197 Abstract The article focuses on the analysis of in- ternal dynamics and controversies of patterns of county-level governance. Seven alternating confi gurations of county governance in Estonia are taken as an empirical case. We also intend to develop further a conceptual framework for analysis of county governance, drawing on the experience of developing county governance in Europe in the last decades. We reveal that dual and fused patterns of county governance have rather different roles in balancing intergovern- mental relations. Hence, when combining dual pattern with elements of fused pattern, it should be done very carefully in order to avoid deep ten- sions in the intergovernmental system. This was ignored in Estonia where, in the 1920s and the beginning of the 1990s, a strong political man- date of county governor and its role as represen- tative of the state were combined. In the 1930s the crisis of county governance was solved with a perfectly balanced fused pattern, established by an authoritarian regime. In the 2000s, Estonia faced a political as well as a structural deadlock in the development of the intergovernmental system because of the deconstruction of coun- ty governance. As a result, the crisis in one link would require a complete reorganization of cen- tral-local relations in Estonia in the second half of the 2010s. Keywords: county governance, county gov- ernor, prefect, fused pattern, dual and split hier- archy, patterns of intergovernmental relations. SOURCES OF INSTITUTIONAL INTEGRITY AND CONTRADICTIONS OF COUNTY GOVERNANCE: ANALYZING THE ESTONIAN EXPERIENCE Georg SOOTLA Liia LAANES Georg SOOTLA (corresponding author) Professor of Public Policy, Institute of Political Science and Governance, Tallinn University, Tallinn, Estonia Tel.: 00372-504.9236 E-mail: sootla.georg@gmail.com Liia LAANES PhD Candidate, Department of Political and Economic Studies, University of Helsinki, Helsinki, Finland E-mail: liia.laanes@helsinki.fi Transylvanian Review of Administrative Sciences, No. 46 E/2015, pp. 197-216 198 1. Introduction In the last decades studies of subnational government have focusedmainly either on the developments of municipal (Amnå and Montin, 2000; Loughlin, 2001; Caul- fi eld and Larsen, 2002; Kersting and Vett er, 2003; Denters and Rose, 2005) or regional tiers of government (Keating, 1998; Loughlin, 2001). Studies of county level gover- nance have focused on its single components – either on the state administration (Ha- jnal and Kovács, 2013; Bogumil and Kuhlmann, 2013) or the institution of the prefect at the county/province level (Knapp and Wright, 2006; DeMontricher, 2000; Bjørnå and Jenssen, 2006) or on county self-governing institutions (Aalbu, Böhme and Uhlin, 2008; Blom-Hansen et al., 2012). There have been only few att empts to analyze coun- ty governance as a system of interaction of institutions and actors (Leemans, 1970; Reigner, 2001; Knapp and Wright, 2006). There are at least two reasons why academic interest in county level government should not fade, especially in Central and Eastern Europe. Firstly, there is increasing need to analyze relations between central and local government from the perspective of multi level governance (Piatt oni, 2009; Peters and Pierre, 2001), in which the county level is the critical link in creating balancing mech- anisms between tiers. Secondly, in many CEE countries the reforms of local govern- ment in the beginning of the 1990s aimed to create the municipal self-government as a democratic counterweight to the state administration (Horváth, 2000). As a result, the capacity of local governance was dissipated among large number of fragmented local authorities (except in Lithuania and Bulgaria) (Lankina, Hudalla and Wollmann, 2008). The capacity of the county government was either intentionally diminished (as in the Baltic States) or it was reorganized, but not so profoundly as the municipal lev- el (Horváth, 2000). In the 2000s profound reorganizations of county level governance were launched (in Latvia, Estonia, Hungary, Slovakia, Eastern Germany) in paral- lel with the amalgamation of municipalities (in Eastern Germany, Latvia, Denmark, Greece) or the establishment of regions (in the Czech Republic, Slovakia, Poland). However, in Denmark, Latvia, Finland and Lithuania counties as administrative units were abolished (Baldersheim and Rose, 2010; Meklinand Pekola-Sjöblom, 2013) and the impact of this measure on intergovernmental relations deserves a special study. Estonia has been within the period of 1917-1940 and since 1989 a great laboratory of experimentation of various patt erns and practices of county governance. In this paper we do not intend per se to introduce the history of Estonian county governance but instead to analyze it as an empirical case in order to understand variables of in- ternal integrity as well as controversies of diff erent patt erns. Besides, we intend to de- velop further a framework for the analysis of county governance as an integrated sys- tem and to identify basic patt erns of interactions between its actors, roles and tasks. Research and data. In the analysis of county governance in Estonia in the interwar period we draw on the legislation, articles and public debates in the press, fi rst of all in local government associations’ journals, and minutes of county council meet- ings. The analysis of post-communist developments is based, supplementary to these sources, on the data archive which our research team (Saar, 2006) started to com- 199 pose from 2006 onwards. To its core is the collection of documents from government archives and from personal archives, concerning local government developments, diff erent research reports, as well as targeted interviews with actors, including with county governors and municipal leaders within last decade. 2. Organization of politico-administrative space at the county level The county, as a governance tier and unit, is not only diff erently named as coun- ty, district, province, department etc., but in EU statistics (NUTS classifi cation) some of them are treated –depending on size of population – as the lower regional units NUTS-3 (German Kreise, French département, etc.) whereas others are considered as upper local level units (LAU 1) (Polish powiat, Estonian maakond) (Eurostat, 2007). Nevertheless, it is the second tier of governance which has specifi c actors, roles and patt erns of interaction clearly diff erent from municipal as well as regional tiers. 2.1. Actors At the county level there are four institutional actors (Knapp and Wright, 2006, p. 352), whose interaction determines the character of governance at the county level. The fi rst (also, historically, the earliest) is the county prefect who is primarily rep- resentative of the state in the county. The symbolic and informal legitimacy of the prefect as the highest state offi cial in the county plays a crucial role in determining its other roles (Knapp and Wright, 2006; De Montricher, 2000). Very important for the le- gitimacy of the prefect is who appoints the prefect: the President as the head of state, like in France, or the Government on the proposal of a minister within routine civil service appointment, like in Estonia. The second set of actors are central government fi eld services which can be or- ganized as a fragmented set of functional agencies of ministries or as a more or less integrated multifunctional state offi ce in a county (like in Hungary and some Länder in Germany). True, sometimes the territorial service area of a fi eld agency and the territory of a county may not coincide. In order to contribute to the development of counties as an integrated space there is a need to coordinate activities of these agen- cies and to adapt ministerial policies to the local needs. The third actor is devolved county self-government, i.e., county council and its executive (Hoene, Baldassare and Shires, 2002; Kemp, 2008; Knapp and Wright, 2006) which accomplishes (1) specifi c local tasks which presume a larger scale than munic- ipalities can provide; (2) the central government tasks delegated to county council’s executive, and (3) assistance of smaller municipalities due to their insuffi cient capaci- ty, especially through professional advice, thus contributing additionally to territorial integrity of administration (Wollmann, 2000). Fourth, Knapp and Wright (2006) consider that important actors are various local private or voluntary organizations, associations, boards, and development councils, which can infl uence the integrity of the development of a county’s politico-adminis- trative space. 