*** 0.715*** 0.722*** lag dep. var. (0.031) (0.030) (0.030) (0.030) i/n -0.195 -0.106 immigrant shr. (0.261) (0.183) f -0.320*** -0.275** fractionalization (0.117) (0.111) yi/n 1.033 1.285 young imm. shr. (1.168) (0.953) pi/n 0.738 1.146 poor imm. shr. (1.193) (1.132) ln(inc) 0.138*** 0.149*** 0.138*** 0.169*** income (0.050) (0.047) (0.046) (0.048) u -0.356*** -0.364*** -0.372*** -0.334*** unemployment (0.120) (0.117) (0.116) (0.120) y/n -1.716*** -1.466*** -1.635*** -1.398*** young pop. (0.370) (0.339) (0.343) (0.330) e/n -0.596 -0.752** -0.575 -0.694* elderly pop. (0.387) (0.379) (0.379) (0.376) r/n -0.365 -0.163 -0.315 -0.162 rural pop. (0.235) (0.218) (0.229) (0.217) ln(n) 0.011 -0.067 -0.040 -0.080 scale (0.088) (0.082) (0.078) (0.077) t 0.002 0.002 0.002 0.001 trend (0.002) (0.002) (0.001) (0.002) c 1.943*** 1.747*** 1.955*** 1.534** constant (0.638) (0.597) (0.600) (0.611) note: significance *** at 1%, ** at 5%, * at 10%. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 24 another concern with our previous results is that there could be non-independence among some observations. this could be due, for example, to autocorrelation. in fact, we know already that this is the case given the high degree of inertia present in the dependent variable --and the high correlation of the dependent variable with many of the right-hand side covariates. furthermore, the residuals of our previous regressions may also be affected by heteroskedasticity across the different cross-sections (countries). therefore, correction of the variance-covariance matrix of residuals may be warranted to account for the potential for non-independence of observations across time and space and the consequent effect on residuals and standard errors in the variance-covariance residuals matrix. in table 7, we show the results of our regressions using huber-white sandwich robust standard errors (huber, 1967; white, 1980), a method equivalent to clustering by state. we run these regressions for total expenditures and revenues, as well as for our three general categories of expenditures. our most important previous conclusions remain qualitatively similar, but in several instances the level of statistical significance falls below the 10% level. table 7. random effect regressions with robust standard errors. dependent variables variables (1) exp (2) rev (3) soc (4) hed (5) oth i/n -0.244 0.152 -0.116 -0.125 -0.628 immigrant shr. -0.338 -0.339 -0.494 -0.423 -0.541 f -0.264 -0.149 -0.356 0.002 -0.255 fractionalization -0.174 -0.153 -0.368 -0.273 -0.22 yi/n -3.286*** 0.507 -3.717** -0.822 -4.477 young imm. shr. -1.07 -1.162 -1.894 -1.305 -2.806 pi/n 0.97 -0.104 -5.025* 4.384 2.905 poor imm. shr. -3.158 -2.612 -3.055 -2.704 -4.015 ln(inc) 0.902*** 0.920*** 0.779*** 0.940*** 0.941*** income -0.094 -0.104 -0.139 -0.0875 -0.121 u 1.164*** 0.251 1.898*** 0.171 1.204*** unemployment -0.188 -0.275 -0.259 -0.211 -0.331 y/n -1.32 -0.45 -2.975 -2.536*** -0.782 young pop. -1.143 -1.164 -2.254 -0.929 -1.392 e/n 0.738 1.707** 1.753 -0.406 0.287 elderly pop. -0.773 -0.723 -1.25 -0.846 -0.972 r/n -0.394** -0.272 -0.252 -0.51 -0.272 rural pop. -0.17 -0.23 -0.457 -0.344 -0.274 ln(n) -0.017 -0.059** 0.009 -0.019 -0.026 scale -0.017 -0.028 -0.043 -0.027 -0.023 t -0.001 -0.002 0.005 0.002 -0.006* trend -0.003 -0.003 -0.006 -0.004 -0.003 c 0.41 0.00559 0.554 -0.941 -0.885 constant -1.233 -1.404 -1.815 -1.102 -1.465 note: significance *** at 1%, ** at 5%, * at 10%. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 25 important qualifications need to be made. in general, across all regressions in table 7, the sandwich procedure increases standard errors to account for a reduction in the effective number of independent observations and the consequent increase in noise in the estimates implied by the loss of degrees of freedom. in general, the variable most affected by the loss of precision in the estimates is ethnic fractionalization, which loses its statistical significance at conventional levels in all regressions compared to the case presented before when adjustment was not made for the non-independence of observations. additionally, other variables also lose their statistical significance at conventional levels, including young and old population shares in column 1; fractionalization, unemployment, young and elderly population shares in column 2; young and elderly population shares in column 3; poor immigrant and rural shares in column 4; and poor immigrant and young population share in column 5. nonetheless, our conclusions about the effect of immigration on the fiscal burden of immigration are not substantially altered. given that most of the variation we rely upon to estimate the covariates in this dataset is within countries over time, we expect we could recover the statistical significance of our coefficient estimates with a longer dataset. such a dataset would help overcome the problem of the clustering of certain observations for each country. when properly accounted for, those clusters of correlated observations reduce the precision of the estimates by effectively reducing the degrees of freedom available. 5.2 the fiscal burden of immigration how do our regression results translate into an estimate of the fiscal burden of immigration? the essential problem is that we cannot separate out what countries should spend to address the costs of public services from what they choose to spend, and that choice in a democratic society is affected by that society’s view of immigration and ethnic difference. we can, however, estimate the observed marginal revenue and marginal expenditure of an immigrant, and compare those to the average revenue and average expenditure of the existing population. we assume that a country’s ethnic fractionalization is an implicit function of the share of foreign-born residents. in a separate random-effects regression, we estimate a marginal coefficient of 𝐹𝐹′=0.747. then using the reported point estimates from our random-effects regressions, we calculate the first derivative for both revenue and total expenditures. the marginal effect of an immigrant (mei) is: 𝑀𝑀𝑀𝑀𝐼𝐼𝑖𝑖,𝑡𝑡 = 𝑑𝑑𝑋𝑋𝑖𝑖,𝑡𝑡 𝑑𝑑𝐼𝐼𝑖𝑖,𝑡𝑡 = 𝑋𝑋𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 �1 + 𝛽𝛽𝐼𝐼 �1 − 𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽𝐹𝐹𝐹𝐹′ �1 − 𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 �� where 𝛽𝛽𝐼𝐼 and 𝛽𝛽𝐹𝐹 are the random effects point estimates for i/n and f. we calculate these marginal effects for each observation in our sample, and then compare them to the observed revenue and expenditure per capita. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 26 for the entire sample, the marginal government expenditure per immigrant averages around $4,600 per year, compared to an average expenditure per capita of $7,100. the average ratio of marginal to averages is 61%. if we instead use the coefficients from our arellano-bond estimates in table 6, marginal expenditures per immigrant average out to 62% of the average expenditure. the marginal government revenue per immigrant averages around $7,400 per year, or about 104% of the average revenue per capita. because the marginal revenue exceeds the marginal cost, our estimates suggest that immigrants are not likely to be a fiscal burden, at least not in general. by relying on the fact that ∂n/∂i = ∂i/∂yi = ∂i/∂pi = 1, as i is a subset of n and both yi and pi are subsets of i, we can also derive the marginal effect per young immigrant (meyi) as: 𝑀𝑀𝑀𝑀𝑌𝑌𝐼𝐼𝑖𝑖 ,𝑡𝑡 = 𝑑𝑑𝑋𝑋𝑖𝑖,𝑡𝑡 𝑑𝑑𝑌𝑌𝐼𝐼𝑖𝑖,𝑡𝑡 = 𝑋𝑋𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 �1 + 𝛽𝛽𝐼𝐼 �1 − 𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽𝐹𝐹𝐹𝐹′ �1 − 𝐼𝐼𝑖𝑖 ,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽𝑌𝑌𝐼𝐼 �1 − 𝑌𝑌𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 �� for young immigrants, the estimated marginal expenditure is estimated to be -263% of total expenditures per capita, a result that is perhaps consistent with research findings that voters reduce educational spending when more immigrants are in schools (e.g., tanaka et al., 2018; davide, 2008; mavisakalyan, 2011). in. using the arellano-bond estimates, the marginal expenditure comes out higher, but the coefficient for 𝛽𝛽𝑌𝑌𝐼𝐼 was statistically insignificant. in contrast, the estimated marginal revenue per young immigrant averages around 154% of average government revenue per capita. we can also derive the marginal effect per poor-country immigrant (mepi) as: 𝑀𝑀𝑀𝑀𝑃𝑃𝐼𝐼𝑖𝑖,𝑡𝑡 = 𝑑𝑑𝑋𝑋𝑖𝑖,𝑡𝑡 𝑑𝑑𝑃𝑃𝐼𝐼𝑖𝑖,𝑡𝑡 = 𝑋𝑋𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 �1 + 𝛽𝛽𝐼𝐼 �1 − 𝐼𝐼𝑖𝑖 ,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽𝐹𝐹𝐹𝐹′ �1 − 𝐼𝐼𝑖𝑖 ,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 � + 𝛽𝛽𝐸𝐸𝐼𝐼 �1 − 𝑃𝑃𝐼𝐼𝑖𝑖,𝑡𝑡 𝑁𝑁𝑖𝑖,𝑡𝑡 �� the estimated marginal expenditure per poor-country immigrant averages 157% of total expenditures per capita, or 135% with the arellano-bond estimates, while the estimated marginal revenue per poor-country immigrant averages 93% of total expenditures per capita. this suggests that immigrants from poorer emerging economies – those officially labeled as less developed economies by the world bank – may be the exception, that fiscal revenues government receive from them or because of them are outweighed by the fiscal cost of the government services they receive. while immigrants in general are not likely to be a fiscal burden for the oecd countries, those from poor countries appear to be. 6. conclusion in this study, we examined government revenue and expenditures for 31 oecd countries over 25 years, in an effort to determine whether immigration created a significant fiscal burden, and whether growing ethnic diversity – as measured by the index of ethnic fractionalization – led to federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 27 less support for public goods. these two effects are potentially contradictory; the immigration rate is correlated with fractionalization, but it is also correlated with the share of school-age and poorer immigrants. to better capture this, we calculate the marginal effects of immigrants in general, as well as the marginal effects of both young and poor immigrants. results over time from a fixed effects model are different from the cross-section regression of country means. just because countries with a greater share of immigrants may have more government spending per capita does not mean that more immigrants will lead to more spending. we try to capture both of these effects with a random effects model. we find that immigration alone is not a statistically significant explanatory variable for spending, but may be significant in the presence of other immigration-related variables. ethnic fractionalization, in particular, does appear to have a consistently negative effect on spending. in addition to comparing revenue and total expenditures, we look at three general categories of government spending. a rising share of immigrants appears to only reduce government expenditures on once social protection, health, and education spending is removed. fractionalization, on the other hand, appears to reduce both all expenditures except those on health and education. immigrants from poorer countries appears to be associated with less spending on social protection but more spending in other areas, while young immigrants are associated with less spending in all areas. consistent with the findings of dustmann and frattini (2014), among others, our estimates suggest that immigrants do not impose a fiscal burden for oecd countries on average. instead, they appear to contribute more to the government’s fiscal budget than native residents, and they absorb fewer expenditures. we find, however, that immigrants from less-developed economies are the exception. their marginal government revenues tend to be smaller than average, and their marginal government expenditures are larger. federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 28 references ajilore, o. (2006). econometric issues in education finance. review of regional studies, 36(2):192–204. alesina, a. & ferrara, e. l. (2005). ethnic diversity and economic performance. journal of economic literature, 43(3):762–800. https://doi.org/10.1257/002205105774431243 alesina, a., devleeschauwer, a., easterly, w., kurlat, s., & wacziarg, r. (2003). fractionalization. journal of economic growth, 8(2):155–194. https://doi.org/10.3386/w9411 alesina, a., glaeser, e., & sacerdote, b. (2001). why doesn’t the us have a european-style welfare system? nber working paper 8524, national bureau of economic research. https://doi.org/10.3386/w8524 alvarado, j. & creedy, j. (1998). population ageing, migration and social expenditure. edward elgar publishing. arellano, m., & bond, s. (1991). some tests of specification for panel data: monte carlo evidence and an application to employment equations. the review of economic studies, 58(2), 277297. https://doi.org/10.2307/2297968 auerbach, a. j. & oreopoulos, p. (1999). analyzing the fiscal impact of us immigration. american economic review, 89(2):176–180. https://doi.org/10.1257/aer.89.2.176 card, d., dustmann, c., & preston, i. (2009). immigration, wages, and compositional amenities. nber working paper 15521, national bureau of economic research. https://doi.org/10.3386/w15521 coen-pirani, d. (2011). immigration and spending on public education: california, 1970–2000. journal of public economics, 95 (11-12):1386–1396. https://doi.org/10.1016/j.jpubeco.2011 .05.006 cutler, d. m., elmendorf, d. w., & zeckhauser, r. j. (1993). demographic characteristics and the public bundle. nber working paper 4283, national bureau of economic research. https://doi.org/10.3386/w4283 davide, d. (2008). low-skilled immigration and education policy with endogenous fertility. discussion paper 2008022, universite catholique de louvain, departement des sciences economiques. dempster, h., leach, a., & hargrave, k. (2020). public attitudes towards immigration and immigrants: what people think, why, and how to influence them. odi working paper 588, overseas development institute. docquier, f., ozden, c., & peri, g. (2013). the labour market effects of immigration and emigration in oecd countries. economic journal, 124 (579):1106–1145. https://doi.org/10. 1111/ecoj.12077 drazanova, l. (2019). historical index of ethnic fractionalization dataset. https://doi.org 10.7910/dvn/4jqrcl dustmann, c. & frattini, t. (2014). the fiscal effects of immigration to the uk. the economic journal, 124 (580): f593–f643. https://doi.org/10.1111/ecoj.12181 dustmann, c., frattini, t., & halls, c. (2010). assessing the fiscal costs and benefits of a8 migration to the uk. fiscal studies, 31(1):1–41. https://doi.org/10.1111/j.1475-5890 .2010.00106.x easterly, w. & levine, r. (1997). africa’s growth tragedy: policies and ethnic divisions. the quarterly journal of economics, 112(4):1203–1250. https://doi.org/10.1162/003355300555 466 edwards, r. & ortega, f. (2017). the economic contribution of unauthorized workers: an industry analysis. regional science and urban economics, 67:119–134. https://doi.org/10.1016/ j.regsciurbeco.2017.09.004 figlio, d. n. & fletcher, d. (2012). suburbanization, demographic change and the consequences for school finance. journal of public economics, 96(11-12):1144–1153. https://doi.org/ 10.1016/j.jpubeco.2011.07.007 hansen, m. f., schultz-nielsen, m. l., & tranæs, t. (2017). the fiscal impact of immigration to welfare states of the scandinavian type. journal of population economics, 30(3):925–952. https://doi.org/10.1007/s00148-017-0636-1 hanson, g. h. (2005). why does immigration divide america? peterson institute. highfill, j. & o’brien, k. (2015). the effect of ethnic diversity on municipal spending. atlantic economic journal, 43(3):305–318. https://doi.org/10.1007/s11293-015-9469-z https://doi.org/10.1257/002205105774431243 https://doi.org/10.3386/w9411 https://doi.org/10.3386/w8524 https://doi.org/10.2307/2297968 https://doi.org/10.1257/aer.89.2.176 https://doi.org/10.3386/w15521 https://doi.org/10.1016/j.jpubeco.2011.05.006 https://doi.org/10.1016/j.jpubeco.2011.05.006 https://doi.org/10.3386/w4283 https://doi.org/10.1111/ecoj.12077 https://doi.org/10.1111/ecoj.12077 https://doi.org/10.1111/ecoj.12181 https://doi.org/10.1111/j.1475-5890.2010.00106.x https://doi.org/10.1111/j.1475-5890.2010.00106.x https://doi.org/10.1162/003355300555466 https://doi.org/10.1162/003355300555466 https://doi.org/10.1016/j.regsciurbeco.2017.09.004 https://doi.org/10.1016/j.regsciurbeco.2017.09.004 https://doi.org/10.1016/j.jpubeco.2011.07.007 https://doi.org/10.1016/j.jpubeco.2011.07.007 https://doi.org/10.1007/s00148-017-0636-1 https://doi.org/10.1007/s11293-015-9469-z federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 29 hopkins, d. (2009). the diversity discount: when increasing ethnic and racial diversity prevents tax increases. the journal of politics, 71(1):160–177. https://doi.org/10.1017/s002238160809 0105 huber, p. j. (1967). the behavior of maximum likelihood estimates under nonstandard conditions. in proceedings of the fifth berkeley symposium on mathematical statistics and probability: weather modification; university of california press: berkeley, ca, usa (vol 1, p. 221-23). jaime-castillo, a. m., marqués-perales, i., & álvarez-gálvez, j. (2016). the impact of social expenditure on attitudes towards immigration in europe. social indicators research, 126(3):1089–1108. https://doi.org/10.1007/s11205-015-0939-0 jofre-monseny, j., sorribas-navarro, p., & vázquez-grenno, j. (2016). immigration and local spending in social services: evidence from a massive immigration wave. international tax and public finance, 23:1004–1029. https://doi.org/10.1007/s10797-016-9399-y lee, r. & miller, t. (2000). immigration, social security, and broader fiscal impacts. american economic review, 90(2):350–354. https://doi.org/10.1257/aer.90.2.350 lee, s., lee, d., & borcherding, t. (2016). ethnic diversity and public goods provision: evidence from u.s. municipalities and school districts. urban affairs review, 52(5):685–713. https://doi.org/10.1177/1078087415587055 mavisakalyan, a. (2011). immigration, public education spending, and private schooling. southern economic journal, 78(2):397–423. https://doi.org/10.4284/0038-4038-78.2.397 mayr, k. (2005). the fiscal impact of immigrants in austria–a generational accounting analysis. empirica, 32(2):181–216. https://doi.org/10.1007/s10663-005-1758-3 mohanty, s. a., woolhandler, s., himmelstein, d. u., pati, s., carrasquillo, o., & bor, d. h. (2005). health care expenditures of immigrants in the united states: a nationally representative analysis. american journal of public health, 95 (8):1431–38. https://doi.org/ 10.2105/ajph.2004.044602 national academies of sciences, engineering, and medicine (2017). the economic and fiscal consequences of immigration. the national academies press, washington, dc. https://doi.org/10.1111/padr.12048 noy, s. & voorend, k. (2016). social rights and migrant realities: migration policy reform and migrants’ access to health care in costa rica, argentina, and chile. international migration and integration, 17:605– 629. https://doi.org/10.1007/s12134-015-0416-2 orrenius, p. (2017). new findings on the fiscal impact of immigration in the united states. working paper 1704, federal reserve bank of dallas, research department. https://doi.org/10.24149/wp1704 ortega, f. & peri, g. (2009). the causes and effects of international migrations: evidence from oecd countries 1980-2005. nber working paper 14833, national bureau of economic research. https://doi.org/10.3386/w14833 parker, e. (2020). migration patterns in eastern europe and the world: a gravity approach. south east european journal of economics and business, 15(1):66–79. https://doi.org/10.2478/jeb2020-0006 poterba, j. (1997). demographic structure and the political economy of public education. journal of policy analysis and management, 16(1):48–66. https://doi.org/10.1002/(sici)15206688(199724)16:1<48::aid-pam3>3.0.co;2-i rasmussen, l. (1997). international migration and health in the european union. in ugalde, a. & cardenas, g., editors, health and social services among international labor migrants, volume ch. 1, pages 1–8. university of texas press, austin, tx. https://doi.org/10.2307/2547335 rugh, j. & trounstine, j. (2011). the provision of local public goods in diverse communities: analyzing municipal bond elections. the journal of politics, 73 (4:1038–1050. https://doi.org/10.1017/s0022381611000776 schaeffer, m. (2013). ethnic diversity, public goods provision and social cohesion: lessons from an inconclusive literature. discussion paper sp vi 2013-103, wissenschaftszentrum berlin für sozialforschung (wzb). smith, j. & edmonston, b. (1997). the new americans: economic, demographic, and fiscal effects of immigration. national academy press, washington, d.c. stichnoth, h. & van der straeten, k. (2013). ethnic diversity, public spending, and individual support for the welfare state: a review of the empirical literature. journal of economic surveys, 27(2):364–389. https://doi.org/10.1111/j.1467-6419.2011.00711.x sunding, d. & zwane, a. (2004). local public goods and ethnic diversity: evidence from the immigration reform and control act. working paper, university of california at berkeley, department of agriculture and resource economics. https://doi.org/10.1017/s0022381608090105 https://doi.org/10.1017/s0022381608090105 https://doi.org/10.1007/s11205-015-0939-0 https://doi.org/10.1007/s10797-016-9399-y https://doi.org/10.1257/aer.90.2.350 https://doi.org/10.1177/1078087415587055 https://doi.org/10.4284/0038-4038-78.2.397 https://doi.org/10.1007/s10663-005-1758-3 https://doi.org/10.2105/ajph.2004.044602 https://doi.org/10.2105/ajph.2004.044602 https://doi.org/10.1111/padr.12048 https://doi.org/10.1007/s12134-015-0416-2 https://doi.org/10.24149/wp1704 https://doi.org/10.3386/w14833 https://doi.org/10.2478/jeb-2020-0006 https://doi.org/10.2478/jeb-2020-0006 https://doi.org/10.1002/(sici)1520-6688(199724)16:1%3c48::aid-pam3%3e3.0.co;2-i https://doi.org/10.1002/(sici)1520-6688(199724)16:1%3c48::aid-pam3%3e3.0.co;2-i https://doi.org/10.2307/2547335 https://doi.org/10.1017/s0022381611000776 https://doi.org/10.1111/j.1467-6419.2011.00711.x federico guerrero and elliott parker / european journal of government and economics 11(1), june 2022, 7-30 30 tanaka, r., farre, l., and ortega, f. (2018). immigration, assimilation, and the future of public education. european journal of political economy, 52:141–165. https://doi.org/10.1016/ j.ejpoleco.2017.06.001 tiebout, c. m. (1956). a pure theory of local expenditures. journal of political economy, 64(5):416–424. https://doi.org/10.1086/257839 trounstine, j. (2013). one for you, two for me: support for public goods investment in diverse communities. working paper, university of california at merced. united nations (1991). provisional central product classification. statistical paper series m 77, united nations publishing division. united nations (2019). international migration. https://www.un.org/en/development/desa/ population/migration/data/. united nations university (2021). government revenue dataset. https://www.wider.unu.edu/ project/governmentrevenue-dataset. https://doi.org/10.35188/unu-wider/grd-2021 wagner, a. (1890). finanzwissenchaft. leipzig: c.f. winter. world bank (2021). world databank: world development indicators. https://databank.worldbank.org. white, h. (1980). a heteroskedasticity-consistent covariance matrix estimator and a direct test for heteroskedasticity. econometrica, 48(4): 817-838. https://doi.org/10.2307/1912934 https://doi.org/10.1016/j.ejpoleco.2017.06.001 https://doi.org/10.1016/j.ejpoleco.2017.06.001 https://doi.org/10.1086/257839 https://www.un.org/en/development/desa/population/migration/data/ https://www.un.org/en/development/desa/population/migration/data/ https://www.wider.unu.edu/project/governmentrevenue-dataset https://www.wider.unu.edu/project/governmentrevenue-dataset https://doi.org/10.35188/unu-wider/grd-2021 https://databank.worldbank.org/ https://doi.org/10.2307/1912934 1. introduction 2. the economic and fiscal impact of immigration 3. model and data 4. regression results 5. interpreting the results 6. conclusion references european journal of government and economics 11(1), june 2022, 73-96 european journal of government and economics issn: 2254-7088 social spending as a development tool: evidence from developing countries deepti ahuja a *, deepak pandit b a indian institute of management, rohtak, india b bml munjal university, india * corresponding author at: dahuja2582@gmail.com abstract. in this paper, we aim to study the interrelationship between social spending, economic growth, and income inequality in developing countries from the year 1990 to 2013. we observed that all the categories of social spending produced a significant reduction in income inequality. further, the impact of health and education spending on economic growth is significant, and that of social protection is insignificant. this indicates that both health and education spending can break the trade-off between equity and efficiency, that is, it can lead to both growth and progressive distributional change. however, given the importance of social welfare measures in reducing income inequality, developing countries need to focus on active social spending like labor market reforms that can increase gross domestic product growth rate and simultaneously reduce income inequality. keywords. social spending; income inequality; gdp growth; simultaneous equation model; asia-pacific countries jel codes. c33, d30, h51, h52, o40 doi. https://doi.org/10.17979/ejge.2022.11.1.7385 1. introduction the increase in income inequality in recent years has drawn considerable attention from academic and policy experts on the relationship between economic growth and income inequality in both developed and developing countries. economic growth and income inequality, being endogenous outcomes of the economic system, are subject to common influences, with respect to both structural changes and macroeconomic policies. structural changes such as improvement in technology facilitate economic development, which is the underlying assumption of kuznet’s curve, and may result in economic inequality. however, macroeconomic policies, particularly fiscal policies are considered an important instrument for achieving goals in terms of equity and efficiency (musgrave, 1959). this paper focuses on the expenditure side of the budget and emphasizes the role of social spending in explaining the dynamics between growth and inequality1. government spending in the social sector received an impetus with the emergence of endogenous growth theory (lucas, 1988; romer, 1994), which largely focused on enhancing human capital development. such policies facilitate the process of innovation, research, knowledge creation, and information 1 in this study, social spending refers to government expenditure on education, health, and social protection. © 2022. this work is licensed under a cc by-nc 4.0 license. mailto:dahuja2582@gmail.com https://doi.org/10.17979/ejge.2022.11.1.7385 https://creativecommons.org/licenses/by-nc/4.0/ deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 74 dissemination and reduce vulnerability to external shocks (gebregziabher & niño-zarazúa, 2014). thus, the government expenditure in the social sector is found to have a positive impact on longterm economic growth and development (benhabib & spiegel, 1994). in addition, such policies play an important role in poverty and inequality reduction (fiszbein et al., 2014). indeed, the millennium development goals that bring poverty reduction, equity, and risk management to the forefront of debates further drew the attention of policymakers and development agencies toward establishing a strong social welfare system, particularly for the poorest and most vulnerable section of the society (baldacci et al., 2008). despite its relevance, few studies have attempted to explore the relationship between income inequality, economic growth, and government spending in the social sector2. most of the empirical studies have analyzed either the impact of government spending in the social sector on economic growth (antonia afonso & alegre, 2011; antónio afonso & furceri, 2010; folster & henrekson, 2001) or its role in poverty and income inequality reduction (cubero & vladkova, 2010; foster, 2012; ospina, 2010; rudra, 2004). further, these studies have not considered the role of gross income inequality in determining the impact of government spending on social spending and how such an impact influences the relationship between economic growth and net income inequality. while gross income inequality is pre-tax and government transfers’ income inequality, net income inequality is post-tax and government transfers’ income inequality. thus, our aim was to study the role of social spending in determining the relationship between economic growth and inequality. first, we analyzed the impact of gross income inequality and other political and economic factors on social spending. second, we examined which categories of social spending (education, health, and social protection) are effective in reducing income inequality and the effects of these policies on economic growth. the remainder of this paper is structured as follows: section 2 provides a brief summary of our literature review, which deals with income inequality, economic growth, and government spending in the social sector. section 3 describes the database and empirical methodology. section 4 delineates the findings and results. section 5 gives the discussion and policy implications. section 6 provides the concluding remarks. 2. literature review since the seminal work of kuznets (1955), the relationship between economic growth and income inequality has drawn a great deal of attention among the academia and policy circles. kuznets established an inverted-u shaped relationship between economic growth and income inequality. he argued that as an economy develops, its structure shifts from the agricultural to the industrial sector, and increases the income of few people. this thus increases the level of inequality. therefore, at the early stages of economic development, there is a positive relationship between economic growth and income inequality. however, at the later stages, inequality diminishes. this 2 arjona et al. (2003) examined the income distribution and social expenditure effects on economic growth. deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 75 indicates a negative relationship between economic growth and inequality. kuznets hypothesis thus implies that redistributive policies have negligible effects on development. till the 1970s, most of the oecd (organization for economic cooperation and development) countries experienced a downward trend in income inequality. this provides evidence of kuznets hypothesis, where a virtuous circle seems to appear: lower inequality fosters economic growth, which in turn lowers the inequality level (aghion et al., 1999). however, during the twentieth century, the developed world experienced a sharp increase in inequalities (gottschalk & smeeding, 1997). this thereby challenged the traditional theory on the economic history perspective proposed by kuznets. as a result, new theories of economic growth evolved during the 1990s to shed light on the effect of growth on income inequality. these theories focused on three new phenomena: (1) trade liberalization, (2) technical change, and (3) the emergence of new organizational forms that were the key components of the growth process (aghion et al., 1999). these factors have widened the income disparity, which in turn has revived the importance of fiscal policy as a redistributive tool to mitigate the negative impact of the growth process in terms of income inequality. with the development of endogenous growth theory, the 1990s have witnessed a resurgence of interest in the determinants of economic growth. there is vast literature related to this, and several studies have examined the impact of inequality on economic growth. one strand of the literature focuses on the pro-growth effects of inequality (forbes, 2000; frank, 2009). forbes (2000) provided evidence of a positive relationship between inequality and economic growth in a large number of countries. similarly, frank (2009) showed a positive inequality-growth relationship that is driven principally by the concentration of income in the upper end of the income distribution across the united states. such a relationship exists because of a) the marginal propensity of the rich to save is greater than that of the poor (kaldor, 1955; lewis, 1954) and b) incentive considerations (mirrlees, 1971). another strand of the literature proposed alternative channels of interaction between income inequality and economic growth, and produced a negative relationship between the two variables (alesina & perotti, 1996; alesina & rodrik, 1994; persson & tabellini, 1994). the channels through which inequality may be harmful to growth include a) political-economy arguments and b) capital market imperfections. persson and tabellini (1994) reported that past inequality is negatively related to the current growth rate of per capita income. they used ordinary least square regressions over a cross-section of nations and offered a political explanation for their result. similarly, alesina and perotti (1996) identified the political channel through which income inequality harms growth. they suggested that an increase in inequality results in socio-political instability, which causes an uncertain political environment. this, in turn, has a negative effect on investment and consequently reduces growth. based on regression analysis, cubero and vladkova (2010) provided evidence of a negative relationship between income inequality and economic growth in the transition economies of central and eastern europe and the commonwealth of independent states. thus, these studies contradict the conventional textbook approach that suggests inequality generates incentives and therefore accelerates growth. thus, deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 76 the extant literature on the relationship between economic growth and income inequality gives mixed results. further, economic growth and income inequality are considered as endogenous outcomes of economic systems. they are influenced by common factors such as structural changes and macroeconomic policies to which both indicators respond simultaneously. however, policy interventions by the government may lead to a loss of economic efficiency that okun (1975) described as the “big trade-off” and used the metaphor “the leaky bucket” to explain this. he argued that government policies, including taxes and transfers, contribute to an equal distribution of income, but this comes at the cost of economic growth. several papers point out that there are some categories of government expenditure that may promote growth and reduce income inequality. other categories may however imply standard equity-efficiency trade-offs that preoccupied okun’s law (benabou, 2000; lopez, 2003; saint-paul & verdier, 1993; seshadri & yuki, 2004). using the dynamic general equilibrium model, seshadri and yuki (2004) showed that redistributive policies, that is, money transfers and educational transfers, result in very large gains in output, but this occurs at the cost of equity. however, within the class of redistributive policies, educational transfers increase economic growth and simultaneously promote equality. sala and roca-sagalés (2011) confirmed the standard equity-efficiency hypothesis which states that current government spending produces non-keynesian effects and has a large negative impact on inequality. however, public investment in infrastructure promotes economic growth and reduces income inequality. as far as taxes are concerned, both direct and indirect taxes generate a positive impact on economic growth. although there is no evidence of an equity-efficiency tradeoff regarding direct taxes, indirect taxes increase inequality. further, lopez (2003) found that improvements in education, infrastructure, and lower inflation levels lead to both growth and progressive distributional change. turnovsky (2015) focused on the role of public investment in determining the dynamics of economic growth and income inequality. two alternative frameworks that were used however gave contrasting results. in the first framework, that is, “representative consumer theory of distribution,” public investment in infrastructure tends to enhance productivity of private capital, and stimulate growth but private capital being more unequally distributed than labor tends to increase inequality. in the second framework, “idiosyncratic productivity shocks,” the degree of income inequality is determined by productivity shocks, which in turn determines growth rate, and implies a negative growth-inequality relationship exists. thus, the extant literature on growth-inequality relationship provides conflicting evidence, which thereby emphasizes the role of fiscal policy as a redistributive tool and as an instrument for promoting economic growth. this paper exclusively focuses on the role of government expenditure in the social sector (education, health, and social security) as a mechanism for determining the relationship between inequality and economic growth. public social expenditure as a source of economic growth has been widely debated in both the developing and developed world. in recent years, several emerging market economies such as indonesia, thailand, and india have undertaken extensive social protection programs that have contributed to economic growth. several developed deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 77 countries including greece, italy, and france however have reduced their social spending after the 2008 crisis as an austerity measure. however, with these diverse policies, both emerging and developed economies have seen an increase in income inequality. this raises the important question of the relationship between public social expenditure aimed at reducing economic inequality and its impact on economic growth. 2.1 social spending, economic growth, and income inequality economists have emphasized the principle that there is “no free lunch,” which implies that even if something appears to be free, there is always a cost to the individual or to society as a whole3. the concept has been applied to the issue of social spending. it was contended by the economists that social spending in the form of health insurance, social welfare programs, and old-age pensions often comes at the cost of economic growth. such spending causes distortions in economic activity because of the disincentives embedded in the structure of welfare spending. lindert (2004) opined that such interventions could nevertheless cause economic externalities that can offset their potential distortions. further, he stated that social spending often positively influences gdp, even if the effects of the taxes that financed the spending are weighed. not only public education spending but also many social transfer programs and health spending increase the gdp per person. as a result, extensive literature that addresses the impact of social welfare spending on economic growth has evolved. although some authors demonstrated that government spending in the social sector could support economic growth ( afonso & alegre, 2011; afonso & jalles, 2014; baldacci et al., 2008; furceri & zdzienicka, 2012; kumlin & rothstein, 2005), others showed that social spending is detrimental to economic growth (folster & henrekson, 2001; im et al., 2011a; persson & tabellini, 1994). studies that provided evidence of the negative effects of social spending have claimed that government expenditure in the social sector affects taxpayers’ incentive to work, and reduces private savings, which thus affects economic growth. for instance, im et al. (2011) examined the economic effects of social spending in less developed nations and compared the situation with that in developed countries. the result suggests that social spending correlates negatively with economic growth in developed countries. in contrast, studies that advocated positive effects of social spending argue that government expenditure in the education and health sector promotes capital accumulation. this can be seen as a productive government expenditure, which enhances economic growth (kumlin & rothstein, 2005; piachaud, 2013). afonso and alegre (2011) found that productivity significantly depended on education expenditure and that social security and health spending played relevant roles in the economic growth in the euro-area panel during 19702006. furceri and aleksandra (2012) assessed the effects of social spending on economic activity in oecd countries from 1980 to 2005. the results indicate that social spending has expansionary effects on gdp. 3 https://en.wikipedia.org/wiki/there_ain%27t_no_such_thing_as_a_free_lunch https://en.wikipedia.org/wiki/there_ain%27t_no_such_thing_as_a_free_lunch deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 78 the government expenditure in the social sector has significant redistributive consequences. foster (2012) found that social spending, that is, on health, education, and social welfare, seems to be redistributive in oecd countries. it has been argued by some scholars that social spending also helped many countries to enjoy the benefits of economic globalization without a large increase in income inequality (gozgor & ranjan, 2017; urata & narjoko, 2017).d’agostino et al. (2020)studied the relationship between social spending and income inequality during the period 1980-2015 in 26 oecd countries by using panel data analysis. the result shows that social spending reduces income inequality. niehues and niehues (2010) explored whether social spending policies lead to less income inequality or whether the redistributive outcomes are offset by behavioral disincentive effects. the regression results suggest that more social spending effectively reduces inequality levels. ospina (2010) used a panel dataset from 1980 to 2000 to analyze the determinants of income inequality in latin american countries, and they paid special attention to education, health, and social security expenditures. these results show that education and health expenditures have a negative impact on income inequality. cammeraat (2020) studied the impact of social spending on economic growth, income inequality and poverty in european union using ols and tsls regressions. the results show a negative impact of social spending on poverty and income inequality. cubero and vladkova (2010) suggested that the redistributive impact of social spending is evident and progressive in central american countries and that it decreases income inequality. bergh et al. (2020) confirm that health spending results in lower-income quality even with economic globalization. thus, the diversity of empirical findings highlights the need for addressing the interaction between social spending, income inequality, and economic growth. although arjona et al. (2003) showed that market income inequality leads to more social spending, which in turn reduces growth in oecd countries, it does not evaluate the effect of social spending on net income inequality. further, it has been theoretically argued that market income inequality tends to influence redistributive policies (meltzer & richard, 1981). this indicates that gross income inequality could be an important determinant of social policy outcomes. few empirical studies have examined the impact of income inequality on social expenditure (sudasinghe & patmasiriwat, 2014). however, these studies did not make a distinction between gross and net income inequality. accordingly, they have not considered the role of gross income inequality in determining its impact on social spending and how such impact influences the relationship between economic growth and net income inequality. therefore, the present study analyzes the relationship between gross income inequality and social spending in light of the newly available and more consistent dataset, swiid that provides two separate series of inequality informationgross and net inequality. thus, our study attempts to develop and estimate an empirical model of joint determinants of social spending, inequality, and economic growth in the context of developing countries (as shown in figure 1). figure 1 shows the main channels of influence discussed in this paper. deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 79 figure 1. framework describing the relationship between social spending, economic growth, and inequality. in this study, the effects of gross income inequality on social spending in asian economies (line a) were first estimated econometrically. next, the impact of social spending on economic growth (line b) and income inequality (line c) was examined in a system. then, to consider the interaction between three endogenous variables (lines d, e, and f), a simultaneous equation model (sem) was used. 3. database and research methodology 3.1 database for the empirical analysis, a panel dataset of 48 developing countries was used. the countries selected for this study and timeframe were determined by data availability. the annual data on various economic, political, and social factors were obtained from different sources for the period 1990-2015. time series data on gdp, investment, trade openness, inflation, population, and urbanization were taken from world development indicators, world bank. the data on market and net income inequality were obtained from swiid (standardized world income inequality database, 201 ). the variables concerning government spending in the social sector were taken from the government finance statistics, international monetary fund. welfare spending is best captured by spending on health, education, and social security, as reported in the international monetary fund’s government finance statistics (rudra, 2004). therefore, in our analysis, social spending included government spending on health, education, and social protection. in line with the work of muinelo-gallo and roca-sagalés (2013), three-year averages of all the variables in our analysis were considered because this helped to capture the effect of changes in deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 80 government expenditure in health, education, and social security on changes in economic growth and income inequality. 3.2 empirical methodology this section discusses the methodological approach that is used to examine the interrelationship between government social spending, income inequality, and economic growth. the basic econometric specification consists of three equations that explore the linkage between three endogenous variables, namely, (1) economic growth, (2) net income inequality, and (3) government social spending. in a neoclassical framework, tax and expenditure policies can be considered as important determinants of economic growth, but they are unlikely to have permanent effects on economic growth. in the endogenous growth model however where investment in human and physical capital affects the growth rate, government policies can play an important role in the growth process (barro, 1996). thus, the aggregate production function in the context of the endogenous growth model is y = f (l, k, g), where y is the aggregated output, l is the labor force, k is the human and physical capital, and g is the fiscal variable (social spending in our case). therefore, the empirical specification of economic growth can be written as (barro, 1996; afonso & jalles, 2014) yit= α + β1( ssit) + η1(zit) + πi+ αt + µit, [1] where yit is the change in growth rate and ssit is the total government social expenditure and respective components (education, health, and social protection). subscripts i and t denote the country and time period, respectively. the equation also includes some macroeconomic factors found in the literature as potential determinants of economic growth. the vector z comprises the following variables: investment to gdp ratio, population growth rate, inflation, and trade openness. these control variables have been used in several studies examining the determinants of economic growth (afonso & jalles, 2014; barro, 1996; muinelo-gallo & roca-sagalés, 2013). investment has been considered as an engine of economic growth (barro, 2003). therefore, the expected sign of these variables is positive. while the tobin-mundell hypothesis stated that anticipated inflation causes portfolio adjustment, thus lowering the rate of interest and raising investment and growth, studies like barro (1991); castelló-climent (2010) and muinelo-gallo & roca-sagalés (2013) found that inflation negatively impacts economic growth. finally, trade expansion is expected to raise economic growth since, as a country specializes and exports, developmental gains from trade can contribute to a too high level of economic growth. however, human capital, which has been considered an important determinant of economic growth, is not included in the growth equation since the variable of interest in this study, that is social spending (education and health expenditure) denotes “human capital investment.” therefore, the inclusion deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 81 of human capital could lead to a multicollinearity problem. the benchmark equation for income inequality is based on the empirical model of ospina (2010), niehues and niehues (2010), odedokun and round (2004), and rudra (2004): niit= α + β1( ssit) + η1(xit) + πi+ αt + µit, [2] where niit is the net income inequality, and ssit is the government social spending and respective components (education, health, and social protection). in line with the empirical approaches of ospina (2010), niehues & niehues (2010), rudra (2004) and muinelo-gallo & roca-sagalés (2013), controls for the inequality equation take into account following variables urbanization, globalization and civil liberty that are strongly associated with income inequality. urbanization is an important determinant of income distribution. on the one hand, it has been argued that an increase in the urban population contributes to a higher middle class and more employment opportunities leading to a shifting in the labor force from agriculture to the urban sector. as a result, the less paid rural job becomes less important, and inequality is expected to decline. on the other hand, urbanization results in better health and education facilities, thus widening the gap between rich and poor. the measure of civil liberty considers the political control of the richest segment of the society and its influence on income distribution. therefore, an increase in this measure tend to reduce income disparity. finally, the equation for government social spending is based on the theoretical framework given by meltzer and richard (1981), who argued that higher inequality would create pressures for redistribution: ssit= α + β1( gii(t-1)) + η1(kit) + πi+ αt + µit, [3] where ssit is the government social expenditure and gii(t-1) is the one-year lagged market income inequality. based on the empirical works of baqir (2002), buracom (2011), huber et al. (2008), and rudra and haggard (2005), many political, economic, and demographic variables were taken as control variables. the social spending equation incorporates the following variables: (i) trade openness, as rodrik (1999) claimed that openness could have a positive effect on government commitments to welfare; (ii) initial per capita income; (iii) age structure of the population captured by the share of the population aged ≤14 years and ≥65 years (pop_15, pop_65); (iv) political system (parliamentary, presidential, and assembly-elected presidential regimes), as persson and tabellini (2002) showed that parliamentary regimes have larger governments, that is, more expenditure than do presidential ones. (v) revenue and debt/gdp ratio: increase in revenue is expected to have a positive impact on social expenditure, while high public debt tends to reduce expenditure on the social sector. given the interdependence between the variables (social spending, economic growth and income inequality), a complete system of simultaneous equation model (sem) has been considered in this study. the three-way linkage between the variables are analyzed by the deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 82 following equations: yit = α + λ(ni1,it) + β( ss1,it) + η1(popit) + η2(globalizationit) + + η4(investmentit) + η5(inflationit)+ π1,i+ α 1,t + µ1,it [4] niit = ρ +ξ (y2,it ) + χ( ss2,it) + ω1(urbanisationit) + + ω2(globalizationit) + ω3(civil libertyit) + π2,i+ α2,t + µ2,it [5] ssit = ψ + ϧ( gi3,i(t-1)) + ϸ1(pop_15it) + ϸ2(pop_65it) + ϸ3(globalizationit) + + ϸ4(political systemit) + ϸ5(revenueit) + ϸ6(public debtit) + + ϸ7(gdp per captiait) + π3,i+ α3,t + µ3,it [6] where yit is the growth rate, niit is the net income inequality and ssit is the government social expenditure (that is education, health and social protection). in all the equations, country and period dummies are included (π & α ) to control for cross country heterogeneity and time-specific fixed effects i.e. global shocks that might affect the dependent variable but are not captured by the explanatory variables. in this study, the sem model was estimated using three-stage least squares (3sls) because it takes into account heteroscedasticity and contemporaneous correlation among the errors across equations. this thus gives more efficient estimations of the relevant explanatory variables4. two conditions need to be satisfied for the identification of equations in sem. these are (a) order condition: “an equation in a sem satisfies the order condition for identification if the number of excluded exogenous variables from the equation is at least as large as the number of endogenous right-hand variables in the equation” and (b) rank condition: “in a model with m equations and m endogenous variables, an equation is identified if at least one nonzero determinant of order (m-1)(m-1) can be composed of the coefficients of variables excluded from that equation but included in other equations in the model” (wooldridge, 2015). the examination of equations (4-5) shows that the number of exogenous variables in the system is sufficient for the order condition to be satisfied (equations are over-identified). the rank condition can be assumed to be satisfied in a model of such a size (greene, 2003). 4. empirical findings this section presents the empirical results for developing countries during 1990-2013. table 1 shows the descriptive statistics of all the variables used in the analysis. table 2 outlines the results of the 3sls simultaneous model, which examines the relationships between economic growth, income inequality, and social spending. 4 for 3sls methodology, refer to zellner and theil (1962) and greene (2003). deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 83 table 1. summary statistics of all variables. variables observations minimum maximum mean s.d. gdp growth 382 -10.12 20.53 4.47 3.20 net income inequality 348 24.75 62.21 42.74 7.67 gross income inequality 348 24.16 72.90 46.43 8.66 social spending 374 0.18 23.10 7.30 4.40 education spending 367 0.14 11.08 3.69 2.15 health spending 371 0.01 6.78 1.74 1.18 social protection 338 0.01 14.27 2.25 2.70 population growth 384 -1.28 6.60 1.66 1.03 investment 381 6.99 60.50 23.95 9.27 globalization 384 9.85 94.21 51.84 14.25 trade 376 16.21 213.78 78.59 38.15 fdi 380 -4.86 32.86 3.05 3.58 inflation 377 -0.99 2237.44 43.20 208.20 population of 15 years or less 384 13.44 51.55 33.84 7.94 population of 65 years or above 384 2.32 18.98 5.28 2.79 revenue 358 7.94 49.51 23.29 8.26 public debt 384 6.52 229.31 53.28 37.06 civil liberties 381 1.00 7.00 3.96 1.49 presidential regime dummy 384 0.00 1.00 0.22 0.41 parliamentary regime dummy 384 0.00 1.00 0.63 0.48 urbanization 384 -3.49 8.99 2.73 1.59 initial per capita income (log) 384 2.29 7.43 4.40 1.22 note: sources are given in the appendix. 4.1 economic growth equation panel a of table 2 shows the impact of social spending and other macroeconomic variables on economic growth. in the case of control variables, it was found that investment has a significant and positive impact on economic growth. this result is in line with the finding of (barro, 1996, 2000), who showed that economic growth depends positively on the investment ratio. population growth, on the other hand, has expected negative sign with significant impact on economic growth. this suggests that, with the increase in population, some proportion of country’s investment that was used to raise capital per worker gets diverted to provide capital for new workers. therefore, the high rate of the population tends to have a negative effect on economic growth. further, it is argued in the literature that countries that are more open to trade are likely to witness high economic growth because the opening of the economy to the global market deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 84 facilitates the transfer of technology and the diffusion of knowledge and helps to exploit the comparative advantage by increasing exposure to competition (petrakos & arvanitidis, 2008). in our empirical analysis, however, though the coefficient of globalization (measured as kof index) is positive, it is statistically insignificant. similarly, the inflation rate appears to be insignificant in the analysis. this can be due to problems of collinearity with the other macroeconomic variables like investment and social spending that are incorporated into the equation. an important result derived from our analysis is that income inequality stimulates economic growth in developing countries. it may be possible that welfare programs of government in developing countries offset the negative impact of income inequality on economic growth. this result is consistent with the conventional textbook argument and empirical findings of banerjee & vashisht (2010); forbes (2000) and frank (2009). according to them, net income inequality has a positive impact on economic growth because of a) incentive considerations and b) high marginal propensity to save of the rich class. however, our results contradict with the empirical studies of alesina & rodrik (1994), persson & tabellini (1994) and perotti (1996), who provides evidence of a negative relationship between income inequality and economic growth. finally, the results indicate that social expenditure significantly influences economic growth with an expected positive sign across developing countries. this shows that social spending in developing countries serves as a means of “making the pie larger” by stimulating gdp growth. this argument is supported by the success experienced by the asian tigers, which made use of social policy as a means of stimulating economic growth (kwon, 2005). through strategic investments in the workforce’s wellbeing, the asian tigers bring social policy to the mainstream of economic development (kwon, 2005). accordingly, these findings may serve as empirical justification for such a strategic approach to social spending. the positive coefficient of social expenditure further supports the argument given by the developmental welfare advocates that government expenditure on the social sector can be instrumental in growth. the developmental welfare proponents claimed that an increase in social expenditure creates a high-quality human capital base, and reduce social conflict, which thereby increases the level of social cohesion of the country and helps the workforce adapt to radically changing industrial structures and technology. this, in turn, contributes to the nation’s economic growth (im et al., 2011b). furthermore, in relation to the components of social spending, the analysis reveals that after considering all the control variables, education and health expenditure have a significant positive impact on economic growth. this suggested that government expenditure in education and health care sector is an important ingredient to economic growth. similar results were found by folster & henrekson (2001) and im et al. (2011). however, the coefficient of social protection is insignificant, indicating that government expenditure on social protection does not affect economic growth in developing countries. these results clearly quashed the claim of neoliberal economic theories that social protection measures harm the nation’s economy and should, therefore, be reduced in order to boost a country’s competitiveness. deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 85 4.2 net income inequality equation the second part of table 2 (panel b) illustrates the inequality equation. the results show that civil liberty index has negative and statistically significant effects on income inequality while urbanization remains statically insignificant. these findings are broadly similar to the results of lin et al. (2009) and muinelo-gallo & roca-sagalés (2013), who reported that an increase in civil liberty constrains the capacity of the rich to influence policy and thus improves income distribution. globalization entered into the regression equation with a significant positive sign. this thus indicates that trade openness exacerbates income inequality in developing countries. the impact of globalization is small but significant at 1% level. these findings lend support to globalization critics who focus on the developing world (bergh & nilsson, 2010; dreher & gaston, 2008; jaumotte et al., 2013; narayan, 2001) it is interesting to note that in the presence of government social expenditure, the impact of economic growth on income inequality becomes insignificant. it could be possible that welfare programs of government in developing countries offset the effects of gdp growth on income inequality. this result is in line with the findings of rudra (2004). panel b of table 2 considered the control variables. it can be observed that the government social expenditure has a significant and sizable negative impact on income inequality. this finding, that government social expenditure is effective in redistributing income, supports recent world bank and ilo reports that argue that government programmes in primary education or health care and social protection measures help alleviate the deep-rooted aspects of inequality by creating opportunities (or what amartya sen calls capabilities). the coefficient of social spending components, i.e. education spending, health spending and social protection, are all negative and significant. this suggests that these types of government expenditures are effective in reducing income inequality. the obtained results lend support to sen’s argument that allocating more funds to education and health sector improves income inequality, which in turn will entail a positive influence on economic growth. the results confirm the findings of foster (2012) and niehues & niehues (2010). 4.3 social spending equation the lowermost part of table 2 gives the results concerning the determinants of social expenditure. in this case, the dependent variable is changed to take into account the components of social expenditure: aggregate social expenditure (column 1), education expenditure (column 2), health expenditure (column 3) and social protection (column 4). panel c of table 2 shows that initial per capita income is insignificant in the case of overall social expenditure, but it has a significant negative impact on education and health spending. these results contradict the empirical findings of persson & tabellini (2002) and muinelo-gallo & roca-sagalés (2013), who observed that in oecd countries, richer economies more intensively carry out distributive expenditures. further, there is a negative and significant deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 86 relationship between trade and social expenditure. these results support the proponents of the efficiency hypothesis that argues that integration with the global market threatens spending on social programs (adserà & boix, 2002; hicks & zorn, 2005; kaufman & segura-ubiergo, 2001; noy, 2011; rudra & haggard, 2005b). they proposed that the quest for international competitiveness places important constraints in social spending and leaves the government with no choice but to cut social spending. concerning the components of social expenditure, while trade openness promotes education spending, it is insignificant in the case of health and social protection expenditure. when demographic variables are considered, aged population (pop > 65) estimates are negative in the case of education spending but have opposite effects on health spending. the increase in the elderly population exerts pressure on the government to increase budget allocations in the health care sector (as reported by sanz & velázquez 2007). furthermore, the population under 14 years of age is significantly related all three components of social expenditure. an increase in pop < 14 pushed up the level of education and health expenditure but lowered social protection spending. similarly, a significant positive impact of revenue is found on social spending. lastly, the significant and positive coefficient of parliamentary and presidential regimes indicates that as compared to assembly elected presidential regimes, both parliamentary and presidential systems are associated with higher levels of social spending. further, the results confirm persson and tabellini's hypothesis in the case of health spending, showing that parliamentary regimes seem to be associated with larger health expenditure, but contradicts persson & tabellini (2002) in the sense that presidential regimes are associated with higher education and social protection expenditure. the results obtained show that after considering the standard determinants of social spending (i.e. controlling for all political, economic and demographic variables), lagged gross income inequality has a significant positive impact on both education and health spending but is insignificant in the case of social protection. this result is consistent with the channel emphasized in the seminal work of meltzer & richard (1981), who argued that a higher market income inequality exerts pressure for more redistribution. deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 87 table 2. sem – regression results panel a growth equation real gdp per capita growth (3sls) real gdp per capita growth (3sls) real gdp per capita growth (3sls) real gdp per capita growth (3sls) net inequality 0.582*** (0.134) 0.479*** (0.119) 0.605*** (0.137) 0.288 (0.192) social spending 0.610*** (0.185) education spending 1.253*** (0.465) health spending 1.636*** (0.683) social protection 0.268 (0.515) population growth -0.754*** (0.352) -0.569** (0.360) -0.821*** (0.336) -0.719*** (0.377) inflation 0.001 (0.001) 0.001 (0.001) 0.001 (0.001) 0.001 (0.001) globalization 0.008 (0.023) 0.021 (0.027) -0.026 (0.024) -0.032 (0.0232) investment 0.120*** (0.031) 0.106*** (0.029) 0.128*** (0.022) 0.141*** (0.024) country effects yes yes yes yes time effects yes yes yes yes adj r2 0.11 0.16 0.06 0.39 no. of observations 314 309 311 290 deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 88 panel b net income inequality equation net income inequality (3sls) net income inequality (3sls) net income inequality (3sls) real gdp per capita growth (3sls) economic growth -0.013 (0.123) -0.018 (0.135) -0.054 (0.132) -0.074 (0.138) social spending -0.264*** (0.122) education spending -0.608*** (0.303) health spending -1.791*** (0.522) social protection -0.888*** (0.285) civil liberties -0.632*** (0.208) -0.571*** (0.226) -0.561*** (0.231) -0.370 (0.257) urbanization 0.227 (0.215) 0.209 (0.237) 0.377 (0.233) 0.292 (0.266) globalization 0.023*** (0.311) 0.015 (0.022) 0.044*** (0.024) 0.053*** (0.025) country effects yes yes yes yes time effects yes yes yes yes adj r2 0.85 0.84 0.84 0.84 no. of observations 314 309 311 290 deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 89 panel c social spending equation aggregate social spending education spending health spending social protection lagged gross income inequality 0.646*** (0.112) 0.121*** (0.044) 0.082*** (0.024) 0.058 (0.052) pop<14 0.439*** (0.079) 0.127*** (0.032) 0.119*** (0.017) -0.164*** (0.041) pop>65 -0.198 (0.331) -0.094*** (0.139) 0.176*** (0.074) 0.006 (0.179) parliamentary regime 1.488 (0.994) 0.304 (0.415) 0.639*** (0.222) 0.470 (0.515) presidential regime 2.026*** (0.782) 0.089*** (0.340) 0.139 (0.174) 1.221*** (0.459) globalization -0.036* (0.026) 0.016*** (0.010) -0.001 (0.001) -0.011 (0.013) revenue 0.050 (0.033) 0.049*** (0.013) 0.020*** (0.007) 0.022 (0.018) initial per capita -0.592 (0.639) -0.470*** (0.267) -0.405*** (0.143) -0.007 (0.342) country effects yes yes yes yes time effects yes yes yes yes adj r2 0.40 0.71 0.75 0.80 no. of observations 322 317 319 297 note panel estimates of 48 developing countries using data for 1990-2013. standard errors are given in parentheses. the statistical significance of the coefficients is as follows: ***significant at the 1% level, **significant at the 5% level and *significant at the 10% level. 