200 2.2. Roles There are three broad sets of roles the actors have at the county level: (1) govern- ing roles; (2) service provision roles; and (3) balancing role in horizontal/territorial and intergovernmental power networks. In the research literature several governing roles are identifi ed. The fi rst is – as indi- cated – representation of the state and assurance of certain uniformity and common values/standards among all actors at the county level (Bjørnå and Jenssen, 2006). The second is supervision of activities of local self-government authorities (Devas and De- lay, 2006; Bjørnå and Jenssen, 2006) as well as of state fi eld agencies and other actors concerning their compliance to laws and regulations. Usually both of these roles are accomplished by the prefect. However,in some countries (Hungary, Germany, Po- land, Estonia), the head of the county council’s executive is charged with the supervi- sory role. Sometimes the county council’s executive body, as we see in Estonia, may be assigned the role of a representative of the state. There is a widespread belief, that – especially in small countries – the merger of diff erent governing roles is justifi ed because of economy of scale. However, as we see later, this may result in controver- sies of the governance and its frequent reshuffl ing without a clear purpose, involving substantial governing gaps. Third, county authorities (usually both the prefect and council) are in charge of allocating central resources (i.e., EU funds) and developing local strategies/priorities for diff erent sectors and municipal authorities (Lidström, 2011; Sharpe, 1993). The fourth important governing role is the coordination of activities of diff erent actors and planning of the county’s development (Sharpe, 1993; Hulst, 2005; Lid- ström, 2011) in order to reduce functional fragmentation and duplication of policies. In some countries both roles are assigned – due to the absence of county level – to the board of municipalities (Finland) or development councils (Lithuania). The fi fth – the county authorities can be responsible for organizing national elections at sub-national level. Hence, the prefect or county governor faces a confl ict between its administra- tive and political roles (Knapp and Wright, 2006; De Montricher, 2000). County level actors are also responsible for provision of public services (Devas and Delay, 2006). This includes two broad classes. First are administrative services, in- cluding technical-legal supervision, enforcing standards, issuing licenses and permits and professional assistance in areas of ministerial responsibility, which are usually accomplished by state fi eld offi ces. Second are local public services, the provision of which presumes larger scale or specifi c competence that is available at the county level (Lidström, 2011, p. 21; Sharpe, 1993; Hoene, Baldassareand Shires, 2002). There are many ways for division of competences and tasks between state agencies and self-governing actors in provision of services at the county level even in a single country (Bogumil and Kuhlmann, 2013). In countries with a strong municipal tier of self-government the second tier has only selected tasks, most often managing health sector, regional roads, high schools, planning/zoning, land issues, etc. In continental countries more tasks may be assigned to the councils, but in service delivery these 201 authorities/offi cials are usually accomplishing fused roles as self-governing as well as state actors. The third role is ensuring balances in intergovernmental power networks. County gov- ernance can form a kind of system of ‘Lagrangian points’, where the county govern- ment is the balancing point or providing mechanisms for merging diff erent levels into a vertically integrated whole and therefore ensuring integrity of the intergovern- mental system. This makes county leaders important interest mediators and surely political actors (Knapp and Wright, 2006; De Montricher, 2000; Reinger, 2001). First, in multilevel governance the county governance can become a crucial actor which ensures an interactive patt ern of central-local relations (Amnå and Montin, 2000) in- stead of insulating the tiers and encouraging competition between them. This can ensure eff ective implementation of national policies in the local space, where a major part of public policies are actually accomplished, and prevent adversarial competi- tion as well as protective-parochial stance of municipalities. At the same time, county level actors may ensure the articulated input of local interests into the national pol- icymaking. Second, county council and politics are necessary channels for upwards mobility of political elites and a tier which can ensure autonomy of local political life vis-á-vis national parties’ headquarters (Sharpe, 1993). Hence, we evidenced how complex a patt ern of actors and roles of county gover- nance can be in a single country, but it is more diverse if we take into account varia- tions of actors and roles in diff erent counties. Therefore we att empt to delineate only general contours of those patt erns of interactions in order to establish main sources of synergy and tensions that diff erent patt erns may contain. 2.3. Patt erns of interaction Leemans (1970) and, later, Smith (1993) and Bennett (1997) distinguished between three basicideal type patt erns of intergovernmental relations which determine the ba- sic logic of the county level governance. The fi rst is the fused or single hierarchy patt ern in which the prefect’s offi ce, government fi eld services, and the representative local gov- ernment are integrated into a more-or-less holistic system. Probably the most fused patt ern was practiced in France before the 1982 reforms, when the prefect was not only the representative of the state but simultaneously the chair and chief executive of the department’s council. A full fusion of hierarchies in the county has not been achieved even in that case because of considerable autonomy of ministerial agencies and impact of elected council. There are a few supplementary mechanisms of fusion: 1. The guidance of central government over civil service in local administration or autonomous service provision organizations (schools, service centers); 2. Delegation of state functions to the county council which results also in more tight supervision by, and information feedback to, state or its agencies; 3. A less formal fusion stemming from rather strong interdependence between state fi eld offi ce staff and council executive and administration (Reigner, 2009), which 202 strengthens the professionalism and capacity of public service at county as well as at municipal level; and 4. County councils are rather dependent on government grants and programs whose provision and management is coordinated with fi eld agencies or/and prefect. There are also specifi c points of fusion of county governance with the fi rst level of local authorities, fi rst of all through formation of statutory cities (the Czech Republic, Austria) or Stadtkreise in Germany. Besides, county council can be formed via indirect elections and consist of representatives or higher offi cials of municipal authorities (Italy, Spain) (UCLG, 2015). The most important thing is that interaction in the framework of the fused system is carried out through complex patt erns of administrative politics between main actors. In horizontal dimension this enables in some cases to reach eff ective balance and coop- eration, but in other cases it results in tensions and competition over the leadership and competences (Reigner, 2001). In vertical dimension the fused and partly holistic hierarchy enables rather eff ectively to ensure policy responsiveness and division of labor between tiers and for county governance to be a balancing point. Decentral- ization trends, however, (like in France starting from Decentralization Act III in last years) (Sharapova, 2014) may diminish the departmental council’s tasks and strength- en regions’ roles, like in Spain and Italy, transforming the central-local relation into the multilevel interactive arrangement. The second is the dual patt ern, in which prefect, central government agencies and rather autonomous local self-governing bodies act as parallel subsystems without di- rect formal administrative links. There is also no subordination or overlap between tiers of government like in the case of the fused patt ern. The government services at county level are coordinated by the prefect via boards (Bjørnå et al., 2006) which are clearly separated from self-government services at county level. This patt ern is prac- ticed in Northern Europe. The dual patt ern establishes more or less fi rm power balance between autonomous actors/levels without any administrative hierarchy and interventions. This patt ern of balance is based, on the one hand, on strong and autonomous municipalities, which provide a major part of public services at subnational level and, on the other hand, on various mechanisms of consensual decision-making and active civic participation, which facilitate the horizontal cooperation in advancing the territorial development fi rst of all via participative planning (Zonneveld and Waterhout, 2010). Instead of administrative politics as main device of coordination, as in the fused system, we observe here a kind of politics of civic, voluntary participative coordination of vari- ous autonomous actors in networks, where the state takes increasingly a role of me- ta-governor (Sørensen and Torfi ng, 2007; Bjørnå and Jenssen, 2006). For these reasons the formal county institutions do not have anymore an important role as intermedi- ate link or lagrangian points in central-local relations and there is a trend to abolish county self-governing authorities in countries which have dual patt ern as in Latvia, 203 Denmark, Finland or to debate on that scenario (Norway, Sweden) without a fear to ruin the central-local balance in the age of multilevel governance. True, those debates are lately held also in continental countries, like in Italy and Spain. Leemans (1970) and Bennett (1997) distinguish also the split hierarchy patt ern, where county governance is accomplished exclusively by state actors where as self-govern- ment (council) is acting only at the fi rst tier. This patt ern can be relevant for analyz- ing developing countries where capacity of municipalities and local democracy are underdeveloped. In countries with councils only at the fi rst level, a wide range of cooperative arrangements of self-governing units at county level are established (like in Finland, Estonia and Slovakia) or municipalities are starting to occupy de facto the county-province space (Denmark, Lithuania, Latvia) whereas state agencies act in the larger administrative districts. For this reason we consider it as a subtype of the dual patt ern in our analysis. 3. Alternating patt erns of county governance in Estonia Estonia was until the beginning of the 20th century a part of the Russian empire and the only experience of governing for Estonians was community governance. The splintering of manorial and community powers started in 1817 and 1819 and by the end of the 19th century self-governance in rural municipalities was rather extensive. Counties and large towns were a strong hold of German elites, and nothing similar with Russian zemstvo (Shutov, 2011) self-government was known on Estonian soil. The att itudes formed by that experience of community self-governance contain the distrust of Estonians towards the government in larger units (including counties) that have been ruled by aliens. This stance revived during the Soviet period and infl u- enced also the governing culture and att itudes towards higher tiers of authority in the Republic of Estonia after the 1990s (Sootla and Lääne, 2012). 3.1. Patt erns of county governance during the interwar period During the interwar period (1917-1940) we can identify at least four patt erns of central-local relations. It is noteworthy that the new County Act was adopted only in 1938; therefore, the fi rst three patt erns were based on temporary and revised imperial laws and on informal customary rules. Estonia gained independence in 1918. The Estonian Constitution of 1920 (Sec. 75) stipulated that ‘State power is exercised at local level through the self-government bodies unless special agencies are created by the law’, thus establishing the fused patt ern. Subnational governments were formally part of holistic state hierarchy, but actually the county self-government enjoyed for a long period a large autonomy be- cause of a strong council (Kohver, 1928, p. 250). However, three consecutive changes in county governance developed rather diff erent confi gurations. The fi rst patt ern of county governance (1917-1920) was a classical fused but rather temporary confi guration. Members of the county council were elected by an electoral college composed of representatives of the municipalities; the council appointed the 204 county governor as chief executive of the collegial government. The prefect (commis- sioner) was appointed as a representative of the central government with the main task of supervision over legality of county governments’ activities. The second patt ern lasted more than a decade (1920-1934). It was a politically auton- omous county government in the framework of formally fused hierarchy. Direct elections of council were established by County Councils’ Election Act of 1920 and the council appointed the county governor and the collegial government. Ex post legal supervi- sion over the county government and cities was assigned to the central government, and supervision over the rural municipalities was assigned to the county government (Temporary Act of 1919). In the patt ern we can identify several sources of tensions. First, the county governor had the role and status of a prefect. However, because of strong political mandate from the council, the tensions between these dual roles of county governor emerged. Second, county departments were chaired by members of government, i.e., a kind of a committ ee system of council-executive was developed (Svara, 2001), which increased the political guidance over the county administration. In case of balanced party competition this patt ern enables the council to focus on stra- tegic issues and to exert eff ective political control over the autonomous executive. However, due to the oligarchic trends in Estonia (Parming, 1975), the real powers at the county level started to transfer into the hands of the governor. Third, the county government had a right to exert not only ex post legal review over decisions of mu- nicipalities, but also to block ex ante some of the councils’ decisions and to scrutiny municipalities’ accounting and management of services (Avikson, 1931). County of- fi cials also assisted municipalities in planning and service delivery because of low administrative