5. discussion and policy implications the 3sls simultaneous equation model gave mixed empirical results for developing countries for the relationship between components of social spending, economic growth, and income inequality during 1990-2013. the analysis shows that social protection, which has been recognized as a crucial mechanism by international organizations (e.g., world bank, ilo, and uno) and policymakers for reducing vulnerability without dampening economic growth, produces a significant reduction in net income inequality. this means that social protection measures that aim at achieving more equality do not harm economic growth in developing countries. in contrast, education and health spending both emerged as important components of social spending that can break the tradeoff between equity and efficiency, that is, it can lead to both growth and progressive distributional change. the results support sen’s approach to income inequality, which was based less on income and more on capability. according to this approach, policies like deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 90 boosting spending on health, education, and active labor reforms not only improve income distribution but also spur economic growth. however, the results defy the theoretical argument pointed by okun (1975), who argued that policy interventions by the government could improve income distribution, but this comes at the cost of distorting incentives for work, investment, and so on, which in turn leads to worse economic performance. besides elaborating on the effects of social expenditure on growth and income inequality, the 3sls estimates also shed light on the determinants of social spending. overall, control variables such as the form of government, population below 15 years, globalization, and revenue, performed as expected across all the specifications. the results suggest that an increase in market income inequality leads to more demand for health and education expenditure. this finding confirms the important role of gross income inequality in determining social spending outcomes, as pointed out by meltzer & richard (1981), who claimed that a wider market inequality creates pressures for redistribution. on the contrary, the results do not give evidence in favor of the hypothesis that gross income inequality has a positive or negative effect on social protection spending these results have important policy implications for developing countries. first, with high budget deficits, the most appropriate policy for reducing income inequality without retarding growth is to increase budget allocation in the health and education sector. such policy interventions by the government will facilitate the process of innovation, knowledge creation, and information and will have positive effects on economic growth (benhabib & spiegel, 1994). the enhancement of human capital development will further contribute to a reduction in income inequality. second, social protection has emerged as an important redistributive policy. therefore, to effectively tackle inequalities and promote economic growth, social protection systems need to be coherently designed in developing countries. social protection policies in these countries need to focus on investment-type social spending, such as on active labor market policies and child education. such policies can help to foster economic growth while enhancing child development, which has long-term payoffs in its own right. however, to do so, policymakers have to find innovative ways to reduce financing constraints and work on strengthening the political system that shapes the social protection policy of the country. 6. conclusions the recent food and economic crisis have brought social spending to the forefront of policy analysis. the interrelated shocks in food, fuel, and financial markets have resulted in a slowing down of the progress that many countries had made toward achieving internationally agreed development goals, including the millennium development goals (united nations escap, 2013). these jolts make a strong case for expanding health and education services and moving toward social protection measures to mitigate the social costs of the crisis without harming economic growth. in this paper, we examined the importance of social spending as a redistributive tool and deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 91 an instrument for promoting economic growth in developing countries. the estimates show that increasing the expenditure on the social sector improves the distribution of income as well as promotes economic growth. as far as the components of social spending are concerned, the effects of social security expenditure on output were found to be statistically insignificant. both increases in education spending and health spending can break the trade-off between efficiency and equity since an increase in education spending reduces inequality while promoting output. on the other hand, social protection spending is found to be redistributive in developing countries. thus, after coalescing these results, we conclude that social expenditure and social spending can be used as a developmental tool in developing economies. references adserà, a., & boix, c. (2002). trade, democracy, and the size of the public sector: the political underpinnings of openness. international organization, 56(2), 229–262. https://doi.org/10.1162/002081802320005478 afonso, a., & alegre, j. g. (2011). economic growth and budgetary components: a panel assessment for the eu. empirical economics, 41(3), 703–723. https://doi.org/ 10.1007/s00181-010-0400-9 afonso, a., & furceri, d. (2010). government size composition, volatility and economic growth. european economic review, 26(4), 517–532. https://doi.org/10.1016/j.ejpoleco.2010.02.002 afonso, a., & jalles, j. t. (2014). fiscal composition and long-term growth fiscal composition and long-term growth. applied economics, 46(3), 349–358. https://doi.org/10.1080/00036846. 2013.848030 aghion, p., caroli, e., & garcía-peñalosa, c. (1999). inequality and economic growth: the perspective of the new growth theories. journal of economic literature, 37(4), 1615–1660. https://doi.org/10.1257/jel.37.4.1615 alesina, a., & perotti, r. (1996). income distribution, political instability, and investment. european economic review, 40(6), 1203–1228. https://doi.org/10.1016/0014-2921(95) 00030-5 alesina, a., & rodrik, d. (1994). distributive politics and economic growth. quarterly journal of economics, 109(2), 465–490. https://doi.org/10.2307/2118470 arjona, r., ladaique, m., & pearson, m. (2003). growth, inequality and social protection. canadian public policy, 29, s119–s139. https://doi.org/10.2307/3552279 baldacci, e., clements, b., gupta, s., & cui, q. (2008). social spending, human capital, and growth in developing countries: implications for achieving the mdgs. world development, 36(8), 1317–1341. https://doi.org/10.1016/j.worlddev.2007.08.003 banerjee, r., & vashisht, p. (2010). the financial crisis: impact on bric and policy response. baqir, r. (2002). social sector spending in a panel of countries. international monetary fund. barro, r. j. (1991). economic growth in a cross section of countries. the quarterly journal of economics, 106(2), 407–443. https://doi.org/10.2307/2937943 barro, r. j. (1996). determinants of economic growth: a cross-country empirical study. national bureau of economic research, no. w5698. https://doi.org/10.3386/w5698 barro, r. j. (2000). inequality and growth in a panel of countries. journal of economic growth, 5, 5–32. barro, r. j. (2003). determinants of economic growth in a panel of countries. annals of economics and finance, 4, 231–274. benabou, r. (2000). unequal societies: income distribution and the social contract. the american economic review, 90(1), 96–129. https://doi.org/10.1257/aer.90.1.96 benhabib, j., & spiegel, m. (1994). the role of human capital in economic development: evidence from aggregate cross-country data. journal of monetary economics, 34(2), 143–173. https://doi.org/10.1016/0304-3932(94)90047-7 bergh, a., mirkina, i., & nilsson, t. (2020). can social spending cushion the inequality effect of globalization? economics & politics, 32(1), 104–142. https://doi.org/10.1111/ecpo.12143 https://doi.org/10.1162/002081802320005478 https://doi.org/%0b10.1007/s00181-010-0400-9 https://doi.org/%0b10.1007/s00181-010-0400-9 https://doi.org/10.1016/j.ejpoleco.2010.02.002 https://doi.org/10.1080/00036846.%0b2013.848030 https://doi.org/10.1080/00036846.%0b2013.848030 https://doi.org/10.1257/jel.37.4.1615 https://doi.org/10.1016/0014-2921(95)%0b00030-5 https://doi.org/10.1016/0014-2921(95)%0b00030-5 https://doi.org/10.2307/2118470 https://doi.org/10.2307/3552279 https://doi.org/10.1016/j.worlddev.2007.08.003 https://doi.org/10.2307/2937943 https://doi.org/10.3386/w5698 https://doi.org/10.1257/aer.90.1.96 https://doi.org/10.1016/0304-3932(94)90047-7 https://doi.org/10.1111/ecpo.12143 deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 92 bergh, a., & nilsson, t. (2010). do liberalization and globalization increase income inequality? european journal of political economy, 26(4), 488–505. https://doi.org/10.1016/ j.ejpoleco.2010.03.002 buracom, p. (2011). the determinants and distributional effects of public education, health, and welfare spending in thailand. asian affairs: an american review, 3(113–142). https://doi.org/10.1080/00927678.2011.565282 cammeraat, e. (2020). the relationship between different social expenditure schemes and poverty, inequality and economic growth. international social security review, 73(2), 101– 123. https://doi.org/10.1111/issr.12236 castelló-climent, a. (2010). inequality and growth in advanced economies: an empirical investigation. journal of economic inequality, 8(3), 293–321. https://doi.org/10.1007/s10888010-9133-4 cubero, r., & vladkova, i. (2010). government spending on education, human capital accumulation, and growth. imf working papers, 1–42. d’agostino, g., luca, p., & margherita, s. (2020). social transfers and income inequality in oecd countries. structural change and economic dynamics, 52, 313–327. https://doi.org/10.1016/j.strueco.2019.11.004 dreher, a., & gaston, n. (2008). has globalization increased inequality? review of international economics, 16(3), 516–536. https://doi.org/10.1111/j.1467-9396.2008.00743.x feenstra, b. r. c., inklaar, r., & timmer, m. p. (2015). the next generation of the penn world table. american, 105(10), 3150–3182. https://doi.org/10.1257/aer.20130954 fiszbein, a., kanbur, r., & yemtsov, r. (2014). social protection and poverty reduction: global patterns and some targets. world development, 61, 167–177. https://doi.org/10.1016/ j.worlddev.2014.04.010 folster, s., & henrekson, m. (2001). growth effects of government expenditure and taxation in rich countries. european economic review, 45(8), 1501–1520. https://doi.org/10.1016/s0014-2921(00)00083-0 forbes, b. k. j. (2000). a reassessment of the relationship between inequality and growth. the american economic review, 90(4), 869–887. https://doi.org/10.1257/aer.90.4.869 foster, j. (2012). income inequality, welfare spending and globalization. frank, m. w. (2009). inequality and growth in the united states: evidence from a new state‐level panel of income inequality measures. economic inquiry, 47(1), 55–68. https://doi.org/10.1111 /j.1465-7295.2008.00122.x furceri, d., & aleksandra, z. (2012). the effects of social spending on economic activity: empirical evidence from a panel of oecd countries. fiscal studies, 33(1), 129–152. https://doi.org/10.1111/j.1475-5890.2012.00155.x furceri, d., & zdzienicka, a. (2012). the effects of social spending on economic activity: empirical evidence from a panel of oecd countries. fiscal studies, 33(1), 129–152. https://doi.org/10.1111/j.1475-5890.2012.00155.x gebregziabher, f., & niño-zarazúa, m. (2014). social spending and aggregate welfare in developing and transition economies. https://doi.org/10.35188/unu-wider/2014/803-2 gottschalk, p., & smeeding, t. m. (1997). cross-national comparisons of earnings and income inequality. journal of economic literature, 35(2), 633–687. gozgor, g., & ranjan, p. (2017). globalisation, inequality and redistribution: theory and evidence. the world economy, 40(12), 2704–2751. https://doi.org/10.1111/twec.12518 greene, w. h. (2003). econometric analysis. pearson education india. hicks, a., & zorn, c. (2005). economic globalization, the macro economy, and reversals of welfare: expansion in affluent democracies, 1978-94. in international organization (vol. 59, issue 3). https://doi.org/10.1017/s0020818305050228 huber, e., mustillo, t., & stephens, j. d. (2008). politics and social spending in latin america. the journal of politics, 70(2), 420–436. https://doi.org/10.1017/s0022381608080407 im, t., cho, w., & porumbescu, g. (2011a). an empirical analysis of the relation between social spending and economic growth in developing countries and oecd members. the asia pacific journal of public administration vol., 33(1), 37–55. https://doi.org/10.1080/ 23276665.2011.10779377 im, t., cho, w., & porumbescu, g. (2011b). an empirical analysis of the relation between social spending and economic growth in developing countries and oecd members. asia pacific journal of public administration, 33(1), 37–55. https://doi.org/10.1080/23276665. 2011.10779377 https://doi.org/10.1016/%0bj.ejpoleco.2010.03.002 https://doi.org/10.1016/%0bj.ejpoleco.2010.03.002 https://doi.org/10.1080/00927678.2011.565282 https://doi.org/10.1111/issr.12236 https://doi.org/10.1007/s10888-010-9133-4 https://doi.org/10.1007/s10888-010-9133-4 https://doi.org/10.1016/j.strueco.2019.11.004 https://doi.org/10.1111/j.1467-9396.2008.00743.x https://doi.org/10.1257/aer.20130954 https://doi.org/10.1016/%0bj.worlddev.2014.04.010 https://doi.org/10.1016/%0bj.worlddev.2014.04.010 https://doi.org/10.1016/s0014-2921(00)00083-0 https://doi.org/10.1257/aer.90.4.869 https://doi.org/10.1111%0b/j.1465-7295.2008.00122.x https://doi.org/10.1111%0b/j.1465-7295.2008.00122.x https://doi.org/10.1111/j.1475-5890.2012.00155.x https://doi.org/10.1111/j.1475-5890.2012.00155.x https://doi.org/10.35188/unu-wider/2014/803-2 https://doi.org/10.1111/twec.12518 https://doi.org/10.1017/s0020818305050228 https://doi.org/10.1017/s0022381608080407 https://doi.org/10.1080/%0b23276665.2011.10779377 https://doi.org/10.1080/%0b23276665.2011.10779377 https://doi.org/10.1080/23276665.%0b2011.10779377 https://doi.org/10.1080/23276665.%0b2011.10779377 deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 93 jaumotte, f., lall, s., & papageorgiou, c. (2013). rising income inequality: technology, or trade and financial globalization? imf economic review, 61(2), 271–309. https://doi.org/ 10.1057/imfer.2013.7 kaldor, n. (1955). alternative theories of distribution. the review of economics studies, 23(2), 83–100. kaufman, r., & segura-ubiergo, a. (2001). globalization, domestic politics, and social spending in latin america. world politics, 53(4), 553–587. https://doi.org/10.1353/wp.2001.0016 kumlin, s., & rothstein, b. o. (2005). making and breaking social capital: the impact of welfarestate institutions. comparative political studies, 38(4), 336–365. https://doi.org/ 10.1177/0010414004273203 kuznets, s. (1955). economic growth and income inequality. american economic review, 45, 1–28. kwon, h.-j. (2005). transforming the developmental welfare states in east asia. development and change, 36(3), 477–497. https://doi.org/10.1111/j.0012-155x.2005.00420.x lewis, w. a. (1954). economic development with unlimited supplies of labour. the manchester school, 22(2), 139–191. https://doi.org/10.1111/j.1467-9957.1954.tb00021.x lin, s.-c., huang, h.-c., kim, d.-h., & yeh, c.-c. (2009). nonlinearity between inequality and growth. studies in nonlinear dynamics and econometrics, 13(2). https://doi.org/ 10.2202/1558-3708.1635 lopez, h. (2003). macroeconomics and inequality. lucas, r. e. (1988). on the mechanics of economic development. journal of monetary economics, 22(1), 3–42. meltzer, a. h., & richard, s. f. (1981). a rational theory of the size of government. journal of political economy, 89(5), 914–927. https://doi.org/10.1086/261013 mirrlees, j. a. (1971). an exploration in the theory of optimum income taxation. the review of economic studies, 38(2), 175–208. https://doi.org/10.2307/2296779 muinelo-gallo, l., & roca-sagalés, o. (2013). joint determinants of fiscal policy, income inequality and economic growth. economic modelling, 30, 814–824. https://doi.org/10.1016/j.econmod.2012.11.009 musgrave, r. a. (1959). theory of public finance. mcgraw-hill book company, inc. narayan, g. (2001). what about the global poor. globalisation from above and below. social alternatives, 20(3), 40. niehues, j., & niehues, j. (2010). social spending generosity and income inequality: a dynamic panel approach (no. 5178). article 5178. https://doi.org/10.2139/ssrn.1675689 noy, s. (2011). new contexts, different patterns? a comparative analysis of social spending and government health expenditure in latin america and the oecd. international journal of comparative sociology, 52(3), 215–244. https://doi.org/10.1177/0020715211408760 odedokun, m. o., & round, j. i. (2004). determinants of income inequality and its effects on economic growth evidence from african countries. african development review, 16(2), 287– 327. https://doi.org/10.1111/j.1017-6772.2004.00093.x okun, a. m. (1975). equality and efficiency: the big trade-off. the brookings institution. ospina, m. (2010). the effect of social spending on income inequality: an analysis for latin american countries. 10-03. perotti, r. (1996). growth, income distribution, and democracy: what the data say. journal of economic growth, 1(2), 149–187. https://doi.org/10.1007/bf00138861 persson, t., & tabellini, g. (1994). is inequality harmful for growth? the american economic review, 84(3), 600–621. persson, t., & tabellini, g. (2002). political economics: explaining economic policy. mit press. petrakos, g., & arvanitidis, p. (2008). determinants of economic growth. economic alternatives, 1, 49–69. piachaud, d. (2013). social protection, redistribution and economic growth. development southern africa, 30(1), 24–38. https://doi.org/10.1080/0376835x.2012.756101 rodrik, d. (1999). where did all the growth go? external shocks, social conflict, and growth collapses. journal of economic growth, 4(4), 385–412. https://doi.org/10.3386/w6350 romer, p. (1994). the origins of endogenous growth. the journal of economic perspectives, 8(1), 3–22. rudra, n. (2004). openness, welfare spending, and inequality in the developing world. international studies quarterly, 48(3), 683–709. https://doi.org/10.1111/j.00208833.2004.00320.x https://doi.org/%0b10.1057/imfer.2013.7 https://doi.org/%0b10.1057/imfer.2013.7 https://doi.org/10.1353/wp.2001.0016 https://doi.org/%0b10.1177/0010414004273203 https://doi.org/%0b10.1177/0010414004273203 https://doi.org/10.1111/j.0012-155x.2005.00420.x https://doi.org/10.1111/j.1467-9957.1954.tb00021.x https://doi.org/%0b10.2202/1558-3708.1635 https://doi.org/%0b10.2202/1558-3708.1635 https://doi.org/10.1086/261013 https://doi.org/10.2307/2296779 https://doi.org/10.1016/j.econmod.2012.11.009 https://doi.org/10.2139/ssrn.1675689 https://doi.org/10.1177/0020715211408760 https://doi.org/10.1111/j.1017-6772.2004.00093.x https://doi.org/10.1007/bf00138861 https://doi.org/10.3386/w6350 https://doi.org/10.1111/j.0020-8833.2004.00320.x https://doi.org/10.1111/j.0020-8833.2004.00320.x deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 94 rudra, n., & haggard, s. (2005a). globalization, democracy, and effective welfare spending in the developing world. comparative political studies, 38(9), 1015–1049. https://doi.org/10.1177/0010414005279258 rudra, n., & haggard, s. (2005b). globalization, democracy, and effective welfare spending in the developing world. comparative political studies, 38(9), 1015–1049. https://doi.org/10.1177/0010414005279258 saint-paul, g., & verdier, t. (1993). education, democracy and growth. journal ofdevelopment economics, 42(2), 399–407. https://doi.org/10.1016/0304-3878(93)90027-k sala, h., & roca-sagalés, o. (2011). government expenditures and the growth-inequality tradeoff: the swedish case. journal of income distribution, 20(2), 38–54. https://doi.org/10.25071/1874-6322.23829 sanz, i., & velázquez, f. j. (2007). the role of ageing in the growth of government and social welfare spending in the oecd. european journal of political economy, 23(4), 917–931. https://doi.org/10.1016/j.ejpoleco.2007.01.003 seshadri, a., & yuki, k. (2004). equity and efficiency effects of redistributive policies. journal of monetary economics, 51, 1415–1447. https://doi.org/10.1016/j.jmoneco.2003.10.006 turnovsky, s. j. (2015). economic growth and inequality: the role of public investment. journal of economic dynamics and control, 61, 204–221. https://doi.org/10.1016/j.jedc.2015.09.009 united nations escap. (2013). economic and social survey of asia and the pacific. urata, s., & narjoko, d. a. (2017). international trade and inequality. international trade and inequality, no.675. wooldridge, j. m. (2015). introductory econometrics: a modern approach. nelson education. zellner, a., & theil, h. (1962). three-stage least squares: simultaneous estimation of simultaneous equations. econometrica, 30(1), 54–78. https://doi.org/10.2307/1911287 https://doi.org/10.1177/0010414005279258 https://doi.org/10.1177/0010414005279258 https://doi.org/10.1016/0304-3878(93)90027-k https://doi.org/10.25071/1874-6322.23829 https://doi.org/10.1016/j.ejpoleco.2007.01.003 https://doi.org/10.1016/j.jmoneco.2003.10.006 https://doi.org/10.1016/j.jedc.2015.09.009 https://doi.org/10.2307/1911287 deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 95 appendix 1. data description and sources. variables description (1990-2013) sources real gdp per capita gdp per capita is the total gross domestic product of a country divided by total population. world development indicators, world bank real gdp growth rate annual growth rate world development indicators, world bank social spending public expenditure on education, health and social security and welfare. it is measured as a percentage of gdp. international monetary fund (gfs) government finance statistics and statistics of public expenditure for economic development database by international food policy research institute tax revenue tax revenue refers to compulsory transfers to the central government for public purposes. it is measured as a % of gdp. world development indicators, world bank public debt public dent refers to a part of the total borrowings by the central government which includes such items as market loans, special bearer bonds, treasury bills and special loans and securities issued by the central bank of the country . it also includes the outstanding external debt. it is measured as central government debt over the gdp. world economic outlook, international monetary fund gross and net income inequality gini’s coefficient of market and net income swiid (standardized world income inequality database, 2014). urbanization urban population (% of total) world development indicators, world bank population annual growth rate of population world development indicators, world bank population<15 years population ages 15 and above as a percentage of the total population world development indicators, world bank population > 65 years population ages 65 and above as a percentage of the total population world development indicators, world bank political regimes parliamentary, assembly elected president and presidential regimes database of political institutions, 2012, world bank civil liberties civil liberties allow for the freedom of expression and belief, associational and organizational rights and rule of law. this index ranges from 1 to 7 scale with 1 representing the higher level and 7 lower level. freedom house inflation annual percentage change in the consumer price index world development indicators, world bank investment total value of the gross fixed capital formation and changes in inventories and acquisitions less disposals of valuables for a unit or sector (as a % of gdp) world economic outlook, international monetary fund globalization covers economic dimension of globalization i.e fdi, fii and terms of trade (kof index). feenstra et al. (2015) , penn world tables http://ebrary.ifpri.org/cdm/ref/collection/p15738coll3/id/85 http://ebrary.ifpri.org/cdm/ref/collection/p15738coll3/id/85 deepti ahuja and deepak pandit / european journal of government and economics 11(1), june 2022, 73-96 96 appendix 2. list of countries in the sample. developing countries (upper and lower middle countries) algeria china jamaica belarus angola fiji mexico egypt botswana malaysia panama ghana iran thailand peru kenya jordan maldives guatemala morocco lebanon brazil venezuela nigeria mauritius colombia russia tunisia namibia costa rica turkey yemen south africa dominica republic bulgaria zambia srilanka bolivia el salvador indonesia mongolia philippines papua new guinea vietnam bangladesh bhutan india pakistan 1. introduction 2. literature review 3. database and research methodology 4. empirical findings 5. discussion and policy implications 6. conclusions references © the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 2 (2025), pages 235-256 https://doi.org/10.17979/ejge.2025.14.2.12026 submitted: may 6, 2025 accepted: aug 18, 2025 published: dec 17, 2025 article the implementation of socially responsible public procurement in greece: barriers and prospects for social economy entities lampros lamprinidis 1, * 1 university of west attica, greece *correspondence: lamprinidislampros@gmail.com abstract. this study explores whether and how social economy entities (sees) in greece engage with socially responsible public procurement (srpp), as introduced by law 4412/2016. it investigates the extent to which sees—specifically koinseps and koispes—are aware of, participate in, and benefit from srpp provisions such as articles 20 and 110, and identifies institutional barriers they encounter. using a mixed-methods approach combining surveys and interviews, the findings reveal that while koispes demonstrate higher awareness and participation, koinseps face legal unfamiliarity, bureaucratic complexity, and liquidity constraints. public authorities, in turn, often lack srpp-related knowledge, contributing to limited implementation. despite these challenges, sees propose concrete reforms, including quotas, administrative simplification, and targeted support structures. the study contributes to the literature on strategic procurement and highlights the importance of institutional capacity, legal clarity, and trust in enabling inclusive economic policies. keywords: socially responsible procurement; social economy entities; koinsep; koispe; public contracts; greece; social economy jel classification: h57; l31; i38; d73; o35 1. introduction public procurement represents a strategic tool not only for achieving economic efficiency but also for promoting broader social objectives, such as inclusion, equity, and sustainable development (lamprinidis, 2024, 2025 a & 2025 c). within the european union (eu), this approach has been institutionalized through directive 2014/24/eu, which explicitly allows for the integration of social and environmental criteria in public contracts. this paradigm shift has been described in the literature as a transition from “traditional” to “strategic” or “socially responsible” public procurement (mccrudden, 2004; european commission, 2017). socially responsible public procurement (srpp) refers to the inclusion of social https://creativecommons.org/licenses/by-nc/4.0/ 236 lamprinidis considerations—such as the promotion of employment opportunities, social inclusion, and support for vulnerable groups—into procurement processes (amitsis, 2014). in greece, directive 2014/24/eu was transposed through law 4412/2016, which includes specific provisions for the participation of social economy entities (sees) in public tenders, particularly in articles 20 and 110. these articles provide a legal basis for reserving contracts for social enterprises and including social clauses in procurement procedures. despite this legal framework, empirical evidence regarding the uptake and effectiveness of srpp in greece remains limited. while policy documents such as the national strategy for public procurement (2016–2020) and its successor for the period 2021–2025 emphasize the need to strengthen the social dimension of procurement, implementation has been inconsistent. moreover, the national strategy for socially responsible procurement, required by the 2021–2025 plan, has not been officially issued, revealing a significant gap between legislative intent and policy execution (lamprinidis, 2024, 2025 b & 2025 c). this study seeks to address this empirical gap by investigating how sees in greece— particularly social integration enterprises (koinsep entaxis) and limited liability social cooperatives (koispe)—engage with the srpp framework. it explores their level of awareness of the legal provisions, their participation in public tenders, the barriers they face, and their proposals for improvement. by doing so, the research aims to contribute to both the academic literature on srpp and the policy discourse on how public procurement can be leveraged to support the development of the social and solidarity economy (sse). 2. objective and methodology of the field research 2.1 research objective and rationale public procurement is increasingly recognized as a policy lever capable of achieving not only economic efficiency but also social justice, inclusiveness, and sustainability (amitsis 2014 & 2016 ; apostolopoulos et al, 2019). this transformation is most clearly reflected in the european commission’s promotion of strategic public procurement (european commission, 2017), which encompasses socially responsible public procurement (srpp) as a key pillar. srpp enables contracting authorities to integrate social considerations—such as employment of disadvantaged groups or support for social enterprises—into public tendering processes (geormas, 2013; kassavetes, 2013; ntoulia, 2015). greece, a member state marked by a prolonged economic crisis and deep structural inefficiencies, presents a particularly relevant case for examining the implementation of srpp. although law 4412/2016 transposed directive 2014/24/eu and introduced enabling provisions (notably articles 20 and 110), existing literature indicates a persistent implementation gap (lamprinidis, 2023). this gap raises critical questions regarding the actual engagement of social economy entities (sees) with the procurement system and the state’s institutional capacity to operationalize inclusive procurement policies. the implementation of socially responsible public procurement in greece the main objective of this research is to assess the extent to which sees in greece— particularly social cooperative enterprises for integration1 (koinsep entaxis) and social cooperatives of limited liability2 (koispe)—are aware of, participate in, and benefit from the srpp framework. the study also aims to identify key structural and institutional obstacles and to capture sees' own proposals for improving srpp policy design and implementation. 2.2 theoretical and conceptual framework this research is conceptually grounded in theories of strategic procurement, institutional embeddedness, and social innovation. it draws on the work of mccrudden (2004) and arrowsmith (2010), who describe public procurement as an instrument of distributive social policy, and applies institutional theory (dimaggio & powell, 1983) to understand the interactions between policy design, administrative systems, and third-sector entities. within the srpp context, the “implementation gap” can be conceptualized as a function of: • regulatory ambiguity • limited institutional knowledge • uneven capacity across public and third-sector actors these factors jointly shape whether legal provisions (such as contract reservation mechanisms) become effective tools or remain symbolic commitments (amitsis, 2016; european commission, 2021). 2.3 methodological design this study adopts a mixed-methods approach, combining quantitative and qualitative techniques to explore both behavioral patterns and contextual explanations. this design ensures triangulation and enhances the robustness and depth of findings, particularly in an under-researched policy field such as srpp in greece. 1 koinseps entaxis are a category of social cooperative enterprises established under law 4019/2011, aiming to promote the labor integration of vulnerable social groups such as people with disabilities, former addicts, exprisoners, and others facing social exclusion. they operate as autonomous legal entities with a strong social mission but receive less institutional support compared to koispes. at least 40% of their workforce must come from the targeted vulnerable populations, and their activity spans various sectors, from services to environmental and cultural initiatives. 2 koispes were established under law 2716/1999 as part of greece’s mental health reform and are legally recognized as both social cooperatives and mental health service units. they aim to provide supported employment opportunities for people with mental disorders, promoting their psychosocial rehabilitation and integration into society. koispes are officially linked to the national health system (nhs) and enjoy institutional recognition, allowing for enhanced cooperation with public authorities, including direct awards under specific conditions. 238 lamprinidis 2.3.1 population and sample the research targeted the total population of greek sees eligible under article 20 of law 4412/2016: • 29 koinsep entaxis and • 28 koispe, registered in the national registry of social and solidarity economy as of october 2020. these two categories (tables 1 & 2 in appendix a) were deliberately chosen due to their legal recognition as social integration entities, their theoretical eligibility for srpp mechanisms, and their operational diversity. koispes, governed by law 2716/1999, are older and institutionally more embedded, while koinseps—established under laws 4019/2011 and 4430/2016— represent a newer, more dynamic yet fragile segment of the greek social economy3. a total of 57 entities were contacted via official emails and phone numbers. 23 completed responses were collected, yielding a response rate of 40.35% (table 1 in appendix a), which is deemed acceptable given the challenges posed by the covid-19 pandemic and the limited administrative capacities of the target population. 2.3.2 data collection primary data were collected between october 2020 and january 2021 through a structured questionnaire and follow-up interviews. the survey instrument was divided into four sections: 1. legal awareness (knowledge of relevant laws and directives) 2. tender participation and success (2016–2019) 3. obstacles and challenges (preand post-award phases) 4. policy recommendations a likert-scale was used for rating perceived barriers and funding priorities, while open-ended questions captured narrative insights and recommendations. where possible, respondents were asked to specify the number and type of tenders they participated in under articles 20 and 110 of law 4412/2016. in cases where digital participation was infeasible, semi-structured telephone interviews were conducted to ensure inclusivity and improve data quality. 3 for the examined social economy entities, entrepreneurship represents the primary source of income, followed by public procurement (figure 1 in appendix b). however, due to their greater recognition by the greek public administration, koispes participate more actively in the public procurement market—where such contracts constitute a comparatively more important revenue stream than for koinseps (figure 2). the implementation of socially responsible public procurement in greece 2.3.3 data analysis quantitative data were analyzed using spss v26, applying descriptive statistics, crosstabulations, and subgroup comparisons between koispes and koinseps. qualitative responses were thematically coded, enabling the identification of patterns regarding institutional knowledge, perceived barriers, and adaptive strategies. the combination of both data types enabled the research to explore not only what is happening (e.g., participation rates) but also why (e.g., perceived exclusion, lack of institutional trust). 2.4 methodological limitations several limitations must be acknowledged: • small sample size limits statistical generalization, though the study is exhaustive within its targeted population. • self-reporting bias may affect certain responses, especially regarding sensitive topics such as legal knowledge or trust in institutions. • the study reflects the pre-pandemic institutional context (2016–2019), although it includes pandemic-related data collection challenges. nonetheless, the research offers rare and valuable empirical insight into the lived experiences of sees navigating the srpp landscape in greece. 3. awareness of legal framework and participation in public tenders a core objective of this study was to investigate the degree of familiarity among social economy entities (sees) with the legal and institutional framework governing socially responsible public procurement (srpp) in greece. awareness of such provisions is not merely informative but critically determines the capacity of sees to access, interpret, and utilize procurement opportunities. the survey results reveal a considerable knowledge gap, particularly among koinsep entaxis. approximately 34.78% of all respondents stated they were unaware of directive 2014/24/eu, which serves as the foundation of srpp in the eu and has been transposed into greek legislation through law 4412/2016. when disaggregated by organizational type, the gap becomes stark: over half of koinseps (54.55%) were unfamiliar with the directive, compared to only 16.67% among koispes. a similar pattern emerged regarding awareness of the srpp concept itself: 54.55% of koinseps indicated they were unfamiliar with srpp, while the vast majority of koispes demonstrated adequate knowledge. these discrepancies may stem from differing degrees of 240 lamprinidis institutional maturity, access to information channels, and the extent of engagement with public authorities. more granular analysis concerning specific articles of law 4412/2016 further supports this observation. while only 13.04% of respondents were unaware of article 20—which allows for the reservation of public contracts for sees—this group consisted solely of koinseps. for article 110, which allows for the incorporation of social clauses in tender evaluation criteria, 26.09% of all respondents declared unfamiliarity, with the figure again disproportionately higher among koinseps (36.36%) than koispes (16.67%). these findings illustrate an institutional asymmetry in legal literacy that risks reinforcing structural exclusion. organizations with limited awareness are less likely to pursue tenders, perceive themselves as eligible, or navigate complex administrative procedures. participation data further reflect this reality. between 2016 and 2019: • 17 sees reported participating in tenders under article 20, with 16 winning at least one contract (figure 3 in appendix b). • 60% of participants engaged in up to four tenders (figure 3), and a similar percentage secured up to four wins (figure 4). • when broken down by type: 12 koispes and 8 koinseps participated (figure 5); 11 koispes and 5 koinseps were successful (figure 6). • notably, 37.5% of koinseps did not participate in any tender, compared to 0% among koispes (figure 5). • furthermore, no koinsep had success in more than 4 contracts (figure 6), whereas koispes showed stronger engagement, with 24.9% participating in over 15 tenders (figure 5) and 18.2% winning more than 12 contracts (figure 6). under article 110, a similar trend emerged. the average participation per entity was two tenders, with most success concentrated in low-volume engagement (figures 7 & 8). again, koispes outperformed koinseps, suggesting not only a higher degree of legal and procedural familiarity but possibly greater institutional trust from contracting authorities (figures 9 & 10). collectively, these findings confirm that legal awareness, institutional positioning, and engagement with srpp are deeply intertwined, shaping the actual ability of sees to access state resources via procurement. 4. obstacles faced by social economy entities in public procurement tenders to better understand the barriers that hinder participation, the survey asked sees to identify obstacles encountered during procurement processes. responses revealed a consistent pattern of structural and operational barriers that impede meaningful access to public contracts (figure 14 in appendix b). the most frequently cited challenge was bureaucratic complexity (21.21%). sees described overwhelming documentation requirements, non-standardized administrative procedures, and a the implementation of socially responsible public procurement in greece lack of digital interoperability as persistent hindrances. these burdens fall disproportionately on small-scale sees, which often lack administrative personnel and procurement expertise. financial constraints represented the second most prominent barrier (18.18%). the requirement to submit bid and performance guarantees, combined with limited access to credit or banking services, effectively excludes many sees from the procurement landscape. this liquidity issue is exacerbated for newly established or early-stage cooperatives. a critical third barrier (15.15%) was identified as the lack of awareness and capacity among public authorities. many contracting officials either do not fully understand the srpp framework or are hesitant to implement its provisions due to legal uncertainty or administrative inertia. this leads to conservative tender design that favors established private suppliers over socially oriented actors. other frequently mentioned obstacles included: • lack of procurement knowledge within sees, making it difficult to prepare competitive bids without costly external consultants; • opaque eligibility criteria and legalistic language in tender calls; • perceptions of closed procurement networks, with sees believing that certain contracts are routinely awarded to the same actors; • delayed payments, further weakening the financial sustainability of successful sees. these findings echo broader concerns in the literature about the mismatch between srpp ambitions and institutional readiness, particularly in countries where procurement systems remain compliance-oriented rather than impact-driven (european commission, 2017). 5. obstacles faced by sees during contract execution even for sees that succeed in public tenders, the post-award phase presents new layers of challenge (figure 15 in appendix b). the most prevalent issue reported was delayed payments by contracting authorities (32.14%). such delays—often extending beyond the contractual deadlines—create serious cash flow constraints, especially for sees operating on limited reserves or dependent on timely payments for payroll and operational costs. beyond financial delays, institutional mistrust and lack of understanding were also noted (14.29%). several sees highlighted tensions with public officials who failed to appreciate the distinctive nature of socially driven service delivery, particularly when projects involved vulnerable groups (e.g., persons with disabilities, mental health service users). in many cases, sees were monitored under rigid performance frameworks more suited to for-profit contractors, resulting in strained cooperation and procedural friction. interestingly, 17.88% of respondents stated that they did not experience significant problems, indicating that institutional barriers are not uniform and may vary depending on the maturity of the see, its prior relationships with public authorities, and the local procurement culture. additional but less frequent challenges included: • ambiguity in contract terms; • inflexible timelines unsuited to social service delivery; • lack of responsiveness from contracting bodies; 242 lamprinidis • difficulty in quantifying social outcomes under standard reporting templates. these findings underscore the multi-phase vulnerability of sees in public procurement—not only during tendering but also throughout contract execution. effective srpp implementation therefore requires holistic support, including adaptive contract management and administrative cultures capable of engaging with mission-driven actors. 6. proposal for improving the srpp institutional framework the field research revealed a wide range of proposals by social economy entities (sees) aimed at addressing the structural barriers that currently hinder their effective participation in public procurement processes. these suggestions were not anecdotal but reflected a detailed understanding of the practical, institutional, and legal deficits that surround srpp implementation in greece (figure 16 in appendix b): 1. procedural simplification: the most frequently proposed reform—endorsed by over 20% of respondents—was the streamlining of procurement procedures. sees emphasized the need to: • minimize documentation requirements, • harmonize eligibility criteria across contracting authorities, • develop user-friendly e-procurement platforms, • reduce administrative duplication in repeated tenders. simplification would particularly benefit smaller sees, which typically operate with limited internal capacity and lack dedicated legal or administrative staff. 2. reserved contracts and minimum quotas: many respondents proposed the establishment of mandatory quotas for contracts reserved for sees, under article 20 of law 4412/2016. this measure would: • ensure regular and predictable access of sees to procurement opportunities, • legitimize their presence in the market, • and reflect a real commitment to social objectives through public spending. the quotas could be expressed as a percentage of total procurement budgets or as fixed numbers of tenders in specific sectors (e.g., care services, supported employment, waste management). 3. capacity building for contracting authorities: a recurrent theme in the responses was the limited institutional knowledge of srpp among public officials. to address this, sees proposed: • development of training modules on srpp and social clauses, • issuance of legal guidelines and templates, • and the inclusion of srpp objectives in public sector performance indicators. capacity building would empower public officials to act with legal certainty and avoid defaulting to risk-averse procurement practices. 4. targeted incentives and regulatory flexibility: some entities called for incentive mechanisms for contracting authorities to increase srpp uptake, such as: • accelerated payment schedules, • bonus points in tender evaluation criteria for social impact, the implementation of socially responsible public procurement in greece • and exemption from certain audit obligations for contracts awarded to sees. these measures could mitigate concerns about administrative burden or legal vulnerability. 5. financial instruments and support: given the liquidity issues reported, sees advocated for: • public guarantee schemes to replace traditional bank guarantees, • pre-financing mechanisms for contracts awarded to sees, • and easier access to microcredit or procurement-specific financing tools. such financial infrastructure would enable sees to meet bidding requirements and deliver services effectively without excessive cash flow stress. 6. establishment of a central srpp coordination body: several sees emphasized the lack of coordination across ministries and procurement authorities. they proposed the creation of a dedicated inter-ministerial unit or helpdesk that would: • oversee srpp implementation, • track compliance with article 20/110 provisions, • and serve as a national focal point for both sees and public buyers. 7. conclusions public procurement, once confined to the technicalities of contract execution and budget efficiency, is now increasingly reframed as a vehicle for achieving broader social objectives. within this evolving landscape, socially responsible public procurement (srpp) holds transformative potential— particularly for countries like greece, where structural inequalities and institutional fragmentation limit the reach of inclusive economic policies. this study has sought to interrogate not only the legal scaffolding of srpp in the greek context, but also its operational viability and resonance among those it purports to serve: the social economy entities (sees). through a focused empirical investigation into the experiences of koinsep entaxis and koispes, the research reveals a persistent implementation gap between the formal provisions of srpp and its real-world impact. although the legal foundation—principally law 4412/2016—is ostensibly aligned with eu directive 2014/24/eu, its potential remains largely unexploited due to a convergence of regulatory ambiguity, administrative inertia, and asymmetrical institutional capacity. 7.1 key findings first, the study documents a striking asymmetry in legal awareness. more than half of koinseps were unfamiliar with core legislative instruments governing srpp, a reality that severely constrains their ability to identify, assess, or pursue public tenders. in contrast, koispes—more embedded in institutional networks—demonstrated greater familiarity and success in procurement processes, underscoring the role of organizational maturity and historical legitimacy in navigating state systems. second, even when sees are aware of srpp provisions, they encounter multi-layered structural barriers: excessive bureaucracy, opaque eligibility criteria, financial guarantees 244 lamprinidis inaccessible to small entities, and recurrent delays in payment. these obstacles not only reduce participation but often negate the very social purpose procurement policies are designed to advance. third, contracting authorities themselves emerge as critical yet underprepared actors. lacking technical guidance and policy clarity, many opt for risk-averse approaches that default to conventional procurement practices. this institutional conservatism reproduces a cycle of exclusion and limits the diffusion of social criteria into mainstream tendering procedures. nevertheless, the study also captures a more hopeful dimension. sees—despite their marginalization—articulated concrete, pragmatic policy proposals to enhance srpp implementation. these include the introduction of reserved quotas, the creation of technical support structures, and simplified administrative pathways. such responses reflect not only operational need but a strategic willingness to co-shape the policy environment. 7.2 theoretical contribution by centering the lived experience of sees within a formalized but under-activated institutional context, the study contributes to strategic procurement theory in two important ways. first, it illustrates that legal transposition is insufficient without administrative translation—laws require enabling conditions, not just existence, to become effective instruments of policy. second, it demonstrates that trust asymmetries and perceived institutional legitimacy are not epiphenomenal, but central to how policies are internalized and acted upon by third-sector actors. moreover, the greek case problematizes assumptions of policy convergence across the eu. it shows that even within a shared regulatory architecture, local institutional cultures, capacity gaps, and path-dependent legacies shape the actual trajectory of implementation. 7.3 policy implications for srpp to function not as rhetorical ornament but as a pillar of inclusive economic governance, the following imperatives emerge: • the greek state must move from normative proclamation to operationalization, ensuring that articles 20 and 110 are not symbolic but actively utilized across procurement authorities. • capacity-building must be systemic, encompassing not only training for public officials but the institutionalization of knowledge management, procedural clarity, and legal certainty. • a national coordination mechanism—possibly situated within an inter-ministerial structure—should be established to monitor implementation, support sees, and streamline procurement practices with social objectives. • lastly, the inclusion of targeted financial instruments (such as guarantee schemes) could mitigate the liquidity barriers faced by sees, enabling them to compete on more equitable terms. the implementation of socially responsible public procurement in greece 7.4 limitations and future research this study, while rich in empirical insight, acknowledges its limitations. the sample is limited in size and scope, reflecting only a segment of the broader see landscape. moreover, the time frame (2016– 2019) does not account for subsequent institutional developments or policy shifts. future research should: • analyze judicial interpretations of srpp-related disputes, as legal precedent increasingly shapes administrative behavior. • conduct comparative studies across eu member states to identify best practices and common failures. • assess the social return on investment (sroi) of srpp contracts in order to substantiate their broader developmental value. 7.5 final reflection the findings of this study call into question the assumption that legal reform alone can drive institutional change. they also challenge the idea that sees are passive beneficiaries of social policy. on the contrary, greek social enterprises have shown they are critical agents, capable of diagnosing policy failures and proposing viable alternatives. what remains is for the public sector to respond—not through temporary measures or pilot programs, but through structural inclusion, co-governance, and a new procurement ethos anchored in solidarity, sustainability, and strategic impact. acknowledgments this article is derived from research conducted as part of my doctoral dissertation at the university of west attica. the material has been further developed and adapted to reach a broader international academic audience. i would like to express my sincere gratitude to my academic supervisors for their valuable guidance throughout the research process, as well as to the participating social economy entities for their cooperation. special thanks are also extended to the support structures and research staff who contributed to the fieldwork. this work received no external funding. 246 lamprinidis references amitsis, g. (2014). the collective challenge of active inclusion of vulnerable groups: developmental lessons from social europe [in greek]. papazisis. amitsis, g. (2016). emblematic initiatives for the preservation of social cohesion in the era of the memoranda: the model of the national strategy for social inclusion [in greek]. papazisis. (series: the reform of the social state, no. 1) apostolopoulos, n., dermatis, z., & liargovas, p. (2019). social economy and social entrepreneurship: the european and greek experience [in greek]. patakis publications. arrowsmith, s. (2010). the law of public and utilities procurement. sweet & maxwell. dimaggio, p. j., & powell, w. w. (1983). the iron cage revisited: institutional isomorphism and collective rationality in organizational fields. american sociological review, 48(2), 147–160. https://doi.org/10.2307/2095101 directive 2014/24/eu of the european parliament and of the council. (2014). on public procurement and repealing directive 2004/18/ec. official journal of the european union, l 94, 65–242. https://eurlex.europa.eu/legal-content/en/txt/?uri=celex:32014l0024 doulia, t. (2015). social economy and social entrepreneurship [in greek]. oselotos publications. european commission. (2017). making public procurement work in europe and for europe (com (2017) 572 final). https://www.eumonitor.eu/9353000/1/j9vvik7m1c3gyxp/vki7eq1l0yr4 european commission. (2021). building an economy that works for people: an action plan for the social economy (swd (2021) 373 final). https://ec.europa.eu/social/blobservlet?docid=24986&langid=en geormas, k. (2013). introductory note. in k. geormas (ed.), social economy: theory, experience and perspectives [in greek]. alternative publications. kassavetes, d. (2013). social entrepreneurship in greece. in k. geormas (ed.), social economy: theory, experience and perspectives (pp. 45–81) [in greek]. alternative publications. lamprinidis, l. (2023a). the eu model of socially responsible public procurement. journal of public policy and administration, 8(3), 1–6. https://doi.org/10.47604/jppa.1852 lamprinidis, l. (2023b). challenges of the social economy action plan for greece in light of socially responsible public procurement. journal of public policy and administration, 8(2), 55–66. https://doi.org/10.47604/jppa.1843 lamprinidis, l. (2024). socially responsible public procurement as a tool for the development of the social economy (doctoral dissertation). university of west attica. https://doi.org/10.12681/eadd/56439 lamprinidis, l. (2025a). socially responsible public procurement as a tool for the development of the social economy. papazisis publications. (series: reforming the welfare state, vol. 13) lamprinidis, l. (2025b). socially responsible public procurement and the social economy: european and global institutional approaches. journal of public policy and administration, 10(1), 46–62. https://doi.org/10.47604/jppa.3297 lamprinidis, l. (2025c). social entrepreneurship and social security: synergies for sustainable social policy. market economy: modern management theory and practice, 24(2), 9–28. law 2716/1999. development and modernization of mental health services and other provisions [in greek]. (1999). https://www.elinyae.gr/ethniki-nomothesia/n-27161999-fek-96a-1751999 https://doi.org/10.2307/2095101 https://eur-lex.europa.eu/legal-content/en/txt/?uri=celex:32014l0024 https://eur-lex.europa.eu/legal-content/en/txt/?uri=celex:32014l0024 https://www.eumonitor.eu/9353000/1/j9vvik7m1c3gyxp/vki7eq1l0yr4 https://ec.europa.eu/social/blobservlet?docid=24986&langid=en https://doi.org/10.47604/jppa.1852 https://doi.org/10.47604/jppa.1843 https://doi.org/10.12681/eadd/56439 https://doi.org/10.47604/jppa.3297 https://www.elinyae.gr/ethniki-nomothesia/n-27161999-fek-96a-1751999 the implementation of socially responsible public procurement in greece law 4019/2011. fourth amendment – law no. 15 – law 4019/2011 [in greek]. (2011). https://www.eiead.gr/wp-content/uploads/2013/11/publications_docs_4h-tropopoihshno.15-n.-4019-2011.pdf law 4412/2016. public works, procurement and services contracts – harmonisation with directives 2014/24/eu and 2014/25/eu [in greek]. (2016). https://ia37rg02wpsa01.blob.core.windows.net/fek/01/2016/20160100147.pdf law 4430/2016. law on social and solidarity economy [in greek]. (2016). http://elib.aade.gr/elib/view?d=/gr/act/2016/4430 mccrudden, c. (2004). using public procurement to achieve social outcomes. natural resources forum, 28(4), 257–267. https://doi.org/10.1111/j.1477-8947.2004.00099.x https://www.eiead.gr/wp-content/uploads/2013/11/publications_docs_4h-tropopoihsh-no.15-n.-4019-2011.pdf https://www.eiead.gr/wp-content/uploads/2013/11/publications_docs_4h-tropopoihsh-no.15-n.-4019-2011.pdf https://ia37rg02wpsa01.blob.core.windows.net/fek/01/2016/20160100147.pdf http://elib.aade.gr/elib/view?d=/gr/act/2016/4430 https://doi.org/10.1111/j.1477-8947.2004.00099.x 248 lamprinidis appendix a – tables table 1. sample of the field research. type of entity total contacted responded koinsep entaxis 29 11 koispe 28 12 total 57 23 % of koispes responded 42,86% % of koinsep entaxis responded 37,93% % of total entitities responded 40,35% table 2. geographical distribution of sees in the field research. type of entity east macedo nia & thrace attica north aegean western greece western macedo nia thessal y ionian islands central macedo nia crete pelopon nese koispe 1 4 1 1 0 0 1 1 3 0 koinse p entaxis 1 5 0 1 1 1 0 1 0 1 total 2 9 1 2 1 1 1 2 3 1 the implementation of socially responsible public procurement in greece appendix b – figures figure 1. sources of funding for entities. figure 2. sources of funding for sees by importance and entity category. 250 lamprinidis figure 3. participation in tenders under article 20 of law 4412/2016. figure 4. success in tenders under article 20 of law 4412/2016. the implementation of socially responsible public procurement in greece figure 5. participation in tenders under article 20 of law 4412/2016 by entity category. figure 6. success in tenders under article 20 of law 4412/2016 by entity category. 252 lamprinidis figure 7. participation in tenders under article 110 of law 4412/2016. figure 8. success in tenders under article 110 of law 4412/2016. the implementation of socially responsible public procurement in greece figure 9. success in tenders under article 110 of law 4412/2016 by entity category. figure 10. success in tenders under article 110 of law 4412/2016 by entity category. 254 lamprinidis figure 11. see’s evaluation of public authorities’ familiarity based on field research. figure 12. opinions of koispe on the familiarity of public authorities with their operation. the implementation of socially responsible public procurement in greece figure 13. opinions of koinsep entaxis on the familiarity of public authorities with their operation. figure 14. barriers to participation in public tenders. 256 lamprinidis figure 15. problems during the execution of public contracts. 1. introduction 2. objective and methodology of the field research 3. awareness of legal framework and participation in public tenders 4. obstacles faced by social economy entities in public procurement tenders 5. obstacles faced by sees during contract execution 6. proposal for improving the srpp institutional framework 7. conclusions references ©the authors 2024. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 13, no. 2 (2024), pages 136-164 https://doi.org/10.17979/ejge.2024.13.2.10003 submitted: oct 26, 2024 accepted: jul 8, 2024 published: dec 3, 2024 article the role of the water framework directive in enhancing water use efficiency in the eu amelia pérez zabaleta1, 2, monica borrat sanjuan2,* 1 aquae chair of water economics 2 school of economics, national university of distance education (uned), avda. senda del rey, 11. planta 3.7. 28040 madrid (spain) *correspondence: monica@biruji.org abstract. the increasing economic activity, population growth and urbanisation are placing increasing stress on europe’s freshwater resources. the european union's water framework directive (wfd) aimed to establish measures to foster efficient use of this valuable natural resource while simultaneously protecting the environment. this study allows for an assessment of the directive's application of public policies to a natural resource, assuming that the efficiency of water use is measured by water productivity. for the purposes of this investigation, purely economic variables and specific variables pertaining to the implementation of the wfd will be considered. the final econometric model indicates that variables with a positive impact, including research and development (r&d) expenditure and the governance index, necessitate the updating of facilities, the implementation of public control of the resource, and the encouragement of citizen interest in influencing eu policies. conversely, variables with a negative impact, including population density and water consumption, indicate that as the utilization of a given resource intensifies, the efficacy of that utilization diminishes. the analysis by country indicates that the northern and more industrialised economies have more efficient water use levels. the evolution over time demonstrates that the wfd is being implemented more extensively in regions where it is most needed, resulting in increased productivity values in areas where they are currently lower. keywords: water; water framework directive; water productivity jel classification:i1; d53; g15; g12; c22 1. introduction in economic terms, water can be considered a public good, a scarce economic good and an essential component of the environment. these characteristics make water a highly valuable resource that must be managed carefully (saleth, 2002). most specialists acknowledge that there is a positive correlation between freshwater withdrawals and economic development. nevertheless, the distribution of these withdrawals by country varies considerably, as evidenced by revenga et al., 2000. the geographical distribution of the resource does not ensure that all consumptive requirements have access to an adequate quantity and quality of water ("wmo”, world https://creativecommons.org/licenses/by-nc/4.0/ mailto:monica@biruji.org the role of the water framework directive in enhancing water use efficiency in the eu meteorological organization, 1997). the distribution of water demands may be concentrated in regions with a limited national supply of resources (cobacho, 2000). faced with the problem of scarcity, water management seeks to integrate the three basic dimensions of water, with the objective of ensuring sustainable and efficient management of infrastructure (cosgrove and loucks, 2015). a 2015 european commission study indicates that the eu can reduce water consumption by 40% through measures in different sectors, including the use of recycled water in factories, improvements in irrigation systems, switching to more drought-resistant crops and repairing leaks in public distribution systems. (eea, european environment agency, 2001). eu water policies encourage member states to implement better water demand management practices. this response to the mounting pressures on water resources is being driven by the necessity to improve the existing supply-demand balance and to bring about a more water-efficient society. this is being achieved through the revision of management practices on water demand (eea, 2017). one of the objectives of both the european union circular economy plan and the spanish 2030 circular economy strategy is to enhance water use efficiency by 10%. the indicator designated as 6.4.1 of sdg 6 is employed to quantify the alterations in water use efficiency and is aimed at addressing the financial aspect of sdg target 6.4 (españa circular, 2030). barraqué (1995, 1998, 2000a, 2000b) is the french academic who has published extensively on water management policy in europe. his work constitutes a valuable reference point for understanding the evolution of european water economics. barbier (2000) also analyses the evolution of consumption patterns and best practices. these two authors, in conjunction with aguilera klink (2000, 2006), arrojo and martínez gil (1999), arrojo (1999, 2005, 2010) at the spanish level, are the pioneers of water economics research. a number of studies have been conducted that analyse water use in europe on a comparative and joint basis. cabrera-marcet et al. (2002) carried out an evaluation of the implementation of the water framework directive (wfd) at a european-wide level. this study highlighted the differences in behaviour between countries in northern and southern europe and focused on three key areas: water consumption; contributions to improvements in pollution from runoff; and water stress. the study concluded that in countries where water resources are scarcest, the management of water is the worst in terms of environmental impact. in their study (wichelns, 2014), wichelns estimated that the water productivity indicator enhances the understanding of farm-level water management in europe. in their study, (viaggi et al., 2014) examined the evolving economic perspectives on water in europe. gutiérrez-martín et al. (2017) conducted an economic analysis of water use in the context of the eu water framework directive (wfd), employing an environmental-economic framework for accounting for water based on a case study of the guadalquivir river basin in spain. in a special issue of the public policy analysis of integrated water resource management (iwrm) in europe, ingold and tosun (2020) present a detailed examination of the topic. mugambi et al. (2021), on the other hand, utilize an econometric model to investigate the relationship between economic growth and the sustainability of water consumption in european union countries. the objective of this article is to examine the efficiency with which water is used within the context of the eu-28. to this end, water productivity is employed as a key indicator, with the aim of 138 amelia pérez zabaleta and monica borrat sanjuan identifying the variables that explain this use from an economic standpoint and in accordance with the implementation of the directives set out in the wfd. the following hypotheses are proposed in order to inform the study: 1) which of the macroeconomic variables that affect the normal functioning of the economy also affect water productivity? 