capacity of municipalities. Fourth, major service provision functions (infrastructure-roads, health and health protection, sanitation, etc.) were the respon- sibility of county administration, but actually accomplished by autonomous public organizations (hospitals, service provision units, centers of professionals). The county budgets were, in comparison to other tiers of local government, relatively large and more dependent on the central government fi nancing than municipal budgets1. Fifth, municipalities were in the role of ‘younger brother’ at joint municipal regional and national associations of rural municipalities. These institutional features make rela- tions between municipalities and counties clearly hierarchical and political. In this way, various tensions in county governance and intergovernmental rela- tions have developed. Municipal councils became interested in applying for support to central authorities to maintain their autonomy against intervention of politically biased county administration. Representatives of central government, who should 1 The total share of local authorities in general public revenues of Estonia was 19.4%, out of that total, the revenues of eleven county governments comprised 3.8% as compared to the 15.6% of revenues for all 398 municipalities. In county government budget the proportion of tax revenues was 35%, whereas in cities’ and towns’ budgets – 41% and in rural municipalities’ budgets it was as much as 67% (Avikson, 1931, pp. 174-175). 205 supervise and direct (along sectoral lines) the collegial county government, which has a strong political mandate from the council, also need the political support of municipalities in curbing the political ambitions and bias of the intractable county’s leadership. As a result, during the 1920s, instead of further legitimizing the county self-government as a key actor, it became in the eyes of municipal and central gov- ernment elites as an unnecessary mediator between central and municipal authori- ties (Avikson, 1931, p. 167). At the end of the 1920s an institutional crisis of Estonian democracy started (Parming, 1975). County governments were accused of excessive politicization (Kohver, 1928) and debates over abolishment of the county self-govern- ment intensifi ed. In January 1934, the Estonian Parliament abolished county councils (Act of 1934). In March 1934, after the military coup, an authoritarian regime was established in Estonia. In the third (interim) patt ern – collegial county government without the council – the county government was directly subordinated to the Minister of Interior, which could appoint and dismiss not only members of government (e.g., governor) but also the se- nior staff of a county administration (Act of 1934). The decisions which were formerly in the competence of the council were simply re-assigned to the county government, but those decisions had to be approved by relevant branch ministers. The patt ern was even more peculiar than previously. First of all, central government offi cials were not able to eff ectively check the extensive amount of tasks accomplished by and decisions of the county government, which were previously checked by the council. The central government was not able, despite explicit intentions (Act of 1934), to redistribute ex- cessive responsibilities of county councils to central government agencies or munic- ipal authorities (Ministry of Interior, 1938). Hence a lesson: the county council with optimal political discretion could serve also as a supplementary balance in ensur- ing eff ective administrative coordination and oversight of actors at the county level (Reigner, 2001). The fourth –a newfused patt ern of county governance was established by the Coun- ty Act of 1938, after the new Constitution of Estonia (adopted in 1937) restored the two-tier local self-government system. It was designed rather carefully in order to abolish major controversies of former ones. The Act sought ‘on the one hand, to strengthen the position and capacity of the county self-government’s bodies and, on the other hand, to streamline the links between the county self-government and cen- tral government’ (Ministry of Interior, 1938). The new system was diff erent in several aspects. On the one hand, the council was appointed by and from the members of the plenary of the rural municipality elders and city majors. It was introduced to increase the proximity of the county govern- ment and the rural municipalities and city governments (Maddisoo, 1938, p. 135), i.e., to prevent the politically biased relations between tiers. On the other hand, the Presi- dent appointed the county governor who, as representative of the state in the county, was also the chief executive and head of the county government and the chairman of the county council. This ensured double check, from both state and municipal level, 206 over the previously too autonomous and politically unpredictable collegial county government. The county governor was supervising the activities of rural munici- palities and towns and had suspensive veto power over decisions of county council. However, if the county council did not amend its decision, the Minister of Interior decided whether to enforce the veto or not. This was an executive-council patt ern (Svara, 2001) mediated by active intervention of central authorities in case of confl icts between diff erent actors of county level governance. The new patt ern was designed as a truly fused multilevel amalgam, where at the bott om was the plenary of municipal leaders and at the top was the county governor eff ectively checked by the central government. These mechanisms ensured an eff ec- tive administrative balance of powers and also eff ective control by the authoritarian government over the self-government. We have to acknowledge that the Estonian governing elite became experienced to be able to establish highly consistent fused patt ern of county governance. However, we cannot assure how it works in practice as two years later Estonia lost its independence. This experience was however neglected after the restoration of independence. 3.2. Patt erns of county governance after the new independence The main roles and competences of the sub-national government during the Sovi- et time were concentrated to the county level whereas the municipal level self-gov- ernment was de facto abolished. In the second half of the 1980s the democratization in USSR was paralleled by some decentralization in economy and signs of market economy. The control over local economic and administrative resources went part- ly into the hands of county elites, the majority of whom became what Przeworski (1988) calls communist regime ‘soft-liners’ (Sootla and Katt ai, 2010). The county level became once more in Estonian history a crucial link in public administration in the period of deep crisis as well as a target of political contest between elites of diff erent tiers of government. 3.2.1. The dual patt ern of county governance with strong elements of fused system In 1989, because of strong bott om up pressure by citizens and direct assistance of Nordic local government associations, a diff erent – as compared to the interwar period – dual patt ern of central-local relations and two-tier self-government was es- tablished. However, due to the impact of interwar as well as Soviet periods, strong el- ements of fused organization of county governance were introduced, which installed – like in the 1920s – implicit contradiction into the county governance. The main purpose of the reform and transition period (planned from 1989 to 1994) was to restore a strong municipal level of self-government and to transfer all local tasks from the county government to the municipal level, except those which munic- ipalities cannot manage alone due to extensive scale or scope of resources and pro- fessionalism needed. County government was considered as a kind of substitute for central government and municipalities in compensating their temporary defi ciencies. 