2) what directives have the wfd established to promote higher water productivity? 2. materials and methods at the european level, the wfd introduces the economic principles and methods for the management of european waters, and promotes a more efficient use of the resource. however, the directive does not have direct tools that enable the measurement of the results obtained from improvement initiatives established in accordance with the directive. the european commission (ec, 2001), the european parliament (ep, 2000), avellà et al. (2009), bueno hernández (2007), embid irujo (2007), and carles (2001) all provide valuable insights into this subject matter. the conclusions of the world economic forum (wef, 2002) posit that northern european countries exhibit greater water capture potential due to climatic conditions and a reduced dependence on agriculture, given their advanced economic status and greater focus on the secondary and tertiary sectors than on the primary sector. in consequence, the degree of water stress in these regions differs considerably from southern european countries, where the opposite is the case. the climatological characteristics of these regions result in lower water abstraction rates, coupled with higher rates of water utilisation in agriculture, which represents a significant component of their respective economies. the european environment agency (eea, 2001) reported that the average water availability in greece, spain, and portugal was 2,847 cubic metres per capita. in denmark, germany, and the united kingdom, however, this figure was lower at 2,313 cubic metres per capita. however, as the quantity of water used for agricultural purposes is inconsequential in northern regions, the resultant stress is relatively low (with an average of 9.3). conversely, the opposite is true in the south (average 61.67), where agricultural water demand is clearly significant. this evidence suggests that the need for sustainable water management is much higher in the south than in the north. the implementation of the wfd is therefore particularly important in the weaker areas, while the work carried out in the danube river basin is of particular importance, giving specific impetus to south-east europe. there are examples of how some of the strongest european economies, defined as countries with a high level of gdp and an economic structure based on high technology and service sectors, use water efficiently and the path they have taken in order to achieve this. this includes studies by solanes (1996, 1998), solanes and gonzález-villarreal (1996). in contrast, in the united kingdom (uk) many residential customers lacked the benefit of water meters and their water consumption charges were not related to their usage levels. in england, a government strategy from 2008 permitted the introduction of metering as a means of promoting more effective use of this resource, with the objective of ensuring the availability of sufficient water supplies in the face of anticipated the role of the water framework directive in enhancing water use efficiency in the eu future trends in the population and climate change (hm government; defra department for the environment, food and rural affairs, 2008). in france, a similar approach was taken with respect to irrigation systems. once the limits of abstraction were exceeded, water counters were installed. the proportion of irrigation systems equipped with water counters between the years 2000 and 2003 increased from 54% to 71%, representing 85% of the total irrigated area (rouillard, 2020). as outlined by beecher et al. (1998), these countries have long been aware that water conservation is not a standalone goal, but rather one component of a comprehensive strategy to guarantee a dependable and secure supply of potable water. significant differences are also evident between northern and southern europe at the urban level. given that water stress is a pressing concern for many southern european countries within the european union, there is a clear need to foster greater social awareness about the importance of water conservation and efficient water usage (barraqué, 1995). the eu-28 member states employ a range of policies and measures to enhance water use efficiency. one key instrument is the productivity variable, which serves to quantify the efficiency of water use in terms of its relationship with economic output. each country reports annually to eurostat on the economic output produced per cubic metre of water abstracted, inclusive of all water abstraction sources and exclusive of water utilised in the generation of electricity by hydropower plants. it would be beneficial to perform a global test of this indicator in order to ascertain the correlation between this indicator and the application of the wfd or other macroeconomic variables for which absolute, harmonised values are available for the eu. this could help to identify the key factors that influence the optimal utilisation of water resources across europe. 2.1 data: base composition for the purposes of this study, we utilised data from the official data source of the european union, eurostat. in all instances, the annual data published by each country was employed as an annual average for the period between 2006 and 2015. this information was obtained from the eu-28 dataset. there is no data available which can be associated with governance. consequently, we have employed an alternative data source, the wbgi (world bank governance indicators), which contains governance indicators published by the world bank and which has been compiled as a yearly average for each country between 1996-2017. 2.2 bibliographic justification of the selected variables -y: water productivity (€/m3) there are a number of studies that demonstrate the correlation between the water productivity indicator and efficient water use. for example, arreguín-cortés et al. (2004) present techniques for efficient water use at the household, industrial, municipal and basin levels, and conclude, on the basis of the evidence presented, that although there are more sophisticated techniques and equipment available, the correlation between water productivity and efficient water use remains. the actions are typically 140 amelia pérez zabaleta and monica borrat sanjuan implemented in isolation, with only limited integration into broader programmes. this is contingent on a number of factors, and his final recommendation is that water efficiency programmes should be supported at the river basin level, with a comprehensive delineation of the roles and responsibilities of all water users. conversely, several authors have established a relationship between the water productivity indicator and various economic parameters. in particular, water is mainly investigated in two areas: firstly, in relation to agriculture, where the majority of consumption occurs; and secondly, in the analysis of wastewater, where it has a greater environmental impact. ibarra (2014) and miracle (2006) provide detailed analysis of these two areas. secondly, in the context of wastewater analysis, where it has a greater environmental impact. in their study, cruz león and bielsa callau (2001) cite specific references to illustrate their argument. their focus is on the search efficiency, sustainable use of water resources and land management in the zaragoza region. the authors highlight that improvements in efficiency are not limited to technological change in the narrow sense of substituting one irrigation technique for another. in contrast to the aforementioned, they often relate to measures that do not necessitate the undertaking of costly investments. instead, they pertain to the comprehension of hydric demand. consequently, there is considerable scope for enhancement of the design and networks aimed at meeting the forecasting needs of users in the requisite amount and at the required time. in the context of industrial economics, a study conducted by cruz et al. (2003) examined the marginal productivity of water in the manufacturing industry within the colombian context. the study’s principal conclusion is that numerous industrial sectors, deemed water-intensive, have succeeded in significantly reducing their water access costs (abstraction, treatment, and distribution). moreover, these sectors demonstrate a relatively low private marginal willingness to pay for the use of water as a raw material. a "wuf, water use fee" that incorporates the private marginal willingness to pay for the use of water as an input could facilitate a more efficient distribution of this natural resource among the various industrial activities within the manufacturing sector. this is particularly relevant when one considers that the study found that the own price elasticity of water as a raw material is approximately -1. vargas ovando (2015) employs a marginal factor productivity approach in order to determine the economic value of water within the context of the chilean manufacturing industry. the aforementioned approach considers a number of variables, including the quantity of water, capital, labour, energy and intermediate materials, among others. in the urban domain, sánchez garcía and blanco jiménez (2012) investigate the potential of using tariffs to regulate water usage in urban centres. their findings indicate that the existing pricing system has become inadequate in its ability to oversee water consumption. this necessitates a reassessment of pricing policy and the potential introduction of a novel tariff structure in spain. this investigation connects the examination of water with the implementation of the wfd. this requires the implementation of pricing systems that recuperate the costs of water resources and the establishment of national pricing policies that contribute to achieving sustainable water use. water tariffs should be employed as a means of regulating consumption, with the aim of promoting efficiency and sustainable utilisation of the resource. the role of the water framework directive in enhancing water use efficiency in the eu x1: government effectiveness (% country ranking) as zetland and gasson (2012) observe, governance can influence the management of operators. furthermore, it has been demonstrated that adequate public intervention is correlated with both the quality of the water supplied and the extent and condition of the water supply and sanitation network. consequently, investment in infrastructure and governance would be found to be positively correlated. in particular, it can be reasonably argued that the dimension of governance that is most likely to influence the water sector is the regulatory quality of the states. therefore, this will be the explanatory variable incorporated into the empirical model. the analysis will include data from a national source for the specified period, namely the wbwgi (kaufmann et al., 2010). di vaio et al. (2021) propose that collaboration, coordination, and stakeholder engagement are essential elements that should be incorporated into water governance models in order to effectively address the challenge of global sustainability. x2: total water consumption (millions of m3) in the econometric model developed by colmenárez and salazar (2018) for estimating domestic water production, the exogenous variables included are per capita consumption, population density, and the investment level made in potable water and wastewater. del villar (2010) establishes the relationship between water prices and the demand for domestic use in spain, taking into account not only consumption but also other variables such as the level of investment, the population growth rate, and per capita income. in their study (martín-ortega et al., 2008), the authors characterised water usage within the guadalquivir demarcation area, applying the principles of the wfd. one key variable identified was the impact of agricultural water consumption on discharge. martínez and goetz (2007) consider the specificities of the water markets in the irrigation area, with variables including operating costs, consumption in the area and the productivity of the farms. in their study on energy efficiency and economic regulation in drinking water and sewerage services in latin america and the caribbean (ferro & lentini, 2015), the authors identify and analyse the conditioning factors of consumption in the final uses of water. this is done from the perspectives of both the supply and the demand of the resource, with particular emphasis on the price of water and the costs of treating waste generated. x3: r&d expenditure (millions of €) moro et al. (2018) performs a thorough panel data analysis comparing china and the eu, thereby corroborating previous studies which demonstrate the association between innovation in the water sector and water use efficiency. the eu has promoted water-related research and development mainly through three initiatives: the eu framework programmes horizon 2020 (h2020), life+ programme and “interreg, programme and european territorial cooperation programmes” have constituted the main vehicles for eu funding for water-related research and development. in the period studied (2014-2019), a total of 1,561 water-related projects were implemented. funding for these initiatives 142 amelia pérez zabaleta and monica borrat sanjuan was provided by a number of sources, including the horizon 2020 programme, the life+ programme, and the “interreg” programme, with a total budget of 1.3 billion, 800 million, and 1 billion euros, respectively. for instance, the ecwrti, imetland, remeb and powerstep projects represent pioneering efforts in wastewater treatment processes. their objective is to recycle and reuse wastewater in industry and agriculture, reduce the operational and energy costs of wastewater treatment and increase the amount of energy that can be produced in sewage treatment plants. other projects are similarly striving to enhance the efficacy and efficiency of water and energy use in irrigation systems within the agricultural sector, such as the moses and maslowaten projects. the objective of the centaur project is to reduce the risk and consequence of flooding in urban areas. the cyto-water project has developed a novel technology platform for rapid microbial detection in industrial and environmental waters. finally, subsol is developing commercialised technologies for the use of sub-surface coastal water. reground, on the other hand, is proposing a novel nanogeotechnology for the immobilisation of toxic metals in aquifers and river filtration sites. -x4: population density (people/km2) the distribution of water as a resource throughout europe is not uniform. moreover, the quality and quantity of this resource varies considerably across the continent. in iceland, the average resident is allocated 600,000m³ per year, while in norway, the figure is 100,000m³ per capita per year, and in sweden and finland, 20,000m³. at the other extreme are eastern european countries, such as moldova and ukraine, which have quantities of less than 4,000 m³ per inhabitant per year, while in southern europe; spain, italy and greece, with 3,000 m³ per inhabitant per year, have even smaller endowments. malta has a water endowment of only 100 m³/inhabitant/year, in some cases derived from seawater desalination. xiaoming liu et al. (2021) demonstrate a direct correlation between the consumption of water resources and population density in their comparative study of china's coastal and inland regions. in olmos salvador (2018), both per capita gdp and population density were included as variables in an econometric model for determining residential water tariffs at the international level. in his 2016 study, schreiber (2016) investigates the influence of population density on water usage, particularly in the context of municipal drought situations. domestic economies account for approximately 12% of global water consumption (fao, 2018). chang et al. (2017) demonstrate that an increase in residential housing density is associated with a reduction in total municipal water use. conversely, ghavidelfar et al. (2018) demonstrate that per capita water consumption varies according to social class, challenging the underlying assumption of contemporary urban policy that densifying central city areas can result in significant savings in water use. sampson et al. (2022) propose that the efficiency gains in water use with increased density are not uniform and may be nonlinear, depending on the specific bt movements. the simulation outputs indicate that the greatest gains in water savings per unit change in duua can be achieved with short movements over the lowest density classes (e.g., lsf to a ssf). kim et al. (1998) examine the economic consequences of spatial integration within urban water service markets in korea, applying the concept of economies of scale in terms of costthe role of the water framework directive in enhancing water use efficiency in the eu effectiveness. it may be inferred from the results that changes in population density would have only a minimal effect on the potential for economies of scale. -x5: gross domestic product (millions of € at current prices) gdp and main components (output, expenditure and income); gross domestic product at market prices expressed at current prices, million euro. the national data taken directly from eurostat expressed in millions of constant euros for the year 2015 will be included. gdp at constant prices will be included in the model as a variable representing the degree of economic development. gross domestic product (gdp) is a metric that serves as a proxy for the level of economic activity in a given country. it is defined as the value of all goods and services produced, net of any goods or services utilised in their creation. the calculation of the annual growth rate of gdp volume allows for comparisons of the dynamics of economic development over time and between economies of different sizes, as well as for the analysis of the impact of various factors on the growth of an economy over time. in order to ascertain the growth rate of gdp in terms of volume, the gdp at current prices is valued in the prices of the previous year. the computed volume changes are then imposed on the level of a reference year. this methodology is known as a chain-linked series. consequently, fluctuations in prices will not affect the calculated growth rate. it should be noted that the income components of eurostat gdp, as well as other income measures, are only available at current prices. this is due to the fact that pure monetary flows cannot be decomposed into price and volume components. nevertheless, it is possible to convert these figures to real terms by applying an appropriate deflator. however, the data series on current prices are susceptible to inflationary effects. the relationship between water productivity and gdp has been explored by numerous authors, who have employed gdp at constant and current prices in an interchangeable manner. it follows that inflation is not considered an important factor in analysing this correlation. the study period from 2006-2015 did not include instances of deflation, or negative inflation, within the eu-28 countries. it may be concluded that the estimated prices for each country, expressed in local currency, were consistently higher than the constant price estimates in subsequent years, starting from the reference year. however, it is not currently possible to report the precise impact of this on each individual country. the inverse u-shaped relationship between economic growth rate and relative water scarcity of a territory is quantified by barbier (2004). this allows the optimal rate of water abstraction to be determined, which is defined by the maximum possible growth rate. his conclusion is that while current water consumption levels in many economies do not yet impede growth, there are a select few countries where even moderate or extreme water scarcity can have a markedly adverse impact on economic growth. orloci et al. (1985) conducted an empirical analysis of the correlation between per capita gross domestic product (gdp) and water usage for a sample of 50 countries over three distinct periods. the results indicate an initially positive correlation between the variables. this implies that, at the outset of economic growth, substantial quantities of water are required to elevate gdp. consequently, access to water represents a limiting factor in the advancement of economic growth within developing countries. as a nation progresses to a higher stage of development, the relationship between per capita gdp and water use may weaken and potentially become negative. 144 amelia pérez zabaleta and monica borrat sanjuan this is due to the implementation of water-saving techniques in irrigation, as well as the scarcity of water resources. in his study of the annual growth rate of per capita gdp (g) at constant prices (1990), olmeda (2006) examines the relationship between economic growth and water, analysing the latter from an economic perspective. the country's level of economic development may influence the productivity of its water resources through a number of interrelated mechanisms. in countries with a higher income, wages are typically higher, resulting in increased operational costs for the relevant parties. moreover, it is anticipated that more advanced nations will allocate greater resources to technological advancement, infrastructure construction, and the maintenance of their water supply systems. consequently, this will lead to an increase in costs. additionally, another issue that could also be affected is the tariff system aimed at cost recovery and incentivising resource savings, which is more entrenched in economically developed environments. -x6: public water as a percentage of total water abstracted (%) this variable is defined as a quotient that includes the total public water (wat_proc: water abstraction for public water supply: surface freshwater (millions of cubic metres)) in its numerator and the total water abstracted (annual freshwater abstraction by source and sector [env_wat_abs]. wat_proc, total gross abstraction: fresh surface water and groundwater (million cubic metres) in the denominator. this variable indicates the specific weight of the public sector with regard to water management. the application of the wfd suggests that the implementation of cost recovery principles, particularly the recovery of environmental costs, may positively impact water productivity. accordingly, the proportion of supply, encompassing the fixed charge related to the provision of supply services and the variable amount contingent upon the quantity consumed, would exhibit a marín (2009) conducts a comprehensive comparison between public and privately managed projects, demonstrating that private management is more efficient with respect to water use efficiency than public management. pérard (2009) offers a comprehensive analysis of the economic discrepancies between the public and private sectors, with a focus on 45 countries belonging to the organization for economic co-operation and development. the author posits that the relative efficiency of one model over another is not always clear-cut; rather, the relative efficiency of one model over another is contingent upon a number of factors, including the cost of public funds, taxes and the influence exerted by private operators. furthermore, the costs associated with privatisation implementation and the technical distinctions between the public and private sectors must be taken into account. it has been observed since 1990 that numerous national and local governments in developing countries have entered into contractual relationships with private companies with the intention of allowing them to operate or manage their water utilities under the auspices of the public-private partnership (ppp) model. the underlying assumption was that the private sector would enhance the performance of utilities by introducing new capital, raising the level of staff expertise and improving the costeffectiveness and efficiency of operations. the role of the water framework directive in enhancing water use efficiency in the eu in 1993, lynk reached the conclusion that the privately owned water sector in the uk exhibited significantly higher levels of inefficiency than the publicly owned sector. -x7: total public water supplied to industrial activities and services (millions of m3) the total volume of water supplied to the public via the municipal water supply system, including all nace activities and households. the quantity of water is expressed in million cubic metres. the european statistical office (eurostat) employs this indicator to gauge the quantity of water supplied and managed by public bodies, encompassing both industrial activities and households. this variable indicates the extent to which public entities are involved in the provision of services related to the full water cycle. this indicator permits the identification of countries where water management is provided by public entities, as opposed to those where it is subcontracted by the administration to other private companies. moreover, it permits the correlation between the impact of governmental management and the water used in economic activity associated with industrial and service activities. anwandter and ozuna (2002) found that neither the decentralisation of water operations to municipal levels nor the establishment of an independent regulatory body had a positive effect on the efficiency of mexican water utilities. wackerbauer (2009) pointed out that in germany, the provision of water supply and sanitation services are considered to be core tasks of public services of general interest, with the responsibility for these services resting with the municipalities. -x8: percentage of population connected to the public service (%) the term "access to drinking water" is defined as a connection to a water supply system (typically a piped system) that is accessible to the general public. all figures are presented as percentages of the resident population. access to potable water, sanitation (sewage) and wastewater treatment represent fundamental services that citizens of developed societies expect to receive in a modern context. of the countries for which data are available, only four (estonia, poland, slovakia and romania) have a population coverage of less than 90%. this does not imply that all segments of the populace lack access to potable water. some may have access to it through alternative avenues, such as a private well. however, they remain unconnected to the broader supply network. the incorporation of this variable into the model not only considers the implementation of the wfd, but also extends beyond this, as the un itself is committed to ensuring access to safe and potable water. in accordance with sdg6, universal and equitable access to safe drinking water at an affordable price for all must be achieved by 2030. a study by benito et al. (2015) examines the comparative efficiency of the spanish water sector, concluding that public sector utilities exhibit a higher level of productivity compared to those in the private sector. -x9: percentage of gdp associated to circular economy (%) % gdp: private investments, jobs and gross value added related to circular economy sectors. the indicator includes “gross investment in tangible goods”, “number of persons employed” and “value 146 amelia pérez zabaleta and monica borrat sanjuan added at factor costs” in the following three sectors: the recycling sector, repair and reuse sector and rental and leasing sector. frérot (2014) makes a connection between the principles of the circular economy in europe and their impact on water efficiency. the return to economic growth, the preservation of the environment and the reduction of our dependence on raw materials and energy require a profound transformation of our production and consumer practices. it follows that this transformation will necessarily entail the adoption of the circular economy paradigm, which is defined as a model for the economic activity that sustains human life on a planet of finite resources, and in which the environmental, social and economic systems are in equilibrium. the optimal transformation, recovery and treatment of energy, waste and water are fundamental aspects of economic and environmental sustainability. -x10: percentage of environmental taxes (%) the percentage of total tax revenues comprised of shares of environmental and labour taxes. the indicator quantifies the relative importance of environmental and labour taxes within the broader context of tax revenue, encompassing both direct taxes and social contributions. the term "environmental tax" is employed to describe a tax that has a specific negative impact on the environment, and whose tax base is a physical unit (or a proxy for one) of that impact. environmental tax revenues are derived from four distinct tax types: energy taxes, which contribute approximately three-quarters of the total, transport taxes, which account for approximately one-fifth of the total, and pollution and resource taxes, which contribute approximately 4% of the total. in the context of taxation, the term "labor tax" is defined as all personal income taxes levied on labor income, encompassing both employed and non-employed individuals. this encompasses a diverse array of taxes, including those applied on both employees and employers, such as payroll taxes and social contributions, respectively. the directive places particular emphasis on the definition of environmental objectives (art. 4 and annex v), requiring member states to achieve good ecological and chemical status in all surface waters, all groundwater and marine waters. moreover, the directive sets out the normative criteria for defining good chemical status of waters in terms of compliance with quality standards. as carles and garcía (2001) observed, "water is not a commercial good like any other" and therefore requires a different approach to its management. it is therefore important to "place the emphasis on the nature of water as an eco-social asset". munguía-lópez et al. (2019) demonstrate a positive relationship between water taxes and the optimisation of water consumption and increased aquifer recharge at a water treatment plant in the sonoran desert (mexico). 3. results the econometric model utilised is a panel data analysis where y represents water productivity, with i denoting the individual (eu country) and t signifying the observed period (year). furthermore, other panel data are taken into consideration as a point of reference in order to gain a more the role of the water framework directive in enhancing water use efficiency in the eu comprehensive understanding of the subject matter under analysis. the following references have been consulted: greene (2000), wooldridge (2002), torres-reyna (2007), beck (2001), breitung and das (2005). the panel data for the natural logarithms (ln) are taken for all variables in the model except for those variables that have already been defined as a percentage (%). this approach allows an elasticities analysis to be carried out. 3.1. descriptive analysis of the main variables upon examination of the descriptive analysis of the principal dependent and independent variables, it becomes evident that the number of observations pertaining to each variable differs. in essence, the total number of observations (n) in each case does not reach the requisite 280. this may potentially constrain the applicability of certain multivariate techniques. given the high degree of dispersion observed between the standard deviations of the dependent and independent variables, logarithmic transformations were applied to the latter, where possible, in order to normalise or reduce the dispersion. the resulting variables exhibited a more uniform standard deviation, thereby facilitating interpretation of the results and reducing the dispersion between standard deviations. the graphs in appendix 1 demonstrate considerable variability in productivity between countries, with no discernible evolution over time to determine a trend towards increasing water productivity. the absence of data in certain cases, either total (1: austria, 9: finland, 22: portugal) or partial (2: belgium, 8: estonia, 10: france, 11: germany, 14: ireland, 18: luxembourg, 27: sweden), makes it challenging to draw generalisations about the european union countries as a whole. the application of logarithms ensures that the data are treated uniformly due to the elimination of scale effects. 148 amelia pérez zabaleta and monica borrat sanjuan table 1. statistical description of the study variables. variable units average value standard deviation min max obs y: water productivity (€/m3) 114.125 global 180.012 5.200 1,017.900 n 189.00 enter 186.552 6.470 919.928 n 25.00 intra 20.050 9.397 212.097 t-bar 7.56 x1: government efectiveness (% country ranking) 81.773 global 12.780 44.075 100.00 n 280.00 enter 12.722 47.651 98.946 n 28.00 intra 2.339 74.152 91.768 t-bar 10.00 x2: total water consumption (millions of m3) 5,277.964 global 8,808.575 0.160 35,476.2 n 247.00 enter 9,843.178 77.469 33,322.45 n 28.00 intra 2,157.195 -11,053.42 14,442.75 t-bar 8.82 x3: r&d expenditure (millons of €) 9,148.942 global 16,283.39 31.253 88,781.82 n 280.00 enter 16,406.56 46.368 73,159.16 n 28.00 intra 2,157.873 -5,043.518 24,771.6 t-bar 10.00 x4: population density (people /km2) 171.922 global 245.128 17.300 1,408.4 n 280.00 enter 249.047 17.680 1,326.37 n 28.00 intra 8.019 129.852 253.9521 t-bar 10.00 x5: gross domestic product (millons of € at 2015 current prices) 1,849.161 global 5,286,747 5,386.1 3,44 exp 7 n 280.00 enter 4,697,630 11,174.02 2,51 exp 7 n 28.00 intra 2,567,844 -2,32exp-7 3,22 exp 7 t-bar 10.00 x6: percentage of public water to total water abstracted (%) 21.729 global 12.212 1.799045 48.95259 n 180.00 enter 12.946 3.404095 48.1831 n 24.00 intra 2.281 13.81413 34.12986 t-bar 7.50 x7: total public water supplied to industrial activities and services (millions of m3) 975.490 global 1,327.252 24.74 5,533.4 n 157.00 enter 1,649.514 25.815 5,382.8 n 25.00 intra 100.698 336.7903 1,338.49 t-bar 6.28 x8: total public water supplied (%) 92.011 global 11.520 49.2 100 n 145.00 enter 10.350 57.47778 100 n 23.00 intra 1.640 83.73415 98.23415 t-bar 6.30 x9: percentage of gpd associated to circular economy (%) 0.942 global 0.187 0.36 1.35 n 168.00 enter 0.188 0.405 1.21875 n 24.00 intra 0.077 0.655 1.246 t-bar 7.00 x10: percentage of environmental taxes (%) 7.394 global 1.678 4.32 11.75 n 280.00 enter 1.569 4.459 10.145 n 28.00 intra 0.659 4.896571 9.458571 t-bar 10.00 the role of the water framework directive in enhancing water use efficiency in the eu 3.2 econometric interpretation of the model the final model selected through the application of panel data with fixed effects is as shown below. 𝐿𝐿𝐿𝐿 = 12,56618 + (0,0053926 ∗ 𝑋𝑋1) − (0,2096545 ∗ 𝐿𝐿𝑋𝑋2) + (0,1398058 ∗ 𝐿𝐿𝑋𝑋3) − (1,735613 ∗ 𝐿𝐿𝑋𝑋4) + (0,120062 ∗ 𝑋𝑋6) + (0,4596479 ∗ 𝐿𝐿𝑋𝑋7) + (0,03114292 ∗ 𝑋𝑋8) − (0,0411553 ∗ 𝑋𝑋10) table 2. estimation results of the econometric fixed-effects model. number of observations number of groups r2 r2 intra r2 global 105 17 0.7501 0.0569 0.0682 sigma_u sigma_e rho corr(u_i, xb) f test (16, 80) prob>f 2.0038387 0.070 0.998 0.882 56.81 0 variable (1) coefficient t-ratios p> (t) [95% confidence interval] constant 12.566 1.76 0.082 -0.001 0.011 x1 0.005 4.39 -0.305 0.115 lx2 0.210 3.04 0.003 0.048 0.231 lx3 0.140 3.80 -2.645 -0.826 lx4 1.735 2.16 0.034 -0.001 0.023 x6 0.012 2.49 0.015 -0.823 -0.092 lx7 0.460 2.69 0.009 0.008 0.055 x8 0.031 2.00 0.049 -0.082 0.0002 x10 0.041 5.10 7.663 17.472 the data pertaining to the variables x5 (gross domestic product) and x9 (percentage of gdp associated with the circular economy) have been excluded from the model due to their lack of statistical significance. in the first instance, it is posited that there is an issue of collinearity with other variables of an economic nature that are more closely related to water use. these include variables representing consumption, such as that indicated in the model by x2 (total water consumption). in the second case, with regard to the concept of the circular economy, it could be postulated that since it encompasses a multitude of factors that are not directly related to water, this variable may be regarded as an indication of sustainability. however, it is not significant. the exclusion of these variables allows for the creation of a fully specified panel data model with fixed effects. all of the variables were found to be significant individually, with their respective p-value falling below a threshold of 0.05. furthermore, the joint significance of these variables was also confirmed, with the prob > f value falling below 0.05. at the level of econometrics, the final model selected is a fixed-effects panel, as indicated by the results of the hausmann test. it does not incorporate temporal effects, and heteroscedasticity and first-order serial autocorrelation are present. however, these issues are alleviated to some extent when applying the feasible generalized least squares (fgls) method. in other cases, authors calculate these statistics with the aim of validating their models. for instance, im et al. (2003) and 150 amelia pérez zabaleta and monica borrat sanjuan blackburne and frank (2007), along with hadri (2000), have provided examples of this approach. the implementation of the distinct examinations has yielded the following results: the prevalence of the panel data model versus ols: random effects are relevant and it is preferable to use the estimate of random effects instead of the one grouped according to the result of the application of the breusch-pagan la-grange multiplier test (lm)). the prevalence of fixed effects over random effects according to the result of the hausmann test considering the different panel data models, with fixed effects and random effects. the non-existence of temporary fixed effects according to the result of the test-parm test our study is located in the so-called micro panels. that implies: the impossibility of carrying out the evaluation of transversal dependency neither by applying the pesaran cd test nor by breusch-pagan due to insufficient observations. the presence of heteroscedasticity, either between countries or over time. the possibility of applying non-robust models such as the panel of fixed effects according to the negative result after applying the robustness test. the presence of first order or serial autocorrelation, which is resolved more efficiently with feasible generalized least squares (fgls). 3.3 economic interpretation of the model signs it is important to note that the final model selected allows an analysis of elasticity, rather than the direct impact of independent variables on the dependent variable. therefore, variables must be ordered according to their contribution to the final model. in this context, the most significant variables are those that contribute the most to population density, public water consumed in industrial and service activities, total consumption, percentage of public water, percentage of environmental taxes, percentage of the population connected, and governance. the effects of agglomeration and consumption economies are particularly pronounced at the industrial and service levels, as well as at the aggregate level. all of these factors contribute to economic development, although not in accordance with the guidelines set by the wfd, which occupy a subordinate position. nevertheless, they are of significant importance within the model. four variables have been identified as having a positive impact on the efficient use of water. these are presented in order of importance based on the highest coefficient they present in the model. the variables that have the greatest positive impact on the efficient use of water are research and development expenditure, the percentage of public water, the percentage of the population connected to the water supply, and governance. these variables are presented in the wfd guidelines, based on the objective of modernising facilities, implementing public control of the resource, guaranteeing access to the population and encouraging public interest in influencing eu policies. there are also four variables that have a negative impact on water productivity, which are, in order of importance: population density, public water supplied to industrial and service activities, total consumption and the percentage of environmental taxes. it shows that as the use of the resource increases, its efficiency decreases, but it also proves that a purely punitive policy does not solve the situation. the role of the water framework directive in enhancing water use efficiency in the eu the presence of greater population density is associated with lower levels of economic productivity. this is a more nuanced phenomenon to analyse, as while greater population density may facilitate a more concentrated water supply, it may also lead to greater water contamination. the so-called agglomeration diseconomies, whereby the benefits of increased density are outstripped by the negative effects, must be considered. a greater total volume of public water supplied to industrial activities and services is associated with lower productivity. this implies that, for water productivity to be high, it is not necessary for water management to be high; private management is more productive. this outcome would be the inverse of the previous one, yet it could yield a parallel conclusion. in developed countries where water consumption is higher due to higher rates of growth, increased industrial activity, and a greater number of cities, the engines of the economy itself are set in motion, leading to a more efficient use of the resource without the need for further public intervention. the consumption of greater quantities of water is associated with a reduction in productivity. in more affluent nations, there is greater utilisation of water for the fulfilment of all productive requirements, thus maintaining a stable economic status. in addition to its use in the agricultural sector, water is also employed at an industrial level. furthermore, the creation of urban settlements in this type of economic system also results in greater water consumption, making the efficient utilisation of the resource challenging. it can be reasonably assumed that increased investment in research and development (r&d) will lead to an enhancement of overall productivity levels. this indicates the necessity for the continued enhancement of investments in technologies that reduce the consumption of water. a reduction in the level of environmental taxation is conducive to an increase in overall productivity. these findings demonstrate that an ambitious taxation strategy is not as effective as a targeted subsidy approach. a greater proportion of the population connected to public services will result in increased productivity. this conclusion aligns with one of the objectives of the wfd, which aims to ensure the supply to the population. the proportion of public water abstracted is positively correlated with productivity levels. in general, public water management has precedence over private water management. this is a common occurrence in economies that are not particularly sustainability-oriented. the implementation of more robust governance mechanisms is conducive to enhanced water productivity. this outcome is in close alignment with the objectives of the wfd, which demonstrates the effectiveness of the oecd water governance programme in improving water resource efficiency. this is achieved by employing appropriate policy interventions to manage this vital natural resource. the gdp, when subjected to a correlation analysis, has been observed to exhibit collinearity with variables of an economic nature that have been demonstrated to exert a more pronounced influence. among these, variables such as consumption have emerged as particularly noteworthy due to their direct bearing on water utilisation. the circular economy is a concept that encompasses a multitude of factors not directly related to water. while it could be interpreted as an indication of sustainability if this variable were significant, it has been discarded due to its lack of significance compared to the other variables. 152 amelia pérez zabaleta and monica borrat sanjuan 4. discussion as discussed in the previous section, all the variables in the final model are individually and jointly significant in the fixed effects model under consideration. a parsed analysis of the variables seeking for the individual effect, which has strictly economic variables and the variables associated with the application of public policies predicted by the “wfd”, has shown that water productivity is better explained by purely economic variables than by variables related to the “wfd”. according to the model, r&d expenditure, the percentage of public water, the percentage of population connected and the governance index are relevant and have a positive impact on water use efficiency, highlighting the need to modernise facilities, implement public control of the resource, guarantee access to the population and encourage citizens' interest to influence eu policies. in turn, population density, public water supplied to industrial and service activities, total consumption and the share of environmental taxes have a negative impact on water productivity, showing that as water use increases, water efficiency decreases, but also that a purely punitive policy does not solve the situation. some of the ue-28 more developed countries, such as denmark, finland, and the united kingdom, have higher levels of water use efficiency, i.e., higher water productivity values. denmark, the 40th largest economy by gdp, is a modern economy with a high-tech agricultural sector, worldleading companies in pharmaceuticals, shipping, and renewable energy, and a high dependence on foreign trade. finland is the 48th largest economy by gdp, and has a highly industrialised economy, based on large forest resources and high levels of capital investment. it now has one of the most technologically advanced economies in the world, although it was basically agrarian 50 years ago. the uk economy is the sixth largest in the world, in terms of market exchange rates, and the second largest in europe. its economy is considered to be highly developed and service sector-oriented. the effect of agglomeration economies and consumption at the total level and in industrial and service activities, all of which are variables that participate in the development of the economy, promote the highest water productivity indexes. netherlands, france, ireland and germany are other examples of highest water productivity values. population density affects the water productivity index in particularly small countries such as belgium, malta, cyprus, and luxembourg, providing better water productivity indexes. governance has come last. it is a fairly modern concept that should be promoted. the notion of public policies "in the form of governance" calls for a state that is present and a society that participates, fostering public governors and administrators in dialogue with private leaders (business and civil) and with citizens who do not belong to any sector. the 2010-2014 situation of budgetary austerity should not be a brake on the development of actions within the framework of public policy evaluation, since this should be understood as an investment and not as a cost. therefore, in 2016, the euro area economy was still on deflationary dangers, the insight of both the public and policymakers as to the necessity of a macroeconomic policy change increased. the calls for a more expansionary fiscal stance, above all for a boost to public – or publically supported – investment, become louder, with the investment for europe plan (juncker-plan) as the most prominent official policy reaction and previous initiatives – as the introduction of the so-called ‘investment clause’ the role of the water framework directive in enhancing water use efficiency in the eu under the stability and growth pact (sgp) – to support and protect public investment. however, those initiatives failed, and public investment in the euro area has decreased substantially since the onset of the crisis. in the periphery countries, public investment expenditures shrunk dramatically as a result of the austerity policies imposed on those member states. and despite all efforts, growth forecasts for the euro area was stagnating since summer 2014. the covid-19 pandemic led to a sharp economic collapse in the eurozone in 2020 and forced the authorities in these countries to employ keynesian methods to sustain the economy. as a result of the severe recession and the temporary emergency measures taken, public finances in the member states suffered significantly. the mutually reinforcing effects of the eurozone’s fiscal policy and monetary policy were crucial for mitigating the effects of the covid-19 crisis in its countries and supporting the economic recovery thereof in 2021– 2022. the rapid development of the crisis has revealed the difficulties in the application of fiscal policy assessment indicators. changes in the surveillance should lead to improved regulatory clarity and reduced regulatory complexity. there is a local effect associated, on the one hand, with the climate and, on the other hand, with the origin of the water. it is important to bear in mind that water productivity also depends on water quality and that water is a common element that is transferred from one country to another through the different river basins. northern european countries have a greater water catchment capacity due to their climatic conditions and less water use in agriculture because they are more developed economies that are more focused on the secondary and tertiary sectors than on the primary sector. water stress in these areas is much lower than in southern european countries, where the opposite is true. countries where specific policies associated with the wfd have been applied more intensively, such as the czech republic, hungary, bulgaria, poland, romania, slovakia, and lithuania, (the majority of them belonging to the danube and rhine river basin) are in the last positions but are in any case significant within the model. it is also no coincidence that these countries are geographically located mostly in south-eastern europe. again, there are both climatological and sustainable awareness reasons for this. although it has not been possible to observe a temporal evolution, it is important to note that the initial information was derived from a micro-panel comprising data from 10 years of time. in order to perform the requisite analysis on this variable, it is essential that the calculation program in question has access to a minimum dataset of at least 25 years. 5. conclusions water productivity in the eu-28 depends mainly on purely macroeconomic variables such as population density, water consumption, environmental taxes, and government efficiency. some authors have previously demonstrated these direct dependencies between water productivity and specific variables in various areas of water use. these include agriculture (martínez and goetz, 2007), industry (cruz et al., 2003; vargas ovando, 2015), services (del villar, 2010), and urban consumption (kim et al., 1998, sampson et al., 2022). however, this study presents a comprehensive analysis of these variables, incorporating harmonised data for the eu-28. in its dummy application, 154 amelia pérez zabaleta and monica borrat sanjuan the distribution by country allows the econometric model to be correlated geographically. the country analysis has highlighted the differences between rich and poor economies and their location between northern and southern europe. global economic drivers, as well as climatic conditions and the quality of the water to be used, could explain these indices. in their respective works, marin (2009) and pérard (2009) examine the distinctions between public and privately managed water resources. at the european level, lynk (1993) focuses on the united kingdom, wackerbauer (2009) on germany, and benito et al. (2015) on spain. the studies in question are local in scope and focus on a single variable, whereas our model is based on the analysis of a global indicator, namely water productivity. the model has demonstrated that water productivity is contingent upon variables that can be linked to the implementation of the wfd. these include total public water supply, environmental taxes, population connected to public services, and public water to total abstraction. a relationship can be established between the application of a public policy, such as the wfd, and an objective indicator, such as water productivity. it is challenging to translate the efficacy of public economic policies into an economic indicator, given that their effects frequently depend on numerous correlated variables. the effectiveness of the implementation of the wfd can be assessed both geographically and in aggregate using the objective indicator of water productivity in the eu-28. the application of the wfd has had an impact on the efficient use of water, and this is an indication that the application of this public policy model to this natural resource has been effective. however, the economic drivers themselves have had a greater impact on the efficient use of water than the application of the public policy itself the implementation of the wfd has been more strongly promoted where it was most needed. the implementation of the guidelines set by the wfd has not yet led to a direct improvement in the water productivity indices. some authors, such as sánchez garcía and blanco jiménez (2012) and munguía-lópez et al. (2019), correlate price with water consumption. this would have been a macroeconomic variable to include in the model if eurostat had harmonised data on water prices in europe. however, the actors involved in the water cycle apply water prices locally and in a relatively arbitrary manner. this is one of the reasons why it is difficult to have public policies in place that have an impact on the overall results. one of the main lines of action of the wfd is based on the water cost recovery policy. however, it has not been possible to correlate or link it to any of the variables of the established econometric model. the new public policies in the water sector should have objective indicators, based on harmonised macro-economic indices in the eu-28, which would make it possible to determine their effectiveness. water quality as a variable was not included in the model. it was not possible to determine the effect of river basin management in the different countries. given that recovery costs include a large part of wastewater treatment costs, this variable is very important. it is important to know not only where water use is most efficient but also where water use efficiency is most needed. it is necessary to determine where it is appropriate to focus public water policy efforts so that more efficient water use takes place where water stress is greatest. it is quite possible that in the future, we will be able to carry out a new analysis of the application of the wfd, but at the moment, there is not enough information to attribute some of the the role of the water framework directive in enhancing water use efficiency in the eu measures taken to productivity. likewise, it is necessary to broaden the study of water from a perspective that goes beyond the efficient use of the resource and considers the impact on the environment. references aguilera klink, f. (2000). value, use and price of water: the protection of water resources and the role of economic analysis in the 2000 directive. in the framework directive perspectives in portugal and spain: ii iberian congress on water planning and management. zaragoza: new water culture foundation & institución fernando el católico. aguilera klink, f. (2006). towards a new water economy: fundamental issues. polis. revista latinoamericana, (14). anwandter, l., & ozuna, t. jr. (2002). can public sector reforms improve the efficiency of public water utilities? environment and development economics, 7(4), 687-700. arreguín-cortés, f. i., & alcocer-yamanaka, v. h. (2004). systemic modelling of the efficient use of water. ingeniería hidráulica en méxico, 19, 83-102. arrojo, p. (1999). la nueva cultura del agua ante el reto de un nuevo plan hidrológico nacional. economistas. colegio de madrid, 80, 441-450. arrojo, p. (2005). the ethical challenge of the new water culture: functions, values, and rights at stake. barcelona: paidós. arrojo, p. (2010). crisis global del agua. cuadernos 168. barcelona: cristianisme i justicia. arrojo, p., & martínez gil, f. j. (1999). debating water at the university: towards a new culture of water. fernando el católico institution. avellà, l. l., carles, j., & sanchis, c. (2009). the water framework directive: an alternative model of water use? the economics of irrigation water in spain. almería: fundación cajamar. barbier, e. (2004). water and economic growth. economic record, 80, 116. https://doi.org/10.1111/j.1475-4932.2004.00121.x barbier, j. m. (2000). l'analyse des pratiques: questions conceptuelles. in blanchard-laville, c., & fablet, d. (eds.) l'analyse des pratiques professionnelles (pp. 35-58). paris: l’harmattan. barraqué, b. (1995). les politiques de l'eau en europe. paris: editions la découverte. barraqué, b. (1998). europäisches antwort auf john briscoes bewertung der deutschen wasserwirtschaft. gwf wasser-abwasser, 139(6). barraqué, b. (2000a). assessing the efficiency of economic instruments: reforming the french agences de l'eau. in m. s. andersen & r. u. sprenger (eds.), market-based instruments for environmental management. cheltenham: edward elgar. https://doi.org/10.4337/9781782543831.00023 barraqué, b. (2000b). are hydrodinosaurs sustainable? the case of the rhone-to-barcelona water transfer. in e. vlachos & f. n. correia (eds.), shared water systems and transboundary issues, with special emphasis on the iberian peninsula. luso-american foundation and colorado state university. beck, n. (2001). time-series-cross-section data: what have we learned in the past few years? annual review of political science, 4, 271-293. http://dx.doi.org/10.1146/annurev.polisci.4.1.271 beecher, j. a., flowers, j. e., & matzke, c. s. (1998). water conservation guidelines and the dwsrf. journal of the awwa, may 1998, 60-67. https://doi.org/10.1002/j.1551-8833.1998.tb08434.x benito, b., guillamón, m. d., & albaladejo, f. (2015). public versus private in municipal services management. lex localis. http://dx.doi.org/10.4335/13.3.995-1018(2015) blackburne, e. iii, & frank, m. (2007). estimation of nonstationary heterogeneous panels. stata journal. https://doi.org/10.1177/1536867x0700700204 breitung, j., & das, s. (2005). panel unit root tests under cross-sectional dependence. statistica neerlandica, 59, xx-xx. http://dx.doi.org/10.1111/j.1467-9574.2005.00299.x bueno hernández, f. (2007). the water framework directive. ingeniería y territorio. revista del colegio de ingenieros de caminos, canales y puertos, 80, 4-11. cabrera-marcet, e., cobacho jordán, r., almandoz berrondo, j., cabrera rochera, e., & arregui de la cruz, f. (2002). water management in the countries of the european union: the paradigm between north and south. in the water framework directive realities and futures (pp. 39-47). sevilla: universidad de sevilla. carles, j. (2001). public administration in the face of the new water policies of the framework directive. in fundación nueva cultura del agua (ed.), the framework directive perspectives in portugal and spain: ii iberian congress on water planning and management. institución fernando el católico. carles, j., & garcía, m. (2001). european water policy: a new framework for sustainable management. in https://doi.org/10.1111/j.1475-4932.2004.00121.x https://doi.org/10.4337/9781782543831.00023 http://dx.doi.org/10.1146/annurev.polisci.4.1.271 https://doi.org/10.1002/j.1551-8833.1998.tb08434.x http://dx.doi.org/10.4335/13.3.995-1018(2015) https://doi.org/10.1177/1536867x0700700204 http://dx.doi.org/10.1111/j.1467-9574.2005.00299.x 156 amelia pérez zabaleta and monica borrat sanjuan a. basanta (ed.), sustainable use of water: from discussion to decision. emasesa. chang, h., bonnette, m. r., stoker, p., crow-miller, b., & wentz, e. (2017). determinants of single-family residential water use across scales in four western us cities. science of the total environment, 596, 451-464. https://doi.org/10.1016/j.scitotenv.2017.03.164 cobacho, r. (2000). demand management in the context of a new comprehensive water policy (doctoral thesis). polytechnic university of valencia. colmenárez, a., & salazar, e. (2018). modelo para la estimación de la producción de agua de uso doméstico. caso municipio palavecino del estado lara. compendium, 19(36), 533. https://revistas.uclave.org/index.php/compendium/article/view/56 cosgrove, w. j., & loucks, d. p. (2015). water management: current and future challenges and research directions. water resources research, 51(6), 48234839. https://doi.org/10.1002/2014wr016869 cruz, m., uribe, e., & coronado, h. (2003). the value of the marginal productivity of water in the colombian manufacturing industry. bogotá: universidad de los andes. del villar, a. (2010). the prices of water services: a prospective analysis of demand on domestic uses. applied economics studies, 28(2), 333-355. di vaio, a., trujillo, l., d'amore, g., & palladino, r. (2021). water governance models for meeting sustainable development goals: a structured literature review. utilities policy, 72, 101255. https://doi.org/10.1016/j.jup.2021.101255 ec, european commission. (2001, may 2). common strategy on the implementation of the water framework directive. brussels: european commission. eea, european environment agency. (2001). sustainable water use in europe. demand management. environmental issue report 19. eea, copenhagen. eea, european environment agency. (2017). service contract no 3415/b2015/eea.56130 for the european environment agency 2017: pricing and non-pricing measures for managing water demand in europe. embid irujo, a. (2007). the water framework directive and some of the problems of its implementation process in spain and other european countries. ingeniería y territorio. revista del colegio de ingenieros de caminos, canales y puertos, 80, 20-27. ep, european parliament. (2000, december 22). establishing a framework for community action in the field of water policy. official journal of the european communities. españa circular (2030). estrategia española de economía circular. madrid: ministerio para la transición ecológica y el reto demográfico. retrieved from https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacionambiental/temas/economia-circular/espanacircular2030_def1_tcm30-509532_mod_tcm30509532.pdf fao food and agriculture organization of the united nations. (2018). information system on water use in agriculture (aquastat) 51, 9. determinants of international residential water pricing. retrieved from www.fao.org/nr/water/aquastat/main/indexesp.stm ferro, g., & lentini, e. (2015). energy efficiency and economic regulation in drinking water and sewage services. natural resources and infrastructure series. united nations cepal. frérot, a. (2014). circular economy and efficiency in the use of resources: an engine of economic growth for europe. cuestiones sobre europa, 331. parís: fundación robert schumann. ghavidelfar, s., shamseldin, a. y., & melville, b. w. (2018). evaluating the determinants of high-rise apartment water demand through integration of water consumption, land use and demographic data. water policy, 20(5), 966-981. greene, w. h. (2000). econometric analysis (4th ed.). upper saddle river, nj: prentice hall. gutiérrez-martín, c., borrego-marín, m., & berbel, j. (2017). the economic analysis of water use in the water framework directive based on the system of environmental-economic accounting for water: a case study of the guadalquivir river basin. water, 9(3), 180. https://doi.org/10.3390/w9030180 hadri, k. (2000). testing for stationarity in heterogeneous panel data. the econometrics journal, 3(2), 148-161. https://doi.org/10.1111/1368-423x.00043 hm government; defra department for environment food and rural affairs. (2008). future water. the government’s water strategy for england. ibarra, m. (2014). water productivity in corn grown under irrigation systems and dry land in san luis, argentina [final degree project, university of córdoba]. im, k. s., pesaran, m. h., & shin, y. (2003). testing for unit roots in heterogeneous panels. journal of econometrics, 115(1), 53-74. https://doi.org/10.1016/s0304-4076(03)00092-7 ingold, k., & tosun, j. (2020). special issue “public policy analysis of integrated water resource management.” water, 12(9), 2321. https://doi.org/10.3390/w12092321 kaufmann, d., kraay, a., & mastruzzi, m. (2010). the worldwide governance indicators: methodology and https://doi.org/10.1016/j.scitotenv.2017.03.164 https://revistas.uclave.org/index.php/compendium/article/view/56 https://doi.org/10.1002/2014wr016869 https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/economia-circular/espanacircular2030_def1_tcm30-509532_mod_tcm30-509532.pdf https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/economia-circular/espanacircular2030_def1_tcm30-509532_mod_tcm30-509532.pdf https://www.miteco.gob.es/content/dam/miteco/es/calidad-y-evaluacion-ambiental/temas/economia-circular/espanacircular2030_def1_tcm30-509532_mod_tcm30-509532.pdf http://www.fao.org/nr/water/aquastat/main/indexesp.stm https://doi.org/10.1016/s0304-4076(03)00092-7 https://doi.org/10.3390/w12092321 the role of the water framework directive in enhancing water use efficiency in the eu analytical issues. world bank working paper series, no. 5430 kim, e., & lee, h. (1998). spatial integration of urban water services and economies of scale. review of urban & regional development studies, 10(1), 3-18. https://doi.org/10.1111/j.1467940x.1998.tb00084.x lynk, e. l. (1993). privatisation, joint production and the comparative efficiencies of private and public ownership: the uk water industry case. fiscal studies, 14(2), 98-116. marín, p. (2009). public-private partnerships for urban water utilities. a review of experiences in developing countries. washington dc: the international bank for reconstruction and development / the world bank. martínez, y., & goetz, r. (2007). efficiency gains versus transaction costs of the water markets. revista de economía aplicada, xv(43), 49-70. martín-ortega, j., gutiérrez-martín, c., & berbel, j. (2008). characterization of water uses in the guadalquivir demarcation in application of the water framework directive. estudios regionales journal, 81, 45-76. miracle, m. r. (2006). consideraciones y caos en torno al ciclo del agua. polis revista latinoamericana 14. http://journals.openedition.org/polis/5105 moro, m., mcknight, u., smets, b., min, y., & andersen, m. (2018). the industrial dynamics of water innovation: a comparison between china and europe. international journal of innovation studies, 2, 14-32. https://doi.org/10.1016/j.ijis.2018.03.001 mugambi, p., blanco, m., ogachi, d., ferasso, m., & bares, l. (2021). analysis of the regional efficiency of european funds in spain from the perspective of renewable energy production: the regional dimension. international journal of environmental research and public health, 18(9), 4553. https://doi.org/10.3390/ijerph18094553 munguía-lópez, a., gonzález-bravo, r., & ponce-ortega, j. (2019). evaluation of carbon and water policies in the optimization of water distribution networks involving power-desalination plants. applied energy, 236, 927-936. 10.1016/j.apenergy.2018.12.053 olmeda pascual, j. a. (2006). water and analysis from the economic view: an application for economic growth [paper presented at viii reunión de economía mundial, alicante, spain]. universidad de castilla-la mancha. olmos salvador, l. (2018). determinants of international residential water pricing. aquae papers, 8. orloci, j., szesztay, k., & várkonyi, l. (1985). national infrastructures in the field of water resources: principles and alternatives in establishing infrastructures for water management (vol. 99). institute for water management. pérard, v. (2009). water supply: public or private? an approach based on cost of funds, transaction costs, efficiency and political costs. policy and society, 27(3), 193-219. https://doi.org/10.1016/j.polsoc.2008.10.004 revenga, c., brunner, j., henninger, n., kassem, k., & payne, r. (2000). pilot analysis of global ecosystems: freshwater systems. washington dc: world resources institute. rouillard, j. (2020). tracing the impact of agricultural policies on irrigation water demand and groundwater extraction in france. in m. giupponi & j. bjornlund (eds.), sustainable groundwater management (pp. 461-479). cham: springer. https://doi.org/10.1007/978-3-030-32766-8_24 saleth, r. m. (2002). water resources and economic development. cheltenham glos, uk: edward elgar publishing. sampson, d. a., quay, r., & horrie, m. (2022). building type, housing density, and water use: denver water data and agent-based simulations. journal of the american water resources association, 58(2), 355-369. https://doi.org/10.1111/1752-1688.13009 sánchez garcía, v. e., & blanco jiménez, f. j. (2012). l uso sostenible del agua en núcleos urbanos: lastarifas como herramienta de control del consumo. observatorio medioambiental, 15, 35-59. schreiber, c. e. (2016). water use patterns: examining the impact of population density on municipal drought response (doctoral dissertation). solanes, m. (1996). water markets: institutional rights. revista cepal, 59. solanes, m. (1998). integrated management of water resources, from the perspective of the dublin principles. revista cepal, 64. solanes, m., & gonzález-villarreal, f. (1996). the dublin principles reflected in a comparative assessment of institutional and legal arrangements for integrated water management. global water partnership. tack background papers, nº 3. torres-reyna, o. (2007). panel data analysis fixed and random effects using stata (v. 4.2). princeton, nj: princeton university. vargas ovando, l. j. (2015). economic value of water in the chilean manufacturing industry [master's thesis, universidad de concepción]. postgraduate management faculty of economics and administrative sciences master's program in economics of natural resources and the environment. https://doi.org/10.1111/j.1467-940x.1998.tb00084.x https://doi.org/10.1111/j.1467-940x.1998.tb00084.x http://journals.openedition.org/polis/5105 https://doi.org/10.1016/j.ijis.2018.03.001 https://doi.org/10.3390/ijerph18094553 https://www.google.com/url?sa=e&source=gmail&q=https://doi.org/10.1016/j.apenergy.2018.12.053 https://www.google.com/url?sa=e&source=gmail&q=https://doi.org/10.1007/978-3-030-32766-8_24 https://doi.org/10.1111/1752-1688.13009 158 amelia pérez zabaleta and monica borrat sanjuan viaggi, d., zanni, g., & raggi, m. (2014). changing perspectives on the economics of water. water, 6(10), 2969–2977. http://dx.doi.org/10.3390/w6102969 wackerbauer, j. (2009). the water sector in germany. université de liège: ciriec working papers, 911. wef, world economic forum. (2002). environmental sustainability index. retrieved from https://sedac.ciesin.columbia.edu/es/esi/esi2002_21mar02tot.pdf wichelns, d. (2014). do estimates of water productivity enhance understanding of farm-level water management? water, 6(4), 778–795. http://dx.doi.org/10.3390/w6040778 wooldridge, j. m. (2002). econometric analysis of cross section and panel data. cambridge, ma: mit press. world meteorological organization. (1997). comprehensive assessment of freshwater resources of the world. stockholm environment institute. retrieved from https://library.wmo.int/idurl/4/51224 xiaoming liu, asad iqbal, ji dai, & guanghao chen (2021). economic and environmental sustainability of the optimal water resources application for coastal and inland regions. journal of cleaner production, 296, 126247. https://doi.org/10.1016/j.jclepro.2021.126247 zetland, d., & gasson, c. (2012). a global survey of urban water tariffs—are they sustainable, efficient and fair? international journal of water resources development, 29, 1-16. http://dx.doi.org/10.3390/w6102969 https://sedac.ciesin.columbia.edu/es/esi/esi2002_21mar02tot.pdf http://dx.doi.org/10.3390/w6040778 https://library.wmo.int/idurl/4/51224 https://doi.org/10.1016/j.jclepro.2021.126247 the role of the water framework directive in enhancing water use efficiency in the eu appendix 1 the following figures present the data of the different study variables on which the model is based. these ad hoc data were provided by the eu-28 countries for the period from 2006 to 2015. figure 1. water productivity. note: this variable represents the quantity of economic output produced for a specific quantity of water collected. the variable measures the efficiency of water use. the total quantity of water extracted encompasses all sources of water withdrawal, encompassing both fresh water and water that has been reused, either temporarily or indefinitely. the data also includes water used in mining and water that has been extracted from the ground, as well as rainwater harvesting. notwithstanding the aforementioned considerations, the quantity of water employed by hydroelectric power stations for the generation of electricity is to be excluded. 200.00 400.00 600.00 800.00 1,000.00 1,200.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m water productivity (€/m3) 2006 2007 2008 2009 2010 2011 2012 2013 160 amelia pérez zabaleta and monica borrat sanjuan figure 2. government effectiveness. note: the concept of government effectiveness is defined by the perception of the quality of public services, the quality of civil services and the degree of independence from political pressures, the quality of policy formulation and implementation, and the credibility of politicians' actions, in terms of their alignment with the implementation of their respective policies. a percentile is distributed across all countries worldwide, with a range from 0, representing the lowest level, to 100, representing the highest level. figure 3. total water consumption. note: the indicator encompasses the total consumption of water utilized across a multitude of sectors. these include agriculture, industry, mining, service, and domestic use. the indicator quantifies the total volume of water consumed, expressed as millions of cubic metres of surface water and groundwater, in all of these sectors. 10.00 20.00 30.00 40.00 50.00 60.00 70.00 80.00 90.00 100.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m government efectiveness (%) 2006 2007 2008 2009 2010 2011 2012 2013 5,000.00 10,000.00 15,000.00 20,000.00 25,000.00 30,000.00 35,000.00 40,000.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m total water consumption (millions of m3) 2006 2007 2008 2009 2010 2011 2012 2013 the role of the water framework directive in enhancing water use efficiency in the eu figure 4. r&d expenditure. note: eurostat collates data related to expenditure on research and development, expressed in millions of euros at 2005 prices. at the level of research and development, the segment of companies related to water comprises technology and component companies that provide services and products to infrastructures. figure 5. population density. note: the population density of a given country has been calculated by dividing the number of inhabitants by the country's area in square kilometres, as provided by the eurostat database. conversely, this constrains the length of the network for a given population, which may be linked to the costs associated with its construction and maintenance. consequently, a positive correlation between productivity and population density would be anticipated. 10,000.00 20,000.00 30,000.00 40,000.00 50,000.00 60,000.00 70,000.00 80,000.00 90,000.00 100,000.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m r&d expenditure (millions of €) 2006 2007 2008 2009 2010 2011 2012 2013 200.00 400.00 600.00 800.00 1,000.00 1,200.00 1,400.00 1,600.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m population density (people/m2) 2006 2007 2008 2009 2010 2011 2012 2013 162 amelia pérez zabaleta and monica borrat sanjuan figure 6. gross domestic product. figure 7. % public water. note: this variable indicates the proportionate importance of public control activity in the context of various services related to the integrated water cycle. it differentiates between countries where the management of water is carried out by public entities, those where it is subcontracted by the administration to other entities of a private nature and how this management affects the water used in the economic activities related to industrial activities and services. furthermore, this differentiation allows us to ascertain the impact this difference in management has on water productivity. 500,000.00 1,000,000.00 1,500,000.00 2,000,000.00 2,500,000.00 3,000,000.00 3,500,000.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m gross domestic product (millions of € at 2015 current prices) 2006 2007 2008 2009 2010 2011 2012 2013 10.00 20.00 30.00 40.00 50.00 60.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m public water as a percentage of total water (%) 2006 2007 2008 2009 2010 2011 2012 2013 the role of the water framework directive in enhancing water use efficiency in the eu figure 8. public water to industrial activities & services. note: eurostat employs this indicator to quantify the water resources managed and supplied by public entities, encompassing all industrial and domestic activities. figure 9. % population connected to wwt. note: the eurostat indicator is employed to ascertain the degree to which the population has access to drinking water, taking into consideration the extent of access to the public service network. the data presented represents the percentage of the resident population that is connected to the public water supply network. 1,000.00 2,000.00 3,000.00 4,000.00 5,000.00 6,000.00 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d… total public water supplied to industrial activities and services (millions of m3) 2006 2007 2008 2009 2010 2011 2012 2013 0 10 20 30 40 50 60 70 80 90 100 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m percentage of population connected to public service (%) 2006 2007 2008 2009 2010 2011 2012 2013 164 amelia pérez zabaleta and monica borrat sanjuan figure 10. % gdp to circular economy. note: this variable represents the percentage of gdp associated with the circular economy, encompassing both private investment and jobs, as well as the gdp generated within sectors directly or indirectly linked to the circular economy. the indicator encompasses substantial financial investments in tangible products, the size of the employed workforce and the value added to assets in sectors pertaining to recycling, reuse, rental and leasing. the nace classification of sectors has been used to define these sectors as being related to the circular economy, with investments in tangible goods calculated to have been made for more than one year. the number of employees has also been considered, expressed as a percentage of the total employees in each country, in order to provide a more detailed picture of the sectors involved. figure 11. % environmental taxes graphic. note: this indicator measures both defined environmental taxes and rates associated with environmental fines, with the latter being primarily related to energy, transportation and atmospheric environmental issues. additionally, it encompasses taxes associated with workers as social contributions for environmental conservation. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1: a us tr ia 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m percentaje of gpd associated to circular economy (%) 2007 2008 2009 2010 2011 2012 2013 2014 2.00 4.00 6.00 8.00 10.00 12.00 14.00 2: b el gi um 3: b ul ga ria 4: c ro at ia 5: c yp ru s 6: c ze ch ia 7: d en m ar k 8: e st on ia 9: f in la nd 10 : f ra nc e 11 : g er m an y 12 : g re ec e 13 : h un ga ry 14 : i re la nd 15 : i ta ly 16 : l at vi a 17 : l ith ua ni a 18 : l ux em bo ur g 19 : m al ta 20 : n et he rla nd s 21 : p ol an d 22 : p or tu ga l 23 : r om an ia 24 : s lo va ki a 25 : s lo ve ni a 26 : s pa in 27 : s w ed en 28 : u ni te d ki ng do m percentage of environmental taxes (%) 2006 2007 2008 2009 2010 2011 2012 2013 1. introduction 2. materials and methods 3. results 4. discussion 5. conclusions references european journal of government and economics 12(1), june 2023, 39-57 this work is licensed under a creative commons attribution-noncommercial 4.0 international license. european journal of government and economics issn: 2254-7088 assessing the fairness of the eu council qualified majority voting. a voting power critical perspective of the liberal intergovernmentalist accounts oana andreea ion a * a affiliation: national university of political studies and public administration (snspa), romania. * corresponding author at: oana.andreea.ion@dri.snspa.ro abstract. the qualified majority voting (qmv) used by the council of the european union developed to a high degree of complexity from one modifying treaty to another, until the latest definition stipulated in the treaty of lisbon. this paper analyses this eu intra-institutional voting method using a rational choice approach and emphasizes that there are situations when not even the institutions, as rational actors, can avoid a collective irrational outcome even when they are addressing subjects such as voting power distribution. it also addresses several shortcomings of the liberal intergovernmentalist explanatory framework focusing on the insufficiently developed level of credible institutional commitments. the core part of the article consists in investigating several types of eu council internal decision-making options, proposing how they can be designed to be considered in the same time fair and efficient, and in analysing how close this voting power ideal type configuration is to the current decisional system. keywords. eu council; input and output legitimacy; qualified majority voting; rational choice; voting power. jel codes. d7, d8, o2 doi. https://doi.org/10.17979/ejge.2023.12.1.9335 1. introduction 1.1 input and output legitimacy in the eu council: a voting power analysis various crises experienced by the european union (refugees, brexit, reaction to the ukrainian conflict, etc.) have invigorated those criticisms of the european union (eu) that underline the difficulty of creating or operating with an inter-institutional or intra-institutional decision-making system that is both representative and efficient, but mainly fair for the multiple actors involved. the importance of the eu on the international stage and the special place that the eu council occupies in the institutional architecture of the union are two arguments to justify the need for this study which offers a political science perspective on a subject usually approached only from the political practitioners’ perspective. establishing as the unit of analysis the rational agent defined here as any member state of the european union, the focus of this paper is therefore on the https://creativecommons.org/licenses/by-nc/4.0/ mailto:oana.andreea.ion@dri.snspa.ro https://doi.org/10.17979/ejge.2023.12.1.9335 oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 40 decision-making system of the european union council, more precisely, the qualified majority voting (qmv), in order to describe this system and to prove that there are situations when not even the institutions can avoid a collective irrational outcome. the specific case of providing a new set of institutions for promoting coordination of activities by finding a commonly agreed set of rules (ostrom, 1990) reflects the very wide spectrum of voting rules. scholars have tried to offer a systematic approach to this field, classifying them according to different criteria broadly connected to the inputs, the processes and the outputs of these voting systems (shepsle, 2010). one voting method can observe only a part of the criteria stated for all the categories and if several actors have different opinions on this subject, reaching a consensus becomes a tough job, as revealed by the european union’s experience in this field. the present article, recognizing the merits of liberal intergovernmentalism (li) against its classic state-centred intergovernmentalist predecessors, aims to address some of the very shortcomings of the li (moravcsik 1993; 1997; moravcsik and schimmelfennig, 2018) and strengthening the rational choice component of this theoretical approach, it focuses on the criticisms connected to its third explanatory level, that of credible institutional commitments, which i consider as insufficiently developed, especially in terms of daily politics decisions analysed through voting power lenses. the paper refers to the broader literature on eu’s decision-making process and it is focused on the eu council internal decisional framework, being conceived on two interconnected research questions: firstly, what kind of fair decision-making system can be designed to be considered in the same time representative and efficient (input and output legitimate, in other words) according to a rational choice theory based on a li logic? and secondly, how close is this ideal type to the current decisional system? one central concept is the fairness of the decision-making process in the council. following verba’s seminal contribution (2006), i also agree that fairness is an essential attribute of a democracy and that the important features of fairness in politics are always connected to ensure an “equal voices” base for the decisional process, therefore augmenting its input legitimacy, as well as – even if it is difficult to accomplish them simultaneously – equal outcomes (to be understood, in fact, as outcomes that are responsive for its members). indeed, the above understanding could be applied to various decision-making bodies, and in the case of the eu it can be applied either to the logic of inter-institutional decisions or to that of intra-institutional decisions. thus, although at the eu level, the political game of the major institutional actors is extremely important, i believe that, in order to understand the bigger picture, which represents a whole greater than the sum of its parts, its constitutive units, i.e. each individual institution, must be carefully analysed. and the case of the council is a special one, considering the central role it played in the history of the eu. therefore, as indicated below, within the fairness framework, i emphasize in this paper the equal distribution of power criterion and i have opted to start my investigation by considering penrose’s voting system (1946) as an example of the fair rule and applying it to the eu council’s case, by analysing the intricate relation between voting power and voting weights. oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 41 even if in the case of the european union it is very important the inter-institutional decisional dynamics, the purpose of this paper is to provide an in-depth picture of some aspects specifically related to the council's decision. it is important to understand the mechanisms leading to a specific common position of the council because that particular design may influence its negotiations with the european parliament when the ordinary legislative procedure, for example, is in place; moreover, the council is still the main voice in the other special legislative procedures and their more tied relation with the consensus practice, which is a logic that informally affects also the ordinary legislative procedure framework (anonymous 2017). in order to offer pertinent answers to the above questions, after a brief description of the eu council decision-making architecture, this study will be developed with a particular interest in the nice and lisbon configurations and implications, in the context of a voting power analysis. the findings of the paper indicate the distance between the ideal and current eu council decisional system and the conclusions add to the existing debates several arguments on the chances the ideal type has for being implemented in the near or medium future. 1.2 liberal intergovernmentalism and rational choice theory, or from individual to institutions li was elaborated by andrew moravcsik (1993; 1997) based on the model offered by robert putnam (1988), being a justification of the european construction process from the perspective of the simultaneous existence of a so-called demand and supply for this integration phenomenon. the representation of the demand highlights the preferences of the various rational actors present in the internal political process and whose interactions determine the appearance of the political agenda of the state, according to the neoliberal international theory of the formation of state preferences. the offer is instead interested in how these states behave on the international stage as unitary actors engaged on the basis of their respective agendas in a series of intergovernmental negotiation processes. therefore, is underlined the presence in the li of a strong influence of the rational choice theory, manifested both at the level of internal political groups and at the level of the state engaged as a unitary whole in negotiation processes (ion, 2013). in fact, moravcsik's intergovernmentalism can be considered not only a two-level (demand and offer) game, but a three-level one, given the process of institutional choice (pollack, 2020). however, contrary to the impact that liberal intergovernmentalism has had (and still has) on the study of the european union, the most important criticisms it has received have been aimed at an insufficient role accorded to the same institutions that are barely recognized by the classic intergovernmental discourse. the role of creating and ensuring effective institutional mechanisms would be, on the one hand, to guarantee compliance by states with the commitments made during the negotiation, and, on the other, to provide a satisfactory motivation for the continuity and development of the cooperation practice. however, unlike his followers, who connected rationalist or historical institutionalism with intergovernmentalism, moravcsik gave little interest to the oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 42 supranational level in the european construction process. in a 2018 contribution (together with schimmelfennig), he argued that in the absence of a complete harmonization of the preferences of the member states (ms), none of the post-maastricht treaties has fundamentally innovated the decisional process, not even the constitutional treaty or the lisbon treaty (moravcsik and schimmelfennig, 2018). in this respect, the paper challenges this perspective by focusing on the day to day politics. the basic assumptions in the before mentioned rational choice theory can be summarized as follows: one individual is said to act rational if, taking into account the fact that she possesses ordered preferences, complete information and a perfect internal device, she chooses the action that will satisfy her preferences in such a manner that her anticipate utility would be maximized (see suzumura, 2016). the discrepancies observed between individual rationality and collective rationality determined scholars to search for solutions of what became the classic issue of collective action. one of the attempts to surmount this obstacle is the rational choice approach to institutions, which can also be found in the before mentioned liberal intergovernmentalist assumptions. in this light, in order to correct collective irrationality that may occur from the simple aggregation of individual preferences, institutions are conceived as a sum of rules and incentives, positive and negative motivations planned to produce collectively rational outcomes (miroiu, 2007, pp. 164-172). the incentive to connect to an institution can be found at different levels of the social structure. peters advances the example of european countries who decided to form and later join eu in order to constrain the behaviour of the other member states interested in maximizing their own utility (see peters, 2019). these theoretical assumptions will be investigated in the context of the eu council qmv case. 2. materials and methods 2.1 the importance of vote weighting and vote power in the qmv in the eu council, the actual interest of a state is its real voting power in the general decisional process, and the qmv definition (constantly refined from rome eec treaty) is just an expression of the result of the negotiations of the states interested in maximizing their share. one’s voting weight is different from its voting power (leech and aziz, 2011; kirsch, 2011; nurmi, 2011; warntjen, 2017, p. 675), therefore – before lisbon – the weighting of votes in the council was actually connected to the voting power that each of the members of the council possessed in the decisions adopted by qm, involving each states level of influence in shaping the results of the decision: “voting weight refers to the relative voting strength accorded to any individual member state under a system of weighted votes, that is, the percentage of the total vote held by each member of the council. voting power refers to the ability of any individual member of the council to cast a decisive vote for adopting or blocking a decision” (galloway, 2001: 59). in a voting power analysis, one should remind that one of the democratic principles requires, oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 43 under the well-known “one person, one vote” slogan, formal equality of the a priori voting power of all citizens. the a priori voting power is defined as a constituent part of a posteriori (or actual) voting power, deduced by the voters only from the decision regulation and the type of issues submitted to vote. in other words, it is a conventional power foreseen by collective constitutive norms, and the analyses should concentrate on this very constitutional architecture of a specific decision-making structure. in this theory, founded by penrose in the ‘40s, a priori voting power is conceived as “a voter's degree of influence over the outcome under a specified decision rule of a division of a decision-making body: […] the a priori probability a that a decision-making body acting under a given decision rule will adopt a bill rather than blocking it” (felsenthal et al., 2003, p. 476). but what about those decision-making processes involving states instead of individuals? in the light of the rational choice assumed approach, should one apply the same voting power principles? “the voting power of each nation in a world assembly should be proportional to the square root of the number of people on each nation's voting list” wrote penrose in 1946 when analysing majority voting and representation in the un (penrose, 1946, p. 57). in fact, penrose’s square root law is considered the ideal type of democratic formula (see van der ploeg, 2008 for an exhaustive list of the scholars supporting penrose’s voting system as a fair rule). basically, when applying it to today’s eu, it means that the vote of any european citizen, irrespective of its member state provenience, should count equally in the general power game, on the basis of establishing each state’s input proportionally to the square root of its population (see also pukelsheim, 2011). that is why in the contemporary approaches of penrose’s ideas, the attention was usually held by the weighted voting systems where the constitutional design caused, by different weights and rules, a different degree of influence of the decision-makers, quantified in power indices, more specifically in a priori power indices. defined as a “statistical measure for summarizing certain properties of a given voting game” (felsenthal et al., 2003, p. 487), power indices are very useful instruments in conceiving weighted political systems or in determining the fairness of the existing procedures and offering solutions for a possible equalizing process of the voting power. their role is extremely important as the inequalities in the distribution of voting power are disturbing both agenda-setting and decision outcomes in favour of the actors with great voting power. in this respect, especially in cooperative games, “power indices, in general, represent the probability that each voter will determine the outcome of a particular voting game” (passarelli and barr, 2007, p. 43). two observations here: firstly, the influence of an agent is only measurable when there is no other agent able to reach by itself a majority (no matter what kind of majority) of votes; secondly, to underline that a winning coalition means that it has reached the threshold to approve a decision. in fact, the penrose method and the penrose square root law were not accepted immediately by the academic community. the principles became famous later, when they were rediscovered by banzhaf, and, after that, by coleman (see leech and aziz, 2011; machover, 2011; slomczynski and zyczkowski, 2011); in fact, the index that banzhaf developed is known either as the penrose-banzhaf index or simple as banzhaf index. nowadays, the most well-known and oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 44 used indices (see table 1) – despite some limitations such as blindness towards any intervening factors (kirsch, 2011) – are the before mentioned penrose-banzhaf (pb) index and the shapleyshubik (ss) index. besides them, van der ploeg (2008, pp. 13-14) also mentions other indexes such as deegan-packel index and holler-index. slomczynski and zyczkowski (n.d.) also introduce – connected to banzhaf index – coleman preventive power index and coleman initiative power index. table 1. various descriptions of the main used indexes penrose-banzhaf index shapley-shubik index “it calculates the number of swing possibilities of each country. a swing possibility is the number of times that a representative can have a pivotal position. this means that if the representative leaves the winning coalition side, the remaining coalition becomes a losing coalition. it considers each coalition equally likely”1. “it deals with so called sequential coalitions. (…) this index shows the fraction of the coalitions in which an actor is pivotal. a pivotal actor is the first voter whose vote makes a coalition a winning coalition”1. “the penrose measure of voter v's voting power (under a given decision rule) is the a priori probability of v being decisive; that is, the probability of the other voters being so divided that v is in a position to determine the outcome of the division”2. “the shapley-shubik index is a measure of the relative frequency with which a member country is in a pivotal position, i.e., in the position of ``swinging'' a losing coalition into a winning one, where all possible coalitions of a fixed number of member states are equally likely to occur. this measure is, in general, some function of the number of votes and the majority threshold”4. “the banzhaf index […] aims to calculate the power of individual players by finding the ratio between the coalitions an individual player can make to win and all winning coalitions”3. “the shapley-shubik index calculates the share of coalitions, which are winning due to the presence of party i in all coalitions, and prescribes a weight to a coalition in these calculations depending on its size”3. source: 1. van der ploeg, 2008, pp. 12-13; 2. felsenthal et al., 2003, p. 477; 3. aleskerov et al., 2002, p. 382; 4. barr and passarelli, 2009, p. 340. therefore, an analysis of how voting power considerations influenced the latest nice and lisbon treaty contexts in terms of drafting the final texts of the treaties and implementing their wordings will be further presented. 2.2 nice, qmv and voting power in light of the previous arguments, one better understands why, at the beginning of the 2000s, when preparing the union’s institutional architecture for the enlargement challenges, the older eu members feared that the newcomers would disturb to such an extent the existing distribution of power that the usual balances and the functioning of the decision-making procedures would have been weakening their particular national interests. the simple reapportionment of votes would have led the newcomers, small or medium-sized countries, to an over-weighted status, accumulating nearly 40% of the total power, a situation unfavourable to the ‘big players’. so, the question of how to adjust the existing distribution of power in such a manner that each actor would consider it has reached an optimal balance received a solution in the so-called nice triple majority criteria (simple majority of the member states, qualified majority of the weighted votes, demographic clause). many scholars, inspired by the voting power theory, were interested in learning how is affecting their new power distribution in the council the outcome of the decision-making. in this respect, using the shapley-shubik and banzhaf indices (see table 2), by determining the number of possible winning coalitions under the new nice stipulations, they proved that the possibility of a oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 45 decision to pass was almost zero and if the threshold of the qmv remained untouched, the council would be in fact blocked (felsenthal et al., 2003, p. 476). other areas of research were the equal voting weights of similar member states, deriving from the principle of a priori voting power, but also the correlation of this issue with the idea of equalizing the a priori voting power of the eu citizens, as a criterion of fairness (a purpose that would have led to an alternative decisionmaking design). the principle of fairness requires the before-described idea of having weighted votes’ distribution in accordance with a proportional correspondence between one country’s power index and its squared root of the population, in order to equalize the eu citizens voting power. analyses of the nice and, later, post-nice distribution of weights revealed that it has serious shortcomings in the fairness problem, as the constitutional treaty was also criticized (by the so-called “mathematician protest”) for not respecting the same criteria of fair weights. following the method of a priori voting power analysis, another research aim was to propose a standardized procedure of easily and fairly establishing voting weights when new enlargement waves would have to occur; votes for newcomers, but also for older members, as it implied a recalculation of their influence in the new extended formula (leech, 2002, p. 438). table 2. power distribution in the council from 1995 until nice and in case of enlargement (eu27). countries shapley-shubik index (1995) banzhaf index (1995) shapley-shubik index (eu27) banzhaf index (eu27) germany 0.119 0.112 0.0872 0.0778 united kingdom 0.119 0.112 0.0870 0.0778 france 0.119 0.112 0.0870 0.0778 italy 0.119 0.112 0.0870 0.0778 spain 0.093 0.092 0.0800 0.0742 poland 0.0799 0.0742 romania 0.0399 0.0426 netherlands 0.056 0.059 0.0368 0.0397 greece 0.056 0.059 0.0340 0.0368 czech republic 0.0340 0.0368 belgium 0.056 0.059 0.0340 0.0368 hungary 0.0340 0.0368 portugal 0.056 0.059 0.0340 0.0368 sweden 0.044 0.049 0.0281 0.0309 bulgaria 0.0281 0.0309 austria 0.044 0.049 0.0281 0.0309 slovakia 0.0195 0.0218 denmark 0.033 0.036 0.0195 0.0218 finland 0.033 0.036 0.0195 0.0218 ireland 0.033 0.036 0.0195 0.0218 lithuania 0.0195 0.0218 latvia 0.0110 0.0125 slovenia 0.0110 0.0125 estonia 0.0110 0.0125 cyprus 0.0110 0.0125 luxembourg 0.021 0.023 0.0110 0.0125 malta 0.0082 0.0094 source: adapted from aleskerov et al., 2002, pp. 384-385 oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 46 figure 1. power distribution in the council as reflected by the two well-known competing indices, before and after eu enlargement. source: own production. figure 1, based on table 2, offers a clearer image of the power distribution in the council as reflected by the two well-known competing indices. after comparing the two situations, eu 15 and eu 27, it is emphasized that, even if the power indices of the previous eu members were decreased after enlargement, the proportion between their levels of influence continued to be similar and the acceptance of the new members was counterbalanced by the big players with significant less vote shares in the power distribution. this is why nice was also dubbed as a “broadly equitable” system (leech and aziz, 2011). moberg (2011) argued that the swedish team in charge with nice negotiations proposed even back then a new voting configuration based on penrose’s formula, an idea rejected due to the increased weight it would have offered germany, while the rest of the large ms preferred to fight for their increased influence. in the end, nevertheless, moberg sustained that nice was the system as closest possible to penrose’s one. the advocates of nice would dismiss the criticisms pointing towards (a) the voting weights lacking a direct proportional link to populations (as the underrepresentation of larger states is, in fact, a constitutive principle of the late communities), (b) the high thresholds (despite them, effective qmv voting was not impeded due to the so strong culture of consensus), or (c) the impossibility to replicate it in case of future enlargement waves. the latter idea was, in fact, an accepted criticism as an opening towards the jagiellonian compromise proposal, as it will be later presented (leech and aziz, 2011). oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 47 2.3 lisbon, qmv and voting power. from voting weights to “fair enough”? it is simple to understand why “[w]hat a voting system `should` be is a political choice” (moberg, 2011). generally, small states are in favour of the weighted votes architecture, while the large ones, tending to capitalize on their population, support the double majority system. in fact, moberg argues that the lisbon treaty offers large countries the same weight they had in the eu12 configuration. different comparisons of the impact that different voting rules (pre-nice, nice, lisbon, penrose’s law) would have (van der ploeg, 2008) indicate this previous weighted vote bias favouring the smaller states, a situation changed by lisbon, also considering the big states’ interests in the context of the latest and future enlargements; that is why they were so eager to preserve the trump card of the population criteria, while the small states became interested on the threshold aspects. one of the main critics addressed the idea of supermajority voting methods following the argumentation line in favour of the simple majority rule. if the probability that a majority is right increases exponentially, then, obviously, the probability that the minority is wrong becomes very high. a supermajority rule demands that a specific level of the minority can block the process of decision-making, meaning that the higher the supermajority threshold required, the lower the minority that can veto the course of the action (dahl, 1991). however, the defenders of the supermajority frame claim that there are no permanent majorities or permanent minorities. the medium states – and here is a larger debate about properly identifying this cluster – were also advantaged by the previous nice system and, as some subsequent debates revealed, even argued for a different type of eu council voting. conveniently, “the code words for these demands have often been `democratic legitimacy` or `efficient institutions`” (moberg, 2011). the leader figure of the medium states asking for a revision of the constitutional treaty – and, later, reform treaty – stipulations on the qm definition was poland. the progressive eurosceptic attitudes of the late 2005 new polish government and the stalemate it induced in the 2007 attempts to renegotiate the late constitutional treaty are detailed in phinnemore (2013, pp. 82-84, 121-124), where the attention is also directed towards the polish concerns on the impact that the new double majority voting within the council would have on warsaw’s role in the european power game, as well as on supporting the jagiellonian compromise, also known as the equal influence system, considered to be more compatible with article i-45 requirements which foreseen “[t]he principle of democratic equality. in all its activities, the union shall observe the principle of the equality of its citizens, who shall receive equal attention from its institutions, bodies, offices and agencies”; the german council presidency's role in completing the negotiation on the final draft of the treaty, while assuring the polish support through several concessions on the double majority entering into force (a sort of a new ioannina compromise), is also widely presented by phinnemore (2013). on the polish position, the threat of not ratifying the reform treaty settlement and the ten years de facto postponement of the new qmv enforcement, slapin (2011, p. 119) also has a welldocumented input. a few words about this so-called jagiellonian compromise (after the name of its first supporters, scholars at the jagiellonian university of krakow). as i previously stated, the voting oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 48 power and the voting weights of any country are not identical notions. while the voting weights are easy to understand, when one has the formula that determines them, the voting power means “the extent to which it may influence decisions taken by the council when all possible coalitions between different countries are taken into consideration” (slomczynski and zyczkowski, 2011). as the authors are arguing, 51% of the shares of a company mean 51% of the votes (as weight) but 100% decisional power. the same idea can be found in plurality electoral systems which grants as winner the person receiving the highest number of votes even if, in a fierce competition involving more than two competitors, the total number of votes not received by this candidate could have been overwhelming. in the eu’s case, due to the increased number of member states and the number of coalitions which is in the order of millions, the voting power of a specific actor is calculated using one of the before mentioned power indexes and one of the most commonly used is the penrose-banzhaf index. one starts from penrose’s arguments that “the a priori voting power of a country is proportional to the probability that its vote will be decisive in a hypothetical ballot: should this country decide to change its vote, the winning coalition would fail to satisfy the qualified majority condition” (slomczynski and zyczkowski, 2011); its jagiellonian upgraded form stipulates country indices whose “sum is equal to unity it is easy to show that the voting power held by a given country depends not only on its voting weight but also on the distribution of the weights among all the remaining member states of the eu” (slomczynski and zyczkowski, 2011). for example, let’s imagine another fictional international organization with 5 member states which applies in one of its institutions the same voting rule as the council of the eu (population share data) and which has the following configuration of the votes: ms1 = 9; ms2 = 4, ms3 = 9; ms4 = 16; ms5 = 25. in this case, ms2 has 6.35% of the votes, and if the square root is applied one has: ms1 = 3; ms2 = 2, ms3 = 3; ms4 = 4; ms5 = 5. ms2 has now 11.76% of the votes. let’s see another configuration of the same imagined international organization, with a different distribution of weights, for the same total amount of the population: ms1 = 4; ms2 = 4, ms3 = 4; ms4 = 9; ms5 = 42. here, ms2 still has 6.35% of the vote and, if the square root is applied, one has: ms1 = 2; ms2 = 2, ms3 = 2; ms4 = 3; ms5 = 6.49. in this case, ms2 has 12.91% of the vote (see table 3 and figure 2). table 3. power distribution in two configurations of the votes for a virtual international organization. member state configuration 1 configuration 2 distribution of weights percent of votes sqrt percent of votes distribution of weights percent of votes sqrt percent of votes ms1 9 14.29 3 17.65 4 6.35 2 12.92 ms2 4 6.35 2 11.76 4 6.35 2 12.92 ms3 9 14.29 3 17.65 4 6.35 2 12.92 ms4 16 25.40 4 23.53 9 14.29 3 19.38 ms5 25 39.68 5 29.41 42 66.67 6.5 41.86 total 63 100.00 17 100.00 63 100.00 15.5 100.00 oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 49 figure 2. graphical example of the voting power held by a given country (ms2) showing how it depends not only on its voting weight but also on the distribution of the weights among all the remaining member states. in other words, “one should choose the weights to be proportional to the square root of the population and then find such an optimal quota that would produce the maximally transparent system, that is, a system under which the voting power of each member state would be approximately equal to its voting weight” (slomczynski and zyczkowski, n.d.). that is why the lisbon image of the double majority system as a result of a simultaneous support for the equal influence of the member states and of the citizens was rejected by some analysts due to the effects that would have brought benefits to the large (on a population basis) and small (on the simple majority quota basis) countries, on the expense of the medium-sized ones; these countries, and especially poland, were promoting instead the jagiellonian compromise: an equal influence power for each citizen, “transparent, easy to implement, efficient to use, and will readily accommodate any future extensions of the eu” (slomczynski and zyczkowski, 2011). the supporters of a direct link between the weights and the total population figures were dismissed by the appeal to the constitutive treaties, and the original degressive proportionality system stipulated there (moberg, 2011), as it was considered that “the idea of dividing votes proportionally to the square root of population (…) is in fact the simplest mathematical implementation of the principle of degressive proportionality and lies exactly between two extremes: ‘one country-one vote’ (…) and votes proportional to population” (slomczynski and zyczkowski, 2011). the jagiellonian scholars also dismissed, in fact, the implementation of the original penrose proposal of connecting a country’s voting power to the blunt square root of the population, especially in a weighted votes context where the allocation of weights should be associated with the setting out of a threshold for qm. instead, they proposed that these weights be allocated so that they reflect proportionally themselves the square root of the population (rounded to the oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 50 nearest integer), while the discussions about where to set the threshold became subsequent: „the optimal quota enabling the computed voting power of each country to be practically equal to the attributed voting weight, and so to be proportional to the square root of the population” (slomczynski and zyczkowski, 2011; see also pukelsheim, 2011). in other words, jagiellonian compromise “consists of a single criterion only, and it is determined by the following two rules: a. the voting weight attributed to each member state is proportional to the square root of its population. b. the decision of the voting body is taken if the sum of the weights of members of a coalition exceeds a 61.6 per cent quota”; adopting an act by using this system was praised by its supporters as “simple, objective, representative, transparent, easily extendible, moderately efficient and moderately conservative” (slomczynski and zyczkowski, 2011). the authors (and also kirsch, 2011 who speaks of a 61.4 per cent quota when mentioning 2004 and 2006 older contributions of slomczynski and zyczkowski) also suggested a modified system introducing the simple or qualified majority of the member states as a secondary criterion; it would not have altered the objective of the square root law, but it would have reinforced the double “union of people” and “union of states” eu nature, also being in favour of the less populous states. nevertheless, in other opinions, “[t]he choice of a quota is a technical means suitable for attaining a given political end” (sozanski, 2011). in this interpretation, the quota is not influencing the power distribution, but the efficiency of the decisional system in terms of a possible number of majorities able to pass a piece of legislation. therefore, it is simpler to understand the interest that the political spectrum has for blocking coalitions. the supporters of the jagiellonian system also underlined that it would succeed to harmonize the eu citizens power shares so that they have an – even if indirectly – equal influence on the eu council decisions. the direct voting power of a citizen is, of course, the one reflected in the domestic elections. leech and aziz (2011) argue that the total voting power of an eu citizen – or the personal penrose power index – represents the product of the power his/her country has in the eu council and his/her power at the national level. when applying the jagiellonian formula to different eu scenarios, the results indicated that the relative citizen voting powers had only minor variations, contrary to the constitutional / lisbon treaty. for this reason, pukelsheim, for example, saw jagiellonian proposal as a “rather sophisticated idealisation of democratic equality” (pukelsheim, 2011). comparative to the double majority system (praised mainly for its capacity to increase the decisiveness of the council), several other analyses also pointed jagiellonian compromise as “remarkably equitable” (especially for the medium-sized countries, not favoured by the current settlement) irrespective of several future enlargements considered by the authors (2011). machover (2011) reminded that poland’s 2007 efforts to support a decisional structure based explicitly on the jagiellonian compromise failed, while the new lisbon system – as data indicated – was even more distant than nice when one refers to the square root law as an ideal standard. therefore, machover insisted that any further negotiation of the square root theme to be held separately on the application of the rule, on the one hand, and the issue of establishing a specific quota, in the other. oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 51 nevertheless, sometimes the arguments of the jagiellonian supporters tend to be rather ideologically influenced than scientifically based. as many of these arguments are elaborated by polish scientists and disseminated by other non-polish scholars in collective works coordinated by the members of the polish academia as is the case of cichocki and zyczkowski volume (2011) -, they seem strongly biased for supporting its cause even if, sometimes, they add spain in the same category in order to be able to refer to a broader “medium-sized countries” case, as van der ploeg (2008) argues that spain also tried to block the convention talks due to the council decision provisions. in the aspects regarding the jagiellonian arguments, on the one side, it is underlined eu’s necessity to get closer to the democratic benchmark by offering each citizen the same decisional influence. as a consequence of this argument, one seems (only seems, as in reality, they are not connected) to also find the idea that one state’s impossibility to split its votes accordingly to various national positions on a specific issue would qualify the weight votes allotment on a population basis to be unfair and favourable to large states (kirsch, 2011). on the other side, nor in the jagiellonian system the votes are split, and this would be possible only if each citizen were able to vote directly on each eu-level decision, which is utopian. the wording of the theme is, most of the time, confusing. as previously indicated, some argue that qmv debates are about the equal distribution of power within the eu council: “it is a fundamental principle of the eu that all citizens should have equal rights, whatever country they happen to live in”, and the arguments are constructed by assuming the citizens’ equality axiomatically right as “a benchmark against which to compare the fairness of the distribution of voting power” (leech and aziz, 2011). but is it true? in fact, the broad mathematical views on the subject are simple to understand: (a) should each citizen have the same power input irrespective of its member state provenience or (b) allotting one vote per each citizen should transform into an increased power quota for larger states? nevertheless, beyond mathematics, various criteria have been considered when analysing voting formulas. among them, van der ploeg (2008) notes equity, efficacy and comprehensibility or, in other words, a fair distribution of power in comparison with a member state’s total population, the capacity of a specific voting option to encourage a fluent decision-making process and, respectively, an open to public scrutiny and easy to understand system. according to her analysis, a square root law-based system or one built on the jagiellonian compromise would indeed better observe the before-mentioned criteria in comparison with the lisbon one. the political choices, however, are different, and there is no sign of a concrete dialogue between the parts. one cause of the misunderstandings accompanying the subject is the mutual contempt between mathematicians and politicians. for example, sozanski (2011) taxes as amateurish the approach of politicians to support one rule or another, arguing that the mathematicians are those really carrying the burden "by translating given voting rules from the legal language used by politician into the set-theoretical language of mathematics" and analysing these mathematical models as various voting games. i consider that there is more behind the political choices than that unfair and reductionist oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 52 “amateurship” label. indeed, this is an important question: if it is so providential, when considering each citizen’s input and the possibility of extrapolating its mechanisms to future eu enlargements, why the other states do not intend to switch towards using the jagiellonian formula, not even those who seem to have the same voting power in the lisbon context when compared to the jagiellonian simulations? as revealed by the tables, the number of states for whom the jagiellonian proposal is not only fair, but also favourable compared to lisbon, is close to 2/3 of the members. so, why are they not supporting this change, both due to net benefits and fairness of the results? and if is so ideal, so to speak, why do the majority of its supporters still favour additional conditions, such as the agreement of the simple or qualified majority of the member states? arguing in favour of the jagiellonian compromise and claiming that “[t]he shifts of seats and power would seem trivial compared to the significant gain in democratic substance – a consequence of putting citizens first” (pukelsheim, 2011) might seem rather a political and not a scientific approach, if one cannot find proper answers. in my opinion, this jagiellonian issue reflects the li power politics in the eu and one could argue that the compromise might be ideal for everybody, and that is why not “ideal enough” for the big players. then why all the fuss? follow the money, as one example (among many others) could be found in the context of the solidarity/cohesion discourse and relocation of the supporting fund. in fact, the situation is more nuanced, and it remains to be seen where one can draw the line between discussing the justice of a system, whatever it may be, and the differentiated impact that system could have in the event of a fluctuation of the member states. for example, kirsch (2016) reveals brexit’s impact on the eu, both in a 27 members’ scenario (without uk) and in a 28 one (with scotland). his research indicates, using the banzhaf index on the current lisbon amended framework, that small members’ states (under 4.5 million citizens) will be the losers of a simple no-uk configuration, where the qmv requires 15 states and “[t]he blocking minority must include at least four council members representing more than 35% of the eu population” (european council, 2022). middle-sized countries, like poland, would strongly benefit, alongside the larger ones. a scenario that would contain scotland would lead to different results, as the first seven most populous countries would see their influence reduced, while the over 2/3 remaining member states would be the main beneficiaries of the new power equilibrium. hence, a system that could be considered fair regardless of these fluctuations of the eu componence could represent a solution. the post-brexit eu power distribution – also calculated on a square root basis – is said to be in favour of the most populous member states (significant for the first five and satisfactory for the next two), due to the decrease in the number of states with a population beyond the eu average. so, the role of the larger states is set to be strengthened due to their future increased influence on the success of a coalition (see the results of kirsch et al., 2018, pp. 24, 27). in corollary, due to the diminishing of the number of states threshold for passing a decision, the role of the less populous ms will be further diminished (kleinowski, 2019). oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 53 figure 3. comparison between banzhaf indices for lisbon system and jagiellonian compromise indices before and after brexit. source: own production employing data from kirsch et al. (2018). nevertheless, the results of kirsch et al. (2018), graphically represented in figure 3, show that a possible use of the jagiellonian compromise would be in benefit of all countries in a post-brexit context (kirsch et al., 2018). here one can mention again the delicate issue of the number of residents used in population statistics relevant for the eu decisional process, as the “old union” (and not exactly in the most / medium populous states, such as poland) is getting advantages from this flux (kleinowski, 2019). oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 54 3. discussions and conclusions conceived as key points in the institutional development of the eu, the treaty of nice and, later, the lisbon treaty (via its constitutional treaty inheritance) proved finally to be extremely contested for their provisions regarding the structural changes envisaged for an enlarged union. institutional arrangements for deciding the best frame for collective decision should take into account the context of the decision process and the outcomes coming up from the political game. qmv, depending on its definition, can imply (a) privileges for the status-quo; the status-quo, as dahl (1991) proves, is not a neutral alternative not for the simple majority, (b) the exiting of a constant majority that can dominate the system or, (c) by contrary, a strong minority that can abusively use its blocking power despite the collective interest. finally, one can draw some conclusions referring to the initial research questions about the institutional dimension (the third li level) and the issue of rational collective output. various changes in the regulations may have occurred, but sometimes they are, in reality, marginal. “the institutions frequently becom[e] reified as rational actors themselves, rather than the reflections of the collective actions of the individuals within them” (peters, 2019, p. 63). if one compares the negotiation process in both nice and constitutional/lisbon treaty, one notices that the alternatives envisaged by the legislators were minimum, and the degree of difference between the final outcomes was not very high, justifying the idea that the final objectives were not enough clear from the beginning and that the change of the status quo was, in fact, incremental if one thinks at the values that would have had to support this important process. actually, more than 60 years later after arguing it, it is still valid that “incrementalism in the large becomes potentially a major political orientation” (lindblom, 1958, p. 312). this problem of collective action transferred into the game theory would face some challenges in describing the type of game played by the council’s members. if it is common sense to agree that one should speak about coordination games (the repetition of the game being a factor that could lead to an optimal equilibrium between the parts), the reality confronts us with a de facto “prisoner’s dilemma” (summing up the individual results offers a lower figure than the one that would have been obtained by a collective action). the difficulty of the cooperation between all the players could have been influenced by the labels that each actor inevitably had (axelrod, 1984). in the eu negotiations, the rival labels “major power” – “minor power” negatively influenced the whole process indeed, enforcing one part’s status and constantly trying to disadvantage the other. nevertheless, cooperation was intense between the two playing parts. the small players and also the medium-sized spain and poland succeeded in improving their condition by joint continuous actions designed to sanction the defection. therefore, a rational collective outcome of the eu council is indeed far away from being reached, and that future institutional arrangements (a consequence of a new treaty, of the withdrawal of some members or of a new enlargement wave) should be treated with more consideration regarding the power distribution into the voting system. oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 55 so, what are the chances for the ideal type to be implemented in the near or medium future? politics is a never-ending story, the political issues being discussed and reviewed constantly accordingly to some given principles. however, the issue of power distribution in a voting system should be treated with more consideration for the idea of a rational collective outcome. nevertheless, the current divisions between the member states (various dichotomies as old versus new member states, northern europe versus southern europe, western europe versus eastern europe, larger versus smaller member states, supranationalist versus intergovernmentalist supporters, etc.) amplified during the latest crises (financial aspects, migration, populism, foreign policy, uk withdrawal, coronavirus disease, russia’s invasion of ukraine) are a strong impediment for the willingness of the leaders to embark on a trip towards a new treaty (stratfor, 2018), irrespective of the fairness it might bring, as the possible spill-back (niemann, lefkofridi and schmitter, 2018) consequences are considerable higher. so, should we just try to limit ourselves to coping with the current lisbon framework, even if this third li level needs serious updates in order to ensure democratic legitimacy when respecting the assumed engagements? in the end, one should mention that between april 2021 and may 2022, the eu organized the conference on the future of europe, a pan-eu democratic exercise where citizens expressed their vision regarding the priorities to be set and challenges to be overcome in a common future. considering the attributes of each major institution involved within this initiative (the european commission, the european parliament and the council of the european union), as well as the aim of this approach to give “europeans a greater say on what the union does and how it works for them” (european commission, 2020), even in this context little can be expected to be translated in concrete measures in the strictly intergovernmental nature of the voting power discussions. besides a vague call to „reopening the discussion about the constitution”, the citizens courageously asked for „reassessing decision-making and voting rules in the eu institutions” in an attempt to strengthen the qualified majority voting to the detriment of unanimity while ensuring a fair calculation of voting 'weights' so that small countries' interests are protected” (art. 7, proposal 39, eu, 2022). nevertheless, alternatives such as the jagiellonian compromise do not seem to have immediate future success, considering that their general fairness in terms of voting power is contradictory to the current stronger influence of the larger states, which would not generate their support for a future decisional framework change. acknowledgements the author thanks professor adrian miroiu for his careful reading of the initial version of the manuscript and for his insightful suggestions regarding a more detailed explanation of the effects of the distribution of the weights. the author is also grateful to the anonymous reviewer whose valuable comments and suggestions substantially contributed to an improved and clearer version of the manuscript. oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 56 references anonymous. (2017). not dead yet. revisiting the 'luxembourg veto' and its foundations. european constitutional law review, 13(1), 1-12. https://doi.org/10.1017/s1574019617 000013 aleskerov, f., avci, g., iakouba, v., and umut türem, z. (2002). european union enlargement: power distribution implications of the new institutional arrangements. european journal of political research, 41 (3), 379-394. https://doi.org/10.1111/1475-6765.t01-1-00016 axelrod, r. (1984). the evolution of cooperation. new york: basic books. barr, j., and passarelli, f. (2009). who has the power in the eu? mathematical social sciences, 57 (3), 339-366. https://doi.org/10.1016/j.mathsocsci.2008.12.007 cichocki, m.a., and życzkowski, k., eds. (2011). institutional design and voting power in the european union. 1 ed. london: routledge. dahl, r. (1991). democracy and its critics. yale university press. european commission. (2020). shaping the conference on the future of europe. communication from the commission to the european parliament and the council. brussels, 22.1.2020 com(2020) 27 final. european council / council of the european union. (2022). qualified majority. https:// www.consilium.europa.eu/en/council-eu/voting-system/qualified-majority/ last reviewed on 13 october 2021. accessed on january 2022. european union. (2022). conference on the future of europe. report on the final outcome (may 2022). felsenthal, d., leech, d., list, c., and machover, m. (2003). in defence of voting power analysis. responses to albert. european union politics, 4 (4), 473 497. https://doi.org/10.1177/1465116503004004005 galloway, d. (2001). the treaty of nice and beyond. realities and illusions of power in the eu. sheffield: sheffield academic press. ion, o.a. (2013). guvernanta uniunii europene. abordari actuale. iasi: polirom. kirsch, w., slomczynski, w., stolicki, d., and zyczkowski, k. (2018). double majority and generalized brexit: explaining counterintuitive results. arxiv:1812.07048 [physics.soc-ph]. accessed on october 2021. kirsch, w. (2011). the distribution of power in the council of ministers of the european union. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 93-109). london: routledge. kirsch, w. (2016). brexit and the distribution of power in the council of the eu. centre for european policy studies. available at: https://www.ceps.eu/ceps-publications/brexit-anddistribution-power-council-eu/ accessed on september 2021. kleinowski, m. (2018). the impact of brexit on the voting power in the council of the european union. przegląd europejski, 4, 95-117. https://doi.org/10.5604/01.3001.0013.3456 leech, d., and aziz, h. (2011). the double majority voting rule of the eu reform treaty as a democratic ideal for an enlarging union: an appraisal using voting power analysis. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 59-74). london: routledge. leech, d. (2002). designing the voting system for the council of the european union. public choice, 113 (3), 437 464. https://doi.org/10.1023/a:1020877015060 lindblom, c.e. (1958). policy analysis. the american economic review, 48 (3), 298-312. machover, m. (2011). penrose's square-root rule and the eu council of ministers: significance of the quota. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 35-42). london: routledge. miroiu, a. (2007). fundamentele politicii. rationalitate si actiune colectiva. 2nd volume. iasi: polirom. moberg, a. (2011). is the double majority really double? the voting rules in the lisbon treaty. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 19-34). london: routledge. moravcsik, a., and schimmelfennig, f. (2018). liberal intergovernmentalism. in a. wiener, t.a. börzel and t. risse (eds.). european integration theory (pp. 64-84). 3rd edition. oxford: oxford university press. https://doi.org/10.1093/hepl/9780198737315.003.0004 moravcsik, a. (1993). preferences and power in the european community: a liberal intergovernmentalist approach. journal of common market studies, 31 (4), 473-524. https://doi.org/10.1017/s1574019617000013 https://doi.org/10.1017/s1574019617000013 https://doi.org/10.1111/1475-6765.t01-1-00016 https://doi.org/10.1016/j.mathsocsci.2008.12.007 http://www.consilium.europa.eu/en/council-eu/voting-system/qualified-majority/ https://doi.org/10.1177/1465116503004004005 https://www.ceps.eu/ceps-publications/brexit-and-distribution-power-council-eu/ https://www.ceps.eu/ceps-publications/brexit-and-distribution-power-council-eu/ https://doi.org/10.5604/01.3001.0013.3456 https://doi.org/10.1023/a:1020877015060 https://doi.org/10.1093/hepl/9780198737315.003.0004 oana andreea ion / european journal of government and economics 12(1), june 2023, 39-57 57 https://doi.org/10.1111/j.1468-5965.1993.tb00477.x moravcsik, a. (1997). taking preferences seriously: a liberal theory of international politics. international organization, 51 (4), 513-553. https://doi.org/10.1162/002081897550447 niemann, a., lefkofridi, z., and schmitter, p.c. (2018). neofunctionalism. in a. wiener, t.a. börzel and t. risse (eds.). european integration theory (pp. 43-63). 