207 The actual devolution of county’s tasks to municipalities thus varied depending on optimism or cautiousness of county governors about the capacity of municipalities (Saar, 2006). Similarly, extensive tasks were left to the competence of counties because central government was not able to take responsibility for these due to substantial reforms of central institutions. Therefore, the dual patt ern was obviously a temporary arrangement. Moreover, because county government was a stronghold of communist soft-liners, but the new political elite considered the municipal level as a basis for its political power resource, a strong political cleavage – besides structural tensions – between tiers shaped the context for evolution of county governance. These tensions created an image of county governments and councils as temporary arrangements or even as obstacles of democratic reforms (CCA, 1997). The rhetoric over the expediency of county self-government, instead of analyzing its actual and possible role, became a key focus of local government developments at the outset and it lasts up to the pres- ent days. Like in the 1920s, the county governor and collegial government were appointed by the directly elected council which had considerable legislative autonomy. At the same time the governor was assigned the role of representative of the state. Thus, the patt ern combined extensive autonomy of county self-government with fused roles of head of county self-government and prefect, which is inherently a controversial amal- gam. Designers of the reform were not ready at the time of democratic enthusiasm to introduce less politicized versions of council (indirectly elected or executive-coun- cil patt ern) which could have decreased tension between dual and fused patt ern ele- ments. There were several other dimensions where the mechanisms of fusion of state/ self-governing roles were installed. First, similarly to the Soviet period the fi ve largest cities had simultaneously the status of counties which substantially decreased the failures of public service provi- sion in cities. Later this arrangement, which would bett er balance local autonomy and fusion of roles, was abolished. Second, collegial county government was established, appointed by the council (and majority coalition), however the chief of county offi ce (county secretary) was appointed by the Cabinet (central government) at the proposal of county governor. Third, county government was primarily a generalist offi ce for accomplishing many subnational tasks, which were still in the responsibility of cen- tral government. Fourth, county governor and council had in this period extensive discretion in ex ante legal supervision over the decisions of county level public (state) and private actors, including supervision over municipalities. Consequent disputes between diff erent tiers’ authorities on the legality of decisions were solved by the Supreme Council. Similarly the Cabinet had discretion to veto the decisions of the county council. This strengthened the tensions in central-local relations because the arbiter in confl icts was not the court but the administrative and representative bodies (LGFA, 1989). The county governor was responsible for the coordination of all public authorities in the county to ensure balanced and integrated local/regional develop- 208 ment. Moreover, the county governor had discretion (LGFA, 1989, Section 12) to give consent or not to the appointments of heads of ministerial fi eld offi ces and organiza- tions. Governors had also a right to participate at the Cabinet sessions with advisory vote (LGFA, 1989). These powers make the county governor extremely powerful but at the same time a very autonomous actor vis-á-vis the central government. Thus, like in the 1920s, the county governance in Estonia contained profound ten- sions caused by the mix of principles of dual patt ern with rather strong elements of fused patt ern. The most controversial was the combination of roles of prefect (as rep- resentative of the state and coordinator of state agencies) and county governor (who has a strong political mandate from the council). These tensions amplifi ed by contro- versies between old and new elites which had dominant positions in diff erent tiers of local governance. 3.2.2. Strong prefect in charge of generalist offi ce After the independence in 1991 the discussions at the Constituent Assembly in 1991 as well as the Constitution of 1992 did not yet defi nitively exclude the possibil- ity of county self-government. However, on 12 May 1993, before the Local Govern- ment Act was adopted (on 3 June 1993),the Parliament introduced a single-tier local self-government in Estonia. This was a victory of new political elites. Though, they still intended to develop in Estonia a dual patt ern of central-local relations based (as in Finland) on strong one-tier self-government and on inter-municipal cooperation arrangements (Jürgenson, 2014). The new patt ern of county government was provided by a temporary act adopted in June 1993 (CGA 1993) because the county government status was to be regulated in the Government of the Republic Act (‘GRA’). The GRA was adopted in 1995 by another coalition in which dominated the old elite. For that reason there were some principal diff erences in the status and roles of county government in those pieces of legislation. Counties were considered in previous laws and practices as historical and inte- grated territorial communities. The CGA of 1993clearly redefi ned the county as a unit of the state administration and abolished the county council, whereas the coun- ty governor became appointed by the Cabinet as a higher civil servant. In 1993 was established, as a compromise, a board of municipal leaders2 which had the right of consultative approval of the candidate of the county governor. It is important that the county government was still identifi ed as a generalist central government offi ce, which was responsible for the implementation of a large amount of tasks of central government at the county level, like infrastructure development, planning, environ- ment protection, emergency service, labor market services etc. and the county gov- ernor was the head of that offi ce. His/her role as prefect (responsible for supervision over autonomous municipal level and coordination of functional ministerial offi ces) 2 This body was abolished in 1994. 209 was listed as subordinated to the administrative-managerial role. Hence, it was not yet the classical prefect but more like a governor without council, which, as we saw in the 1930s, was a rather controversial patt ern. The collegial government was replaced by the governor although the term ‘county government’ is used up to the present day designating not the collegial cabinet but the governor’s offi ce. In the GRA, adopted in 1995, the county government was defi ned primarily as the offi ce which assists the county governor (i.e., the prefect) – as representative of the central government at the county level. The GRA did not includ eanymore the role of county government as responsible for accomplishing service provision tasks in the county. From 1993 the county governors had high status not only because they were ap- pointed by the Cabinet at the proposal of the Prime Minister, but they enjoyed high legitimacy largely inherited from the 1980s. They convened in regular (informal) county governors’ board which was a veto as well as a pressure group in articulating and channeling local level interests at the time when the legitimacy of national local government associations already declined. Strong informal and collective powers of county