3rd edition. oxford: oxford university press. https://doi.org/10.1093/hepl/9780198737315.003.0003 nurmi, h. (2011). on the relative unimportance of voting weights: observations on agendabased voting procedures. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 171-180). london: routledge. ostrom, e. (1990). governing the commons: the evolution of institutions for collective action, political economy of institutions and decisions. cambridge university press. https://doi.org/10.1017/cbo9780511807763 passarelli, f., and barr, j. (2004). preferences, the agenda setter and the distribution of power in the eu. social choice and welfare, 28 (1), 41-60. https://doi.org/10.1007/s00355-006-01556 penrose, l.s. (1946). the elementary statistics of majority voting. journal of the royal statistical society, 109 (1), 53-57. https://doi.org/10.2307/2981392 peters, g.b. (2019). institutional theory in political science: the new institutionalism. edward elgar publishing. 4th edition. isbn: 978 1 78643 792 1. phinnemore, d. (2013). the treaty of lisbon. origins and negotiation. palgrave macmillan. https://doi.org/10.1057/9781137367877 pollack, m.a. (2020). theorizing eu policy-making. in h. wallace, m.a. pollack, c. roedererrynning and a.r. young (eds.). policy making in the european union (pp. 13-42). 8th edition. oxford university press. isbn: 9780198807605. https://doi.org/10.1093/hepl/ 9780198807605.003.0002 pukelsheim, f. (2011). putting citizens first: representation and power in the european union. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 235-253). london: routledge. putnam, r.d. (1988). diplomacy and domestic politics: the logic of two-level games. international organization, 42 (3), 427-460. https://doi.org/10.1017/s0020818300027697 shepsle, k.a., and boncheck, m.s. (2010). analyzing politics. rationality, behavior and institutions. 2nd edition. new york: w.w. norton & company. slapin, j.b. (2011). veto power. institutional design in the european union. the university of michigan press. https://doi.org/10.3998/mpub.2012704 slomczynski, w., and zyczkowski, k. (2011). jagiellonian compromise an alternative voting system for the council of the european union. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 43-58). london: routledge. slomczynski, w., and zyczkowski, k. (n.d.) jagiellonian compromise an alternative voting system for the council of the european union. available at: https://chaos.if.uj.edu.pl/~karol/pdf/jagcom07.pdf. accessed on august 2021. sozanski, t. (2011). the voting power in the voting systems. in m.a. cichocki and k. zyczkowski (eds.). institutional design and voting power in the european union (pp. 75-92). london: routledge. stratfor. (2018). the eu and the unanimity trap. available at: https://worldview.stratfor.com /article/eu-and-unanimity-trap. accessed on november 2021. suzumura, k. (2016). choice, preferences, and procedures. a rational choice theoretic approach. london: harvard university press. https://doi.org/10.4159/9780674726307 van der ploeg, c. (2008). the distribution of power in the council of the european union. bwi werkstuk (vrije universiteit amsterdam). available at: https://www.semanticscholar.org/ paper/the-distribution-of-power-in-the-council-of-the-werkstuk-ploeg/1d1b04803dc1ada2a 58a12f90449e8d080225661 accessed in december 2021. verba, s. 92006). fairness, equality, and democracy: three big words. social research: an international quarterly of social sciences, 73(2), 499-540. https://doi.org/10.1353/ sor.2006.0029 warntjen, a. (2017). do votes matter? voting weights and the success probability of member state requests in the council of the european union. journal of european integration, 39 (6), 673-687. https://doi.org/10.1080/07036337.2017.1332057 https://doi.org/10.1111/j.1468-5965.1993.tb00477.x https://doi.org/10.1162/002081897550447 https://doi.org/10.1093/hepl/9780198737315.003.0003 https://doi.org/10.1017/cbo9780511807763 https://doi.org/10.1007/s00355-006-0155-6 https://doi.org/10.1007/s00355-006-0155-6 https://doi.org/10.2307/2981392 https://doi.org/10.1057/9781137367877 https://doi.org/10.1093/hepl/%0b9780198807605.003.0002 https://doi.org/10.1093/hepl/%0b9780198807605.003.0002 https://doi.org/10.1017/s0020818300027697 https://doi.org/10.3998/mpub.2012704 https://chaos.if.uj.edu.pl/%7ekarol/pdf/jagcom07.pdf.%20accessed%20on%20august%202021 https://doi.org/10.4159/9780674726307 https://www.semanticscholar.org/%0bpaper/the-distribution-of-power-in-the-council-of-the-werkstuk-ploeg/1d1b04803dc1ada2a%0b58a12f90449e8d080225661 https://www.semanticscholar.org/%0bpaper/the-distribution-of-power-in-the-council-of-the-werkstuk-ploeg/1d1b04803dc1ada2a%0b58a12f90449e8d080225661 https://www.semanticscholar.org/%0bpaper/the-distribution-of-power-in-the-council-of-the-werkstuk-ploeg/1d1b04803dc1ada2a%0b58a12f90449e8d080225661 https://doi.org/10.1353/%0bsor.2006.0029 https://doi.org/10.1353/%0bsor.2006.0029 https://doi.org/10.1080/07036337.2017.1332057 1. introduction 2. materials and methods 3. discussions and conclusions references © the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 2 (2025), pages 147-163 https://doi.org/10.17979/ejge.2025.14.2.11181 submitted: mar 9, 2024 accepted: jul 15, 2025 published: dec 17, 2025 article public debt and fiscal consolidation in the western balkans: a panel analysis of growth effects artina kamberi,1 abdylmenaf bexheti 1 1 south east european university, north macedonia *correspondence: ak19520@seeu.edu.mk abstract. this study examines the impact of public debt on economic growth and the effects of fiscal consolidation efforts in the western balkan countries (wb6)—albania, bosnia and herzegovina, kosovo, montenegro, north macedonia, and serbia. utilizing a dual methodology of descriptive and panel data analysis, the research investigates whether higher public debt adversely affects economic growth and if fiscal consolidation measures lead to improved economic outcomes. the descriptive analysis identifies fiscal consolidation periods based on improvements in the cyclically-adjusted primary budget balance and subsequent changes in the public debt-to-gdp ratio. the econometric analysis employs both fixed-effects and random-effects panel regression models, using data from international monetary fund (imf) world economic outlook databases covering the period 2000-2023. the findings reveal that higher levels of public debt are detrimental to economic growth across the wb6 countries. conversely, successful fiscal consolidation, characterized by reductions in government expenditure relative to gdp, correlates with enhanced economic growth. the study underscores the importance of maintaining prudent debt levels and implementing effective fiscal policies. recommendations for policymakers include prioritizing debt reduction strategies, improving revenue collection, and investing savings in infrastructure, education, and innovation to support long-term economic stability and growth. keywords: western balkans; debt-to-gdp ratio; fiscal consolidation; economic growth; panel data analysis jel classification: i1; d53; g15; g12; c22 1. introduction public debt and fiscal policy have been central issues in economic discourse, particularly in the context of emerging economies such as the western balkan countries (wb6). the wb6 countries— comprising albania, bosnia and herzegovina, kosovo, montenegro, north macedonia, and serbia— have experienced varying degrees of economic challenges and growth trajectories since the early 2000s. amidst these challenges, the sustainability of public finances and the implications of fiscal consolidation efforts have garnered significant attention from policymakers and researchers alike. research question: what is the impact of public debt on economic growth in the western balkan countries (wb6), and how do fiscal consolidation efforts affect this relationship? https://creativecommons.org/licenses/by-nc/4.0/ 148 kamberi and bexheti hypothesis: based on existing literature and theoretical frameworks, we propose the following hypotheses: h0: higher levels of public debt in the wb6 countries negatively impact economic growth. h1: fiscal consolidation measures aimed at reducing public debt-to-gdp ratios lead to improved economic growth outcomes. this study aims to contribute empirical evidence to inform policymakers and stakeholders about the implications of public debt and fiscal consolidation on economic performance in the western balkans. 2. literature review the literature on fiscal consolidation and debt management provides valuable insights into the challenges and strategies pertinent to the western balkan countries. historically, these nations have navigated complex economic transitions and geopolitical shifts, influencing their fiscal policies and debt dynamics (blejer & leone, 2020; world bank, 2021). post-conflict reconstruction efforts and institutional reforms have shaped their fiscal landscapes, impacting revenue generation, expenditure priorities, and debt sustainability (international monetary fund [imf], 2019). scholars highlight the theoretical foundations of fiscal consolidation, emphasizing the trade-offs between short-term stabilization measures and long-term fiscal sustainability (alesina & perotti, 1994, 1999; wyplosz, 2013). effective consolidation strategies typically involve a balanced approach of expenditure rationalization, revenue enhancement, and structural reforms (von hagen & harden, 1995). case studies from diverse economies underscore the importance of political commitment and institutional capacity in achieving successful fiscal adjustment (cottarelli & giannini, 2017; european commission, 2020). in the context of the western balkans, fiscal challenges are exacerbated by high levels of public debt and vulnerabilities to external shocks (imf, 2022). countries such as serbia and montenegro have implemented fiscal consolidation programs aimed at reducing deficits and stabilizing debtto-gdp ratios amidst varying degrees of political stability and institutional capacity (milosavljevic, 2018; world bank, 2020). the european union integration process plays a pivotal role in shaping fiscal policies in the region, influencing governance standards and economic convergence goals (european commission, 2021). regional initiatives, such as the western balkans investment framework, aim to foster economic resilience and infrastructure development through targeted financial assistance and policy coordination (european investment bank, 2019; regional cooperation council, 2020). empirical evidence suggests that credible fiscal reforms can enhance investor confidence, stimulate economic growth, and improve social outcomes (acemoglu et al., 2019; ebrd, 2021). however, challenges persist in achieving sustainable fiscal outcomes amidst political instability, corruption risks, and demographic pressures (marinov et al., 2017; world bank, 2023). public debt and fiscal consolidation in the western balkans 3. methodology to investigate the impact of fiscal consolidation on economic growth in western balkan countries, this study employs a dual approach: descriptive analysis and panel data analysis. the descriptive analysis first outlines the criteria for identifying periods of fiscal consolidation, following the framework established by alesina and ardagna (1998) and adapted for post-transition economies by mirdala (2013). fiscal consolidation periods are defined by either: a one-year improvement in the cyclically-adjusted primary budget balance of at least 1.5% of gdp, or three consecutive years where this balance does not deteriorate by more than 0.5% of gdp. a fiscal consolidation is considered successful if, over the three-year period following the consolidation: the cyclically-adjusted primary deficit decreases on average by at least 2% relative to its level in the consolidation year, or the public debt-to-gdp ratio is at least 5% lower than in the consolidation year. additionally, fiscal consolidation is classified as expansive if the average gdp growth rate during the consolidation period and the subsequent two years surpasses the average growth rate prior to consolidation (alesina and ardagna, 1998; mirdala, 2013). government expenditure as a percentage of gdp is used as a proxy for fiscal consolidation, following similar approaches in empirical literature. however, we recognize this is a simplified representation. ideally, fiscal consolidation would be measured using structural balance changes or cyclically-adjusted primary balances, but such data are not consistently available across wb6 countries. as such, this proxy provides a workable— though limited—approximation of fiscal effort. this study employs an econometric approach to investigate the relationship between public debt and economic growth in the western balkan countries (wb6). the research utilizes panel data analysis, which is well-suited for examining longitudinal data across multiple countries over time, allowing for the control of individual country-specific effects. data for this study are sourced primarily from international databases such as the world bank, international monetary fund (imf), eurostat, and national statistical offices of the western balkan countries. the dataset includes annual observations of key variables from 2000 to 2023, capturing macroeconomic indicators such as gdp growth rates, public debt as a percentage of gdp, inflation rates, unemployment rates, and government expenditure. variables dependent variable: gdp growth rate (gdp_growth): annual percentage change in real gdp, representing the economic growth rate of each country in the western balkans. independent variables: public debt (public_debt): the ratio of gross government debt to gdp, reflecting the fiscal burden on the economy. control variables: includes variables such as inflation rate (inflation), unemployment rate (unemployment) and government expenditure as a percentage of gdp (gov_expenditure). 150 kamberi and bexheti (i) econometric model specification given the panel nature of the data, the analysis employs both fixed-effects and random-effects panel regression models to assess the impact of public debt on economic growth while controlling for other factors. 𝐺𝐺𝐺𝐺𝐺𝐺_𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔ℎ𝑖𝑖𝑔𝑔 = 𝛽𝛽0 + 𝛽𝛽1𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑖𝑖𝑝𝑝_𝑑𝑑𝑑𝑑𝑝𝑝𝑔𝑔𝑖𝑖𝑔𝑔 + 𝛽𝛽2𝑝𝑝𝑔𝑔𝑐𝑐𝑔𝑔𝑔𝑔𝑔𝑔𝑝𝑝_𝑣𝑣𝑣𝑣𝑔𝑔𝑖𝑖𝑣𝑣𝑝𝑝𝑝𝑝𝑑𝑑𝑣𝑣𝑖𝑖𝑔𝑔 + 𝜖𝜖𝑖𝑖𝑔𝑔 where: 𝐺𝐺𝐺𝐺𝐺𝐺_𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔𝑔ℎ𝑖𝑖𝑔𝑔: dependent variable representing gdp growth rate for country i in year t. 𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑝𝑖𝑖𝑝𝑝_𝑑𝑑𝑑𝑑𝑝𝑝𝑔𝑔𝑖𝑖𝑔𝑔: independent variable denoting public debt as a percentage of gdp for country i in year t. 𝑝𝑝𝑔𝑔𝑐𝑐𝑔𝑔𝑔𝑔𝑔𝑔𝑝𝑝_𝑣𝑣𝑣𝑣𝑔𝑔𝑖𝑖𝑣𝑣𝑝𝑝𝑝𝑝𝑑𝑑𝑣𝑣𝑖𝑖𝑔𝑔: vector of control variables influencing gdp growth in country i in year t. 𝛽𝛽0,𝛽𝛽1,𝛽𝛽2: coefficients to be estimated. 𝜖𝜖𝑖𝑖𝑔𝑔: error term capturing unobserved factors affecting gdp growth. (ii) explanation for choosing the model panel data approach: the use of panel data allows for the examination of both cross-sectional and time-series variations across the wb6 countries. this approach is beneficial because it considers country-specific effects while controlling for unobserved heterogeneity that may affect economic growth. fixed-effects vs. random-effects models: fixed-effects model: by controlling for time-invariant country-specific effects, such as cultural, institutional, or geographical factors, this model helps isolate the impact of changes in public debt on economic growth within each country over time. random-effects model: assuming no correlation between country-specific effects and public debt, this model provides average effects across the wb6 countries, providing broader insights into the relationship between public debt and economic growth. while the pooled ols and random effects models are presented for comparison, the fixed effects model is preferred based on the hausman test (p = 0.03). these additional models help to assess the consistency and robustness of the core relationship, though their underlying assumptions are less appropriate for capturing unobserved heterogeneity across wb6 countries. control variables: including variables such as inflation, unemployment, and government expenditure allows for a comprehensive analysis by accounting for other factors that may influence economic growth independently of public debt. public debt and fiscal consolidation in the western balkans 4. estimations and results table 1. effects of fiscal consolidation on public finances in western balkans country years of fiscal consolidation effects on public finances albania 2014-2016 2018-2019 reduced budget deficit from 5.2% to 2.4% of gdp (20142016) improved tax collection, reduced informal employment (2018-2019) bosnia and herzegovina 2012-2014 2016-2017 narrowed budget deficit, modest decrease in public debt (2012-2014) more balanced budget, gradual decline in debt levels (20162017) kosovo 2015-2017 reduced budget deficit, maintained manageable debt level (2015-2017) montenegro 2017-2019 reduced budget deficit, curbed rapid growth of public debt (2017-2019) north macedonia 2012-2014 2017-2019 lowered budget deficit, stabilized public debt (2012-2014) reduced fiscal deficit, stabilized debt-to-gdp ratio (20172019) serbia 2014-2016 2017-2019 drastically reduced deficit from 6.6% to below 2% of gdp, decreased public debt (2014-2016) budget surplus by 2017, public debt declining steadily (2017-2019) source: author’s findings and calculations based on imf, world economic outlook databases the consolidation factor index is a theoretical construct that is be used to compare the effectiveness and intensity of fiscal consolidation efforts among western balkan countries. this index would ideally take into account various indicators, such as reductions in budget deficits (bdr), changes in public debt, improvements in tax collection, and structural reforms. cfi is designed as a comprehensive metric to assess the effectiveness and sustainability of fiscal consolidation efforts within a country. cfi = bdr + pds + tci + sr + es / n(v) the structural reform (sr) component of the consolidation factor index (cfi) captures key policy changes in public administration, pension systems, and labor market efficiency. the evaluation is based on imf article iv consultations, world bank public finance reviews, and national reform agendas. each reform effort was rated on a scale from 1 to 10: (i) 1–3: minimal reform, mostly policy discussion or partial implementation (ii) 4–6: moderate reform with partial impact and institutional uptake (iii) 7–8: substantial reform with clear policy shifts and some measurable effects (iv) 9–10: deep structural transformation with broad economic impact scores were averaged and triangulated using cross-institutional reports and policy briefs for consistency. 152 kamberi and bexheti table 2. cfi* in western balkan countries country years of fiscal consolidation (bdr)* (pds) * (tci)* (sr)* (es)* consolidation factor index albania 2014-2016 2018-2019 9 7 8 7 8 7.8 bosnia and herzegovina 2012-2014 2016-2017 7 6 6 6 6 6.2 kosovo 2015-2017 7 8 7 6 7 7 montenegro 2017-2019 7 7 6 7 6 6.6 north macedonia 2012-2014 2017-2019 8 7 7 7 7 7.2 serbia 2014-2016 2017-2019 10 9 8 9 9 9 source: author’s findings and calculations based on imf, world economic outlook databases it is important to note that interpreting tax revenue, expenditures, and debt levels as a percentage of gdp requires careful consideration of both the numerator (fiscal variable) and denominator (gdp). for instance, a decline in tax revenue as a percentage of gdp might reflect either a real drop in collections or a rapid increase in gdp. similarly, during economic downturns, gdp contractions can exaggerate fiscal burdens even if revenue collection remains stable. thus, when analyzing trends across the tables in this section, we account for these dynamics and contextualize them within broader macroeconomic developments such as regional recessions, structural reforms, and external shocks (e.g., the 2008 financial crisis, covid-19 pandemic). table 3. tax revenue as a percentage of gdp (pre-consolidation) in western balkan countries country pre-consolidation period tax revenue (% of gdp) albania before 2014-2016 & 2018-2019 19.5 bosnia and herzegovina before 2012-2014 & 2016-2017 30.3 kosovo before 2015-2017 24.8 montenegro before 2017-2019 26.1 north macedonia before 2012-2014 & 2017-2019 18.5 serbia before 2014-2016 & 2017-2019 21.3 source: author’s findings and calculations based on imf, world economic outlook databases table 3 provides insights into the tax revenue as a percentage of gdp for several countries in the western balkans region during specific pre-consolidation periods. the data is crucial for understanding fiscal trends and policy implications in these economies. firstly, albania, with a tax revenue of 19.5% of gdp before 2014-2016 and 2018-2019, indicates a moderate level of revenue collection compared to its gdp, potentially reflecting its tax policy effectiveness and economic structure during those periods. bosnia and herzegovina recorded a tax revenue of 30.3% of gdp before 2012-2014 and 2016-2017, suggesting a relatively higher tax burden on its economy during these years. this could signify significant fiscal efforts or structural characteristics influencing tax collection. kosovo's tax revenue stood at 24.8% of gdp before 2015public debt and fiscal consolidation in the western balkans 2017, indicating its fiscal landscape during this period and providing a basis for understanding its revenue mobilization efforts and economic dynamics. montenegro's tax revenue was 26.1% of gdp before 2017-2019, highlighting its fiscal stability and the effectiveness of tax policies during that timeframe, potentially influencing its economic growth and public finance management. north macedonia reported a tax revenue of 18.5% of gdp before 2012-2014 and 2017-2019, indicating variations in tax collection effectiveness and economic circumstances over these periods. serbia's tax revenue was 21.3% of gdp before 20142016 and 2017-2019, reflecting its tax policy dynamics and economic conditions during these specific pre-consolidation periods. table 4. tax revenue as a percentage of gdp (during consolidation) in western balkan countries. country consolidation periods tax revenue (% of gdp) albania 2014-2016 & 2018-2019 18.2 bosnia and herzegovina 2012-2014 & 2016-2017 26.6 kosovo 2015-2017 25.3 montenegro 2017-2019 23.5 north macedonia 2012-2014 & 2017-2019 17.9 serbia 2014-2016 & 2017-2019 20.5 source: author’s findings and calculations based on imf, world economic outlook databases table 4 presents an analysis of tax revenue as a percentage of gdp for six countries in the western balkans region during specific consolidation periods. albania during the consolidation periods of 2014-2016 and 2018-2019, albania's tax revenue constituted 18.2% of gdp. this figure reflects a slight decrease compared to the preconsolidation periods, indicating potential shifts in fiscal policy priorities or economic conditions affecting revenue collection. bosnia and herzegovina in the consolidation periods of 2012-2014 and 2016-2017, reported tax revenue equivalent to 26.6% of gdp. this data highlights continued fiscal robustness and effective tax administration strategies in maintaining revenue levels amidst economic challenges or reforms during these periods. kosovo for the consolidation period of 20152017, tax revenue amounted to 25.3% of gdp. this figure indicates stability in revenue mobilization efforts and suggests sustained fiscal discipline and economic management during the specified timeframe. montenegro during the consolidation period of 2017-2019, recorded tax revenue at 23.5% of gdp. this data underscores the country's ability to maintain adequate revenue levels amid economic adjustments or reforms implemented during this period. in the consolidation periods of 2012-2014 and 2017-2019, north macedonia reported tax revenue of 17.9% of gdp. this figure shows a marginal decrease compared to the pre-consolidation periods, reflecting potential challenges in maintaining revenue collection amidst economic fluctuations or policy adjustments. as for the consolidation periods of 2014-2016 and 2017-2019, serbia's tax revenue constituted 20.5% of gdp. this data highlights serbia's fiscal stability and the impact of tax policy reforms or economic conditions on revenue collection during the specified periods. 154 kamberi and bexheti table 5. changes in tax revenue (% of gdp) during consolidation (western balkan countries). country percentage change albania 7.14% (increase) bosnia and herzegovina 13.91% (increase) kosovo -1.98% (decrease) montenegro 11.06% (increase) north macedonia 3.35% (increase) serbia 3.90% (increase) source: author’s findings and calculations based on imf, world economic outlook databases table 5 presents the percentage changes in tax revenue as a percentage of gdp for several western balkan countries during their respective consolidation periods. bosnia and herzegovina experienced a significant increase of 13.91% in tax revenue as a percentage of gdp during its consolidation periods of 2012-2014 and 2016-2017. this substantial rise indicates effective fiscal reforms or enhanced tax administration strategies that bolstered revenue collection relative to economic output. the marked improvement underscores bosnia and herzegovina's commitment to fiscal consolidation and economic stability during these pivotal periods. montenegro also demonstrated notable growth in tax revenue, showing an 11.06% increase during the consolidation period of 2017-2019. this upward trend suggests successful fiscal management and policy adjustments that contributed to higher revenue mobilization relative to gdp. montenegro's ability to sustain and enhance tax revenue highlights its proactive approach to economic governance and fiscal sustainability in a changing economic landscape. albania, despite a modest increase of 7.14% in tax revenue as a percentage of gdp during its consolidation periods of 2014-2016 and 2018-2019, showed incremental progress in fiscal performance. this suggests ongoing efforts to strengthen tax policies or improve revenue collection efficiency amidst economic challenges or reforms. albania's consistent but moderate growth reflects its cautious approach to fiscal management and economic stabilization efforts during the specified periods. conversely, kosovo experienced a slight decrease of -1.98% in tax revenue as a percentage of gdp during its consolidation period of 2015-2017. this decline highlights potential challenges in sustaining revenue collection amidst economic fluctuations or evolving fiscal priorities. kosovo's experience underscores the importance of adaptive fiscal policies and economic resilience in maintaining stable revenue streams relative to economic activity. north macedonia and serbia both demonstrated positive growth in tax revenue as a percentage of gdp, with increases of 3.35% and 3.90%, respectively, during their consolidation periods. these moderate yet positive changes indicate steady progress in fiscal consolidation and economic recovery efforts. north macedonia and serbia's strategies likely focused on improving tax administration efficiency and optimizing fiscal policies to enhance revenue generation while supporting economic growth. public debt and fiscal consolidation in the western balkans table 6. expenditures as a percentage of gdp (pre-consolidation) in western balkan countries. country pre-consolidation period expenditures (% of gdp) albania before 2014-2016 & 2018-2019 29.8 bosnia and herzegovina before 2012-2014 & 2016-2017 44.1 kosovo before 2015-2017 32.7 montenegro before 2017-2019 41.5 north macedonia before 2012-2014 & 2017-2019 32.5 serbia before 2014-2016 & 2017-2019 39.6 source: author’s findings and calculations based on imf, world economic outlook databases table 6 reveals varying levels of government expenditures relative to gdp across the western balkan countries before their consolidation periods. higher percentages, such as those in bosnia and herzegovina and montenegro, suggest greater fiscal challenges that consolidation efforts aimed to mitigate. in the case of albania, before the consolidation periods (2014-2016 and 2018-2019), government expenditures accounted for approximately 29.8% of gdp. this indicates a moderate level of government spending relative to the size of the economy. bosnia and herzegovina prior to its consolidation periods (2012-2014 and 2016-2017), government expenditures were about 44.1% of gdp. this high level of spending suggests significant fiscal challenges that consolidation efforts aimed to address. kosovo before its consolidation period of 2015-2017, government expenditures were 32.7% of gdp. this reflects a moderate level of spending contextually considered during fiscal consolidation. before the consolidation period of 2017-2019, montenegro’s government expenditures were approximately 41.5% of gdp. this high expenditure level highlights the need for fiscal consolidation to manage spending and improve sustainability. as for the north macedonia before its consolidation periods (2012-2014 and 2017-2019), government expenditures accounted for about 32.5% of gdp. this moderate level of spending underscores the context in which fiscal consolidation efforts were undertaken. before its consolidation periods (2014-2016 and 20172019), serbia’s government expenditures were about 39.6% of gdp. this indicates significant government spending relative to gdp, emphasizing the importance of consolidation measures for fiscal stability. table 7. government expenditures as a percentage of gdp (during consolidation) in western balkan countries country consolidation periods expenditures (% of gdp) albania 2014-2016 & 2018-2019 29.1 bosnia and herzegovina 2012-2014 & 2016-2017 42.3 kosovo 2015-2017 30.2 montenegro 2017-2019 41.3 north macedonia 2012-2014 & 2017-2019 32.2 serbia 2014-2016 & 2017-2019 38.2 source: author’s findings and calculations based on imf, world economic outlook databases 156 kamberi and bexheti table 7 reveals varying levels of government expenditures relative to gdp across the western balkan countries during their consolidation periods. during the consolidation periods of 2014-2016 and 2018-2019, albania's government expenditures accounted for approximately 29.1% of gdp. this suggests a moderate level of government spending relative to economic output during the years focused on fiscal consolidation. bosnia and herzegovina recorded government expenditures equivalent to 42.3% of gdp during its consolidation periods of 2012-2014 and 20162017. this indicates a high level of spending relative to the size of its economy, highlighting significant fiscal challenges and the scale of efforts required for consolidation. kosovo's government expenditures during the consolidation years of 2015-2017 were approximately 30.2% of gdp. this reflects efforts to manage government spending while consolidating fiscal policies to enhance economic stability. montenegro's government expenditures were about 41.3% of gdp during the consolidation period of 2017-2019. this high expenditure level indicates substantial fiscal commitments during consolidation, aiming to address fiscal imbalances and improve sustainability. during its consolidation periods of 2012-2014 and 2017-2019, north macedonia's government expenditures accounted for approximately 32.2% of gdp. this moderate level of spending suggests efforts to align expenditures with economic growth and fiscal sustainability goals. serbia recorded government expenditures equivalent to 38.2% of gdp during its consolidation periods of 2014-2016 and 2017-2019. this indicates significant government spending relative to gdp, underscoring the challenges and objectives of fiscal consolidation efforts. table 8. changes in government expenditures (% of gdp) during consolidation (western balkan countries). country expenditure (% of gdp) change from pre-consolidation percentage change albania 29.1% -0.7% -2.35% bosnia and herzegovina 42.3% -1.8% -4.08% kosovo 30.2% -2.5% -7.63% montenegro 41.3% -0.2% -0.48% north macedonia 32.2% -0.3% -0.92% serbia 38.2% -1.4% -3.54% source: author’s findings and calculations based on imf, world economic outlook databases table 8 revels the changes in government expenditures (% of gdp) during consolidation (western balkan countries). albania experienced a moderate decrease of 0.7 percentage points in government expenditure as a percentage of gdp during its consolidation periods (2014-2016 & 2018-2019). this decrease reflects efforts to tighten fiscal policies and improve fiscal sustainability, resulting in a 2.35% reduction relative to the pre-consolidation period. bosnia and herzegovina recorded a significant decrease of 1.8 percentage points in government expenditure as a percentage of gdp during its consolidation periods (2012-2014 & 2016-2017). this substantial reduction of 4.08% reflects rigorous fiscal consolidation efforts aimed at addressing high fiscal deficits and improving economic stability. kosovo saw a notable decrease of 2.5 percentage points in government expenditure as a percentage of gdp during its consolidation period (2015-2017). this significant public debt and fiscal consolidation in the western balkans reduction of 7.63% underscores effective fiscal management efforts to align expenditures with economic growth and enhance fiscal sustainability. montenegro experienced a minor decrease of 0.2 percentage points in government expenditure as a percentage of gdp during its consolidation period (2017-2019). this slight reduction of 0.48% indicates efforts to maintain fiscal discipline and manage expenditures amidst consolidation efforts. north macedonia recorded a slight decrease of 0.3 percentage points in government expenditure as a percentage of gdp during its consolidation periods (2012-2014 & 2017-2019). this modest reduction of 0.92% reflects balanced fiscal policies aimed at sustaining economic growth while improving fiscal stability. serbia saw a decrease of 1.4 percentage points in government expenditure as a percentage of gdp during its consolidation periods (2014-2016 & 2017-2019). this reduction of 3.54% indicates significant efforts to control government spending and enhance fiscal resilience amidst consolidation initiatives. 4.1. descriptive statistics this analysis investigates the relationship between public debt and economic growth across a span of 23 years (2000-2023) in six western balkan countries. the study incorporates controls for inflation and unemployment to offer a thorough examination of how public debt influences economic growth in this region. table 9. descriptive statistics. variable obs mean std. dev. min max gdp_growth 143 3.365273 3.34405 -15.307 13.043 public_debt 136 46.20361 24.82498 5.569 213.781 inflation 143 4.894825 9.721578 -2.419 80.744 unemployment 133 22.78814 10.02008 9.396 57 gov_expend 142 36.69305 8.706383 8.506 56.306 source: author’s findings and calculations based on imf, world economic outlook databases the average gdp growth rate over the period is 3.37%, indicating moderate economic growth. however, the large standard deviation suggests significant variability in growth rates, with a sharp contraction of -15.31% at the minimum and a robust growth of 13.04% at the maximum. public debt averages at 46.20% of gdp across the wb6 countries, but this figure masks significant inter-country variation and differences in debt sustainability thresholds. for instance, while serbia and montenegro may manage higher debt ratios due to more diversified economies and stronger institutional frameworks, smaller economies like kosovo or north macedonia may face greater challenges in servicing even moderate debt levels. while international norms often consider public debt levels below 60% of gdp to be sustainable, country-specific research shows that optimal debt thresholds vary significantly across transition economies. for north macedonia, bexheti et al. (2020) estimate an optimal public debt threshold in the range of 52–55% of gdp, beyond which additional debt may hinder economic growth. this reflects the specific economic and institutional context of small transition economies, where limited absorptive capacity and market depth may amplify debt risks. thus, applying generalized benchmarks may obscure important national vulnerabilities, and fiscal targets should reflect local economic realities. 158 kamberi and bexheti additionally, the regional context—characterized by transitional economies, limited capital markets, and ongoing eu convergence efforts—amplifies the fiscal vulnerability tied to rising debt levels. these structural and institutional factors must be considered alongside numerical debt thresholds when evaluating sustainability in the western balkans. the minimum value of 5.57% reflects instances of very low public debt, while the maximum of 213.78% suggests cases of extremely high indebtedness, which could pose sustainability concerns. with an average inflation rate of 4.89%, the region has experienced moderate inflation. the high standard deviation points to considerable fluctuations, including deflationary periods (minimum -2.42%) and episodes of very high inflation (maximum 80.74%), which could indicate economic instability or external price shocks. the average unemployment rate is high at 22.79%, reflecting significant labor market challenges in the region. the variability is also notable, with the lowest unemployment at 9.40% and the highest at 57.00%. persistent high unemployment rates can hinder economic growth and social stability. government expenditure averages 36.69% of gdp, indicating a substantial role of the public sector in the economy. the range is wide, from a low of 8.51% to a high of 56.31%, suggesting differing levels of government involvement in economic activities across the countries and over time. 4.2. correlation matrix the correlation matrix provides insights into the relationships between the key variables in our study: gdp growth, public debt, inflation, unemployment and government expenditure. table 10. correlation matrix with p-values. variables gdp growth public debt inflation unemployment gov. expend gdp growth 1.000 -0.010 -0.029 0.003 -0.025 p-value — 0.911 0.745 0.977 0.782 public debt — 1.000 -0.121 0.162 0.011 p-value — 0.177 0.070 0.903 inflation — — 1.000 0.160 -0.047 p-value — 0.073 0.599 unemployment — — — 1.000 0.012 p-value — 0.898 gov. expend — — — — 1.000 p-value — source: author’s findings and calculations based on imf, world economic outlook databases government expenditure (gov_expend): the correlation between government expenditure and gdp growth is moderately negative (r = -0.2468). while this may appear counterintuitive, it likely reflects the countercyclical nature of fiscal policy. during economic downturns, when gdp growth is already low or negative, governments tend to increase spending to stabilize the economy. thus, this relationship may indicate timing effects rather than inefficiency — spending increases in response to economic contraction, not as a cause of it. public debt (public_debt): the negative correlation between public debt and public debt and fiscal consolidation in the western balkans unemployment (r = -0.162) — although marginally significant — suggests that government borrowing may have supported job-creating activities. this aligns with the use of stimulative fiscal measures, such as public investment or social transfers, to boost employment. hence, debt-financed spending may play a developmental and stabilizing role, particularly during times of crisis or structural transition. these results are not contradictory. rather, they reveal the multifaceted function of fiscal policy in the western balkans — where both spending and borrowing are used as tools to counteract economic volatility and foster long-term development. the correlation matrix offers preliminary insights into these dynamics, but it is essential to rely on regression analysis for causal interpretation. implications (i) unemployment and economic growth: the strong negative association between unemployment and gdp growth underscores the urgency of reducing unemployment as a key growth driver in the region. (ii) public debt and inflation: the moderate positive relationship between public debt and inflation indicates the importance of managing debt levels to preserve price stability. (iii) public debt and unemployment: the observed (though weak) negative correlation suggests a role for debt in employment support, particularly during economic downturns. (iv) government expenditure: mixed and generally weak correlations emphasize the need for targeted, efficient spending, rather than across-the-board increases or reductions, to support growth and macroeconomic stability. 4.3. regression analysis results to examine the impact of public debt and fiscal strategies on economic growth in the western balkan countries (wb6), we estimated three panel regression models: pooled ols, fixed effects (fe), and random effects (re). the regression model focuses on structural determinants of economic growth and excludes dynamic elements such as lagged dependent variables or country-specific time trends. year fixed effects were not included to preserve degrees of freedom, given the limited panel size. however, we acknowledge that including time fixed effects or dynamic specifications may reveal further nuances, and we recommend this as an extension in future research. public debt across all models, public debt as a percentage of gdp is negatively associated with gdp growth. the effect is statistically significant in the fixed effects model (p < 0.05), supporting the hypothesis that high public debt burdens constrain economic growth in the wb6. similar negative associations between debt and growth have been observed in reinhart and rogoff (2010), though recent debates emphasize context-dependency (panizza and presbitero, 2013). 160 kamberi and bexheti table 10. regression analysis results. variable pooled ols coefficients fixed effects coefficients random effects coefficients public debt -0.0321 (0.0170) -0.0526 (0.0197)** -0.0321 (0.0170) inflation 0.0828 (0.0354)** 0.0711 (0.0411) 0.0828 (0.0354)** unemployment -0.0458 (0.0433) 0.0041 (0.0492) -0.0458 (0.0433) gov. expenditure -0.1032 (0.0388)*** -0.3579 (0.0947)*** -0.1032 (0.0388)*** constant 9.3276 (2.1762)*** 18.8246 (4.0566)*** 9.3276 (2.1762)*** r-squared: 0.1073 (pooled ols), 0.1777 (fe within), 0.1157 (re within) observations: 127 hausman test: χ² (4) = 10.71, p = 0.0300 → fe model preferred breusch-pagan test: χ² (1) = 0.00, p = 1.000 government expenditure (proxy for fiscal consolidation) government expenditure, used here as a proxy for fiscal consolidation, is negatively and significantly associated with gdp growth in all models, particularly under fixed effects. this suggests that, in the wb6 context, reductions in government spending may be associated with improved economic outcomes— potentially due to enhanced investor confidence and restored fiscal discipline. the case of north macedonia provides further nuance. as bexheti et al. (2021) argue, much of the country’s public expenditure is directed toward unproductive categories, such as recurrent administrative costs and inefficient subsidies, which are unlikely to yield growth-enhancing effects. in contrast, investment in infrastructure, education, and innovation tends to generate long-term economic gains. thus, the negative relationship observed may reflect spending inefficiency and composition, rather than the volume of expenditure per se. this finding partially aligns with alesina and ardagna (2010), who argue that spendingbased fiscal consolidations—particularly those that cut unproductive expenditures—can be less detrimental to growth than tax-based adjustments. however, it contrasts with blanchard and leigh (2013), who find that fiscal multipliers in european economies were generally positive during downturns, suggesting that spending cuts in such periods may hinder recovery. control variables (i) inflation: a modest positive effect, significant in pooled ols and re, indicating mild inflation may accompany growth. (ii) unemployment: negative but not statistically significant, though its inclusion improves model fit and controls for labor market conditions. public debt and fiscal consolidation in the western balkans 4.4. regression diagnostics to ensure that the regression results are statistically valid and can serve as a reliable basis for policy recommendations, we conducted the following diagnostic tests. table 11. multicollinearity – variance inflation factors (vif). variable vif public debt 1.02 inflation 1.03 unemployment 1.01 gov. expenditure 1.02 all vif values are far below the critical threshold of 5, indicating that multicollinearity is not present and the variables are not linearly dependent on each other. table 12. heteroskedasticity – breusch-pagan test test component value lagrange multiplier statistic 5.65 p-value 0.130 f-value 1.39 f p-value 0.238 the p-values exceed the 0.05 threshold, so we fail to reject the null hypothesis of homoskedasticity. this indicates constant variance in the residuals, validating the standard error estimates. all diagnostic tests confirm the statistical adequacy of the fixed effects regression model. there is no multicollinearity, heteroskedasticity, or autocorrelation detected. consequently, the results presented can be considered robust and form a credible empirical foundation for the policy recommendations proposed in the next section. 5. conclusion this study contributes to the empirical literature on the macroeconomic consequences of public debt and fiscal consolidation in transitional economies by examining the western balkans (wb6) over the period 2000–2023. through a rigorous panel data approach utilizing fixed effects estimation— validated by robust diagnostic testing—this research provides clear evidence in support of both central hypotheses: (i) elevated public debt levels negatively affect economic growth in the wb6, and (ii) expenditure-based fiscal consolidation contributes positively to growth performance. the analysis demonstrates that public debt, when exceeding sustainable thresholds, exerts a statistically and economically significant drag on gdp growth. this effect is most pronounced in economies characterized by limited institutional capacity and shallow capital markets—conditions prevalent across the wb6. importantly, the negative debt-growth relationship persists even after controlling for inflation, unemployment, and government expenditure, underscoring the structural vulnerability associated with excessive sovereign indebtedness. 162 kamberi and bexheti complementing this finding, the study shows that fiscal consolidation, proxied by reductions in government expenditure relative to gdp, is associated with improved growth outcomes, particularly under fixed effects estimation. the negative and significant coefficients on government spending in the growth regressions indicate that expenditure-based adjustments—rather than ad hoc revenue measures—are more conducive to restoring fiscal balance while supporting macroeconomic stability. this result aligns with established findings in the literature emphasizing the long-run benefits of credible and rule-based fiscal frameworks. moreover, the construction of the consolidation factor index (cfi) introduces a novel evaluative lens through which to assess the depth and effectiveness of consolidation episodes. countries with higher cfi scores—such as serbia and albania—exhibited more successful consolidation trajectories, marked by durable improvements in debt dynamics and sustained growth. conversely, weaker performers reflected either insufficient fiscal effort or a lack of accompanying structural reform, reinforcing the centrality of policy design and implementation capacity. the broader policy implication is clear: fiscal consolidation, when anchored in institutional reform and expenditure rationalization, can coexist with—if not facilitate—economic growth. for the wb6 countries, where fiscal space remains constrained and external vulnerabilities persist, this strategy is not only desirable but necessary. as these countries advance toward european integration and seek to bolster macroeconomic credibility, the prioritization of sustainable debt paths and disciplined fiscal governance should be at the forefront of national policy agendas. future research should build on these findings by employing dynamic panel models or threshold regression techniques to capture potential non-linearities in the debt-growth relationship and assess state-contingent fiscal multipliers. moreover, incorporating political economy variables could yield insights into the conditions under which consolidation efforts are most likely to succeed. in sum, this study offers a robust empirical foundation for rethinking the fiscal-growth nexus in emerging european economies, with direct relevance to both policymakers and academic inquiry. references acemoglu, d., johnson, s., robinson, j. a., & thaicharoen, y. (2019). institutional causes, macroeconomic symptoms: volatility, crises and growth. journal of monetary economics, 56(4), 305–322. acemoglu, d., naidu, s., restrepo, p., & robinson, j. a. (2019). democracy does cause growth. journal of political economy, 127(1), 47–100. https://doi.org/10.1086/700936 alesina, a., & ardagna, s. (1998). tales of fiscal adjustment. economic policy, 13(27), 487–545. https://doi.org/10.1111/1468-0327.00039 alesina, a., & ardagna, s. (2010). large changes in fiscal policy: taxes versus spending. in tax policy and the economy (vol. 24, pp. 35–68). university of chicago press. https://doi.org/10.1086/649828 alesina, a., & perotti, r. (1994). the political economy of budget deficits. nber working paper no. 4637. https://www.nber.org/papers/w4637 https://doi.org/10.1086/700936 https://doi.org/10.1111/1468-0327.00039 https://doi.org/10.1086/649828 https://www.nber.org/papers/w4637 public debt and fiscal consolidation in the western balkans alesina, a., & perotti, r. (1999). fiscal adjustments in oecd countries: composition and macroeconomic effects. imf staff papers, 44(2), 210–248. https://doi.org/10.2307/3867543 bexheti, a., fetai, b., avdimetaj, k., & malaj, a. (2020). threshold effect of public debt on economic growth: an empirical analysis in the european transition countries. zbornik radova ekonomskog fakulteta u rijeci, 38(2), 381–406. https://doi.org/10.18045/zbefri.2020.2.381 bexheti, a., fetai, b., alili-sulejmani, l., & ramadani, v. (2021). does fiscal policy accelerate economic growth in transition economies? the case of north macedonia. international journal of public sector performance management, 7(1), 87–97. https://doi.org/10.1504/ijpspm.2021.111969 blanchard, o., & leigh, d. (2013). growth forecast errors and fiscal multipliers. american economic review, 103(3), 117–120. https://doi.org/10.1257/aer.103.3.117 blejer, m. i., & leone, a. j. (eds.). (2020). fiscal policy formulation and implementation in oil-producing countries. international monetary fund. cottarelli, c., & giannini, c. (2017). fiscal adjustment: a look at some recent international experiences. international monetary fund. european bank for reconstruction and development. (2021). transition report 2021/2022: system upgrade – delivering the digital dividend. ebrd. european commission. (2020). fiscal sustainability report 2020. european union. european commission. (2021). western balkans economic integration and enlargement: economic reforms and trade integration. european union. european investment bank. (2019). western balkans investment framework: annual report. european investment bank. international monetary fund. (2019). fiscal monitor: public debt in western balkans. imf. international monetary fund. (2022). western balkans: economic outlook and fiscal sustainability. imf. marinov, v., christov, g., & arevyan, a. (2017). fiscal discipline in the western balkans. economic policy review, 29(2), 87–102. milosavljevic, m. (2018). fiscal consolidation and economic growth in serbia: challenges and prospects. serbian journal of management, 13(2), 245–264. mirdala, r. (2013). fiscal imbalances and financial instability in the post-transition period: case of slovakia. procedia economics and finance, 5, 737–746. panizza, u., & presbitero, a. f. (2013). public debt and economic growth in advanced economies: a survey. swiss journal of economics and statistics, 149(2), 175–204. https://doi.org/10.1007/bf03399388 reinhart, c. m., & rogoff, k. s. (2010). growth in a time of debt. american economic review, 100(2), 573–578. https://doi.org/10.1257/aer.100.2.573 regional cooperation council. (2020). western balkans: economic and social analysis. regional cooperation council. von hagen, j., & harden, i. (1995). budget processes and commitment to fiscal discipline. european economic review, 39(3–4), 771–779. https://doi.org/10.1016/0014-2921(94)00084-d wyplosz, c. (2013). fiscal rules: theoretical issues and historical experiences. in a. alesina & f. giavazzi (eds.), fiscal policy after the financial crisis (pp. 495–525). university of chicago press. https://doi.org/10.7208/chicago/9780226018584.003.0013 world bank. (2020). western balkans public finance review. world bank. world bank. (2021). western balkans: economic challenges and prospects. world bank. world bank. (2023). fiscal adjustment in transitional economies. world bank. https://doi.org/10.2307/3867543 https://doi.org/10.18045/zbefri.2020.2.381 https://doi.org/10.1504/ijpspm.2021.111969 https://doi.org/10.1257/aer.103.3.117 https://doi.org/10.1007/bf03399388 https://doi.org/10.1257/aer.100.2.573 https://doi.org/10.1016/0014-2921(94)00084-d https://doi.org/10.7208/chicago/9780226018584.003.0013 1. introduction 2. literature review 3. methodology 4. estimations and results 5. conclusion references © the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 2 (2025), pages 185-218 https://doi.org/10.17979/ejge.2025.14.2.11713 submitted: feb 4, 2025 accepted: jun 26, 2025 published: dec 17, 2025 article management impression and bank’s performance with nlp for chairperson’s statement tam phan huy1*, quan tran bich1, vang nguyen thi mai1, vy do tuong1, linh nguyen thi khanh1 1 university of economics and law and vietnam national university, ho chi minh city, vietnam *correspondence: tamph@uel.edu.vn abstract. impression management within chairperson’s statements is a prominent area of study, particularly in the banking sector. this research seeks to understand the nuanced strategies banks deploy in these statements to shape stakeholder perceptions regarding their financial performance. utilizing a mixed-method approach that combines qualitative content analysis with algorithmic techniques, the study applies natural language processing (nlp) including sentiment analysis, topic modeling, word vectorization, and readability scoring to systematically examine chairperson’s statements, using data from 2012 to 2022 of commercial banks in vietnam. the results revealed that banks facing financial adversities tend to craft strategic narratives to underscore their resilience and adaptability. crucially, elements such as public visibility and consumer proximity emerged as dominant factors influencing the direction and tone of these narratives. the study underscores the pivotal role of chairperson’s statements in molding and upholding a bank's image. the effectiveness of these statements is contingent upon their alignment with stakeholder expectations and the prevailing market dynamics, providing invaluable insights for investors, bank executives, and regulatory institutions. keywords: impression management; chairperson’s statement; textual analysis; vietnam jel classification: g21; m41; c80 1. introduction this study aims to investigate how impression management strategies embedded in chairperson’s statements affect bank performance, with particular attention to the role of chairperson characteristics in shaping narrative tone and content in the vietnamese banking sector. impression management, particularly as expressed in chairperson's statements within annual reports, serves as a focal point of corporate disclosure practices, potentially influencing stakeholders' perceptions of a firm's image and legitimacy. goffman (2016) posits that organizations, much like individuals, engage in self-presentation to cultivate and maintain favorable images to their audiences. such practices in the corporate world are critical, as the narratives and tones set in chairperson's statements can sway investors, consumers, employees, and other key stakeholders (giglioni, 2017; li et al., 2019). https://creativecommons.org/licenses/by-nc/4.0/ https://doi.org/10.17979/ejge.2025.14.2.11713 186 tam phan huy et al. the chairperson's statement not only communicates financial performances but also offers insights into the company's values, strategic orientations, and future outlook (merkl-davies & brennan, 2007). scholars such as oliveira et al. (2016) suggest that these statements can serve as a tool for managing and even manipulating perceptions, especially during financially challenging times or corporate crises. furthermore, the significance of impression management intensifies in a globalized business landscape where companies must navigate diverse cultural and regulatory environments (hooghiemstra, 2000). given the importance of such disclosures, it becomes imperative for researchers to scrutinize the motives and strategies behind them, ensuring transparency and accountability in corporate communications. while some researchers have delved into specific geographical contexts, such as the u.s (li et al., 2019; smith & taffler, 1992), malaysia (al-sayani et al., 2020), and portugal (oliveira et al., 2016), a comparative analysis in the south african context (dhludhlu, 2022; phesa, 2021) brought into sharp focus the role of impression management in shaping voluntary disclosures, especially during unprecedented challenges like the covid-19 pandemic. interestingly, the research indicates that companies, whether profitable or unprofitable, have continued to engage in impression management, even in the face of a global crisis. the contemporary landscape of impression management in chairperson's statements paints a picture of strategic communication practices deeply intertwined with the broader corporate governance and disclosure frameworks. recent revelations from studies conducted in hong kong (li et al., 2019) and egypt (zerban & ateia, 2016) stress the significance of chairperson's tone and corporate governance structures in shaping stakeholders' perceptions and subsequent decisionmaking processes. despite the wealth of studies across various global markets, one notices a gaping lacuna when it comes to the vietnam market. diverging from the trajectory of established studies, vietnam remains a relatively unexplored terrain in terms of research on impression management in chairperson's statements. given vietnam's burgeoning economy, distinct corporate culture, and evolving regulatory framework, understanding impression management practices in this context can provide unparalleled insights into the intersection of corporate communication strategies and socioeconomic dynamics in emerging markets. specifically, these insights can guide banks in refining their narrative disclosures to align more closely with stakeholders' expectations, enhance transparency, and build investor trust. for instance, by identifying prevalent im tactics, banks can develop training programs for executives to promote more authentic and balanced communication. regulators can also utilize these findings to establish clearer guidelines for narrative reporting, ensuring consistency and reducing the potential for misleading disclosures. moreover, investors and analysts can benefit from frameworks developed through this research to critically assess the tone and content of chairperson’s statements, leading to more informed decision-making. given the profound implications these practices might have on investor trust, stakeholder engagement, and overall corporate transparency in the vietnamese context, this research stands poised to make a significant contribution to both academic discourse and practical applications in the realm of corporate communication. while prior studies have explored impression management in diverse markets such as the u.s. (li et al., 2019; smith & taffler, 1992), malaysia (al-sayani et al., 2020), and portugal (oliveira et management impression and bank’s performance with nlp for chairperson’s statement al., 2016), these economies typically exhibit more mature regulatory environments, higher levels of corporate transparency, and broader adoption of international reporting standards. in contrast, vietnam represents a transitional and emerging economy, characterized by evolving corporate governance frameworks, a less mature capital market, and high levels of state influence in the banking sector. these distinctions are critical, as they may affect both the motivations and techniques used in impression management. for instance, while impression management in developed markets often focuses on shareholder reassurance during earnings volatility, vietnamese bank chairmen may prioritize legitimacy and stability narratives to navigate regulatory scrutiny and stakeholder trust. this contextual divergence reinforces the relevance of examining vietnam as a unique case within the broader literature on corporate narrative strategies. in conclusion, while impression management in chairperson's statements has been studied extensively across different global contexts, the vietnamese market remains largely untouched. this research seeks to bridge that gap, shedding light on a critical aspect of corporate disclosure practices in a rapidly evolving economic landscape. by delving into this uncharted territory, the study aims to offer significant insights that could reshape our understanding of impression management in emerging markets. this study seeks to answer the central research question: how do impression management strategies embedded in chairperson’s statements influence bank performance in the vietnamese banking sector? the banking industry is a particularly relevant context because it is a highly regulated, trust-based sector, where corporate narratives play a critical role in maintaining stakeholder confidence and managing perceptions of stability and risk, making it especially susceptible to impression management. this research is structured into 5 parts: (i) introduction, (ii) literature review, (iii) methodology, (iv) results and discussion, and (v) conclusion and recommendation. 2. literature review 2.1 concepts and measurements 2.1.1 impression management impression management, as a concept, can be traced back to sociologist erving goffman's seminal work in the late 1950s. goffman (1949) describes it as the process through which individuals attempt to influence the perceptions of other people about a person, object, or event. they do so by regulating and controlling information during social interactions. in the context of corporate reporting, this concept has evolved to describe how companies may attempt to shape investors' and stakeholders' perceptions of their performance and prospects (merkldavies & brennan, 2007). from a socio-psychological perspective, impression management is deeply entrenched in individuals' innate inclination to present themselves in a positive light to their peers and the broader society. this urge is not merely superficial; it's intertwined with our self-identity, social 188 tam phan huy et al. integration, and the roles we play within our communities. this intrinsic need can act as a potent driver, propelling individuals to act in ways that might either bolster a desired image or avert any potential negative judgments from others. leary and kowalski (1990) delved deeper into this intricate web of self-presentation and societal expectations by presenting the twocomponent model of impression management. the model bifurcates the concept into "impression motivation" and "impression construction." the former revolves around the 'why' – why individuals might feel the need to manage perceptions. for instance, they might be more motivated to shape impressions when they perceive these impressions as crucial for achieving their personal or professional objectives. another compelling driver for impression motivation is the individual's belief in their capability to effectively manipulate these perceptions to their advantage. on the other hand, "impression construction" touches upon the 'how'. it's about the tactics and strategies individuals deploy to create the desired impression. this could range from verbal cues, such as choosing one's words carefully, to non-verbal cues, including body language and attire. the choices made during the impression construction phase often draw from an individual's goals (whether they're looking for acceptance, admiration, or authority) and their perceived social norms and expectations. but beyond the mechanics of this model, what stands out is the intricate balance individuals must strike. on one hand, there's the genuine desire for authenticity and presenting one's true self. on the other, there's the push and pull of societal expectations, peer pressures, and individual aspirations. navigating this delicate balance becomes an ongoing journey for many, as they calibrate their impression management strategies in response to ever-evolving personal and societal dynamics. switching to an organizational context, organizations are perceived as collective entities that also engage in behavior aimed at shaping stakeholder perceptions. here, impression management can manifest in various forms, from the design of annual reports to public relations campaigns. such actions aim to showcase the organization's best aspects while potentially downplaying or obscuring less favorable information (hooghiemstra, 2000). studies like those by clatworthy and jones (2001) have further examined this from the lens of financial reporting, asserting that firms might intentionally adopt certain linguistic or numerical strategies in their disclosures to influence stakeholder perceptions. furthermore, the critical perspective on impression management is anchored in the belief that it represents a deliberate distortion of reality, wherein organizations actively deploy tactics to manipulate stakeholders' perceptions. critics, firmly rooted in this view, argue that the strategic use of impression management goes beyond mere presentation or framing. instead, it ventures into the realm of deception, painting an unrealistically favorable picture of an organization's performance, actions, or intentions (neu et al., 1998). several scholars have explored this dimension, raising concerns about the underlying motives and the potential implications of such practices. for instance, bowler and woehr (2006) suggest that impression management tactics, especially when excessively or inappropriately used, can blur the line between strategic communication and misinformation. such practices can lead to skewed decision-making, particularly when stakeholders, such as investors or employees, base their decisions on this altered reality. furthermore, the ethical dimension of management impression and bank’s performance with nlp for chairperson’s statement impression management cannot be overlooked. solomon (2020) delve into the moral implications, noting that intentionally misleading stakeholders, whether by omission or commission, raises serious ethical questions. the core issue here revolves around the principles of truthfulness, fairness, and transparency, which are often compromised when organizations indulge in aggressive impression management strategies. a counterargument, as presented by bolino et al. (2008), posits that not all forms of impression management are inherently deceptive. they argue that some tactics might merely involve presenting information in the best possible light without falsifying or hiding the truth. however, distinguishing between these benign forms and their more manipulative counterparts remains a challenge, further complicating the discourse. so, impression management, as a multi-faceted concept, offers various interpretations and nuances depending on the perspective adopted. impression management, as understood from a socio-psychological lens, is seen as an innate human drive influenced by societal norms and personal aspirations, often emerging as natural responses to social stimuli (leary & kowalski, 1990). in contrast, organizations adopt a more strategic stance, using impression management to shape perceptions in line with their objectives, focusing on tangible goals like attracting investments or fostering loyalty (higgins & judge, 2004). however, this tactical maneuvering is not without its critics, who caution against the potential ethical pitfalls, suggesting that organizations might occasionally distort realities for favorable optics (neu et al., 1998). these differing viewpoints set the stage for a deeper debate. while the sociopsychological stance could argue that organizations, being human-led entities, merely amplify natural human tendencies at a larger scale, the critical perspective would counter that the stakes in organizational contexts are significantly higher. after all, misleading a vast array of stakeholders can have profound financial, social, and even environmental consequences. when weaving together these views, it's clear that impression management acts as a two-sided coin. it holds the power to craft identities and steer through social and corporate terrains, yet it also brings inherent challenges. the line between tactical portrayal and intentional misinformation can sometimes be thin, pushing organizations to find a middle ground that maintains honesty while pursuing their objectives. in summary, impression management, as reviewed from various perspectives, is intricately tied to the language and narratives employed by individuals and organizations. the socio-psychological viewpoint emphasizes the intrinsic human behavior to shape perceptions, while the organizational perspective underscores the strategic language use to align with business objectives. the critical view further highlights the potential nuances and subtleties in language that may be employed to toe the line between ethical communication and deception. so, natural language processing (nlp) emerges as a particularly suitable method for extracting information and nuances from the chairperson’s statements in annual reports. nlp, with its ability to decipher textual data, can penetrate beyond the surface level of statements, capturing underlying sentiments, tonal shifts, and rhetorical strategies that might be indicative of impression management tactics. given that impression management is deeply rooted in the language and narratives employed, an analytical approach like nlp that specializes in textual 190 tam phan huy et al. analysis can provide deeper insights and more objective measures of management impressions as conveyed in chairperson's statements. furthermore, as organizations become more sophisticated in their communication strategies, leveraging advanced methodologies like nlp ensures that researchers and stakeholders stay one step ahead, capturing the full spectrum of impression management cues embedded within corporate narratives. this is consistent with prior studies that demonstrate the effectiveness of nlp techniques in extracting sentiment, tone, and strategic content from corporate narratives (henry & leone, 2016; li, 2010; loughran & mcdonald, 2011). nlp-based approaches have been increasingly adopted in accounting and finance research to analyze narrative disclosures and detect impression management tactics (brown & tucker, 2011; huang et al., 2014). 2.1.2 bank’s performance bank performance assessment has continually occupied a prominent space in both academic discourses and industry analytics, given its paramount significance in maintaining financial stability and fostering economic development. historically, a substantial segment of research has anchored bank performance measurement on traditional financial ratios. bourke (1989) emphasized the weightage of financial metrics, specifically mentioning the return on assets (roa), return on equity (roe), and the net interest margin (nim) as critical indicators of a bank's profitability and its efficiency in resource deployment. recent studies continue to uphold this perspective, with nim consistently recognized as a core indicator of bank profitability and efficiency, especially in emerging markets (dietrich & wanzenried, 2014; petria et al., 2015). delving into the undercurrents of the myriad performance metrics used in banking, the net interest margin (nim) undeniably occupies a pivotal position, especially within narrative disclosures such as chairperson's statements. barth et al. (2004) elucidated the value of riskadjusted metrics, and within this framework, nim emerges as a particularly illuminating indicator. this sentiment is echoed by molyneux and thornton (1992), who posit that nim provides an intricate view into a bank's operational efficiency, risk appetite, and overall management proficiency. more recent empirical studies reinforce nim's value as a sensitive and dynamic measure of bank performance, capturing managerial decisions on pricing, funding costs, and risk-taking (claessens et al., 2018; sahul hamid, 2017). fundamentally, nim, which represents the difference between interest incomes earned and interest expended expressed as a percentage of interest-earning assets, encapsulates the essence of banking operations. demirgüç-kunt and huizinga (1999) further elaborate on its significance, highlighting that a higher nim often indicates a bank's capacity to manage its funds adeptly, ensure creditworthiness, and maintain a balance between risk and return. recent research highlights that nim remains particularly relevant in post-crisis regulatory environments and in the current era of compressed interest rates, where banks’ ability to maintain margins is closely scrutinized by investors and regulators (ratanavararak & ananchotikul, 2018). moreover, in emerging markets such as vietnam, nim is frequently used in both academic studies and industry practice as a transparent and comparable indicator of management impression and bank’s performance with nlp for chairperson’s statement bank performance (quoc trung, 2021). furthermore, saunders and schumacher (2000) point out the importance of analyzing nim in conjunction with other metrics to gain a comprehensive view of a bank's financial health. they emphasize that while nim offers insights into operational aspects, its interplay with metrics like return on assets or capital adequacy ratios can provide a holistic perspective on the bank's strategic positioning. given this multifaceted value, it comes as no surprise that nim garners pronounced emphasis in annual reports. berger and deyoung (1997) observed a positive correlation between nim and a bank's profitability, further underlining its importance in shaping stakeholder perceptions through chairperson's statements and other financial disclosures. in the vietnamese context, the banking sector has been a focal industry of both interest and intrigue, reflecting the nation's economic metamorphosis and quest for greater integration within global financial systems. indeed, over the last few decades, vietnam's banking landscape has undergone momentous changes, marked by liberalization efforts, technological adoptions, and increased foreign participation. coupled with these are the intense competitions among domestic banks and the continual adaptation to new and often more stringent regulatory frameworks, reflecting both domestic policy shifts and international banking standards. given these shifts, the role of chairperson's statements in annual reports has gained unprecedented significance in vietnam. these statements, in their narrative potency, serve as a beacon of clarity, guiding stakeholders through the often-complex terrains of banking operations, strategies, and performances. the chairperson's statements encapsulate, to a large extent, how a bank perceives its place in the financial ecosystem and its roadmap for the future, offering invaluable insights to a diverse set of stakeholders, from international investors assessing market entry strategies to local customers gauging the stability of their chosen financial institutions. the gravitas of metrics like the net interest margin (nim) within these narratives cannot be overstated, especially given its direct linkage to a bank's core operational efficiency and profitability (le & phan, 2017). considering the pivotal role that nim plays in elucidating a bank's core financial health, and its anticipated prominence in narrative disclosures such as chairperson's statements, it emerges as an apt proxy for bank performance in the context of vietnam's market. employing nim as the focal metric not only aligns with global best practices but also provides a nuanced understanding of bank profitability in a rapidly evolving vietnamese banking ecosystem. 2.1.3 corporate characteristics and disclosure of chairperson’s statements corporate characteristics such as firm size, ownership structure, and governance practices play a pivotal role in shaping the content and tone of chairperson’s statements. larger firms, often under greater public and regulatory scrutiny, are more likely to adopt professionalized communication strategies and emphasize performance consistency, stability, and forwardlooking narratives (brennan & merkl-davies, 2013). in contrast, smaller or less visible firms may 192 tam phan huy et al. prioritize legitimacy-building or reassurance narratives, particularly in the absence of strong financial results. firms with high leverage or weaker financial positions may also engage in more strategic impression management to mitigate stakeholder concerns (clatworthy & jones, 2001). empirical evidence from india provides additional perspective on the role of corporate characteristics in narrative disclosures. for example, singh and singla (2024) find that managers may strategically alter the tone and volume of md&a disclosures based on firm performance, supporting the use of a "management impression strategy" when performance is poor. similarly, singla and singh (2024) show that voluntary disclosures in md&a reports are more strongly influenced by profitability than by governance variables such as board size or the presence of independent directors, suggesting that favorable performance motivates firms to disclose more to reinforce a positive image. these studies reinforce the notion that performance-linked impression management is a cross-market phenomenon, even under mandatory reporting frameworks. additionally, state ownership as a common feature in many vietnamese banks can significantly affect disclosure content. state-affiliated banks may emphasize policy alignment, stability, and national development goals, which in turn influence narrative structure and tone. the presence of politically connected executives or government-appointed chairmen can further influence the strategic direction of disclosures, making them less market-driven and more aligned with regulatory or political expectations (nguyen & nielsen, 2010). regarding disclosure standards, vietnam follows the guidelines outlined in circular no. 155/2015/tt-btc issued by the ministry of finance, which mandates the structure and minimum disclosure requirements for annual reports of listed firms. while the chairperson’s statement is considered a narrative and voluntary disclosure section, it is expected to reflect the overall performance, challenges, and strategic outlook of the firm. however, no strict formatting or content rules exist, giving room for considerable discretion. this flexibility enables impression management practices, especially in markets with weaker enforcement mechanisms or lower levels of investor activism. in summary, the content and presentation of chairperson’s statements are shaped not only by individual chairmen but also by broader corporate characteristics and institutional contexts. understanding these factors is essential to interpret the motivations behind narrative disclosures and their implications for transparency and stakeholder perception. 2.2 background theories to frame the investigation of impression management within chairperson's statements and its relationship to bank performance, this chapter reviews several key theoretical perspectives and related empirical findings. the discussion integrates insights from impression management theory, narrative disclosure theory, signaling theory, and social constructivism, each of which offers a unique lens on how corporate narratives shape stakeholder perceptions and influence financial outcomes. these theories collectively inform the study’s conceptual framework, management impression and bank’s performance with nlp for chairperson’s statement guiding both the selection of variables and the interpretation of results. in addition, the chapter reviews contemporary approaches to measuring impression management and evaluating bank performance, with particular attention to their application in emerging market contexts such as vietnam. by weaving together these diverse strands of literature, the review establishes a solid foundation for the study’s analytical approach and contributes to positioning this research within the broader discourse on corporate communication and financial performance. 2.2.1 impression management theory impression management theory, deeply entrenched in the works of sociological and psychological scholars, serves as a pertinent lens through which corporate behaviors and communications can be evaluated. originating from goffman's dramaturgical perspective, the theory posits that individuals and organizations continuously perform, akin to actors on a stage, to construct and maintain a desired identity in front of their intended audience (goffman, 1949). this performance is characterized by the distinction between the "front stage" – the curated and polished version intended for the audience – and the "backstage" – where the raw, unfiltered preparations occur. in the realm of corporate disclosures, this performance is shaped by goffman’s well-known metaphor of the “front stage” — the curated and polished version intended for the audience — and the “backstage” — where the raw, unfiltered preparations occur. several scholars have ventured into the domain of corporate narratives, leveraging goffman’s impression management principles. for instance, merkl-davies et al. (2011) provide insights into how companies employ retrospective sense-making in their corporate narratives as tools for impression management, echoing goffman's assertions of rewriting and presenting events in a light that favors the firm's image. likewise, bowman (1984) assessed the strategic conveyance of corporate strategy and associated risks through annual reports, subtly drawing upon principles of impression management, even if not directly referencing goffman. in a similar vein, focusing on the environmental dimension, neu et al. (1998) argued for the strategic disclosure of environmental information in annual reports. such disclosures, they suggest, aim to control public perceptions regarding a company's environmental stewardship. the intricate dance between curated narratives and underlying realities becomes even more pronounced when examined against the backdrop of specific contexts. given the transformative landscape of corporate reporting, where narratives are often viewed as holding immense power in shaping investor sentiments and market behaviors, it becomes imperative to recognize and understand the nuances of impression management within these communications. at the same time, it is important to acknowledge that some scholars remain skeptical of the actual influence of such narratives, arguing that sophisticated investors may discount rhetorical devices and rely more heavily on quantitative fundamentals (merkl-davies & brennan, 2007; brennan et al., 2013). this study, therefore, contributes to this ongoing debate 194 tam phan huy et al. by empirically examining the extent to which impression management features in chairperson’s statements are associated with observable performance outcomes in the banking sector. the convergence of these scholarly works points to an undeniable fact: impression management is an intrinsic part of corporate communication. while the underlying motivations might differ from managing stakeholder perceptions (merkl-davies et al., 2011), to strategically conveying risks (bowman, 1984), or showcasing environmental responsibility (neu et al., 1998) the overarching theme remains consistent. companies, intentionally or not, resort to crafting narratives that serve their broader strategic objectives. yet, herein lies the crux: while these narratives can offer valuable insights, they also pose the risk of obfuscating the complete picture. as researchers dive deeper into understanding this delicate balance, tools like natural language processing (nlp) and content analysis can play pivotal roles in demystifying the intricacies of corporate narratives, ensuring transparency and credibility in corporate communications. 2.2.2 narrative disclosure theory the narrative disclosure theory underscores the increasing value attributed to non-quantitative forms of corporate communication, such as chairperson's statements. these narratives have become instrumental in supplementing and contextualizing the purely quantitative data found in annual reports. smith and taffler (2000) emphasize that while numerical financial data remains indispensable, it often falls short in conveying the entirety of an organization's landscape. this gap is bridged by narrative disclosures, which capture the sentiment, contextual undertones, and subtleties that numbers might overlook. similarly, brennan and merkl-davies (2013) postulate that these disclosures facilitate a dialogue, enabling corporations to communicate their stance, intentions, and reflections on various issues, thus fostering a deeper understanding and connection with their stakeholders. narrative disclosure theory has steadily garnered traction in the realm of corporate communications and financial reporting. the premise, hinging on the assertion that quantitative metrics alone cannot encapsulate the multifaceted nature of organizational performance and prospects, resonates profoundly with contemporary research paradigms, such as studies on the role of tone and sentiment in financial disclosures (henry & leone, 2016; loughran & mcdonald, 2011), investigations of narrative readability and complexity (li et al., 2019), and the increasing focus on narrative dimensions of esg and sustainability reporting (huang et al., 2014). chairperson's statements, serving as a cornerstone of these narrative disclosures, are often seen as providing a unique lens through which stakeholders can glean insights into an organization's strategic outlook, challenges, and sentiment. the application of the narrative disclosure theory within the domain of financial research offers a fresh perspective on corporate communications. beyond the traditional metrics and figures that dominate annual reports, chairperson's statements bring forth an management impression and bank’s performance with nlp for chairperson’s statement element of humanization, context, and strategic direction (holland, 2005). recent trends indicate a heightened interest among stakeholders in these narrative components, emphasizing the need for a more comprehensive evaluation framework (beattie et al., 2004). studies by van der laan smith et al. (2005) further corroborate this, noting that in markets characterized by information asymmetry, such as emerging economies, narrative disclosures could play a pivotal role in bridging the knowledge gap. as the relevance of narrative disclosures, particularly chairperson's statements, continues to rise in contexts where qualitative communication remains vital, such as in emerging markets, state-influenced sectors, and relationship-driven industries like banking, recent studies have underscored their ongoing importance for shaping stakeholder perceptions (dhludhlu, 2022; merkl-davies et al., 2011; oliveira et al., 2016). moreover, in periods of heightened uncertainty, such as the covid-19 pandemic, the strategic role of chairperson’s narratives in maintaining investor confidence has been particularly evident (dhludhlu, 2022; phesa, 2021). thus, while global trends toward standardized, data-driven reporting are advancing in some domains, narrative disclosures remain highly relevant in contexts such as the vietnamese banking sector, where transparency, trust, and relational signals conveyed through chairperson’s statements continue to influence market perceptions. 2.2.3 signaling theory signaling theory is anchored in the idea that when there is an imbalance of information between two parties, the better-informed party can convey or "signal" valuable information to the lesserinformed one, often to alleviate concerns or convey credibility (spence, 1978). within the context of corporate disclosures, this theory holds great significance, as companies constantly engage in signaling to manage stakeholders' perceptions and mitigate uncertainties. several scholars have delved into the realm of signaling within corporate disclosures. early foundational work by healy and palepu (2001) highlighted the critical role of financial disclosures in reducing information asymmetries between firms and their stakeholders. however, financial reporting standards and stakeholder expectations have evolved substantially over the past two decades, particularly with the rise of integrated reporting, realtime data dissemination, and esg accountability. more recently, connelly et al. (2011) explored how firms use voluntary disclosures as strategic signals to reduce uncertainties among investors, emphasizing that the quality, clarity, and narrative tone of such disclosures can significantly shape stakeholder perceptions. in the banking sector, signaling becomes even more nuanced. research by nini et al. (2012) demonstrated how banks employ syndicated loans to signal creditworthiness, underscoring the broader strategies banks adopt to communicate stability and reliability. although not directly tied to chairperson’s statements, these findings highlight the critical role of carefully crafted corporate communications in signaling financial health to the market. the vietnamese banking sector, given its rapid transformations and unique socioeconomic dynamics, further emphasizes the importance of signaling. it's not just about 196 tam phan huy et al. disclosing financial metrics but also providing a narrative that assures stakeholders of the bank's resilience and vision for the future. in synthesizing this, signaling theory, with its focus on information asymmetry and the role of disclosures in bridging this gap, provides a robust theoretical lens for investigating chairperson's statements in the banking sector. the vietnamese context, with its blend of traditional banking paradigms and contemporary challenges, makes this exploration all the more pertinent. 2.2.4 social constructivism social constructivism posits that our understanding of the world is not purely objective but is significantly influenced by human interactions, cultural norms, and societal influences. central to this theory is the idea that language and narrative shape and construct our realities berger et al. (1966). instead of passively reflecting the world, language actively constitutes and gives meaning to it. in the realm of corporate communications, the chairperson’s statements are not just mere reflections of an institution's performance. they are, in essence, narratives that construct a particular version of the bank's reality. this "constructed" reality, portrayed through strategic language choices, tonality, and emphasis, shapes stakeholders' perceptions of the bank's health, strategies, growth trajectory, and overall ethos. given the complexity and, at times, opacity of banking operations, stakeholders often rely on these narratives to decode the bank's current status and future outlook. several scholars have underscored the importance of narrative disclosures in shaping organizational realities. for example, beattie and smith (2013) analyzed the strategic use of graphs in corporate annual reports, suggesting that these visual narratives play a pivotal role in crafting a company's story. similarly, cho et al. (2010) discussed how firms leverage language in their sustainability reports to craft a certain image, especially in the face of environmental challenges. extending this to the vietnamese banking context, where trust and credibility are of paramount importance, chairperson's statements take on added significance. they become tools that not only communicate performance but also construct a reality that resonates with cultural, economic, and social nuances specific to vietnam. to conclude, social constructivism provides a valuable lens through which to examine chairperson's statements in the banking sector. it underscores the active role these narratives play in molding stakeholders' perceptions and shaping the bank's socially constructed reality. 2.3 empirical studies impression management within chairperson's statements has garnered significant attention across diverse markets and economic settings. the methodologies, findings, and subsequent limitations vary, offering a rich tapestry of insights and opportunities for further exploration. management impression and bank’s performance with nlp for chairperson’s statement for example, giglioni’s (2017) pivotal research using content analysis highlighted the apologetic strategies prominent in the annual reports of ft30 index companies. this study found a direct relationship between a company's financial performance and the extent of its apologetic discourse, suggesting that firms underperforming financially may engage more intensively in such strategies to curate their corporate image. while the ft30 index encompasses large, internationally active firms, a potential limitation of giglioni’s study lies in its focus on mature, well-established corporations in advanced markets, which may not fully capture impression management dynamics in smaller firms, emerging markets, or state-influenced sectors such as banking in vietnam. in the european context, the intricate dynamics of impression management in chairperson's statements have been significantly illuminated by research emanating from the iberian peninsula. oliveira et al. (2016) spearheaded this exploration, concentrating on the narratives of portuguese managers, especially during periods marked by economic downturns and scarce resources. drawing from the theoretical underpinning of the two-component model of impression management, their research unearthed some compelling findings. namely, they deduced that public visibility and consumer proximity, dimensions related to the public's perception and direct engagement with the company, respectively, emerged as pivotal determinants in shaping impression management tactics. interestingly, these variables appeared to supersede the influence of organizational outcomes, such as profit margins or other financial indicators, on narrative strategies. yet, in parallel to giglioni's study on the ft30 index members, the research by azevedo and borges is not without its limitations. the study's geographical demarcation, primarily centered on portugal, indicates a constrained cultural and economic context. such a confinement underscores the need for caution when attempting to generalize or extrapolate its findings to a more global or diverse setting. further reinforcing the european perspective, brennan and merkl-davies (2013) analyzed chairperson's statements across several ftse 100 companies in the uk. they found that narrative disclosures, often manipulated through impression management techniques such as positive tone bias, selective emphasis on favorable outcomes, and attribution of poor performance to external factors, were used to influence readers' perceptions and interpretations. this observation was consistent with the findings of van der laan smith et al. (2005), who in their examination of european firms found that larger corporations, particularly those with more significant public exposure, leaned towards impression management strategies in their annual reports such as managing the readability and complexity of text, highlighting future-oriented statements, and downplaying risk factors. such strategies were particularly evident during financial downturns, emphasizing the organization's resilience and adaptability. merkl-davies and brennan (2007) undertook a significant exploration into the intricacies of impression management within chairperson's statements, particularly emphasizing the influence of goal-relevance. their research argued that companies often tailor their narrative disclosures based on strategic goals, whether they be related to market position, brand perception, or stakeholder management. one of the standout findings from their research is the influence of company visibility and consumer proximity on impression management 198 tam phan huy et al. strategies. company visibility, as elucidated by merkl-davies and brennan (2007), pertains to the degree to which a company's activities, performances, and decisions are under the scrutiny of the public and relevant stakeholders. firms with higher visibility often have a greater onus to manage impressions effectively, given the amplified repercussions of public perception on their brand and operations. consumer proximity, on the other hand, relates to the closeness or directness of the relationship between the company and its consumer base. companies with a direct consumer interface, such as retail or service-based firms, may be more inclined to employ impression management tactics to curate and maintain a favorable image. in corroborating the research findings of oliveira et al. (2016), merkl-davies and brennan (2007) not only validated the significance of these determinants but also expanded upon them. they posited that impression management isn't merely a reactionary corporate exercise; it's a strategic endeavor rooted in a firm's core objectives and market positioning. by understanding the factors that drive firms to utilize specific strategies, these studies together provide a nuanced comprehension of corporate narrative practices and the underlying motivations propelling them. the african market, as examined by phesa (2021) and dhludhlu (2022) for jse listed companies, contributes uniquely to this discourse. phesa's content analysis method revealed strategic impression management practices, particularly pronounced in unprofitable companies. dhludhlu's subsequent study during the covid-19 pandemic era showcased that companies, irrespective of profitability, engaged in impression management. however, the limited temporal scope, focusing on the pandemic period, calls for more extensive studies to discern long-term patterns. li et al. (2019) exploration in the hong kong market is particularly insightful, offering a distinctive methodology that amalgamates both quantitative financial indicators and sentiment metrics. their approach, which primarily zooms in on the tone shifts in chairperson's statements, uncovers a fascinating link: these tone fluctuations have the potential to forecast stock prices over extended periods. this revelation brings to the fore potential inefficiencies within the hong kong stock market, suggesting that sentiment in narrative disclosures might exert more influence on stock valuations than previously acknowledged. however, it's crucial to juxtapose their results with findings from other asian markets for a more comprehensive perspective. in a parallel study, tan et al. (2011) delved deep into the chinese stock market, emphasizing how firms' annual report narratives could influence investor sentiment, especially in a market characterized by a high retail investor base. their results affirmed the persuasive power of these narratives, particularly in shaping stock valuations. meanwhile, suto (2003) conducted an analysis on japanese firms and underscored the crucial role of voluntary narrative disclosures in influencing investor decisions, especially amidst information asymmetries. furthermore, (adil, 2021) in their probe into the pakistan stock market, discerned a noticeable relationship between firms' chairperson's statement tones and their subsequent stock performance, suggesting a common trend across different asian markets. surprisingly, the discourse around impression management in chairperson's statements remains largely untouched in markets like vietnam. this presents a stark contrast, considering management impression and bank’s performance with nlp for chairperson’s statement studies like those by smith and taffler (1992), which broadened the purview to uk-based firms, or by merkl-davies and brennan (2007), which delved into the rhetorical strategies adopted by firms in market commentaries. the empirical landscape on impression management within chairperson's statements showcases a diverse range of methodologies, findings, and inherent limitations. while studies from established markets offer invaluable insights, a glaring research gap persists for emerging markets, notably vietnam. given its distinct socio-economic and corporate backdrop, this represents a promising avenue warranting rigorous academic exploration. 3. methodology 3.1 variables measurement 3.1.1 management impression in the quest to quantify management impression through chairperson's statements in annual reports, this study harnesses the capabilities of natural language processing (nlp). a variety of sophisticated nlp techniques are deployed, each offering unique insights into the narrative fabric of these statements. one of the cornerstone methodologies employed is sentiment analysis, which decomposes text into two main attributes: subjectivity and polarity. subjectivity discerns the extent to which statements are objective or based on personal sentiments and beliefs. on the other hand, polarity gauges the sentiment's direction, determining if the narrative exudes a positive, negative, or neutral tone (liu, 2022). these dimensions together offer a snapshot into the emotional and factual content of the chairperson's pronouncements, revealing potential managerial biases or overall sentiment regarding bank performance. additionally, topic change analysis is employed, focusing on the thematic shifts within the narrative. by computing a ratio that captures these changes, the method provides insights into the variability of topics addressed. a higher ratio indicates a pronounced topic change within the chairperson's discourse. this can be indicative of the management's attempt to highlight a diverse set of achievements or challenges or could signify strategic diversions from sensitive subjects (blei et al., 2003). furthermore, the study employs a word vectorization method, specifically leveraging average word vectors. this technique captures the semantic richness of the text, enabling the identification of recurring themes and the underlying sentiments attached to them. by averaging these vectors, we can glean a holistic understanding of the central themes dominating the chairperson's discourse (mikolov et al., 2013). lastly, the reading ease of the statements is evaluated. a higher score in this metric indicates that the narrative is more accessible and comprehensible to the audience. clear, straightforward statements might reflect management's dedication to transparency, while more 200 tam phan huy et al. convoluted texts could signify obfuscation or a penchant for complexity (kincaid et al., 1975). collectively, these methodologies offer a nuanced lens to interpret and quantify management impressions. the amalgamation of sentiment tone, thematic variability, semantic richness, and readability provides a comprehensive assessment, ensuring that various facets of the chairperson's narrative are captured and analyzed. this multilayered approach not only enhances the validity of the findings but also aligns seamlessly with the overarching objective of understanding management impression in the vietnamese banking context. 3.1.2 bank performance net interest margin (nim) is particularly apt for this research topic as it offers a direct lens into a bank's core operational efficiency in the vietnamese context. as a reflection of the difference between the interest income generated by banks and the amount of interest paid out, nim provides a clear metric of a bank's profitability relative to its interest-earning assets. this ratio encapsulates the essential dynamics of banking operations, making it a pertinent measure for juxtaposing against management impressions conveyed in chairperson's statements (rose & hudgins, 2008). the formula for calculating nim is: nim = (interest income – interest expense) / average earning assets 3.1.3 chairperson characteristic the age of a chairperson is often seen as a reflection of their experience, leadership style, and strategic vision. according to adams and ferreira (2009), older chairmen might tend to rely heavily on their past experiences. this might influence their impression management strategies to echo traditional banking practices and conservative approaches in their statements. conversely, jenter and lewellen (2021) suggest that such accumulated wisdom from older chairmen might provide steady leadership, especially in uncertain economic times, leading to statements that emphasize stability and resilience. in the upper echelons theory proposed by hambrick and mason (1984), it's highlighted that age not only shapes the cognitive bases but also the values which influence strategic choices. this means the framing of annual reports by older chairmen might be crafted from a perspective that reflects cumulative banking wisdom. however, a potential downside, as explored by mahlberg et al. (2013), is that older chairmen might be less receptive to newer market trends or innovations, which could reflect in their statements and indirectly influence the perception of the bank's adaptability, subsequently affecting bank performance. education, serving as a marker of a chairperson's formal training, can significantly shape how they perceive and project the bank's image through their statements. gerasimenko and razumova (2020) note that a chairperson's educational background, both in level and field, can management impression and bank’s performance with nlp for chairperson’s statement alter corporate policies and this extends to how they frame their annual reports. for instance, a chairperson with a background in business or finance might present statements that lean heavily on data-driven successes or financial metrics. nguyen and nielsen (2010) offer an interesting perspective, suggesting that education from elite institutions can influence a more global or expansive outlook, which could reflect in the tone and content of chairperson's statements, emphasizing the bank's positioning in a broader financial ecosystem. furthermore, kim et al. (2007) and daily et al. (2000) provide insights into the risk-taking propensity and diverse decision-making abilities influenced by education. a chairperson's educational background might either embolden them to project ambitious future plans in their statements or might lead to a more conservative, risk-averse narrative. lastly, darmadi (2013) found that chairmen with strong educational backgrounds are more likely to embrace best practices in corporate governance, which could translate into clearer, more transparent annual statements, potentially boosting investor confidence and positively impacting bank performance. in conclusion, a chairperson's age and education not only influence their leadership styles but also the impression management strategies they deploy in their annual statements. both these factors can play pivotal roles in shaping stakeholders' perceptions and can subsequently impact bank performance, either by bolstering investor confidence or by giving insights into the bank's strategic vision and adaptability in a dynamic financial landscape. 3.1.4 control variables the selection of control variables in a study concerning management impression in chairperson's statements and their relationship with bank performance is crucial to account for other factors that might influence bank outcomes, include: size, age, leverage, and growth.. size (total asset): bank size, typically measured by total assets, is a foundational control variable. larger banks might have different operational efficiencies, risk exposures, and market impacts than their smaller counterparts. in the context of impression management, larger banks might adopt different strategies given their wider stakeholder base and more significant market scrutiny. studies such as berger et al. (2005) have routinely included bank size as a control variable, emphasizing its importance in financial models. age (bank age): the age of a bank can indicate its experience, stability, and historical market presence. older banks might have established reputations, which can influence how they engage in impression management. pervan and višić (2012) have employed bank age as a control variable, hypothesizing that older banks might exhibit different patterns in financial disclosures compared to younger, potentially more agile banks. leverage (debt to asset ratio): leverage measures the extent to which a bank is financed by debt. high leverage can indicate higher risk but also potentially higher returns. it's essential to control leverage since banks with different leverage ratios might have varied risk appetites and, subsequently, different strategies for impression management. gropp and heider (2010) incorporated leverage in their study on bank capital structure, suggesting its significance 202 tam phan huy et al. in shaping bank behaviors. growth (change in interest income): a bank's growth trajectory, especially in its core operations like interest income, can significantly impact its communication strategies. rapidly growing banks might prioritize showcasing their success, whereas banks experiencing slower growth might employ more defensive impression management tactics. in their study on the growth effects of bank mergers, beccalli and frantz (2009) considered the change in interest income as an indicator of growth momentum. to conclude, these control variables provide a comprehensive backdrop against which the interplay of management impression and bank performance can be understood. by considering these factors, the study ensures a more robust and nuanced understanding of the primary variables of interest, drawing from a rich tradition of financial research. table 1. main variables. variables proxy formula references bank’s performance net interest margin (nim) (interest income – interest expense) / average earning assets rose and hudgins (2008) management impression (mi) sentiment analysis polarity: measures the sentiment of a statement, indicating whether it's positive, negative, or neutral. subjectivity: assesses how personal or objective a statement is, with higher subjectivity meaning the statement is based more on personal feelings or opinions rather than factual information. liu (2022) topic change variability of topics addressed; higher number indicate wider topic covered or changed blei et al. (2003) word vectorize semantic richness of the text, the higher number indicate the higher level of word spreading mikolov et al. (2013) reading ease a higher score in this metric suggests that the narrative is easier to understand and more relatable to the readers. kincaid et al. (1975) chairperson characteristic (cc) chairperson age (ce) natural logarithm of chairperson age chairperson education (ce) 0: bachelor or below 1: master or equivalent 2: phd or above control variables bank size (size) natural logarithm of total asset berger et al. (2005) bank age (age) natural logarithm of bank age pervan and višić (2012) leverage (lev) total debt / total equity gropp and heider (2010) growth (gro) change in interest income beccalli and frantz (2009) source: author management impression and bank’s performance with nlp for chairperson’s statement 3.2 data in this study, data was sourced from all commercial banks listed in the vietnamese market. primary financial metrics were extracted from the refinitive secondary database. complementing this, the narrative data from chairperson's statements was manually compiled from the annual reports of these banks. instances where banks failed to publish requisite information were excluded from the dataset. from total listed bank in vietnam stock exchange market, the sample size reduces to 27 commercial banks. after exclude observation without available chairperson’s statements, the final compiled dataset comprised 212 observations, spanning a decade from 2012 to 2022. the choice of the 2012–2022 period is deliberate and methodologically appropriate for several reasons. first, this decade captures a full economic cycle in vietnam, including phases of recovery after the global financial crisis and periods of sustained economic growth. second, significant developments in financial regulation and disclosure transparency occurred during this period, notably with the implementation of circular no. 155/2015/tt-btc, which formalized annual report disclosure requirements. additionally, this time frame includes the onset and impact of the covid-19 pandemic (2020–2021), which provides a natural context for examining how impression management strategies evolved under crisis conditions. thus, the selected period offers a rich and relevant context for analyzing both normal and stress conditions in corporate narrative practices. subsequently, the narrative text data was subjected to processing using natural language processing (nlp) techniques. with the assistance of specialized python libraries tailored for nlp, the chairperson's statements were quantitatively analyzed to extract features related to management impression. once processed, these quantified impressions, alongside the financial metrics, were integrated to formulate regression models for further analysis. 3.2 model this investigation is guided by the research model proposed by xie and wang (2023), as detailed below. performancei, t = β0 + β1mii, t -1 + β2cci, t-1 + β3cci, t-1* mii, t-1 + β4sizei, t-1 + β5agei, t-1 + β6levi, t-1 + β7groi, t-1 + ε where: performancei, t: the performance of the bank i at year t, measured by nim. mii, t: the management impression score of bank i at year t-1, measured by subjectivity (sub), polarity (pol), topic change score (top), word vectorized score (vec), and reading ease score (ease). cci, t: chairperson characteristics of bank i at year t-1, include chairperson age (ca) and chairperson education (ce). sizei, t: the equity to asset ratio of bank i at year t-1. 204 tam phan huy et al. agei, t: the loan to deposit ratio of bank i at year t-1. levi, t: the loan to asset ratio of bank i at year t-1. groi, t: the board independence ratio of bank i at year t-1. β0, β1, β2, β3, β4, β5, β6: are coefficients. ε: error term. in this research, the authors have employed the linear regression technique. linear regression is especially apt for this study as it allows us to understand and quantify the relationship between digital transformation and bank performance. given the continuous nature of our dependent variable, which in this case is bank performance, linear regression can effectively capture the direction and strength of the relationship between our predictor variables, notably digital transformation metrics, and the outcome. moreover, this statistical method is renowned for its capacity to provide a clear view of the potential predictors' impact on the outcome, while controlling for other variables. as such, this study seeks to understand the nuanced influence of digital transformation on bank performance amidst other factors, the linear regression technique proves to be a valuable tool. to ensure the validity of the regression results, the study applies pooled ols regression with heteroskedasticity-robust standard errors. prior to estimation, diagnostic tests were conducted to verify model assumptions. the variance inflation factor (vif) was used to assess multicollinearity, confirming that no serious multicollinearity exists among the independent variables (all vifs < 5). the breusch-pagan/cook-weisberg test was applied to detect heteroskedasticity, which was addressed using robust standard errors. additionally, residual plots and kernel density estimations were reviewed to assess the linearity and normality assumptions of residuals. for robustness, alternative model specifications were tested, including year-fixed effects (not reported here but available upon request), which produced consistent results in terms of coefficient signs and statistical significance. these diagnostic procedures support the reliability and robustness of the regression outcomes presented in this study. 4. results and discussion 4.1 descriptive analysis the dataset is a comprehensive collection of 212 observations derived from 27 distinct commercial banks over a decade-long span from 2012 to 2022. this breadth provides a rich backdrop for understanding not only the financial metrics but also the narrative strategies adopted by bank chairperson. management impression and bank’s performance with nlp for chairperson’s statement table 2. descriptive statistics. nim pol sub top vec ease ca size age lev gro count 212 212 212 212 212 212 212 212 212 212 212 mean 0.03 0.15 0.75 1.17 0.06 0.74 52.68 306,473 26.29 0.95 0.13 std 0.02 0.15 0.16 0.37 0.44 0.16 8.8 394,518 11.89 0.25 0.26 min 0.001 -0.4 0 0.3 -1.25 0 32 150 5 0.001 -0.43 25% 0.02 0.08 0.7 0.95 -0.23 0.65 47 74,746 20 0.9 0.02 50% 0.02 0.16 0.85 1.12 0.11 0.77 52 165,135 24 0.92 0.14 75% 0.04 0.25 0.85 1.4 0.35 0.86 59 363,057 28 0.94 0.22 max 0.09 0.5 0.87 2 1.64 1 74 2,080,18 0 65 2.8 2.76 note: unit for size is billion vnd source: author the net interest margin (nim) presents a pivotal insight into a bank's core business profitability. with an average nim of 0.026 and a range from 0.00015 to 0.0876, there is evidence of significant variation. this disparity possibly stems from diverse business models, risk appetites, and operational efficiencies among the banks. the variability might also hint at how different banks have weathered economic challenges over the observed period. exploring the narratives from chairperson's statements provides a unique perspective into the banks' communication strategies. the polarity (pol) measure, with an average score of 0.152, illustrates that most narratives lean towards a positive sentiment. this is understandable given the importance of instilling confidence in stakeholders. however, the range from -0.4 to 0.5 suggests there were instances where a more cautious or even negative tone was adopted, perhaps reflecting challenging fiscal years or significant adverse events. polarity captures the overall sentiment of the chairperson’s statement, indicating whether the language used is positive, neutral, or negative. for example, words and phrases such as “strong growth,” “robust performance,” “resilient strategy” contribute to a higher positive polarity score, while terms like “challenges,” “difficult market conditions,” “unexpected losses” lower the score into negative territory. in contrast, subjectivity reflects the extent to which the statement expresses personal opinions, judgments, or emotions rather than objective facts. highly subjective sentences might include phrases like “we believe the bank is well-positioned” or “we are confident in our future trajectory,” whereas objective statements would be purely factual, such as “net interest margin increased by 2%”. subjectivity (sub) emerges as another compelling metric. with an average score of 0.75, it's evident that chairperson's statements often oscillate between objective presentations and personal insights. this interplay of factual data and subjective viewpoints can be seen as a strategy to balance transparency with the chairperson's vision and interpretation of events. variability in topic change (top), which averages at 1.17, shows that chairmen frequently adjust the subjects they address. this adaptability might reflect the shifting priorities and challenges banks face annually. furthermore, word vectorization (vec), with its mean value of 0.056, offers glimpses into the semantic depth of these statements. a diverse semantic 206 tam phan huy et al. structure might be indicative of a more comprehensive discussion or a strategy to address a broader audience. reading ease (ease) emerges as a vital metric, especially in a sector like banking which can be replete with technical jargon. an average score of 0.736 indicates a general inclination towards making narratives relatable and comprehensible, a strategy likely aimed at ensuring broader stakeholder engagement. chairperson demographics, specifically chairperson age (ca) with an average of 52.68 years, hints at the vast reservoir of experience helming these institutions. this wide age range, spanning from 32 to 74, might correlate with varied leadership styles, strategies, and risk appetites. on the control side of the equation, the data reveals valuable insights. bank size (size), represented by an average asset value of 306,473.2006, showcases the considerable financial heft of these institutions. meanwhile, bank age (age) at an average of 26.29 years highlights the maturity and possibly the resilience of these banks. the leverage (lev) metric, averaging 0.947, gives insights into the capital structure and risk posture of these banks. growth (gro), finally, with an average of 0.1301, sheds light on the banks' ability to expand their interest income streams, a key driver of profitability. figure 1. distribution of chairperson education (ce) in summation, the dataset is not just a reflection of numbers but a mosaic of narratives, strategies, and financial health. it provides an intricate view of the vietnamese commercial banking scene, spotlighting both their financial trajectories and the narrative techniques of their leadership. in the examination of the educational backgrounds of chairmen across the dataset, the majority were found to possess an education level of a bachelor's degree or below, representing 41.5% of the dataset. notably, the data also revealed that a substantial proportion, 31.1%, have attained the highest echelon of academic achievement, holding a phd or a similar advanced degree. meanwhile, those with a master's degree or its equivalent make up 27.4% of the chairmen. this diverse distribution underscores the variability in educational qualifications among the chairmen of these institutions. while a significant portion have foundational academic backgrounds, there is an almost equally large cohort with the highest academic qualifications. this could suggest that while formal education is important, banks also value management impression and bank’s performance with nlp for chairperson’s statement diverse experiences and expertise that might not necessarily correlate with advanced degrees (hambrick & mason, 1984; pyatt, 1966). furthermore, the substantial presence of phd holders at the helm might indicate the increasing complexity and sophistication of the banking industry, requiring a deep, research-oriented understanding of financial systems and markets. 4.2 regression analysis this study delves into the nuanced relationship between the net interest margin (nim) and a blend of bank management impressions and chairperson characteristics. the regression models employed revolve around three central sets. the first, management impression (mi), integrates metrics extracted from annual bank reports, such as subjectivity (sub), polarity (pol), topic change (top), word vectorized score (vec), and reading ease (ease). the second set, chairperson characteristics (cc), considers the chairperson's age (ca) and educational qualifications (ce). additionally, an interaction term (cc*mi) is introduced to discern any potential interplay between the mi and cc variables. to offer a holistic understanding, the analysis also factors in several bank-specific controls, including size (size), age (age), leverage (lev), and growth (gro). with the intent of providing robust insight, the research meticulously crafts combinations of variables from the mi and cc groups, resulting in a total of 10 distinct regression models. this exhaustive approach assures a comprehensive evaluation of the dynamics at play. the regression result table shows the investigation encompassing ten distinct regression models unearthed a myriad of associations between the net interest margin (nim) and an assortment of bank attributes. unpacking the nuances from the results: diving into management impression metrics, pol stands out in model_2 with a negative, significant coefficient. this suggests that banks with a more positive tone in their narratives might witness a decline in nim. similarly, the significant positive coefficient for top in model_4 hints at the value of diverse topics in management discourse. a diverse range of topics might indicate thoroughness and holistic management approaches that resonate with stakeholders. meanwhile, the negative significance of ease in model_9 underscores that more complex, less easily digestible narratives could adversely affect the nim. one could argue that challenging texts might deter certain stakeholders, possibly influencing financial indicators like nim. chairperson characteristics manifest varied significances. for ce, its positive significance in models 2, 5, and 7 highlights the value of advanced education. a possible interpretation is that chairmen with higher educational qualifications bring advanced management and strategic perspectives, subsequently improving the nim. conversely, the age (ca) emerges significant in model_5, hinting that older chairperson, perhaps with more experience, might steer banks towards better financial performance. 208 tam phan huy et al. table 3. regression result. model 1 model 2 model 3 model 4 model 5 model 6 model 7 model 8 model 9 model 10 interce pt 0.12 0.01 0.05 0.02 -0.17* 0.01 -0.01 0.02** 0.28** 0.03* sub -0.16 0.01 pol -0.38* -0.00 top 0.14* 0.00 vec 0.03 -0.00 ease -0.34* -0.01 ca -0.03 -0.01 0.04* 0.01 -0.06 ce 0.01* 0.00** 0.00 0.00** 0.01* sub*ca 0.04 sub*ce -0.01 pol*ca 0.10* pol*ce 0.00 top*ca -0.03 top*ce 0.00 vec*ca -0.01 vec*ce 0.01** ease*c a 0.08* ease*c e -0.01 size -0.001** 0.001** -0.001** -0.001** 0.001** 0.001** -0.001* -0.001** -0.001** -0.001* age 0.001** 0.001** 0.001** 0.001* 0.001** 0.001** * 0.001** -0.001** -0.001* -0.001** lev 0.001* 0.001* 0.001** 0.001* -0.001** 0.001** -0.001** 0.001* -0.001** 0.001** gro 0.01*** 0.01*** 0.01*** 0.01*** 0.01*** 0.01*** 0.01** 0.01** 0.01** 0.01** note: ***, ** and * indicate that the p_value is lower than 0.01, 0.05 and 0.1 source: author the interaction terms deserve special mention. the positive significant interaction between pol and ca in model_3 suggests that the age or experience of the chairperson modulates the effect of narrative polarity on nim. perhaps, seasoned chairmen can effectively use positive tones to the bank's advantage. the significant positive interaction of vec and ce in model_8, and ease and ca in model_9, further underscore that the effect of management impressions on nim isn't isolated but interplays dynamically with chairperson characteristics. the control variables, namely size, age, lev, and gro, display almost ubiquitous significance across the models, underscoring their foundational roles. especially, gro emerges significant across all models, suggesting that bank growth is an unequivocal determinant of nim, resonating with conventional financial wisdom. management impression and bank’s performance with nlp for chairperson’s statement in essence, while numerous variables wield significance in specific models, the omnipresence of control variables underlines their criticality. additionally, the complex interplay between management narratives and chairperson attributes delineates the multifaceted nature of influences on financial outcomes. these findings advocate for a deeper introspection by banks into their narrative strategies, chairperson selections, and how these facets could be harmonized for better financial performance. 4.3 discussion of findings this study set out to examine how impression management strategies, as embedded in chairperson’s statements and quantified using natural language processing (nlp) techniques, are associated with bank performance in vietnam. the core research objective was to investigate whether narrative features, such as sentiment, topic diversity, semantic depth, and readability, alongside chairperson characteristics, could meaningfully predict bank performance, proxied by net interest margin (nim). this research contributes to the growing literature on narrative disclosures by providing evidence from an emerging economy with distinctive governance dynamics. the regression findings reveal several notable patterns. first, narrative polarity (pol) is negatively associated with nim. this counterintuitive result supports earlier findings by singh and singla (2022) and vijay and singla (2023), who observed that firms tend to emphasize overly positive narratives during periods of weaker financial performance, a form of "impression management strategy." similarly, giglioni (2017) reported that companies with poor financial performance engage in apologetic or overly optimistic language to manage stakeholder expectations. this suggests that positive tone in chairperson’s statements may not necessarily reflect actual performance but rather serve as a compensatory mechanism to protect corporate image. in contrast, topic variability (top) shows a positive and statistically significant effect on performance, which aligns with oliveira et al. (2016) and merkl-davies & brennan (2007), who found that diverse narratives are more persuasive and indicative of strategic clarity and adaptive management. a broader range of discussion points may signal transparency and thoroughness, enhancing stakeholder trust. reading ease (ease) exhibits a negative association with nim, indicating that overly simplistic (or possibly overly technical) statements may be viewed with skepticism. this result parallels the concerns raised by clatworthy and jones (2001), who suggested that readability manipulation could either obscure information or present it too superficially, reducing stakeholder confidence. with regard to chairperson characteristics, chairperson education (ce) is positively associated with bank performance in several models. this supports findings by darmadi (2013) and nguyen and nielsen (2010), who argue that more educated leaders tend to adopt better governance practices and more credible disclosure strategies. the finding also aligns with the 210 tam phan huy et al. upper echelons theory (hambrick & mason, 1984), which posits that top executives’ backgrounds influence strategic outcomes. importantly, the interaction effects such as pol*ca and ease*ca underscore that narrative strategies do not operate in isolation; rather, their effectiveness is conditioned by who delivers them. the positive interaction between chairperson age and tone suggests that seasoned leaders may be more effective in using narrative tools to positively influence stakeholder interpretation, echoing adams and ferreira (2009) on the signaling power of leadership experience. the results also reinforce key theoretical underpinnings: • impression management theory: the strategic use of tone and narrative structures confirms that disclosures are not merely descriptive but are deliberately framed to influence perception (merkl-davies et al., 2011; neu et al., 1998). • narrative disclosure theory: the findings validate the view that qualitative disclosures provide value-added context beyond financial metrics (beattie et al., 2004; holland, 2005). • signaling theory: the interaction terms highlight how disclosure effectiveness depends on the credibility of the signaler (spence, 1978; healy & palepu, 2001). implications for companies include the importance of maintaining consistency between tone and actual performance. misalignment such as overly positive narratives during weak financial years may be interpreted as manipulative and erode trust. furthermore, firms should consider how leadership characteristics, especially education and experience, impact the effectiveness of narrative strategies. boards should be mindful of how these disclosures reflect on corporate governance quality and investor communication. 5. conclusions and recommendations 5.1 conclusions impression management within chairperson's statements has long been recognized as a critical tool for shaping perceptions among stakeholders. this research furthered the exploration into this realm by analyzing how the tone and nature of these statements, juxtaposed with the characteristics of the chairperson, impact the net interest margin (nim) of banks. drawing from various theoretical frameworks and empirical studies, the findings here bridge several research gaps and reinforce some established paradigms. this study set out to investigate the research question: how do impression management strategies within chairperson’s statements affect bank performance in the vietnamese banking sector? specifically, it aimed to examine whether key narrative features, such as sentiment, topic diversity, and readability, as well as chairperson characteristics, influence banks’ net interest management impression and bank’s performance with nlp for chairperson’s statement margin (nim), a core indicator of performance. through the application of nlp techniques and regression analysis on data from 27 vietnamese banks over a 10-year period, the findings provide clear evidence that impression management features in chairperson’s statements are significantly associated with variations in bank performance, thus directly addressing and answering the central research question. firstly, the significance of management narratives, particularly those involving polarity (pol), topic change (top), and ease of reading (ease), affirms several hypotheses posited in previous literature. giglioni (2017) observations on the correlation between a company's financial narratives and its performance resonate well with the results of our study. just as giglioni highlighted the apologetic strategies prominent in annual reports of certain companies, the research suggests that polarity in narratives, representing the overall sentiment, plays a pivotal role in influencing the nim. this finding corroborates the idea that firms, particularly those facing financial challenges, might resort to specific narrative tactics, underpinned by impression management theories, to strategically curate their corporate image. furthermore, the findings relating to chairperson characteristics, including age and education, and their intersection with narrative strategies, present an enriched understanding of corporate communication. while previous research by oliveira et al. (2016) emphasized the overarching role of public visibility and consumer proximity in shaping corporate narratives, this study delves deeper. it highlights that not only does the nature of the narrative matter, but so does the chairperson's personal attributes. moreover, the way these attributes intertwine with the narrative can profoundly influence core financial metrics like the nim. such insights offer a more granular perspective, suggesting that the art of impression management is deeply entrenched in the tapestry of a corporation's governance and leadership dynamics. the consistency of these findings with broader asian studies, like those by li et al. (2019) and tan et al. (2011), offers a harmonized perspective on the significance of chairperson's narratives across diverse markets. both these studies illuminated the consequential role of chairperson's statements in shaping financial outcomes and investor sentiments. here, the pronounced significance of variables such as sub, pol, top, and ease, and their interplay with chairperson attributes (ca and ce), adds more depth to this perspective. it amplifies the argument that impression management, far from being a superficial endeavor, is an intricate, strategic exercise embedded within a firm's broader corporate strategy. drawing from the theoretical frameworks of the two-component model of impression management, this study has empirically validated how public perception, entwined with chairperson attributes, molds narrative strategies. in light of the theories propounded by merkl-davies and brennan (2007), the current findings underscore the overarching influence of company visibility and consumer proximity on impression management strategies. firms with a more pronounced public profile face an inherent onus to adeptly manage stakeholder impressions, given the ramifications of public perception on both their brand and operational viability. this study accentuates that impression management strategies extend beyond just shaping narratives; they interlace with a firm's leadership dynamics, notably the characteristics of the chairperson. in conclusion, this research has succeeded in offering a multifaceted view of the world 212 tam phan huy et al. of impression management within chairperson's statements. by weaving together, the threads of narrative strategies, chairperson attributes, and financial metrics, it offers a more holistic understanding of corporate communications in the banking sector. as global markets evolve, and as stakeholder perceptions become increasingly pivotal in shaping corporate trajectories, the insights from this study spotlight the inextricable linkage between impression management, leadership, and financial outcomes. as with all research, while this study bridges existing gaps, it also illuminates avenues for future exploration, particularly in understanding how these dynamics play out in other sectors and diverse economic landscapes. 