governors were cautiously perceived, on the one hand, by municipal elites whose power resources declined due to declining popular democratic enthusiasm and due to increasing need for capacities which many of municipalities did not have. On the other hand, county governors were important actors in the central policymak- ing and were perceived by central government as contesting their supremacy. At the end of the 1990s a strong critique of county governors’ double and balancing role was launched in the press (e.g., Postimees, 1998). County governors’ att endance of Cabinet sessions was cancelled in 1999. This could be considered as a symbolic turning point in the transformation of county gov- ernor as the balancing and mediating actor between local and central government into the mere administrative offi cial of a ministerial unit. 3.2.3. Deconstructing the county governance Already in the mid-1990s was developed a kind of informal agreement between central and municipal elite to re-allocate the administrative and service provision tasks of the county governor’s offi ce to the county’s local government associations. This might result in the establishment of a kind of dual patt ern in which the county prefect’s state representative and supervisory roles and the county level self-govern- ing roles are institutionally separated (Ministry of Interior, 1998). The structures of supra-municipal cooperation as substitute for county self-government are used also in Finland. In Estonia the county association has a status of NGO and its board is composed of mayors of municipalities, who represent municipalities’ interests. In Finland joint municipal authorities are public agencies whose boards (and also re- gional (maakunta) councils) are elected by councilors according to overall party repre- sentation of municipalities (LGA Finland, Chapter 10) and thus they can focus on the issues and interests of counties as a whole. This specifi city of Estonian county’s asso- ciation has been one of the main arguments against assigning them meaningful pub- 210 lic service provision responsibilities. However, the central government has resisted changing its status also. When the central government started to reduce powers and services provided by the county government, it transferred these not to the county as- sociations but to branch ministries and their fi eld offi ces. This process was contested by the Minister of Regional Aff airs in 2003 (Toila, 2003), who was one of the leaders of local elites. Four scenarios were proposed and debated, which included a completely dual patt ern as well as fused patt erns and also the introduction of the second tier of self-government. But proposals did not cross even the doorsill of the Cabinet: central government elites have preferred purely functional organization of intermediate lev- el governance. A similar package was proposed by the next Minister of Regional Af- fairs in 2007 (Reima, 2007) which resulted in his immediate dismissal. This is a good evidence that in case the (controversial) dual patt ern is not able to hold the balance of power between tiers the position and interests of the more powerful side – central government – starts to prevail and to shape the bias in central-local relations. The process of actual deconstruction of county politico-administrative space start- ed in 2000 with the reorganization of administrative units of county administration, which commenced with the reorganization of environment protection units into min- isterial fi eld offi ces and subordination of county level service agencies (rescue service, labor market service, environment protection fund, etc.) to branch ministries. Hospi- tals became de facto subordinated to the national health insurance fund. True, nurs- ing homes for the elderly were transferred to municipalities. The decisions upon the state budget investments for local and county level were transferred from the board of mayors presided by the county governor to the competence of ministerial units and bureaucrats. By the 2003 parliamentary elections the county governments had lost ma- jority of their administrative tasks and many county governors entered into politics. In 2004 several amendments of GRA were adopted which reduced considerably the autonomy of the county governor and its offi ce (GRA as of 2 April 2004). First, a provi- sion was introduced which enabled to dismiss the governor in case of lack of smooth cooperation with the central government, i.e., for political reasons. Second, the selec- tion of candidates for county governors was formally re-assigned from Government Offi ce to the Minister of Regional Aff airs, but actually this was made in party head- quarters based on political deals of coalition. Third, as a result, the county governor’s offi ce became a deconcentrated unit to the Ministry, and the governor became directly supervised by the Ministry’s senior staff . In the middle of the 2000s, after EU member- ship, the abolishment of autonomous state fi eld agencies at the county level and the transfer of these agencies to four large city-regions have started. Currently only few fi eld agencies are located at the county level. Hence the county governor lost also the real capacity to coordinate the activities of state fi eld agencies. Also the role of the pre- fect as a representative of the state in the county is diminished. In 2008 the Ministry of Interior commissioned an evaluation report which made a practical recommendation to decrease in the governor’s daily agenda the share of activities related to the repre- sentation of the state, and the Ministry has approved it. 211 In September 2010, at the parliamentary hearings ‘Partnership of central and local governments’ it was acknowledged that after the considerable diminishing of county governance capacity there have been important steps also towards the erosion of ca- pacity of municipalities (Estonian Parliament, 2010). In 2009, the Government unilat- erally cut the proportion of personal income tax transferred to the municipal budget and de facto blocked local government borrowing except for EU funds co-fi nancing. Since the end of the 2000s the Government started to establish state-run upper sec- ondary schools at counties’ centres, thus unilaterally taking over one of the core func- tions of municipalities. The President of Estonia at the ‘Day of rural municipalities and cities’ (March 2014) found that central government should be responsible for the provision of basic education (Ilves, 2014). Finally, in 2014 the post of Minister of Re- gional Aff airs was abolished. Surprisingly however, the most obvious scenario to abolish an already largely symbolical actor – county governor and its offi ce – is not seriously debated in Estonia. Why? First, it has become obvious in practice that functionally strong government fi eld offi ces cannot eff ectively accomplish these local (regional) tasks, which presume a ter- ritorial dimension and well-oiled coordination between county level actors to achieve expected outcomes like in regional transport and roads, matching education to the needs of the labor market, ensuring professional advice on social issues, developing local business environment etc. (Sootla and Katt ai, 2013). As indicated in an OECD report (OECD, 2011) public administration in Estonia needs urgently eff ective coor- dination mechanisms to overcome functional fragmentation not only at national but also at subnational level. Second, diff erently from those EU countries where coun- ties were abolished, in Estonia it is not possible to delegate those tasks and roles to municipalities because of their insuffi cient capacity; nor to the county’s associations which, as demonstrated, are not capable to articulate the interests of the county as a whole and do not have administrative capacity to accomplish consistent admin- istration at the county level. Third, the county governor’s offi ce has lost not only its administrative capacity but also its legitimacy in the eyes of local actors, especially after the political appointments of governors. Local actors have few incentives and litt le trust to cooperate under the leadership of the politically affi liated governor and also to assign county level tasks to the governor’s offi ce. Fourth, those tasks of territo- rial administration could not be delegated to regional self-government either, as it is currently happening in European countries, because this level of self-government is absent in Estonia. 