5.2 recommendations the recent findings, drawing on the intricate dynamics of impression management within chairperson's statements, offer crucial insights for various stakeholders, particularly investors, managers, and regulatory agencies. here's how these stakeholders can utilize and benefit from this research: for investors: in today's complex investment landscape, the ability to discern and navigate corporate narratives is paramount. it's evident from the study that firms often employ specific narrative strategies in their chairperson's statements, especially during challenging financial times. a direct relationship exists between a company's financial performance and the nature of its discourse. hence, astute investors should look beyond the numerical data, giving due weight to the tone and content of such statements. they can serve as a potential indicator of underlying financial health and the company's future trajectory. moreover, by understanding the nuances of the two-component model of impression management, investors can make more informed decisions, especially in markets characterized by high retail investor engagement. for managers and corporate leaders: managers and corporate leaders stand to gain significantly by integrating the principles of impression management into their communication strategies. the research underscores the profound influence of company visibility and consumer proximity on narrative strategies. therefore, managers should craft statements that not only reflect the organization's financial stance but also resonate with its broader stakeholder base. this requires a deep understanding of both the firm's internal dynamics and the external market landscape. furthermore, periodic communication audits can be immensely beneficial. by assessing the effectiveness and alignment of leadership narratives with organizational goals, managers can refine their communication strategies, ensuring they resonate with their intended audience. emphasizing transparency, especially during financial downturns, can also foster trust and loyalty among stakeholders. for regulatory and related agencies: the research provides a treasure trove of insights for regulatory bodies and related agencies. recognizing the significant influence of chairperson's statements on stakeholder perceptions, regulatory bodies can develop guidelines that promote transparency and authenticity in corporate narratives. by ensuring that firms adhere to best practices in impression management, these agencies can level the playing field, management impression and bank’s performance with nlp for chairperson’s statement ensuring that stakeholders have access to accurate and unbiased information. moreover, the study's findings underscore the importance of regular stakeholder engagement. regulatory bodies can facilitate this by creating platforms where firms can interact directly with their stakeholders, gathering feedback, and addressing concerns in real-time. the findings of this study hold important implications for enhancing corporate governance standards and disclosure regulations in vietnam. regulatory bodies, such as the state bank of vietnam and the ministry of finance, could leverage insights from this research to develop more detailed guidelines on narrative disclosures, particularly in chairperson’s statements. for instance, standardizing requirements on tone clarity, readability, and thematic consistency could reduce information asymmetry and foster greater transparency. additionally, the demonstrated influence of chairperson characteristics on impression management highlights the need for more robust board appointment criteria, emphasizing educational diversity and communication competence. introducing formal training or disclosure accountability mechanisms for board-level communication may help align public narratives with actual performance, thus strengthening stakeholder trust and contributing to a more transparent financial system. in conclusion, the research on impression management within chairperson's statements provides a rich tapestry of insights for a diverse range of stakeholders. by understanding and internalizing these insights, investors, managers, and regulatory bodies can optimize their strategies, ensuring alignment with evolving market dynamics and stakeholder preferences. this not only fosters a cohesive and robust brand identity but also builds a foundation of trust and mutual respect among all stakeholders involved. 5.3 limitations and further research one of the primary limitations of this study pertains to its geographical focus. while the research offers deep insights into the intricacies of impression management within chairperson's statements in certain markets, these findings might not be universally applicable. regions or countries with distinct cultural, economic, or corporate governance mechanisms might display divergent impression management strategies. this inherent constraint mirrors the challenge in giglioni's study, where concentration on particular types of firms—whether demarcated by size, sector, or market reach—might introduce bias. the behaviors and communicative nuances of entities like smaller firms, startups, or those operating in niche industries could greatly differ from the predominant subjects of this research. furthermore, the study's methodological reliance on the content analysis of chairperson's statements, though invaluable, might not encompass the entire spectrum of impression management practices. corporate communication is vast, spanning beyond just chairperson's statements to include mediums such as press releases, investor presentations, and even informal interactions. the potential richness in these communications remains largely untapped in our current endeavor. 214 tam phan huy et al. looking ahead, there's a horizon of opportunities for research expansion. it's imperative to venture beyond established geographical confines, with regions like vietnam or other burgeoning economies waiting to be academically explored. such diversification would provide a richer, more rounded global perspective. delving deeper into a broader array of firms, especially those previously overshadowed like startups or smes, or even ones marking the frontline of industry evolution, can furnish a granulated understanding of varied impression management practices. future investigative avenues could also adopt a dual-method approach, amalgamating the depth of qualitative content analyses with the objectivity of quantitative methods, such as sentiment analysis. such synergy could unravel layers of corporate communication hitherto unexplored. equally captivating would be an exploration into stakeholder perception. understanding the reception, interpretation, and consequent decision-making influenced by chairperson's statements would be enlightening. this could be realized through stakeholder surveys or immersive focus groups. lastly, a temporal lens, scrutinizing the evolution of impression management tactics across extended timelines, especially during globally significant epochs or economic recalibrations, could elucidate the adaptive nature of these strategies. funding the research is funded by the university of economics and law, vietnam national university, ho chi minh city, vietnam. references adams, r. b., & ferreira, d. (2009). women in the boardroom and their impact on governance and performance. journal of financial economics, 94(2), 291–309. https://doi.org/10.1016/j.jfineco.2008.10.007 adil, m. (2021). the impact of audit committee characteristics on the risk-taking behavior of firms in pakistan. south asian review of business and administrative studies, 3(1), 33–54. al-sayani, y. m., mohamad nor, m. n., & amran, n. a. (2020). the influence of audit committee characteristics on impression management in chairperson statement: evidence from malaysia. cogent business & management, 7(1), 1774250. https://doi.org/10.1080/23311975.2020.1774250 barth, j. r., caprio jr, g., & levine, r. (2004). bank regulation and supervision: what works best? journal of financial intermediation, 13(2), 205–248. https://doi.org/10.1016/j.jfi.2003.06.002 beattie, v., mcinnes, b., & fearnley, s. (2004). a methodology for analysing and evaluating narratives in annual reports: a comprehensive descriptive profile and metrics for disclosure quality attributes. accounting forum. https://doi.org/10.1016/j.accfor.2004.07.001 https://doi.org/10.1016/j.jfineco.2008.10.007 https://doi.org/10.1080/23311975.2020.1774250 https://doi.org/10.1016/j.jfi.2003.06.002 https://doi.org/10.1016/j.accfor.2004.07.001 management impression and bank’s performance with nlp for chairperson’s statement beattie, v., & smith, s. j. (2013). value creation and business models: refocusing the intellectual capital debate. the british accounting review, 45(4), 243–254. https://doi.org/10.1016/j.bar.2013.06.001 beccalli, e., & frantz, p. (2009). m&a operations and performance in banking. journal of financial services research, 36, 203–226. https://doi.org/10.1007/s10693-008-0051-6 berger, a. n., & deyoung, r. (1997). problem loans and cost efficiency in commercial banks. journal of banking & finance, 21(6), 849–870. https://doi.org/10.1016/s0378-4266(97)00003-4 berger, a. n., miller, n. h., petersen, m. a., rajan, r. g., & stein, j. c. (2005). does function follow organizational form? evidence from the lending practices of large and small banks. journal of financial economics, 76(2), 237–269. https://doi.org/10.1016/j.jfineco.2004.06.003 berger, p. l., luckmann, t., blanchard, d. c., & bunot, j. (1966). the social construction of reality: a treatise in the sociology of knowledge. anchor books. blei, d. m., ng, a. y., & jordan, m. i. (2003). latent dirichlet allocation. journal of machine learning research, 3, 993–1022. bolino, m. c., kacmar, k. m., turnley, w. h., & gilstrap, j. b. (2008). a multi-level review of impression management motives and behaviors. journal of management, 34(6), 1080–1109. https://doi.org/10.1177/0149206308324325 bourke, p. (1989). concentration and other determinants of bank profitability in europe, north america and australia. journal of banking & finance, 13(1), 65–79. https://doi.org/10.1016/03784266(89)90020-4 bowler, m. c., & woehr, d. j. (2006). a meta-analytic evaluation of the impact of dimension and exercise factors on assessment center ratings. journal of applied psychology, 91(5), 1114. https://doi.org/10.1037/0021-9010.91.5.1114 bowman, e. h. (1984). content analysis of annual reports for corporate strategy and risk. interfaces, 14(1), 61–71. https://doi.org/10.1287/inte.14.1.61 brennan, n. m., & merkl-davies, d. m. (2013). accounting narratives and impression management. in the routledge companion to accounting communication (pp. 109–132). brown, s. v., & tucker, j. w. (2011). large-sample evidence on firms’ year-over-year md&a modifications. journal of accounting research, 49(2), 309–346. https://doi.org/10.1111/j.1475679x.2010.00396.x cho, c. h., roberts, r. w., & patten, d. m. (2010). the language of u.s. corporate environmental disclosure. accounting, organizations and society, 35(4), 431–443. https://doi.org/10.1016/j.aos.2009.10.002 claessens, s., coleman, n., & donnelly, m. (2018). “low-for-long” interest rates and banks’ interest margins and profitability: cross-country evidence. journal of financial intermediation, 35, 1–16. https://doi.org/10.1016/j.jfi.2017.05.004 clatworthy, m., & jones, m. j. (2001). the effect of thematic structure on the variability of annual report readability. accounting, auditing & accountability journal, 14(3), 311–326. https://doi.org/10.1108/09513570110399890 daily, c. m., certo, s. t., & dalton, d. r. (2000). international experience in the executive suite: the path to prosperity? strategic management journal, 21(4), 515–523. darmadi, s. (2013). board members’ education and firm performance: evidence from a developing economy. international journal of commerce and management, 23(2), 113–135. https://doi.org/10.1108/10569211311324911 https://doi.org/10.1016/j.bar.2013.06.001 https://doi.org/10.1007/s10693-008-0051-6 https://doi.org/10.1016/s0378-4266(97)00003-4 https://doi.org/10.1016/j.jfineco.2004.06.003 https://doi.org/10.1177/0149206308324325 https://doi.org/10.1016/0378-4266(89)90020-4 https://doi.org/10.1016/0378-4266(89)90020-4 https://doi.org/10.1037/0021-9010.91.5.1114 https://doi.org/10.1287/inte.14.1.61 https://doi.org/10.1111/j.1475-679x.2010.00396.x https://doi.org/10.1111/j.1475-679x.2010.00396.x https://doi.org/10.1016/j.aos.2009.10.002 https://doi.org/10.1016/j.jfi.2017.05.004 https://doi.org/10.1108/09513570110399890 https://doi.org/10.1108/10569211311324911 216 tam phan huy et al. demirgüç-kunt, a., & huizinga, h. (1999). determinants of commercial bank interest margins and profitability: some international evidence. the world bank economic review, 13(2), 379–408. https://doi.org/10.1093/wber/13.2.379 dhludhlu, n. r. (2022). a comparative analysis of impression management in chairperson’s statements amongst profitable and unprofitable jse-listed companies during the covid-19 pandemic. dietrich, a., & wanzenried, g. (2014). the determinants of commercial banking profitability in low-, middle-, and high-income countries. the quarterly review of economics and finance, 54(3), 337– 354. https://doi.org/10.1016/j.qref.2014.03.001 gerasimenko, v. v., & razumova, t. o. (2020). digital competencies in management: a way to superior competitiveness and resistance to changes. serbian journal of management, 15(1), 115–126. https://doi.org/10.5937/sjm15-23865 giglioni, c. (2017). annual reports of the ft30 index members: deployment of apologetic strategies in ceo and chairperson’s statements. international journal of language and linguistics, 4, 88–94. goffman, e. (1949). presentation of self in everyday life. american journal of sociology, 55, 6–7. goffman, e. (2016). the presentation of self in everyday life. in social theory re-wired (pp. 482–493). routledge. gropp, r., & heider, f. (2010). the determinants of bank capital structure. review of finance, 14(4), 587– 622. https://doi.org/10.1093/rof/rfp030 hambrick, d. c., & mason, p. a. (1984). upper echelons: the organization as a reflection of its top managers. academy of management review, 9(2), 193–206. https://doi.org/10.2307/258434 henry, e., & leone, a. j. (2016). measuring qualitative information in capital markets research: comparison of alternative methodologies to measure disclosure tone. the accounting review, 91(1), 153–178. https://doi.org/10.2308/accr-51161 higgins, c. a., & judge, t. a. (2004). the effect of applicant influence tactics on recruiter perceptions of fit and hiring recommendations: a field study. journal of applied psychology, 89(4), 622. https://doi.org/10.1037/0021-9010.89.4.622 holland, j. (2005). a grounded theory of corporate disclosure. accounting and business research, 35(3), 249–267. https://doi.org/10.1080/00014788.2005.9729990 hooghiemstra, r. (2000). corporate communication and impression management: new perspectives why companies engage in corporate social reporting. journal of business ethics, 27, 55–68. https://doi.org/10.1023/a:1006400707757 huang, x., teoh, s. h., & zhang, y. (2014). tone management. the accounting review, 89(3), 1083–1113. https://doi.org/10.2308/accr-50684 jenter, d., & lewellen, k. (2021). performance-induced ceo turnover. the review of financial studies, 34(2), 569–617. https://doi.org/10.1093/rfs/hhaa069 kim, k. a., kitsabunnarat-chatjuthamard, p., & nofsinger, j. r. (2007). large shareholders, board independence, and minority shareholder rights: evidence from europe. journal of corporate finance, 13(5), 859–880. https://doi.org/10.1016/j.jcorpfin.2007.09.001 kincaid, j. p., fishburne jr, r. p., rogers, r. l., & chissom, b. s. (1975). derivation of new readability formulas (automated readability index, fog count and flesch reading ease formula) for navy enlisted personnel. https://doi.org/10.21236/ada006655 le, t. p. v., & phan, t. b. n. (2017). capital structure and firm performance: empirical evidence from a small transition country. research in international business and finance, 42, 710–726. https://doi.org/10.1016/j.ribaf.2017.07.012 https://doi.org/10.1093/wber/13.2.379 https://doi.org/10.1016/j.qref.2014.03.001 https://doi.org/10.5937/sjm15-23865 https://doi.org/10.1093/rof/rfp030 https://doi.org/10.2307/258434 https://doi.org/10.2308/accr-51161 https://doi.org/10.1037/0021-9010.89.4.622 https://doi.org/10.1080/00014788.2005.9729990 https://doi.org/10.1023/a:1006400707757 https://doi.org/10.2308/accr-50684 https://doi.org/10.1093/rfs/hhaa069 https://doi.org/10.1016/j.jcorpfin.2007.09.001 https://doi.org/10.21236/ada006655 https://doi.org/10.1016/j.ribaf.2017.07.012 management impression and bank’s performance with nlp for chairperson’s statement leary, m. r., & kowalski, r. m. (1990). impression management: a literature review and two-component model. psychological bulletin, 107(1), 34–52. https://doi.org/10.1037/0033-2909.107.1.34 li, f. (2010). the information content of forward-looking statements in corporate filings: a naïve bayesian machine learning approach. journal of accounting research, 48(5), 1049–1102. https://doi.org/10.1111/j.1475-679x.2010.00382.x li, j., luo, w., & deng, x. (2019). the effect of chairperson’s statement tone changes in annual reports from hong kong. journal of physics: conference series. https://doi.org/10.1088/17426596/1168/3/032024 liu, b. (2022). sentiment analysis and opinion mining. springer nature. loughran, t., & mcdonald, b. (2011). when is a liability not a liability? textual analysis, dictionaries, and 10-ks. the journal of finance, 66(1), 35–65. https://doi.org/10.1111/j.1540-6261.2010.01625.x mahlberg, b., freund, i., cuaresma, j. c., & prskawetz, a. (2013). the age-productivity pattern: do location and sector affiliation matter? the journal of the economics of ageing, 1, 72–82. https://doi.org/10.1016/j.jeoa.2013.08.002 merkl-davies, d. m., & brennan, n. m. (2007). discretionary disclosure strategies in corporate narratives: incremental information or impression management? journal of accounting literature, 27, 116– 196. merkl-davies, d. m., brennan, n. m., & mcleay, s. j. (2011). impression management and retrospective sense-making in corporate narratives: a social psychology perspective. accounting, auditing & accountability journal, 24(3), 315–344. https://doi.org/10.1108/09513571111124036 mikolov, t., sutskever, i., chen, k., corrado, g. s., & dean, j. (2013). distributed representations of words and phrases and their compositionality. advances in neural information processing systems, 26. molyneux, p., & thornton, j. (1992). determinants of european bank profitability: a note. journal of banking & finance, 16(6), 1173–1178. https://doi.org/10.1016/0378-4266(92)90065-8 neu, d., warsame, h., & pedwell, k. (1998). managing public impressions: environmental disclosures in annual reports. accounting, organizations and society, 23(3), 265–282. https://doi.org/10.1016/s0361-3682(97)00008-1 nguyen, b. d., & nielsen, k. m. (2010). the value of independent directors: evidence from sudden deaths. journal of financial economics, 98(3), 550–567. https://doi.org/10.1016/j.jfineco.2010.07.004 oliveira, j., azevedo, g., & borges, f. (2016). impression management and self-presentation dissimulation in portuguese chairperson’s statements. corporate communications: an international journal, 21(3), 388–408. https://doi.org/10.1108/ccij-11-2015-0074 pervan, m., & višić, j. (2012). influence of firm size on its business success. croatian operational research review, 3(1), 213–223. petria, n., capraru, b., & ihnatov, i. (2015). determinants of banks’ profitability: evidence from eu 27 banking systems. procedia economics and finance, 20, 518–524. https://doi.org/10.1016/s22125671(15)00104-5 phesa, m. (2021). impression management observation in chairperson statements in jse top 40 listed companies. pyatt, g. (1966). human capital: a theoretical and empirical analysis, with special reference to education. oxford university press. https://doi.org/10.2307/2229541 quoc trung, n. k. (2021). determinants of bank performance in vietnamese commercial banks: an application of the camels model. cogent business & management, 8(1), 1979443. https://doi.org/10.1080/23311975.2021.1979443 https://doi.org/10.1037/0033-2909.107.1.34 https://doi.org/10.1111/j.1475-679x.2010.00382.x https://doi.org/10.1088/1742-6596/1168/3/032024 https://doi.org/10.1088/1742-6596/1168/3/032024 https://doi.org/10.1111/j.1540-6261.2010.01625.x https://doi.org/10.1016/j.jeoa.2013.08.002 https://doi.org/10.1108/09513571111124036 https://doi.org/10.1016/0378-4266(92)90065-8 https://doi.org/10.1016/s0361-3682(97)00008-1 https://doi.org/10.1016/j.jfineco.2010.07.004 https://doi.org/10.1108/ccij-11-2015-0074 https://doi.org/10.1016/s2212-5671(15)00104-5 https://doi.org/10.1016/s2212-5671(15)00104-5 https://doi.org/10.2307/2229541 https://doi.org/10.1080/23311975.2021.1979443 218 tam phan huy et al. ratanavararak, l., & ananchotikul, n. (2018). bank profitability and risk-taking in a low interest rate environment: the case of thailand. puey ungphakorn institute for economic research discussion paper, 89. rose, p. s., & hudgins, s. c. (2008). bank management and financial services. mcgraw-hill. sahul hamid, f. (2017). the effect of market structure on banks’ profitability and stability: evidence from asean-5 countries. international economic journal, 31(4), 578–598. https://doi.org/10.1080/10168737.2017.1408668 saunders, a., & schumacher, l. (2000). the determinants of bank interest rate margins: an international study. journal of international money and finance, 19(6), 813–832. https://doi.org/10.1016/s0261-5606(00)00033-4 smith, m., & taffler, r. (1992). the chairperson’s statement and corporate financial performance. accounting & finance, 32(2), 75–90. https://doi.org/10.1111/j.1467-629x.1992.tb00187.x smith, m., & taffler, r. j. (2000). the chairperson’s statement: a content analysis of discretionary narrative disclosures. accounting, auditing & accountability journal, 13(5), 624–647. https://doi.org/10.1108/09513570010353738 solomon, j. (2020). corporate governance and accountability. wiley. spence, m. (1978). job market signaling. in uncertainty in economics (pp. 281–306). elsevier. https://doi.org/10.1016/b978-0-12-214850-7.50025-5 suto, m. (2003). capital structure and investment behaviour of malaysian firms in the 1990s: a study of corporate governance before the crisis. corporate governance: an international review, 11(1), 25–39. https://doi.org/10.1111/1467-8683.00299 tan, h., wang, s., & welker, m. (2011). analyst following and forecast accuracy after mandated ifrs adoptions. journal of accounting research, 49(5), 1307–1357. https://doi.org/10.1111/j.1475679x.2011.00422.x van der laan smith, j., adhikari, a., & tondkar, r. h. (2005). exploring differences in social disclosures internationally: a stakeholder perspective. journal of accounting and public policy, 24(2), 123– 151. https://doi.org/10.1016/j.jaccpubpol.2004.12.007 xie, x., & wang, s. (2023). digital transformation of commercial banks in china: measurement, progress and impact. china economic quarterly international, 3(1), 35–45. https://doi.org/10.1016/j.ceqi.2023.03.002 zerban, a. m., & ateia, w. (2016). corporate governance in the banking sector (empirical study on the effect of separating chairperson and chief executive officer (ceo) positions on financial performance). accounting and finance research, 5(3), 37–47. https://doi.org/10.5430/afr.v5n3p37 https://doi.org/10.1080/10168737.2017.1408668 https://doi.org/10.1016/s0261-5606(00)00033-4 https://doi.org/10.1111/j.1467-629x.1992.tb00187.x https://doi.org/10.1108/09513570010353738 https://doi.org/10.1016/b978-0-12-214850-7.50025-5 https://doi.org/10.1111/1467-8683.00299 https://doi.org/10.1111/j.1475-679x.2011.00422.x https://doi.org/10.1111/j.1475-679x.2011.00422.x https://doi.org/10.1016/j.jaccpubpol.2004.12.007 https://doi.org/10.1016/j.ceqi.2023.03.002 https://doi.org/10.5430/afr.v5n3p37 1. introduction 2. literature review 3. methodology 4. results and discussion 5. conclusions and recommendations references ©the author(s) 2025. this is an open access article distributed under the terms of the creative commons attribution-noncommercial 4.0 international license vol. 14, no. 1 (2025), pages 21-54 https://doi.org/10.17979/ejge.2025.14.1.11102 submitted: jul 26, 2024 accepted: jan 17, 2025 published: jun 20, 2025 article behaviorally-informed policies to reduce materialism among youth in egypt: experimental evidence sidrah khalil,1,* sarah mansour,1 marwa shibl biltagy 1 1 faculty of economics and political sciences, cairo university, egypt *correspondence: sedra_nasr2019@feps.edu abstract. materialism, characterized by prioritizing material possessions over intrinsic values, has been shown to negatively impact well-being, social behavior, and economic activities. however, limited research addresses interventions to reduce materialism in non-weird societies, particularly in the middle east. this study examines the effectiveness of behavioral interventions in lowering personal materialism among egyptian youth, a demographic displaying significant materialistic tendencies. utilizing a survey experiment with 296 participants from egypt’s 25 governorates, the study employed richins and dawson’s material values scale (1992) to measure materialism. participants were randomly assigned to one of three interventions: boosting self-esteem, fostering gratitude, or promoting empathy, in addition to a control group. results indicate that interventions targeting self-esteem and empathy significantly reduced materialism overall score (p < 0.05) and (p < .01), respectively, and materialism score was associated with centrality, while self-esteem, gratitude, and empathy notably lower materialism in dimensions associated with success. however, no significant impact was observed on materialism's score. these findings highlight the potential of governance measures and behaviorally-informed policies in addressing materialism and suggest tailored interventions to promote intrinsic values among youth. keywords: behavioral interventions; government policies; materialism; egypt; youth jel classification: c9; z13; z18 1. introduction materialism is a process of preference. it is the tradeoff between materialistic and non-materialistic values and gains; the inclination to consider material acquisitions, possessions, and physical comfort as of higher importance than spiritual values (fournier & richins, 1991). the concept is a purely humanistic perspective as it entails a deep unmet psychological need and insecurity in which people use material objects to fulfil it. high levels of materialism can stimulate consumer desire to some degree and stimulate achievement motivation, which in turn stimulates commodity demand (sirgy et al., 2013). for consumers, younger people with higher levels of materialism tend to shop more often, have more knowledge about products and services, and are most sensitive to advertising and promotions (goldberg et al., 2003). however, the economic literature shows that this phenomenon https://creativecommons.org/licenses/by-nc/4.0/ 22 khalil, mansour and biltagy has a negative impact on happiness and wellbeing (roberts & clement, 2007; dittmar et al. 2014; kasser, 2018). it leads to irrational behaviors (consumerism, compulsive buying behaviors, low social giving, environmental neglect) and negative economic and social consequences (high indebtedness and low satisfaction from work and marriage). even though materialism is not dependent upon affluence (ger and belk, 1996), it has been proven that levels of materialism are higher in poorer countries than in richer ones (delhey, 2010; moldes & ku, 2020). focusing on egypt, the most populous country in the mena region and the second largest economy in africa (imf, 2024), there are several manifestations of high levels of materialism. as per the world values survey (wvs), which uses the post-materialism index of inglehart (1971), egyptians show extreme skewness towards wanting their government to focus on materialistic goals at the expense of post-materialistic ones. the survey was designed to show the level of consistency in choosing materialistic items at the expense of post-materialistic items in several groups of questions. data from the latest world values survey in 2020 for 54 countries shows that egypt is the top country in terms of the percentage of respondents who chose only materialistic national goals as a priority; egypt (29.9%) is followed by russia (21.0%), china (22.2%), tajikistan (18.9%), ethiopia (18.8%), tunisia (18%), jordan (17%), taiwan (16.9%), zimbabwe (14.7%), myanmar (13.2%), greece (13%), hong kong (12.6%), and argentina (11.6%). data on personal materialism – linking materialism to one’s personal view and practice of life – in egypt is rarely found. however, there are manifestations of high levels of materialism in the egyptian society, such as the belief that material possessions represent success and self-fulfillment (adib & el-bassiouny, 2012), the high level of consumerism (abaza, 2001; abaza, 2006; el-bassiouny et al., 2011; zayed, 2019), and the high level of social hatred and the rupture of social ties (okasha et al., 2009). in addition, spiritual and religious dimensions are receding (arab barometer, 2019) in favor of the idea of physical satisfaction, and happiness is becoming highly linked with material consumption, as cited in adib and el-bassiouny (2012). especially for youth, who constitute around 21% of the total egyptian population, levels of materialism are getting higher. unnecessary personal needs are turning into indispensable necessities, and the aim of consumption is becoming more for the fulfillment of social and psychological needs (el din & el sahn, 2013). in egypt, material prestige is expressed through a package of expensive goods and services, an actual or fake luxurious lifestyle, and special entertainment venues, for instance, in food (ammar et al., 2016; dawoud, 2014) and clothing (talaat, 2020; ahmed, 2018). add to that, religious holidays are becoming of less importance to focus on spiritual and human implications, but shifted to an occasion for extra consumption of unnecessary items; for instance, consumption in ramadan, a month of fasting to teach patience and asceticism, nearly doubles, according to the internal trade development authority itda (2020). while the reasons behind this phenomenon vary, access to technology leading to continuous social comparisons is a major recent induction that caused many to financially exhaust themselves. connections between material acquisition, and happiness and success are forming at the expense of personal mental health and family duties (henry & elwy, 2020). this study focuses on egyptian youth as a research object for several reasons. first, egypt is a non-weird (western, educated, industrialized, rich, democratic) context, and much of the existing literature on materialism has been concentrated in weird societies, leaving a gap in understanding behaviorally-informed policies to reduce materialism the phenomenon in culturally distinct settings. second, egyptian youth are particularly vulnerable to the effects of materialism due to the rapid socio-economic transformations, increased exposure to globalized media, and heightened societal pressures to conform to materialistic ideals. third, young people represent a critical demographic for shaping future consumption patterns, social attitudes, and cultural norms, making them an essential group for targeted interventions aimed at reducing materialism. lastly, focusing on youth is especially important given their role in driving societal change; their attitudes and behaviors can have long-lasting implications on economic stability, social cohesion, and cultural values. this study uses data collection methods to explore personal materialism and imply behavioral interventions through an experiment aimed at testing interventions that can play a role in reducing the level of personal materialism among youth. the hypothesis of this study is that the interventions used, which will mainly focus on priming youth with spiritual and pro-social values and boosting their self-esteem and sense of security, will reduce the level of materialism. these hypotheses are derived from behavioral economics and the nudge theory, and the results are aimed to help formulate behaviorally-informed policies to reduce materialism in egypt. the paper is divided into six sections. following this introduction, the paper presents a literature review, followed by a section for data analysis, and a section presenting the results. the paper then provides a section discussing the results, and a section for conclusions. 2. literature review the concept of materialism theories of materialism define materialism as a failure to meet a psychological need of a higher order, such as a healthy relationship or a healthy view of oneself (kasser 2002; kasser and ryan 1993; wong et al., 2003). thus, purchasing a high-status object is a way of compensation that people with feelings of professional or personal inadequacies use (sivanathan & pettit, 2010; rucker & galinsky, 2008). the definition of insecurity is that it is a state of being open to danger or threat and a feeling of inadequacy and uncertainty. having an insecurity pushes people to try to fill it with material objects that cannot replace love or affection, emotional support, or empathy (burroughs and rindfleisch 2002; kasser 2002). as a result, people often find themselves isolated and unhappy (kasser 2002; richins and dawson 1992). theories of basic needs the need from which all other needs stem – are many. however, the self-determination theory (deci & ryan, 1985; ryan & deci, 2000), the acquired needs theory (mcclelland, 1987), and maslow’s needs hierarchy (maslow, 1970) are the three most relevant to the study of materialism. all three theories have one idea in common: there is a set of core needs in every human being. if they are not met, insecurity emerges. these needs can be classified into physical and economic security, self-needs, and social needs. in terror management theory (arndt et al., 2004; rindfleisch et al., 2009), researchers added existential needs and spirituality. the ‘crowding out’ theory of kasser (2002) explains how materialistic people focus their energies and interest on getting material objects at the expense of (crowding out) satisfying healthy needs, which deepens the insecurity by creating a vicious cycle that causes subjective well-being and happiness to 24 khalil, mansour and biltagy diminish. building on that, in order to reduce materialism, interventions should focus on countering these psychological insecurities and fulfilling these needs. consequences of materialism regarding social consequences, the primary consequence of materialism is that it fails to create real happiness or satisfaction. the endless pursuit of material matters is similar to drug addiction (slater, 1980). similar to drugs, individuals often find themselves in need of larger doses and more frequent consumption to maintain the false temporary mood lifting (schudson, 1984), keeping people in a trap such that happiness becomes, in the end, unattainable (brickman & campbell, 1971). materialism has a significant negative correlation with satisfaction in many life domains (roberts & clement, 2007). specifically, there is a negative robust relationship between materialism and subjective well-being (dittmar et al. 2014). using different ways to measure materialism, it is constantly found to have a negative relationship with wellbeing (puente-díaz & cavazos-arroyo, 2019); materialism measured as a trait (ahuvia & wong, 2002; belk, 1984; 1985; dawson, 1988); materialism measured as a personal value (burroughs & rindfleisch, 2002; richins & dawson, 1992; shrum et al., 2011; sirgy et al., 2012); and materialism measured as extrinsic goal attainment (sheldon & kasser, 1995, 1998; kasser, 2002). it appears that merely priming thoughts about luxury, for example, just by having people view luxury products, seems to cause people to be less happy (shrum et al., 2022). in addition, measuring materialism as a trait was found to be positively associated with experiences that are considered depressive, such as dependency or harsh self-criticism (wachtel & blatt, 1990) and with social anxiety (schroeder & dugal, 1995). when measured as a personal value, it was also found to be positively associated with neuroticism (burroughs & rindfleisch, 2002; mick, 1996), with depression, anxiety, and high levels of stress (burroughs & rindfleisch, 2002). materialism as an extrinsic goal pursuit also correlates positively with a host of problems, including negative affectivity (kasser et al., 2014), depression (kasser & ryan, 1993, 1996; yamaguchi & halberstadt, 2012), and anxiety (kasser et al., 2014). pieters (2013) concluded that loneliness promotes materialism, which in turn contributes to greater loneliness. moreover, lebaron et al. (2018) found that materialism was negatively associated with perceptions of marriage's importance and was associated with lower marital satisfaction. carroll et al. (2011) found that married couples who are high in the degree of materialism have significantly lower quality marriages when compared to married couples who are both low in the degree of materialism or when compared to couples with different degrees of materialism. king & datu (2017) showed that materialistic students have lower motivation, engagement, and achievement. in addition, highly materialistic children and adolescents perform poorer at school (goldberg et al., 2003), and are more likely to use tobacco or marijuana (williams et al., 2000). they are also more likely to adopt unethical thoughts and beliefs, in addition to behaviors, compared to their peers with lower degrees of materialism (gentina et al., 2018a, 2018b). a study conducted on children by chan (2004) showed that children perceive possessions as something that brings fun and friends, but also selfishness, arrogance, and envy. highly materialist people are more likely to consider themselves spenders; they are more behaviorally-informed policies to reduce materialism likely to spend rather than save; and they are more likely to borrow (watson, 2003; maison & adamczyk, 2020). even with equal income levels, brown et al. (2016) found that high materialists made more frequent purchases, spent more on their purchases, and had higher levels of “letdown” after spending compared to lower materialists. materialism predicts a person's likelihood of incurring consumer debt (ponchio & aranha, 2008); additionally, personal debt and account balances were found to be positively associated with materialism (nepomuceno & laroche, 2015). de matos et al. (2019) found that materialism has a significant effect on consumer indebtedness, in which higher impulsiveness triggers materialism and influences debt levels. dittmar (2005) and ridgway et al. (2008) found a positive correlation between materialism and compulsive buying. mueller et al. (2011) found that materialism and depression jointly influence compulsive buying, and donnelly et al. (2013) also found that materialism correlates positively with compulsive buying and that a lack of money management is a mediator between materialism and compulsive buying. gararsdóttir & dittmar (2012) showed that people who have higher materialistic values have greater financial worries, worse skills in money management, and a greater inclination towards compulsive buying and spending. as for the consequences in the labour market, deckop et al. (2010) found that materialistic values are negatively associated with multiple indicators of personal wellbeing, such as job satisfaction, career satisfaction, and reward satisfaction. in line with this, unanue et al. (2017) found that materialism at work is negatively associated with work satisfaction and engagement and more positively with turnover intentions and burnout. on the environmental consequences, kilbourne & pickett (2008) and kasser (2018) found that materialism is negatively associated with pro-social and pro-environmental attitudes and wellbeing. research also suggests that materialism is associated with greater environmental neglect. materialistic individuals consider the protection of natural environments as a low priority (clump et al., 2002; kilbourne & pickett, 2008) and are more likely to consume a higher amount of natural resources (winter, 2004). experimental studies studies have used experimental tools to understand what induces or dampens materialistic values and behaviors. studies that explain what induces materialism have concluded that threatening selfesteem (chang & arkin, 2002) or threatening personal power (rucker & galinsky, 2008; rustagi et al., 2016) promotes materialism and the desire for luxury and status goods and the willingness to pay more for them (sivanathan & pettit, 2010), and increases consumption in general (lee & shrum, 2012; lee et al., 2017; rucker & galinsky, 2009). boosting self-esteem, on the other hand, decreases materialism (chaplin & john, 2007; park & john, 2011), especially in self-threatening situations (yang et al., 2021). similar to that, when people’s feelings towards the meaning of life were threatened by reminding them of their death (mandel & heine, 1999; kasser & sheldon, 2000) or by making them feel ignored (lee & shrum, 2012; lee et al., 2017), people showed higher signs of greediness, a higher desire to purchase luxury products, and higher logos. in addition, threatening people’s feelings of belonging increases loneliness, which is compensated for through spending on other people (pieters, 2013). also, social exclusions have been found to increase preferences for nostalgic products (loveland et al., 2010), spending on partners (mead et al., 2011), and on charity 26 khalil, mansour and biltagy (lee & shrum, 2012; lee et al., 2017). materialistic concerns, when activated, cause disturbances in psycho-social mechanisms that are vital for building healthy social ties; trust in other people; desire to engage socially; and feelings of responsibility (bauer et al., 2012). lens et al. (2010), through experiments, showed that advertisements induce materialism, especially among individuals who can afford the advertised product. in addition, the study showed that richer individuals have fewer problems associated with materialism as they have higher self-esteem. interestingly, when consumers imagined themselves already purchasing the products, the advertisement had no impact. sayah (2024), found that advertising negatively impacts children from lower social classes, leading to lower self-satisfaction, lower self-esteem and higher materialistic tendencies. however, the impact of materialism was found to be stronger amongst the wealthier sample included. zhao et al. (2023), using a sample from lebanon, found that lower self-esteem, anxious attachment and poorer quality of family relationships predicted higher levels of materialism, with self-esteem showing the strongest association. similarly, using a sample from uk, children from deprived backgrounds were found to be more materialistic than wealthier ones (nairn & opree, 2021). similarly, trzcińska et al. (2024) found that materialism in preschool children is influenced by selfesteem. studies have also tackled interventions aimed at decreasing materialism. materialistic goal orientation can decrease if individuals are encouraged to set intrinsic goals and keep reflecting on self-transcendence values (parker et al., 2020). lekavičienė et al. (2022) demonstrated that emotional intelligence training effectively reduces consumer materialism. studies show that materialism can decrease when interventions that incorporate inward reflections or incentivize intrinsic values are used, or when interventions that lead individuals to rethink and analyse the “consumer culture” or detach from it are used (kasser, 2018). results from short-term experiments (dechesne et al., 2003; weinstein et al., 2009; fritsche et al., 2010; stillman et al., 2012) have found that spiritual, pro-environmental, and/or pro-social values can help decrease materialism. lekes et al. (2012) sent multi-week emails to the intervention group reminding them to reflect their intrinsic values. the study found that participants decreased in materialism and increased in wellbeing. in what the authors called a "diminishing of desire," joye et al. (2020) found that exposure to nature compared to urban environments weakened materialism. sheldon et al. (2003) found that when people are allowed to reflect inwardly by pausing and thinking, they reconsider their values and place higher importance on intrinsic ones. in addition, mindfulness is a type of inward reflection that decreases materialism (brown & kasser, 2005). it boosts satisfaction with financial situations and well-being (brown et al., 2009). moreover, mindfulness is a therapeutic technique that has been used to impact materialism. mindfulness is a state of mind of acceptance, which is the acceptance of experiences and emotions. some research has found that deep, continuously inward reflections on death can reduce materialism (cozzolino et al., 2004; lykins et al., 2007), particularly among people who are more open to new experiences (prentice et al., 2017). in a similar vein, gratitude as an affective trait, mood, or emotion has been linked to lower materialism in a number of studies. it is considered an affective trait that, in general, lowers one’s probability of experiencing negative emotions (rosenberg, 1998). feelings of gratitude entail a recognition of having a positive outcome and that an external agent is responsible for it. feelings of gratitude are implicitly prosocial, as they entail a recognition of other people’s actions and kindness behaviorally-informed policies to reduce materialism in making one’s life better and may encourage similar behaviours in return (bertocci & millard, 1963; mccullough et al., 2001). that is why it is sometimes called the "empathic emotion" (lazarus & lazarus, 1994). gratitude reflections cause people to de-prioritize materialistic values (lambert et al., 2009). saleem et al. (2022) found that gratitude positively impacts subjective well-being among youth, but this relationship is moderated by materialism, with higher levels of materialism weakening the positive effects of gratitude. lee et al. (2022) found that gratitude reduces materialism through decreased entitlement and enhanced perceived resources. in their study, chaplin et al. (2019) showed that there is a negative association between gratitude and materialism in children and adolescents, and that keeping a gratitude journal reduces materialism and also eases its negative effect on generosity; participants who were exposed to the intervention donated 60% more. similarly, unanue et al. (2021) found that gratitude at work prospectively reduces workplace materialism. polak & mccullough (2006) discussed how materialistic strivings are lower in people who are grateful because, usually, grateful people are more complete and more secure. in addition, evidence shows that gratitude is negatively associated with envy and with materialistic behaviours and positively with spirituality and pro-social attitudes (mccullough et al., 2002). this is all because gratitude is a positive emotion (mayer et al., 1991; ortony et al., 1990; weiner, 1986) and is associated with higher wellbeing (emmons & mccullough, 2003; watkins et al., 2003). from another perspective, kasser et al.'s (2014) study showed that adolescents decreased in materialism and increased in self-esteem when they, and their parents, received sessions to help them create a spending plan that is based on values and were encouraged to question the consumer culture around them. buijzen (2007) and buijzen and valkenburg (2005) also showed that children’s levels of materialism decline when their parents explain and criticise advertising intentions. policy interventions many governmental attempts have been made to combat materialism levels mainly through campaigns such as the “buy nothing day” in sweden, finland, the uk, and north america, and the “international downshifting week” and the “turn off your tv week”. in many countries, governments and banks have given more attention to financial literacy programmes and credit counselling to rationalise consumer spending, in addition to passing the necessary laws to discourage materialism. levying taxes on excess consumption or on luxury and status goods that have an impact on income distribution, as cited by bagwell and bernheim (1996), was one of the failed trials. the failure of this policy and similar polices is due to the lack of understanding of the behavioral analysis of materialism. policies that do not target fulfilling the original insecurity that has caused the high level of materialism fail against the massive effort exerted by companies seeking profits (burroughs et al., 2013). levitt and dubner (2005) criticise tax levying on status goods because taxes give people a “moral license” to consume, as they believe they have already paid more, so they deserve to continue in their behavior. moreover, combating materialism through narrowing the wealth gap may also be ineffective, as it triggers competition and encourages people to link their worth to what they acquire (ordabayeva & chandon, 2011). it is also worth noting that encouraging charitable behavior as a policy doesn’t necessarily reduce materialism, as charitable behavior can 28 khalil, mansour and biltagy co-exist with materialism (thibaut & kelley, 1959; homans, 1958; emerson, 1976; pitts & skelly 1984; mathur, 1996; clary et al., 1998; park & john, 2010; mathur, 2013; bock, 2018). however, there are several policies that are considered effective, such as increasing parental time with children and limiting children's exposure to advertising and marketing (burroughs et al., 2013). as for increasing parental time, danish companies are positive examples as they have some of the most family-friendly work policies in the world; denmark has 34 days of paid vacation and holiday leave per year compared to the united states, which gives only 10 days (www.mercer.com). similarly, denmark has an average of 52 paid weeks of parental leave compared to 24 weeks in the us (http://www.cepr.net). as for children's exposure to advertising and marketing offers, kasser & linn (2016) believe that the most effective policy would be for governments to prohibit all marketing and advertising aimed at children, which is a policy adopted by quebec (a canadian province) and some few countries such as norway, sweden, and brazil. however, burroughs et al. (2013) believe that it may not be realistic to completely prohibit marketing and advertising to children, but it can be regulated. for example, issuing legal acts that require companies to disclose the existence of any marketing or advertising in television programmes or movies or video games and all other media platforms presented to children (the us children’s television act of 1990 even prohibits the incorporation of products in any scenes or animated characters) or eliminating incentives or imposing taxes on companies that create marketing materials for children. kasser and linn (2016) advocate for the restriction of promotional messages in physical places frequented by children; there are some existing acts that require advertising of unhealthy products, such as alcohol and cigarettes, to be placed a certain distance away from schools; and france, brazil, vermont (a state in the united states), grenoble, and sao paolo have significantly limited outdoor advertising. 3. data analysis hypotheses on many occasions, neglecting the behavioral insights can cause the public policy to fail (hansen & jespersen, 2013). this happens because, from the behavioral economics view, the process of decision making sometimes leads individuals to fail in acting on their own well-informed selfinterest, or society’s general interests (thaler et al., 2013). this issue has been addressed by richard thaler, a nobel prize winner in economics, who introduced the “nudge theory." the theory uses human beings' stereotypes, biases, and errors as a tool to implicitly alter their behaviour (dolan et al., 2010). it explains that positive nudges to the environment around individuals can effectively influence their behaviour and decision-making through cost-effective and behaviorally informed policies. thaler defines a "nudge" as “any aspect of the choice architecture that alters people’s behaviour in a predictable way without forbidding any options or significantly changing their economic incentives” (thaler & sunstein, 2008, p. 6). in his proposition, policy makers can become choice architects, “a choice architect has the responsibility for organising the context in which people make decisions” (thaler & sunstein, 2008, p. 3). behaviorally-informed policies to reduce materialism theories of materialism define materialism as a failure to meet a psychological need of a higher order, such as a healthy relationship or a healthy view of oneself (kasser 2002; kasser and ryan 1993; wong et al., 2003). the unfulfilled need creates a sense of insecurity that people try to fill with material objects that cannot replace love or affection, emotional support, or empathy, which fulfil their basic needs (burroughs and rindfleisch 2002; kasser 2002). as a result, people often find themselves isolated and unhappy (kasser 2002; richins and dawson 1992). building on that, in order to reduce materialism, interventions should focus on countering these psychological insecurities and fulfilling these needs. the choice of interventions is based on the literature review of previous experiments used to discourage materialism. interventions tested are going to be helpful in suggesting the formulation of targeted policies or encouraging the activation of existing ones in order to raise awareness and/or fight against materialism. building on the empirical and theoretical frameworks, this thesis sets the following hypotheses: h0: exposure to the intervention does not reduce personal materialism. h1: exposure to intervention 1, boosting self-esteem, reduces the level of personal materialism. h2: exposure to intervention 2, exercising gratitude, reduces the level of personal materialism. h3: exposure to intervention 3, stimulating empathy, reduces the level of personal materialism. data collection scholars have relied on surveys to measure materialism correctly. belk (1984) and dawson (1992) were the early researchers who designed fundamental surveys to measure materialism. belk’s (1984) questionnaire measures materialism based on the existence of three personality traits: nongenerosity, envy, and possessiveness. however, richins and dawson’s (1992), also known as the material values scale (mvs), measure materialism by looking at the extent to which people think that (i) material goods cause happiness, (ii) material goods are a central part of life, and (iii) possessions and success are related, see table 1. richins and dawson’s (1992) survey is utilized in this study. the data collection procedures are presented in table 2. this study uses a survey experiment to identify incentives that can have an impact on personal materialism among youth. the study uses online random selection from various online social media groups. the survey has 18 items covering three major dimensions: success, centrality, and happiness. the items are scored on a 5-point likert scale from strongly agree to strongly disagree, which are summed within each component and re-summed to form an overall score. all collected observations were appropriately coded. some items had to be recorded as negative values to correctly sum the overall score of materialism. using google forms, links were sent online, randomly assigning people to control and different intervention groups. targeted respondents are 18 to 29 years old males and females, who have at least completed high school, and egyptians residing in any egyptian governorate. the reason behind the choice of the sample is the fact that young people's relationships with money, possessions, and physical aspects of life are likely to persist when they get older (borg et al., 2017). concepts and perceptions developed in the early stages of life 30 khalil, mansour and biltagy only practically manifest and become tested when individuals start being financially independent by the age of 18. youths between the ages of 18 and 29 are expected to take full or partial control of their finances and develop a spending pattern that cannot be easily changed later in life (krosnick & alwin, 1989; milfont et al., 2016). table 1. richins and dawson’s survey of personal materialism (1992) *,** success centrality happiness 1. i admire people who own expensive homes, cars, and clothes. 2. some of the most important achievements in life include acquiring material possessions. 3. i don't place much emphasis on the amount of material objects people own as a sign of success. 4. the things i own say a lot about how well i'm doing in life. 5. i like to own things that impress people. 6. i don't pay much attention to the material objects other people own. 1. i usually buy only the things i need. 2. i try to keep my life simple, as far as possessions are concerned. 3. the things i own aren't all that important to me. 4. i enjoy spending money on things that aren't practical. 5. buying things gives me a lot of pleasure. 6. i like a lot of luxury in my life. 7. i put less emphasis on material things than most people i know. 1. i have all the things i really need to enjoy life. 2. my life would be better if i owned certain things i don't have. 3. i wouldn't be any happier if i owned nicer things. 4. i'd be happier if i could afford to buy more things. 5. it sometimes bothers me quite a bit that i can't afford to buy all the things i'd like. *the non-bold items range from 1 to 5 and the bold items are negative items ranging from -1 to -5. **highest materialism score possible is 48, and lowest materialism score possible is -24. **from -24 to 0 is considered low materialism, 1 to 24 is medium, and 25 to 48 is high. table 2. data collection procedures 1. translate richins and dawson’s survey into arabic 2. create four google forms that have sections a, b, and c as follows: form 1 section a (control): form 2 section a (others): form 3 section a (surroundings): form 4 section a (oneself): not applied task: view images with written messages on them that activate empathy (specifically wanting to help) and prosocial feelings and answer questions about how it made them feel task: write down 10 brief sentences about people/agencies/god they feel grateful for task: rewrite 10 scrambled sentences that have deep meanings about one’s self value section b: translated richins and dawson’s questionnaire section b: translated richins and dawson’s questionnaire section b: translated richins and dawson’s questionnaire section b: translated richins and dawson’s questionnaire section c: demographic questionnaire section c: demographic questionnaire section c: demographic questionnaire section c: demographic questionnaire 3. send to participants behaviorally-informed policies to reduce materialism the e-links of the four survey experiments were sent directly on social media channels such as, whatsapp and facebook (after getting the permission from the group admins). these groups were chosen based on the targeted age group. examples include but are not limited to: groups dedicated to finding jobs for fresh graduates, student unions, graduate studies, youth volunteering, youth sports, etc. in order to avoid a subject accessing more than one group, respondents were asked to enter their phone number on the first link that they choose, the same phone number is not accepted again in any of the other links. once one survey of the four is filled with 74 respondents, it can no longer accept more participants experimental methods in the field of economics have been widely used over the past few decades. these methods give researchers an opportunity to quantify what was previously considered unquantifiable. in its traditional form, an experiment would be conducted by gathering participants in a room and exposing them to a stimulus. nowadays, these experiments can be conducted in a computer laboratory. the main aim of an experiment is to find causation rather than correlation. the causation is found by assigning two different groups, one group being a control group that does not receive any intervention, and the other being a treatment group that receives the designed intervention. the outcomes of the two groups are expected to differ only with respect to the explanatory values of the experiment; the controlled variation of one variable while keeping the other environmental conditions fixed. survey experiments are one of the experimental methods widely used in economics. they are cost-effective, easier to conduct, and cover bigger sample sizes as they do not require in-person contact to be implemented (hainmueller et al., 2014). there are several types of survey experiments. this study uses the priming survey experiment, which is an experiment that is conducted within a survey used to collect sensitive information. (druckman et al., 2011). this research uses the betweensubjects design, in which there are no before and after estimations. the target of this tool is to compare two groups of participants; one group to which an intervention is administered (for example, exposed to a stimulus) and the other group that receives no intervention. a behavioral experiment does not need to be generalised to represent samples of the population. instead, scholars of experimental economics replicated their experiments on other samples. the results of randomised control trials have a high level of internal validity, achieved through randomization of the sample. internal validity is the extent to which a study establishes a trustworthy cause-and-effect relationship between a treatment and an outcome. whereas regression models find correlations, randomization of an experiment ensures finding clean casualties. randomization ensures no dependence between the assignment in the intervention group and the characteristics of the subjects. however, a balance test should be used before conducting the analysis. the baseline balance test ensures that allocation of subjects to intervention or control groups is left purely to chance; in other words, that the groups are similar to each other. individuals are randomly assigned to both the control and treatment groups. while materialism is not a very sensitive topic compared to other sensitive topics such as racism or illegal practices, many sensitive topics appear so sensitive that the study subjects' conscious and explicit behaviour may be different from what they implicitly hold inside (greenwald & banaji, 1995). in many cases, certain topics are considered beyond the subject's conscious awareness, even if they are not considered sensitive (nisbett & wilson, 1977). priming experiments 32 khalil, mansour and biltagy work by guiding the conscious mind away from topic a and towards topic b so that subjects unconsciously change their minds about topic a (macrae et al., 1994; schwarz & clore, 1983). the three interventions employed focus on three basic dimensions: others, surroundings, and oneself. the first dimension, "others,” is tested by exposing participants to prosocial and public service advertising before taking the survey. prosocial advertising is aimed at achieving a specific goal that benefits society. advertising should not be about raising money; it should incentivize empathy among the respondents. for example, an image of people in rural areas who need volunteers to help them read and write, or people with disabilities who need physical help to move around the city. drążkowski and trepanowski (2022) found that practicing acts of kindness significantly reduces materialism over time. chao (2019) studied the impact of advertising in china. the study concluded that public service advertising does moderate the relationship between viewing television and the level of materialism. researchers use videos as an effective tool for eliciting emotions. some videos commonly used are some movie or television scenes. usually, a clip is watched and then participants answer a few questions about how informative and interesting it was. video techniques are generally well received by people. it is also beneficial as participants don’t know what the hidden message is or what they are going to watch. it is very important to ask the participants whether they have seen the video before, because if they have, the effect will be dampened. advertising can also take the form of multiple nonmoving images. similar to videos, this tool has been used by researchers to activate certain emotions. commonly, it is used with a cover story. in this study, only three images are included with three messages; helping a senior, donating blood, and listening to a friend. the second dimension, "surroundings,” is tested through fostering gratitude. several studies have linked gratitude scale results to materialism scale results. however, few of them have used direct interventions. froh et al. (2011) conducted their study on youth, lambert et al. (2009) and chaplin et al. (2019), on children and adolescents. they all found that fostering gratitude reduces materialism. participants in this study’s intervention group write down at least 10 sentences about people or agencies or god or whoever they feel grateful for and why, then they take the survey. writing tasks are effective ways to elicit emotions in human beings (cohen et al., 2008). it can manipulate both basic and social emotions (kardes et al., 2019). the writing lets participants recall certain experiences and also use their imagination. the third dimension, "oneself,” is tested by boosting self-worth. participants of this group are given simple mixed words that should be rearranged to form several sentences. the sentences have deep meanings about one’s self-value and worth (for example, i accept who i am: the good and the bad; i am alive and so i have worth; i’m human, not perfect; etc.). they then take the survey. the idea behind mixing the words is to make the participant write down these sentences, which stimulates the brain into believing in them, and it also gives a sense of achievement after correctly forming the sentence. in psychology, this technique is called “supraliminal goal activation" in which participants are exposed to a stimulus they are unaware of that is being used to influence their behaviour (bargh & chartrand, 2014). the scrambled sentence task – developed by costin (1969) – is part of it and has long been used in research. this type of intervention has not been used before in boosting self-esteem in materialism studies. however, generally, several studies have concluded that lower feelings of insecurity are associated with lower levels of materialism. however, few studies have used interventions to boost self-esteem. all of these studies concluded that boosting implicit behaviorally-informed policies to reduce materialism subconscious self-esteem decreases materialism in adults (chaplin & john, 2007; park & john, 2011), and in adolescence (kasser et al., 2014). each of these interventions focuses on different but complementary psychological mechanisms. boosting self-esteem addresses internal self-worth (which reduces insecurities); practicing gratitude shifts focus from external material goods to the present moment (which reduces dissatisfaction with life) and relational values; and stimulating empathy focuses on social and relational connections (which reduces self-centeredness). these interventions could work in harmony through applicable governmental policies to shift youth away from materialistic pursuits toward more intrinsically rewarding goals and directly address core emotional and cognitive processes that fuel materialism. model specification the aim of the study is to find the statistical significance of each intervention utilized on the overall score of personal materialism and each materialism score associated with each component of the survey. using ols, the dependent variable personal_materialismi represents the score attained on the mvs scale, with higher scores representing higher levels of materialism. the independent variables are represented by three dummy variables corresponding to the three experimental treatments, and the control group represents the omitted category. the self-esteem, gratitude, and empathy variables represent the three utilized interventions as follows: 𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃𝑃_𝑀𝑀𝑃𝑃𝑀𝑀𝑃𝑃𝑃𝑃𝑀𝑀𝑃𝑃𝑃𝑃𝑀𝑀𝑃𝑃𝑀𝑀𝑖𝑖= 𝛼𝛼0 + 𝛽𝛽0𝑁𝑁𝑃𝑃𝑁𝑁𝑃𝑃𝑀𝑀𝑃𝑃𝑃𝑃𝑁𝑁𝑃𝑃𝑃𝑃𝑀𝑀𝑀𝑀𝑃𝑃𝑃𝑃0 + 𝛽𝛽1𝑃𝑃𝑃𝑃𝑃𝑃𝑠𝑠 − 𝑃𝑃𝑃𝑃𝑀𝑀𝑃𝑃𝑃𝑃𝑀𝑀1 + 𝛽𝛽2𝑔𝑔𝑃𝑃𝑃𝑃𝑀𝑀𝑀𝑀𝑀𝑀𝑔𝑔𝑔𝑔𝑃𝑃2 + 𝛽𝛽3𝑃𝑃𝑀𝑀𝑒𝑒𝑃𝑃𝑀𝑀ℎ𝑦𝑦3 + 𝛽𝛽4𝑋𝑋4 …𝛽𝛽𝑛𝑛𝑋𝑋𝑛𝑛 + ɛ𝑖𝑖 where 𝛽𝛽4 to 𝛽𝛽𝑛𝑛 are the coefficients of the control variables collected at the end of each survey experiment through a post-experiment questionnaire. the three intervention groups would enter the equation as independent variables. an observation would take 1 in one treatment group and 0 otherwise. the study assumes a linear regression in the coefficients and the error term. the model is applied assuming that the error term has a population mean of zero, the absence of heteroscedasticity, the absence of a correlation between the observations of the error term, and the absence of correlation between all independent variables and the error term. 4. findings using richins and dawson’s survey (1992), data were collected from 296 participants (74 per group) over the course of 4 consecutive months from november 2023 to february 2024. responses came from 25 different governorates across egypt (no responses received from south sinani, new valley, and suez). sample descriptive statistics are presented in table 3. answers were reviewed per response and tasks were insured to be completed correctly. observations failed to complete the task in each intervention group were dropped from the sample. stata 17 software was used to conduct the statistical analysis. 34 khalil, mansour and biltagy table 3. sample definitions and descriptive statistics variable definition obs. mean std. dev. min max materialism combined score of rating survey items (strongly agree (5), agree (4), neutral (3), disagree (2), strongly disagree (1) ) 296 10.442 7.222 -10 33 intervention_1_selfesteem completing the task (1), not completing task (0) 296 0.25 0.433 0 1 intervention_2_gratitude completing the task (1), not completing task (0) 296 0.25 0.433 0 1 intervention_3_empathy completing the task (1), not completing task (0) 296 0.25 0.433 0 1 control_group no completed task (1), completed any task (0) 296 0.25 0.433 0 1 age from 18 to 29 296 21.331 2.651 18 29 gender male (1), female (0) 296 0.537 0.499 0 1 financial_satisfaction very satisfied (5), satisfied (4), neutral (3), not satisfied (2), not satisfied at all (1) 296 2.888 1.183 1 5 life_satisfaction very satisfied (5), satisfied (4), neutral (3), not satisfied (2), not satisfied at all (1) 296 3.304 0.954 1 5 income_5000_orless monthly income of 5000 or more (6), less than 5000 and more than 4000 (5), less than 4000 and more than 3000 (4), less than 3000 and more than 2000 (3), less than 2000 and more than 1000 (2), less than 1000 (1) 296 2.668 1.539 0 6 education bachelor degree student (1), bachelor degree graduate (2), master’s degree student (3), master’s degree graduate (4), phd student or more (5) 296 1.405 0.701 1 5 employed employed (1), unemployed (0) 296 0.371 0.484 0 1 urban_governorate belongs to an urban governorate (1), belong to a rural governorate (0) 296 0.594 0.491 0 1 single being single (1), otherwise (0) 296 0.966 0.180 0 1 living_area_class living in a high-class area (5), middleclass area (4), average area (3), lower than average area (2), common area (1) 296 1.783 0.772 1 4 health excellent (5), very good (4), good (3), poor (2), very poor (1) 296 3.412 0.973 1 5 success the combined materialism score of only items related to the success part of the survey 296 4.070 3.136 -5 13 centrality the combined materialism score of only items related to the centrality part of the survey 296 -4.425 3.384 -12 5 happiness the combined materialism score of only items related to the happiness part of the survey 296 10.797 3.249 1 19 behaviorally-informed policies to reduce materialism table 3 provides sample definitions and descriptive statistics. as shown in the table, the mean score of materialism among youth surveyed across the different intervention and nonintervention groups is positive at 10.44, which indicates a medium score of materialism. the mean age of the sample is about 21 years old, and around 53% of the sample are males. subjective scores of financial satisfaction showed a lower mean (2.8) than life satisfaction (3.3). the average income level is between 2000 and 3000, and the average level of education of the sample is a bachelor's degree. only 37% of the sample are employed, 59% of the sample live in urban governorates, and 96% are single. the average participant either lives in a lower-than-average area or a common area, and reports a good or very good health status. materialism scores for each component of the survey vary greatly. the average in the success component is 4 , the centrality component is 4.4 , and the happiness component is 10.7. table 4 shows the difference in means of materialism scores between interventions and control groups. the highest average mean score is amongst the control group (11.9) followed by intervention number 2 related to fostering gratitude (10.6), followed by intervention number 1 related to incentivizing higher self-esteem (10.2), followed by the lowest intervention related to incentivizing empathy towards others (8.8). table 4. materialism descriptive statistics per group variable obs mean std. dev. min max intervention_1_selfesteem 74 10.2973 7.679772 -9 33 intervention_2_gratitude 74 10.63514 6.4523 -3 28 intervention_3_empathy 74 8.851351 7.325471 -10 25 control_group 74 11.98649 7.173951 -5 28 table 5. variance inflation factors of all variables variable vif 1/vif age 2.54 0.394 education 2.34 0.428 intervention_3_empathy 1.68 0.595 intervention_1_self-esteem 1.59 0.630 financial_satisfaction 1.57 0.636 intervention_2_gratitude 1.54 0.649 employed 1.50 0.664 life_satisfaction 1.45 0.691 single 1.34 0.746 gender 1.30 0.770 living_area_class 1.23 0.812 health 1.21 0.828 income_5000_orless 1.15 0.867 urban_governorate 1.13 0.882 36 khalil, mansour and biltagy balance tests (randomization tests) were conducted prior to performing regression analysis. to compare groups across continuous and ordinal variables, the kruskal-wallis equality-ofpopulations rank test (the non-parametric alternative to anova) was used. p-values for these variables were: age (0.8419), education (.8144), income_5000_orless (0.1541), life_satisfaction (0.7372), financial_satisfaction (0.6765), health (0.1250), and living_area_class (.0524). for dummy variables, a chi-square test was performed. p-values for these variables were: gender (0.924), employed (0.119), urban_governorate (0.718), and single (0.937). these results confirm that our randomization was successful. table 5 shows the variance inflation factors of all variables. according to the table there is no evidence of multicollinearity. tables 6, 7, 8, and 9 present the ols regression results. table 6 displays the impact of the applied interventions on the total materialism score. according to the table, the interventions selfesteem and empathy show a significant negative impact on materialism at p<.05 and p<.01, respectively. similarly, table 6 shows the negative significance of the same interventions of selfesteem and empathy on the score of the centrality component of the survey at p<.10 and p<.01, respectively. however, from table 7, the three interventions, self-esteem, gratitude, and empathy all showed a significant negative impact at p<.10, p<0.01, and p<.01, respectively, on the score of the success components of the survey. table 9 shows no evidence of the impact of any of the interventions on the materialism score of the happiness component of the survey. behaviorally-informed policies to reduce materialism table 6. ols regression output for the impact of the interventions on the total score of materialism. (1) (2) (3) (4) (5) (6) (7) (8) intervention_1_selfesteem -0.194 -0.122 -1.689 -1.689 -1.698 -1.594 -1.693 -1.957* (1.009) (1.079) (1.222) (1.222) (1.221) (1.218) (1.185) (1.149) intervention_2_gratitude 0.216 -1.351 -1.351 -1.343 -1.331 -1.289 -1.524 (0.965) (1.122) (1.122) (1.125) (1.122) (1.090) (1.065) intervention_3_empathy -3.135*** -3.135*** -3.136*** -3.113*** -3.145*** -3.544*** (1.192) (1.192) (1.194) (1.189) (1.167) (1.124) i.control_group age -0.066 3.094 3.23 3.964** (0.162) (2.100) (2.050) (1.917) age_square -0.069 -0.074* -0.093** (0.0457) (0.0446) (0.0419) gender 2.378*** 2.232*** (0.834) (0.795) financial_satisfaction -1.879*** (0.375) life_satisfaction income_5000orless education employed urban_governorate single living_area_class health constant 10.49*** 10.42*** 11.99*** 11.99*** 13.40*** -21.97 -23.98 -25.02 (0.476) (0.609) (0.834) (0.834) (3.661) (23.85) (23.31) (21.80) obs. 296 296 296 296 296 296 296 296 r-squared 0 0 0.024 0.024 0.025 0.033 0.059 0.15 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 38 khalil, mansour and biltagy table 6 (cont.). ols regression output for the impact of the interventions on the total score of materialism. (9) (10) (11) (12) (13) (14) (15) (16) intervention_1_selfesteem -1.814 -2.100* -2.191* -2.375** -2.386** -2.430** -2.283** -2.298** (1.148) (1.132) (1.133) (1.135) (1.135) (1.140) (1.130) (1.133) intervention_2_gratitude -1.524 -1.53 -1.605 -1.63 -1.651 -1.641 -1.496 -1.675 (1.060) (1.063) (1.068) (1.067) (1.066) (1.068) (1.078) (1.085) intervention_3_empathy -3.533*** -3.511*** -3.686*** -3.887*** -3.892*** -3.927*** -3.650*** -3.729*** (1.132) (1.130) (1.132) (1.144) (1.146) (1.146) (1.155) (1.180) i.control_group age 4.145** 3.891** 3.899** 4.196** 4.184** 4.534** 4.443** 4.460** (1.879) (1.832) (1.813) (1.804) (1.802) (1.812) (1.815) (1.813) age_square -0.096** -0.091** -0.095** -0.101*** -0.101*** -0.110*** -0.107*** -0.108*** (0.040) (0.039) (0.038) (0.038) (0.038) (0.038) (0.039) (0.039) gender 2.173*** 1.641** 1.732** 2.100** 2.122** 2.119** 2.074** 1.989** (0.799) (0.801) (0.813) (0.848) (0.848) (0.849) (0.853) (0.873) financial_satisfaction -1.549*** -1.667*** -1.635*** -1.591*** -1.593*** -1.549*** -1.617*** -1.648*** (0.435) (0.428) (0.429) (0.430) (0.430) (0.439) (0.438) (0.432) life_satisfaction -0.77 -0.712 -0.745 -0.77 -0.778 -0.825* -0.855* -0.872* (0.501) (0.471) (0.475) (0.482) (0.481) (0.488) (0.487) (0.493) income_5000orless 0.381 0.374 0.467 0.459 0.458 0.390 0.396 (0.298) (0.296) (0.303) (0.305) (0.312) (0.308) (0.308) education 0.933 1.037 1.034 1.045 0.868 0.891 (0.816) (0.821) (0.823) (0.828) (0.850) (0.856) employed -1.268 -1.292 -1.376 -1.297 -1.226 (0.959) (0.958) (0.962) (0.956) (0.962) urban_governorate 0.276 0.269 0.020 0.009 (0.785) (0.788) (0.807) (0.805) single -2.294 -1.943 -2.009 (2.075) (2.055) (2.074) living_area_class 0.673 0.656 (0.558) (0.565) health 0.182 (0.424) constant -25.7 -24.27 -23.8 -27.68 -27.57 -28.81 -28.85 -29.36 (21.46) (20.89) (20.77) (20.76) (20.74) (20.80) (20.78) (20.69) obs. 296 296 296 296 296 296 296 296 r-squared 0.157 0.187 0.19 0.195 0.196 0.198 0.203 0.203 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 behaviorally-informed policies to reduce materialism table 7. ols regression output for the impact of the interventions on the total success component of the survey. (1) (2) (3) (4) (5) (6) (7) (8) intervention_1_selfesteem 0.374 0.196 -0.703 -0.703 -0.702 -0.64 -0.664 -0.741 (-0.418) (-0.446) (-0.508) (-0.508) (-0.509) (-0.504) (-0.498) (-0.487) intervention_2_gratitude -0.534 -1.432*** -1.432*** -1.433*** -1.426*** -1.416*** -1.485*** (-0.441) (-0.503) (-0.503) (-0.504) (-0.501) (-0.496) (-0.494) intervention_3_empathy -1.797*** -1.797*** -1.797*** -1.784*** -1.791*** -1.909*** (-0.504) (-0.504) (-0.505) (-0.504) (-0.501) (-0.492) i.control_group age 0.005 1.880** 1.912** 2.128*** (-0.064) (-0.835) (-0.826) (-0.790) age_square -0.041** -0.042** -0.047*** (-0.017) (-0.017) (-0.016) gender 0.561 0.517 (-0.363) (-0.356) financial_satisfaction -0.554*** (-0.168) life_satisfaction income_5000orless education employed urban_governorate single living_area_class health constant 3.977*** 4.155*** 5.054*** 5.054*** 4.940*** -16.03 -16.51* -16.82* (0.211) (0.262) (0.357) (0.357) (1.458) (9.745) (9.662) (9.319) observations 296 296 296 296 296 296 296 296 r-squared 0.003 0.008 0.049 0.049 0.049 0.064 0.072 0.114 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 40 khalil, mansour and biltagy table 7 (cont.). ols regression output for the impact of the interventions on the total success component of the survey. (9) (10) (11) (12) (13) (14) (15) (16) intervention_1_selfesteem -0.67 -0.755 -0.799 -0.876* -0.874* -0.899* -0.903* -0.927* (0.491) (0.490) (0.492) (0.493) (0.495) (0.498) (0.501) (0.499) intervention_2_gratitude -1.485*** -1.487*** -1.523*** -1.534*** -1.530*** -1.525*** -1.528*** -1.579*** (0.491) (0.492) (0.490) (0.491) (0.497) (0.499) (0.503) (0.507) intervention_3_empathy -1.903*** -1.897*** -1.982*** -2.066*** -2.065*** -2.085*** -2.092*** -2.219*** (0.493) (0.494) (0.497) (0.503) (0.504) (0.501) (0.519) (0.529) i.control_group age 2.219*** 2.143*** 2.147*** 2.271*** 2.272*** 2.473*** 2.475*** 2.502*** (0.785) (0.783) (0.788) (0.791) (0.790) (0.796) (0.797) (0.793) age_square -0.049*** -0.048*** -0.050*** -0.052*** -0.052*** -0.057*** -0.057*** -0.058*** (0.016) (0.016) (0.016) (0.016) (0.016) (0.016) (0.016) (0.016) gender 0.488 0.329 0.374 0.527 0.524 0.522 0.523 0.387 (0.355) (0.360) (0.357) (0.382) (0.386) (0.385) (0.385) (0.395) financial_satisfaction -0.389* -0.424** -0.409** -0.390** -0.390* -0.365* -0.363* -0.413** (0.203) (0.201) (0.200) (0.198) (0.199) (0.202) (0.202) (0.200) life_satisfaction -0.385 -0.368 -0.384* -0.395* -0.394* -0.420* -0.420* -0.446* (0.240) (0.232) (0.232) (0.231) (0.232) (0.235) (0.235) (0.235) income_5000orless 0.274 0.277 0.297 0.298 0.334 0.334 0.337 (0.124) (0.126) (0.127) (0.128) (0.135) (0.136) (0.135) education 0.455 0.498 0.499 0.505 0.509 0.547* (0.307) (0.313) (0.313) (0.315) (0.324) (0.325) employed -0.527 -0.524 -0.572 -0.574 -0.459 (0.415) (0.419) (0.418) (0.417) (0.421) urban_governorate -0.0391 -0.043 -0.0366 -0.0544 (0.367) (0.368) (0.376) (0.372) single -1.31 -1.319 -1.427 (0.894) (0.912) (0.925) living_area_class -0.0173 -0.0449 (0.235) (0.237) health 0.293 (0.205) constant -17.16* -16.73* -16.50* -18.11* -18.13* -18.84** -18.83** -19.66** (9.194) (9.156) (9.211) (9.255) (9.255) (9.227) (9.239) (9.229) observations 296 296 296 296 296 296 296 296 r-squared 0.123 0.137 0.142 0.147 0.147 0.151 0.151 0.158 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 behaviorally-informed policies to reduce materialism table 8. ols regression output for the impact of the interventions on the total centrality component of the survey. (1) (2) (3) (4) (5) (6) (7) (8) intervention_1_selfesteem -0.135 -0.0608 -0.865 -0.865 -0.873 -0.853 -0.875 -0.924 (0.464) (0.496) (0.588) (0.588) (0.585) (0.586) (0.579) (0.577) intervention_2_gratitude 0.223 -0.581 -0.581 -0.574 -0.571 -0.562 -0.605 (0.462) (0.559) (0.559) (0.561) (0.561) (0.559) (0.557) intervention_3_empathy -1.608*** -1.608*** -1.609*** -1.605*** -1.612*** -1.685*** (0.555) (0.555) (0.557) (0.557) (0.555) (0.554) i.control_group age -0.0612 0.541 0.571 0.706 (0.0794) (1.055) (1.052) (1.029) age_square -0.0132 -0.0143 -0.0178 (0.0233) (0.0233) (0.0228) gender 0.528 0.501 (0.388) (0.385) financial_satisfaction -0.344* (0.181) life_satisfaction income_5000orless education employed urban_governorate single living_area_class health constant -4.392*** -4.466*** -3.662*** -3.662*** -2.356 -9.097 -9.544 -9.735 (0.225) (0.285) (0.423) (0.423) (1.777) (11.80) (11.78) (11.57) observations 296 296 296 296 296 296 296 296 r-squared 0.000 0.001 0.029 0.029 0.032 0.033 0.039 0.053 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 42 khalil, mansour and biltagy table 8 (cont.). ols regression output for the impact of the interventions on the total centrality component of the survey. (9) (10) (11) (12) (13) (14) (15) (16) intervention_1_selfesteem -0.888 -0.999* -1.104* -1.136** -1.168** -1.161** -1.038* -1.041* (0.578) (0.572) (0.565) (0.567) (0.566) (0.567) (0.563) (0.566) intervention_2_gratitude -0.605 -0.608 -0.694 -0.698 -0.758 -0.760 -0.639 -0.645 (0.556) (0.562) (0.556) (0.556) (0.547) (0.548) (0.549) (0.553) intervention_3_empathy -1.682*** -1.674*** -1.876*** -1.912*** -1.924*** -1.919*** -1.687*** -1.703*** (0.557) (0.553) (0.542) (0.543) (0.544) (0.544) (0.546) (0.570) i.control_group age 0.751 0.653 0.662 0.715 0.681 0.626 0.549 0.553 (1.023) (1.010) (0.965) (0.963) (0.962) (0.979) (0.973) (0.971) age_square -0.018 -0.016 -0.021 -0.022 -0.022 -0.020 -0.018 -0.018 (0.022) (0.022) (0.020) (0.020) (0.020) (0.021) (0.021) (0.021) gender 0.486 0.281 0.387 0.451 0.514 0.515 0.477 0.460 (0.387) (0.395) (0.400) (0.414) (0.411) (0.411) (0.412) (0.437) financial_satisfaction -0.263 -0.308 -0.271 -0.264 -0.268 -0.275 -0.332 -0.338 (0.206) (0.204) (0.202) (0.202) (0.200) (0.203) (0.203) (0.205) life_satisfaction -0.191 -0.168 -0.207 -0.211 -0.232 -0.225 -0.250 -0.253 (0.230) (0.225) (0.228) (0.229) (0.228) (0.230) (0.228) (0.229) income_5000orless 0.356 0.361 0.370 0.362 0.353 0.328 0.328 (0.167) (0.161) (0.165) (0.164) (0.168) (0.166) (0.166) education 1.084** 1.103** 1.094** 1.092** 0.943** 0.948** (0.438) (0.442) (0.443) (0.441) (0.441) (0.444) employed -0.223 -0.290 -0.277 -0.211 -0.197 (0.463) (0.460) (0.464) (0.457) (0.457) urban_governorate 0.765* 0.766* 0.559 0.557 (0.397) (0.398) (0.406) (0.408) single 0.361 0.654 0.641 (0.887) (0.849) (0.858) living_area_class 0.562** 0.559** (0.273) (0.275) health 0.0368 (0.223) constant -9.904 -9.352 -8.799 -9.482 -9.180 -8.985 -9.017 -9.120 (11.54) (11.34) (10.92) (10.92) (10.89) (10.93) (10.81) (10.72) observations 296 296 296 296 296 296 296 296 r-squared 0.055 0.075 0.098 0.098 0.110 0.110 0.124 0.124 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 behaviorally-informed policies to reduce materialism table 9. ols regression output for the impact of the interventions on the total happiness component of the survey. (1) (2) (3) (4) (5) (6) (7) (8) intervention_1_selfesteem -0.432 -0.257 -0.122 -0.122 -0.123 -0.100 -0.154 -0.292 (0.433) (0.461) (0.528) (0.528) (0.529) (0.535) (0.526) (0.509) intervention_2_gratitude 0.527 0.662 0.662 0.663 0.666 0.689 0.566 (0.462) (0.529) (0.529) (0.530) (0.531) (0.516) (0.490) intervention_3_empathy 0.270 0.270 0.270 0.275 0.258 0.0493 (0.539) (0.539) (0.539) (0.538) (0.524) (0.488) i.control_group age -0.010 0.673 0.747 1.130 (0.071) (0.914) (0.883) (0.837) age_square -0.015 -0.017 -0.027 (0.019) (0.019) (0.018) gender 1.290*** 1.213*** (0.378) (0.355) financial_satisfaction -0.981*** (0.162) life_satisfaction income_5000orless education employed urban_governorate single living_area_class health constant 10.91*** 10.73*** 10.59*** 10.59*** 10.82*** 3.165 2.072 1.529 (0.219) (0.269) (0.372) (0.372) (1.593) (10.40) (10.07) (9.558) observations 296 296 296 296 296 296 296 296 r-squared 0.003 0.008 0.009 0.009 0.009 0.011 0.049 0.171 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 44 khalil, mansour and biltagy table 9 (cont.). ols regression output for the impact of the interventions on the total happiness component of the survey. (9) (10) (11) (12) (13) (14) (15) (16) intervention_1_selfesteem -0.256 -0.347 -0.288 -0.363 -0.344 -0.370 -0.342 -0.330 (0.508) (0.504) (0.506) (0.510) (0.509) (0.513) (0.513) (0.514) intervention_2_gratitude 0.566 0.564 0.612 0.602 0.638 0.644 0.671 0.697 (0.491) (0.488) (0.489) (0.490) (0.492) (0.492) (0.499) (0.501) intervention_3_empathy 0.0521 0.0590 0.172 0.0903 0.0976 0.0771 0.130 0.194 (0.490) (0.492) (0.497) (0.508) (0.507) (0.507) (0.523) (0.538) i.control_group age 1.176 1.095 1.090 1.211 1.231 1.436 1.419 1.405 (0.828) (0.816) (0.821) (0.820) (0.819) (0.846) (0.848) (0.848) age_square -0.028 -0.026 -0.024 -0.026 -0.026 -0.031 -0.031 -0.030 (0.018) (0.017) (0.017) (0.017) (0.017) (0.018) (0.018) (0.018) gender 1.199*** 1.030*** 0.971*** 1.121*** 1.085*** 1.083*** 1.074*** 1.143*** (0.356) (0.368) (0.370) (0.382) (0.380) (0.380) (0.383) (0.388) financial_satisfaction -0.897*** -0.935*** -0.955*** -0.937*** -0.935*** -0.909*** -0.922*** -0.897*** (0.196) (0.195) (0.194) (0.196) (0.196) (0.200) (0.202) (0.197) life_satisfaction -0.194 -0.176 -0.154 -0.164 -0.152 -0.180 -0.185 -0.172 (0.227) (0.221) (0.218) (0.222) (0.222) (0.226) (0.228) (0.228) income_5000orless 0.292 0.288 0.309 0.313 0.350 0.344 0.343 (0.132) (0.133) (0.134) (0.134) (0.139) (0.137) (0.137) education -0.606 -0.563 -0.558 -0.552 -0.585 -0.604 (0.369) (0.366) (0.368) (0.370) (0.376) (0.380) employed -0.517 -0.478 -0.527 -0.512 -0.570 (0.393) (0.395) (0.398) (0.399) (0.402) urban_governorate -0.451 -0.455 -0.502 -0.493 (0.350) (0.351) (0.358) (0.359) single -1.344 -1.277 -1.223 (1.041) (1.049) (1.049) living_area_class 0.128 0.142 (0.261) (0.262) health -0.148 (0.188) constant 1.357 1.809 1.500 -0.083 -0.261 -0.987 -0.994 -0.580 (9.496) (9.364) (9.458) (9.479) (9.467) (9.606) (9.605) (9.608) observations 296 296 296 296 296 296 296 296 r-squared 0.173 0.188 0.195 0.200 0.204 0.208 0.209 0.211 robust standard errors in parentheses. *** p<0.01, ** p<0.05, * p<0.1 behaviorally-informed policies to reduce materialism these results show the ability of these interventions on detaching the link between success and materialism among youth, and also on lowering the level of centrality of material matters in youths’ life, but not on detaching the link between materialism and happiness. in addition, results of the impact of the control variables on the overall score of materialism show a significant positive impact of age that lessens with older ages, and a positive impact of gender (being male) on the score of materialism. financial satisfaction and life satisfaction, however, showed a negative impact on materialism. 5. discussion the results obtained from the findings section confirm what the study has hypothesized. the three behavioral interventions—boosting self-esteem, fostering gratitude, and stimulating empathy— would reduce materialism levels among egyptian youth. the results confirm these hypotheses to varying extents. first, boosting self-esteem significantly reduced total materialism scores and had a notable impact on the centrality and success dimensions of materialism. these findings confirm that boosting self-esteem is an effective method for lowering materialistic tendencies. second, gratitude reduced materialism in the success dimension but showed limited effects on other dimensions, suggesting its role in helping individuals value intrinsic achievements over material acquisitions. third, among the three interventions, empathy had the strongest and most consistent impact, significantly reducing total materialism and influencing both the centrality and success dimensions. this underscores empathy’s critical role in shifting focus from self-centered material goals to communal and prosocial values. interestingly, none of the interventions succeeded in detaching the link between materialism and happiness. this indicates the inadequacy of these interventions in lowering the association of material possessions with emotional satisfaction. this link is recommended for further study in future research. however, the interventions were effective in weakening the association between success and materialism, as well as lowering the centrality of material matters in youths' lives. these findings emphasize the potential of psychological and social interventions in altering materialistic attitudes among young populations. these results align with the work of chaplin and john (2007) and park and john (2011), who found that boosting self-esteem in adolescents reduces materialism by addressing feelings of insecurity. additionally, similar effects were observed by froh et al. (2011) and lambert et al. (2009), who concluded that fostering gratitude can diminish materialistic tendencies by shifting focus away from material possessions towards more meaningful, intrinsic goals. the study’s finding that empathy has a significant negative impact on materialism supports previous research by chao (2019), who demonstrated that prosocial interventions could moderate the relationship between media consumption and materialism. this aligns with the broader understanding that empathy reduces the focus on material goods by encouraging individuals to consider the needs of others over personal acquisition. 46 khalil, mansour and biltagy the results demonstrate the practical implications of these interventions for formulating youth-focused policies in egypt. these policies should aim to address the psychological roots of materialism by promoting self-worth and communal responsibility. developing online and offline platforms that provide youth with opportunities to express their thoughts and participate in community-building activities can foster self-esteem and purpose, for example, through organized youth forums and leadership workshops. educational programs can also be altered to include self-worth, discussions on gratitude, and empathy-building exercises in school and university curricula. focusing more on extracurricular activities such as volunteering or community service projects can cultivate empathy and intrinsic values as well. in addition, partnering with religious institutions to organize youth workshops and campaigns promoting moderate religiosity and spirituality through initiatives that focus on gratitude, inner contentment and volunteerism can reinforce the importance of non-material success and well-being. nationwide campaigns celebrating acts of gratitude and empathy and showcasing stories of individuals contributing to their communities or overcoming challenges through intrinsic values, can inspire broader societal change. collaborating with influencers and public figures to model and promote behaviors centered on gratitude and empathy creates aspirational norms among youth. community centers or youth hubs can be established to facilitate activities designed to build selfesteem and empathy, such as creative workshops, team-based problem-solving events, and volunteer initiatives. partnering with local ngos to create mentorship programs pairing youth with role models who emphasize the importance of personal growth and societal contributions over material success can further reinforce these values. these policies can help detangle the strong associations between materialism and success while reducing the centrality of material possessions in youths' lives. by promoting self-worth, gratitude, and empathy, such interventions foster a culture of intrinsic values and collective wellbeing. 6. conclusions materialism is a false belief that material matters and physical well-being hold greater significance than spiritual or postmaterialist values. several studies have shown negative impacts of high materialism on individuals’ personal life, societies, and markets. this study addresses a significant research gap by focusing on a non-weird country, specifically egypt, where data on personal materialism and interventions targeting it were previously absent. conducting this research on egypt will allow the testing of existing theories on a different cultural context, which enhances the understanding of how behavioral interventions can address materialism in developing nations. the study involved 296 egyptian youth, aged 18 to 29, from 25 governorates out of egypt’s 28 governorates. the study uses experimental interventions designed to enhance self-esteem, foster gratitude, and stimulate empathy. boosting self-esteem and fostering empathy significantly reduced materialism, while gratitude showed limited but noteworthy effects. although the findings are behaviorally-informed policies to reduce materialism promising, the study’s limitations include the small sample size and reliance on self-reported data rather than behavioral measures, which may cause social desirability concerns. nevertheless, the fact that we have three treatments and a control has, to a great extent, eliminated the effect of bias. this research underscores the importance of culturally relevant, behaviorally informed policies that prioritize intrinsic values, social connection, and emotional security. by addressing the psychological roots of materialism, these interventions offer a pathway for fostering well-being and reducing the excessive prioritization of material acquisition. the study recommends embedding the three tested dimensions of this study in schools’ curricula on different levels; encouraging volunteerism and community service; engaging youth in empowerment programs; and facilitating cultural and artistic initiatives to allow youth to express themselves creatively and gain recognition for their achievements. in addition, encouraging modest religious activities that promote mindfulness, thankfulness and care for others. future research is recommended to tackle the impact of lowering materialism on consumption, market dynamics, and saved social benefit. future research should also explore larger more diverse samples, and employ longitudinal designs to assess the sustainability of these interventions, as well as integrate objective behavioral measures. additionally, investigating the interactions between these interventions and broader socioeconomic or cultural factors could yield deeper insights. references abaza, m. (2001). shopping malls, consumer culture and the reshaping of public space in egypt. theory, culture & society, 18(5), 97–122. https://doi.org/10.1177/02632760122051986 abaza, m. (2006). the changing consumer cultures of modern egypt: cairo's urban reshaping. leiden: brill. https://doi.org/10.1163/9789047410478 adib, h., & el-bassiouny, n. (2012). materialism in young consumers: an investigation of family communication patterns and parental mediation practices in egypt. journal of islamic marketing, 3(3), 255-282. https://doi.org/10.1108/17590831211259745 ahmed, j. m. y. (2018). antecedents and consequences of adolescents' materialism in egypt (doctoral dissertation, faculty of commerce, cairo university). ahuvia, a., & wong, n. (2002). personality and values based materialism: their relationship and origins. journal of consumer psychology, 12(4), 389–402. https://doi.org/10.1016/s10577408(16)30089-4 ammar, n., el-bassiouny, n., & hawash, r. (2016). materialism and healthy food consumption: can health education play a role? social business, 6(4), 377–401. https://doi.org/10.1362/204440816x14811339919813 arndt, j., solomon, s., kasser, t., & sheldon, k. m. (2004). the urge to splurge: a terror management account of materialism and consumer behavior. journal of consumer psychology, 14(3), 198–212. https://doi.org/10.1207/s15327663jcp1403_2 bagwell, l. s., & bernheim, b. d. (1996). veblen effects in a theory of conspicuous consumption. the american economic review, 86(june), 349–373. bargh, j. a., & chartrand, t. l. (2014). the mind in the middle: a practical guide to priming and automaticity research. in h. t. reis & c. m. judd (eds.), handbook of research methods in social and personality psychology (pp. 311–344). cambridge, united kingdom: cambridge university press. https://doi.org/10.1017/cbo9780511996481.017 bauer, m. a., wilkie, j. e. b., kim, j. k., & bodenhausen, g. v. (2012). cuing consumerism: situational materialism undermines personal and social well-being. psychological science, 23(5), 517–523. https://doi.org/10.1177/0956797611429579 https://doi.org/10.1177/02632760122051986 https://doi.org/10.1163/9789047410478 https://doi.org/10.1108/17590831211259745 https://doi.org/10.1016/s1057-7408(16)30089-4 https://doi.org/10.1016/s1057-7408(16)30089-4 https://doi.org/10.1362/204440816x14811339919813 https://doi.org/10.1207/s15327663jcp1403_2 https://doi.org/10.1017/cbo9780511996481.017 https://doi.org/10.1177/0956797611429579 48 khalil, mansour and biltagy belk, r. w. (1984). three scales to measure constructs related to materialism: reliability, validity, and relationships to measures of happiness. in t. kinnear (ed.), advances in consumer research (vol. 11, pp. 291–297). urbana, il: association for consumer research. bertocci, p. a., & millard, r. m. (1963). erich fromm: man’s search for freedom. in personality and the good: psychological and ethical perspectives (pp. 69-93). new york, ny: david mckay company. bock, d. e., eastman, j. k., & eastman, k. l. (2018). encouraging consumer charitable behavior: the impact of charitable motivations, gratitude, and materialism. journal of business ethics, 150(4), 1213– 1228. https://doi.org/10.1007/s10551-016-3203-x borg, i., hertel, g., & hermann, d. (2017). age and personal values: similar value circles with shifting priorities. psychology and aging, 32(7), 636–649. https://doi.org/10.1037/pag0000196 brickman, p., & campbell, d. t. (1971). hedonic relativism and planning the good society. in m. h. appley (ed.), adaptation-level theory (pp. 287–302). new york, ny: academic press. brown, k. w., & kasser, t. (2005). are psychological and ecological well-being compatible? the role of values, mindfulness, and lifestyle. social indicators research, 74, 349–368. https://doi.org/10.1007/s11205-004-8207-8 brown, k. w., kasser, t., ryan, r. m., & konow, j. (2016). materialism, spending, and affect: an eventsampling study of marketplace behavior and its affective costs. journal of happiness studies, 17(6), 2277–2292. https://doi.org/10.1007/s10902-015-9694-9 brown, k. w., kasser, t., ryan, r. m., linley, a. p., & orzech, k. (2009). when what one has is enough: mindfulness, financial desire discrepancy, and subjective well-being. journal of research in personality, 43(5), 727–736. https://doi.org/10.1016/j.jrp.2009.07.002 buijzen, m. (2007). reducing children's susceptibility to commercials: mechanisms of factual and evaluative advertising interventions. media psychology, 9(2), 411–430. https://doi.org/10.1080/15213260701291361 buijzen, m., & valkenburg, p. m. (2005). parental mediation of undesired advertising effects. journal of broadcasting & electronic media, 49(2), 153–165. https://doi.org/10.1207/s15506878jobem4902_1 burroughs, j. e., & rindfleisch, a. (2002). materialism and well-being: a conflicting values perspective. journal of consumer research, 29(3), 348–370. https://doi.org/10.1086/344429 burroughs, j. e., chaplin, l. n., pandelaere, m., norton, m. i., ordabayeva, n., gunz, a., & dinauer, l. (2013). using motivation theory to develop a transformative consumer research agenda for reducing materialism in society. journal of public policy & marketing, 32(1), 18–31. https://doi.org/10.1509/jppm.10.046 carroll, j. s., dean, l. r., call, l. l., & busby, d. m. (2011). materialism and marriage: couple profiles of congruent and incongruent spouses. journal of couple & relationship therapy: innovations in clinical & educational interventions, 10(4), 287–308. https://doi.org/10.1080/15332691.2011.613306 chang, l., & arkin, r. m. (2002). materialism as an attempt to cope with uncertainty. psychology & marketing, 19(5), 389–406. https://doi.org/10.1002/mar.10016 chao, q. (2019). the impacts of public service advertising on materialism among young children in urban china. in s. liu (ed.), 5th international conference on economics, management and humanities science (ecomhs 2019) (pp. 1106-1111). london: francis academic press. https://doi.org/10.25236/ecomhs.2019.234 chaplin, l. n., & john, d. r. (2007). growing up in a material world: age differences in materialism in children and adolescents. journal of consumer research, 34(4), 480–493. https://doi.org/10.1086/518546 chaplin, l. n., john, d. r., rindfleisch, a., & froh, j. j. (2019). the impact of gratitude on adolescent materialism and generosity. the journal of positive psychology, 14(4), 502–511. https://doi.org/10.1080/17439760.2018.1497688 clary, e. g., snyder, m., ridge, r. d., copeland, j., stukas, a. a., haugen, j., & meine, p. (1998). understanding and assessing the motivations of volunteers: a functional approach. journal of personality and social psychology, 74(6), 1516–1530. https://doi.org/10.1037/0022-3514.74.6.1516 clump, m. a., brandel, j. m., & sharpe, p. j. (2002). differences in environmental responsibility between materialistic groups. psychologia, 45(3), 155–161. https://doi.org/10.2117/psysoc.2002.155 cohen, j. b., pham, m. t., & andrade, e. b. (2008). the nature and role of affect in consumer behavior. in c. p. haugtvedt, p. m. herr, & f. r. kardes (eds.), handbook of consumer psychology (pp. 297–348). new york, ny: taylor & francis group. costin, f. (1969). the scrambled sentence test: a group measure of hostility. educational and psychological measurement, 29(2), 461–468. https://doi.org/10.1177/001316446902900224 cozzolino, p. j., staples, a. d., meyers, l. s., & samboceti, j. (2004). greed, death, and values: from terror management to transcendence management theory. personality & social psychology bulletin, 30(3), 278–292. https://doi.org/10.1177/0146167203260716 https://doi.org/10.1007/s10551-016-3203-x https://doi.org/10.1037/pag0000196 https://doi.org/10.1007/s11205-004-8207-8 https://doi.org/10.1007/s10902-015-9694-9 https://doi.org/10.1016/j.jrp.2009.07.002 https://doi.org/10.1080/15213260701291361 https://doi.org/10.1207/s15506878jobem4902_1 https://doi.org/10.1086/344429 https://doi.org/10.1509/jppm.10.046 https://doi.org/10.1080/15332691.2011.613306 https://doi.org/10.1002/mar.10016 https://doi.org/10.1086/518546 https://doi.org/10.1080/17439760.2018.1497688 https://doi.org/10.1037/0022-3514.74.6.1516 https://doi.org/10.2117/psysoc.2002.155 https://doi.org/10.1177/001316446902900224 https://doi.org/10.1177/0146167203260716 behaviorally-informed policies to reduce materialism dawoud, s. d. (2014). econometric analysis of the changes in food consumption expenditure patterns in egypt. journal of development and agricultural economics, 6(1), 1–11. https://doi.org/10.5897/jdae2013.0511 dawson, s. (1988). trait materialism: improved measures and an extension to multiple domains of life satisfaction. in s. shapiro & a. h. walle (eds.), the proceedings of the ama winter educators conference (pp. 478–481). chicago, il: american marketing association. de matos, c. a., vieira, v., bonfanti, k., & mette, f. m. b. (2019). antecedents of indebtedness for low-income consumers: the mediating role of materialism. journal of consumer marketing 36(1), 92-101. https://doi.org/10.1108/jcm-09-2017-2352 dechesne, m., pyszczynski, t., arndt, j., ransom, s., sheldon, k. m., van knippenberg, a., & janssen, j. (2003). literal and symbolic immortality: the effect of evidence of literal immortality on self-esteem striving in response to mortality salience. journal of personality & social psychology, 84(4), 722– 737. https://doi.org/10.1037/0022-3514.84.4.722 deci, e. l., & ryan, r. m. (2013). intrinsic motivation and self-determination in human behavior. new york, ny: springer science & business media. deckop, j. r., jurkiewicz, c. l., & giacalone, r. a. (2010). effects of materialism on work-related personal well-being. human relations, 63(7), 1007–1030. https://doi.org/10.1177/0018726709353953 delhey, j. (2010). from materialist to post-materialist happiness? national affluence and determinants of life satisfaction in cross-national perspective. social indicators research, 97(1), 65–84. https://doi.org/10.1007/s11205-009-9558-y dittmar, h. (2005). a new look at "compulsive buying": self discrepancies and materialistic values as predictors of compulsive buying tendency. journal of social and clinical psychology, 24(6), 832– 859. https://doi.org/10.1521/jscp.2005.24.6.832 dittmar, h., bond, r., hurst, m., & kasser, t. (2014). the relationship between materialism and personal well-being: a meta-analysis. journal of personality and social psychology, 107(5), 879–924. https://doi.org/10.1037/a0037409 donnelly, g., ksendzova, m., & howell, r. t. (2013). sadness, identity, and plastic in over-shopping: the interplay of materialism, poor credit management, and emotional buying motives in predicting compulsive buying. journal of economic psychology, 39, 113–125. https://doi.org/10.1016/j.joep.2013.07.006 drążkowski, d., & trepanowski, r. (2022). a longitudinal experimental study examining how and whether practicing acts of kindness affects materialism. international journal of environmental research and public health, 19(23), 16339. https://doi.org/10.3390/ijerph192316339 druckman, j. n., green, d. p., kuklinski, j. h., & lupia, a. (2011). an introduction to core concepts. in j. n. druckman, d. p. green, j. h. kuklinski, & a. lupia (eds.), the cambridge handbook of experimental political science (pp. 15–26). cambridge, united kingdom: cambridge university press. https://doi.org/10.1017/cbo9780511921452.002 el din, d. g., & el sahn, f. (2013). measuring the factors affecting egyptian consumers' intentions to purchase global luxury fashion brands. the business & management review, 3(4), 44. el-bassiouny, n., adib, h., karem, s., hammad, h., ammar, n., & brunner, c. (2011). slaves of consumerism: highlights of egypt post 25 january 2011. in proceedings of the international association for business and society, 22, 22–31. emerson, r. m. (1976). social exchange theory. annual review of sociology, 2, 335–362. https://doi.org/10.1146/annurev.so.02.080176.002003 emmons, r. a., & mccullough, m. e. (2003). counting blessings versus burdens: an experimental investigation of gratitude and subjective well-being in daily life. journal of personality and social psychology, 84(2), 377–389. https://doi.org/10.1037/0022-3514.84.2.377 fournier, s., & richins, m. l. (1991). some theoretical and popular notions concerning materialism. journal of social behavior and personality, 6, 403–414. fritsche, k., jonas, e., kayser, d. n., & koranyi, n. (2010). existential threat and compliance with proenvironmental norms. journal of environmental psychology, 30(1), 67–79. https://doi.org/10.1016/j.jenvp.2009.08.007 froh, j. j., fan, j., emmons, r. a., bono, g., huebner, e. s., & watkins, p. (2011). measuring gratitude in youth: assessing the psychometric properties of adult gratitude scales in children and adolescents. psychological assessment, 23(2), 311–325. https://doi.org/10.1037/a0021590 garðarsdóttir, r. b., & dittmar, h. (2012). the relationship of materialism to debt and financial well-being: the case of iceland's perceived prosperity. journal of economic psychology, 33(3), 471–481. https://doi.org/10.1016/j.joep.2011.12.008 gentina, e., shrum, l. j., & lowrey, t. m. (2018a). coping with loneliness through materialism: strategies matter for adolescent development of unethical behaviors. journal of business ethics 152(1). https://doi.org/10.1007/s10551-016-3329-x https://doi.org/10.5897/jdae2013.0511 https://doi.org/10.1108/jcm-09-2017-2352 https://doi.org/10.1037/0022-3514.84.4.722 https://doi.org/10.1177/0018726709353953 https://doi.org/10.1007/s11205-009-9558-y https://doi.org/10.1521/jscp.2005.24.6.832 https://doi.org/10.1037/a0037409 https://doi.org/10.1016/j.joep.2013.07.006 https://doi.org/10.3390/ijerph192316339 https://doi.org/10.1017/cbo9780511921452.002 https://doi.org/10.1146/annurev.so.02.080176.002003 https://doi.org/10.1037/0022-3514.84.2.377 https://doi.org/10.1016/j.jenvp.2009.08.007 https://doi.org/10.1037/a0021590 https://doi.org/10.1016/j.joep.2011.12.008 https://doi.org/10.1007/s10551-016-3329-x 50 khalil, mansour and biltagy gentina, e., shrum, l. j., lowrey, t. m., vitell, s., & rose, g. (2018b). an integrative model of the influence of parental and peer support on consumer ethical beliefs: the mediating role of self-esteem, power, and materialism. journal of business ethics 150(4). https://doi.org/10.2139/ssrn.3235207 ger, g., & belk, r. w. (1996). cross-cultural differences in materialism. journal of economic psychology, 17(1), 55–77. https://doi.org/10.1016/0167-4870(95)00035-6 goldberg, m. e., gorn, g. j., peracchio, l. a., & bamossy, g. (2003). understanding materialism among youth. journal of consumer psychology, 13(3), 278–288. https://doi.org/10.1207/s15327663jcp1303_09 greenwald, a. g., & banaji, m. r. (1995). implicit social cognition: attitudes, self-esteem, and stereotypes. psychological review, 102(1), 4–27. https://doi.org/10.1037/0033-295x.102.1.4 hainmueller, j., hopkins, d. j., & yamamoto, t. (2014). causal inference in conjoint analysis: understanding multidimensional choices via stated preference experiments. political analysis, 22(1), 1–30. https://doi.org/10.1093/pan/mpt024 hansen, p. g., & jespersen, a. m. (2013). nudge and the manipulation of choice: a framework for the responsible use of the nudge approach to behaviour change in public policy. european journal of risk regulation, 4(1), 3–28. https://doi.org/10.1017/s1867299x00002762 henry, h. m., & elwy, m. (2020). egyptian rearing practices: takafol and observance of family rituals. in b. k. ashdown, & a. n. faherty (eds.), parents and caregivers across cultures (pp. 247–257). cham, switzerland: springer. https://doi.org/10.1007/978-3-030-35590-6_17 homans, g. c. (1958). social behavior as exchange. american journal of sociology, 63(6), 597–606. https://doi.org/10.1086/222355 inglehart, r. (1971). the silent revolution in europe: intergenerational change in post-industrial societies. american political science review, 65(4), 991–1017. https://doi.org/10.2307/1953494 joye, y., bolderdijk, j. w., köster, m. a., & piff, p. k. (2020). a diminishment of desire: exposure to nature relative to urban environments dampens materialism. urban forestry & urban greening, 54, 126783. https://doi.org/10.1016/j.ufug.2020.126783 kardes, f. r., herr, p. m., & schwarz, n. (eds.). (2019). handbook of research methods in consumer psychology. new york, ny: routledge. https://doi.org/10.4324/9781351137713 kasser, t. (2002). the high price of materialism. cambridge, ma: mit press. https://doi.org/10.7551/mitpress/3501.001.0001 kasser, t. (2018). materialism and living well. in e. diener, s. oishi, & l. tay (eds.), handbook of well-being. salt lake city, ut: def publishers. kasser, t., & linn, s. (2016). growing up under corporate capitalism: the problem of marketing to children, with suggestions for policy solutions. social issues and policy review, 10(1), 122–150. https://doi.org/10.1111/sipr.12020 kasser, t., & ryan, r. m. (1993). the dark side of the american dream: correlates of intrinsic and extrinsic goals. journal of personality and social psychology, 65(2), 410–422. https://doi.org/10.1037/0022-3514.65.2.410 kasser, t., & ryan, r. m. (1996). further examining the american dream: differential correlates of intrinsic and extrinsic goals. personality and social psychology bulletin, 22(3), 280–287. https://doi.org/10.1177/0146167296223006 kasser, t., & sheldon, k. m. (2000). of wealth and death: materialism, mortality salience, and consumption behavior. psychological science, 11(4), 348–351. https://doi.org/10.1111/1467-9280.00269 kasser, t., rosenblum, k. l., sameroff, a. j., deci, e. l., ryan, r. m., et al. (2014). changes in materialism, changes in psychological well-being: evidence from three longitudinal studies and an intervention experiment. motivation and emotion, 38, 1–22. https://doi.org/10.1007/s11031013-9371-4 kilbourne, w., & pickett, g. (2008). how materialism affects environmental beliefs, concern, and environmentally responsible behavior. journal of business research, 61(9), 885–893. https://doi.org/10.1016/j.jbusres.2007.09.016 king, r. b., & datu, j. a. d. (2017). materialism does not pay: materialistic students have lower motivation, engagement, and achievement. contemporary educational psychology, 49, 289–301. https://doi.org/10.1016/j.cedpsych.2017.03.003 krosnick, j. a., & alwin, d. f. (1989). aging and susceptibility to attitude change. journal of personality and social psychology, 57(3), 416–425. https://doi.org/10.1037/0022-3514.57.3.416 lambert, n. m., fincham, f. d., stillman, t. f., & dean, l. r. (2009). more gratitude, less materialism: the mediating role of life satisfaction. journal of positive psychology, 4(1), 32–42. https://doi.org/10.1080/17439760802216311 lazarus, r. s., & lazarus, b. n. (1994). passion and reason: making sense of our emotions. oxford: oxford university press. https://doi.org/10.1093/oso/9780195087574.001.0001 https://doi.org/10.2139/ssrn.3235207 https://doi.org/10.1016/0167-4870(95)00035-6 https://doi.org/10.1207/s15327663jcp1303_09 https://doi.org/10.1037/0033-295x.102.1.4 https://doi.org/10.1093/pan/mpt024 https://doi.org/10.1017/s1867299x00002762 https://doi.org/10.1007/978-3-030-35590-6_17 https://doi.org/10.1086/222355 https://doi.org/10.2307/1953494 https://doi.org/10.1016/j.ufug.2020.126783 https://doi.org/10.4324/9781351137713 https://doi.org/10.7551/mitpress/3501.001.0001 https://doi.org/10.1111/sipr.12020 https://doi.org/10.1037/0022-3514.65.2.410 https://doi.org/10.1177/0146167296223006 https://doi.org/10.1111/1467-9280.00269 https://doi.org/10.1007/s11031-013-9371-4 https://doi.org/10.1007/s11031-013-9371-4 https://doi.org/10.1016/j.jbusres.2007.09.016 https://doi.org/10.1016/j.cedpsych.2017.03.003 https://doi.org/10.1037/0022-3514.57.3.416 https://doi.org/10.1080/17439760802216311 https://doi.org/10.1093/oso/9780195087574.001.0001 behaviorally-informed policies to reduce materialism lebaron, a. b., kelley, h. h., & carroll, j. s. (2018). money over marriage: marriage importance as a mediator between materialism and marital satisfaction. journal of family and economic issues, 39(2), 337–347. https://doi.org/10.1007/s10834-017-9563-2 lee, h., chugani, s., & namkoong, j. (2022). the role of entitlement and perceived resources in gratitude's effect on materialism: longitudinal and situational effects. journal of business research, 139, 993– 1003. https://doi.org/10.1016/j.jbusres.2021.10.029 lee, j., shrum, l. j., & yi, y. (2017). the role of cultural communication norms in social exclusion effects. journal of consumer psychology, 27(1), 108–116. https://doi.org/10.1016/j.jcps.2016.05.006 lekavičienė, r., antinienė, d., nikou, s., rūtelionė, a., šeinauskienė, b., & vaiciukynaitė, e. (2022). reducing consumer materialism and compulsive buying through emotional intelligence training amongst lithuanian students. frontiers in psychology, 13, 932395. https://doi.org/10.3389/fpsyg.2022.932395 lekes, n., hope, n. h., gouveia, l., koestner, r., & philippe, f. l. (2012). influencing value priorities and increasing well-being: the effects of reflecting on intrinsic values. journal of positive psychology, 7(3), 249–261. https://doi.org/10.1080/17439760.2012.677468 lens, i., pandelaere, m., & warlop, l. (2010). effects of advertising exposure on materialism and selfesteem: advertised luxuries as a feel-good strategy? acr north american advances, 37, 43–48. levitt, s. d., & dubner, s. j. (2005). freakonomics: a rogue economist explores the hidden side of everything. new york, ny: harper collins. loveland, k. e., smeesters, d., & mandel, n. (2010). still preoccupied with 1995: the need to belong and preference for nostalgic products. journal of consumer research, 37(3), 393–408. https://doi.org/10.1086/653043 lykins, e. l. b., segerstrom, s. c., averill, a. j., evans, d. r., & kemeny, m. e. (2007). goal shifts following reminders of mortality: reconciling posttraumatic growth and terror management theory. personality and social psychology bulletin, 33(8), 1088–1099. https://doi.org/10.1177/0146167207303015 macrae, c. n., bodenhausen, g. v., milne, a. b., & jetten, j. (1994). out of mind but back in sight: stereotypes on the rebound. journal of personality and social psychology, 67(5), 808–817. https://doi.org/10.1037/0022-3514.67.5.808 mandel, n., & heine, s. j. (1999). terror management and marketing: he who dies with the most toys wins. in e. arnould & l. scott (eds.), advances in consumer research (vol. 26, pp. 527–532). urbana, il: association for consumer research. maslow, a. h. (1970). motivation and personality (2nd ed.). new york, ny: harper & row. mathur, a. (1996). older adults' motivations for gift giving to charitable organizations: an exchange theory perspective. psychology and marketing, 13(1), 107–123. https://doi.org/10.1002/(sici)15206793(199601)13:1<107::aid-mar6>3.0.co;2-k mathur, a. (2013). materialism and charitable giving: can they co-exist? journal of consumer behaviour, 12(3), 149–158. https://doi.org/10.1002/cb.1404 mayer, j. d., salovey, p., gomberg-kaufman, s., & blainey, k. (1991). a broader conception of mood experience. journal of personality and social psychology, 60(1), 100–111. https://doi.org/10.1037/0022-3514.60.1.100 mcclelland, d. c. (1987). human motivation. cambridge, united kingdom: cambridge university press. mccullough, m. e., emmons, r. a., & tsang, j. a. (2002). the grateful disposition: a conceptual and empirical topography. journal of personality and social psychology, 82(1), 112–127. https://doi.org/10.1037/0022-3514.82.1.112 mccullough, m. e., kilpatrick, s. d., emmons, r. a., & larson, d. b. (2001). is gratitude a moral affect? psychological bulletin, 127(2), 249–266. https://doi.org/10.1037//0033-2909.127.2.249 mead, n. l., baumeister, r. f., stillman, t. f., rawn, c. d., & vohs, k. d. (2011). social exclusion causes people to spend and consume strategically in the service of affiliation. journal of consumer research, 37(5), 902–919. https://doi.org/10.1086/656667 mick, d. g. (1996). are studies of dark side variables confounded by socially desirable responding? the case of materialism. journal of consumer research, 23(2), 106–119. https://doi.org/10.1086/209470 milfont, t. l., milojev, p., & sibley, c. g. (2016). values stability and change in adulthood: a 3-year longitudinal study of rank-order stability and mean-level differences. personality and social psychology bulletin, 42(5), 572–588. https://doi.org/10.1177/0146167216639245 moldes, o., & ku, l. (2020). materialistic cues make us miserable: a meta-analysis of the experimental evidence for the effects of materialism on individual and societal well-being. psychology & marketing, 37(10), 1396–1419. https://doi.org/10.1002/mar.21387 mueller, a., mitchell, j. e., peterson, l. a., faber, r. j., steffen, k. j., crosby, r. d., & claes, l. (2011). depression, materialism, and excessive internet use in relation to compulsive buying. comprehensive psychiatry, 52(4), 420–424. https://doi.org/10.1016/j.comppsych.2010.09.001 https://doi.org/10.1007/s10834-017-9563-2 https://doi.org/10.1016/j.jbusres.2021.10.029 https://doi.org/10.1016/j.jcps.2016.05.006 https://doi.org/10.3389/fpsyg.2022.932395 https://doi.org/10.1080/17439760.2012.677468 https://doi.org/10.1086/653043 https://doi.org/10.1177/0146167207303015 https://doi.org/10.1037/0022-3514.67.5.808 https://doi.org/10.1002/cb.1404 https://doi.org/10.1037/0022-3514.60.1.100 https://doi.org/10.1037/0022-3514.82.1.112 https://doi.org/10.1037/0033-2909.127.2.249 https://doi.org/10.1086/656667 https://doi.org/10.1086/209470 https://doi.org/10.1177/0146167216639245 https://doi.org/10.1002/mar.21387 https://doi.org/10.1016/j.comppsych.2010.09.001 52 khalil, mansour and biltagy nairn, a., & opree, s. j. (2021). tv adverts, materialism, and children's self-esteem: the role of socioeconomic status. international journal of market research, 63(2), 161–176. https://doi.org/10.1177/1470785320970462 nepomuceno, m. v., & laroche, m. (2015). the impact of materialism and anti-consumption lifestyles on personal debt and account balances. journal of business research, 68(3), 654–664. https://doi.org/10.1016/j.jbusres.2014.08.006 nisbett, r. e., & wilson, t. d. (1977). telling more than we can know: verbal reports on mental processes. psychological review, 84(3), 231–259. https://doi.org/10.1037/0033-295x.84.3.231 okasha, a., al-bayoumi, e. h., & hashem, a. o. (2009). vision wings: towards a positive pattern of social values and soaring egyptians to the horizon of the future vision 2030. institute of national planning, egypt. ordabayeva, n., & chandon, p. (2011). getting ahead of the joneses: when equality increases conspicuous consumption among bottom-tier consumers. journal of consumer research, 38(1), 27–41. https://doi.org/10.1086/658165 ortony, a., clore, g. l., & collins, a. (1990). the cognitive structure of emotions. cambridge, united kingdom: cambridge university press. park, j. k., & john, d. r. (2010). more than meets the eye: the influence of implicit and explicit self-esteem on materialism. journal of consumer psychology, 21(1), 73–87. https://doi.org/10.1016/j.jcps.2010.09.001 parker, n., kasser, t., bardi, a., gatersleben, b., & druckman, a. (2020). goals for good: testing an intervention to reduce materialism in three european countries. european journal of applied positive psychology, 4(9), 1–15. pieters, r. (2013). bidirectional dynamics of materialism and loneliness: not just a vicious cycle. journal of consumer research, 40(4), 615–631. https://doi.org/10.1086/671564 pitts, r. e., & skelly, g. u. (1984). economic self-interest and other motivational factors underlying charitable giving. journal of behavioral economics, 12(2), 93–109. https://doi.org/10.1016/0090-5720(84)90005-6 ponchio, m. c., & aranha, f. (2008). materialism as a predictor variable of low-income consumer behavior when entering into installment plan agreements. journal of consumer behaviour, 7(1), 21–34. https://doi.org/10.1002/cb.234 prentice, m., kasser, t., & sheldon, k. m. (2017). openness to experience predicts intrinsic value shifts after deliberating one's own death. death studies, 42(4), 205-215. https://doi.org/10.1080/07481187.2017.1334016 puente-díaz, r., & cavazos-arroyo, j. (2019). influence of gratitude and materialism on two different conceptualizations of subjective well-being. universitas psychologica, 18(3), 1–9. https://doi.org/10.11144/javeriana.upsy18-3.igmt richins, m. l., & dawson, s. (1992). a consumer values orientation for materialism and its measurement: scale development, and validation. journal of consumer research, 19(3), 303–316. https://doi.org/10.1086/209304 ridgway, n. m., kukar-kinney, m., & monroe, k. b. (2008). an expanded conceptualization and a new measure of compulsive buying. journal of consumer research, 35(4), 622–639. https://doi.org/10.1086/591108 roberts, j. a., & clement, a. (2007). materialism and satisfaction with over-all quality of life and eight life domains. social indicators research, 82(1), 79–92. https://doi.org/10.1007/s11205-006-9015-0 rosenberg, e. l. (1998). levels of analysis and the organization of affect. review of general psychology, 2(3), 247–270. https://doi.org/10.1037/1089-2680.2.3.247 rucker, d. d., & galinsky, a. d. (2008). desire to acquire: powerlessness and compensatory consumption. journal of consumer research, 35(2), 257–267. https://doi.org/10.1086/588569 rucker, d. d., & galinsky, a. d. (2009). conspicuous consumption versus utilitarian ideals: how different levels of power shape consumer behavior. journal of experimental social psychology, 45(3), 549– 555. https://doi.org/10.1016/j.jesp.2009.01.005 rustagi, n., shrum, l. j., & lowrey, t. m. (2016, july). undermining the potential of within-domain compensatory consumption: a product’s explicit identity connection impedes self-repair. paper presented at the society for consumer psychology boutique conference on identity and consumption, chicago, il. ryan, r. m., & deci, e. l. (2000). intrinsic and extrinsic motivations: classic definitions and new directions. contemporary educational psychology, 25(1), 54–67. https://doi.org/10.1006/ceps.1999.1020 saleem, m., muntaha, s. t., durrani, a. k., & shafique, m. (2022). impact of gratitude on subjective wellbeing among pakistani youth: moderating role of materialism. clinical and counselling psychology review, 3(1), 39–58. https://doi.org/10.32350/ccpr.31.04 https://doi.org/10.1177/1470785320970462 https://doi.org/10.1016/j.jbusres.2014.08.006 https://doi.org/10.1037/0033-295x.84.3.231 https://doi.org/10.1086/658165 https://doi.org/10.1016/j.jcps.2010.09.001 https://doi.org/10.1086/671564 https://doi.org/10.1016/0090-5720(84)90005-6 https://doi.org/10.1002/cb.234 https://doi.org/10.11144/javeriana.upsy18-3.igmt https://doi.org/10.1086/209304 https://doi.org/10.1086/591108 https://doi.org/10.1007/s11205-006-9015-0 https://doi.org/10.1037/1089-2680.2.3.247 https://doi.org/10.1086/588569 https://doi.org/10.1016/j.jesp.2009.01.005 https://doi.org/10.1006/ceps.1999.1020 https://doi.org/10.32350/ccpr.31.04 behaviorally-informed policies to reduce materialism sayah, r. (2024). social class, smartphones advertising exposure, self-esteem, and materialism: the case of children in lebanon. digest of middle east studies, 33(3). https://doi.org/10.1111/dome.12329 schroeder, j. e., & dugal, s. s. (1995). psychological correlates of the materialism construct. journal of social behavior and personality, 10(1), 243–253. schudson, m. (1984). advertising, the uneasy persuasion. new york, ny: basic books. schwarz, n., & clore, g. l. (1983). mood, misattribution, and judgments of well-being: informative and directive functions of affective states. journal of personality and social psychology, 45(3), 513– 523. https://doi.org/10.1037/0022-3514.45.3.513 sheldon, k. m., & kasser, t. (1995). coherence and congruence: two aspects of personality integration. journal of personality and social psychology, 68(3), 531–543. https://doi.org/10.1037/00223514.68.3.531 sheldon, k. m., & kasser, t. (1998). pursuing personal goals: skills enable progress, but not all progress is beneficial. personality and social psychology bulletin, 24(12), 1319–1331. https://doi.org/10.1177/01461672982412006 sheldon, k. m., arndt, j., & houser-marko, l. (2003). in search of the organismic valuing process: the human tendency to move towards beneficial goal choices. journal of personality, 71(5), 835–869. https://doi.org/10.1111/1467-6494.7105006 shrum, l. j., chaplin, l. n., & lowrey, t. m. (2022). psychological causes, correlates, and consequences of materialism. consumer psychology review, 5(1), 69–86. https://doi.org/10.1002/arcp.1077 shrum, l. j., lee, j., burroughs, j. e., & rindfleisch, a. (2012). an online process model of second-order cultivation effects: how television cultivates materialism and its consequences for life satisfaction. human communication research, 37(1), 34–57. https://doi.org/10.1111/j.14682958.2010.01392.x shrum, l. j., wong, n., arif, f., chugani, s. k., gunz, a., lowrey, t. m., nairn, a., pandelaere, m., ross, m. s., ruvio, a., scott, k., & sundie, j. (2013). reconceptualizing materialism as identity goal pursuits: functions, processes, and consequences. journal of business research, 66(8), 1179–1185. https://doi.org/10.1016/j.jbusres.2012.08.010 sirgy, m. j., gurel-atay, e., webb, d., cicic, m., husic, m., ekici, a., et al. (2012). linking advertising, materialism, and life satisfaction. social indicators research, 107(1), 79–101. https://doi.org/10.1007/s11205-011-9829-2 sirgy, m. j., gurel-atay, e., webb, d., cicic, m., husic-mehmedovic, m., ekici, a., & johar, j. s. (2013). is materialism all that bad? effects on satisfaction with material life, life satisfaction, and economic motivation. social indicators research, 110(1), 349–366. https://doi.org/10.1007/s11205-0119934-2 sivanathan, n., & pettit, n. c. (2010). protecting the self through consumption: status goods as affirmational commodities. journal of experimental social psychology, 46(3), 564–570. https://doi.org/10.1016/j.jesp.2010.01.006 slater, p. (1980). wealth addiction. new york, ny: e. p. dutton. stillman, t. f., fincham, f. d., vohs, k. d., lambert, n. m., & phillips, c. a. (2012). the material and immaterial in conflict: spirituality reduces conspicuous consumption. journal of economic psychology, 33(1), 1–7. https://doi.org/10.1016/j.joep.2011.08.012 talaat, r. m. (2020). fashion consciousness, materialism and fashion clothing purchase involvement of young fashion consumers in egypt: the mediation role of materialism. journal of humanities and applied social sciences, 4(2), 132–154. https://doi.org/10.1108/jhass-02-2020-0027 thaler, r. h., & sunstein, c. r. (2008). nudge: improving decisions about health, wealth, and happiness. new haven, ct: yale university press. thaler, r. h., sunstein, c. r., & balz, j. p. (2013). choice architecture. in e. shafir (ed.), the behavioral foundations of public policy (pp. 428–439). princeton, nj: princeton university press. https://doi.org/10.2307/j.ctv550cbm.31 thibaut, j. w., & kelley, h. h. (1959). the social psychology of groups. new york, ny: wiley publishing. trzcińska, a., podsiadłowski, w., golus, p., & wieleszczyk, j. (2024). self-esteem and materialism in preschool children: the role of theory of mind and parental material indulgence. osf preprints. https://doi.org/10.31219/osf.io/7x2ef unanue, j., oriol, x., oyanedel, j. c., rubio, a., & unanue, w. (2021). gratitude at work prospectively predicts lower workplace materialism: a three-wave longitudinal study in chile. international journal of environmental research and public health, 18(7), 3787–3803. https://doi.org/10.3390/ijerph18073787 unanue, w., rempel, k., gómez, m. e., & van den broeck, a. (2017). when and why does materialism relate to employees' attitudes and well-being: the mediational role of need satisfaction and need frustration. frontiers in psychology, 8, 1755. https://doi.org/10.3389/fpsyg.2017.01755 wachtel, p. l., & blatt, s. j. (1990). perceptions of economic needs and of anticipated future income. journal of economic psychology, 11(3), 403–415. https://doi.org/10.1016/0167-4870(90)90020-a https://doi.org/10.1111/dome.12329 https://doi.org/10.1037/0022-3514.45.3.513 https://doi.org/10.1037/0022-3514.68.3.531 https://doi.org/10.1037/0022-3514.68.3.531 https://doi.org/10.1177/01461672982412006 https://doi.org/10.1111/1467-6494.7105006 https://doi.org/10.1002/arcp.1077 https://doi.org/10.1111/j.1468-2958.2010.01392.x https://doi.org/10.1111/j.1468-2958.2010.01392.x https://doi.org/10.1016/j.jbusres.2012.08.010 https://doi.org/10.1007/s11205-011-9829-2 https://doi.org/10.1007/s11205-011-9934-2 https://doi.org/10.1007/s11205-011-9934-2 https://doi.org/10.1016/j.jesp.2010.01.006 https://doi.org/10.1016/j.joep.2011.08.012 https://doi.org/10.1108/jhass-02-2020-0027 https://doi.org/10.2307/j.ctv550cbm.31 https://doi.org/10.31219/osf.io/7x2ef https://doi.org/10.3390/ijerph18073787 https://doi.org/10.3389/fpsyg.2017.01755 https://doi.org/10.1016/0167-4870(90)90020-a 54 khalil, mansour and biltagy watkins, p. c., woodward, k., stone, t., & kolts, r. l. (2003). gratitude resentment and appreciation test (grat) [database record]. apa psyctests. https://doi.org/10.1037/t60955-000 watson, j. j. (2003). the relationship of materialism to spending tendencies, saving, and debt. journal of economic psychology, 24(6), 723–739. https://doi.org/10.1016/j.joep.2003.06.001 weinstein, n., przybylski, a. k., & ryan, r. m. (2009). can nature make us more caring? effects of immersion in nature on intrinsic aspirations and generosity. personality and social psychology bulletin, 35(10), 1315–1329. https://doi.org/10.1177/0146167209341649 williams, g. c., hedberg, v. a., cox, e. m., & deci, e. l. (2000). extrinsic life goals and health-risk behaviors in adolescents. journal of applied social psychology, 30(8), 1756–1771. https://doi.org/10.1111/j.1559-1816.2000.tb02466.x winter, d. d. n. (2004). shopping for sustainability: psychological solutions to overconsumption. in t. kasser & a. d. kanner (eds.), psychology and consumer culture: the struggle for a good life in a materialistic world (pp. 161–178). washington, dc: american psychological association. https://doi.org/10.1037/10658-005 wong, n., rindfleisch, a., & burroughs, j. e. (2003). do reverse-worded items confound measures in crosscultural consumer research? the case of the material values scale. journal of consumer research, 30(1), 72–91. https://doi.org/10.1086/374697 world values survey. (2020). world values survey: 2020 [worldwide dataset]. http://www.worldvaluessurvey.org yamaguchi, m., & halberstadt, j. (2012). goals and well-being in new zealand. new zealand journal of psychology, 41(2), 5–10. yang, b., chen, s., su, s., & chen, f. (2021). outwardly strong but inwardly weak, pretensions to wealth? exploring the impact of heterogeneous high self-esteem on materialism in a self-threat situation. acta psychologica sinica, 53(6), 667–680. https://doi.org/10.3724/sp.j.1041.2021.00667 zayed, a. a. (2019). popular culture and consumerism in underdeveloped urban areas: a study of the cairene quarter of al-sharrabiyya. in d. s. teixeira (ed.), popular culture and political identity in the arab gulf states (pp. 287–312). new york, ny: routledge. https://doi.org/10.4324/9780429035289-12 zhao, j., tibber, m. s., & butler, s. (2023). the association between materialism and perceived relationship quality in young adults. current psychology, 42(27), 23437–23447. https://doi.org/10.1007/s12144-022-03353-y https://doi.org/10.1037/t60955-000 https://doi.org/10.1016/j.joep.2003.06.001 https://doi.org/10.1177/0146167209341649 https://doi.org/10.1111/j.1559-1816.2000.tb02466.x https://doi.org/10.1037/10658-005 https://doi.org/10.1086/374697 http://www.worldvaluessurvey.org/ https://doi.org/10.3724/sp.j.1041.2021.00667 https://doi.org/10.4324/9780429035289-12 https://doi.org/10.1007/s12144-022-03353-y 1. introduction 2. literature review 3. data analysis 4. findings 5. discussion 6. conclusions references