4. Discussion and conclusion Reforms of county governance in Estonia provide rather rich and instructive les- sons of institutionalization of county governance. These developments are summa- rized in Table 1. We have made in the course of analysis of these patt erns some obser- vations and some tentative conclusions. 212 In Estonia the county level governance had high capacity and legitimacy in times of profound regime change when counties were subsidiary to the weak central and municipal government and ensured an extensive set of public services. With the sta- bilization in the country both – local as well as central government elites – started to contest the status and powers of county government. It was fi rst of all because estab- lished patt erns of actors and roles of county governance contained deep internal con- tradictions, which started to erode power position, capacity and legitimacy of county governance in Estonia in the end of the 1920s and in the 2000s. We revealed that fused and dual patt erns should be combined very carefully and to limited extent because they are based on rather diff erent mechanisms of balancing intergovernmental rela- tions. First of all, it is risky to assign to county governor simultaneously the role of representative of the state and executive of county council when the governor has strong direct political mandate from the electorate (council). Second, in the 1920s and in the 1990s a strong political mandate was paralleled with strong mandate of admin- istrative interventions into the every day aff airs of municipalities, which got political tone in the eyes of municipal elites. As a result, both central government as well as municipal elite’s pressures resulted in the abolishment of county council and in the establishment of a perfectly balanced and integrated fused patt ern, true, by authori- tarian regime, and the dual patt ern in 1993. The same miscalculation, i.e. the institu- tional merger of state’s representation and electorate representation roles was made in the course of the 1989 reform. This controversy was amplifi ed by political rivalry between tiers, when elites with Soviet background dominated at the county level and the municipal level became a stronghold for new political elites. This resulted in 1993 in abolishment of county council and establishment of a strong prefecture. Prospects for the development of specifi c dual patt ern of county governance emerged through the strengthening of county’s municipal association’s executive roles. This patt ern of dual system, which was successful in diff erent versions in Finland and enables to balance central-local relations, did not materialize in Estonia. Because Estonian lo- cal governance does not meet two important preconditions for balances: strong and increasingly capable municipalities and consensual politics, and cooperation among them. Besides, the pressures from municipal and central elites were targeted to weak- en powers and legitimacy of a strong prefect. As a result of these trends we evidence today simultaneously, on the one hand, formally autonomous municipalities (i.e., formal dual patt ern) which due to decreas- ing capacity cannot ensure anymore central-local balances that is a function and sense of dual patt ern of governance at the county level. This becomes especially obvious from the beginning of 2001 after the central government failed to accomplish amal- gamation reform in order to increase the capacity and political weight of municipali- ties, which in Denmark (2007) and Latvia (2009) enabled to abolish the second tier of government. In Estonia the power balance became increasingly biased towards the central government and has resulted in its overwhelming dominance. On the other 213 hand, we evidence the emergence of a vacuum in the politico-administrative space at county level which cannot ensure integrated intergovernmental relations in the age of multilevel governance through tools of administrative politics and coordination, appropriate to fused patt ern of governance. It is a kind of deadlock which developed step by step by unintentional as well as politically motivated decisions. It is an in- structive case: neglecting some of the basic premises in the design of one of the links in the chain of the politico-administrative system – county governance – can result in the need for a complete redesign of the whole governance machine. The need for redesign has got increasing support from diff erent parties and elites, e.g., Estonian Employers’ Confederation, Estonian Chamber of Commerce and Industry, and trade unions presented a joint statement on state reform to the Government in March 2015. Possible solutions have been put forward by the state reform program issued by the Estonian Cooperation Assembly (2014) at the President’s offi ce. Table 1: Summary of patterns of county governance in Estonia Unit / dimension 1917-20 1920-34 1934-38 1938-40 1989-93 1993-99 1999 … General pattern Second tier indirect self- government and separate prefect Second tier self-govern- ment merged with prefect’s role Prefect as the head of generalist fi eld offi ce Prefect as the head of generalist offi ce and indirect self- government Second tier local self- government and generalist offi ce merged with prefect’s role Prefect as the head of generalist fi eld offi ce Prefect as solely a representative of the state Formal pattern of central-local relations Fused Fused Fused Fused Dual Dual (Split hierar- chy) Dual (Split hierar- chy) Head of government Governor, Collegial Governor, Collegial Prefect, Collegial Prefect, Collegial Governor, collegial Prefect Prefect Council Indirectly elected Direct elec- tions No Indirectly elected Direct elec- tions No No References: 1. Aalbu, H., Böhme, K. and Uhlin, Å., Administrative Reform – Arguments and Values, Stockholm: Nordregio, 2008. 2. Act on Temporary Governance of Counties of 1934, published in Riigi Teataja, 1934, 5, 40. 3. Amnå, E. and Montin, S., Towards a New Concept of Local Self-government?: Recent Local Government Legislation in Comparative Perspective, Bergen: Fagbokforlaget, 2000. 4. Avikson, H., ‘Teise astme maa omavalitsuse reform’, 1931, Maaomavalitsus, vol. 11, pp. 167-186. 5. Baldersheim, H. and Rose L.E. (eds.), Territorial Choice: The Politics of Boundaries and Borders, Houndsmill: Palgrave Macmillan, 2010. 6. Bennett , R., Local Government in Post-Socialist Cities, LGI Discussion papers no. 2, 1997. 7. Bjørnå, H. and Jenssen, S., ‘Prefectoral Systems and Central-Local Government Rela- tions in Scandinavia’, 2006, Scandinavian Political Studies, vol. 29, no. 4, pp. 308-332. 8. Blom-Hansen, J., Munk Christiansen, P., Fimreite,A.L. and Selle, P., ‘Reform Strategies Matt er: Explaining the Perplexing Results of Regional Government Reforms in Nor- way and Denmark’, 2012, Local Government Studies, vol. 38, no. 1, pp. 71-90. 214 9. Bogumil, J. and Kuhlmann, S., Territorial Administration in Germany: Institutional Vari- ants, Reforms and Actors at the Meso-Level of Government, Conference paper, EGPA An- nual Conference, Edinburgh, UK, September 11-13, 2013. 10. Caulfi eld, J. and Larsen, H.O., Local Government at the Millennium, Opladen: Leske + Budrich, 2002. 11. CCA [Constitution and Constitutional Assemble], Põhiseadusja Põhiseaduse Assamblee, Tallinn: Juura, 1997. 12. CGA [County Government Act] of 1993, published in Riigi Teataja I 1993, 51, 696. 13. County Act of 1938, published in Riigi Teataja 1938, 43, 405. 14. De Montricher, N., ‘The Prefect and State Reform’, 2000, Public Administration, vol. 78, no. 3, pp. 657-678. 15. Denters, B. and Rose, L.E., Comparing Local Governance: Trends and Developments, Houndmills: Palgrave Macmillan, 2005. 16. Devas, N. and Delay, S., ‘Local Democracy and the Challenges of Decentralising the State: An International Perspective’, 2006, Local Government Studies, vol. 32, no. 5, pp. 677-695. 17. Estonian Cooperation Assembly, State Reform Program, 2014. 18. Estonian Parliament, Discussion on ‘Partnership of Central and Local Governments’, Sep- tember 23, 2010, [Online] available at htt p://www.riigikogu.ee/?op=steno&stcom mand=stenogramm&date=1285225500, accessed on March 1, 2015. 19. Eurostat, Regions in the European Union. Nomenclature of Territorial Units for Statistics NUTS 2006 /EU-27, Luxembourg: Offi ce for Offi cial Publications of the European Communities, 2007. 20. Hajnal, G. and Kovács, É., Coordination at the Middle: Reforms of the County Level State Administration in Hungary (1990-2012), Conference paper, EGPA Annual Conference, Edinburgh, UK, September 11-13, 2013. 21. Hoene, C., Baldassare, M. and Shires, M., ‘The Development of Counties as Munici- pal Governments: A Case Study of Los Angeles County in the Twenty-First Century’, 2002, Urban Aff airs Review, vol. 37, no. 4, pp. 575-591. 22. Horváth, T.M., Decentralization: Experiments and Reforms, Budapest: LGI, 2000. 23. Hulst, R., ‘Regional Governance in Unitary States: Lessons from the Netherlands in Comparative Perspective’, 2005, Local Government Studies, vol. 31, no. 1, pp. 99-120. 24. Ilves, T.H., ‘Speech at the “Day of Cities and Rural Municipalities”’, 2014, [Online] available at htt p://president.ee/et/ametitegevus/koned/10001-2014-03-26-11-30-14/in dex.html, accessed on March 1, 2015. 25. Jürgenson, K., Põhiseaduse Assambleest ja kohaliku omavalitsuse süsteemi taasloomisest, Pa- per presented at Polis Summer Seminar, Tallinn-Kiili, August 19, 2014. 26. Keating, M., The New Regionalism in Western Europe: Territorial Restructuring and Politi- cal Change, Cheltenham: Edward Elgar, 1998. 27. Kemp, R., ‘County Government: Past, Present, and Future’, 2008, National Civic Review, vol. 97, no. 4, pp. 52-54. 28. Kersting, N. and Vett er, A., Reforming Local Government in Europe, Opladen: Leske + Budrich, 2003. 29. Knapp, A. and Wright, V., The Government and Politics of France, 5th edition, London: Routledge, 2006. 215 30. Kohver, A., ‘Maavalitsuste ümberkorraldamise vajaduse põhjendused ja kavad’, 1928, Maaomavalitsus, vol. VI, no.16, p. 250. 31. Lankina, T., Hudalla, A. and Wollmann, H., Local Governance in Central and Eastern Europe Comparing Performance in the Czech Republic, Hungary, Poland and Russia, Hound- mills: Palgrave Macmillan, 2008. 32. Leemans, A., Changing Patt erns of Local Government, Hague: IUAL, 1970. 33. Lidström, A., ‘Regional Self-government and Democracy’, in Herrschel, T. and Tall- berg, P. (eds.), The Role of Regions? Networks, Scale, Territory, Kristianstads Boktryckeri, 2011, pp. 21-33. 34. Local Government Act – LGA Finland, [Online] available at htt p://www.fi nlex.fi /fi / laki/kaannokset/1995/en19950365.pdf, accessed on March 1, 2015. 35. Local Government Foundation Act of 1989, LGFA, published in ÜVT 1989, 34, 517. 36. Loughlin, J., Subnational Democracy in the European Union: Challenges and Opportunities, New York: Oxford University Press, 2001. 37. Maddisoo, E., ‘Uuendatud maaomavalitsused’, 1938, Maaomavalitsus, vol. 16, no. 9, pp. 131-135. 38. Meklin, P. and Pekola-Sjöblom, M. (eds.), ‘The Reform to Restructure Municipalities and Services in Finland: A Research Perspective’, 2013, Evaluation Research Programme ARTTU Studies No. 23. 39. Ministry of Interior, Conception of Public Administration Development, Tallinn: Ministry of Interior, 1998. 40. Ministry of Interior, Seletuskiri maakonnaseaduse juurde [Explanatory Memorandum to County Act], 1938, State Archives: ERA 31.3.2830. 41. OECD, Estonia: Towards a Single Government Approach, OECD Public Governance Re- views, OECD Publishing, 2011. 42. Parming, T., The Collapse of Liberal Democracy and the Rise of Authoritarianism in Estonia, London: SAGE Publications, 1975. 43. Peters, G.B. and Pierre, J., ‘Developments in Intergovernmental Relations: Towards Multi-level Governance’, 2001, Policy & Politics, vol. 29, no. 2, pp. 131-135. 44. Piatt oni, S., ‘Multi-level Governance: A Historical and Conceptual Analysis’, 2009, Journal of European Integration, vol. 31, no. 2, pp. 163-180. 45. Przeworski, A., ‘Some Problems in the Study of the Transition to Democracy’, in O’Donnell, G., Schmitt er, P.C. and Whitehead, L. (eds.), Transitions from Authoritari- an Rule: Comparative Perspectives, Baltimore: Johns Hopkins University Press, 1988, pp. 47-63. 46. Reigner, H., ‘Multi-level Governance or Co-administration? Transformation and Con- tinuity in French Local Government’, 2001, Policy & Politics, vol. 29, no. 2, pp. 181-192. 47. Reima, V., Basic Standpoints on the Reorganization of Regional Level of Administration, 2007 (unpublished). 48. Saar, M., Patt erns of Interaction of Politicians and Civil Servants in the Reform Process: A Case of Local Government Reforms 1989-2003, Tallinn University (MA thesis, supervised by G. Sootla), 2006. 49. Sharapova, O., ‘Decentralization Act III: the Era of New Relations between Central and Local Self-Government in France’, 2014, Socialiniai Tyrimai, vol. 34, no. 1, pp. 106-116. 216 50. Sharpe, L.J., ‘The European Meso: An Appraisal’, in Sharpe, L.J. (ed.), The Rise of Meso Government in Europe, London: Sage Publications, 1993, pp. 1-39. 51. Shutov, A., Rossiiiskoe zemstvo ievropeiskietradit s iimestnogosamoupravlenii a, Mos- cow: MGU, 2011. 52. Smith, G., Politics in Western Europe: A Comparative Analysis, 5th edition, Aldershot: Dartmouth, 1993. 53. Sootla, G. and Katt ai, K., ‘Estonia: Challenges and Lessons of the Development of Lo- cal Autonomy’, in Loughlin, J., Hendriks, F. and Lidström, A. (eds.), The Oxford Hand- book of Local and Regional Democracy in Europe, Oxford: Oxford University Press, 2010, pp. 576-595. 54. Sootla, G. and Katt ai, K., ‘Strategy on Consolidation of Capacity of Municipalities in Rapla County’, 2013, [Online] available at htt p://rol.raplamaa.ee/att achments/029_ ROLi%20projekti%20lõppdokument%20-%20suutlikkuse%20strateegia%202013.pdf, accessed on March 1, 2015. 55. Sootla, G. and Lääne, S., ‘Keskvalitsuse ja kohalikuomavalitsusesuhted’, in Vetik, R., Eestipoliitikajavalitsemine 1991-2011, Tallinn: TLÜ Kirjastus, 2012, pp. 278-324. 56. Sørensen, E. and Torfi ng, J., Theories of Democratic Network Governance, Houndmill: Pal- grave McMillan, 2007. 57. Svara, J., ‘Do We Still Need Model Charters? The Meaning and Relevance of Reform in the Twenty-First Century’, 2001, National Civic Review, vol. 90, no. 1, pp. 19-33. 58. Temporary Act on the Self-Government Supervision of 1919, published in Riigi Teata- ja 1919, 78, 149. 59. Toila, 2003, Proceedings of Conference ‘Regional Government Reform in Estonia’, Sep- tember 2-3, 2003, Toila (personal archive). 60. UCLG, ‘Decentralization and Local Governments in the World. Country Profi les’, [On- line] available at htt p://www.cities-localgovernments.org/gold/country_profi le_euro pa.asp#c57, accessed on March 1, 2015. 61. Wollmann, H., ‘Local Government Modernization in Germany: between Incremental- ism and Reform Waves’, 2000, Public Administration, vol. 78, no. 4, pp. 915-936. 62. Zonneveld, W. and Waterhout, B., Implications of Territorial Cohesion: An Essay,paper prepared for the RSA Annual Conference, Pécs, Hungary, May 24-26, 2010.